From X99Lynx at aol.com  Thu Jul  1 07:31:10 1999
From: X99Lynx at aol.com (X99Lynx at aol.com)
Date: Thu, 1 Jul 1999 03:31:10 EDT
Subject: indoeuropean
Message-ID: <THU.1.JUL.1999.033110.EDT.>

In a message dated 6/29/1999 12:17:30 AM, Sunnet at worldnet.att.net writes:

<<Correction: Slavic *renka/ronka means "arm", not "hand". The word for "hand"
in Russian is /kist'/ - looks like it wasn't lost there after all.

Gene>>

On the road but checking the list and I couldn't let this one go by.

'Ruka/ruce' (Czech) and 'reka/rece' (Pol) definitely are hand/hands and have
been for many centuries.  Wherever <<*renka = arm>> comes from, it
demonstrates again that Russian and 'Slavic' are not equivalents.

In Czech, 'Naruci' will translate as 'arms'.  But this is no real surprise.
I recall that 'Cheir' itself can mean either or both hand and arm in the
Illiad.

In a message dated 6/28/1999 6:16:14 PM, georg at letmail.let.LeidenUniv.nl
writes:

<</manus/ seems to be able to claim an IE pedigree as well, with as yet
unknown semantic differences between it and *ghes-r-.... The other
possibility is that the more abstract meanings (circling around "power,
ability to protect athl.")
are older for this etymon, *ghes-r- being the anatomical term from the
beginning....

Other languages have replaced this apparently oldest word for "hand", to wit
Latin manus, Gothic handus,Baltic and Slavic *renka/ronka (from a verb
meaning "to grasp", cf.Lithuanian /rinkti/.>>

There is possibly another explanation for 'manus.'  'Mane^s' appears in Greek
as early as Aristophanes as a common word for a slave.  It is explained by
Strabo later on as being a name borrowed from the Phrygian.  Strabo explains
that the self-names of foreign captives became the nick-names used by the
Greeks - 'mane^s' being one of them given as an example.  'Mancips' of course
became the Latin word for a slave, particularly a captive.  The obvious
connection between hands and manual labor - be they farmhands or otherwise -
might explain the multiple sense.  And might even be the source of the
Germanic 'mann'.  The relation between 'androo'/'aner' and 'hand'/'manus'
might also be something to look into.

[ Moderator's note:
  These words obviously cannot be related.
  --rma ]

'Cheir' itself in C. Greek suggests the same relation between status and
'hand' - 'cheiron' in Homer refers to a person of 'meaner rank', inferior.

'*ghes-r-' as the original form for hand seems to be associated in Homer
(aside from cheir?) with such things as handcrafts - 'kestoros',(embroidery),
'keiro' (carving) cf. theros (harvest), ge^s (of the earth), keranos
(pottery, anything made of earth), kiste, ches/chernips/chersos (cups and
such), ar[i]/isteros (left handed).  Perhaps this was not THE original word,
but only one tradition among a craftsperson class.

[ Moderator's note:
  These words obviously cannot be related.
  --rma ]

There is also an odd thing that happens in early Greek.  'Dexios' refering
mainly only to the right hand, but with an obvious (to me at least) relation
to 'deka' (ten) suggesting the name for the count of fingers on both hands
was shifted to describe one hand but not the other.  'Endexios' in Homer
however refers to 'from one hand to the other', literally between hands,
possibly reflecting an earlier usage.

Finally there is again the Greek 'keiro' (to cut, carve) that would explain
churgeon/surgeon much more adequately than a direct connection to /cheir/.
In Aristophanes The Birds, 'keirulos' describes a barber.

[ Moderator's note:
  These words obviously cannot be related.
  --rma ]

Regards,
Steve Long


From edsel at glo.be  Thu Jul  1 14:18:51 1999
From: edsel at glo.be (Eduard Selleslagh)
Date: Thu, 1 Jul 1999 16:18:51 +0200
Subject: Latin perfects and Fluent Etruscan in 30 days!
Message-ID: <THU.1.JUL.1999.161851.0200.>

-----Original Message-----
From: Macia Riutort Riutort <mrr at astor.urv.es>
Date: Tuesday, June 29, 1999 3:39 AM

>Es gibt eine kleine Anzahl von Wörtern im Spanischen, die im Kastilischen
>-rd- aufweisen, im Katalanischen jedoch -rr-. Zum Beispiel:
>Katalanisch: esquerre - esquerra - Spanisch izquierdo - izquierda
>Katalanisch: cerra - Spanisch cerda ("Borste")
>Katalanisch: marrà - Spanisch mardano ("Schafbock") usw.
>Da dies hauptsächlich in Wörtern aus dem Substrat vorkommt, wird
>angenommen, dass -rd- bzw. -rr- die Anpassung an das vorkastilische bzw.
>vorkatalanische Lautsystem eines ihm unbekannten Lautes -oder Lautgruppe- der
>gebenden Substratsprache.

>M.R.

>>Yes, though there is a phonological problem here, since Basque <ezker>
>>`left', definite form <ezkerra>, should not have yielded Castilian
>><izquierdo> (m.), <izquierda> (f.).  Since there is evidence that Basque
>>once had a word-forming suffix *<-do>, meaning something like `bad
>>thing', it is possible that an unrecorded Basque derivative *<ezkerdo>
>>was borrowed into Castilian before being lost from Basque itself.
>>Nobody knows.

>>Larry Trask

[Ed Selleslagh]

I just re-discovered an article you're probably interested in:

Mary Carmen Iribarren-Argaiz*: "Los vocablos en -rr- de la lengua sarda.
Conexiones con la península ibérica", Fontes Linguae Vasconum
(Pamplona/Iruñea, Navarra/Nafarroa), Nº 76, pp. 335-354, Sept.-Dic. 1997.
It has a huge bibliography.
*University of New Mexico (in 1997).

It is based upon her doctoral dissertation at the University of Florida,
Gainesville, 1995: "Origen y desarrollo de la sufijación ibero-romance
en -rr-: vinculaciones y contrastes con las otras lenguas".

Resumen:
El presente estudio examina cómo el sardo, especialmente en sus dialectos
centrales, contiene elementos relacionables con las lenguas ibéricas.  Entre
tales elementos, para los estudiosos del euskera resulta de particular
interés la existencia de antiguos vocablos sardos con formaciones en -rr- en
sus sílabas finales. En sardo estas formaciones han quedado fosilizadas y
resulta más evidente su carácter de elemento prerromano, como hemos dicho
relativamente fosilizado, de poca evolución y de muy escasa o inexistente
productividad sufijal románica.
Ciertas circunstancias históricas como colonizaciones, expediciones
militares u otros contactos de naturaleza más esporádica hacen verosímil la
hipótesis de contactos de lenguas en los que hablantes de vasco pueden haber
estado implicados.  La exposición histórica y los argumentos lingüísticos
ofrecen interesantes sugerencias en la línea de la expansión mediterránea de
vocablos vascos.

Ed.

[ moderator re-encoded ]

-----Original Message-----
From: Macia Riutort Riutort <mrr at astor.urv.es>
Date: Tuesday, June 29, 1999 3:39 AM

>Es gibt eine kleine Anzahl von W{\"o}rtern im Spanischen, die im Kastilischen
>-rd- aufweisen, im Katalanischen jedoch -rr-. Zum Beispiel:
>Katalanisch: esquerre - esquerra - Spanisch izquierdo - izquierda
>Katalanisch: cerra - Spanisch cerda ("Borste")
>Katalanisch: marr{\`a} - Spanisch mardano ("Schafbock") usw.
>Da dies haupts{\"a}chlich in W{\"o}rtern aus dem Substrat vorkommt, wird
>angenommen, dass -rd- bzw. -rr- die Anpassung an das vorkastilische bzw.
>vorkatalanische Lautsystem eines ihm unbekannten Lautes -oder Lautgruppe- der
>gebenden Substratsprache.

>M.R.

>>Yes, though there is a phonological problem here, since Basque <ezker>
>>`left', definite form <ezkerra>, should not have yielded Castilian
>><izquierdo> (m.), <izquierda> (f.).  Since there is evidence that Basque
>>once had a word-forming suffix *<-do>, meaning something like `bad
>>thing', it is possible that an unrecorded Basque derivative *<ezkerdo>
>>was borrowed into Castilian before being lost from Basque itself.
>>Nobody knows.

>>Larry Trask

[Ed Selleslagh]

I just re-discovered an article you're probably interested in:

Mary Carmen Iribarren-Argaiz*: "Los vocablos en -rr- de la lengua sarda.
Conexiones con la pen{\'i}nsula ib{\'e}rica", Fontes Linguae Vasconum
(Pamplona/Iru{\~n}ea, Navarra/Nafarroa), N{^o} 76, pp. 335-354, Sept.-Dic.
1997.
It has a huge bibliography.
*University of New Mexico (in 1997).

It is based upon her doctoral dissertation at the University of Florida,
Gainesville, 1995: "Origen y desarrollo de la sufijaci{\'o}n ibero-romance
en -rr-: vinculaciones y contrastes con las otras lenguas".

Resumen:
El presente estudio examina c{\'o}mo el sardo, especialmente en sus dialectos
centrales, contiene elementos relacionables con las lenguas ib{\'e}ricas.
Entre tales elementos, para los estudiosos del euskera resulta de particular
inter{\'e}s la existencia de antiguos vocablos sardos con formaciones en -rr-
en sus s{\'i}labas finales. En sardo estas formaciones han quedado fosilizadas
y resulta m{\'a}s evidente su car{\'a}cter de elemento prerromano, como hemos
dicho relativamente fosilizado, de poca evoluci{\'o}n y de muy escasa o
inexistente productividad sufijal rom{\'a}nica.  Ciertas circunstancias
hist{\'o}ricas como colonizaciones, expediciones militares u otros contactos de
naturaleza m{\'a}s espor{\'a}dica hacen veros{\'i}mil la hip{\'o}tesis de
contactos de lenguas en los que hablantes de vasco pueden haber estado
implicados.  La exposici{\'o}n hist{\'o}rica y los argumentos
ling{\"u\'i}sticos ofrecen interesantes sugerencias en la l{\'i}nea de la
expansi{\'o}n mediterr{\'a}nea de vocablos vascos.

Ed.


From X99Lynx at aol.com  Tue Jul  6 06:48:05 1999
From: X99Lynx at aol.com (X99Lynx at aol.com)
Date: Tue, 6 Jul 1999 02:48:05 EDT
Subject: Chronology of the breakup of Common Romance/CELTS
Message-ID: <TUE.6.JUL.1999.024805.EDT.>

In a message dated 6/18/99 10:36:23 PM, mclssaa2 at fs2.mt.umist.ac.uk wrote:

<<I have seen a suggestion that:-
  (1) The Romanians are not descended from the Dacians, but from Vlachs
(scattered nomadic mountain shepherds who are found over much of the Balkans)
who found what is now Romania....>>
<<(2) The Vlachs are descended from Latinized Illyrians who fled into the
mountains when the Avars marauded in;
  in which case the ancestors of the Romanians started learning Latin when
Rome invaded Illyria, not when Rome invaded Dacia, and that alters the
linguistic timetable a bit.>>

There's another element in this piece of history that could skew everything.

"Vlach" or "Wallach" seems to be descended from a name that was commonly
applied to Celts.  Appearing as "Walh" or "Walah" in OHG, it has interpreted
as meaning "foreigner", sometimes Roman, but in usage it is closely
associated with Celts and regions of Celtic habitation - e.g., "Wales",
"Walloon".  I seem to recall an suggestion that there may be a tie here to L.
vallum (fortified wall, the earliest meaning of "wall") and refer to the
Celtic or Romano-Celtic oppidum or walled town.

The strong Celtic presence along the Danube is attested by Classical writers
well before the common era.  E.g., Alexander fights them before turning
against the Persians and meets Celts who are from Illyria.  Galicia in
southern Poland is a region whose name remembers a Celtic presence even
farther north.  I remember an old passage where Vlachs are identified as "the
shepherds of the Romans" and in this role they may also have been imported
help as they were in northern Italy.

<<in which case the ancestors of the Romanians started learning Latin when
Rome invaded Illyria, not when Rome invaded Dacia, and that alters the
linguistic timetable a bit.>>

The ancient ethnic designations in that part of the world are a little
difficult to follow,  but it seems clear that the whole region from Illyria
to present day Romania was under Rome by 250 ace.  And there is some
possibility that Vlachs represented Romanized Celts across those regions.  In
any case, the small difference in time between Rome's entry into Illyria and
into Dacia would be a de minimis factor.

In a message dated 6/20/99 5:55:33 PM, rmccalli at sunmuw1.MUW.Edu wrote:

<<Rumanian does share a common pre-Romance substrate with Albanian
which presumibly should help pinpoint the origin of Rumanian.>>

The suggested Thraco-Illyrian-Albanian substrate has always been called
tentative, especially in light of potential borrowing across short distances
where there would have been much opportunity for contact, particularly prior
to the Slavic invasions of the Balkans.

This brings up that familiar problem of finding a Celtic substrate, one that
might be fruitful.

E.g., one example of the Albanian substrate often given (per G. Mallinson)
has been Rom. "abure"/ Alb. "avull" (steam.)  But in Gaelic I see "co-bur"
(foam), "to-bar" (well) and "bruich" (boil) - add the Gaelic "a-" (out of) to
get /a-bruich (out of boiling- steam?) and the Celtic seems as possible as
the Albanian as a source.  Another substrate example has been Rom. "vatra"/
Alb. "vatre" (hearth), but perhaps forms like the Gaelic "fadadh" (kindling)
and "bradhadair" (blazing fire) offers evidence of a common origin and
original meaning for these later similarities.

I haven't found any recent consideration of the Celtic remnants in this area,
but I would think it might offer some real possibilities.

In a message dated 6/20/99 5:55:33 PM, rmccalli at sunmuw1.MUW.Edu wrote:

<<BUT there is also an argument that Albanian is descended from Thracian
speakers who fled to the west>>

An important question would be when this migration would have occured.

<<I seem to remember reading that while Albanian has a strong substrate, that
it is significantly lacking in Ancient or Classical Greek substrate. If
this is true, then it tells us where not to look...>>

But the problem is that Rumanian does have a substantial Greek substrate (and
had a larger one centuries ago before Rumanian was "purified" in recent
times.)  Much of this is attributed to Old Church Slavonic carrying Greek
lexicon and to Greek administration under the Turks. (The highly-related
Arumanian particularly has morphological and syntactic similarities to Greek,
e.g, the use of compound pluperfect.)  But as one scholar pointed out, "The
desire to improve and purify Rumanian... has complicated the already
difficult task of carrying out research on the origins of Rumanian
vocabulary."

Furthermore, given the extreme amount of intercourse between Greek and Thrace
during those many early centuries, wouldn't one expect Thracian to have a
strong and ancient Greek substrate, especially with regard to trade items and
such?  If Albanian doesn't have such substrate, it may not be Thracian.

Regards,
S. Long


From JoatSimeon at aol.com  Wed Jul  7 04:46:51 1999
From: JoatSimeon at aol.com (JoatSimeon at aol.com)
Date: Wed, 7 Jul 1999 00:46:51 EDT
Subject: accusative and ergative languages
Message-ID: <WED.7.JUL.1999.004651.EDT.>

>proto-language at email.msn.com writes:

>So what? It only requires us to expand our universe of applicable data. The
>idea that vocabulary is irretrievably lost is jejeune.

-- actually, it's obvious.  Information vanishes as entropy increases.  If we
didn't have written or other artificially preserved examples of now-extinct
languages, much of the PIE vocabulary we have would be completely
unrecoverable.

Try reconstructing PIE using _nothing_ but contemporary Albanian and English.
You'd be hard-put to prove that such a language even existed.


From proto-language at email.msn.com  Wed Jul  7 05:18:47 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Wed, 7 Jul 1999 00:18:47 -0500
Subject: PIE vs. Proto-World (Proto-Language)
Message-ID: <WED.7.JUL.1999.001847.0500.>

Dear Steven and IEists:

 ----- Original Message -----
From: Steven Schaufele <fcosw5 at mail.scu.edu.tw>
Sent: Tuesday, June 29, 1999 1:08 PM

> I would, however, reject any suggestion that our knowledge of PIE is
> ipso facto on a par with our knowledge (or lack thereof) of
> `Proto-World'.  Though i admit frankly to being little more than an
> amateur at this game, i am quite confident of *most* of what we claim to
> know about PIE.  Although i'm not prepared to go as far as Calvert
> Watkins (i think it was?) who composed a fable in PIE, i certainly do
> not doubt that, in principle, it could be done with our current state of
> knowledge.  Whereas i regard `Proto-World' as little more than an
> entertaining fantasy.

There are others who are working on reconstructing "Proto-World" from a
totally different methodology.

What I have been working on, I call "Proto-Language".

I simply do not understand why some find it difficult to understand that
reconstructing the Proto-Language is only primarily different from
reconstructing Indo-European in the wider selection of source languages for
data. Of course, secondarily I have labored to reconstruct the underlying
monosyllables by analysis of attested compounds.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From edsel at glo.be  Wed Jul  7 10:35:36 1999
From: edsel at glo.be (Eduard Selleslagh)
Date: Wed, 7 Jul 1999 12:35:36 +0200
Subject: 50% Spanish or German, 50% Chinese
Message-ID: <WED.7.JUL.1999.123536.0200.>

[snip]

>At the risk of beating an undead horse, i would very much like it
>reiterated that not all of us have software capable of recognizing the
>means by which some lucky subscribers' software can encode diacritics
>and other expanded character sets.  And worse, some of us have software
>that recognizes such codes, but interprets them quite differently from
>the way they are intended.

[Ed Selleslagh]

I know of no software (e.g. Windows, text processors, browsers, e-mail
programs, etc.) that are incapable of handling the full 256 characters
(using 8 bits) of code page 437 or 850, unless you are using a '70's IBM
mainframe.

[ Moderator's comment:
  Try a 1990's XKL mainframe.  One that uses 7-bit ASCII natively, and barfs
  on anything else.  Like the one running this mailing list.
  --rma ]

It is simply a matter of settings which you can change easily. Unless... See
below.

>For instance, the software we have here at SCU automatically converts
>all such codes into representations of various Chinese characters.  The
>result is that i have recently received many postings including extended
>and richly-exemplified discussions of Spanish vocabulary, and most of
>the words in question have been printed half in Spanish and half in
>Chinese, to the point that i have been unable to make any sense out of
>the posting at all, and have regretfully decided i must automatically
>dump & ignore the whole discussion.

[Ed]

In the English speaking countries a number of servers, but far from all,
use only 7-bit encoding, i.e. only 128 characters.

In the case of 8-bit transmission, local (on your own PC) switching to the
right character set (code page) will solve all the problems (e.g. in
Windows: 'international settings'). You clearly use another code page in
which only the first 128 characters are recognized as classic ASCII. BUT:
languages like Chinese usually use 16-bit characters; I guess your PC is set
for a 16-bit (256 x 256 = 65,536 characters) code page, the first 128 of
which are reserved for the classic ASCII characters like in all non-Latin
code pages (e.g. Greek, Cyrillic, etc.).

Apparently, the *.xkl.com server correctly transmits 8-bit (256) codes,
which can locally (your PC) be recognized correctly using the right
settings, since I receive the diacritics correctly, using code page 850 (437
works equally well).

[ Moderator's comment:
  No, the server here transmits only 7-bit ASCII.  MIME-quoted-printable
  messages, which use only 7-bit ASCII, are translated at the *receiving* end
  into 8-bit (which I cannot read with any mail program available to me, to
  check on this).

  Further, you are assuming a Windows system, with your discussion of "code
  pages" and the like.  Neither Macintosh nor Unix systems agree with Windows
  on 8-bit conventions.  Therefore, it is unfriendly to use 8-bit characters
  in mailing list messages.
  --rma ]

Of course, there may be some people on this list that depend on a local
(somewhat old-fashioned) server that only transmits 7 bits, in which case
there is no solution to the problem.

[ Moderator's comment:
  Your moderator, for instance.
  --rma ]

It is not obvious that setting the right code page would eliminate your
'Chinese problem' because of the 16-bit setting, which creates a problem on
a deeper level than that among European-type PC's (actually their software).

>I've recently begun noticing similar problems with postings in German.

[Ed]

The solution is the same for all Western European languages (code page 437
or 850).

[ moderator snip ]

>[ Moderator's comment:
>  I have pointed this out in the past, and been roundly excoriated for my
>  point of view, taken somehow to be "English-only".  I will once again
>  suggest that we adopt a modified TeX-like accent-writing system, in which
>  the accent (in the typographical sense, which includes umlaut/diaeresis/
>  trema and the like) is written next to the character affected.  (In TeX
>  systems, it must precede, but I think that context can disambiguate for
>  human readers.)  Should I send out a list of the TeX conventions, for those
>  unused to them?
>  --rma ]

[Ed]

The list would be very welcome.
However, there is a problem: on most non-US keyboards, either the diacritics
are on dead keys (How do you write a diacritic without a letter under it? I
tried the dead key plus Spacebar: it works) or the letters with diacritics
are normal keys (Maybe you have to use ALT+number codes?).

[ Moderator's response:
  The diacritics used in TeX are 7-bit ASCII characters.  I will post a list of
  them shortly.  Let's get caught up on the backlog first.
  --rma ]


From X99Lynx at aol.com  Wed Jul  7 13:02:14 1999
From: X99Lynx at aol.com (X99Lynx at aol.com)
Date: Wed, 7 Jul 1999 09:02:14 EDT
Subject: PIE vs. Proto-World
Message-ID: <WED.7.JUL.1999.090214.EDT.>

Pat Ryan <proto-language at email.msn.com> writes:

<<Of course we have no data on the earliest human language in the same
way we have no data on IE>>

There's a slight problem here in our use of the word "data."  Data is
strictly speaking raw and awaiting interpretation.  In that sense, there is
data on both PIE and earlier languages.  The level of certainty about our
conclusions based on this data however is the real question.

Because neither PIE or early human languages are directly observable, we are
always dealing with inferences from data.  Much as with radiation or the
atom, we draw conclusions from secondary events.  These secondary events
would be the consequences of there having been a PIE or an early universal
human language (if there was one.)

The data pool in which we look for evidence of PIE is a large one - starting
with the attributes of every known IE language over thousands of years of
time.  But as large as that potential mass of data is, it is nothing compared
to the data we need to examine in pursuit of a universal human language.  How
many times as large?  Hundreds, thousands?  Maybe more?

And there is also the fact that the comparative is not available.  A
universal language excludes no past or present language, so there is no way
to differentiate x from y.  This makes the null hypothesis (there was never a
universal language) difficult to test.

Two hundred years ago it was not apparent that there had been such a thing as
PIE.  The pool of data was there, but the hypothesis of a single parent
tongue had not been made.

There is nothing inappropriate about searching for evidence of a proto-world
language.  But it should be obvious that the task is many, many times the
size of tracking PIE.

In a message dated 7/6/99 10:42:34 PM, fcosw5 at mail.scu.edu.tw responded:

<<Although i'm not prepared to go as far as Calvert Watkins (i think it was?)
who composed a fable in PIE, i certainly do not doubt that, in principle, it
could be done with our current state of knowledge.>>

I'm beginning to suspect this isn't quite true.  There are two factors of
interference with our certainty about PIE.  One is endemic to any historical
science.  As in geology, we are dealing with strata that can overlap,
disappear or appear out of chronological context.  The relationship between
Latin and Romance languages is a good example.  A second factor is semantic
shift.  This appears to be the weak link in identifying cognition and
relationships between forms in time.

regards,
Steve Long


From thorinn at diku.dk  Wed Jul  7 13:37:12 1999
From: thorinn at diku.dk (Lars Henrik Mathiesen)
Date: Wed, 7 Jul 1999 15:37:12 +0200
Subject: 50% Spanish or German, 50% Chinese
In-Reply-To: <377909D2.1335@mail.scu.edu.tw> (message from Steven Schaufele on
 Tue, 29 Jun 1999 11:00:50 -0700)
Message-ID: <WED.7.JUL.1999.153712.0200.>

> Date: Tue, 29 Jun 1999 11:00:50 -0700
> From: Steven Schaufele <fcosw5 at mail.scu.edu.tw>

> [... I] have recently received many postings including extended and
> richly-exemplified discussions of Spanish vocabulary, and most of
> the words in question have been printed half in Spanish and half in
> Chinese [...]

Check to see if the mail program and/or the terminal program you are
using has a character set setting --- and set it to Latin-1 to read
list mail. If it doesn't, enter a bug report.

(Technically, for this to happen, the software has to recognize the
MIME format, and convert from a pure-ASCII representation to 8-bit
data. The MIME format clearly and unambiguously indicates which
character set the data are in (e.g., Latin-1) --- but the software
(another bit of it, perhaps) then turns around and pretends that the
data are really in another character set, presumably the standard
Taiwanese multi-byte character set.)

> I've recently begun noticing similar problems with postings in German.

It's unavoidable that when the technology exists to use national
characters, they will be used. The dominant mail platform in the
Internet today is a browser on Windows, where users don't even know
which characters are ASCII and which aren't --- as witnessed by many
posts using the double left and right quote characters (not even part
of Latin-1).

Unicode is on its way in too --- if Microsoft Internet Explorer 5 does
not already by default send mail that is unreadable in a non-MIME
reader, I'm sure that will happen in the Windows 2000 version.

So, it may be possible to make this list into a backwater where people
learn TeX codes and refrain from using accented characters out of
politeness. But it's only a question of time before people have to
upgrade anyway, if only to be able to read mail from aunt Agatha.

Lars Mathiesen (U of Copenhagen CS Dep) <thorinn at diku.dk> (Humour NOT marked)

[ Moderator's comment:
  The mail system on which this list is operated is 7-bit ASCII only, and will
  remain so for the foreseeable future.  I'm sorry, but the lowest common
  denominator is preferable to Windows-only capabilities.  As I have noted
  elsewhere, not even other 8-bit-capable systems agree with Windows encodings.
  --rma ]


From cjustus at mail.utexas.edu  Wed Jul  7 15:32:06 1999
From: cjustus at mail.utexas.edu (Carol F. Justus)
Date: Wed, 7 Jul 1999 10:32:06 -0500
Subject: PIE vs. Proto-World
Message-ID: <WED.7.JUL.1999.103206.0500.>

August Schleicher composed the PIE Fable based on the hypothesis that
Sanskrit was closer to PIE. Hermann Hirt revised it updating with the
knowledge that Sanskrit 'a' came from PIE 'e' and 'o' among other things,
and Lehmann & Zgusta (in the Szemerenyi Fs. ca. 1980) further revised it
based on late 20th century views. I reviewed it in Language some time
thereafter suggesting that our even newer knowledge of the fact that *kwi-
was a focus marker might be integrated into yet another revision.

I'm not sure that these Fable writings reflect a confidence in our
knowledge of what PIE was like, but more likely a synthesis of the
hypothesis so far about the kind of details people usually feel more
comfortable commenting on.

I realize that opinions differ as to what one conceives the goal of
reconstruction to be. Statements like 'The reconstruction X has not been
proven', of course, fly in the face of the face of the assumption that
reconstruction is intended to reflect the hypothesis so far. Continued
revisions of the Fable, however, would seem to be a good example of the
fact that at least some IEists view reconstruction as the most plausibly
account of the known facts so far.

Carol Justus


From alderson at netcom.com  Wed Jul  7 16:26:40 1999
From: alderson at netcom.com (Richard M. Alderson III)
Date: Wed, 7 Jul 1999 09:26:40 -0700
Subject: PIE vs. Proto-World
In-Reply-To: <37790BA6.252E@mail.scu.edu.tw> (message from Steven Schaufele on
 Tue, 29 Jun 1999 11:08:38 -0700)
Message-ID: <WED.7.JUL.1999.092640.0700.>

Steven Schaufele wrote [inter alia]:

>Although i'm not prepared to go as far as Calvert Watkins (i think it was?)
>who composed a fable in PIE,

I'm not familiar with a Watkins fable, though it's possible.  Perhaps you're
thinking of Schleicher's fable (in his very Sanskrit-like PIE), which was
re-worked by Hirt (his version is filled with reduced vowels, but otherwise
looks very Neogrammarian), and most recently to my knowledge by Lehmann and
Zgusta (with Lehmann's versions of the laryngeals and of PIE syntax).

								Rich


From petegray at btinternet.com  Wed Jul  7 19:28:14 1999
From: petegray at btinternet.com (petegray)
Date: Wed, 7 Jul 1999 20:28:14 +0100
Subject: PIE vs. Proto-World
Message-ID: <WED.7.JUL.1999.202814.0100.>

Steven said:
>  Although i'm not prepared to [compose] a fable in PIE, i certainly do
> not doubt that, .. it could be done with our current state of
> knowledge.

Yes it has, quite recently.   A number of linguists were asked to give their
versions, and they were published together in a special volume.   I'm afraid
they prove you wrong - the diversity was enormous, with some linguists using
a more-or-less Brugmannian approach, and others being much more glottalic or
laryngeal.   The variety is great fun to see, and very educative!

Peter

[ Moderator's request:
  Would you mind providing bibliographical information for the volume in
  question?  I for one would be fascinated.
  --rma ]


From petegray at btinternet.com  Wed Jul  7 19:41:28 1999
From: petegray at btinternet.com (petegray)
Date: Wed, 7 Jul 1999 20:41:28 +0100
Subject: accusative and ergative languages
Message-ID: <WED.7.JUL.1999.204128.0100.>

Pat said:
> The
> idea that vocabulary is irretrievably lost is jejeune.

The idea that it is not lost is equally jejune (if I may use the English
spelling - I mean no offence).   One of the problems in PIE is identifying
which isolates are PIE and which are not,  or, where there are similar words
in two languages, which are inherited cognates, and which are loans (yes, I
know there are techniques for this - but there are still big problems).

Proto-proto- languages simply cannot be recovered with the same degree of
certainty as languages nearer to our attested texts.

Peter


From evenstar at mail.utexas.edu  Thu Jul  8 00:05:37 1999
From: evenstar at mail.utexas.edu (Shilpi Misty Bhadra)
Date: Wed, 7 Jul 1999 19:05:37 -0500
Subject: TeX conventions
In-Reply-To: <377909D2.1335@mail.scu.edu.tw>
Message-ID: <WED.7.JUL.1999.190537.0500.>

[ moderator snip ]

I am unfamilar to the TeX conventions. Would you please send me a list of
it. Thank you.

Shilpi Misty Bhadra
University of Texas at Austin
Classics/Ancient History/Humanities major
evenstar at mail.utexas.edu
512-474-2368

[ Moderator response:
  I will post a suggested style in a couple of days.  I'm trying to get the
  recent backlog under control.


From ECOLING at aol.com  Thu Jul  8 23:03:18 1999
From: ECOLING at aol.com (ECOLING at aol.com)
Date: Thu, 8 Jul 1999 19:03:18 EDT
Subject: 50% Span. or German, 50% Chinese
Message-ID: <THU.8.JUL.1999.190318.EDT.>

Steven Schaufele: English Department
Soochow University, Waishuanghsi Campus,
Taipei 11102, Taiwan, ROC

That is a very interesting problem you describe.
Unique I think, so the problem may lie in some quirk of
the software or mailers you use, or which your university uses.
It would be best I think if you directed your request to
the managers of those mailers or of that software.

I was accustomed to receiving unintelligible nonsense,
or at least material readable with difficulty,
but that seems to be decreasing, and I believe that
precisely the continued USE of the accented European vowels
will cause the makers of software to make the necessary
corrections to their software.
(Probably, in the Taiwan case, providing a mode to
select a segment of text, and interpret it as from a
different code standard than that containing Chinese characters
using the "upper ASCII" codes (128 to 255 more or less).

My recommendation is therefore just the opposite of yours.

Lloyd Anderson
Ecological Linguistics


From Anthony.Appleyard at umist.ac.uk  Fri Jul  9 07:13:49 1999
From: Anthony.Appleyard at umist.ac.uk (Anthony Appleyard)
Date: Fri, 9 Jul 1999 08:13:49 +0100
Subject: connected text in PIE; Proto-World language
Message-ID: <FRI.9.JUL.1999.081349.0100.>

[ moderator re-formatted ]

  Steven Schaufele <fcosw5 at mail.scu.edu.tw> wrote (Subject:      PIE vs.
Proto-World):-

> ... everything we have on PIE has to be starred because it's reconstructed,
> ergo hypothetical. ... Calvert Watkins (i think it was?) who composed a fable
> in PIE, i certainly do not doubt that, in principle, it could be done with
> our current state of knowledge.  Whereas I regard `Proto-World' as little
> more than an entertaining fantasy.

  (1) When neural-net computers get advanced enough for people to
simulate better than now human brain processes including language, it
would be interesting to simulate language evolution and see over how
many centuries and by what steps a language changes so much that all
trace of common ancestry vanishes behind the `noise' of accidental
resemblances.

  (2) What is the source of that fable in PIE? Has anyone made a good
collection of PIE grammar and accidence and vocabulary, enough to learn
the language from, to write connected text in it? I once heard of
someone writing a fable not in PIE but in `Aryan', i.e. the common
ancestor of Iranian and Sanskrit, with such special regional features as
turning PIE `e' and `o' into `a'. Who else has had a go at writing
connected text in PIE?

[ Moderator's response:
  August Schleicher.  Details have been noted in other answers to the original
  posting.
  --rma ]


From Anthony.Appleyard at umist.ac.uk  Fri Jul  9 07:28:38 1999
From: Anthony.Appleyard at umist.ac.uk (Anthony Appleyard)
Date: Fri, 9 Jul 1999 08:28:38 +0100
Subject: hand (was: Re: indoeuropean)
Message-ID: <FRI.9.JUL.1999.082838.0100.>

[ moderator re-formatted ]

Stefan Georg wrote:
> ...Other languages have replaced this apparently oldest word for "hand", to
> wit Latin manus, Gothic handus, Baltic and Slavic *renka/ronka (from a verb
> meaning "to grasp", cf. Lithuanian /rinkti/.  The limited distribution of
> this core-vocabulary item in IE has once given rise to the bromide that the
> early Indo-Europeans did >have feet but no hands (mocking at linguistic
> palaeontology, of course).

Compare the amount of modern English slang words for people's hands:
`grabs', `paws', `dooks', etc. Likely one relevant factor here is how
much a particular object or process is likely to accumulate slang names
for itself.


From pagos at bigfoot.com  Fri Jul  9 08:16:06 1999
From: pagos at bigfoot.com (pagos at bigfoot.com)
Date: Fri, 9 Jul 1999 10:16:06 +0200
Subject: Latin perfects and Fluent Etruscan in 30 days
In-Reply-To: <l03130309b39e90640b98@[209.147.211.55]>
Message-ID: <FRI.9.JUL.1999.101606.0200.>

At 10.13 29/06/99 -0500, Rick Mc Callister wrote:

>Mi suegro es agricultor en Costa Rica y llama "hijos" los retoños
>Vástago, para mí, es la talla de cualquier planta en
>cualquier etapa de crecimiento.

Muy estimados Señores y Amigos!

With all respect, I have to inform you that the following dictionary may
prove helpful in your postings:

Guenther Haensch - Gisela Haberkamp de Anto'n: Woerterbuch der
Landwirtschaft (Dictionary of Agriculture; Dictionnaire Agricole;
Diccionario de Agricultura; Dizionario di Agricoltura;
Sel6skoxoziajstvennyij slovar6). BLV Verlagsgesellschaft. Muenchen - Wien -
Zuerich, 1987.

It contains all the specific agricultural technical terms you might ever
need in German, English, French, Spanish, Italian and Russian. No need to
cite more knowledgeable informants, i.e. your father-in-law, the very
father of your own sons aut similia.

In Spanish you talk of _planta madre_, exactly as in English you say
"parent plant", "mother plant", "stock plant" (German "Mutterpflanze",
French "plante m`ere" or "pied m`ere", etc.). Yet I never ran across a
_[planta] hija_ (f.) in Spanish.

If you really happened to hear the word "hijo", it might be due -- IMHO --
to an improper usage and/or a rare "modismo hispanoamericano".

Cheers

       Paolo Agostini


From pagos at bigfoot.com  Fri Jul  9 08:14:59 1999
From: pagos at bigfoot.com (pagos at bigfoot.com)
Date: Fri, 9 Jul 1999 10:14:59 +0200
Subject: TeX-encoding
In-Reply-To: <l03130309b39e90640b98@[209.147.211.55]>
Message-ID: <FRI.9.JUL.1999.101459.0200.>

>[ moderator re-encoded (experimental) ]
>hab{\'i}a, reto{\~n}os, v{\'a}stago, m{\'i}, ma{\'i}z

I use TeX, but I think that this way of encoding diacritics might not be
painless for those who don't. That's why I dare suggest to use a simpler
encoding as follows:

a' e' i' o' u'
a` e` i` o` u`
a^ e^ i^ o^ u^
a" e" i" o" u"
a~ e~ i~ o~ u~
n' n^ n~
s' s^

etc.

Best regards

            Paolo Agostini

[ Moderator's response:
  The problem with a simplified method is that it makes multiple diacritics
  difficult.  Bear with me; I will post a set of suggestions shortly.
  --rma ]


From jer at cphling.dk  Fri Jul  9 13:18:22 1999
From: jer at cphling.dk (Jens Elmegaard Rasmussen)
Date: Fri, 9 Jul 1999 15:18:22 +0200
Subject: accusative and ergative languages
In-Reply-To: <Pine.GSO.4.03.9906291119320.10157-100000@rsunx.crn.cogs.susx.ac.uk>
Message-ID: <FRI.9.JUL.1999.151822.0200.>

On Tue, 29 Jun 1999, Larry Trask wrote:

[Pat Ryan, quoting LT:]

>>> For Basque, and for other ergative languages, the "passive" view of
>>> transitive sentences can be shredded, point by devastating point.

>>> Mutila jo zuen.
>>> `He hit the boy.'

[LT:]

> In English, the utterance `He hit the boy' is *only* possible in a
> context in which `he' has already been identified: otherwise it's
> gibberish.

> And the same is true of Basque <Mutila jo zuen>: it is only possible in
> a context in which the identity of the hitter is already known, and
> otherwise it's gibberish.  In no context whatever could it be
> interpreted as `The boy was hit'.  There *must* be an identified hitter
> in the discourse.

> To express `The boy was hit', Basque uses other constructions.  One
> possibility is <Mutila jo zuten>.  This is literally `They hit the boy',
> and it can be used to mean this, when the identity of `they' is known.
> But equally it can mean `The boy was hit', in a context in which the
> identity of the hitters is unknown.  In this case, it is functionally,
> though not formally, identical to English `The boy was hit'.

> But Basque also has an overt passive: <Mutila jo zen>.  This means
> literally `The boy was hit', and it can be used with no hitter
> identified.  Moreover, this construction does not allow the addition of
> an overt agent: the Basque passive permits no agent.

Excuse my interruption, but I think your discussion is missing the only
point of any interest in this context: Can "Mutila jo zuen" not *come
from* something which *originally* meant, not 'the boy was hit' pure and
simple, but specifically 'the boy was hit by him (e.g., by the one we're
talking about)'? Nobody is claiming that the passive constructions out of
which the ergative grew in a number of languages were restricted to
impersonal use. Incidentally, if they were, there would not have been
anything to put in the case that subsequently got interpreted as a
"transitive-subject, i.e. ergative case".

I don't know a first thing about Basque, though I have been intrigued by
it on many occasions, especially since it offers such a good parallel to
Old Irish in the verb where you apparently have to memorize practically
all forms (which are many) to be able to say even the simplest of things -
and that of course was also what made me stop every time I got started.
>From the primitive and casual books at my disposal I do see that "zuen"
and "zuten" mean 'he had him' and 'they had him' resp. I also believe I
see that such auxiliaries are combined with a particularly short form of
the participles, referred to by Schuchardt as the root of the participle;
and "jo" is 'stick; beat' in its shortest form, says my little dictionary;
and "mutil-a" is 'boy' with the article "-a", but without case or number
marking. Therefore my persistent question: Why can't "mutila jo zuen" and
"mutila jo zuten" reflect a construction that was earlier meant to express
'the boy, he had him hit', 'the boy, they had him hit'? Schuchardt also
gives "zen" to mean 'he was', so that if you gloss "mutila jo zen" as 'the
boy was hit', it seems there is quite a bit of agreement that the verbal
root is a participle by itself. I do not see in what way this makes the
*diachronic* interpretation of "mutila jo zuen" any different from the
Hindi preterites that are based on Sanskrit constructions of the type
"A-Nominative + B-Genitive + PPP/nom." meaning earlier "A was (verb)-ed by
B", but now simply "B (verb)-ed A." Where am I wrong?

I am not saying you have not addressed that question properly elsewhere,
only we are some on this list who do not keep abreast of scholarly
discussion concerning Basque. And the question belongs here right now -
could we have your response to it?

Cheers,
Jens


From proto-language at email.msn.com  Fri Jul  9 13:34:46 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Fri, 9 Jul 1999 08:34:46 -0500
Subject: accusative and ergative languages
Message-ID: <FRI.9.JUL.1999.083446.0500.>

Dear Larry and IEists:

 ----- Original Message -----
From: Larry Trask <larryt at cogs.susx.ac.uk>
Sent: Tuesday, June 29, 1999 5:36 AM

[ moderator snip ]

[LT]

> OK, then -- try Japanese.  Japanese does not code subjects in the verb,
> and yet omission of the subject is perfectly normal in Japanese, but an
> object is still interpreted as an object.

[PR]

In my experience, many Japanese speaking English make "excessive" use of the
passive which reflects the fact that the subject is deleted in the Japanese
from which they are roughly mentally translating.

So, if it is the point about "Noun+acc. Verb" being "ungrammatical" in an
accusative language without an expressed Noun+nom. (including
cross-reference in the verb), I guess we should ask: "Is Japanese
accusative?"

Thirdly, an object in Japanese, is normally marked by [o] but. In the proper
context,

e{'}iga ga (subjective) suki{i} desu

will be translated as "I like movies" even though [ga] identifies a subject.

Now, with only a slight difference of nuance,

go{'}han o (objective) tabema{'}shita and

go{'}han wa (topical) tabema{'}shita

will be translated as: "I have eaten dinner".

I really do think that Japanese is not the best language to illustrate your
point.

<snip>

 [PR]

>> In the sentence mentioned above:  "noun(B)+abs. verb", which is
>> interpreted as an 'activity is performed by an unspecified agent on
>> B' --- this construction perfectly meets the definition: "a
>> construction in which an intrinsically transitive verb is construed
>> in such a way that its underlying object appears as its surface
>> subject"; accordingly, it is "passive".

[LT]

> No, not so.  See below.

[PR]
That definition is from your own dictionary.

> [LT]

>>> For Basque, and for other ergative languages, the "passive" view of
>>> transitive sentences can be shredded, point by devastating point.

> [PR]

>> Perhaps we should reopen the question of where you have "shredded,
>> point by devastating point" the view that "for Basque(, and for
>> other ergative languages,) the "passive" view of transitive
>> sentences". I saw nothing that I recognized as doing this in your
>> Basque grammar.

[LT]

> That's probably because I haven't written a Basque grammar.
> I have, however, written elsewhere on this point.

[PR]

Well, _The History of Basque_ will generally pass for a grammar in my
opinion. Well, where did you do the "shredding"? In a Basque cookbook
(:-{})?

[LT]

> [on my Basque example]

>>> Mutila jo zuen.
>>> `He hit the boy.'

> [PR]

>> Keine Endung ist auch eine Endung. Surely you must have run across
>> that someplace. And your translation of "mutila jo zuen" as '(he)
>> hit the boy' is not preferable to 'the boy was hit'.

[LT]

> Sorry, not so -- not so at all.

> In English, the utterance `He hit the boy' is *only* possible in a
> context in which `he' has already been identified: otherwise it's
> gibberish.

[PR]

Rather overly broad! I have heard conversations that go along the lines of
'He hit the boy' with the adressee responding, 'Who hit him, John or Phil?'
Unless you consider clarification a gibberish-process.

[LT]

> And the same is true of Basque <Mutila jo zuen>: it is only possible in
> a context in which the identity of the hitter is already known, and
> otherwise it's gibberish.  In no context whatever could it be
> interpreted as `The boy was hit'.  There *must* be an identified hitter
> in the discourse.

[PR]

Also not: 'The boy was hit (by someone known)'?

[LT]

> To express `The boy was hit', Basque uses other constructions.  One
> possibility is <Mutila jo zuten>.  This is literally `They hit the boy',
> and it can be used to mean this, when the identity of `they' is known.
> But equally it can mean `The boy was hit', in a context in which the
> identity of the hitters is unknown.  In this case, it is functionally,
> though not formally, identical to English `The boy was hit'.

[PR]

Well, obviously, according to you, Basque makes a fine distinction in
definiteness between singular and plural constructions.

[LT]

> But Basque also has an overt passive: <Mutila jo zen>.  This means
> literally `The boy was hit', and it can be used with no hitter
> identified.  Moreover, this construction does not allow the addition of
> an overt agent: the Basque passive permits no agent.

[PR]

I suspect but cannot prove that there is *also* a modal nunace here.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From petegray at btinternet.com  Fri Jul  9 19:39:08 1999
From: petegray at btinternet.com (petegray)
Date: Fri, 9 Jul 1999 20:39:08 +0100
Subject: indoeuropean/hand
Message-ID: <FRI.9.JUL.1999.203908.0100.>

> It's an interesting metaphor
> I was also interested in why the "masculine" form manus and not *mana?

The form is only apparent.   It is actually a -u stem, whereas the
masculines that you are thinking of are thematic (-o/e).   The -u stems can
be any gender, although masculines do predominate.   There are also a few -
a very few -  feminine thematic stems.

Peter


From jer at cphling.dk  Fri Jul  9 22:00:54 1999
From: jer at cphling.dk (Jens Elmegaard Rasmussen)
Date: Sat, 10 Jul 1999 00:00:54 +0200
Subject: accusative and ergative languages
In-Reply-To: <Pine.GSO.4.03.9906291609240.10157-100000@rsunx.crn.cogs.susx.ac.uk>
Message-ID: <SAT.10.JUL.1999.000054.0200.>

On Tue, 29 Jun 1999, Larry Trask wrote:

[...]

> There are two separate issues here, the synchronic one and the
> diachronic one.

> (1) Is a particular ergative construction "really" passive in nature?

That was not meant to be my question.

> (2) Does a particular ergative construction descend by reanalysis from
> an earlier passive?

That was what I was driving at, albeit only as a possibility, not as a
pervasive solution of all ergative.

> For the diachronic question (2), the answer "yes" has been defended in
> some particular cases, including Indo-Iranian.  But a passive origin for
> the Indo-Iranian ergative constructions has also been disputed, and I am
> not aware that there exists a consensus among specialists.

> Since Indo-Iranian is a rare case in which we have millennia of
> documentation of the intervening stages, and since the right answer to
> the question is still not obvious, then it must be very much harder to
> answer the question in respect of other cases for which little history
> is available.  At present, there appear to be few cases in which the
> origin of an ergative construction is fully understood and beyond
> controversy.
[...]

I'd say the cases I have seen of ergativity in Indic and Iranian languages
so clearly reflect underlying/earlier passive circumlocutions that
controversy is absurd. Yet, in that field controversy is to be expected
over anything.

I grant you (quoting Dixon) that Hittite is not this way: making neuters
animate when they are subjects of transitive verbs is indeed not the
reflex of a passive transformation. I have no knowledge of the Amazon
language Pari, let alone of its history. But note that I never claimed
that ergative always comes from passive, only that there are very clear
examples that it sometimes does - at least, that was what I meant.

Jens


From PIE at AN3039.spb.edu  Fri Jul  9 23:15:17 1999
From: PIE at AN3039.spb.edu (PIE at AN3039.spb.edu)
Date: Sat, 10 Jul 1999 02:15:17 +0300
Subject: accusative and ergative languages
In-Reply-To: <Pine.GSO.4.03.9906291609240.10157-100000@rsunx.crn.cogs.susx.ac>; from "Larry Trask" at Jun 29,
 99 6:49
Message-ID: <SAT.10.JUL.1999.021517.0300.>

Tue, 29 Jun 99 18:49 +0300 MSK Larry Trask wrote to Indo-European at xkl.com:

> But Dixon also cites examples of ergative constructions which, in his
> view, have very clearly *not* developed from passives, but from other
> constructions.  His examples are Hittite and Pari.

Pardon my ignorance, but i wonder what Hittite constructions can be
claimed as ergative? Could you clarify Dixon's opinion?

Does he mean the tendency to avoid using inanimates as 1st actants of
two-argument predicate and to transform them to animates in -nt-, like

    eshanants inan karapzi,
    witenants eshar parkunut,    etc?

_________
Alex


From jer at cphling.dk  Fri Jul  9 23:02:04 1999
From: jer at cphling.dk (Jens Elmegaard Rasmussen)
Date: Sat, 10 Jul 1999 01:02:04 +0200
Subject: Momentary-Durative
In-Reply-To: <006701bec26f$07f7eec0$b1fdabc3@xpoxkjlf>
Message-ID: <SAT.10.JUL.1999.010204.0200.>

On Tue, 29 Jun 1999, petegray wrote:

> Jens said:

>>    I wonder how else anybody would understand these data - except by
>> ignoring their being just that.

> Alas, Jens, I do not see data, but hypothesis and ideology.   The data is
> that some IE langs use nasal presents for a particular root where others do
> not, so your evidence from Latin cannot count for much in Sanskrit.    The
> data is that root aorists are associated with a variety of present
> formations, including nasal infixes.

I resent the use of the word "ideology"; is the common occurrence of a
nasal-infix structure from *tewd- 'thrust' in Skt. tundate and Lat. tundo
not part of our "data"? Do I have to add that the impression that the
Indo-European languages are related and descended from a common source is
not based on ideology, but on inspection and common-sense interpretation
of the data. And it takes outright rejection of the relationship between
the IE languages to make evidence from one IE language irrelevant in
another. There are so many instances of the _same_ verbal root turning up
with a nasal present in IE branches that have nothing else in common than
the mere fact that they are accepted as Indo-European. Is it a coincidence
that *k^lew- 'hear' forms a nasal present in Indo-Iranian and Celtic?
*telH2- 'endure', in Tocharian, Armenian, Italic and Celtic? *g^neH3-
'know', in Indo-Iranian and Germanic? This could go on - as Strunk's book
(and indeed already Brugmann's Grundriss) does. Does that not indicate
that the assigment of the nasal-infix structure as the present of certain
roots was fully lexicalized in the protolanguage?

> The hypothesis which has become an ideology, is that all root aorists must
> have had a nasal present.

Oh no, it works only in the direction that a nasal present has a root
aorist beside it in practically all cases. Also the reduplicated presents
and the y-presents generally form root aorists. I suppose there were some
old distinctions that only made sense with durative actions. However, by
PIE, each verb had mostly picked one form for its present. The massive
concord among the IE languages in this respect can be explained in no
other way that I can think of - while the even greater discord is easily
explained by continued normal language change.

For example, you said:

>> Strunk has shown that nasal
>> presents go with root aorists, thus we would like to derive tuda'ti from
>> an original root aorist if that is in any way possible.

> I see no reason to, since I do not share your ideology.   You may or may not
> be right - the important thing is that what you offer is not data, and so
> you should not insult those of us who do not agree with it.

Hey, hey, how can it be an insult to anybody that I try to account for the
greates amount of correspondences between the IE languages with the least
amount of force? I do not see you offering any understanding of the data
you apparently do not want me to talk about; I point out that they fit a
well-argued theory that is already on the market, and I add that the same
theory also accomodates the data you did choose to talk about; if that
is offensive, I'm afraid I may soon be doing it again.

> You said:

>> Likewise we would like to have a root aorist beside the nasal present
>> vinda'ti ... and so we
>> have a strong motivation to derive the thematic aorist a'vidat from a root
>> aorist.

> Traditionally, these are taken from different roots.

"Different roots"? But there is only one underived verbal paradigm, and
that is as trivial as nasal-infix present + root aorist if *(e-)wid-e/o-
is explained in exactly the same way as the Greek thematic reflexes of IE
root aorists (the e-lipon type). That there is a perfect beside it,
is normal; that it means 'know' rather than 'have caught sight of' is a
case-story of the most usual kind. The stem *wid-eH1- 'be looking' is a
derivative stem which can in principle be made from all verbs. There is no
need to project the Indic distinction between vid- 'know' and vid- 'find'
back into the protolangue, the functional range is delivered free of
charge by the large IE verbal system already.

Jens


From gordonselway at gn.apc.org  Fri Jul  9 23:57:40 1999
From: gordonselway at gn.apc.org (Gordon Selway)
Date: Sat, 10 Jul 1999 00:57:40 +0100
Subject: role of verb endings
Message-ID: <SAT.10.JUL.1999.005740.0100.>

At 16:25:19 -0500 on Tue, 29 Jun 1999, Patrick C. Ryan
<proto-language at email.msn.com>

>Larry indicated that verbal inflections should not be considered an
>expression of the subject in languages like Latin.

>I consider this position unjustifiable.

Why?

Gordon Selway
<gordonselway at gn.apc.org>


From gordonselway at gn.apc.org  Fri Jul  9 23:57:42 1999
From: gordonselway at gn.apc.org (Gordon Selway)
Date: Sat, 10 Jul 1999 00:57:42 +0100
Subject: indoeuropean/hand
Message-ID: <SAT.10.JUL.1999.005742.0100.>

At 10:15 am 29/6/1999, Rick Mc Callister wrote:

>It's an interesting metaphor

>I was also interested in why the "masculine" form manus and not *mana?

Various answers, or points of departure:

a.  it's a u- stem, not an o- stem
b.  there are several other grammatically f u- stems (domus, quercus &c)
c.  common enough with o- stems in Greek (hodos, nesos ktl)

Gordon
<gordonselway at gn.apc.org>


From proto-language at email.msn.com  Sat Jul 10 01:47:55 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Fri, 9 Jul 1999 20:47:55 -0500
Subject: accusative and ergative languages
Message-ID: <FRI.9.JUL.1999.204755.0500.>

Dear Peter and IEists:
 ----- Original Message -----
From: petegray <petegray at btinternet.com>
Sent: Tuesday, June 29, 1999 3:07 PM

<snip>

Peter:

> IE roots of the kind CRC had the ablauts CeRC, CoRC, CR.C.   The first
> appears in the Germanic present, the second in the past singular, the third
> in the past plural and the past participle (remember that PIE /o/ appears as
> /a/ in Germanic).  After standardisation of the vowel of the past, we find
> in modern German:
>     werfen    warf    geworfen

Pat:

Yes, but if a new verb were formed, e.g. webben (from English 'web'), there
is no possibility that it would be conjugated webben, wabb, gewobben.

What I consider 'internal inflection'  is Arabic yaktubu, kataba, katibun,
etc. which applies to any and all verbs, old, and those taken new into the
language.

 Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From proto-language at email.msn.com  Sat Jul 10 02:19:26 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Fri, 9 Jul 1999 21:19:26 -0500
Subject: accusative and ergative languages
Message-ID: <FRI.9.JUL.1999.211926.0500.>

Dear Wolfgang and IEists:

 ----- Original Message -----
From: Wolfgang Schulze <W.Schulze at lrz.uni-muenchen.de>
Sent: Wednesday, June 30, 1999 9:44 AM

W:

> Let me briefly comment upon some of the recent arguments:

> "Patrick C. Ryan" wrote:

>> A sequence like: "Me hurt. Tommy did it" is a virtual ergative so far as I
>> would judge. I have heard children speak in this way.

> This surely is not ERG-like! It's present tense, we ought to have "me hurts
> - Tommy does it" (or so) which shows that "me" still is in the ACC case
> (without AGR on the verb - "-s" has a "dummy agent" as a trigger. The
> same is true with structures like German "mich friert" etc...

Pat:

In spite of the grammar correction, what I have heard is "Me hurt". I assume
from this and other examples I have heard that 'me', regardless of its
function in standard English, is being used absolutively; and 'hurt', in
this phrase, I would think of more as a stative rather than transitive form.

Pat, previously:

>> (3)
>> However, in Language A,
>>         noun(B)+abs. verb
>> will be interpreted as an activity is performed by an unspecified agent--

Wolfgang:

> This is true for only some ERG systems. In such cases, the verb very
> often gets a plural morphology to install AGR with a "hidden" agent. In
> many other ERG systems, the omission of the "agent" leads to new
> intransitive structures with the inference ABS > AGENT (cf. I-ERG
> boy-ABS bring > boy-ABS come etc.).

Pat:

I have attempted to point out the salience of dividing these sentences into
two-element and three-element divisions rather than transitive and
intransitive categories. Obviously, you do not agree. To my analysis,
'boy-ABS. bring' is a transitive sentence with the object unspecified.
'boy-ABS come' is, unless we regard verbs of motion as reflexives (moves
himself to) as I would prefer, are, for most, intransitive --- perhaps
because many people regard a characteristic of transitivity as affecting in
some real way the properties of an object.

[ Moderator's comment:
  In an ergative system, a sentence "boy-ABS bring" would be transitive with an
  upspecified *subject*, as absolute marking on "boy" would require it to be
  the object.
  --rma ]

Wolfgang:

> Discussing the possible PASS background of ERG structures, Jens finally
> asks:

>> A truly innocent question for information: Are there other avenues that
>> are _known_ to have led to the creation of an ergative than the one
>> starting from an old passive?

> Sure, there are plenty such avenues! Most of them have to do with the
> Silverstein Hierarchy (or its expansion). The less a potential agent in
> (stronger) transitive scenarios is thought to bear inherent agentive
> features, the more it becomes likely that this "light agent" is marked
> by something that strengthens its agentive role. Some options are:
> (alienable) genitives (which can be extended to "heavy agents" on an
> inalienable basis), locatives (esp. with very light (or secondarily
> lightened) agents or weather phenomena etc...), instrumentals that are
> metaphorized to "agent" markers in the context of anthropomorphization),
> true "agentive" markers that are grammaticalized from e.g. a deicitic
> source, topic markers... Another possibility (though related to the
> strategies mentioned so far) is the reanalysis of 'active' structures in
> an ergative perspective (note that I do not want to suggest any 'active'
> typology for IE here, see my earlier postings): the S-split would then be
> harmonized on an S=0 level. All these strategies are based on the
> semantic or functional grading of the agentive (in terms of the
> manipulation of "lightness" and "heaviness"). Naturally, ERG techniques
> can also evolve from a special 'treatment' of the objective: One of the
> most prominent one is that of syntactic and/or pragmatic foregrounding
> which means that O is syntactically referred to as an intransitive S.
> Such a technique may be equivalent to passive strategies, however, this
> is only ONE of the many possible inferences.
> The syntactic/lexicical interface is touched upon when causatives of
> intransitives form the basis for newly established ERG features: Here,
> the morphosyntax of the causer can be introduced in the paradigm of
> other 'true' agentives via analogy.
> Finally, agreement strategies may play an important role in the game.
> If, for instance, agreement is coupled with some kind of person
> hierarchy, the presense of any SAP in a clause may condition agreement
> irresepctive their functional or semantic role. Hence, a scheme nSAP:A >
> SAP:O would necessariliy produce an erg-like AGR pattern (in case AGR
> becomes active), whereas SAP:A >nSAP:O would produce ACC-AGR. ERG AGR
> patterns may also result from the reinterpretion of clausal layers, e.g.
> the structure SV // AOV could be read as SV // A[OV] which means that O
> becomes some kind of closer attribute to the (participle-like) verb...

Pat:

All this theorization is undoubtedly interesting but, IMHO, does not answer
Jens' question. As for lessened degrees of animacy, most ergative languages
have antipassives to indicate this.

Although Dixon is certainly a man who has devoted much thought to
ergativity, I find something inherently problematical in combining ergative
and accusative features in one sentence (a little schzophrenic) which he is
forced to do by analyzing pronominal and nominal structures differently when
they occur in the same sentence.

I think it is likelier that, because of perceived greater animacy (or
definiteness), pronouns have a different method of marking that can still be
interpreted within an ergative context.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From JoatSimeon at aol.com  Sat Jul 10 03:12:19 1999
From: JoatSimeon at aol.com (JoatSimeon at aol.com)
Date: Fri, 9 Jul 1999 23:12:19 EDT
Subject: `cognate'
Message-ID: <FRI.9.JUL.1999.231219.EDT.>

>larryt at cogs.susx.ac.uk writes:

>This posting will not be news to the linguists on this list, who may prefer
>to stop reading at this point, unless they want to offer criticisms of the
>definitions below.  But I have noticed that a significant number of
>non-linguists on this list have apparently misunderstood the sense of our
>technical term `cognate': many of them
>appear to believe that `cognates' means something like `words of similar
>form

- that's odd; I always assumed it meant "derived from a common ancestral
word".  You know, like Tiwaz and dyaus.


From proto-language at email.msn.com  Sat Jul 10 14:24:11 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Sat, 10 Jul 1999 09:24:11 -0500
Subject: accusative and ergative languages
Message-ID: <SAT.10.JUL.1999.092411.0500.>

Dear Ralf-Stefan and IEists:

 ----- Original Message -----
From: Ralf-Stefan Georg <Georg at home.ivm.de>
Sent: Tuesday, June 29, 1999 7:37 AM

Pat wrote:

>> "Overt" means 'open and observable', and the overt subject of the Spanish
>> phrase is "he/she" indicated by -0 on [ha]; of the Latin phrase, 'he/she'
>> indicated by -(i)t. I would think you might have understood that I was
>> referring to languages which do not code the subject with affixes on the
>> verb.

R-S responded:

> No, it was not clear that you were referring to this kind of languages.
> Let's have a look at one of those, though:

> KhalkhaMongolian (no subject affixes on the verb):

> Nom chamd yavuulav.

>  (He, she, it, someone, nobody, I, we, you and whatnot) -"book (indefinite
> acc.)", "to-you", "sent". (Someone) sent you the book. "Someone" is not the
> translation, but the dummy for every agent you wish and which can be made
> clear by the context; it is overtly expressed by nuffin'. The
> extra-syntactic context, that is. A perfectly normal sentence in context.
> It is true that Khalkha does use the personal pronouns in examples like
> this to disambiguate, but it doesn't have to. The construction is certainly
> not ungrammatical.

Pat responds:

I do not readily have access to material on Khalkha Mongolian so I cannot
comment in detail on your example.

However, in general, I would hazard the guess that, regardless of its
apparently accusative construction, for all practical purposes, this
sentence could be just as easily translated: 'A book was sent to you'.  I am
also a little sceptical of an "accusative" in such a context which is marked
by -0.

But, having said that, I am willing to concede that for *some* accusative
languages, this construction, provided a nominative subject can be mentally
supplied from context, is grammatical.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From rmccalli at sunmuw1.MUW.Edu  Sat Jul 10 21:42:54 1999
From: rmccalli at sunmuw1.MUW.Edu (Rick Mc Callister)
Date: Sat, 10 Jul 1999 16:42:54 -0500
Subject: Chronology of the breakup of Common Romance [long]
In-Reply-To: <4835aed4.24b300a5@aol.com>
Message-ID: <SAT.10.JUL.1999.164254.0500.>

[snip]

>"Vlach" or "Wallach" seems to be descended from a name that was commonly
>applied to Celts.

	It has long been proposed that the name Volcae, a Gaulish tribe,
was the source of Anglo-Saxon Wealh, Balkans Vlakh, Polish WLochy, etc.
	The point I've generally seen is something to the extent that
supposedly the Volcae were the first Gaulish tribe to fall under Roman rule.
	Some popular writers have also proposed Volcae as the source of
English "folk" but I've never seen this proposal from professional
linguists.

>Appearing as "Walh" or "Walah" in OHG, it has interpreted
>as meaning "foreigner", sometimes Roman, but in usage it is closely
>associated with Celts and regions of Celtic habitation - e.g., "Wales",
>"Walloon".  I seem to recall an suggestion that there may be a tie here to L.
>vallum (fortified wall, the earliest meaning of "wall") and refer to the
>Celtic or Romano-Celtic oppidum or walled town.

	You'd have to account for the ending in Vlakh, WLochy, etc.

>The strong Celtic presence along the Danube is attested by Classical writers
>well before the common era.  E.g., Alexander fights them before turning
>against the Persians and meets Celts who are from Illyria.  Galicia in
>southern Poland is a region whose name remembers a Celtic presence even
>farther north.  I remember an old passage where Vlachs are identified as "the
>shepherds of the Romans" and in this role they may also have been imported
>help as they were in northern Italy.

	Isn't the term Vlakh anachronistic in this sense? Vlakh is first
documented in the middle ages, isn't it?

	[snip]

>The ancient ethnic designations in that part of the world are a little
>difficult to follow,  but it seems clear that the whole region from Illyria
>to present day Romania was under Rome by 250 ace.  And there is some
>possibility that Vlachs represented Romanized Celts across those regions.  In
>any case, the small difference in time between Rome's entry into Illyria and
>into Dacia would be a de minimis factor.

	My guess is that any Celtic element in the Balkan would have been
extremely thin.
	The Celts raided the Balkans and stormed through the region on the
way to Galatia but my understanding is that they lived north of the Balkans
in Galatia and Pannonia

>In a message dated 6/20/99 5:55:33 PM, rmccalli at sunmuw1.MUW.Edu wrote:

><<Rumanian does share a common pre-Romance substrate with Albanian
>which presumibly should help pinpoint the origin of Rumanian.>>

>The suggested Thraco-Illyrian-Albanian substrate has always been called
>tentative, especially in light of potential borrowing across short distances
>where there would have been much opportunity for contact, particularly prior
>to the Slavic invasions of the Balkans.

>This brings up that familiar problem of finding a Celtic substrate, one that
>might be fruitful.

>E.g., one example of the Albanian substrate often given (per G. Mallinson)
>has been Rom. "abure"/ Alb. "avull" (steam.)  But in Gaelic I see "co-bur"
>(foam), "to-bar" (well) and "bruich" (boil) - add the Gaelic "a-" (out of) to
>get /a-bruich (out of boiling- steam?) and the Celtic seems as possible as
>the Albanian as a source.

	Following Larry Trask's post that "boil" words tend to be
onomatopoeic --largely based on "b-l, f-r" [e.g. "boil, bubble", fervire",
etc.] I'd be real careful with this one

>Another substrate example has been Rom. "vatra"/
>Alb. "vatre" (hearth), but perhaps forms like the Gaelic "fadadh" (kindling)
>and "bradhadair" (blazing fire) offers evidence of a common origin and
>original meaning for these later similarities.

>I haven't found any recent consideration of the Celtic remnants in this area,
>but I would think it might offer some real possibilities.

>In a message dated 6/20/99 5:55:33 PM, rmccalli at sunmuw1.MUW.Edu wrote:

[snip]

><<I seem to remember reading that while Albanian has a strong substrate, that
>it is significantly lacking in Ancient or Classical Greek substrate. If
>this is true, then it tells us where not to look...>>

>But the problem is that Rumanian does have a substantial Greek substrate (and
>had a larger one centuries ago before Rumanian was "purified" in recent
>times.)

	As I understand, its substrate is from Byzantine and Modern Greek
but according to what I remember reading, it has little or no substrate
from Classical Greek.

[snip]

>Furthermore, given the extreme amount of intercourse between Greek and Thrace
>during those many early centuries, wouldn't one expect Thracian to have a
>strong and ancient Greek substrate, especially with regard to trade items and
>such?  If Albanian doesn't have such substrate, it may not be Thracian.

	Thracian was not necessarily a single entity. Some dialects may
have been influenced by Greek while others may have not. But I defer to
those who actually know something about Thracian.

	Here's what I have in my notes on the subject. My apologies for
spacing.

	Graham Mallinson, "Rumanian," 391-419. Martin Harris and Nigel
Vincent, eds. The Romance Languages. London: Croom Helm, 1988.

	There is very little substrate vocabulary in Rumanian, and its
origin is unknown. Since the grouping of Dacian, Thracian and Illyrian
within IE is uncertain, it is questionable whether Albanian and Rumanian
shared a substrate language, although they do have cognates:

[ Moderator's note:
  The following 8-bit entities are translated to TeX notation at the end of
  the message.
  --rma ]

	ALBANIAN			RUMANIAN
	avull "steam"			abure
	mës "colt"			mînz
	përrua "brook"		pîrîu
	shkrump "ash"			scrum
	vatrë "hearth"		vatra [413]

L. R. Palmer, The Latin Language London: Faber, 1954.

	The Illyrians left geonyms in Apulia and Calabria:
Brundisium < Brendou elaphon "Stag's horn," see Albanian bri,ni "horn,"
Salapia and Salapitani < Selepitani < sal "salt," + ap "water,"
Otranto < Odruntum contains Messapic odra "water." [39]
	Illyrian and/or Messapic words in Latin include
blatea < Illyrian balta "swamp,"
deda "nurse,"
paró "small ship,"
gandeia "gondola,"
hóreia "small fishing boat,"
mannus < Illyrian manda- "pony," see Messapic Menzana, to whom horses were
sacrificed. [40]
Lama "swamp," occurs as an element of Illyrian geonyms in Italy. [41]

A. Rosetti. Brève histoire de la langue roumaine des origines à nos jours.
The Hague: Mouton, 1971.

	The Thracians, known as Gètes by Greek writers and Dacians by Latin
writers, left geonyms ending in -dava, -upa, -sara, -para, -bria. [51]

	Illyrian words included
Bardylis (proper name) > Albanian i bardhë "white,"
dranis (Greek elaphos) > Gheg dreni, Tosk drëri "stag,"
Dalmatia, Delminium (probably Bosnian Duvno) > Albanian dele, delme "schaf,"
Dardani > Albanian dardhë "pear, pear tree, compare Illyrian dardá "pear,"
Messapian Menzana "Jupiter," > Albanian mës, mëz-i "mule," Daco-Romanian
mînz "poulain." [52]

	Thracian words include
bría "city," vs. Tokharian ri, riye "city,"
deva, dava "city, village," e.g. Pupuldeva > Bulgarian Plovdiv;
bur-, bour- > Albanian burrë "male"
Drénis (proper name) > Albanian drê, drêni "stag," zeiz, zis, ziz >
Albanian i zi, e zëzë "black,"
manteía, mantua, mantia "Brombeere, Brombeerebaum (blackberry) > Albanian
man, mand "Maulbeerbaum, Maulbeere,"
xarpáies (Carpathians) < carpa > Albanian karpë "rock," xarpoi (Tribe from
north of Thrace) > Albanian karpë "Fels,"
xorpíloi (Tribe from south of Thrace),
Maluensis, Maluese, Maluntum (geonyms) > Albanian mal "mountain, mountain
range, Daco-Romanian mal "escarpment, mountain, riverbank, seashore,"
pinon, pinos > Albanian pi "drink," Albanian pirë "drunk,"
Dacian sxiáre "chardon" > Albanian shqer "reisse auseinander, zerreiser,
kratze." [52]

	The Daco-Romanian suffix -esc > Thracian -isk. [53]
	Certain words are the same or cognate in Albanian and Daco-Romanian:

		Albanian			Romanian
	buzë "lip, edge"		buzâ "lip, edge"
	këlbazë, gëlbazé		câlbeazâ, gâlbeazâ
	(sheep illness)			(clavelée)
	këpushë "tick"			câpus,â "tick"
	avull "Dunst"			abur "steam"
	bres, brez "Gürtel"		brîu "waist, belt"
	ragal "hat"			argea "voûte d'une cave"
	baç "fromager"			baci "fromager"
	bollë "dragon, monster"	balaur "dragon, monster"
	baigë "Veternarian"		balegâ "fiente"
	baltê				baltâ "lagoon, pond, lake"
	i bardhê "white"		barzâ "stork"
	baskë "Fliess"			bascâ "toison, laine"
	pellg "Weiher, Regenpfütze"	bîlc "flaque, mare, bourbier"
	berr "Schaf, Weidevieh"		bîr (sheep call)
	bredh "Tanne"			brad "sapin"
	byk "pàglia"			buc "balle du blé"
	bukurê "pretty"			bucurie "joy"
	bungu "Eiche"			bunget "thick part of forest"
	kêsulê cap, stocking cap	câciulâ "bonnet de fourrure"
	katund "village"		câtun "hamlet"
	qafë "Kehle, Hals"		ceafâ "nuque"
	çok "Fussfessel, Schnabel"	cioc "beak, point, prow"
	sorrë "Krähe"			ciocâ "summit"
	çufkë "Quaste, Maishaar"	ciuf "tignasse, huppe"
	thump "Stachel"			ciump "bout, chicot, moignon"
	çupis "picke"			ciupi "picorer"
	shut "ohne hörner"		ciut "écorné"
	koqë "berry"			coacâzâ "berry"
	kopil "young man"		copil "child"
	kulpërë "clématite sauvage"	curpen "clématite des Alpes"
	kurthë "Falle, Fangeisen"	cursâ "trappe, piège, embûche"
	droe "fruit"			droiae "band, troop, many"
	dru, drutë "wood, stick"	druete "short thick piece of wood"
	thërrimë "crumb, debris"	fârîmâ "debris"
	gardh "Hecke, Zaun"		gard "enclosure"
	gat "bereit"			gata "near"
	gjëmle "thorn"			ghimpe "thorn, needle"
	gjon "Nacheteule"		ghionoaie "pivert"
	gjysh "grandfather"		ghuij "old"
	gogë "Gespenst, Wauwau"		gogâ "mitten"
	grep "Haken"			grapâ "herse"
	gërresë "Schabholz, Schabeisen" gresie "grès, pierre a aiguiser"
	gurmaz "Kehle"			grumaz "neck, throat"
	grundë "Kleie"			grunz "grumeau, pelote"
	gropë "Grube, Graben, Grab"	graopâ "tomb, ditch"
	gushë "Hals, cock's dewlap"	gus,â "throat, goiter"
	hamës "scarfer"			hames, "avid"
	gjymës "half"			jumâtate "half"
	i lehtë "leicht" 		lete "slowly"
	hudhër "Knoblauch"		leurdâ "wild garlic"
	magulë "hill"			mâgurâ "hill, height"
	mal "mountain (range)"		mal "escarpment, edge, shore"
	i math, i mall "great, big"	mare "great, big"
	maraj "fennel"			mârar "fennel"
	modhullë "peas"			mazâre "peas"
	moshë "Greis, Alter"		mos, "old (man), grandfather"
	mugull "Pfropfreis, Spross"	mugur "bourgeon"
	murg "dark, black, gray"	murg "brown"
	mushk "Maulesel"		mus,coi "mulet"
	nëpërke "viper"			nâpîrcâ "viper"
	ujü,"water" ujane "ocean"	noian "ocean, vastness"
	përroa "riverbed, brook"	pârâu "torrent"
	pupëzë "bird" 			pupë "Wiedehopf", pupazâ "huppe"
	rëndës "Lab"			rînzâ "gésier, cow's stomach"
	thabët "sour"			sarbât "sour, curdled"
	shkrep "schlage Funken, Feuer"	scâpâra "to spark, to kindle"
	shkrumb "alles Verkohlte"	scrum "ash, cinder, coal"
	thumbëz "Knopf"			sîmbure "nugget, nougat, seed"
	shpendër "hellebore"		spînz "hellebore"
	shpuzë "glowing coal"		spuzâ "hot coal"
	shtrep "worm"			strepede "worm"
	shtrungë Abteilung des Pferchs" "strungâ "defile"
	shapi "eidechse"		s,opîrlâ "lizard"
	cjap "Ziegenbock"		t,ap "buck, stag"
	thark "pen, corral"		t,arc "pen, menagerie"
	thep "point, sharp rock"	t,eapâ "point"
	udhos "cheese"			urdâ "cottage cheese"
	votër, vatrë "fireplace"	vatrâ "fireplace"
	vjedull "burrowing mammal"	viezure "blaireau"
	dhallë "sour milk"		zarâ "curdled milk"
	shkardhë "Hundekette"		zgardâ "dog collar" [58-62]

John A. C. Greppin Grolier Electronic Encyclopedia

Phrygian
	Phrygian, a language of west central Anatolia, had two literary
periods, Old Phrygian (730-430 BC) & New Phrygian (AD 100-350). The later
stage used a Greek-like script; the earlier had an eclectic alphabet based
on Northwest Semitic models. Though of Indo-European origin, Phrygian is
poorly understood, especially in the writing of the earlier period. It
seems to be more closely related to Greek, but it also shows certain
affinities with Armenian. An older theory, no longer tenable, related
Phrygian to Thracian, & posited a Thraco-Phrygian language family.

Thracian
	Thracian was spoken in what is now Bulgaria & parts of Greece &
Turkey. No significant inscriptions exist but numerous words are known from
Greek & Roman texts. A large number of personal & place names have also
been recorded. Thracian is Indo-European, but its affinities to any
language other than Dacian & perhaps Phrygian are vague.

Dacian
	Also referred to as Getic, Dacian was spoken in what is now
Romania. Like Thracian, it is known both from words mentioned in Greek &
Latin texts, & also from proper names. It has recently been shown that
Dacian became distinct from Thracian, but this took place probably after
1500 BC. Some scholars believe that a Dacian layer underlies Albanian, &
that perhaps Dacian, rather than Illyrian, was the original form of that
language.

Illyrian
	Illyrian was spoken north & west of Greece during the Greco-Roman
period. Scholars are unsure whether the term refers to just one language or
many. Most evidence for Illyrian comes from proper names; the core of the
material is now called Messapic. Illyrian has been considered an ancient
form of Albanian, but this view is losing favor.

Rick Mc Callister
W-1634
Mississippi University for Women
Columbus MS 39701

[ Moderator re-encoding (experimental) ]

	ALBANIAN			RUMANIAN
	avull "steam"			abure
	m{\"e}s "colt"			m{\^i}nz
	p{\"e}rrua "brook"		p{\^i}r{\^i}u
	shkrump "ash"			scrum
	vatr{\"e} "hearth"		vatra [413]

L. R. Palmer, The Latin Language London: Faber, 1954.

	The Illyrians left geonyms in Apulia and Calabria:
Brundisium < Brendou elaphon "Stag's horn," see Albanian bri,ni "horn,"
Salapia and Salapitani < Selepitani < sal "salt," + ap "water,"
Otranto < Odruntum contains Messapic odra "water." [39]
	Illyrian and/or Messapic words in Latin include
blatea < Illyrian balta "swamp,"
deda "nurse,"
par{\'o} "small ship,"
gandeia "gondola,"
h{\'o}reia "small fishing boat,"
mannus < Illyrian manda- "pony," see Messapic Menzana, to whom horses were
sacrificed. [40]
Lama "swamp," occurs as an element of Illyrian geonyms in Italy. [41]

A. Rosetti. Br{\`e}ve histoire de la langue roumaine des origines {\`a} nos
jours.  The Hague: Mouton, 1971.

	The Thracians, known as G{\`e}tes by Greek writers and Dacians by Latin
writers, left geonyms ending in -dava, -upa, -sara, -para, -bria. [51]

	Illyrian words included
Bardylis (proper name) > Albanian i bardh{\"e} "white,"
dranis (Greek elaphos) > Gheg dreni, Tosk dr{\"e}ri "stag,"
Dalmatia, Delminium (probably Bosnian Duvno) > Albanian dele, delme "schaf,"
Dardani > Albanian dardh{\"e} "pear, pear tree, compare Illyrian dard{\'a}
"pear," Messapian Menzana "Jupiter," > Albanian m{\"e}s, m{\"e}z-i "mule,"
Daco-Romanian m{\^i}nz "poulain." [52]

	Thracian words include
br{\'i}a "city," vs. Tokharian ri, riye "city,"
deva, dava "city, village," e.g. Pupuldeva > Bulgarian Plovdiv;
bur-, bour- > Albanian burr{\"e} "male"
Dr{\'e}nis (proper name) > Albanian dr{\^e}, dr{\^e}ni "stag," zeiz, zis, ziz >
Albanian i zi, e z{\"e}z{\"e} "black,"
mante{\'i}a, mantua, mantia "Brombeere, Brombeerebaum (blackberry) > Albanian
man, mand "Maulbeerbaum, Maulbeere,"
xarp{\'a}ies (Carpathians) < carpa > Albanian karp{\"e} "rock," xarpoi (Tribe from
north of Thrace) > Albanian karp{\"e} "Fels,"
xorp{\'i}loi (Tribe from south of Thrace),
Maluensis, Maluese, Maluntum (geonyms) > Albanian mal "mountain, mountain
range, Daco-Romanian mal "escarpment, mountain, riverbank, seashore,"
pinon, pinos > Albanian pi "drink," Albanian pir{\"e} "drunk,"
Dacian sxi{\'a}re "chardon" > Albanian shqer "reisse auseinander, zerreiser,
kratze." [52]

	The Daco-Romanian suffix -esc > Thracian -isk. [53]
	Certain words are the same or cognate in Albanian and Daco-Romanian:

		Albanian			Romanian
	buz{\"e} "lip, edge"		  buz{\^a} "lip, edge"
	k{\"e}lbaz{\"e}, g{\"e}lbaz{\'e}  c{\^a}lbeaz{\^a}, g{\^a}lbeaz{\^a}
		(sheep illness)			(clavel{\'e}e)
	k{\"e}push{\"e} "tick"		  c{\^a}pus,{\^a} "tick"
	avull "Dunst"			  abur "steam"
	bres, brez "G{\"u}rtel"		  br{\^i}u "waist, belt"
	ragal "hat"			  argea "vo{\^u}te d'une cave"
	ba{\c c} "fromager"		  baci "fromager"
	boll{\"e} "dragon, monster"	  balaur "dragon, monster"
	baig{\"e} "Veternarian"		  baleg{\^a} "fiente"
	balt{\^e}			  balt{\^a} "lagoon, pond, lake"
	i bardh{\^e} "white"		  barz{\^a} "stork"
	bask{\"e} "Fliess"		  basc{\^a} "toison, laine"
	pellg "Weiher, Regenpf{\"u}tze"	  b{\^i}lc "flaque, mare, bourbier"
	berr "Schaf, Weidevieh"		  b{\^i}r (sheep call)
	bredh "Tanne"			  brad "sapin"
	byk "p{\`a}glia"		  buc "balle du bl{\'e}"
	bukur{\^e} "pretty"		  bucurie "joy"
	bungu "Eiche"			  bunget "thick part of forest"
	k{\^e}sul{\^e} cap, stocking cap  c{\^a}ciul{\^a} "bonnet de fourrure"
	katund "village"		  c{\^a}tun "hamlet"
	qaf{\"e} "Kehle, Hals"		  ceaf{\^a} "nuque"
	{\c c}ok "Fussfessel, Schnabel"	  cioc "beak, point, prow"
	sorr{\"e} "Kr{\"a}he"		  cioc{\^a} "summit"
	{\c c}ufk{\"e} "Quaste, Maishaar" ciuf "tignasse, huppe"
	thump "Stachel"			  ciump "bout, chicot, moignon"
	{\c c}upis "picke"		  ciupi "picorer"
	shut "ohne h{\"o}rner"		  ciut "{\'e}corn{\'e}"
	koq{\"e} "berry"		  coac{\^a}z{\^a} "berry"
	kopil "young man"		  copil "child"
	kulp{\"e}r{\"e} "cl{\'e}matite sauvage"
					  curpen "cl{\'e}matite des Alpes"
	kurth{\"e} "Falle, Fangeisen"	  curs{\^a} "trappe, pi{\`e}ge,
						emb{\^u}che"
	droe "fruit"			  droiae "band, troop, many"
	dru, drut{\"e} "wood, stick"	  druete "short thick piece of wood"
	th{\"e}rrim{\"e} "crumb, debris"  f{\^a}r{\^i}m{\^a} "debris"
	gardh "Hecke, Zaun"		  gard "enclosure"
	gat "bereit"			  gata "near"
	gj{\"e}mle "thorn"		  ghimpe "thorn, needle"
	gjon "Nacheteule"		  ghionoaie "pivert"
	gjysh "grandfather"		  ghuij "old"
	gog{\"e} "Gespenst, Wauwau"	  gog{\^a} "mitten"
	grep "Haken"			  grap{\^a} "herse"
	g{\"e}rres{\"e} "Schabholz, Schabeisen"
					  gresie "gr{\`e}s, pierre a aiguiser"
	gurmaz "Kehle"			  grumaz "neck, throat"
	grund{\"e} "Kleie"		  grunz "grumeau, pelote"
	grop{\"e} "Grube, Graben, Grab"	  graop{\^a} "tomb, ditch"
	gush{\"e} "Hals, cock's dewlap"	  gus,{\^a} "throat, goiter"
	ham{\"e}s "scarfer"		  hames, "avid"
	gjym{\"e}s "half"		  jum{\^a}tate "half"
	i leht{\"e} "leicht" 		  lete "slowly"
	hudh{\"e}r "Knoblauch"		  leurd{\^a} "wild garlic"
	magul{\"e} "hill"		  m{\^a}gur{\^a} "hill, height"
	mal "mountain (range)"		  mal "escarpment, edge, shore"
	i math, i mall "great, big"	  mare "great, big"
	maraj "fennel"			  m{\^a}rar "fennel"
	modhull{\"e} "peas"		  maz{\^a}re "peas"
	mosh{\"e} "Greis, Alter"	  mos, "old (man), grandfather"
	mugull "Pfropfreis, Spross"	  mugur "bourgeon"
	murg "dark, black, gray"	  murg "brown"
	mushk "Maulesel"		  mus,coi "mulet"
	n{\"e}p{\"e}rke "viper"		  n{\^a}p{\^i}rc{\^a} "viper"
	uj{\"u},"water" ujane "ocean"	  noian "ocean, vastness"
	p{\"e}rroa "riverbed, brook"	  p{\^a}r{\^a}u "torrent"
	pup{\"e}z{\"e} "bird" 		  pup{\"e} "Wiedehopf", pupaz{\^a}
						"huppe"
	r{\"e}nd{\"e}s "Lab"		  r{\^i}nz{\^a} "g{\'e}sier, cow's
						stomach"
	thab{\"e}t "sour"		  sarb{\^a}t "sour, curdled"
	shkrep "schlage Funken, Feuer"	  sc{\^a}p{\^a}ra "to spark, to kindle"
	shkrumb "alles Verkohlte"	  scrum "ash, cinder, coal"
	thumb{\"e}z "Knopf"		  s{\^i}mbure "nugget, nougat, seed"
	shpend{\"e}r "hellebore"	  sp{\^i}nz "hellebore"
	shpuz{\"e} "glowing coal"	  spuz{\^a} "hot coal"
	shtrep "worm"			  strepede "worm"
	shtrung{\"e} "Abteilung des Pferchs"
					  strung{\^a} "defile"
	shapi "eidechse"		  s,op{\^i}rl{\^a} "lizard"
	cjap "Ziegenbock"		  t,ap "buck, stag"
	thark "pen, corral"		  t,arc "pen, menagerie"
	thep "point, sharp rock"	  t,eap{\^a} "point"
	udhos "cheese"			  urd{\^a} "cottage cheese"
	vot{\"e}r, vatr{\"e} "fireplace"  vatr{\^a} "fireplace"
	vjedull "burrowing mammal"	  viezure "blaireau"
	dhall{\"e} "sour milk"		  zar{\^a} "curdled milk"
	shkardh{\"e} "Hundekette"	  zgard{\^a} "dog collar" [58-62]

[ End of moderator re-encoding (experimental) ]


From edsel at glo.be  Sun Jul 11 10:31:58 1999
From: edsel at glo.be (Eduard Selleslagh)
Date: Sun, 11 Jul 1999 12:31:58 +0200
Subject: Latin perfects and Fluent Etruscan in 30 days!
Message-ID: <SUN.11.JUL.1999.123158.0200.>

-----Original Message-----
From: petegray <petegray at btinternet.com>
Date: Friday, July 09, 1999 11:03 PM

>Ed connected Latin amicus and ambo, and suggested a derivation of them both
>from Etruscan.

>Max and I both pointed that ambo has a good PIE pedigree, so this can't
>work.

>Ed then said:

>>And I don't see why any relationship with 'amicus', 'amare', etc. should be
>>excluded a priori: ....  Couldn't it be a double
>>transfer: early Lat./IE 'amb(i)-' > Etr. 'am(e)-' > later Lat. 'am-'?

>Anything's possible!  But it's easier to take the word we know to be IE as
>IE, and leave open the possibility of an Etruscan origin for amicus, if you
>want to explore that.   Don't be mislead by the initial "am-".   In ambo it
>derives from syllabic /m./, which cannot be the case in amicus.

[Ed Selleslagh]

There seems to be a slight misunderstanding here: with "later Lat. 'am-'" I
meant the beginning of words like 'amare', 'amicus', etc. , i.e. without
the -b-, which - according to the reasoning above - would have transited via
Etruscan, as opposed to those that came straight from PIE and preserved the
(a)m(.)b-.

Apart from all that, I never suggested that 'ambo', ambi-', Grk. 'amphi'
(and Cat. 'amb', which proved to be unrelated) etc. were of Etruscan origin,
only that they should be added to the data pool when looking at possible
relationships of Etr. 'am(e)-' and Lat. 'amicus' etc. It seems to me that
this led to an interesting discussion that yielded the possibility of the
existance of two parallel paths: one 'directly' from PIE to Latin and one
via Etruscan.

Ed


From X99Lynx at aol.com  Sun Jul 11 17:58:46 1999
From: X99Lynx at aol.com (X99Lynx at aol.com)
Date: Sun, 11 Jul 1999 13:58:46 EDT
Subject: indoeuropean (correction)
Message-ID: <SUN.11.JUL.1999.135846.EDT.>

In a message dated 7/9/99 6:06:11 PM, I wrote:

<<Finally there is again the Greek 'keiro' (to cut, carve) that would explain
churgeon/surgeon much more adequately than a direct connection to /cheir/.
In Aristophanes The Birds, 'keirulos' describes a barber.

[ Moderator's note:
  These words obviously cannot be related.
  --rma ]>>

Inexcusibly, I suggested here that 'cheirourgia', 'cheirourgio^', etc., might
be directly from 'keiro' (cut) instead of 'cheir'/cheiroo' (hand).   Though
the ultimate connection between those two basic forms (in some way or other)
doesn't seem a bad idea to me, the source of 'cheirougia' in Classic Greek is
clearly 'cheir'.

The earliest appearances of 'cheirourgio^' apparently refer not to medicine
or surgery, but to handicraft and the skills of the artisan.  Plato uses it
in reference to musicians.  The later extension to a skilled surgeon would
only be natural.

Sorry for not going to the books first and I hope I'm correcting this piece
of misinformation here.

Regards,
Steve Long


From petegray at btinternet.com  Sun Jul 11 18:45:59 1999
From: petegray at btinternet.com (petegray)
Date: Sun, 11 Jul 1999 19:45:59 +0100
Subject: accusative and ergative languages
Message-ID: <SUN.11.JUL.1999.194559.0100.>

> Pat writes:

> Well, on page 40 of Chao's Mandarin Primer, are listed "Affixes":
> 11 are listed; ...

I simply cannot let this go.    Verbal compounds are rife in modern Chinese,
and the second elements ("affixes" if we are using terms inappropriate for
Chinese) are derived from a large number of concepts.   The fact that Chao
lists 11 only means that he has selected 11, and does not reflect the actual
nature of the Chinese language.   Things are not as simple as that grammar
makes them seem!

Peter


From CONNOLLY at LATTE.MEMPHIS.EDU  Sun Jul 11 20:49:29 1999
From: CONNOLLY at LATTE.MEMPHIS.EDU (CONNOLLY at LATTE.MEMPHIS.EDU)
Date: Sun, 11 Jul 1999 15:49:29 -0500
Subject: accusative and ergative languages
Message-ID: <SUN.11.JUL.1999.154929.0500.>

After some illuminating discussion of Germanic ablaut, petegray wrote:

>Some Germanic ablaut is indeed the result of "vowel harmony" with a vowel in
>an ending which has since been lost - but this example is not one.

Germanic ablaut *never* results from vowel harmony.  You're thinking of umlaut,
in which a stressed vowel is assimilated to the articulation of a following
segment: in the most common instance, back vowels before /i i: j/ are fronted
and become [-low]: PGmc. gasti:z > OHG gesti > German G"aste 'guests', PGmc.
*fo:dijan > OE f"odan > later OE fe:dan > modern feed.  Umlaut is a much later
process (it's an on-going process in Old High German) than ablaut, which was
conditioned in part by stress but not by vowel harmony.

Leo

Leo A. Connolly                         Foreign Languages & Literatures
connolly at latte.memphis.edu              University of Memphis


From CONNOLLY at LATTE.MEMPHIS.EDU  Sun Jul 11 21:00:54 1999
From: CONNOLLY at LATTE.MEMPHIS.EDU (CONNOLLY at LATTE.MEMPHIS.EDU)
Date: Sun, 11 Jul 1999 16:00:54 -0500
Subject: Ergative vs. accusative
Message-ID: <SUN.11.JUL.1999.160054.0500.>

Pat said, while discussing "ergative" languages:

>> [3]
>>> However, in Language A, noun(B)+abs. verb
>>> will be interpreted as an activity is performed by an unspecified
>>> agent on B

>> [4]
>>> whereas in Language B: noun(B)+acc. verb is *ungrammatical*.

Leo replied -- too obscurely, it would seem:

>> what you said about combining this verb with the patient alone simply isn't
>> true.  Some ergative languages happily omit the agent, others do not.  So
>> what I've numbered [3] above will be legal in some ergative languages but
>> not in others.

>> Conversely, [4] is perfectly grammatical in many accusative languages.
>> Couldn't think of an example good enough to convince you.  But look at
>> this post.  Must've seen stuff like this before, right?

Pat responded:

>Well, I agree with your first comment. But *most* ergative languages treat
>the agentive as a missible adverbial adjunct of the verb. Perhaps there may
>be a question of these languages being "truly" ergative?

>I am unaware of any accusative language in which this contsruction is
>grammatical. As you know by now, Larry indicated that verbal inflections
>should not be considered an expression of the subject in languages like
>Latin. I consider this position unjustifiable.

Should've looked a bit closer, Pat.  In discussing [4], I intentionally used
several sentences which lack overt subjects, as I just did again.  And the
subject is not marked on any of the verbs.  Are they grammatical?  Rather not
decide myself; let the rest of our merry group decide.

Leo

Leo A. Connolly                         Foreign Languages & Literatures
connolly at latte.memphis.edu              University of Memphis


From CONNOLLY at LATTE.MEMPHIS.EDU  Mon Jul 12 05:01:00 1999
From: CONNOLLY at LATTE.MEMPHIS.EDU (CONNOLLY at LATTE.MEMPHIS.EDU)
Date: Mon, 12 Jul 1999 00:01:00 -0500
Subject: Interpreting ergative sentences
Message-ID: <MON.12.JUL.1999.000100.0500.>

Dear Pat and everyone else,

I've been reading the recent (well, actually not so recent) discussion
of ergativity with more than a little interest, since I've been working
on the problem off and on for years -- strictly from the synchronic
point of view.  Please bear with me -- this will be long, since I have
to do some theoretical stuff to work through, some of which will be old
hat to many among us.

Much of the discussion has centered around the interpretation of
ergative sentences in Sumerian, of all things, which only Pat seems to
know at all.  (I sure don't.)  But I think it would be worth while to
consider various phenomena in modern languages, some of them ergative,
others not.

First, what is a "subject"?  Pat and the others have been talking past
each other on this, to a considerable extent.  The reason is simple:
there's great disagreement on what it means.

Dixon, in his work on ergativity, has consistently equated "subject"
with "morphological subject".  That is, if a language has a rule that
it must (usually) have one noun or pronoun which gets some specified
treatment not accorded to other items in the sentence, that item is the
morphological subject.  If there is case, it will be called
"nominative" in "accusative" languages, but (usually, but IMHO
unnecessarily) "absolutive" in "ergative" languages. -- Not all
languages have such a rule.  Those that do not are called "active"
languages, where casemarking (or equivalent) is not uniform for
one-argument sentences.

Pat definitely agrees with Dixon here, and so (but for very different
reasons) do I.  Most modern syntacticians do not: for them, the subject
is the noun phrase (NP) which, in an accusative language, would (a) be
the morphological subject and (b) have the *syntactic* privileges of a
subject.  That is, it would occupy "subject position", if the language
has such a thing, or it might be the only item to which a reflexive
pronoun might refer, or the only one which might be replaced by a
relative pronoun -- this is all highly language-specific, of course.

The catch is: for some verbs in (as far as I can tell) all languages,
and for many or most transitive verbs in ergative languages, the thing
that "ought" to be the subject might not be.  Then some funny things
can happen, which again are language specific.  But first: how can we
tell which NP "ought" to be the subject?

"Case Grammar" of the sort first proposed by Charles Fillmore is
helpful here.  It assumes that various NPs have semantico-syntactic
relations to verbs, which he foolishly called "deep cases".  There is a
hierarchy of these cases, which I believe is as follows:

	1.  Agent
	2.  Experiencer; beneficiary (including owner)
	3.  Instrument; effector
	4.  Patient (aka "theme")
	5.  Oblique cases (including place, time, goal, and a few
	    others)

In an accusative language, the NP which ranks highest in this hierarchy
is normally made the morphological subject.  In addition, it has the
syntactic privileges associated with it.  For convenience, I call this
one "HRNP", which stands for "highest-ranking NP".

But not all verbs in these languages actually "select" (to use
Fillmore's term) the normal subject.  Repeatedly we see that verbs with
experiencers or beneficiaries instead of agents may make the patient
the morphological subject.  It then often turns out that while the
morphogical subject (patient) has some syntactic privileges, the HRNP
has others.  Some examples:

English _own_ and _belong_ have Beneficiary and Patient as arguments.
_Own_ selects the Beneficiary as subject, but _belong_ has the Patient.
The result: sentences with _own_ can be made passive, but those with
_belong_ cannot.

	The Rockefellers own that railroad. ==>
	That railroad is owned by the Rockefellers.

	That railroad belongs to the Rockefellers.
	*The Rockefellers are belonged to by that railroad.

Why no passive for _belong_?  I suspect the reason is that
passivization serves to demote the "correct" subject (HRNP) and use the
"wrong" one instead -- but _belong_ already has the wrong subject!

German _helfen_ 'help' and _schmecken_ 'taste (good)' both have dative
objects.  Normally, German nominative nouns must precede dative nouns
when the NPs are contiguous:

	Will der Professor dem Studenten helfen?
	     Agent         Beneficiary
	'Does the professor wish to help the student?'

	*Will dem Studenten der Professor helfen?

But watch what happens with _schmecken_ and many other verbs that
select Patient as subject over the higher-ranking Experiencer:

	?Hat der Kaffee dem Professor geschmeckt?
	       Patient  Experiencer
	'Did the coffee taste good to the professor?'

	Hat dem Professor der Kaffee geschmeckt?
	'Did the professor like the coffee?'

The "normal" subject+object version of this sentence, while surely
acceptable, is simply not as good as the "abnormal" object+subject
version.  But why?   The reason is simply that the "abnormal" version
puts the HRNP ahead of the lower-ranking Patient-subject: the NP that
"ought" to be the subject is put where the subject "ought" to be!  And
it's no accident that the English gloss which puts _professor_ ahead of
_coffee_ sounds better than the other -- for exactly the same reason.

Accusative languages very greatly in which syntactic provileges are
attached to the morphological subject and which to the HRNP.  It may
well be true that in accusative languages, if any NP is marked on the
verb, the morphological subject must be; but otherwise it's a very
mixed bag.  In Icelandic, the HRNP seems to have *all* the syntactic
"subject" properties, regardless of what case it's in.  In English, the
morphological subject is king (except re. passivization).

A similar situation obtains in ergative constructions.  In Dyirbal, the
HRNP is has no syntactic privileges whatsoever; everything depends on
the morphological subject.  But most ergative languages do exactly the
opposite: the HRNP has all the syntactic properties, and in some it is
even the only one which controls verb agreement.  The Patient is then
still the "morphological subject" in that it's in the absolutive case,
but that's all.

That being said, we must ask what is the most appropriate way to
translate a "normal" transitive sentence in a "typical" ergative
language?  Larry Trask has already pointed out that Basque (like many
other ergative languages) has a passive formation.  But since the
Patient is already the morphological subject, what does the passive
accomplish?  It simply demotes the HRNP, so that it no longer has any
syntactic "subject" properties and must (in many other ergative
languages: may) be omitted.

Many also have an "antipassive" formation.  This makes the HRNP the
morphological subject, demoting the Patient.  The HRNP now has all the
syntactic subject properties, as it typically does in typical sentences
in accusative languages.  Different languages have different reasons
for doing this; Dyirbal uses it mainly as a technique for stringing
sentences together that can share a single absolutive.

A "normal" ergative sentence -- no passive or antipassive, and agent NP
present in the ergative case -- corresponds most closely to the English
active voice; the agent should then be the English subject.  A passive
translation would be inappropriate, since passivization downplays the
agent and makes it syntactically irrelevant, making the translation
quite different in tone (and content) than the original.  But a passive
sentence in an ergative language might best be translated with an
English passive.

This is actually quite like what often happens with experiencer and
beneficiary verbs in accusative languages.  The fact that e.g. Spanish
makes the patient the morphological and syntactic subject of a verb
does not mean that an English translator must do the same.  _No me
gusta la mu'sica_ does not mean 'the music doesn't please me' -- that
translation suggests that someone was trying to please (or perhaps
annoy) me with the music.  Rather, it means, very precisely, 'I don't
like the music': the Spanish doesn't insinuate that anyone was thinking
of me, so neither should the translation.

Similarly, I had no hesitation above in glossing _Hat dem Professor der
Kaffee geschmeckt?_ as 'Did the professor like the coffee?', even
though (a) the two languages have different morphological subjects, and
(b) _like_ is not an exact translation of _schmecken_, which always
means 'taste'. I also changed the tenses: present perfect in German,
past in English, since each is the normal tense for past time in the
respective language.  And this is legitimate, since in tone and meaning
it most closely approximates the German original.  Another example:

	Der Wagen ist nicht mehr zu reparieren.
	the car   is  not   more to repair

A good translation would be 'The car can't be fixed any more' -- or
'The car is beyond repair' -- or 'They can't fix the car this time'
etc.  They're all pretty good, but they're not literal renderings.

I could have said instead: 'The car is no longer to be repaired.'  That
seems closer -- except that I will have used a *passive* infinitive
instead of the German active.  And let's face it: it sounds stilted,
while the original doesn't.

I can take these "liberties" because German is a modern language that I
know well.  There are very few who could do this for a dead language,
even one as well known and thoroughly studied as Latin or Greek.  And
that's the trap: because we don't know the nuances, we want to be as
literal as possible, in the hope that a literal interpretation might be
correct.  Sumerian is even worse, since we don't have all that much of
it (no poetry, no menus, nothing that seems to be finely nuanced
speech.  And on top of that, it's ergative, and most of us (including
me) are *not* used to dealing with such languages.

So how should we render Sumerian sentences of the type

	verb = 'distributed'
	NP absolutive = 'camels'
	NP dative = 'heirs'

(Sorry I didn't cite the actual sentence, but this time it really is
the thought that counts!)  Not knowing Sumerian nuances, we can only
draw on seemingly similar sentences in other ergative languages.
Despite what Pat says, it is far from clear that the English must have
a passive of the type 'the camels were distributed to the heirs by some
unknown agent'.  That is a plausible attempt at a rendering -- but
since we do not know the effect of omitting the agent, and there seems
to be no hint of a passive transformation being done, there's no reason
to think it's "better" or "more precise" than any of the following:

	The heirs shared the camels.
	The heirs divided the camels aong them.
	The camels were shared by the heirs.
	"They" divided the camels among the heirs.
	All the heirs got some camels.
	The camels (jewels, books etc.) went to the heirs.

We don't know, because we don't know Sumerian the way we know English
and other modern languages.

So please, caution!  And take due note of what has been discovered
about *living* ergative languages.  You can't hope to understand the
dead ones without them.

Leo

Leo A. Connolly                         Foreign Languages & Literatures
connolly at latte.memphis.edu              University of Memphis


From Georg at home.ivm.de  Sat Jul 10 10:03:34 1999
From: Georg at home.ivm.de (Ralf-Stefan Georg)
Date: Sat, 10 Jul 1999 12:03:34 +0200
Subject: accusative and ergative languages
In-Reply-To: <002b01bec282$df10a6a0$dfd3fed0@patrickcryan>
Message-ID: <SAT.10.JUL.1999.120334.0200.>

>> Having to be rendered by the passive in English is not the same thing as
>> "being passive in nature".

>Pat responds:

>How about explaining "passive in nature"?  Is that a Platonic idea?

As you might have observed, I put double quotes around this expression, the
most widespread functions of which are:

- the expression is not mine, but used by others, and I use it only not to
complicate the discussion, assuming everyone knows what I'm talking about

or

- the expression is handy, but admittedly imprecise, I only use it since I
don't want to put too much effort in terminological precision at this
point, having different things to communicate (admittedly a vice, but
normally tolerable when speaking to attentive and knowledgable people)

I was reacting to some claim of yours, namely that ergatives are - now what
was your expression ? I don't recall, was it "really passives", "always
former. i.e. reanalyzed p.s", "p.s in nature", "p.s by nature". Hell, I
dunno. Something to this effect, at least. The important thing is that I
managed to point out that this, i.e. the ergative-as-passive-claim, however
phrased, is wrong, and Larry Trask, who has also a well-known publishing
record on the issue, has spoken the definite word about this here.
Pat, this is, alas, a familiar pattern: when one's theory gets into dire
straits, some aside discussion on irrelevant points is opened, obviously
with the intention to show weaknesses of whatever kind in the opponent's
standing, hoping that this will cloud the correct and justified remarks he
might have made on the relevant points before. This may work in some
election campaign, but not here, sorry.

To answer your question: "passives in nature", as used by me, is neither a
Platonic idea, nor a Wittgensteinean prototype, nor a correct description
of anything I may happen to think, it is just shorthand for a bundle of
concepts I don't share. And, above all, it is bad English. Should we open a
new thread about this, or shouldn't we better return to ergativity and your
misconceptions about it, such as "there are ergative languages without any
splits" athl. ?

Stefan

Stefan Georg
Heerstrasse 7
D-53111 Bonn
FRG
+49-228-69-13-32


From Georg at home.ivm.de  Sat Jul 10 10:27:00 1999
From: Georg at home.ivm.de (Ralf-Stefan Georg)
Date: Sat, 10 Jul 1999 12:27:00 +0200
Subject: accusative and ergative languages
In-Reply-To: <003f01bec285$ac207600$dfd3fed0@patrickcryan>
Message-ID: <SAT.10.JUL.1999.122700.0200.>

>Pat responds:

>Well, on page 40 of Chao's Mandarin Primer, are listed "Affixes":

>11 are listed; of these 6 qualify as related to "inflections" :
>modal -m(en), phrase marker -le; completed action -le; progressive
>action -j(y/e); possibility or ability -de; subordination -de.

>Undoubtedly, a historical grammar might provide a few more but I consider
>this a pretty simple system.

This is getting weary. I think no sane linguist will be unaware of the fact
that there are languages with fewer purely morphological means than others.
Also, "simpler" and "not-so simple" phonological systems have been heard of.
But, if I'm not completely mistaken, there was talk about sthl. like
general, overall simplicity of languages, which is a pre-scientific notion,
quite simply, if this pun is allowed.

If we assume, as of course we have to, that human languages are
problem-solving devices which face a set of possible communicative
problems, which is the same for all linguistic communities (don't come with
obvious cultural differences, which pertain to vocabulary only; I'm aware
of the fact that not all linguistic communities of the planet need to talk
meaningfully about the various kinds of fish in the Yenissey, or the
Sepik). Languages make different choices regarding the set of functions
they grammaticalize (tense vs. aspect or both, inferentiality, relative or
absolute tense, number, participant-identification, reference-tracking,
pragmatic categories, you name it), the hows and whys of which are the
object of linguistic typology. One language uses grammaticalized
bound-morphology for, say, inferentiality, others have to use different
means to convey the idea. But they are all able to convey the idea, however
elegant, or however clumsy. The idea is that they *use means* to do it, or
that speakers may find these means, even if the category is weakly or not
at all grammaticalized in their language (your inflection paradigm may be
my intonation pattern, your inferentiality affix may be my expletive
adverb; to convey the idea of the Dakota sentence-particle /yelo/ or Thai
/khrap/ I may even be forced to say each time "I am male"; clumsy, but
possible, if I face the necessity). The description of these means is the
task of the grammarian. Some grammarians are aware of this task, and write
a 600-page grammar of Chinese, some aren't (or simply want to produce a
primer to help you get along abroad) and write a 60-pp. treatment of
Russian. So what ?

I will not deny that notions of simplicity vs. complexity may be useful
distinctions when talking about subsystems. For the characterization of
whole languages they are definitely not useful. The grammar-school like
rote learning of paradigms, which your Chinese teacher will be able to
spare you, will be compensated by having to digest subtle and difficult
rules of syntax.

Stefan

Stefan Georg
Heerstrasse 7
D-53111 Bonn
FRG
+49-228-69-13-32


From adahyl at cphling.dk  Mon Jul 12 12:27:17 1999
From: adahyl at cphling.dk (Adam Hyllested)
Date: Mon, 12 Jul 1999 14:27:17 +0200
Subject: `cognate'
In-Reply-To: <Pine.GSO.4.03.9906301059430.22695-100000@rsunx.crn.cogs.susx.ac.uk>
Message-ID: <MON.12.JUL.1999.142717.0200.>

On Wed, 30 Jun 1999, Larry Trask wrote:

> I am still willing to accept suggestions for revisions, though I may
> be unable to consider large revisions.
> (...)
> `father', all of these being descended from PIE *<pater>.

I'm sure it says PIE *<p at ter> in your manuscript (+ diacritics on the
<e>).

Adam Hyllested


From vidynath at math.ohio-state.edu  Mon Jul 12 12:43:49 1999
From: vidynath at math.ohio-state.edu (Vidhyanath Rao)
Date: Mon, 12 Jul 1999 08:43:49 -0400
Subject: Momentary-Durative
Message-ID: <MON.12.JUL.1999.084349.0400.>

Jens Elmegaard Rasmussen <jer at cphling.dk> wrote:

> I guess the widespread iterative value of sk-formations has started with
> those that were reduplicated.

Just how widespread are the reduplicated forms with -sk-? It is found Greek.
Hittite has -sk- forms both with and without reduplication in many roots.
Others?


From vidynath at math.ohio-state.edu  Mon Jul 12 13:43:34 1999
From: vidynath at math.ohio-state.edu (Vidhyanath Rao)
Date: Mon, 12 Jul 1999 09:43:34 -0400
Subject: Momentary-Durative
Message-ID: <MON.12.JUL.1999.094334.0400.>

Due to the long break from reading this list, I am not sure if (and how
much) I am starting to repeat myself. Please forgive me if the two new posts
do contains little new information/question. It is just that I didn't see
them addressed.

Jens Elmegaard Rasmussen <jer at cphling.dk> wrote:

> I do not think the chain of reformulations is longer than many other
> stories generally accepted (or even as long as some known to be true). For
> the type _tuda'ti_, the key example itself has apparently replaced a nasal
> present still seen in tundate and Lat. tundo.

The tuda'ti type, with primary endings, is somewhat rare in RV. Perhaps,
tuda'ti type was a replacement for injunctives, i.e., tuda'ti was a general
present and not progressive present. Just a thought.

> Strunk has shown that nasal presents go with root aorists,

Reduplicated forms seem to go with root aorists as well. Anybody have info
on statistical comparisons between the two groups?

This also bears on the other proposal, going back to Kurylowicz at least,
for the origin of (asigmatic) aorist forms: namely, these are old preterits
that became limited to aorist function due to the rise of new presents.

Incidentally, what are your thoughts on the origin of nasal presents? The
rarity of infixes in PIE has led to the proposal that nasal presents
originated from a double affix, that is *wined- was really *wi-n-ed- (or
*wi-ne-d). This would mean that originally win(e)d- and w(e)id (or wide/o)
were not grammatically associated. In case of dehmi lehmi etc, it then
becomes a question of when the roots came into existence. I am not sure if
we have enough information to decide this.

This brings me to a general question. There seem to be two camps about the
category system of the PIE verb. One believes that the aorist-imperfect
distinction, to be equated to perfective-imperfective distinction,
``always'' existed in PIE and Hittie lost this distinction, while Vedic
changed things around. The other considers the aspectual distinction to
postdate the separation of Anatolian.

There are certain nagging questions about the first thesis: The change in
Vedic is not explained and how it came about without the prior loss of
aspect has, AFAIK, not been explained. Those who adhere to this also feel
the need to explain away as much as possible of root presents. But
there are enough of them remain in Hittite and Vedic to raise doubts.

Sigmatic aorists, as forming perfectives for root presents, present some
problems too. There are roots that look like older roots extended with an s.
This suggests that sigmatic aorists came from grammatization of a root
extention and fits in with the general opinion that sigmatic aorist as a
>grammatized< formation (as opposed to -s as derivational affix for forming
new verbs) is late PIE.

There is another question that typology suggests: There are basically two
forms of perfectives (Dahl, ``Tense and Aspect systems; Bybee and Dahl, in
Studies in Language, 13(1989) pp.51--103; Bybee et al ``The evolution of
grammar''): The first seems to come from old preterits that become limited
to perfectives due to the rise of new present/imperfective. The second is,
as in Russian, from ``bounders'', originally adverbs that denote the
attainment of a limit. In the first, aorist is unmarked and refers to action
considered as a whole, even if it is atelic. In the second, the perfective
does not fit that well with atelic events. The difference is said to be
evident in Bulgarian which has both the aorist-imperfect distinction (of the
first type) and perfective-imperfective distinction using prefixes (the
bounders).

Now, in PIE, some verbs seem to be like the first system (root aorist vs
marked presents) while we cannot escape the presence of the second type as
well (root present with sigmatic aorist). Sometimes Greek situation is used
to argue that aorist looks older due to the number of irregularities. But in
Sanskrit, as can be seen from looking at the lists in grammars, root
presents show the most irregularity while aorists (and class 3, 7
presents) are regular exemplars of internal sandhi however complicated the
rules may seem. Again we are faced with the question of who is more archaic
and who regularized.

One thing needs to be said about a different attempt to explain this, namely
the traditional (in IE studies) equation durative=atelic=imperfective,
momentary=perfective. Limiting perfectives to telic/momentary situations is
decidely the minority option among languages. In Dahl's survey it was found
only in Slavic, Finnish (based on acc vs partative), Japanese, Hindi,
Mandarin, Bandjalang and Cebuano. It is even absent in aorist-imperfect
distinction of Bulgarian. Positing this restriction for PIE requires
appropriately strong evidence, not just pointing to Russian.

The equation durative=telic is equally atypical. In such a language, one
could not just say ``John walked home'' or ``John ate one piece of bread''.
Instead one must say something like ``John walked, reached home'', or have
two verbs for every transitive durative, one used only for past with
definite objects, one used for past with indefinite/mass objects and for all
presents. Again, we cannot just assume such a property, especially given
that this does not seem all that common in contemporary languages.

>  for the functional change is quite small: it only takes the use
> of the aorist form as an imperfect, then the rest follows by itself.

The change of syntactic categories does not strike me as a small change. I
would like to know more about incontrovertable cases of such changes before
arriving at any conclusions.


From vidynath at math.ohio-state.edu  Mon Jul 12 13:40:14 1999
From: vidynath at math.ohio-state.edu (Vidhyanath Rao)
Date: Mon, 12 Jul 1999 09:40:14 -0400
Subject: Momentary-Durative
Message-ID: <MON.12.JUL.1999.094014.0400.>

petegray <petegray at btinternet.com> wrote:

> [Earlier, in the Vedic hymns,] The aorist can even
> be used where a present would be expected.

I suspect that augmented aorist was preferred for performatives. Another
thing which seems to occur a few times is that augmented aorist being used
for future perfect (where, in English, simple present can be used). [A
similar case occurs in Pali, where both pluperfect and future perfect are
expressed the same way.] But, an aorist for action in progress?
[Injunctives, being tenseless, do not count.]

> In the later language the perfect is simply a preterit or past tense
> equivalent with the imperfect and
> fully interchangeable, and sometimes co-ordinated with it.   Different
> authors appear to prefer different tenses.

This pays no attention to difference in genres and influence of Prakrit
syntax, as I pointed our earlier and which others have: for example Speyer,
``Sanskrit Syntax''.

> Older grammars (e.g. Whitney), modern grammars by Indian linguists (e.g.
> Misra) and modern lingistics books (e.g. Hewson & Bubenik "Tense and
> aspect in IE") all say the same kind of thing.

Unfortunately people sometimes just repeat what they have been told.

Incidentally, I would like to know what the list members think of Hewson and
Bubenik. I find the chapter on Sanskrit confused. Tables repeatedly list
`akarot' and `agacchat' as imperfective, but once admitting that it is the
regular tense of narration and is not really imperfective.

>> I don't see any parts of Sans lit in which aorist has resultative
>> meaning.

> An example from the RV (sorry I don't have the exact reference):
>   putrasya na:ma grhanti praja:m eva anu sam atanat.

> "He gives the son's name; and thus _he has extended_ his race."

There is no such sentence in RV. (it doesn't scan!) Perhaps you are thinking
of Maitra:yan.i Sam.hita (somewhere in bk 1) ``putrasya na:ma gr.hn.a:ti;
praja:m eva:nu sam atani:t''.

Let me first define what a resultative is to me: It is a form that conveys
that the result of the action indicated by the verb obtains at the time of
reference, irrespective of whether the action is recent or not, or if it can
be repeated or not, etc.

The trouble is that the differences between ``hot news'' forms and
resultatives (of recent events) are purely subjective and are not ameable to
derivation by decoding (that is the so-called scientific approach that
denying the understanding of native speakers a la Whitney). It is not easy
to come up with an example of hot news situation for which resultative is
out of the question. Only eamples I can come up with are momentary
light/noise production: for example, to say that ``the bell has rung'', I
can only say ``man.i at.t.ittu vi.t.tatu''. ``.. irukkiratu'' would be
interpreted with an inferential sense (due to a rule that if resultative and
experiential senses are ruled out, then present of iru- will be given an
inferential
meaning) [resultative sense is absent here because when I start talking, the
bell is not audible.] It is not surprising that such examples are hard to
come by in extant Vedic texts.

Even worse, in modern Western European languages, resultatives and passives
cannot be readily distinguished, and perfect and resultatives are also
similar. Thus we take ``John has eaten dinner'' to be resultative, but
``John has eaten ostrich meat'' as experiential perfect. ``The door is
closed'' can be resultative or passive, while ``The door is closed by John''
is passive; the resultative has be expressed as ``The door is closed because
of John'' or some such. We should not attribute labels to the constructions
of another language based on how we translate into the more familiar
languages. [I mention passives here because the question of possible passive
origin of ergatives discussed in a different thread.]  If we use
translations, we must use a language in which the distinction is obligatory.
So I would ask, for example, how to translate the above into Tamil:  Is it
``makanin peyaray upayokikra:n; (atana:l) kutumpat tod.arai ni:t.t.i
vid.ukira:n'' or ``... ni:t.t.iy iruppavana:y a:kira:n''? [But this seems to
disappear in post-Vedic times. It certainly cannot be made morphologically
in Pali or medieval Sanskrit.]  In Vedic the ta-adjective is resultative. It
is not clear that it is interchangeable with the aorist. This is often
claimed, but the evidence comes from much later form of Sanskrit and is not
applicable to Vedic. So the best we can say is that the evidence is murky.
But there are examples where the resultative sense is questionable if not
absent: Where aorist is used with jyok as in RV 10.124.1 ``jyog eva
di:rgham. tama
a:s'ayis.t.ha:h.'', or na + aorist, which denies occurance at any point in
time (``never have ...'') rather than just denying that the result obtains.

> I happily grant that there might be distinctions;  but if there are, they
> are subtle, and not present in all cases.    This does not weaken my
> argument that Sanskrit and Greek do not agree in the meaning and the
> function of these tenses, even if they do agree on the formation.

I agree with the second sentence and the first part of the first sentence.
But to check if a difference is present, we must look at the examples the
difference is palpable and not at the examples where pragmatics makes it
hard to decide. After all, we design experiments to remove the influence of
confounding variables. And the lack of agreement between different languages
should not stop us from looking for possbile common starting point(s) from
which the different systems may have evolved. Studies of grammaticization
that cover a wide spectrum (both areally and genetically) that are becoming
more popular should help here.


From ALDERSON at mathom.xkl.com  Thu Jul 15 01:36:36 1999
From: ALDERSON at mathom.xkl.com (Rich Alderson)
Date: Wed, 14 Jul 1999 18:36:36 -0700
Subject: [Meta-discussion] Backlogs and antique systems
Message-ID: <WED.14.JUL.1999.183636.0700.>

Dear Readers:

We are now caught up from the backlog caused by my failure to take into account
a limitation of the Tops-20 filesystem, the absolute maximum number of files in
a single directory.

There are subscribers on about 170 different systems worldwide; to make list
processing simpler, for each outgoing message I create a separate queue file
for each system in the database; if there are enough messages, I create more
queue files than the queue directory can hold.  This occurred late last week.

Because of the logjam thus created, I could not respond to some of the early
queries about the list; I apologize to those who may have felt slighted by the
lack of an answer.

And so it goes...

								Rich Alderson
								list owner and
								moderator

P. S.  If there are any responses to this message, I will post a summary or a
hand-created digest of any that seem to be of general interest.  --rma


From ALDERSON at mathom.xkl.com  Thu Jul 15 02:47:14 1999
From: ALDERSON at mathom.xkl.com (Rich Alderson)
Date: Wed, 14 Jul 1999 19:47:14 -0700
Subject: [Meta-discussion] TeX-style diacritics
Message-ID: <WED.14.JUL.1999.194714.0700.>

Dear Readers:

Recently, a reader noted that he was having trouble reading posts containing
8-bit (non-ASCII) characters in Spanish and German, because his mail system
interpreted them as the first part of a 16-bit Chinese encoding.  I have, as an
experiment in usability, re-coded the portions of some posts which contained
8-bit characters to use TeX-style diacritics; I hope those have been useful to
those who do not use Windows systems to read this list.

  (Aside:  I had to play games with a Windows system to create a cheat sheet
  for myself to do this.  As I have noted in the past, the Indo-European and
  Nostratic lists are maintained and processed on an XKL Toad-1 system running
  Tops-20, which is aggressively 7-bit-only, and it will be a long while before
  I can take the time to learn a different system to move the lists.  I'd like
  to spend that time on the lists themselves, of course.)

I have long advocated the use of TeX-style diacritics in e-mail and in Usenet
newsgroup postings, because both of these modes of information transfer are
significantly unfriendly to 8-bit character encodings in many systems; further,
even 8-bit-friendly systems disagree in the placement of accented characters
within the character set (Windows differs from Macintosh differs from various
flavours of Unix differ from Windows ...).

TeX is a text formatting system which, with its extensive facilities, allows
the creation of document formatting systems of great richness.  (I personally
think that all linguists should learn and use it, for its wonderful capacity to
use multiple character sets organized into easily-accessed fonts, but this is
not the forum for that discussion; I mention it only as background.)

In standard TeX, accents are created by the following sequences of characters:

	\'	acute accent
	\`	grave accent
	\^	"hat"-style circumflex accent
	\~	tilde or "swing"-style circumflex accent
	\"	umlaut/trema/diaeresis
	\=	macron/overline
	\.	superior dot
	\b	underline/bar-under
	\c	cedilla
	\d	inferior dot
	\u	breve accent
	\v	h{\'a}\v{c}ek
	\H	Hungarian long umlaut
	\t	superior tie

I usually enclose accented letters in braces {} to organize them.  In TeX,
spaces after any command (like \' or \v) are ignored, so {\v c} is one way to
write <c-with-hac^ek> (to use another mode of writing certain accents); this
can be hard to read, even when one is a TeX user, so it is preferable to use
the alternate I used above, \v{c}.  Thus, <c-with-cedilla> could be written
\c{c}.

NB:  This is standard TeX; I would actually prefer to use \, for the cedilla
accent, although it has a different meaning in standard TeX.  Since anything
in TeX is ultimately user-definable, we can adopt that as an alternate to \c .

Finally, in TeX superscripts and subscripts respectively are indicated by
^{super} and _{sub}, where braces are not needed for a single character in this
context:  *k{^w}is vs. *ekwos is unambiguous as to meaning--the former has a
labiovelar initial, the latter a medial cluster.  The laryngeals can be written
as *H_1, *H_2, *H_3, or \'x, x, x{^w} (as was Cowgill's wont), or @_1, @_2, @_3
(using the "ASCII IPA" symbol <@> for shwa), or some other mode if you prefer.

It was suggested, by a TeX user, that we use an abbreviated set of accents; in
many cases this is fine, but for full generality I would like to encourage the
readers to use the TeX notation when necessary.

I will collect the responses to this message and summarize them, or create a
digest by hand, rather than filling the list with further non-Indo-European
topics.  Simple replies will keep the "[Meta-discussion]" subject header and
make it easy for me to do so.

								Rich Alderson

References for those who would like to know more about TeX:

Lamport, L.  _LaTeX:  A Document Preparation System_, 2nd edition, (1994:
Addison-Wesley, ISBN 0-201-52983-1) describes the most commonly used variant
of TeX.

Snow, W.  _TeX for the Beginner_ (1992: Addison-Wesley, ISBN 0-201-54799-6) is
a hands-on tutorial which concentrates on "Plain TeX", with references to major
differences from LaTeX.

Kopka, H.  _LaTeX: Eine Einfuehrung_ (1994: Addison-Wesley) is a 3-volume work
on the basic LaTeX system and the large number of extension packages which have
been created for it.  I don't own a copy and can't get the ISBN.

Goossens, M., Mittelbach, F., and Samarin, A.  _The LaTeX Companion_ (1994:
Addison-Wesley, ISBN 0-201-54199-8) describes a large number of extensions to
the LaTeX system.

Knuth, D.  _The TeXbook_ (1986:  Addison-Wesley, ISBN 0-201-13447-0) is the
description of "plain TeX" by the gentleman who created it.  Difficult and
dense, but worth looking through for the erudite quotations at chapter ends,
and the subtle jokes throughout.
-------


From JoatSimeon at aol.com  Tue Jul 13 04:09:46 1999
From: JoatSimeon at aol.com (JoatSimeon at aol.com)
Date: Tue, 13 Jul 1999 00:09:46 EDT
Subject: PIE vs. Proto-World (Proto-Language)
Message-ID: <TUE.13.JUL.1999.000946.EDT.>

In a message dated 7/12/99 9:28:59 PM Mountain Daylight Time,
proto-language at email.msn.com writes:

>I simply do not understand why some find it difficult to understand that
>reconstructing the Proto-Language is only primarily different from
>reconstructing Indo-European in the wider selection of source languages for
>data.

-- Temporal distance.  Loss of information.  Entropy.

A distinction between 5,000 years of unrecoverable loss and 50,000 years of
unrecoverable loss.

Words _vanish_.  A certain percentage of vocabulary just ceases to exist in
every century.


From jer at cphling.dk  Tue Jul 13 11:39:25 1999
From: jer at cphling.dk (Jens Elmegaard Rasmussen)
Date: Tue, 13 Jul 1999 13:39:25 +0200
Subject: Recoverability
In-Reply-To: <d04dd9cf.24b435bb@aol.com>
Message-ID: <TUE.13.JUL.1999.133925.0200.>

On Wed, 7 Jul 1999 JoatSimeon at aol.com wrote:

> [...]
> If we
> didn't have written or other artificially preserved examples of now-extinct
> languages, much of the PIE vocabulary we have would be completely
> unrecoverable.

> Try reconstructing PIE using _nothing_ but contemporary Albanian and English.
> You'd be hard-put to prove that such a language even existed.

But that is not what you do in cases where only modern languages are
available: You do not pick two and ignore the rest, you take all there is.
In the case of IE, you would take _all_ modern Germanic languages, if need
be with their dialects; you take all Slavic languages, you don't forget
about Lithuanian and Latvian or Modern Greek which gives quite a lot; you
add four Celtic languages (Welsh, Breton, Irish and Scottish Gaelic with
due attention to their dialects); you get most of Latin by taking all of
the Romance languages and the many learned words of Latin origin preserved
in many languages (you get much of Ancient Greek the same way); you take
Albanian in at least two varieties; Armenian likewise; and you add the
total effort of comparative linguistics applied to the countless Modern
Indic and Modern Iranian (and Nuristani) languages. This is a LOT - and
NOBODY could even thgen be in doubt that this is indeed a family of
related languages sprung from a common source. True, we would still be in
doubt or indeed ignorant about many a finer point which is only added by
languages of old texts - and of course even they do not contain
everything, so doubt will remain - but the general utline and subgrouping
of the family would stand firm even on the sole basis of modern languages.

Jens


From proto-language at email.msn.com  Tue Jul 13 05:21:22 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Tue, 13 Jul 1999 00:21:22 -0500
Subject: accusative and ergative languages
Message-ID: <TUE.13.JUL.1999.002122.0500.>

[ moderator re-formatted ]

Dear Joat and IEists:

 ----- Original Message -----
From: <JoatSimeon at aol.com>
Sent: Tuesday, July 06, 1999 11:46 PM

>> proto-language at email.msn.com writes:

>> So what? It only requires us to expand our universe of applicable data.
>> The idea that vocabulary is irretrievably lost is jejeune.

> -- actually, it's obvious.  Information vanishes as entropy increases.  If we
> didn't have written or other artificially preserved examples of now-extinct
> languages, much of the PIE vocabulary we have would be completely
> unrecoverable.

> Try reconstructing PIE using _nothing_ but contemporary Albanian and
> English.  You'd be hard-put to prove that such a language even existed.

It is thuis type of illogic which stands in the way of real progress in
historical linguistics.

Larry Trask and others have done an excellent job, IMHO, of reconstructing
the language from which present Basque derives --- from only one language.

I also am certain that if we had reason to know that Albanian and English
were descendants of a common language, we would eventually discover a kind
of IE that would be inaccurate in some ways but essentially write.

Naturally, the more languages we added to the equation, the better
reconstruction of IE we would achieve.

Same principle.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From Georg at home.ivm.de  Tue Jul 13 08:55:38 1999
From: Georg at home.ivm.de (Ralf-Stefan Georg)
Date: Tue, 13 Jul 1999 10:55:38 +0200
Subject: connected text in PIE; Proto-World language
In-Reply-To: <3785A12D.9CF0DD1@umist.ac.uk>
Message-ID: <TUE.13.JUL.1999.105538.0200.>

>  (1) When neural-net computers get advanced enough for people to
>simulate better than now human brain processes including language, it
>would be interesting to simulate language evolution and see over how
>many centuries and by what steps a language changes so much that all
>trace of common ancestry vanishes behind the `noise' of accidental
>resemblances.

Sure we will learn a great deal from those neural-net computers. But, in
order to gain a better understanding of language change from them, we
should be able to bring them to interact socially with all that goes with
it.

Stefan Georg
Heerstrasse 7
D-53111 Bonn
FRG
+49-228-69-13-32


From larryt at cogs.susx.ac.uk  Tue Jul 13 10:24:45 1999
From: larryt at cogs.susx.ac.uk (Larry Trask)
Date: Tue, 13 Jul 1999 11:24:45 +0100
Subject: accusative and ergative languages
In-Reply-To: <Pine.SO4.4.05.9907091402440.15900-100000@rask>
Message-ID: <TUE.13.JUL.1999.112445.0100.>

On Fri, 9 Jul 1999, Jens Elmegaard Rasmussen wrote:

> Excuse my interruption, but I think your discussion is missing the
> only point of any interest in this context: Can "Mutila jo zuen" not
> *come from* something which *originally* meant, not 'the boy was
> hit' pure and simple, but specifically 'the boy was hit by him
> (e.g., by the one we're talking about)'? Nobody is claiming that the
> passive constructions out of which the ergative grew in a number of
> languages were restricted to impersonal use. Incidentally, if they
> were, there would not have been anything to put in the case that
> subsequently got interpreted as a "transitive-subject, i.e. ergative
> case".

Well, this suggestion is possible, I suppose, but I know of no evidence
to support it, while there is some evidence against it.  Let's look at
the structure of the Basque sentence.

Here <mutil> is `boy', and <-a> is the article.  This NP, being a direct
object, stands in the absolutive case, which has case-suffix zero.

The item <jo> is the perfective participle of the verb `hit'.

And <zuen> is the auxiliary verb-form.  Now, third person is usually
marked by zero in Basque verbs.  So, the usual absolutive agreement
slot, which is the first slot in the auxiliary, cannot be filled by an
agreement marker, and it is filled instead by <z->, a redundant marker
of past tense.  The next element, <-u->, is a reduced form of the verbal
root <-du->, from *<edun> `have'; this is the usual transitive
auxiliary.  Again, the ergative agreement slot, which follows the verbal
root, is empty, because the ergative (subject) NP is third-person.
Finally, <-en> marks past tense.

The whole thing is thus this:

	boy-the-Abs hit-Perf Past-have-0-Past

Or, roughly, `(She/he/it) had hit the boy.'

But it translates English `S/he hit the boy' (before today).  This is a
periphrastic form comparable to the ones we find in Romance and
Germanic.  Nobody knows if such periphrastic forms are calqued on
Romance or are an independent development in Basque.

Now, I can see little scope here either for an original passive
interpretation of the form or for any way of reading `by him' into the
auxiliary form.

> I don't know a first thing about Basque, though I have been
> intrigued by it on many occasions, especially since it offers such a
> good parallel to Old Irish in the verb where you apparently have to
> memorize practically all forms (which are many) to be able to say
> even the simplest of things - and that of course was also what made
> me stop every time I got started. >From the primitive and casual
> books at my disposal I do see that "zuen" and "zuten" mean 'he had
> him' and 'they had him' resp. I also believe I see that such
> auxiliaries are combined with a particularly short form of the
> participles, referred to by Schuchardt as the root of the
> participle;

It is simply the perfective participle of the verb.  Most participles in
Basque carry an overt suffix proclaiming them as such, but <jo> happens
to be one of the exceptions: it has no participial suffix, though I
suspect that it once did, and that the suffix has been lost by a
combination of phonological change and analogical readjustment.

> and "jo" is 'stick; beat' in its shortest form, says my little
> dictionary;

Actually, `hit', `strike', `beat' -- not `stick', which is a noun.

> and "mutil-a" is 'boy' with the article "-a", but without case or
> number marking.

Not quite.  The form <mutila> is marked as absolutive by its zero
suffix, and as singular by its singular article <-a>.  The plural
article is <-ak>, and `the boys' is <mutilak>, in the absolutive.

> Therefore my persistent question: Why can't "mutila jo zuen" and
> "mutila jo zuten" reflect a construction that was earlier meant to
> express 'the boy, he had him hit', 'the boy, they had him hit'?

Well, I can't rule that out, but I can't see any evidence to support it.
Note in particular that <mutila> exhibits *no* subject properties in
modern Basque, or in Basque of the historical period.  If it ever was a
subject, as this proposal requires, the reanalysis must have been
carried to completion a long time ago.

Note also that intransitive verbs are likewise conjugated
periphrastically but with the intransitive auxiliary <izan> `be'.

> Schuchardt also gives "zen" to mean 'he was', so that if you gloss
> "mutila jo zen" as 'the boy was hit', it seems there is quite a bit
> of agreement that the verbal root is a participle by itself.

The lexical verb stands in the form of its perfective participle in all
periphrastic past-tense forms, and also in all periphrastic perfects.
The perfect form corresponding to <Mutila jo zuen> is <Mutila jo du>,
which differs only in that the auxiliary is now present-tense.  This
form translates both English `He has hit him' and English `He hit him'
(earlier today) -- much as in European Spanish.

> I do not see in what way this makes the *diachronic* interpretation
> of "mutila jo zuen" any different from the Hindi preterites that are
> based on Sanskrit constructions of the type "A-Nominative +
> B-Genitive + PPP/nom." meaning earlier "A was (verb)-ed by B", but
> now simply "B (verb)-ed A." Where am I wrong?

Well, in the Indic case, we have several thousand years of texts to
consult, so that we can get an idea how the ergative construction arose.
With Basque, we are not so lucky.

Note also that, in Hindi, as in Indic generally (I think), the ergative
occurs only in the past tense, as is common with ergatives that have
arisen from perfective or passive constructions.  In Basque, however,
the ergative construction is used in all circumstances, without
exception.

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk


From stevegus at aye.net  Tue Jul 13 14:28:12 1999
From: stevegus at aye.net (Steven A. Gustafson)
Date: Tue, 13 Jul 1999 10:28:12 -0400
Subject: indoeuropean/hand
Message-ID: <TUE.13.JUL.1999.102812.0400.>

petegray wrote:

> The form is only apparent.   It is actually a -u stem, whereas the
> masculines that you are thinking of are thematic (-o/e).   The -u stems can
> be any gender, although masculines do predominate.

There are a fair number of Latin -u stems that are feminine, especially
considering how few -u stems there are to begin with: acus, domus,
cornus, idus, and tribus (as well as manus) are the ones that come to
mind.

--
Steven A. Gustafson, attorney at law
Fox & Cotner:  PHONE (812) 945 9600   FAX (812) 945 9615
http://www.foxcotner.com

Ecce domina quae fidet omnia micantia aurea esse, et scalam in
caelos emit.  Adveniente novit ipsa, etiamsi clausae sint portae
cauponum, propositum assequitur verbo.


From Georg at home.ivm.de  Tue Jul 13 12:20:10 1999
From: Georg at home.ivm.de (Ralf-Stefan Georg)
Date: Tue, 13 Jul 1999 14:20:10 +0200
Subject: accusative and ergative languages
In-Reply-To: <001801beca0f$f9125140$399ffad0@patrickcryan>
Message-ID: <TUE.13.JUL.1999.142010.0200.>

>[PR]

>Rather overly broad! I have heard conversations that go along the lines of
>'He hit the boy' with the adressee responding, 'Who hit him, John or Phil?'
>Unless you consider clarification a gibberish-process.

The explanation for this is, obviously, that speaker A had reasons to
believe (or was unattentive to reasons against this) that the reference was
clear enough for B from any kind of context (previous discourse, situation,
accompanying gesture, or whatnot). The gist is that, unless you want to be
deliberately opaque, you use anaphoric pronouns to refer to a referent
which is somehow identifyable by your addressee. You may, however, be wrong
about this, and then you may expect a clarifying question. The very fact
that this clarifying question makes up for a meaningful and
often-encountered micro-dialogue in your example bespeaks the
less-than-fully-acceptable nature of a communication strategy which opens a
conversation with anaphora, without making sure that everyone knows who you
refer to by "he".

And, by the way, this makes up for a crucial difference between person
marking on the verb and overt pronouns. Though both have comparable
functions (reference identification), the former is hardly ever explicitly
anaphoric. It would be interesting to watch out for a language with verbal
affixal person marking, which would only be used for anaphora, and
suppressed in other cases. I'm not aware of such a language, but I may be
missing the very obvious at the moment. Any ideas, someone ?

St.G.

Stefan Georg
Heerstrasse 7
D-53111 Bonn
FRG
+49-228-69-13-32


From ECOLING at aol.com  Tue Jul 13 15:51:00 1999
From: ECOLING at aol.com (ECOLING at aol.com)
Date: Tue, 13 Jul 1999 11:51:00 EDT
Subject: PIE vs. Proto-World (Proto-Language)
Message-ID: <TUE.13.JUL.1999.115100.EDT.>

Patrick Ryan writes:

>I simply do not understand why some find it difficult to understand that
>reconstructing the Proto-Language is only primarily different from
>reconstructing Indo-European in the wider selection of source languages
>for
>data. Of course, secondarily I have labored to reconstruct the underlying
>monosyllables by analysis of attested compounds.

>Pat

I strongly support work on earlier language families not part of the
standard doctrine today.
Against those who say (very nearly) that everything which
can be discovered has already been discovered.
That is of course a charicature, but with some truth to it.
(Or they say this while specifying "using the comparative method",
and defining it circularly to mean only the existing tools,
and only those ways of using those tools,
which are well known today.)

However, the above statement by Patrick Ryan I find highly surprising.

Of course it is going to get RELATIVELY more difficult
to reconstruct to greater time depths.
The mistake of those who reject all Nostratic and similar work
is in drawing a sharp fixed line,
saying that short of that time depth it can be done,
beyond that time depth it cannot be done.

The mistake of Patrick Ryan in the quotation above is to neglect
that it does get substantially more difficult as the time depths increase,
or in particular language families, because of the specific nature
of the sound-changes and grammatical changes which occurred,
or etc.

There is no sharp break.
There is a gradually increasing difficulty with time depth and
    with depth of changes (the two correlated but not the same concept).
There is ample room for expanding the tools used,
and for empirical studies of what sorts of changes each tool is
capable of penetrating beyond, and to what degree.

To give merely one example, one factor, sound symbolsim:

If there are sound-symbolic ideal forms for lexemes having
certain meanings,
then the more arbitrary ones (farther from those sound-symbolic
ideal forms)
have greater value as evidence for historical connections
of specific languages or language families,
simply because they are less likely to have resulted from
later pressure towards the sound-symbolic ideal.

This is of course a terribly difficult circularity,
because it means that look-alikes which are more widely attested
for a given meaning or set of related meanings
may be EITHER relics of an earlier historical unity
(whose changes were perhaps ALSO retarded by sound-symbolic forces),
OR the results of pressure towards some sound-symbolic ideal forms,
from diverse original and unrelated forms.
Much more subtle and difficult reasoning is therefore needed to
establish what are results of sound-symbolism and what are results
of historical common origins.
EVEN when we have a suprisingly widespread statistical sound-meaning
correlation.

The usual procedure is also circular.
Simply taking a sample of purportedly unrelated languages
and attempting to determine how many look-alikes word lists contain
is a bit naive,
because the languages may not be totally unrelated,
because chance resemblances may be more common in certain
meaning or sound ranges,
because the biases of different types of sound-system structures
are not yet well handled,
and for many other reasons.

Lloyd Anderson


From Georg at home.ivm.de  Tue Jul 13 12:24:17 1999
From: Georg at home.ivm.de (Ralf-Stefan Georg)
Date: Tue, 13 Jul 1999 14:24:17 +0200
Subject: accusative and ergative languages
In-Reply-To: <Pine.SO4.4.05.9907092339440.18724-100000@rask>
Message-ID: <TUE.13.JUL.1999.142417.0200.>

>I grant you (quoting Dixon) that Hittite is not this way: making neuters
>animate when they are subjects of transitive verbs is indeed not the
>reflex of a passive transformation. I have no knowledge of the Amazon
>language Pari, let alone of its history. But note that I never claimed
>that ergative always comes from passive, only that there are very clear
>examples that it sometimes does - at least, that was what I meant.

Sorry for playing the smartass, but P"ari is Nilotic, one of the celebrated
cases where ergativity has been found on the African continent. It is
particularly rare there.

St.G.

Stefan Georg
Heerstrasse 7
D-53111 Bonn
FRG
+49-228-69-13-32


From rmccalli at sunmuw1.MUW.Edu  Tue Jul 13 17:14:43 1999
From: rmccalli at sunmuw1.MUW.Edu (Rick Mc Callister)
Date: Tue, 13 Jul 1999 12:14:43 -0500
Subject: Latin perfects and Fluent Etruscan in 30 days
In-Reply-To: <3.0.6.32.19990709101606.00798950@box5.tin.it>
Message-ID: <TUE.13.JUL.1999.121443.0500.>

	hijo for "shoot from a dead plant, plant baby" is very common and
is a normal usage
	I've heard Spaniards as well as Latin Americans use it when talking
about "spider plants"
	Be care ful with dictionaries because so many of them use only
technical terms that agricultural engineers use while skipping many of the
actual terms farmers use

[snip]

>If you really happened to hear the word "hijo", it might be due -- IMHO --
>to an improper usage and/or a rare "modismo hispanoamericano".

>Cheers

>       Paolo Agostini

Rick Mc Callister
W-1634
Mississippi University for Women
Columbus MS 39701


From proto-language at email.msn.com  Tue Jul 13 13:16:18 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Tue, 13 Jul 1999 08:16:18 -0500
Subject: accusative and ergative languages
Message-ID: <TUE.13.JUL.1999.081618.0500.>

[ moderator re-formatted ]

Dear Peter and IEists:

 ----- Original Message -----
From: petegray <petegray at btinternet.com>
Sent: Wednesday, July 07, 1999 2:41 PM

> Pat wrote:

>> The idea that vocabulary is irretrievably lost is jejeune.

Pat writes:

Undoubtedly, a comment loses some pungency when the key concept is
misspelled.

Peter wrote:

> The idea that it is not lost is equally jejune (if I may use the English
> spelling - I mean no offence).

Pat writes:

How could I possibly be offended seeing that the language of the list is
English? I acknowledge the mistake.

Peter wrote further:

> One of the problems in PIE is identifying which isolates are PIE and which
> are not, or, where there are similar words in two languages, which are
> inherited cognates, and which are loans (yes, I know there are techniques for
> this - but there are still big problems).

> Proto-proto- languages simply cannot be recovered with the same degree of
> certainty as languages nearer to our attested texts.

Pat writes:

Cannot disagree with a word above. But, based on the study I have made of
supposedly unrelated languages, I am simply amazed that so many CVC(V) roots
seem to be preserved more or less semantically intact in so many of them.

When the languages in question developed CVCVC roots, the picture changes
considerably. CVCVC roots apparently were developed after the major language
families separated so that that level we can either 1) notice retention of
the CVC core; 2) say that CVCVC was not preserved. Is the glass half full or
half empty?

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From martinez at eucmos.sim.ucm.es  Tue Jul 13 17:15:16 1999
From: martinez at eucmos.sim.ucm.es (Javier Martinez)
Date: Tue, 13 Jul 1999 19:15:16 +0200
Subject: Schleicher's fable
Message-ID: <TUE.13.JUL.1999.191516.0200.>

> thinking of Schleicher's fable (in his very Sanskrit-like PIE), which was
> re-worked by Hirt (his version is filled with reduced vowels, but otherwise
> looks very Neogrammarian), and most recently to my knowledge by Lehmann and
> Zgusta (with Lehmann's versions of the laryngeals and of PIE syntax).

cf. also the trilarygalictic version of the fable by Martin Peters in H.
Birkhan, Etymologie des Deutschen, Bern, 1985 (Germanistische Lehrbuchsammlung,
15), p. 308.

j.m.
--
  Javier Martínez ~ Filología Griega y Lingüística Indoeuropea (A-327)
  Facultad de Filología  ~  Universidad Complutense ~  E-28040 Madrid
  Fax: +34- 9131 49023~ Tlf. +34- 91314 4471~ (secret.)+34- 91394 5289
       http://www.ucm.es/info/griego/
  ~ TITUS-Projekt: Vergleichende Sprachwissenschaft ~ Postfach 111932
  Universität Frankfurt ~ D-60054 Frankfurt
       http://titus.uni-frankfurt.de

[ Moderator's re-encoding, with apologies.  Not TeX-style ]

  Javier Marti'nez ~ Filologi'a Griega y Lingu"i'stica Indoeuropea (A-327)
  Facultad de Filologi'a  ~  Universidad Complutense ~  E-28040 Madrid
  Fax: +34- 9131 49023~ Tlf. +34- 91314 4471~ (secret.)+34- 91394 5289
       http://www.ucm.es/info/griego/
  ~ TITUS-Projekt: Vergleichende Sprachwissenschaft ~ Postfach 111932
  Universita"t Frankfurt ~ D-60054 Frankfurt
       http://titus.uni-frankfurt.de


From mcv at wxs.nl  Tue Jul 13 18:44:13 1999
From: mcv at wxs.nl (Miguel Carrasquer Vidal)
Date: Tue, 13 Jul 1999 18:44:13 GMT
Subject: accusative and ergative languages
In-Reply-To: <Pine.SO4.4.05.9907091402440.15900-100000@rask>
Message-ID: <TUE.13.JUL.1999.184413.GMT.>

Jens Elmegaard Rasmussen <jer at cphling.dk> wrote:

>>From the primitive and casual books at my disposal I do see that "zuen"
>and "zuten" mean 'he had him' and 'they had him' resp. I also believe I
>see that such auxiliaries are combined with a particularly short form of
>the participles, referred to by Schuchardt as the root of the participle;
>and "jo" is 'stick; beat' in its shortest form, says my little dictionary;
>and "mutil-a" is 'boy' with the article "-a", but without case or number
>marking. Therefore my persistent question: Why can't "mutila jo zuen" and
>"mutila jo zuten" reflect a construction that was earlier meant to express
>'the boy, he had him hit', 'the boy, they had him hit'? Schuchardt also
>gives "zen" to mean 'he was', so that if you gloss "mutila jo zen" as 'the
>boy was hit', it seems there is quite a bit of agreement that the verbal
>root is a participle by itself. I do not see in what way this makes the
>*diachronic* interpretation of "mutila jo zuen" any different from the
>Hindi preterites that are based on Sanskrit constructions of the type
>"A-Nominative + B-Genitive + PPP/nom." meaning earlier "A was (verb)-ed by
>B", but now simply "B (verb)-ed A." Where am I wrong?

Not necessarily wrong, but a complicating factor is the
simultaneous existence in Basque of a set of synthetic verb forms
(first and foremost of the auxiliary verbs "to be" and "to
have"), which, with one exception, are also absolutive/ergative
in nature.  A diachronic interpretation in terms of ancient
passives of the periphrastic forms alone does not explain all of
the ergative verbal morphology of Basque.

We need to demonstrate that the synthetic forms too can be
explained as old passives, and I'm not sure the evidence quite
points that way.  But I think it is possible to derive the Basque
synthetic forms from older periphrastic ones.

Although not quite on topic, let me explain.

If we look at the more or less regular verbs first, we have:

ETORRI "to come"
(present)  (past)
na-torr    nen-torr-en  "I come, came"
ha-torr    hen-torr-en
da-torr    ze-torr-en
ga-toz     gen-toz-en
za-toz     zen-toz-en
da-toz     ze-toz-en

EKARRI "to bring"
na-karr-ERG    ninde-karr-ERG-(e)n     "ERG bring<s>/brought me"
ha-karr-ERG    hinde-karr-ERG-(e)n
da-karr-ERG    --
ga-kartza-ERG  ginde-kartza-ERG-(e)n
za-kartza-ERG  zinde-kartza-ERG-(e)n
da-kartza-ERG  --

where the ergative endings are:
1sg.   -t -da-
2sg.m  -k -ga-
2sg.f. -n -na-
3sg.   -0 -0-
1pl.   -gu -gu-
2pl.   -zu -zu-
3pl.   -te -te-

With 3rd. person abolutive object, however, we have in the
transitive past tense:

ne-karr-en      ne-kartza-n  "I brought it/them"
he-karr-en      he-kartza-n
(z)e-karr-en    (z)e-kartza-n
gene-karr-en    gene-kartza-n
zene-karr-en    zene-kartza-n
(z)e-karr-te-n  (z)e-kartza-te-n

So we have the following pronominal affix sets:

PRES  PAST (itr) PAST (tr1)   PAST (tr2)  ERG      DAT     [PP]
na-   nen-       ninde-       ne-         -t -da-  -t -da-  ni
ha-   hen-       hinde-       he-         -k -ga-  -k -ga-  hi
,,    ,,         ,,           ,,          -n -na-  -n- na-  ,,
da-   ze-        --           e-          -0       -o       --
ga-   gen-       ginde-       gene-       -gu      -gu      gu
za-   zen-       zinde-       zene-       -zu      -zu      zu
da-   ze-        --           e-          -te      -ote/-e  --

My proposal is that the first three sets contain an auxiliary
element *DA (present) / *DE(N) (past), as follows:

na- < *na-da-   nen-/ninde- < *nen-de-
ha- < *ha-da-   hen-/hinde- < *hen-de-
da- < *# -da-   ze-         < *    de-
ga- < *ga-da-   gen-/ginde- < *gen-de-
za- < *za-da-   zen-/zinde- < *zen-de-

As proof, I offer the paradigm of IZAN "to be", which in the 1st
and 2nd pp. sg. is composed regularly out of prefix + root
IZA(N), but in the other persons is composed of the auxiliary
DA/DEN alone (reduplicated in the plural):

naiz (naz, niz)   < *na-a-iz(a) < *na-da-iza
haiz (haz, hiz)   < *ha-a-iz(a) < *ha-da-iza
---
da                              < *#-da
gara (gera, gira) < *ga-ira     < *ga-di.da
zara (zera, zira) < *za-ira     < *za-di.da
dira                            < *di.da

nintzen (nintzan)      < *nen-de-izan-en
hintzen (hintzan)      < *hen-de-izan-en
---
zen (zan)              < *dan-en
ginen (ginan, gindan)  < *gen-di.dan-en
zinen (zinan, zindan)  < *zen-di.dan-en
ziren (ziran)          < *di.dan-en

We see that *d has suffered various phonological developments,
tentatively:

z-  in absolute initial
-d- in sandhi after an absolutive (pro)noun
-r- between vowels, before stressed?
-0- between vowels, before unstressed?

The -ai- in <naiz>, <haiz> (vs. -e- or -a- in <gara>, <zara>)
points to an earlier long vowel (*naaiz < *na-da-iza), just as in
the case of <nau->, <hau-> ("have me, have you") vs. <du-> "have
it" from *na-da-du (> *naau-), *ha-da-du (> *haau-) vs. *da-du (>
*dau-).

Another case where the "auxiliary" -d- resurfaces in the
transitive past is the verb JOAN "to go", whose past tense goes:
ninDoan, hinDoan, zihoan, ginDoazen, zinDoazen, zihoazen.

Reduplication (instead of the usual suffixation of *-(t)z(a)) to
denote plurality of the absolutive might also underlie the forms
of the verb "to have", if we allow for metathesis of the -i-:

nau-   < *naau-    < *na-da-du-
hau-   < *haau-    < *ha-da-du-
du-    < *dau-     < *#- da-du-
gaitu- < *gaaiddu- < *ga-da-di.du-
zaitu- < *zaaiddu- < *za-da-di.du-
ditu-  < *daiddu-  < *#- da-di.du-


The transitive past with 3rd. person object apparently contains
no auxilary.  I would reanalyze the forms as following:

*n(e) e-karr-en
*h(e) e-karr-en
*     e-karr-en
*gen  e-karr-en
*zen  e-karr-en
*     e-karr-en

The e- is possibly the same prefix as in the participle (e.g.
e-karr-i).  Assuming a word break between pronoun and verb avoids
the necessity of reconstructing 1 & 2 pl. as *genne-, *zenne- in
Pre-Basque.  That leaves the interesting question of why the 1st
and 2nd sg. forms have -n (nen-, hen-) in one set of past tense
prefixes, but not in the other (although in Gipuzkoan at least,
variants like <netorren>, <hetorren> do exist).  The 2sg.fem. in
-n (which we see in the ergative) might explain it for the 2nd.
person (cf. a similar phenomenon in Hausa, where the 2nd. person
feminine behaves morphologically like a plural in some personal
prefixes).

Finally, I notice that in this reconstruction, the auxiliary *DA
would have been used in both transitive and intransitive forms
(except 3sg. and pl. of the verb "to be", and the acc/nom.
transitive past).  I further suspect that the same *DA is present
in the causative formant -ra- (e.g. e-ra-man "to carry" <
*e-da-oan "to cause to go", etc.).  Originally "to do, make" [cf.
English "to do" as an auxiliary]?

=======================
Miguel Carrasquer Vidal
mcv at wxs.nl


From colkitto at sprint.ca  Thu Jul 15 01:47:14 1999
From: colkitto at sprint.ca (Robert Orr)
Date: Wed, 14 Jul 1999 21:47:14 -0400
Subject: PIE vs. Proto-World
Message-ID: <WED.14.JUL.1999.214714.0400.>

-----Original Message-----
From: petegray <petegray at btinternet.com>
Date: Tuesday, July 13, 1999 2:22 AM

>Steven said:

>> Although i'm not prepared to [compose] a fable in PIE, i certainly do not
>> doubt that, .. it could be done with our current state of knowledge.

>Yes it has, quite recently.  A number of linguists were asked to give their
>versions, and they were published together in a special volume.  I'm afraid
>they prove you wrong - the diversity was enormous, with some linguists using a
>more-or-less Brugmannian approach, and others being much more glottalic or
>laryngeal.  The variety is great fun to see, and very educative!

>Peter

>[ Moderator's request:
>  Would you mind providing bibliographical information for the volume in
>  question?  I for one would be fascinated.
>  --rma ]

So would I - Robert Orr


From Georg at home.ivm.de  Tue Jul 13 22:22:59 1999
From: Georg at home.ivm.de (Ralf-Stefan Georg)
Date: Wed, 14 Jul 1999 00:22:59 +0200
Subject: accusative and ergative languages
In-Reply-To: <006c01beca76$3bff6ea0$ec142399@patrickcryan>
Message-ID: <WED.14.JUL.1999.002259.0200.>

>Pat:

>Yes, but if a new verb were formed, e.g. webben (from English 'web'), there
>is no possibility that it would be conjugated webben, wabb, gewobben.

>What I consider 'internal inflection'  is Arabic yaktubu, kataba, katibun,
>etc. which applies to any and all verbs, old, and those taken new into the
>language.

The pattern lost productivity in German. So what ?
Also, I hardly understand your claim, which basically amounts to refusing
to speak of technique X in language L, unless it isn't present in each and
every item in which it theoretically could be present. So, you would refuse
to speak of, say, vowel harmony in, say, Turkish, because some borrowed
(and some native) words fail to obey this rule ? ("What I call VH is ...,
which applies to any and all ... Period. There is an exception, or the
phenomenon is gradually losing ground, therefore it doesn't extst at all."
?????).

Taking your example with the artificial verb "webben" (maybe meaning
something  like "to surf the web", for which we already adopted "surfen").
You are right that it could be expected to be inflected according to the
weak pattern. But, if some people choose to use the strong past participle
(leaving aside the simple preterite, which is rarely used in colloquial
speech nowadays), maybe out of some whim, maybe with humorous intentions,
and manage to place this strategically in the mass-media, say, in
online-service-ads, I see no stringent reason why this shouldn't fall on
fertile soil and get used gradually. First as a joke, later as a habit. The
advertising industry *is* able to do this and other things to your or my
language. The pattern is there, and any existent pattern can trigger
analogy.Maybe not too likely, I admit in this case, but certainly not
impossible. Analogy is not eternally confined to pre-modern times.

Stefan Georg
Heerstrasse 7
D-53111 Bonn
FRG
+49-228-69-13-32


From larryt at cogs.susx.ac.uk  Wed Jul 14 10:19:08 1999
From: larryt at cogs.susx.ac.uk (Larry Trask)
Date: Wed, 14 Jul 1999 11:19:08 +0100
Subject: `cognate'
In-Reply-To: <ef25a75f.24b81413@aol.com>
Message-ID: <WED.14.JUL.1999.111908.0100.>

On Fri, 9 Jul 1999 JoatSimeon at aol.com wrote:

[LT]

>> But I have noticed that a significant number of
>> non-linguists on this list have apparently misunderstood the sense of our
>> technical term `cognate': many of them
>> appear to believe that `cognates' means something like `words of similar
>> form

> - that's odd; I always assumed it meant "derived from a common
> ancestral word".  You know, like Tiwaz and dyaus.

Yes, but non-linguists often misunderstand the term, and I've noticed
several queries on this list in the last few weeks from people who had
apparently misunderstood it.

By the way, the following article presents a small but spectacular
sample of cognates which do not resemble one another at all:

John Lynch (1999), `Language change in southern Melanesia: linguistic
aberrancy and genetic distance'. In Roger Blench and Matthew Spriggs
(eds), Arachaeology and Language IV, pp. 149-159, London: Routledge.

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk


From jer at cphling.dk  Wed Jul 14 01:04:18 1999
From: jer at cphling.dk (Jens Elmegaard Rasmussen)
Date: Wed, 14 Jul 1999 03:04:18 +0200
Subject: accusative and ergative languages
In-Reply-To: <006801bec8b4$e04f1460$bdf9abc3@xpoxkjlf>
Message-ID: <WED.14.JUL.1999.030418.0200.>

On Wed, 7 Jul 1999, petegray wrote:

> [...]
> Proto-proto- languages simply cannot be recovered with the same degree of
> certainty as languages nearer to our attested texts.

At the risk of shocking some and destroying what little reputation I may
have, I'd say: That depends on where the lines meet.

Thus, on the basis of a handful of Modern Indo-European languages taken
from different branches you can hope to get little glimpses of the
language of the youngest period common to all of them, i.e. the IE
protolanguage. You cannot get Russian + English + Persian + Greek +
Italian + Albanian + Irish to tell you what 'name' was in Old Russian, or
in Proto-Balto-Slavic; your only result, if any, would apply to
Proto-Indo-European.

Therefore, in many respects we are better informed about very old periods
than about the intervening stages closer to the attested languages. But do
go easy with this truth: it is immediately misused to discredit our most
objective scholarship if told to the wrong people.

Jens


From larryt at cogs.susx.ac.uk  Wed Jul 14 10:20:17 1999
From: larryt at cogs.susx.ac.uk (Larry Trask)
Date: Wed, 14 Jul 1999 11:20:17 +0100
Subject: `cognate'
In-Reply-To: <Pine.SO4.4.05.9907121416130.5019-100000@rask>
Message-ID: <WED.14.JUL.1999.112017.0100.>

On Mon, 12 Jul 1999, Adam Hyllested wrote:

> On Wed, 30 Jun 1999, Larry Trask wrote:

[LT]

>> `father', all of these being descended from PIE *<pater>.

> I'm sure it says PIE *<p at ter> in your manuscript (+ diacritics on the
> <e>).

Yes, it does, and my posting should have read `p at ter', but I slipped up.

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk


From proto-language at email.msn.com  Wed Jul 14 01:20:09 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Tue, 13 Jul 1999 20:20:09 -0500
Subject: accusative and ergative languages
Message-ID: <TUE.13.JUL.1999.202009.0500.>

Dear Peter and IEists:

 ----- Original Message -----
From: petegray <petegray at btinternet.com>
Sent: Sunday, July 11, 1999 1:45 PM

>> Pat wrote:

>> Well, on page 40 of Chao's Mandarin Primer, are listed "Affixes":
>> 11 are listed; ...

Peter wrote:

> I simply cannot let this go.    Verbal compounds are rife in modern Chinese,
> and the second elements ("affixes" if we are using terms inappropriate for
> Chinese) are derived from a large number of concepts.   The fact that Chao
> lists 11 only means that he has selected 11, and does not reflect the actual
> nature of the Chinese language.   Things are not as simple as that grammar
> makes them seem!

Pat responds:

Fine, Peter. Why not list two or three additional elements that function in
Chinese equivalent to inflectional affixes?

 Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From edsel at glo.be  Wed Jul 14 10:21:19 1999
From: edsel at glo.be (Eduard Selleslagh)
Date: Wed, 14 Jul 1999 12:21:19 +0200
Subject: Chronology of the breakup of Common Romance [long]
Message-ID: <WED.14.JUL.1999.122119.0200.>

[ moderator re-formatted ]

-----Original Message-----
From: Rick Mc Callister

>> "Vlach" or "Wallach" seems to be descended from a name that was commonly
>> applied to Celts.

> It has long been proposed that the name Volcae, a Gaulish tribe, was the
> source of Anglo-Saxon Wealh, Balkans Vlakh, Polish WLochy, etc.  The point
> I've generally seen is something to the extent that supposedly the Volcae
> were the first Gaulish tribe to fall under Roman rule.

[Ed Selleslagh]

That seems highly unlikely since the term was originally used exclusively by
non-Romans, non-Celts, and applied to Celtic and Latin languages speaking
peoples.  That excludes a Roman origin of the word.
All this sounds like a typical 19th century invention by Latin-worshippers

> Some popular writers have also proposed Volcae as the source of
> English "folk" but I've never seen this proposal from professional
> linguists.

[Ed]

Even less likely : the peoples that used the word had probably never heard
of the Volcae, not even second-hand.

>> Appearing as "Walh" or "Walah" in OHG, it has interpreted as meaning
>> "foreigner", sometimes Roman, but in usage it is closely associated with
>> Celts and regions of Celtic habitation - e.g., "Wales", "Walloon".  I seem
>> to recall an suggestion that there may be a tie here to L. vallum (fortified
>> wall, the earliest meaning of "wall") and refer to the Celtic or
>> Romano-Celtic oppidum or walled town.

> You'd have to account for the ending in Vlakh, WLochy, etc.

[Ed]

Indeed, and a derivation from a Latin word is unlikely for the reasons given
above.

Note that the form with -ch is East-European, while the 'Wal-' form is
Western, mainly Germanic and in loanwords of Germanic origin, e.g. in Latin
languages (Walen-Wallons, Wallis-Valais, Wales...).

>> The strong Celtic presence along the Danube is attested by Classical writers
>> well before the common era.  E.g., Alexander fights them before turning
>> against the Persians and meets Celts who are from Illyria.  Galicia in
>> southern Poland is a region whose name remembers a Celtic presence even
>> farther north.  I remember an old passage where Vlachs are identified as
>> "the shepherds of the Romans" and in this role they may also have been
>> imported help as they were in northern Italy.

> Isn't the term Vlakh anachronistic in this sense? Vlakh is first
>documented in the middle ages, isn't it?

[Ed]

Indeed. The medieval use of 'Vlach' probably referred to the 'Latin'
Rumanians, cf. Wallachia.
In modern Greece, the word is also used to refer to the local nomads/gypsies
of unclear (i.e. to me) ethnic origin (maybe Rumanian Gypsies, 'Roma', or
from Pannonia? Maybe Albanians?).

>> The ancient ethnic designations in that part of the world are a little
>> difficult to follow, but it seems clear that the whole region from Illyria
>> to present day Romania was under Rome by 250 ace.  And there is some
>> possibility that Vlachs represented Romanized Celts across those regions.
>> In any case, the small difference in time between Rome's entry into Illyria
>> and into Dacia would be a de minimis factor.

> My guess is that any Celtic element in the Balkan would have been extremely
> thin.

> The Celts raided the Balkans and stormed through the region on the way to
> Galatia but my understanding is that they lived north of the Balkans in
> Galatia and Pannonia

[Ed]

See my last remark above.

[snip]

Ed. Selleslagh<edsel at glo.be>


From proto-language at email.msn.com  Wed Jul 14 01:52:32 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Tue, 13 Jul 1999 20:52:32 -0500
Subject: Interpreting ergative sentences
Message-ID: <TUE.13.JUL.1999.205232.0500.>

Dear Leo and IEists:

 ----- Original Message -----
From: <CONNOLLY at LATTE.MEMPHIS.EDU>
Sent: Monday, July 12, 1999 12:01 AM

<snip>

> Sumerian is even worse, since we don't have all that much of
> it (no poetry, no menus, nothing that seems to be finely nuanced
> speech.  And on top of that, it's ergative, and most of us (including
> me) are *not* used to dealing with such languages.

 <snip>

> So please, caution!  And take due note of what has been discovered
> about *living* ergative languages.  You can't hope to understand the
> dead ones without them.

Pat comments:

I think everyone will agree that this is an interesting posting. And the
cautions that Leo advises should be taken to heart by anyone dealing with
languages of the past.

But, Leo, there is quite a lot of Sumerian poetry --- some quite intriguing.

Pat


From edsel at glo.be  Wed Jul 14 11:34:17 1999
From: edsel at glo.be (Eduard Selleslagh)
Date: Wed, 14 Jul 1999 13:34:17 +0200
Subject: `cognate'
Message-ID: <WED.14.JUL.1999.133417.0200.>

-----Original Message-----
From: JoatSimeon at aol.com <JoatSimeon at aol.com>
Date: Wednesday, July 14, 1999 12:43 AM

>>larryt at cogs.susx.ac.uk writes:

>>This posting will not be news to the linguists on this list, who may prefer
>>to stop reading at this point, unless they want to offer criticisms of the
>>definitions below.  But I have noticed that a significant number of
>>non-linguists on this list have apparently misunderstood the sense of our
>>technical term `cognate': many of them
>>appear to believe that `cognates' means something like `words of similar
>>form

>- that's odd; I always assumed it meant "derived from a common ancestral
>word".  You know, like Tiwaz and dyaus.

[Ed Selleslagh]

That's what I meant when I caused all this by calling Grk. 'cheir' (N.Grk.
'cheri') and Georgian (Kartvelian) 'cheli' cognates. I should have said
'possible cognates' since I was asking the list members whether these
actually were or could be cognates.

Sorry for all the fuss, but at least things got clearer (1. they probably
aren't cognates, 2. we, the non-professionals,  have a better understanding
of what cognates are).

Ed.


From proto-language at email.msn.com  Wed Jul 14 07:49:56 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Wed, 14 Jul 1999 02:49:56 -0500
Subject: accusative and ergative languages
Message-ID: <WED.14.JUL.1999.024956.0500.>

Dear Ralf-Stefan and IEists:

 ----- Original Message -----
From: Ralf-Stefan Georg <Georg at home.ivm.de>
Sent: Saturday, July 10, 1999 5:03 AM

R-S wrote previously:

>>> Having to be rendered by the passive in English is not the same thing as
>>> "being passive in nature".

>> Pat responded:

>> How about explaining "passive in nature"?  Is that a Platonic idea?

R-S then responded:

> As you might have observed, I put double quotes around this expression, the
> most widespread functions of which are:

> - the expression is not mine, but used by others, and I use it only not to
> complicate the discussion, assuming everyone knows what I'm talking about

> or

> - the expression is handy, but admittedly imprecise, I only use it since I
> don't want to put too much effort in terminological precision at this
> point, having different things to communicate (admittedly a vice, but
> normally tolerable when speaking to attentive and knowledgable people)

Pat responds:

All very well and good. Having explained next to nothing, perhaps you will
now tell us how precisely *you* were using "passive in nature" and what in
God's name it is supposed to mean to you, to those you may be quoting, or
anyone else including me.

R-S continued:

> I was reacting to some claim of yours, namely that ergatives are - now what
> was your expression ? I don't recall, was it "really passives", "always
> former. i.e. reanalyzed p.s", "p.s in nature", "p.s by nature". Hell, I
> dunno. Something to this effect, at least. The important thing is that I
> managed to point out that this, i.e. the ergative-as-passive-claim, however
> phrased, is wrong, and Larry Trask, who has also a well-known publishing
> record on the issue, has spoken the definite word about this here.
> Pat, this is, alas, a familiar pattern: when one's theory gets into dire
> straits, some aside discussion on irrelevant points is opened, obviously
> with the intention to show weaknesses of whatever kind in the opponent's
> standing, hoping that this will cloud the correct and justified remarks he
> might have made on the relevant points before. This may work in some
> election campaign, but not here, sorry.

Pat responds:

Yes, you are absolutely correct. This is a familar pattern indeed. When
challenged for an argument, you sidestep the issues of the question by
conveying that someone you like (here, Larry Trask) "has spoken the definite
word about this here". If Larry could speak the definite word about
everything, then this list would be a waste of time. We could just subscribe
to his newsletter for the latest ex cathedra rulings on all our troublesome
questions.

Now, when Larry recently quoted Dixon about the nature of the ergative, he
conveniently neglected to mention that Dixon acknowledged that there were
currently practising linguists --- not amateur linguists like myself ---
still defending the passive interpretation of ergative constructions. I
asked Larry where he had "shredded" this interpretation, and to my
knowledge, got no answer. If I missed the "shredding", perhaps you will be
kind enough to rehearse his performance for us. I have seen nothing by
Larry's vehemence and your allegiance to support the idea that the ergative
should not be interpreted as a passive.

R-S then continued:

> To answer your question: "passives in nature", as used by me, is neither a
> Platonic idea, nor a Wittgensteinean prototype, nor a correct description
> of anything I may happen to think, it is just shorthand for a bundle of
> concepts I don't share. And, above all, it is bad English. Should we open a
> new thread about this, or shouldn't we better return to ergativity and your
> misconceptions about it, such as "there are ergative languages without any
> splits" athl. ?

Pat concludes:

That Ralf-Stefan is incapable of defining the term he introduced: "passive
in nature". By the way, being a native speaker of English, I can assure you
that in my dialect, "passive in nature" is *not* bad English.

And to answer your -- I hope not purposeful -- distortion of what I wrote,
let me say explicitly that I did not assert "there are ergative languages
without (any) splits". I asserted that Thomsen did not, at least in her
grammar, identify splits in Sumerian, which you seemed to think she had. I
then invited you to identify them in Sumerian if you *could*.

You might review your own procedures for employing quotation marks. Your use
of them on "there are ergative languages without any splits" strongly and
falsely implies that I wrote this in the context of a judgment on the
question.

 Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From rmccalli at sunmuw1.MUW.Edu  Wed Jul 14 19:25:03 1999
From: rmccalli at sunmuw1.MUW.Edu (Rick Mc Callister)
Date: Wed, 14 Jul 1999 14:25:03 -0500
Subject: Momentary-Durative
In-Reply-To: <001901becc6c$a3651e80$8471fe8c@lucent.com>
Message-ID: <WED.14.JUL.1999.142503.0500.>

Spanish distinguishes them
by resorting to

estar --a "momentary" verb--
[actually a verb indicating condition] for resultant conditions--

and ser--a "durative" verb
[actually a verb indicating  characteristics]
for passive constructions.

e.g. The store

Past-tense resultant conditions are usually descriptive and usually in the
imperfect while past-tense passive voices usually deals with completed
actions and is usually in the preterite

e.g.

The was closed. (resultant condition)
La tienda estaba cerrada.

The store was closed. (passive)
La tienda fue cerrada.

Curiously, in the present tense Spanish tends to use
a medio-passive type of construction,
instead of passive voice.

The store is closing/being closed. (medio-passive)
Se cierra la tiendas.

The store is closed. (resultant condition)
La tienda esta/ cerrada.

Does anyone know why?

[snip]

>Even worse, in modern Western European languages, resultatives and passives
>cannot be readily distinguished, and perfect and resultatives are also
>similar. Thus we take ``John has eaten dinner'' to be resultative, but
>``John has eaten ostrich meat'' as experiential perfect. ``The door is
>closed'' can be resultative or passive, while ``The door is closed by John''
>is passive; the resultative has be expressed as ``The door is closed because
>of John'' or some such.

[snip]

Rick Mc Callister
W-1634
Mississippi University for Women
Columbus MS 39701


From proto-language at email.msn.com  Wed Jul 14 07:58:06 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Wed, 14 Jul 1999 02:58:06 -0500
Subject: accusative and ergative languages
Message-ID: <WED.14.JUL.1999.025806.0500.>

Dear Ralf-Stefan and IEists:

 ----- Original Message -----
From: Ralf-Stefan Georg <Georg at home.ivm.de>
Sent: Saturday, July 10, 1999 5:27 AM

>> Pat wrote:

>> Well, on page 40 of Chao's Mandarin Primer, are listed "Affixes":

>> 11 are listed; of these 6 qualify as related to "inflections" :
>> modal -m(en), phrase marker -le; completed action -le; progressive
>> action -j(y/e); possibility or ability -de; subordination -de.

>> Undoubtedly, a historical grammar might provide a few more but I consider
>> this a pretty simple system.

R-S sighed:

> This is getting weary. I think no sane linguist will be unaware of the fact
> that there are languages with fewer purely morphological means than others.

Pat sympathizes:

Apparently not.

R-S continued:

> Also, "simpler" and "not-so simple" phonological systems have been heard of.
> But, if I'm not completely mistaken, there was talk about sthl. like
> general, overall simplicity of languages, which is a pre-scientific notion,
> quite simply, if this pun is allowed.

> If we assume, as of course we have to, that human languages are
> problem-solving devices which face a set of possible communicative
> problems, which is the same for all linguistic communities (don't come with
> obvious cultural differences, which pertain to vocabulary only; I'm aware
> of the fact that not all linguistic communities of the planet need to talk
> meaningfully about the various kinds of fish in the Yenissey, or the
> Sepik). Languages make different choices regarding the set of functions
> they grammaticalize (tense vs. aspect or both, inferentiality, relative or
> absolute tense, number, participant-identification, reference-tracking,
> pragmatic categories, you name it), the hows and whys of which are the
> object of linguistic typology. One language uses grammaticalized
> bound-morphology for, say, inferentiality, others have to use different
> means to convey the idea. But they are all able to convey the idea, however
> elegant, or however clumsy.

Pat interjects:

Yes, some people use a drill-stick and bow to make a fire, and some use
matches.

R-S continued explaining:

> The idea is that they *use means* to do it, or
> that speakers may find these means, even if the category is weakly or not
> at all grammaticalized in their language (your inflection paradigm may be
> my intonation pattern, your inferentiality affix may be my expletive
> adverb; to convey the idea of the Dakota sentence-particle /yelo/ or Thai
> /khrap/ I may even be forced to say each time "I am male"; clumsy, but
> possible, if I face the necessity). The description of these means is the
> task of the grammarian. Some grammarians are aware of this task, and write
> a 600-page grammar of Chinese, some aren't (or simply want to produce a
> primer to help you get along abroad) and write a 60-pp. treatment of
> Russian. So what ?

> I will not deny that notions of simplicity vs. complexity may be useful
> distinctions when talking about subsystems. For the characterization of
> whole languages they are definitely not useful. The grammar-school like
> rote learning of paradigms, which your Chinese teacher will be able to
> spare you, will be compensated by having to digest subtle and difficult
> rules of syntax.

Pat concludes:

In my opinion, dearth of morphological devices is validly characterised as
"simple".

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From petegray at btinternet.com  Wed Jul 14 19:21:14 1999
From: petegray at btinternet.com (petegray)
Date: Wed, 14 Jul 1999 20:21:14 +0100
Subject: Momentary-Durative
Message-ID: <WED.14.JUL.1999.202114.0100.>

Vidhyanath's two postings on aspect show clearly what the issue of aspect in
PIE may be - we simply don't have a clear definition of what we are talking
about.

Greek usage does not agree with Vedic, and where traces of aorists exist in
other languages (e.g. Armenian, OCS, Albanian, Latin, Baltic), we cannot
convincingly recover an original aspectual meaning.

The formations are clearly PIE, but I believe I can continue to claim that
the distinctions between them, if recoverable, are certainly not those we
find in Greek.   The Armenian-Greek- I-I sprachbund has certain known
innovations, such as vocabulary and the augment, and we should not be miseld
into accepting as original what may be another innovation, namely the
development of new distinctions based on an existing variety of forms.

Peter


From Georg at home.ivm.de  Wed Jul 14 08:52:08 1999
From: Georg at home.ivm.de (Ralf-Stefan Georg)
Date: Wed, 14 Jul 1999 10:52:08 +0200
Subject: accusative and ergative languages
In-Reply-To: <001001becadf$e28d9580$759ffad0@patrickcryan>
Message-ID: <WED.14.JUL.1999.105208.0200.>

>Pat responds:

>I do not readily have access to material on Khalkha Mongolian so I cannot
>comment in detail on your example.

>However, in general, I would hazard the guess that, regardless of its
>apparently accusative construction, for all practical purposes, this
>sentence could be just as easily translated: 'A book was sent to you'.

If I had to *translate* a Mongolian novel into English, I might choose the
passive as well, in *English*, since the "agentlessness" of this sentence
is something which goes well with English (and other) passives. But that's
*translation*, and not *analysis*. I chose Mongolian, because it has a
passive available. The active construction, as I gave it, will be most
naturally be found, when an agent is inferable from the
pragmatic/textual/situational context, passive being preferred when it is
unknown or information on it deliberately suppresses. there are other
parameters which govern the choice of diathesis, mostly of a pragmatic
nature and not entirely explored. The question was about grammaticality of
such a construction in an acc. language.

> I am
>also a little sceptical of an "accusative" in such a context which is marked
>by -0.

I made a slight mistake, thanks for your attentiveness. I should have
rendered the sentence in English with *a book*, not *the book*, the overt
accusative marker being used with definite patients only. With an
indefinite one, it is -0.

>But, having said that, I am willing to concede that for *some* accusative
>languages, this construction, provided a nominative subject can be mentally
>supplied from context, is grammatical.

OK, thanks, we are getting somewhere.

Stefan

Stefan Georg
Heerstrasse 7
D-53111 Bonn
FRG
+49-228-69-13-32


From petegray at btinternet.com  Wed Jul 14 19:29:06 1999
From: petegray at btinternet.com (petegray)
Date: Wed, 14 Jul 1999 20:29:06 +0100
Subject: Momentary-Durative
Message-ID: <WED.14.JUL.1999.202906.0100.>

>>> I don't see any parts of Sans lit in which aorist has resultative
>>> meaning.

If you didn't like my first example, how about:
  Ramayana 2:64:52
  yah s'aren.aikaputram tvam aka:rsi:r aputrakam
 with one arrow you have rendered me childless, who had but one son

(Sorry you don't like my transliteration - it's due to using ASCII)

or Mahabharata 5:3:10
  tad akars.it prajagaram
  that has produced sleepiness

or Mahabharata 2:60:7
  kim nu pu:rvam para:jais.i:t a:tma:nam
 whom you have lost first, yourself....

In all of these examples the result of the action is the point of the
statement - and it is an aorist which is used to describe it.

Peter


From W.Schulze at lrz.uni-muenchen.de  Wed Jul 14 10:18:26 1999
From: W.Schulze at lrz.uni-muenchen.de (Wolfgang Schulze)
Date: Wed, 14 Jul 1999 12:18:26 +0200
Subject: accusative and ergative languages
Message-ID: <WED.14.JUL.1999.121826.0200.>

"Patrick C. Ryan" schrieb:

>.... As for lessened degrees of animacy, most ergative languages
> have antipassives to indicate this.

Before you invite people to subscribe to this claim you should first
demonstrate that a) "most ergative languages have antipassives". This
claim suggests that "ergative" is a substantial attribute that can be
used with the referent "language". In earlier postings I have tried to
show that "ergativity" (as well as "accusativity") represents a label
for a structural BEHAVIOR of single paradigms WITHIN a language system.
For instance (as I have said) a language system may be ergative in its
agreement paradigm, accusative in owrd order, accusative in case
marking, ergative in discourse cohesion etc. (to give a fictive
example). Hence, there are NO "ergative (or "accusative") languages" or
only, if you use this term in a very informal sense. Now, if you talk
about antipassives, you should make clear to which morphosyntactic
domain you allude to. Moreover, your claim suggests that most 'ergative
languages' are reference dominated in the sense of Role and Reference
Grammar. Only if a given language system ("operating systems in terms of
the Grammar of Scenes and Scennarios" (GSS)) uses actant encoding
devices to indicate fore- and/or backgrouding (instead of - for instance
- smeanti coles such as agent/patient...) we can expect some kind of
diathesis be it passive or antipassive (note that again passives and
antipassives represent two poles on a much more complex scale that also
involes bi-absolutives, pseudo-passives and many more structures).
	The fact, however, is that many 'ergative' languages lack an
antipassive. For instance, there are nearly 30 East Caucasian languages
all of them using some ergative strategies in at least parts of their
operating systems. But only a handfull of them (five or six, to be
precise) have true antipassives (only one has some kind of
"pseudo-passive"). The same is true for accusative systems (as you
probably know). Hence, antipassives are a possible extension of ergative
stretagies, they cannot serve for any kind of typological
generalization.

b) Antipassives have rarely to do with the "lessening of animacy". The
most common inferences that allow antipassive structures are:

- Reduction of 'activity' (that is the degree to which an actant is
thought to be 'active' during a *specifc* (and single) event. From this
another inference is given:

- Habitual, durative action (-> imperfectiveness).....

- The event becomes less discrete, hence less transitive. Another
inference: The 'patient' looses its referentiality: It cannot be
subjected to wh-questions, it cannot be counted, very often such
referents are mass nouns or collectives...

- Antipassives are part of the discourse cohesion strategies (most
famous example is Dyirbal): Here, antipassives are neither semantically
nor syntactically motivated, but merely a pragmatic feature of topic
chaining.

There are much more functional options that are carried out by
antipassives. In fact, all these options NEVER allow such a claim as
quoted above (rather, they contradict it).

> Although Dixon is certainly a man who has devoted much thought to
> ergativity, I find something inherently problematical in combining ergative
> and accusative features in one sentence (a little schzophrenic) which he is
> forced to do by analyzing pronominal and nominal structures differently when
> they occur in the same sentence.

WHY? It all depends from how you interpret erg. and acc. features.
Confer for instance the following (one!) sentence from (informal) Lak
(East Caucasian):

t:ul b-at:-ay-s:a-ru zu
I:ERG I:PL-hit-PART:PRES-ASS-SAP:PL you:PL:ABS
'I surely hit you (plural)'

[For the expert: Standard (Literary) Lak would have 'na bat:ays:aru zu'].

This sentence is:

ACC with respect to word order [*zu t:ul bat:ays:ara would be ERG]
ERG with respect to case marking ['neutral with 'na' for "I" is also
possible]
ERG with respect to class agreement (b- = (here) class [+hum;+plural])
ERG with respect to personal (or, better, speech act participant)
agreement (-ru is SAP:PL and agrees with 'zu' "you:PL").

But if you say "I am surely hitting you (plural)", you get:

na b-at-la-ti-s:a-ra zu
I:ABS I:PL-hit1-DUR-hit2-ASS-SAP:SG you:PL:ABS

Here we have:

ACC with respect to word order
ERG with respect to calss agreement
ACC (or neutral) with respect to case marking)
ACC with respect to SAP agreement (-ra is triggered by 'I:ABS').

Now, please tell me: Is Lak an 'ergative' or an 'accusative' language?
[Please note that I did not include (among others) strategies of
discourse cohesion, reflexivization and logophization].

> I think it is likelier that, because of perceived greater animacy (or
> definiteness), pronouns have a different method of marking that can still be
> interpreted within an ergative context.

This a (very simplified) 'on-dit' that stems from the earlier version of
the Silverstsein hierachy. Again, we have to deal with the question,
whether a 'pronoun' (I guess you mean some kind of 'personal pronouns')
can behave 'ergatively' or 'accusatively'. The list below gives you a
selection of SAP case marking in East Caucasian languages with respect
to ABS/ERG:

	ABS vs. ERG	ABS = ERG
	ALL		---
	Singular	Plural
	Plural		Singular
	1.Incl.		Rest
	1:SG		Rest
	2:SG		Rest
	1:SG/PL		Rest
	---		ALL	

This list (aspects of personal agreement NOT included!) shows that SAP
pronouns may behave different within the same paradigm. Any
generalization like that  one quoted above does not help to convey for
these data...

[Please note new phone number (office) :+89-2180 5343]
___________________________________
| Prof. Dr. Wolfgang Schulze
| Institut fuer Allgemeine und Indogermanische Sprachwissenschaft
| Ludwig-Maximilians-Universitaet Muenchen
| Geschwister-Scholl-Platz 1
| D-80539 Muenchen
| Tel:	+89-21802486 (secr.)
|      	+89-21805343 (office) NEW ! NEW !
| Fax:	+89-21805345
| Email: W.Schulze at mail.lrz-muenchen.de
| http://www.lrz-muenchen.de/~wschulze/
_____________________________________________________


From petegray at btinternet.com  Wed Jul 14 20:19:23 1999
From: petegray at btinternet.com (petegray)
Date: Wed, 14 Jul 1999 21:19:23 +0100
Subject: Momentary-Durative
Message-ID: <WED.14.JUL.1999.211923.0100.>

Thank you for your long reply, Jens.   Firstly, I apologise for any offence
in the word "ideology".   I intended only to indicate what I see (rightly or
wrongly) as assumptions based on theories.

Now, going through your posting in detail:

>  is the common occurrence of a
> nasal-infix structure from *tewd- 'thrust' in Skt. tundate and Lat. tundo
> not part of our "data"?

Yes - but you have made the assumption that this root always shows the nasal
in the present.   This is not the case.
(a) Albanian shows two forms, one with, and one without the nasal.
(b) Old Irish shows no nasal.
(c) Greek a-tuzomai shows no nasal.
(d) If Pokorny is right to link the root with stud- then we have no nasal in
Latin studeo, nor in the various reflexes of it in Germanic.

>  There are so many instances of the _same_ verbal root turning up
> with a nasal present in IE branches that have nothing else in common ...

And there are very many instances of roots showing nasal presents in one IE
language, and not in another, or indeed appearing in both forms in the same
language, as Latin cumbo and cubo, or Greek leipo: and limpano:

>. Is it a coincidence
> that *k^lew- 'hear' forms a nasal present in Indo-Iranian and Celtic?

Yes.   It has no nasal in Greek or Latin or some others. (and so on with the
rest of your list)

I cannot agree with you that certain verbal rootts always had nasal presents
in all IE languages.   This is simply factually untrue.   But without that
assumption, your argument collapses.

>. Does that not indicate
> that the assigment of the nasal-infix structure as the present of certain
> roots was fully lexicalized in the protolanguage?

No, for the reasons just stated.   There are enough counter-examples to show
that whatever the conditions were for the "selection" of nasal, full grade
or suffixed present, they are now as unrecoverable as they are within Latin,
Greek, or Sanskrit, where very little difference of meaning can be
consistently shown.

> Oh no, it works only in the direction that a nasal present has a root
> aorist beside it in practically all cases.

Not a surprise, if you ignore all the langauges where this is not the case.
Likewise, not a surprise, since root aorists were an old form.   The same
could be said of full grade presents, or suffixed presents.   So the
statement really is without significant meaning.

> Also the reduplicated presents
> and the y-presents generally form root aorists.

Yes - so your argument (that a root aorist implies a nasal present) is
really rather weak.   Yet without this argument, you cannot derive tuda'ti
the way you do.

> [in] PIE, each verb had mostly picked one form for its present. The massive
> concord among the IE languages in this respect

There is no such concord.  Even a single language shows variety of
formations of the present, and the variety across the languages is
considerable.    But it would be good to have firm data on this - anyone got
three weeks to spare going through the text books?

> - while the even greater discord is easily
> explained by continued normal language change.

No.   Normal language change would not turn a nasal present into a
reduplicated one or vice versa, nor would it turn a -sk suffixed one into a
nasal present, etc, etc.

Peter


From edsel at glo.be  Wed Jul 14 10:54:06 1999
From: edsel at glo.be (Eduard Selleslagh)
Date: Wed, 14 Jul 1999 12:54:06 +0200
Subject: Interpreting ergative sentences
Message-ID: <WED.14.JUL.1999.125406.0200.>

-----Original Message-----
From: CONNOLLY at LATTE.MEMPHIS.EDU <CONNOLLY at LATTE.MEMPHIS.EDU>
Date: Wednesday, July 14, 1999 3:06 AM

>Dear Pat and everyone else,

[snip]

>This is actually quite like what often happens with experiencer and
>beneficiary verbs in accusative languages.  The fact that e.g. Spanish
>makes the patient the morphological and syntactic subject of a verb
>does not mean that an English translator must do the same.  _No me
>gusta la mu'sica_ does not mean 'the music doesn't please me' -- that
>translation suggests that someone was trying to please (or perhaps
>annoy) me with the music.  Rather, it means, very precisely, 'I don't
>like the music': the Spanish doesn't insinuate that anyone was thinking
>of me, so neither should the translation.

[Snip]

>Leo

[Ed Selleslagh]

First of all, thank you for this excellent piece of work.

Speaking about Castilian, I would like to have your views on a peculiarity,
or rather a tendency, that is still productive at least in popular speech
('le-ismo'), namely the tendency to use the indirect subject form where all
other Latin (and other West-European) languages use the direct object form,
and almost exclusively with persons (animate), e.g. "le ví" ("le vi' "), "I
saw him".
It has sometimes be suggested that this was a substrate influence from
Basque, an ergative language (that lacks an accusative, of course), but
opinions are extremely divided on this subject. On the other hand, it seems
to me this could (but I don't know) also be related to some of the arguments
presented in the discussion on ergative/accusative concerning the distinct
role of animates.

Ed.


From jer at cphling.dk  Thu Jul 15 02:20:50 1999
From: jer at cphling.dk (Jens Elmegaard Rasmussen)
Date: Thu, 15 Jul 1999 04:20:50 +0200
Subject: Momentary-Durative
In-Reply-To: <001a01becc6c$a4de02e0$8471fe8c@lucent.com>
Message-ID: <THU.15.JUL.1999.042050.0200.>

On Mon, 12 Jul 1999, Vidhyanath Rao wrote:

  [snip of my (Jens') message]

> The tuda'ti type, with primary endings, is somewhat rare in RV. Perhaps,
> tuda'ti type was a replacement for injunctives, i.e., tuda'ti was a general
> present and not progressive present. Just a thought.

Hardly so, for the injunctive is not a stem-formation, but an inflectional
category that can be formed from all verbal stems.

> [Jens:]

>> Strunk has shown that nasal presents go with root aorists,

> [VR:]

> Reduplicated forms seem to go with root aorists as well. [...]

Indeed they do, the percentage is _very_ high in both types. In other
words, the nasal and teh reduplicated present share the same aorist type,
and in fact they have another rival in the y-presents. I suppose
durativity could be of different kinds, e.g. duration, repetition, or a
mere attempt.

> This also bears on the other proposal, going back to Kurylowicz at least,
> for the origin of (asigmatic) aorist forms: namely, these are old preterits
> that became limited to aorist function due to the rise of new presents.

If my observation that there is an alliance between the sk^-present type
and the s-aorist is correct, and the morpheme *-sk^e/o- is a development
of the expected *-s-ye/o- (implementing the "lengthening s" also known
from the nom.sg., which must earlier have been a different phoneme from
other sources of IE /s/), then the s-aorist was originally inchoative in
function. I'd say that makes very good sense, for the s-aor. is also
widely used with verbs that form radical or thematic presents, as
*weg^h-e/o- 'drive', whose aor. *we:g^h-s- will then originally have meant
'start driving, set out (by carriage)'.

> Incidentally, what are your thoughts on the origin of nasal presents? The
> rarity of infixes in PIE has led to the proposal that nasal presents
> originated from a double affix, that is *wined- was really *wi-n-ed- (or
> *wi-ne-d). This would mean that originally win(e)d- and w(e)id (or wide/o)
> were not grammatically associated. In case of dehmi lehmi etc, it then
> becomes a question of when the roots came into existence. I am not sure if
> we have enough information to decide this.

The double suffix theory has little to recommend it, and for some root
types the nasal-infix structure shows surprises that can only be
reasonably understood as the result of metathesis. Thus roots ending in
*-yH and *-Hy (the later being the "long diphthongs" like *dheH1y- 'suck')
both form nasal presents with the -y- before the nasal infix and the -H-
after it. The natural origin of an infix is of course either a prefix or a
suffix which got displaced and moved into the interior af an adjacent
morpheme by simple metathesis. The nasal infix is located before the final
root consonant, therefore it must have moved in from behind which makes it
an old suffix. I take it that a present-stem structure like *li-ne-{k}-,
*{kw}ri-ne-H2- and *dhi-ne'-H1- from *ley{kw}- 'leave', *{kw}reyH2- 'buy',
*dheH1y- 'suck' proceed from old structures in which /n/ was simply added
to the root, i.e.

   *ley{kw}-n-      *{kw}reyH2-n-     *dheH1y-n-,

and that these structures were subsequently adjusted to some more widely
"preferred syllable types" by simple metathesis, which apparently gave

   *leyn{kw}-       *{kw}reynH2-      *dheynH1-.

Then, as generally in IE, stems ending in three consonants were alleviated
by insertion of an /e/ before the last consonant (my own observation);
that gave:

    *leyne{kw}-     *{kw}reyneH2-     *dheyneH1-.

The new vowel took the accent, so that the strong forms acquired a span
/-ne'-/; on that basis, the weak form which had syllabic endings, moved
the accent one slot further, leaving /-n-/. The product was then:

3sg *li-ne'-{kw}-t  *{kw}ri-ne'-H2-t  *dhi-ne'-H1-t
3pl *li-n-{kw}-e'nt *{kw}ri-n-H2-e'nt *dhi-n-H1-e'nt.

> This brings me to a general question. There seem to be two camps about the
> category system of the PIE verb. One believes that the aorist-imperfect
> distinction, to be equated to perfective-imperfective distinction,
> ``always'' existed in PIE and Hittie lost this distinction, while Vedic
> changed things around. The other considers the aspectual distinction to
> postdate the separation of Anatolian.

There are these two camps, yes, and I am in no doubt that camp one is
right. There is no way the specific forms of the aspect stems could have
been formed secondarily in "the rest of IE" left after the exodus of (or
from) the Anatolians. At the very least, all the _forms_ must be assigned
to a protolanguage from which also Anatolian is descended. And what would
the forms be there for, if the functions that go with them only developed
later? We find practically all the IE verbal stems in Anatolian, many only
in a few lexicalized remains; are we to assume they have had totally
enigmatic earlier functions, and that they were later dug up by "the rest
of IE" and given totally new functions there? Clearly the only unforced
interpretation is that their functions in the common protolanguage were
the ones with which they are found where they do survive with a palpable
functional identity.

> There are certain nagging questions about the first thesis: The change in
> Vedic is not explained and how it came about without the prior loss of
> aspect has, AFAIK, not been explained. Those who adhere to this also feel
> the need to explain away as much as possible of root presents. But
> there are enough of them remain in Hittite and Vedic to raise doubts.

I don't follow - what change in Vedic are you talking about? Why would
anyone want to explain away root presents where they are securely
reconstructible? One would do that only to avoid having a language
combining a root-present with a root-aorist, for in that case the two
aspect stems are identical. That is why I am so sceptical about the
authenticity of the Vedic root presents lehmi and dehmi, because for these
verbs we have nasal presents in some other IE languages pointing to the
existence of a root aorist; thus leh- deh- look like displaced aorists.
But not so for eti 'goes' or asti 'is': these are durative verbs, and so
their unmarked form could function as a durative (socalled "present")
stem.

> Sigmatic aorists, as forming perfectives for root presents, present some
> problems too. There are roots that look like older roots extended with an s.
> This suggests that sigmatic aorists came from grammatization of a root
> extention and fits in with the general opinion that sigmatic aorist as a
> >grammatized< formation (as opposed to -s as derivational affix for forming
> new verbs) is late PIE.

If my wild guess analyzing the inchoative present suffix *-sk^e/o- as
**-z-ye/o- (with a special sibilant, whence later IE /s/) is correct,
there is no way this could belong to any period we would call "late". I
don't expect everybody to accept this idea, however (though I don't see
why they shouldn't).

> There is another question that typology suggests: There are basically two
> forms of perfectives (Dahl, ``Tense and Aspect systems; Bybee and Dahl, in
> Studies in Language, 13(1989) pp.51--103; Bybee et al ``The evolution of
> grammar''): The first seems to come from old preterits that become limited
> to perfectives due to the rise of new present/imperfective. The second is,
> as in Russian, from ``bounders'', originally adverbs that denote the
> attainment of a limit. In the first, aorist is unmarked and refers to action
> considered as a whole, even if it is atelic. In the second, the perfective
> does not fit that well with atelic events. The difference is said to be
> evident in Bulgarian which has both the aorist-imperfect distinction (of the
> first type) and perfective-imperfective distinction using prefixes (the
> bounders).

> Now, in PIE, some verbs seem to be like the first system (root aorist vs
> marked presents) while we cannot escape the presence of the second type as
> well (root present with sigmatic aorist). Sometimes Greek situation is used
> to argue that aorist looks older due to the number of irregularities. But in
> Sanskrit, as can be seen from looking at the lists in grammars, root
> presents show the most irregularity while aorists (and class 3, 7
> presents) are regular exemplars of internal sandhi however complicated the
> rules may seem. Again we are faced with the question of who is more archaic
> and who regularized.

Again, I do not think there is any problem in accepting root present as
original for inherently durative verbs, and root aorist as equally
original for inherently punctual verbs.

> One thing needs to be said about a different attempt to explain this, namely
> the traditional (in IE studies) equation durative=atelic=imperfective,
> momentary=perfective. Limiting perfectives to telic/momentary situations is
> decidely the minority option among languages. In Dahl's survey it was found
> only in Slavic, Finnish (based on acc vs partative), Japanese, Hindi,
> Mandarin, Bandjalang and Cebuano. It is even absent in aorist-imperfect
> distinction of Bulgarian. Positing this restriction for PIE requires
> appropriately strong evidence, not just pointing to Russian.

But the IE aorist is not restricted to any special kind of verbs - it is
only _unmarked_ (better: apparently originally unmarked) for inherently
punctual verbs; for other verbs the aorist need a morphological marking,
and the meaning is then some nuance that can be regarded as punctual
("started to -") or it just reports that the action got done (Meillet's
action pure et simple). One important functional point with the aorist,
however, is that it marks a turn of events which creates a new situation,
whereas the "present aspect" stays in the situation already given and
reports another action contributing to that situation. This is seen
remarkably well in the prohibitive use of the prs. vs. aor. injunctive, as
propounded so clearly by Hoffmann.

> The equation durative=telic is equally atypical. In such a language, one
> could not just say ``John walked home'' or ``John ate one piece of bread''.
> Instead one must say something like ``John walked, reached home'', or have
> two verbs for every transitive durative, one used only for past with
> definite objects, one used for past with indefinite/mass objects and for all
> presents. Again, we cannot just assume such a property, especially given
> that this does not seem all that common in contemporary languages.

Still, even verbs generally signifying completed action could form
duratives, indicating e.g. a repetition of the action (give one thing, and
then another) or an as yet unsuccessful attempt (I'm opening the window).

[Jens:]

>> for the functional change is quite small: it only takes the use
>> of the aorist form as an imperfect, then the rest follows by itself.

> The change of syntactic categories does not strike me as a small change. I
> would like to know more about incontrovertable cases of such changes before
> arriving at any conclusions.

We know that this kind of change was small enough in the languages here
concerned to lead to a number of misplaced aspect stems. E.g., the
Armenian aor. eber is an old ipf. Is the jump from "narrative past" to
"recent past" so great? If it is, even great changes happen.

Jens


From larryt at cogs.susx.ac.uk  Wed Jul 14 14:39:07 1999
From: larryt at cogs.susx.ac.uk (Larry Trask)
Date: Wed, 14 Jul 1999 15:39:07 +0100
Subject: accusative and ergative languages
In-Reply-To: <1999Jul10.021517@AN3039.spb.edu>
Message-ID: <WED.14.JUL.1999.153907.0100.>

On Sat, 10 Jul 1999 PIE at AN3039.spb.edu wrote:

[LT]

>> But Dixon also cites examples of ergative constructions which, in his
>> view, have very clearly *not* developed from passives, but from other
>> constructions.  His examples are Hittite and Pari.

> Pardon my ignorance, but i wonder what Hittite constructions can be
> claimed as ergative? Could you clarify Dixon's opinion?

> Does he mean the tendency to avoid using inanimates as 1st actants of
> two-argument predicate and to transform them to animates in -nt-, like
>
>     eshanants inan karapzi,
>     witenants eshar parkunut,    etc?

Dixon appears to be relying entirely on the views of Garrett, as
presented in this article:

A. Garrett (1990), `The origin of NP split ergativity', Language 66:
261-296.

It would appear that Garrett argues for an instrumental origin for the
Hittite construction he (she?) interprets as an ergative.

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk


From obaumann at stud.uni-frankfurt.de  Thu Jul 15 15:51:48 1999
From: obaumann at stud.uni-frankfurt.de (Oliver Baumann)
Date: Thu, 15 Jul 1999 17:51:48 +0200
Subject: FYI: Cimbrish course
Message-ID: <THU.15.JUL.1999.175148.0200.>

Dear members,

We want to inform you that a course in Cimbrish, a very old westgermanic
language spoken in Norther Italy til today, will be held next week on
our mailing list. If there is interrest to submit or to take part,
please inform us with some words:
Cimbrian-List-request at em.uni-frankfurt.de

If there is further interrest in that seriously endangered language,
see:
http://www.diens.de/Zimberland

Gildo Bidese and Oliver Baumann

PS: Sorry for cross posting


From petegray at btinternet.com  Wed Jul 14 18:48:34 1999
From: petegray at btinternet.com (petegray)
Date: Wed, 14 Jul 1999 19:48:34 +0100
Subject: accusative and ergative languages
Message-ID: <WED.14.JUL.1999.194834.0100.>

Pat said:

> What I consider 'internal inflection'  is Arabic yaktubu, kataba, katibun,
> etc. which applies to any and all verbs, old, and those taken new into the
> language.

Like it or not, in the pattern "sing, sang, sung" there is inflection, and
it is internal.   If you mean "applying consistently to all verbs", that's a
different matter.   But "sing, sang, sung" woudl still be internal
inflection, even if it were the only surviving example.

I am quite happy if you want to say that this pattern is no longer
productive in the language.

Peter


From petegray at btinternet.com  Wed Jul 14 19:01:30 1999
From: petegray at btinternet.com (petegray)
Date: Wed, 14 Jul 1999 20:01:30 +0100
Subject: accusative and ergative languages
Message-ID: <WED.14.JUL.1999.200130.0100.>

Leo said:
> Germanic ablaut *never* results from vowel harmony.  You're thinking of
umlaut,

Of course.   Because of the nature of the discussion, I simply didn't bother
to add "but here it is more often called umlaut"

Peter


From proto-language at email.msn.com  Fri Jul 16 03:17:46 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Thu, 15 Jul 1999 22:17:46 -0500
Subject: PIE vs. Proto-World (Proto-Language)
Message-ID: <THU.15.JUL.1999.221746.0500.>

Dear Joat and IEists:

 ----- Original Message -----
From: <JoatSimeon at aol.com>
Sent: Monday, July 12, 1999 11:09 PM

> In a message dated 7/12/99 9:28:59 PM Mountain Daylight Time,
> proto-language at email.msn.com writes:

>> I simply do not understand why some find it difficult to understand that
>> reconstructing the Proto-Language is only primarily different from
>> reconstructing Indo-European in the wider selection of source languages
>> for data.

Joat responded:

> -- Temporal distance.  Loss of information.  Entropy.

> A distinction between 5,000 years of unrecoverable loss and 50,000 years
> of unrecoverable loss.

> Words _vanish_.  A certain percentage of vocabulary just ceases to exist
> in every century.

Pat responds:

Nice theory! But the rate of vocabulary replacement is so variable (not to
mention punctuated equilibrium) that it cannot be used to exclude possible
retention of words for periods longer than 5,000 years.

What is shameful is that when someone does show *regular* similarities, that
perforce exceed that date by far, bogus reasons for dismissing them are
employed: like Larry's favorite onomatopeia, or expressive, or fantastic
loan scenarios.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From kurisuto at unagi.cis.upenn.edu  Fri Jul 16 04:37:43 1999
From: kurisuto at unagi.cis.upenn.edu (Sean Crist)
Date: Fri, 16 Jul 1999 00:37:43 -0400
Subject: Recoverability
In-Reply-To: <Pine.SO4.4.05.9907131322380.11749-100000@rask>
Message-ID: <FRI.16.JUL.1999.003743.0400.>

On Tue, 13 Jul 1999, Jens Elmegaard Rasmussen wrote:

> But that is not what you do in cases where only modern languages are
> available: You do not pick two and ignore the rest, you take all there is.
> In the case of IE, you would take _all_ modern Germanic languages, if need
> be with their dialects; you take all Slavic languages, you don't forget
> about Lithuanian and Latvian or Modern Greek which gives quite a lot;

[...]

> This is a LOT - and
> NOBODY could even thgen be in doubt that this is indeed a family of
> related languages sprung from a common source. True, we would still be in
> doubt or indeed ignorant about many a finer point which is only added by
> languages of old texts - and of course even they do not contain
> everything, so doubt will remain - but the general utline and subgrouping
> of the family would stand firm even on the sole basis of modern languages.

As far as this goes, it is quite right.  However, I don't think the point
was that we ought to actually try to attempt to reconstruct PIE on the
basis of only modern English and Albanian, or otherwise arbitrarily limit
our selection of data.

I think the point was something more like this.  Suppose for the sake of
argument that there is in fact a genetic relationship between
Proto-Indo-European and Proto-Uralic (to take one possible example). Even
if this were true, and given the evidence which we actually do have at
hand, would we expect to be able to _show_ that there is such a relation?

Here, the problem _is_ like trying to reconstruct PIE on the basis of only
modern English and Albanian.  Even tho external evidence might lead us to
guess that there could be a relation between PIE and PU, the cognations
are so obscured by millenia of sound changes, loss of old lexical items,
the noise of new lexical items, etc., that it might well be impossible to
show a genetic relationship on the basis of the available information.

Suppose, for example, that PIE *d corresponds to PU *z (I don't know that
there is a *z in PU; I'm making this up).  Even if there were originally
dozens of words showing this correspondence, it might well be that case
that only one such pair exists between the small subsets of the original
lexicons of PIE and PU which we can reconstruct.  Given the fundamental
assumptions of the Comparative Method, you _can't_ show that the two words
are cognate when you've only got one example of the correspondence.

You need multiple examples of sound correspondences to be able to conduct
the Comparative Method at all; and when the the cognates become as
rarefied as they are this time-depth, the likelihood of having access to
an adequate number of examples to work out the relevant sound changes
becomes proportionately smaller, eventually reaching what for practical
purposes is an impossibility.

  \/ __ __    _\_     --Sean Crist  (kurisuto at unagi.cis.upenn.edu)
 ---  |  |    \ /     http://www.ling.upenn.edu/~kurisuto/
  _| ,| ,|   -----
  _| ,| ,|    [_]
   |  |  |    [_]


From cz at magnet.at  Fri Jul 16 04:42:44 1999
From: cz at magnet.at (Clemens Zeitlhofer)
Date: Fri, 16 Jul 1999 06:42:44 +0200
Subject: Re(2): indoeuropean/hand
Message-ID: <FRI.16.JUL.1999.064244.0200.>

Indo-European at xkl.com writes:
>acus, domus,
>cornus, idus, and tribus (as well as manus) are the ones that come to
>mind.

Don't forget porticus :-)

Just wanted to say hello to everybody (I'm a newbie).

Regards,

c.
********** ********** **********
Clemens J. Zeitlhofer
1096 Wien, Postfach 181
cz at magnet.at

Odi profanas litteras et arceo
********** ********** **********


From X99Lynx at aol.com  Fri Jul 16 06:10:28 1999
From: X99Lynx at aol.com (X99Lynx at aol.com)
Date: Fri, 16 Jul 1999 02:10:28 EDT
Subject: Chronology of the breakup of Common Romance/CELTS
Message-ID: <FRI.16.JUL.1999.021028.EDT.>

I wrote with regard to "Vlakhi", 'Vlachi', etc.:

<<Appearing as "Walh" or "Walah" in OHG, it has interpreted
as meaning "foreigner", sometimes Roman, but in usage it is closely
associated with Celts and regions of Celtic habitation - e.g., "Wales",
"Walloon".

In a message dated 7/13/99 6:16:38 PM, rmccalli at sunmuw1.MUW.Edu replied:

<<You'd have to account for the ending in Vlakh, WLochy, etc.>>

The -kh, -chy endings - as adj. affixes - appear to reflect variants anywhere
the Celt connection was made. E.g., 'Wilisc', 'Wylisc', 'Wielisc' (W. Saxon
7th Cent.) = Celt or Briton. Also, 'Waleis' (Anglo-French)> 'Wallace,' OHG
'walhisc,' 'walesc' = Roman, French, O.E. Chronicle 'wilsc', bryt-'wylsc'
(refering to the language).  By 1100, the forms 'Wylisca cing' and 'Wylsca
biscop' occur among the English. Also, 'Wallych', 'Walsche' and of course
'Welch'.  In MHG, 'walsch'.  Closer to the east, 'Valskr' (ON), (MSw) =
Gauls, Frenchmen.  The various Slavic endings I believe are in line with all
this.  (cf, B-T-W, 'Cornwall', 'Wellfleet', etc.)

The OED calls 'Vlach' a Slavic borrowing from the Germanic above.

I also have this note from the Hungarian scholar who annotated the Gesta
Hungarorum (1975 ed) re the Pannonia re a bad date in that chronicle:  "An
explanation of this erroneous record may have been the Slavic tradition which
held that the Hungarians ousted the Franks - the Volochs in the language of
the Slavs - from the area of the Danube."

Here is another indication that the term "vlach" was originally directed to
the west:  "The Slavic ethnic name of the 10th century, Vlach, was, in its
plural
form: vlasi, borrowed by the Hungarian language with the sense
'Neo-latin, Italian, French', in the form olasz(i); in the period of
Arpád, this was the Hungarian name of the Neo-Latin peoples, thus also
of the French, i.e., the Franks (cf. Latin Frankavilla > Hungarian
Olaszi) and this is also at present the Hungarian name of the Italians."

It is interesting to note that the Franks themselves considered their
original homeland - according to their official historian, Gregory of Tours -
not Germania, but Pannonia.  And when Constantine II founds a military colony
in Moesia at present day Fenehpusta, it is called 'Valcum'.  It will later be
the site of a Frankish fortress destroyed by the Maygars.

I wrote:
<<I seem to recall an suggestion that there may be a tie here to L.
vallum (fortified wall, the earliest meaning of "wall") and refer to the
Celtic or Romano-Celtic oppidum or walled town.>>

This was based I believe on L. 'vallatii', entrenchers, cf 'valla^tio, onis',
an entrenchment, from 'vallo -a^vi, -atum -' (see Cicero, Tacitus and Pliny)
to surround with a rampart.  Nothing I'm sure of, in the least.

<<It has long been proposed that the name Volcae, a Gaulish tribe,
was the source of Anglo-Saxon Wealh, Balkans Vlakh, Polish WLochy, etc.
The point I've generally seen is something to the extent that
supposedly the Volcae were the first Gaulish tribe to fall under Roman rule.>>

The BIG problem with this is that the Romans NEVER call the Celts/Gauls in
general 'Volcae' or anything quite like that.  Caesar does tell us the Volcae
Tectosages invaded Germania, so the contact between this group of Celts and
Germanics was probably direct.  Which may account for why the Romans did not
use the word as a name for other Celts, but the Germans perhaps did.  (B-T-W,
the first time anything like the word 'Volc/-' appears in Latin, it is in
connection with the Etruscans, whose town 'Volci' or 'Vulci' falls to the
Romans in 280 BC.  Velitrae is the town first associated with the Celtic
Volscae.)

<<My guess is that any Celtic element in the Balkan would have been
extremely thin.  The Celts raided the Balkans and stormed through the region
on the way to Galatia but my understanding is that they lived north of the
Balkans
in Galatia and Pannonia.>>

If the Danube is the northern line of the Balkans, then the Celts were right
on it all the way to the Black Sea.  In fact, they seem to be the only ones
on it that founded towns - Gorsium (attested to having been founded by Celtic
Eraviscans), Sungidium (Belgrade) Siscia (at the mouth of the Odra), and many
other sites show that originally Celtic settlers preceded Roman occupation.
Konjic, not far from Sarajevo, decribed as "Illyrian-Celtic" yielded evidence
of centuries of extensive Celtic occupation before the Romans came and some
of these were still present in the 1st Century AD.  Recently I saw a
catalogue with pages of Greek coins found in Thrace (2/1 century BC) that
were described as Celtic imitations.  And around 80 BC, when the Dacians
under Burebista make their move towards Greece and the Black Sea, they are
constantly warring with the Celts.  Even on the north bank of the Danube in
Dacia before the Roman withdrawal, there is still a Celtic presence in the
archaeology.

It seems that in this area, as in France, a significant Celtic cultural
presence seems to disintegrate when the Romans shows up.  And as in France
the obvious conclusion is that the Celts became Romanized on a wholesale
basis.  And one would think therefore there would be some substrate in the
remaining languages in this area.  (Even given the difficulty of finding such
substrates in everything from English and French to Spanish and Bohemian.)

It is odd, isn't it, that Celtic culture left such an 'thin' linguistic trail
across so much territory where its presence was not just attested, but even
dominant?

Regards,
Steve Long


From JoatSimeon at aol.com  Fri Jul 16 16:53:47 1999
From: JoatSimeon at aol.com (JoatSimeon at aol.com)
Date: Fri, 16 Jul 1999 12:53:47 EDT
Subject: PIE vs. Proto-World (Proto-Language)
Message-ID: <FRI.16.JUL.1999.125347.EDT.>

>ECOLING at aol.com writes:

>Of course it is going to get RELATIVELY more difficult to reconstruct to
>greater time depths.

-- true.  And 50,000 years is a whole different ball game compared to 5000
years.

ALL of that extra time-distance is pre-literate, which means no records of
what was going on.  Our picture of PIE would be very different, and much less
detailed, if we had no written records of extinct languages.

Hell, the original insight that the IE languages were derived from a common
ancestor was made by comparting Sanskrit with Classical Greek and Latin --
all extinct languages!

The link between Sanskrit and Latin or Classical Greek is so close that any
layman can see it; whole phrases are nearly identical.

This is, to put it mildly, not the case between, say, English and Bengali.

>If there are sound-symbolic ideal forms for lexemes having certain meanings

-- with minor exceptions (such as "kuku"), words _are_ arbitrary sound
assemblages.


From larryt at cogs.susx.ac.uk  Fri Jul 16 15:35:10 1999
From: larryt at cogs.susx.ac.uk (Larry Trask)
Date: Fri, 16 Jul 1999 16:35:10 +0100
Subject: accusative and ergative languages
In-Reply-To: <001f01becdcd$786928e0$309ffad0@patrickcryan>
Message-ID: <FRI.16.JUL.1999.163510.0100.>

Pat Ryran writes:

> Now, when Larry recently quoted Dixon about the nature of the
> ergative, he conveniently neglected to mention that Dixon
> acknowledged that there were currently practising linguists --- not
> amateur linguists like myself --- still defending the passive
> interpretation of ergative constructions.

No; this is not so.  Read p. 189 of Dixon's 1994 book.

> I asked Larry where he had "shredded" this interpretation, and to my
> knowledge, got no answer. If I missed the "shredding", perhaps you
> will be kind enough to rehearse his performance for us. I have seen
> nothing by Larry's vehemence and your allegiance to support the idea
> that the ergative should not be interpreted as a passive.

Mr. Ryan, I'm afraid I lack the time to reply in detail to all, or even
most, of the 37 or so postings from you that greet me each morning.

But briefly: if a transitive sentence in an ergative language were
"really" a passive, then its absolutive NP would be the subject and
would exhibit subject properties.  But this not the case in most of the
ergative languages I have heard about.  In spite of the ergative
case-marking and/or verbal agreement, it is usually the *ergative* NP
which exhibits subject properties, not the absolutive NP.

Take any ergative language you like.  Assume that the subject of an
intransitive sentence is the sole (absolutive) NP in it.  Examine the
syntactic properties of that NP, and tabulate them.

Then look at transitive sentences.  Tabulate the syntactic properties of
the absolutive and ergative NPs, and compare these with the preceding.

In the great majority of ergative languages, it is the ergative NP which
shares the subject properties of the intransitive subject.

A few examples of typical subject properties:

The subject controls reflexive and reciprocal NPs.

The subject cannot itself be reflexive or reciprocal.

The subject controls the empty NP in an empty-NP complement.

The empty NP in an empty-NP complement is itself a subject.

A subject can be coordinated with another subject.

And so on.  There are also various language-specific tests.  For
example, in some varieties of Basque, the object of a gerund (but not
the subject of a gerund) goes, exceptionally, into the genitive case.
Neither the absolutive subject of an intransitive gerund nor the
ergative subject of a transitive gerund can be genitivized, but the
absolutive object of a transitive gerund can be.

One more test from Basque.  In Basque, the subject of an intransitive
sentence cannot be reflexive or reciprocal.  So, Basques can say,
literally, `Susie and Mike [ABSOLUTIVE] were talking to each other', but
they cannot say *`Each other [ABSOLUTIVE] was talking to Susie and
Mike'.

Now, in a transitive sentence, Basques can say `Susie and Mike
[ERGATIVE] slapped each other [ABSOLUTIVE]', but they cannot say *`Each
other [ERGATIVE] slapped Susie and Mike [ABSOLUTIVE]' -- or, in terms of
the discredited passive theory, they cannot say *`Susie and Mike
[ABSOLUTIVE] were slapped by each other [ERGATIVE]'.

In an intransitive sentence, the absolutive subject cannot be reflexive
or reciprocal.  In a transitive sentence, the ergative subject cannot be
reflexive or reciprocal, but the absolutive object *can* be.

All tests for subjecthood in Basque give the same result: it is the
ergative NP in a transitive sentence which shares the subject properties
of the absolutive NP in an intransitive sentence, and therefore it is
the ergative NP, and not the absolutive NP, which is the subject of a
transitive sentence.

There are unusual languages, of course.  In Dyirbal, and in
Nass-Gitksan, the absolutive NP in a transitive sentence shows at least
some subject properties, while the ergative NP does not show the same
properties -- though it does show other subject properties.  This is the
phenomenon we call `syntactic ergativity'.  But I know of no language
which is syntactically ergative without exception, and I know of few
languages which are syntactically ergative at all.

The "passive" view of ergative languages in general is indefensible.

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk


From Georg at home.ivm.de  Fri Jul 16 09:28:37 1999
From: Georg at home.ivm.de (Ralf-Stefan Georg)
Date: Fri, 16 Jul 1999 11:28:37 +0200
Subject: PIE vs. Proto-World (Proto-Language)
In-Reply-To: <6fa43e91.24bc160a@aol.com>
Message-ID: <FRI.16.JUL.1999.112837.0200.>

[ Moderator's note:
  The following quoted material is from a message by <JoatSimeon at aol.com>.
  Just keeping things straight.
  --rma ]

>In a message dated 7/12/99 9:28:59 PM Mountain Daylight Time,
>proto-language at email.msn.com writes:

>>I simply do not understand why some find it difficult to understand that
>>reconstructing the Proto-Language is only primarily different from
>>reconstructing Indo-European in the wider selection of source languages for
>>data.

>-- Temporal distance.  Loss of information.  Entropy.

>A distinction between 5,000 years of unrecoverable loss and 50,000 years of
>unrecoverable loss.

>Words _vanish_.  A certain percentage of vocabulary just ceases to exist in
>every century.

And, if I may insert this much to Pat's chagrin, the notion that human
language is a monogenetic phenomenon is aprioristic ideology. Even the
notion that every known language, as Basque, Burushaski or whatnot, has to
be related to some other language is ideology. "Having consonants" may be a
universal feature of human language, "Being related to some other lg."
simply is not.

Stefan Georg
Heerstrasse 7
D-53111 Bonn
FRG
+49-228-69-13-32


From JoatSimeon at aol.com  Fri Jul 16 16:43:23 1999
From: JoatSimeon at aol.com (JoatSimeon at aol.com)
Date: Fri, 16 Jul 1999 12:43:23 EDT
Subject: accusative and ergative languages
Message-ID: <FRI.16.JUL.1999.124323.EDT.>

proto-language at email.msn.com writes:

> I also am certain that if we had reason to know that Albanian and English
>were descendants of a common language, we would eventually discover a kind
>of IE that would be inaccurate in some ways but essentially write.

-- no we wouldn't.  Sigh.

As time goes on, any given language gives us less and less information about
previous forms.  Information is _lost_.

Eg., no amount of analysis of Germanic will give you a PIE word for "bear"
because that word dropped out of that family.  All you'll get is variations
on "the brown one".


From larryt at cogs.susx.ac.uk  Fri Jul 16 15:06:31 1999
From: larryt at cogs.susx.ac.uk (Larry Trask)
Date: Fri, 16 Jul 1999 16:06:31 +0100
Subject: PIE vs. Proto-World (Proto-Language)
In-Reply-To: <6fa43e91.24bc160a@aol.com>
Message-ID: <FRI.16.JUL.1999.160631.0100.>

Pat Ryan writes:

> I simply do not understand why some find it difficult to understand
> that reconstructing the Proto-Language is only primarily different
> from reconstructing Indo-European in the wider selection of source
> languages for data.

Others have already made substantial points in reply, but I'd like to
add one more.

Before reconstruction can be attempted, there must be a persuasive
*prima facie* case that the languages under consideration are genuinely
related.  Our predecessors did not begin reconstructing PIE until they
had persuaded themselves that the evidence for a genetic link among the
IE languages was too massive to be ignored.

But you can't just pick some languages that catch your eye and then
"reconstruct" a common ancestor for them.  That is, you can't pick, say,
Zulu, Sumerian and Korean and "reconstruct" Proto-ZSK.  You must first
make a good case that a common ancestor for these languages is
preferable to any other scenario.

And the same goes for "Proto-World", or whatever.  Until a persuasive
case has been made that *all* of the world's 6000 or so known languages
are genuinely related, there is no point in attempting a
"reconstruction" of "Proto-World".  The result of such a rash attempt
can be no more than legerdemain.  Of course you can show that this, that
and the other *might* have a common ancestor, but you can do this in
countless entirely different ways, none of them superior to any other,
and you cannot show that these things really *are* related.

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk


From DFOKeefe at aol.com  Fri Jul 16 09:39:54 1999
From: DFOKeefe at aol.com (DFOKeefe at aol.com)
Date: Fri, 16 Jul 1999 05:39:54 EDT
Subject: PIE vs. Proto-World (Proto-Language)
Message-ID: <FRI.16.JUL.1999.053954.EDT.>

Good Morning I-E-ists,
          I was under the impression that I.E. Roots are pretty much
time-invariant, They can't be created, change, or get lost.  Words derived
from I.E. roots frequently do change, as do languages.  As a matter fact,
entire languages and language families shift in organized manner particular
to each.  But I.E. roots do not. Thus, I.E. roots don't decay, degenerate, or
go out of existence.  In the time span of recorded human history, entropy
does seem to not apply to I.E. roots.  Maybe to words, though.

Best regards,
David O'Keefe
Houston, Texas

[ Moderator's response:
  Your impression is mistaken.  In the reconstructed morphology of PIE, we see
  a theoretical construct which we call the root; this is the morpheme which
  carries the base meaning assigned to the word(s) in which it is found.  It is
  no more time-invariant that the reconstructed stem formants or endings in the
  same words.

  A root can only survive in the history of a language as long as a word which
  contains it.  There is no reflex in modern English, for example, of the root
  *yebh- "fuck", although it is still alive and well in modern Russian (to the
  extent that the name of one IBM mainframe utility is a great joke).  There
  very likely were roots in IE of which we have no knowledge at all because
  they have not survived into *any* extant language in a manner which allows us
  to assume their existence.

  --rma ]


From X99Lynx at aol.com  Fri Jul 16 16:43:18 1999
From: X99Lynx at aol.com (X99Lynx at aol.com)
Date: Fri, 16 Jul 1999 12:43:18 EDT
Subject: The Rumanian Question (was: Chronology of the breakup of Common
 Romance)
Message-ID: <FRI.16.JUL.1999.124318.EDT.>

In a message dated 7/13/99 6:16:38 PM, rmccalli at sunmuw1.MUW.Edu wrote:

<<Graham Mallinson, "Rumanian," 391-419. Martin Harris and Nigel
Vincent, eds. The Romance Languages. London: Croom Helm, 1988.

	There is very little substrate vocabulary in Rumanian, and its
origin is unknown. Since the grouping of Dacian, Thracian and Illyrian
within IE is uncertain, it is questionable whether Albanian and Rumanian
shared a substrate language, although they do have cognates:>>

The question of the origins of the Rumanian language have involved historical
linguists and other historians in some pretty sensitive political issues in
southeast Europe.  The matter of the breakup of Hungary and Romanian national
identity have been debated on historical grounds for some time now.  (Whether
such things should really be relevant is another matter.)

One controversy is the question of "Daco-Roman Continuity."  Essentially,
this revolves around whether Romance was always spoken in Romania and
represents a continuation of what began as the Roman colonization of the
province Dacia in the first centuries of the current era.  At first glance,
this would seem logical enough and it had been the official line of most
historians and linguists for a long time.  But that conclusion has been put
into question from time to time and lately some substantial arguments have
been made to the contrary.

In 1996, Andre duNay published "The Origins of the Rumanians: the Early
History of the Rumanian Language." (Matthias Corvinus, Toronto).  It should
be of special interest to the members of this list for its methodology which
attempts to synchronize an apparently exhaustive mass of past linguistic
research with historical data.

Of particular interest is the way duNay goes beyond mere lexical comparisons
(such as those reproduced by rmccalli from Mallinson and Rosetti in the prior
post) and considers the chronology of syntactical developments, innovations
and changes in Latin, Southern Slavic, Greek, Albanian and Romanian - as well
as archaeology and primary historical records - to reach conclusions about
when linguistic events actually occurred.

duNay addresses a long list of linguistic theories regarding Rumanian
language and does a fairly compelling job of dissembling the Daco-Roman
continuity theory.  Particularly compelling is the linguistic and historical
evidence that Rumanian developed in close proximity to Albanian in the
southern Balkans.  The syntactical and other characteristics by which duNays
identifies Rumanian as a "southern Balkan" language joined with Bulgarian,
Albanian and Greek should again be of particular interest to members of this
list from the point of view of methodology.

The book BTW is reproduced in its entirety on the web, on the Hungarian web
site (for reasons that become more apparent as one gets into this.)  I will
get the URL and post it if there's interest.

I don't know if one must conclude that the Albanian or Proto-Albanian
substrate in Rumanian settles the issue of origins.  I do know that I
appreciate the historicity of this work very much - particularly because it
shows how silly statements like this one are:

<<A. Rosetti: "The Thracians, known as Gètes by Greek writers and Dacians by
Latin writers..." >> (This is the type of bad historical summary that must
pollute any understanding of the situation for linguists relying on it.)

There are forwards in the book by Robert A. Hall at Cornell and Adam Makke
that are also interesting, including the following statement:  "Lack of
knowledge of the Slavic languages,...hindered the Rumanian historians from
expressing the problem clearly and recognizing its significance in the study
of the beginnings of the Rumanian people."

This may explain a bit of the 'basic word' substrate mentioned in the prior
post.  E.g.,:

<<Albanian			Romanian
buzë "lip, edge"		buzâ "lip, edge">>

See 'pusa' (Czech), 'buzia' (Pol.) which I believe follow the variances
expected between Romanian and Slavic found in other -a^ > -ia forms.

In fact, the following makes me think that maybe we are dealing with a
transfer of technology here, rather than proof of common ancestry:

<<The total number of words of significance for the everyday life of
shepherds among these  89 words common to Rumanian and Albanian is at least
60  - 66 % of all.  Of the rest, most denote basic human notions and
conditions, such as  parts of the human body and family relations. >>

Hope this is interesting or helpful.

Regards,
S. Long


From edsel at glo.be  Fri Jul 16 10:04:21 1999
From: edsel at glo.be (Eduard Selleslagh)
Date: Fri, 16 Jul 1999 12:04:21 +0200
Subject: indoeuropean/hand
Message-ID: <FRI.16.JUL.1999.120421.0200.>

[ moderator re-formatted ]

-----Original Message-----
From: Steven A. Gustafson <stevegus at aye.net>
Date: Friday, July 16, 1999 6:07 AM

[Snip]

>There are a fair number of Latin -u stems that are feminine, especially
>considering how few -u stems there are to begin with: acus, domus,
>cornus, idus, and tribus (as well as manus) are the ones that come to
>mind.

>Steven A. Gustafson

[Ed]

Actually it's 'cornu' in the nominative.

All tree names, also those with -u stems like 'quercus' (oak) (and -o stems
like po:pulus) are feminine.  Could there be some relationship here?  I mean
'extremities, branches...'.  That would also account for manus, acus, cornu and
tribus being feminine, but not for domus nor idus.  'Foot, pes...' are probably
not viewed as extremities, but as 'base' to stand on.

Ed.


From JoatSimeon at aol.com  Fri Jul 16 17:42:45 1999
From: JoatSimeon at aol.com (JoatSimeon at aol.com)
Date: Fri, 16 Jul 1999 13:42:45 EDT
Subject: Momentary-Durative
Message-ID: <FRI.16.JUL.1999.134245.EDT.>

>petegray at btinternet.com writes:

<< The Armenian-Greek- I-I sprachbund has certain known
>innovations, such as vocabulary and the augment, and we should not be miseld
>into accepting as original what may be another innovation, namely the
>development of new distinctions based on an existing variety of forms. >>

-- very good point.

A great pity the Linear B sources aren't more extensive.


From proto-language at email.msn.com  Fri Jul 16 13:25:32 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Fri, 16 Jul 1999 08:25:32 -0500
Subject: accusative and ergative languages
Message-ID: <FRI.16.JUL.1999.082532.0500.>

Dear Ralf-Stefan and IEists:

 ----- Original Message -----
From: Ralf-Stefan Georg <Georg at home.ivm.de>
Sent: Tuesday, July 13, 1999 5:22 PM

>> Pat:

>> Yes, but if a new verb were formed, e.g. webben (from English 'web'), there
>> is no possibility that it would be conjugated webben, wabb, gewobben.

>> What I consider 'internal inflection' is Arabic yaktubu, kataba, katibun,
>> etc. which applies to any and all verbs, old, and those taken new into the
>> language.

R-S:

> The pattern lost productivity in German. So what ?
> Also, I hardly understand your claim, which basically amounts to refusing
> to speak of technique X in language L, unless it isn't present in each and
> every item in which it theoretically could be present. So, you would refuse
> to speak of, say, vowel harmony in, say, Turkish, because some borrowed
> (and some native) words fail to obey this rule ? ("What I call VH is ...,
> which applies to any and all ... Period. There is an exception, or the
> phenomenon is gradually losing ground, therefore it doesn't extst at all."
> ?????).

Pat:

The distinction I am trying to make, and you are certainly not required to
make it also, is that the pattern "e-a-o" *never* was really "productive" in
German although one might term it "resultative". There was never a time
during which *any* verb in [e] followed this "pattern", or even *any* verb
in /eR/ followed it (weNden, waNdte, gewaNdt).

 Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From proto-language at email.msn.com  Fri Jul 16 13:05:19 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Fri, 16 Jul 1999 08:05:19 -0500
Subject: PIE vs. Proto-World (Proto-Language)
Message-ID: <FRI.16.JUL.1999.080519.0500.>

Dear Lloyd and IEists:

 ----- Original Message -----
From: <ECOLING at aol.com>
Sent: Tuesday, July 13, 1999 10:51 AM

> Patrick Ryan wrote:

>> I simply do not understand why some find it difficult to understand that
>> reconstructing the Proto-Language is only primarily different from
>> reconstructing Indo-European in the wider selection of source languages
>> for
>> data. Of course, secondarily I have labored to reconstruct the underlying
>> monosyllables by analysis of attested compounds.

Lloyd Anderson wrote:

> I strongly support work on earlier language families not part of the
> standard doctrine today.
> Against those who say (very nearly) that everything which
> can be discovered has already been discovered.
> That is of course a charicature, but with some truth to it.
> (Or they say this while specifying "using the comparative method",
> and defining it circularly to mean only the existing tools,
> and only those ways of using those tools,
> which are well known today.)

> However, the above statement by Patrick Ryan I find highly surprising.

> Of course it is going to get RELATIVELY more difficult
> to reconstruct to greater time depths.
> The mistake of those who reject all Nostratic and similar work
> is in drawing a sharp fixed line,
> saying that short of that time depth it can be done,
> beyond that time depth it cannot be done.

> The mistake of Patrick Ryan in the quotation above is to neglect
> that it does get substantially more difficult as the time depths increase,
> or in particular language families, because of the specific nature
> of the sound-changes and grammatical changes which occurred,
> or etc.

Pat responds:

First, let me say that I agree substantially with everything you have
written in this posting. But I really think the "difficult"  part is
correctly factoring in what may be sound-symbolic influences.

Lloyd Anderson continued:

> There is no sharp break.
> There is a gradually increasing difficulty with time depth and
>     with depth of changes (the two correlated but not the same concept).
> There is ample room for expanding the tools used,
> and for empirical studies of what sorts of changes each tool is
> capable of penetrating beyond, and to what degree.

> To give merely one example, one factor, sound symbolsim:

> If there are sound-symbolic ideal forms for lexemes having
> certain meanings,
> then the more arbitrary ones (farther from those sound-symbolic
> ideal forms)
> have greater value as evidence for historical connections
> of specific languages or language families,
> simply because they are less likely to have resulted from
> later pressure towards the sound-symbolic ideal.

> This is of course a terribly difficult circularity,
> because it means that look-alikes which are more widely attested

Pat interjects:

Here I think you are dangerously introducing the mistaken terminology of the
opposing argument. I do *not* look for "look-alikes"; I only am interested
in "cognates" the phonological forms of which can be supported through
multiple comparisons. For example, one set of interesting IE and Sumerian
"cognates" shows IE *-wey- = Sumerian -g{~}-; interestingly, this
development *is* found in *some* IE languages, like nearby Armenian.

Lloyd Anderson continued:

> for a given meaning or set of related meanings
> may be EITHER relics of an earlier historical unity
> (whose changes were perhaps ALSO retarded by sound-symbolic forces),
> OR the results of pressure towards some sound-symbolic ideal forms,
> from diverse original and unrelated forms.
> Much more subtle and difficult reasoning is therefore needed to
> establish what are results of sound-symbolism and what are results
> of historical common origins.
> EVEN when we have a suprisingly widespread statistical sound-meaning
> correlation.

> The usual procedure is also circular.
> Simply taking a sample of purportedly unrelated languages
> and attempting to determine how many look-alikes word lists contain
> is a bit naive,
> because the languages may not be totally unrelated,
> because chance resemblances may be more common in certain
> meaning or sound ranges,
> because the biases of different types of sound-system structures
> are not yet well handled,
> and for many other reasons.

Pat concludes:

Lloyd, I think the real hinge of this question is how one defines "loss";
and, after calculating "losses", what kind of percentages one arrives at for
"vocabulary loss". Having worked extensively with language-family
comparisons, my statistically unsupported "guess" is that there is amazingly
little vocabulary loss if one allows reasonably semantically-shifted pairs
to be counted as "retained".

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From Georg at home.ivm.de  Fri Jul 16 21:54:08 1999
From: Georg at home.ivm.de (Ralf-Stefan Georg)
Date: Fri, 16 Jul 1999 23:54:08 +0200
Subject: accusative and ergative languages
In-Reply-To: <001f01becdcd$786928e0$309ffad0@patrickcryan>
Message-ID: <FRI.16.JUL.1999.235408.0200.>

>All very well and good. Having explained next to nothing, perhaps you will
>now tell us how precisely *you* were using "passive in nature" and what in
>God's name it is supposed to mean to you, to those you may be quoting, or
>anyone else including me.

It means nothing to me, it was sloppy terminology and has nothing
whatsoever to do with the things we are talking about. I think I made that
clear. It was my incompetent rephrasing of "ergative constructions are, or
invariably come from, passive constructions", a position not endorsed by
me, but apparently by you.

>Pat responds:

>Yes, you are absolutely correct. This is a familar pattern indeed. When
>challenged for an argument, you sidestep the issues of the question by
>conveying that someone you like (here, Larry Trask) "has spoken the definite
>word about this here". If Larry could speak the definite word about
>everything, then this list would be a waste of time. We could just subscribe
>to his newsletter for the latest ex cathedra rulings on all our troublesome
>questions.

Sorry, I've not implied that Larry Trask is the pontifex maximus
ergativitatis, I only stressed that he, in my humble opinion, laid the
issue of "ergative is passive" to rest, with arguments sufficient enough to
show this, and nothing more. That my own position on ergativity happens to
coincide with this and, admittedly, informed my reading of these postings,
is of course instrumental in this. Not who says something matters, but how
well it is argued; believe me or not, I'd even take it from you, if you
only ever said such a thing.

>Now, when Larry recently quoted Dixon about the nature of the ergative, he
>conveniently neglected to mention that Dixon acknowledged that there were
>currently practising linguists --- not amateur linguists like myself ---
>still defending the passive interpretation of ergative constructions.

This is uninteresting. I know practising linguists I wouldn't buy a used
car from. From some I wouldn't buy a *new* car. Who is it and what are
their arguments ? Don't hide behind an anonymous bunch of practising
linguists.

>I asked Larry where he had "shredded" this interpretation, and to my
>knowledge, got no answer. If I missed the "shredding", perhaps you will be
>kind enough to rehearse his performance for us. I have seen nothing by
>Larry's vehemence and your allegiance to support the idea that the ergative
>should not be interpreted as a passive.

How do you live with the fact that some "ergative languages" have
independent passives ?

>That Ralf-Stefan is incapable of defining the term he introduced: "passive
>in nature". By the way, being a native speaker of English, I can assure you
>that in my dialect, "passive in nature" is *not* bad English.

Thanks a bundle for sparing me the one verdict I was really fearing ...

>And to answer your -- I hope not purposeful -- distortion of what I wrote,
>let me say explicitly that I did not assert "there are ergative languages
>without (any) splits". I asserted that Thomsen did not, at least in her
>grammar, identify splits in Sumerian, which you seemed to think she had. I
>then invited you to identify them in Sumerian if you *could*.

I did. I did, out of my die-hard habit, address every single point of your
objections to it, and I addressed them satisfactorily. If not, for heaven's
sake, we will have to go over it again, but I doubt that this time we will
have too many happy faces reading along ...

And, as far as I remember, this whole brouhaha started with me asserting
that all known "ergative languages" have some split, followed by you
prompting we to show such a thing in Sumerian. I'm not a mind-reader, but
this could be interpreted as a challenge aiming at the gist of my
assertion, n'est-ce pas ? I managed to show those splits, although the
discussion got a bit swamped under some hassle over maru:- and
HamTu-conjugations, but I managed to do it. At least I got you to accept,
late in coming, though, that my initial assertian still stands up.

>You might review your own procedures for employing quotation marks. Your use
>of them on "there are ergative languages without any splits" strongly and
>falsely implies that I wrote this in the context of a judgment on the
>question.

I accept the reproach of not being a mind-reader ...

St.G.

Stefan Georg
Heerstrasse 7
D-53111 Bonn
FRG
+49-228-69-13-32


From mcv at wxs.nl  Fri Jul 16 22:58:07 1999
From: mcv at wxs.nl (Miguel Carrasquer Vidal)
Date: Fri, 16 Jul 1999 22:58:07 GMT
Subject: Chronology of the breakup of Common Romance [long]
In-Reply-To: <007101becdef$66b57b40$6701703e@edsel>
Message-ID: <FRI.16.JUL.1999.225807.GMT.>

"Eduard Selleslagh" <edsel at glo.be> wrote:

>Note that the form with -ch is East-European, while the 'Wal-' form is
>Western, mainly Germanic and in loanwords of Germanic origin, e.g. in Latin
>languages (Walen-Wallons, Wallis-Valais, Wales...).

That's merely because the Slavic languages have retained the
"(c)h" in the Germanic loanword which was dropped in the modern
Germanic languages (OE wealh, OHG walh).

=======================
Miguel Carrasquer Vidal
mcv at wxs.nl


From jer at cphling.dk  Fri Jul 16 22:33:52 1999
From: jer at cphling.dk (Jens Elmegaard Rasmussen)
Date: Sat, 17 Jul 1999 00:33:52 +0200
Subject: Ergative & Basque
In-Reply-To: <Pine.GSO.4.03.9907131058110.17292-100000@rsunx.crn.cogs.susx.ac.uk>
Message-ID: <SAT.17.JUL.1999.003352.0200.>

On Tue, 13 Jul 1999, Larry Trask wrote:

> On Fri, 9 Jul 1999, Jens Elmegaard Rasmussen wrote:

>> [...] Can "Mutila jo zuen" not
>> *come from* something which *originally* meant, not 'the boy was
>> hit' pure and simple, but specifically 'the boy was hit by him
>> (e.g., by the one we're talking about)'? [...]

> [LT:]
> Well, this suggestion is possible, I suppose, but I know of no evidence
> to support it, while there is some evidence against it.  Let's look at
> the structure of the Basque sentence.

[JER:]
How can it be possible, if there is evidence against it? This is
getting intriguing now.

[LT:]
> Here <mutil> is `boy', and <-a> is the article.  This NP, being a direct
> object, stands in the absolutive case, which has case-suffix zero.

> The item <jo> is the perfective participle of the verb `hit'.

> And <zuen> is the auxiliary verb-form.  Now, third person is usually
> marked by zero in Basque verbs.  So, the usual absolutive agreement
> slot, which is the first slot in the auxiliary, cannot be filled by an
> agreement marker, and it is filled instead by <z->, a redundant marker
> of past tense.  The next element, <-u->, is a reduced form of the verbal
> root <-du->, from *<edun> `have'; this is the usual transitive
> auxiliary.  Again, the ergative agreement slot, which follows the verbal
> root, is empty, because the ergative (subject) NP is third-person.
> Finally, <-en> marks past tense.

> The whole thing is thus this:

> 	boy-the-Abs hit-Perf Past-have-0-Past

> Or, roughly, `(She/he/it) had hit the boy.'

> But it translates English `S/he hit the boy' (before today).  This is a
> periphrastic form comparable to the ones we find in Romance and
> Germanic.  Nobody knows if such periphrastic forms are calqued on
> Romance or are an independent development in Basque.

> Now, I can see little scope here either for an original passive
> interpretation of the form or for any way of reading `by him' into the
> auxiliary form.

[JER:]
If there is no passive expressed as belonging to the agent in this, what
then is the "have" verb doing here? Can't the underlying construction be
analyzed in a sensible way at all? A silly question from a complete
outsider: is the translation "have" internally motivated? Can it express
the "having" of anything other than a "participled" object?

>> [...] From the primitive and casual
>> books at my disposal I do see that "zuen" and "zuten" mean 'he had
>> him' and 'they had him' resp. I also believe I see that such
>> auxiliaries are combined with a particularly short form of the
>> participles, referred to by Schuchardt as the root of the
>> participle;

> [LT:]
> It is simply the perfective participle of the verb.  Most participles in
> Basque carry an overt suffix proclaiming them as such, but <jo> happens
> to be one of the exceptions: it has no participial suffix, though I
> suspect that it once did, and that the suffix has been lost by a
> combination of phonological change and analogical readjustment.

> [JER:]
>> and "jo" is 'stick; beat' in its shortest form, says my little
>> dictionary;

> [LT:]
> Actually, `hit', `strike', `beat' -- not `stick', which is a noun.

[JER:] In fact, I meant a different mistake, viz. the verb 'to stick,
adhere': I trusted the Spanish dictionary gloss "pegar" to express that
the Basque verb was as semantically broad as the Spanish which, looking in
the Sp.-B. part, I now see it isn't. Let's forget that.

>> and "mutil-a" is 'boy' with the article "-a", but without case or
>> number marking.

> [LT:]
> Not quite.  The form <mutila> is marked as absolutive by its zero
> suffix, and as singular by its singular article <-a>.  The plural
> article is <-ak>, and `the boys' is <mutilak>, in the absolutive.

[JER:] That's what I meant.

>> Therefore my persistent question: Why can't "mutila jo zuen" and
>> "mutila jo zuten" reflect a construction that was earlier meant to
>> express 'the boy, he had him hit', 'the boy, they had him hit'?

> [LT:]
> Well, I can't rule that out, but I can't see any evidence to support it.
> Note in particular that <mutila> exhibits *no* subject properties in
> modern Basque, or in Basque of the historical period.  If it ever was a
> subject, as this proposal requires, the reanalysis must have been
> carried to completion a long time ago.

[JER:] What do you mean mutila has no subject properties - is "the boy" as
the subject of an intransitive verb not precisely mutila? I'm not afraid
of "a long time ago"; I guess it's what we're trying to reach in diachrony
all the time.

[LT:]
> Note also that intransitive verbs are likewise conjugated
> periphrastically but with the intransitive auxiliary <izan> `be'.

> [JER:]
>> Schuchardt also gives "zen" to mean 'he was', so that if you gloss
>> "mutila jo zen" as 'the boy was hit', it seems there is quite a bit
>> of agreement that the verbal root is a participle by itself.

> [LT:]
> The lexical verb stands in the form of its perfective participle in all
> periphrastic past-tense forms, and also in all periphrastic perfects.
> The perfect form corresponding to <Mutila jo zuen> is <Mutila jo du>,
> which differs only in that the auxiliary is now present-tense.  This
> form translates both English `He has hit him' and English `He hit him'
> (earlier today) -- much as in European Spanish.

[JER:] This would fall into place if "have" itself contains "be": Is there
any possibility that the constructions "mutila jo zen" and "mutila jo du"
were originally designed to mean 'the-boy - beaten - he was' = 'the boy
was beaten', and 'the-boy - beaten - for/of-him-he-is' = 'the-boy beaten
he-has-him' = 'he has beaten the boy'/'he beat the boy'? Is there any
solid knowledge excluding such a prehistory?

>> I do not see in what way this makes the *diachronic* interpretation
>> of "mutila jo zuen" any different from the Hindi preterites that are
>> based on Sanskrit constructions of the type "A-Nominative +
>> B-Genitive + PPP/nom." meaning earlier "A was (verb)-ed by B", but
>> now simply "B (verb)-ed A." Where am I wrong?

> [LT:]
> Well, in the Indic case, we have several thousand years of texts to
> consult, so that we can get an idea how the ergative construction arose.
> With Basque, we are not so lucky.

[JER:] But the elements mean the same, one by one, and the semantic sum
total is the same, right? Does that not count for anything?

[LT:]
> Note also that, in Hindi, as in Indic generally (I think), the ergative
> occurs only in the past tense, as is common with ergatives that have
> arisen from perfective or passive constructions.  In Basque, however,
> the ergative construction is used in all circumstances, without
> exception.

[JER:] Looks like a very small difference to me. Can't languages stop at
different points in a process of generalizing a favoured structure? Who
knows that Post-Modern Indic won't introduce the ergative model in the
present tense some day?
   First and last, is translation of the "auxiliary 'have'" demanded by
the language itself? And even if it is, can "be" be contained in it?

Jens


From kurisuto at unagi.cis.upenn.edu  Fri Jul 16 20:15:03 1999
From: kurisuto at unagi.cis.upenn.edu (Sean Crist)
Date: Fri, 16 Jul 1999 16:15:03 -0400
Subject: `cognate'
In-Reply-To: <Pine.GSO.4.03.9907141119250.17570-100000@rsunx.crn.cogs.susx.ac.uk>
Message-ID: <FRI.16.JUL.1999.161503.0400.>

On Wed, 14 Jul 1999, Larry Trask wrote:

>> I'm sure it says PIE *<p at ter> in your manuscript (+ diacritics on the
>> <e>).

> Yes, it does, and my posting should have read `p at ter', but I slipped up.

At least according to Don Ringe, this word should be reconstructed
*pH2te'r.  I should ask him what the arguments are in favor of that
reconstruction, since I don't know myself.

  \/ __ __    _\_     --Sean Crist  (kurisuto at unagi.cis.upenn.edu)
 ---  |  |    \ /     http://www.ling.upenn.edu/~kurisuto/
  _| ,| ,|   -----
  _| ,| ,|    [_]
   |  |  |    [_]


From Georg at home.ivm.de  Fri Jul 16 09:32:01 1999
From: Georg at home.ivm.de (Ralf-Stefan Georg)
Date: Fri, 16 Jul 1999 11:32:01 +0200
Subject: accusative and ergative languages
In-Reply-To: <001001beccef$8ce328e0$40d3fed0@patrickcryan>
Message-ID: <FRI.16.JUL.1999.113201.0200.>

>> Try reconstructing PIE using _nothing_ but contemporary Albanian and
>> English.  You'd be hard-put to prove that such a language even existed.

>It is thuis type of illogic which stands in the way of real progress in
>historical linguistics.

>Larry Trask and others have done an excellent job, IMHO, of reconstructing
>the language from which present Basque derives --- from only one language.

That's because Basque has been for long known to be related to Basque. This
is called internal reconstruction.

>I also am certain that if we had reason to know that Albanian and English
>were descendants of a common language, we would eventually discover a kind
>of IE that would be inaccurate in some ways but essentially write.

The trick is about having reasons to know this, without resorting to some
ideology which tells you before, i.e. without looking at any data.

>Naturally, the more languages we added to the equation, the better
>reconstruction of IE we would achieve.

>Same principle.

>Pat
>
>PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.

[ moderator snip ]

Stefan Georg
Heerstrasse 7
D-53111 Bonn
FRG
+49-228-69-13-32


From mcv at wxs.nl  Fri Jul 16 22:50:18 1999
From: mcv at wxs.nl (Miguel Carrasquer Vidal)
Date: Fri, 16 Jul 1999 22:50:18 GMT
Subject: accusative and ergative languages
In-Reply-To: <001f01becdcd$786928e0$309ffad0@patrickcryan>
Message-ID: <FRI.16.JUL.1999.225018.GMT.>

"Patrick C. Ryan" <proto-language at email.msn.com> wrote:

>And to answer your -- I hope not purposeful -- distortion of what I wrote,
>let me say explicitly that I did not assert "there are ergative languages
>without (any) splits". I asserted that Thomsen did not, at least in her
>grammar, identify splits in Sumerian, which you seemed to think she had.

Of course she has, unless your edition differs from mine: p. 51,
paragraph 42:

	On the morphological level Sumerian has thus an ergative
 	system in the nouns and the intransitive vs. the
	transitive hamTu conjugation [...]. In the pronouns and
	the transitive maru^ conjugation vs. the intransitive
	verb, on the other hand, the system is nominative-
	accusative [...].
	This `split ergativity' is no uncommon phenomenon, in
	fact no ergative language is entirely ergative in both
	syntax and morphology.

=======================
Miguel Carrasquer Vidal
mcv at wxs.nl


From Georg at home.ivm.de  Fri Jul 16 22:16:05 1999
From: Georg at home.ivm.de (Ralf-Stefan Georg)
Date: Sat, 17 Jul 1999 00:16:05 +0200
Subject: accusative and ergative languages
In-Reply-To: <002601becdce$9c77b340$309ffad0@patrickcryan>
Message-ID: <SAT.17.JUL.1999.001605.0200.>

>R-S sighed:

I think I have the sympathy of most people for an occasional sigh during
this thread ...

>> This is getting weary. I think no sane linguist will be unaware of the fact
>> that there are languages with fewer purely morphological means than others.


>Pat concludes:

>In my opinion, dearth of morphological devices is validly characterised as
>"simple".

"Simple" as a characterization of the *morphological subsystem* of a
language. Any dearth of means in one subsystem has to be compensated in
another one to keep the language what it is meant to be, a proble-solving
device. A language is more than its morphology. Anything else is
balderdash, mystification, eighteenth-century stuff. Do you have an English
translation of the Iliad handy ? Do you think it is accurate enough to
convey what the unknown writer going by the name of Homer wanted to say,
its poetic beauties apart, of course ?? Yes ? See what I mean ?
I may have difficulties to come through to you, but, please, you have to
bear with me. I speak a language rich in morphological means, you cannot
expect me to be able to express the richness of my profound thoughts
accurately in a language as simple as English, the best parts are
irretrievably lost, because of the lack of a synthetic dative ;-)
["because-of-the-lack-of", good heavens, take Turkish: noksanlIGIN iCin,
two words instead of five, now *that's* a language !].
(This could also explain the strange ways Marxism went in China, ever
thought of this ? ;-))

Stefan

Stefan Georg
Heerstrasse 7
D-53111 Bonn
FRG
+49-228-69-13-32


From petegray at btinternet.com  Fri Jul 16 18:30:40 1999
From: petegray at btinternet.com (petegray)
Date: Fri, 16 Jul 1999 19:30:40 +0100
Subject: accusative and ergative languages
Message-ID: <FRI.16.JUL.1999.193040.0100.>

> Pat said:
> .... Why not list two or three additional elements that function in
> Chinese equivalent to inflectional affixes?

The most obvious omission from your list is the "directional" affixes,
simple and complex, e.g.: lai (come); qi3-lai; etc
And you can add zhe, guo, zheng4zai4, yao4, ...etc etc all showing different
limitations to the action.

Whether you count these as "inflexions" or as "affixes" or as common second
elements in compound verbs, is probably a matter of taste.

There are books of Chinese verbs that list all these things for you, but
alas, I have none handy.

Peter


From kurisuto at unagi.cis.upenn.edu  Fri Jul 16 20:12:28 1999
From: kurisuto at unagi.cis.upenn.edu (Sean Crist)
Date: Fri, 16 Jul 1999 16:12:28 -0400
Subject: accusative and ergative languages
In-Reply-To: <l03102801b3b1682958b3@[195.247.226.43]>
Message-ID: <FRI.16.JUL.1999.161228.0400.>

Somebody (I've lost track of who) wrote:

> >Yes, but if a new verb were formed, e.g. webben (from English 'web'), there
> >is no possibility that it would be conjugated webben, wabb, gewobben.

This sort of thing actually _does_ happen sometimes; it just isn't that
common.  The English verb "strive" is a loan from Old French; but it is
often declined as if it were an old Class I ablauting strong verb (strove,
striven).  So I'd correct "no possibility" to "much less likely".

  \/ __ __    _\_     --Sean Crist  (kurisuto at unagi.cis.upenn.edu)
 ---  |  |    \ /     http://www.ling.upenn.edu/~kurisuto/
  _| ,| ,|   -----
  _| ,| ,|    [_]
   |  |  |    [_]


From proto-language at email.msn.com  Sat Jul 17 04:14:58 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Fri, 16 Jul 1999 23:14:58 -0500
Subject: Momentary-Durative
Message-ID: <FRI.16.JUL.1999.231458.0500.>

Dear Peter, Vidhyanath, and IEists:

 ----- Original Message -----
From: petegray <petegray at btinternet.com>
Sent: Wednesday, July 14, 1999 2:21 PM

> Vidhyanath's two postings on aspect show clearly what the issue of aspect in
> PIE may be - we simply don't have a clear definition of what we are talking
> about.

> Greek usage does not agree with Vedic, and where traces of aorists exist in
> other languages (e.g. Armenian, OCS, Albanian, Latin, Baltic), we cannot
> convincingly recover an original aspectual meaning.

> The formations are clearly PIE, but I believe I can continue to claim that
> the distinctions between them, if recoverable, are certainly not those we
> find in Greek.   The Armenian-Greek- I-I sprachbund has certain known
> innovations, such as vocabulary and the augment, and we should not be miseld
> into accepting as original what may be another innovation, namely the
> development of new distinctions based on an existing variety of forms.

I do not wish to intrude on the interesting discussion between you in
progress. However, for whatever it may be worth, I have recently finished a
first draft of a small essay on IE aspects, and a few of you may find it of
passing interest:

http://www.geocities.com/Athens/Forum/2803/IE-VerbalInflection.htm

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From proto-language at email.msn.com  Sat Jul 17 04:27:07 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Fri, 16 Jul 1999 23:27:07 -0500
Subject: Momentary-Durative
Message-ID: <FRI.16.JUL.1999.232707.0500.>

Dear Peter and IEists:

 ----- Original Message -----
From: petegray <petegray at btinternet.com>
Sent: Wednesday, July 14, 1999 3:19 PM

<snip>

> And there are very many instances of roots showing nasal presents in one IE
> language, and not in another, or indeed appearing in both forms in the same
> language, as Latin cumbo and cubo, or Greek leipo: and limpano:

<snip>

> No, for the reasons just stated.   There are enough counter-examples to show
> that whatever the conditions were for the "selection" of nasal, full grade
> or suffixed present, they are now as unrecoverable as they are within Latin,
> Greek, or Sanskrit, where very little difference of meaning can be
> consistently shown.

I have found that postulating an ingressive meaning for the affix *-n works
very well. It is my opinion that this *-n was originally added to a root not
infixed.

The example you gave of Latin cubo:, 'lie', and cumbo, 'lie down', will
serve as an excellent example of the ingressive force of the affix.

 Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From proto-language at email.msn.com  Sat Jul 17 05:21:32 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Sat, 17 Jul 1999 00:21:32 -0500
Subject: accusative and ergative languages
Message-ID: <SAT.17.JUL.1999.002132.0500.>

Dear Wolfgang and IEists:

First let me thank you for your very interesting and thought-provoking
comments.

----- Original Message -----
From: Wolfgang Schulze <W.Schulze at lrz.uni-muenchen.de>
Sent: Wednesday, July 14, 1999 5:18 AM

> "Patrick C. Ryan" schrieb:

>> .... As for lessened degrees of animacy, most ergative languages
>>  have antipassives to indicate this.

Wolfgang wrote:

> Before you invite people to subscribe to this claim you should first
> demonstrate that a) "most ergative languages have antipassives".

Pat writes:

I have reviewed Dixon, and apologize for the claim. I inferred incorrectly
that beacuse Dixon so often mentioned antipassives in connection with
ergatives that the frequency of their co-occurence was high.

Wolfgang worte further:

> This
> claim suggests that "ergative" is a substantial attribute that can be
> used with the referent "language". In earlier postings I have tried to
> show that "ergativity" (as well as "accusativity") represents a label
> for a structural BEHAVIOR of single paradigms WITHIN a language system.
> For instance (as I have said) a language system may be ergative in its
> agreement paradigm, accusative in owrd order, accusative in case
> marking, ergative in discourse cohesion etc. (to give a fictive
> example). Hence, there are NO "ergative (or "accusative") languages" or
> only, if you use this term in a very informal sense.

Pat writes:

In the sense you are using these, they seem to be of little value in
describing anything.

Wolfgang wrote further:

> Now, if you talk
> about antipassives, you should make clear to which morphosyntactic
> domain you allude to. Moreover, your claim suggests that most 'ergative
> languages' are reference dominated in the sense of Role and Reference
> Grammar. Only if a given language system ("operating systems in terms of
> the Grammar of Scenes and Scennarios" (GSS)) uses actant encoding
> devices to indicate fore- and/or backgrouding (instead of - for instance
> - smeanti coles such as agent/patient...) we can expect some kind of
> diathesis be it passive or antipassive (note that again passives and
> antipassives represent two poles on a much more complex scale that also
> involes bi-absolutives, pseudo-passives and many more structures).

Pat writes:

I am sorry that I do not agree with the validity of this distinction. For
me, 'actant' is 'agent'. Perhaps you can explain the difference.

Wolfgang wrote:

> The fact, however, is that many 'ergative' languages lack an
> antipassive. For instance, there are nearly 30 East Caucasian languages
> all of them using some ergative strategies in at least parts of their
> operating systems. But only a handfull of them (five or six, to be
> precise) have true antipassives (only one has some kind of
> "pseudo-passive").

Pat writes:

Are you asserting that the majority of ergative languages do not have
anti-passives?

Wolfgang wrote:

> The same is true for accusative systems (as you
> probably know). Hence, antipassives are a possible extension of ergative
> stretagies, they cannot serve for any kind of typological
> generalization.

Pat writes:

I was not aware that I was using anti-passives as a "typological
generalization". I only suggested (I thought) that that was one method
ergative languages employed for degrees of animacy/actancy.

Wolfgang wrote further:

> b) Antipassives have rarely to do with the "lessening of animacy". The
> most common inferences that allow antipassive structures are:

> - Reduction of 'activity' (that is the degree to which an actant is
> thought to be 'active' during a *specifc* (and single) event. From this
> another inference is given:

> - Habitual, durative action (-> imperfectiveness).....

> - The event becomes less discrete, hence less transitive. Another
> inference: The 'patient' looses its referentiality: It cannot be
> subjected to wh-questions, it cannot be counted, very often such
> referents are mass nouns or collectives...

> - Antipassives are part of the discourse cohesion strategies (most
> famous example is Dyirbal): Here, antipassives are neither semantically
> nor syntactically motivated, but merely a pragmatic feature of topic
> chaining.

Pat writes:

After having read Dixon, I can hardly be unaware of these considerations;
however, I can see that degree of actantcy might be defined in this context
meaningfully. But actantcy and animacy are still closely related.

As for the far-reaching conclusions of Dixon based on discourse conhesion
strategies, in my opinion they are flawed because these strategies are
purely conventional.

In English, I can say: "John hit me, and he went away" or "I hit John and he
went away". What is the discourse cohesion strategy for English?

Wolfgang further wrote:

> There are much more functional options that are carried out by
> antipassives. In fact, all these options NEVER allow such a claim as
> quoted above (rather, they contradict it).

Pat wrote earlier:

>> Although Dixon is certainly a man who has devoted much thought to
>> ergativity, I find something inherently problematical in combining ergative
>> and accusative features in one sentence (a little schizophrenic) which he is
>> forced to do by analyzing pronominal and nominal structures differently when
>> they occur in the same sentence.

Wolfgang wrote further:

> WHY? It all depends from how you interpret erg. and acc. features.
> Confer for instance the following (one!) sentence from (informal) Lak
> (East Caucasian):

> t:ul b-at:-ay-s:a-ru zu
> I:ERG I:PL-hit-PART:PRES-ASS-SAP:PL you:PL:ABS
> 'I surely hit you (plural)'

> [For the expert: Standard (Literary) Lak would have 'na bat:ays:aru zu'].

> This sentence is:

> ACC with respect to word order [*zu t:ul bat:ays:ara would be ERG]

Pat writes:

I have to plead ignorance of Lak however, I find an analysis of an otherwise
completely ergative sentence as having an ACC word-order impossible. I
believe this opinion rests on a false analysis of word-order significance.

Wolfgang continued:

> ERG with respect to case marking ['neutral with 'na' for "I" is also
> possible]
> ERG with respect to class agreement (b- = (here) class [+hum;+plural])
> ERG with respect to personal (or, better, speech act participant)
> agreement (-ru is SAP:PL and agrees with 'zu' "you:PL").

> But if you say "I am surely hitting you (plural)", you get:

> na b-at-la-ti-s:a-ra zu
> I:ABS I:PL-hit1-DUR-hit2-ASS-SAP:SG you:PL:ABS

> Here we have:

> ACC with respect to word order
> ERG with respect to calss agreement
> ACC (or neutral) with respect to case marking)
> ACC with respect to SAP agreement (-ra is triggered by 'I:ABS').

Pat writes:

So, agent marked ABS and patient marked ABS is, according to you, ACC with
respect to case marking? Sorry, not convincing at all! And I think the SAP
difference can be explained as indicating seriality of the patients involved
as recipients of the action by *one* agent.

Wolfgang wrote further:

> Now, please tell me: Is Lak an 'ergative' or an 'accusative' language?
> [Please note that I did not include (among others) strategies of
> discourse cohesion, reflexivization and logophization].

Pat writes:

By these examples, I would tell you that Lak is 'ergative'; and that
accusativity is not demonstrable from these examples.

Pat wrote previously:

>> I think it is likelier that, because of perceived greater animacy (or
>> definiteness), pronouns have a different method of marking that can still be
>> interpreted within an ergative context.

Wolfgang wrote:

> This a (very simplified) 'on-dit' that stems from the earlier version of
> the Silverstsein hierachy. Again, we have to deal with the question,
> whether a 'pronoun' (I guess you mean some kind of 'personal pronouns')
> can behave 'ergatively' or 'accusatively'. The list below gives you a
> selection of SAP case marking in East Caucasian languages with respect
> to ABS/ERG:

> ABS vs. ERG ABS = ERG
> ALL ---
> Singular Plural
> Plural Singular
> 1.Incl. Rest
> 1:SG Rest
> 2:SG Rest
> 1:SG/PL Rest
> --- ALL

> This list (aspects of personal agreement NOT included!) shows that SAP
> pronouns may behave different within the same paradigm. Any
> generalization like that  one quoted above does not help to convey for
> these data...

Pat writes:

I do not have a reference book for Lak so that my hands are somewhat tied.
But, I have found that paradigms are often inconsistent in ways that reflect
earlier lost phonological changes, or other lost schemata. The locative
plural terminations of IE are certainly not, in origin, terminations of the
locative plural. Etc.

But, Wolfgang, I have to thank you because you are stretching my mind, and I
like the feeling.

Thank you.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From X99Lynx at aol.com  Sat Jul 17 07:39:22 1999
From: X99Lynx at aol.com (X99Lynx at aol.com)
Date: Sat, 17 Jul 1999 03:39:22 EDT
Subject: Chronology of the breakup of Common Romance [long]
Message-ID: <SAT.17.JUL.1999.033922.EDT.>

In a message dated 7/13/99 6:16:38 PM, rmccalli at sunmuw1.MUW.Edu wrote:
<<Some popular writers have also proposed Volcae as the source of English
"folk" but I've never seen this proposal from professional linguists.>>

In a message dated 7/16/99 2:53:27 AM, edsel at glo.be replied:
<<Even less likely : the peoples that used the word had probably never heard
of the Volcae, not even second-hand.>>

Actually, of the few things we hear of the Volcae Tectosages (versus the
Volcae Arecomici, a more southwestern federation of Celts) one of them is
that they moved into "Germania" and were becoming assimilated.   And we know
this from one of the most credible sources in antiquity (among others of the
same period):

<<And there was formerly a time when the Gauls excelled the Germans in
prowess, and waged war on them offensively, and, on account of the great
number of their people and the insufficiency of their land, sent colonies
over the Rhine.

Accordingly, the Volcae Tectosages, seized on those parts of Germany which
are the most fruitful [and lie] around the Hercynian forest, (which, I
perceive, was known by report to Eratosthenes and some other Greeks, and
which they call Orcynia), and settled there. Which nation to this time
retains its position in those settlements, and has a very high character for
justice and military merit; now also they continue in the same scarcity,
indigence, hardihood, as the Germans, and use the same food and dress; but
their proximity to the Province and knowledge of commodities from countries
beyond the sea supplies to the Gauls many things tending to luxury as well as
civilization.   Accustomed by degrees to be overmatched and worsted in many
engagements, they do not now even compare themselves to the Germans in
prowess.>>

- Caesar, Gallic War, 6.24 (Loeb)

It may be significant that about 150 years later, these Volcae have
completely disappeared in the descriptions of the same territories by Tacitus
and Ptolemy.  If the Volcae could have been assimilated, so could their
self-name.

Currently, the origins of the terms I use to describe my nationality and
other affilations include, first, the name of a 15th Century Italian sailor
(no, not 'Vespucian'). Then there's a Latin name compounding what looks to be
a Celtic one ('Pennsylvanian').  Others have included an anglicizing of a
Delaware name for an even earlier group of Native Americans, names from the
Greek alphabet, a sports team name borrowed from a rather unpleasant African
fauna and something in a Dutch place name that ended up being rendered as
"Brooklynite' - a term of no small pride and self-identification, as
mysterious as that may seem to the outsider.  If 'folk' moved around anything
like any of these other self-names, then I'm sure we should not be surprised
where it came from.

In a message dated 7/13/99 6:16:38 PM, rmccalli at sunmuw1.MUW.Edu wrote:
<<Isn't the term Vlakh anachronistic in this sense? Vlakh is first
documented in the middle ages, isn't it?>>

In a message dated 7/16/99 2:53:27 AM, edsel at glo.be replied:
<<Indeed. The medieval use of 'Vlach' probably referred to the 'Latin'
Rumanians, cf. Wallachia. In modern Greece, the word is also used to refer to
the local nomads/gypsies of unclear (i.e. to me) ethnic origin (maybe
Rumanian Gypsies, 'Roma', or from Pannonia? Maybe Albanians?).>>

But in fact the form 'vlach' appears very late in medieval times - one of the
arguments that the migration into Romania was relatively recent.  There are
even indications that at one point there was a diachronic presence with
'woloch' (Celt, Roman, Frank) still existing alongside of 'vlach' outside of
South and East Slavic.  In fact, before 'vlach' first enters English, it is
preceded by a century by 'Woloch" refering to Vlachs.  See also the Hungarian
form 'olasi' mentioned in my earlier post.  [And I see also that in Western
Slavic, "walska" (Old Bo. and something like it in Pol) is "war."]

I don't think anyone seriously argues with the idea that 'vlach' is derived
from the name applied to Romance-types, particularly Romanized Celts - and
not orginally a Rumanian-speaking group.  (BTW, what is 'balkan' from?)

With regard to the vallum/Volcae connection, edsel at glo.be replied:

<<That excludes a Roman origin of the word.  All this sounds like a typical
19th century invention by Latin-worshippers>>

As opposed to 19th Century Wagnerian Teutonists or Pan-Slavists, who weren't
above some invention of their own.  I got a post that corrects me about the
'wall' theory - apparently a British scholar named Lewis suggested in a RS
paper in the twenties that both vallum ('vallus stake') and waldh, etc.
derived from the  early proto-european [sic] "fala", which the Romans applied
to wall-building and seige scaffling, and created heights in general, fully
aware that it was a borrowed word "quod apud Etruscos significat caelum"; cf.
fulcio, fulsi, fultum: to strenghten, particularly by walls town-places ."

Regards,
S. Long


From jer at cphling.dk  Sat Jul 17 13:50:17 1999
From: jer at cphling.dk (Jens Elmegaard Rasmussen)
Date: Sat, 17 Jul 1999 15:50:17 +0200
Subject: Momentary-Durative
In-Reply-To: <007201bece36$602a66e0$a53bac3e@xpoxkjlf>
Message-ID: <SAT.17.JUL.1999.155017.0200.>

To "petegray" and List,

Thank you, Peter, for a challenging reply. I'll try and do it justice.

On Wed, 14 Jul 1999, petegray wrote:

> [...]
> Yes - but you have made the assumption that this root always shows the nasal
> in the present.   This is not the case.
> (a) Albanian shows two forms, one with, and one without the nasal.
> (b) Old Irish shows no nasal.
> (c) Greek a-tuzomai shows no nasal.
> (d) If Pokorny is right to link the root with stud- then we have no nasal in
> Latin studeo, nor in the various reflexes of it in Germanic.

Sure, verbs can form other derivatives too. Gmc. *staut-i/a- must be an
old intensive (*stu-sto'wd-); studeo is a stative in *-eH1-; atu'zomai 'am
frightened' has a velar in atukhthei's and does not look related; Alb.
shtyj, -yn is no evidence since the nj-present is productive (if the old
form is shtyenj it is hard to understand as related at all). But Old
Irish -tuit 'falls' is *tudeti and identical with tudati. So maybe for
this verb the stem restructuring should be pushed back in to the
prehistory of the protolanguage to give us PIE variant presents - though
palallel development of course must be very common.

>>  There are so many instances of the _same_ verbal root turning up
>> with a nasal present in IE branches that have nothing else in common ...

> And there are very many instances of roots showing nasal presents in one IE
> language, and not in another, or indeed appearing in both forms in the same
> language, as Latin cumbo and cubo, or Greek leipo: and limpano:

Right, and in such cases it is advisable to assume a set of nasal present
and root aorist, since they are so often combined, and are retained
together with so many verbs. Thus Gk. leip-/limp- are inded both presents,
but *ley{kw}-e/o- is a perfectly normal IE root-aorist subjunctive which,
due the closely allied function of that category, often came to compete
with the inherited present.

>>. Is it a coincidence
>> that *k^lew- 'hear' forms a nasal present in Indo-Iranian and Celtic?

> Yes.   It has no nasal in Greek or Latin or some others. (and so on with the
> rest of your list)

Why should it have, if the individual languages went crazy and levelled
almost all differences between the aspect stems?

> I cannot agree with you that certain verbal roots always had nasal presents
> in all IE languages.   This is simply factually untrue.   But without that
> assumption, your argument collapses.

Not "in all IE languages", only in the IE protolanguage, which apparently
had a general (unilateral) solidarity between nasal presents and root
aorists. From this common point of departure the languages developed all
sorts of differences (even closely related lgg. differ in this respect,
oh yes).

>>. Does that not indicate
>> that the assigment of the nasal-infix structure as the present of certain
>> roots was fully lexicalized in the protolanguage?

> No, for the reasons just stated.   There are enough counter-examples to show
> that whatever the conditions were for the "selection" of nasal, full grade
> or suffixed present, they are now as unrecoverable as they are within Latin,
> Greek, or Sanskrit, where very little difference of meaning can be
> consistently shown.

>> Oh no, it works only in the direction that a nasal present has a root
>> aorist beside it in practically all cases.

> Not a surprise, if you ignore all the langauges where this is not the case.
> Likewise, not a surprise, since root aorists were an old form.   The same
> could be said of full grade presents, or suffixed presents.   So the
> statement really is without significant meaning.

No, thematic presents and root-presents typically take the s-aorist (or
suppletion).

>> Also the reduplicated presents
>> and the y-presents generally form root aorists.

> Yes - so your argument (that a root aorist implies a nasal present) is
> really rather weak.   Yet without this argument, you cannot derive tuda'ti
> the way you do.

No again, a root aorist _is implied by_ a nasal present. And since that is
found, there is very good reason to believe that there was a root aorist
in IE. And then, surprise, that offers all the basis you need for a smooth
and natural derivation of the tudati type - smooth at least in comparison
with many stories of verbal stem-formation that can be followed in
attested languages. The story also explains what is perhaps the most
important point of all, which I did not mention in my posting: the
punctual aspect value observed by Renou in the use of the tudati present
type (Louis Renou: Le type ve'dique tuda'ti. Me'langes Vendryes, Paris
1925, 310-16, a classical and oft-quoted philological paper written
completely without comparative or even diachronic bias). - In fact, I am
only saying that tuda-, which is structured like vida-, may have been an
aorist stem just like the latter.

>> [in] PIE, each verb had mostly picked one form for its present. The massive
>> concord among the IE languages in this respect

> There is no such concord.  Even a single language shows variety of
> formations of the present, and the variety across the languages is
> considerable.

Yes, but easily explained as secondary, or not relevant, if the forms
adduced are not from the primary verb, or if they belong to types that got
productive in the separate lives of the individual languages. When that is
subtracted, it is quite strinking how well the old core of the different
branches get to resemble each other. The impression that IE verbal stem
formation just mutates erratically and that no common system is
recoverable is not compatible with current knowledge.

> But it would be good to have firm data on this - anyone got
> three weeks to spare going through the text books?

So you haven't done that? It's quite easy now: LIV (Lexikon
Indogermanischer Verben, by Helmut Rix & al., Wiesbaden 1998) offers all
you need, including reconstructions and developmental tracings.

>> - while the even greater discord is easily
>> explained by continued normal language change.

> No.   Normal language change would not turn a nasal present into a
> reduplicated one or vice versa -

Why not - if they were both equally expected as the companion of a root
aorist? Is that not the way analogy works? Let me give you two
illustrative case stories. The IE status of a reduplicated present from
the two verbs *sed- 'sit down' and *steH2- 'stand up' cannot be questioned
(there is *si-sd-e/o- in Ital., Gk., IIr.; and *sti-steH2- or rather its
old subj. *sti-stH2-e/o- in IIr., Gk., Ital. and Celt.), nor can a root
aorist be avoided for the latter; I accept it for both, of course, but
*sed- as aor. is not so well-preserved. Now, the paradigms *si'-sd-e-ti,
*se'd-t and *sti'-stH2-e-ti *sta'H2-t thus restored had a funny fate in
Balto-Slavic and Germanic: BSl. changed *sisd-e/o- into *sind-e/o-, and
Gmc. back-formed *st at 2-n-t-e/o- from the aor. *staH2-t (drawing the -t- to
the stem, i.e. apparently after an adjustment of *staH2-t to "*staH2t-e"
with the productive ending of the perfect). Why would they do such a
thing? There can be only one answer: because the reduplicated present type
was going out of fashion, while the nasal present was still productive and
still went well with root aorists.

>  -, nor would it turn a -sk suffixed one into a nasal present, etc, etc.

That's right, these do not belong together, and I never said they did.
However, in a few odd cases the whims of chance did end up combining them,
as when Hitt. nasal verbs form durative derivatives with -sk- like all
other verbs do, or Toch. makes itself a prs. formant -na"sk- (-na:sk-),
thereby giving inherited nasal presents a productive present-stem marking.
And if there are both a set "root-aor. + nas.-prs." and an inchoative
"s-aor. + sk-prs.", and the functional difference was lost, you may end up
finding a nasal present combined with an s-aorist, or an sk-prs. with a
root aorist. In some cases one may even suspect that this had happened
before the dissolution of the protolanguage; a prominent case is Gk.
gigno:'sko: with root-aor. e'gno:n, but Gothic kunnai{th} combining nasal
infix with the vocalism (*-n-e:-, not *-n-o:-) of the s-aor. that also
gave the stem of Gmc. *kne:-(j)i/a- 'know', retained with the -s- in Hitt.
ganest 'recognized' (*g^ne:'H3-s-t). In a principled analysis, this
reflects two IE verbs, (1) *g^n.-n¢-H3-ti, aor. *g^no'H3-t; (2)
*g^n.H3-sk^e'-ti, aor. *g^ne:'H3-s-t. The latter set is directly continued
as a synchronic paradigm by Arm. {cv}ana{cv}'em, aor. caneay as seen by
Jasanoff.

The most rewarding experience during the time I have been watching
Indo-European Studies has been to see the protolanguage come alive and
assume an increasingly well-established structure, with archaisms and
productive patterns, just like the living languages we know. I react in
defense of the field when I see somebody turning the clock back to a stage
we have left behind.

Jens


From CONNOLLY at LATTE.MEMPHIS.EDU  Sat Jul 17 16:49:17 1999
From: CONNOLLY at LATTE.MEMPHIS.EDU (CONNOLLY at LATTE.MEMPHIS.EDU)
Date: Sat, 17 Jul 1999 11:49:17 -0500
Subject: Interpreting ergative sentences
Message-ID: <SAT.17.JUL.1999.114917.0500.>

>[Ed Selleslagh]

>Speaking about Castilian, I would like to have your views on a peculiarity,
>or rather a tendency, that is still productive at least in popular speech
>('le-ismo'), namely the tendency to use the indirect subject form where all
>other Latin (and other West-European) languages use the direct object form,
>and almost exclusively with persons (animate), e.g. "le v=ED" ("le vi' "), "I
>saw him".

I don't actually know Spanish; I just know a few examples of how such verbs
work in Spanish and many other languages.

>It has sometimes be suggested that this was a substrate influence from
>Basque, an ergative language (that lacks an accusative, of course), but
>opinions are extremely divided on this subject. On the other hand, it seems
>to me this could (but I don't know) also be related to some of the arguments
>presented in the discussion on ergative/accusative concerning the distinct
>role of animates.

I don't know any Basque either.  But I believe that the same tendency is found
in Sicilian dialects, so I'm a little leery of looking to Basque for an
explanation.  Especially since, as far as I know, personal absolutive NPs with
transitive verbs get the same treatment as impersonal NPs, while beneficiary
"indirect objects" with verbs such as 'give' get a different, "dative"
treatment.  If that's so, then Basque doesn't show the tendency to treat all
personal non-ergative NPs with bi- or trivalent verbs alike.  But better ask
Larry about that.

I should add further that I'm generally suspicious of explanations based on
substrate influence.  Yes, substrates are real, and sometimes they do seem to
provide a plausible explanation of one phenomenon or another.  But languages
are capable of changing strictly on their own, and I think in general we should
look first for language-internal solutions.  Since even American Spanish uses
_a_ before personal direct objects, as it does before indirect objects, it is
hardly surprising that Castillian would tend to use identical pronouns for all
personal objects, whether direct or indirect.

Leo

Leo A. Connolly                         Foreign Languages & Literatures
connolly at latte.memphis.edu              University of Memphis


From petegray at btinternet.com  Sun Jul 18 17:33:54 1999
From: petegray at btinternet.com (petegray)
Date: Sun, 18 Jul 1999 18:33:54 +0100
Subject: Momentary-Durative
Message-ID: <SUN.18.JUL.1999.183354.0100.>

>>> Strunk has shown that nasal presents go with root aorists,

Can you give us a reference please, Jens?   I would like to question this,
for the following reasons:
(a) In Greek it is largely true, but there are exceptions;   so a bald
statement would need qualification.
(b) In Latin it is largely untrue, since the aorists are either sigmatic or
the rare reduplicative  aorist (tango tetigi, claimed by some as an aorist
on the basis of Homer tetago:n, or the lengthened vowel: pango pe:gi (~
perfect pepigi).   I can only find cumbo cubui which supports the claim in
Latin.
(c) In Sanskrit there is a variety of possible presents and possible
aorists.   There are 29 roots listed by Whitney which take nasal presents
(ignoring the class nine presents).  Of these the first retains its nasal in
the aorist, the second has no aorist, the third shows non-nasal present
forms, and has either reduplicated or sigmatic aorist, and so on .....
through the list.
(d) As stated earlier, a root may show nasal and non-nasal presents in
different languages and even in the same language.

So as a bald statement I find it unbelievable.   The evidence doesn't seem
to be there.    I would like to see what Strunk actually says.

> If my observation that there is an alliance between the sk^-present type
> and the s-aorist is correct,

Both forms may be used for some other reason, for example phonetic.   In
both Greek and Latin the sk- ending occurs only after a vowel, which then
takes a sigmatic aorist for purely phonetic reasons.

>> This brings me to a general question. There seem to be two camps about the
>> category system of the PIE verb. One believes that the aorist-imperfect
>> distinction, to be equated to perfective-imperfective distinction,
>> ``always'' existed in PIE and Hittie lost this distinction, while Vedic
>> changed things around. The other considers the aspectual distinction to
>> postdate the separation of Anatolian.

> [Jens] There are these two camps, yes, and I am in no doubt that camp one is
> right. There is no way the specific forms of the aspect stems could have
> been formed secondarily in "the rest of IE" left after the exodus of (or
> from) the Anatolians. At the very least, all the _forms_ must be assigned
> to a protolanguage from which also Anatolian is descended.

I thoroughly agree with you here, Jens.   The forms are scattered over most
of the IE languages.

> And what would
> the forms be there for, if the functions that go with them only developed
> later?

Here I disagree.   There can be variety of form in a language without a
difference in meaning.   Look at German plurals (or English for that
matter).   Sometimes a difference of meaning will emerge, fixing one meaning
on one form and the other on the other (German Woerter and Worte, Dinger and
Dinge), but this does not mean that all variations must necessarily
correspond to a difference in function.    Language simply doesn't work that
way, and there are far too many counter-examples for that claim to stand.

Counter examples in English:
(a) strong versus weak past tenses, e.g. dived ~ dove (origin historical /
analogical)
(b) plural /s/ versus plural /z/ (origin phonetic)
(c) gentive versus preposition e.g. Tom's ~ of Tom (basis stylistic)
(d) Time expressions (believe it or not, 10:45 really is the same as a
quarter to eleven)
and so on.

I see the development of a functional difference as a later phenomenon, and
I see it emerging differently in Sansksrit and Greek and Latin.  I suspect
that some scholars are temporarily misled by the use of Greek names for the
Sanskrit verbal system.

Peter


From ALDERSON at toad.xkl.com  Tue Jul 20 17:01:33 1999
From: ALDERSON at toad.xkl.com (Rich Alderson)
Date: Tue, 20 Jul 1999 10:01:33 -0700
Subject: Hiatus
Message-ID: <TUE.20.JUL.1999.100133.0700.>

Dear Readers:

I will be away from these facilities until 25 July 1999.  There will be
no mailings from the Indo-European or Nostratic lists until I return.
I apologize for any inconvenience this causes.

								Rich Alderson
								list owner and
								moderator
-------


From proto-language at email.msn.com  Mon Jul 19 05:13:46 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Mon, 19 Jul 1999 00:13:46 -0500
Subject: accusative and ergative languages
Message-ID: <MON.19.JUL.1999.001346.0500.>

Dear Larry and IEists:

 ----- Original Message -----
From: Larry Trask <larryt at cogs.susx.ac.uk>
Sent: Friday, July 16, 1999 10:35 AM

> Pat Ryan wrote:

>> Now, when Larry recently quoted Dixon about the nature of the
>> ergative, he conveniently neglected to mention that Dixon
>> acknowledged that there were currently practising linguists --- not
>> amateur linguists like myself --- still defending the passive
>> interpretation of ergative constructions.

Larry responded:

> No; this is not so.  Read p. 189 of Dixon's 1994 book.

Pat responds:

I suggest you read it. "But we do still encounter scholars who insist that
there is a necessary diachronic connection, e.g. Estival and Myhill
(1988:445): 'we propose here the hypothesis that in fact all ergative
constructions have developed from passives'."  Are you suggesting that
Estival and Myhill are not "linguists", or that Shibatani, in whose book
this essay appeared, is not a "linguist"?  The quotation above is from p.
189 of Dixon's 1994 book. What are you playing at?

Pat asked:

>> I asked Larry where he had "shredded" this interpretation, and to my
>> knowledge, got no answer. If I missed the "shredding", perhaps you
>> will be kind enough to rehearse his performance for us. I have seen
>> nothing by Larry's vehemence and your allegiance to support the idea
>> that the ergative should not be interpreted as a passive.

Larry answered:

> Mr. Ryan, I'm afraid I lack the time to reply in detail to all, or even
> most, of the 37 or so postings from you that greet me each morning.

> But briefly: if a transitive sentence in an ergative language were
> "really" a passive, then its absolutive NP would be the subject and
> would exhibit subject properties.  But this not the case in most of the
> ergative languages I have heard about.  In spite of the ergative
> case-marking and/or verbal agreement, it is usually the *ergative* NP
> which exhibits subject properties, not the absolutive NP.

Pat interjects:

Still no word on the scene of the great "shredding" which was *claimed* by
Larry.
And what are subject properties that an absolutive NP never displays?

Larry continued:

> Take any ergative language you like.  Assume that the subject of an
> intransitive sentence is the sole (absolutive) NP in it.  Examine the
> syntactic properties of that NP, and tabulate them.

> Then look at transitive sentences.  Tabulate the syntactic properties of
> the absolutive and ergative NPs, and compare these with the preceding.

> In the great majority of ergative languages, it is the ergative NP which
> shares the subject properties of the intransitive subject.

> A few examples of typical subject properties:

> The subject controls reflexive and reciprocal NPs.

> The subject cannot itself be reflexive or reciprocal.

> The subject controls the empty NP in an empty-NP complement.

> The empty NP in an empty-NP complement is itself a subject.

> A subject can be coordinated with another subject.

Pat responds:

Why do you not give us an example of these so-called subject properties in
some ergative language besides Basque? And how were these properties
selected?

Larry continued:

> And so on.  There are also various language-specific tests.  For
> example, in some varieties of Basque, the object of a gerund (but not
> the subject of a gerund) goes, exceptionally, into the genitive case.
> Neither the absolutive subject of an intransitive gerund nor the
> ergative subject of a transitive gerund can be genitivized, but the
> absolutive object of a transitive gerund can be.

Pat responds:

Dialect-specific and totally unconvincing for ergative languages in general.

Larry continued:

> One more test from Basque.  In Basque, the subject of an intransitive
> sentence cannot be reflexive or reciprocal.  So, Basques can say,
> literally, `Susie and Mike [ABSOLUTIVE] were talking to each other', but
> they cannot say *`Each other [ABSOLUTIVE] was talking to Susie and
> Mike'.

Pat responds:

Since '*Each other was talking to Susie and Mike' is equally ridiculous, I
fail to see any valuable point made. Frankly, I am amazed. A reflexive
requires an agent and a patient, and a reciprocal requires two agents and
two patients. An intransitive verb, by definition, has only one NP element
so any two(or four)-NP-element construction obviously is a contradiction in
terms.

Larry continued:

> Now, in a transitive sentence, Basques can say `Susie and Mike
> [ERGATIVE] slapped each other [ABSOLUTIVE]', but they cannot say *`Each
> other [ERGATIVE] slapped Susie and Mike [ABSOLUTIVE]' -- or, in terms of
> the discredited passive theory, they cannot say *`Susie and Mike
> [ABSOLUTIVE] were slapped by each other [ERGATIVE]'.

Pat responds:

More nonsense! 'Each other' cannot function as an ergative or nominative
subject. So what? That is semantic not grammatical. That it can be used in
some languages in an oblique case (as in English above) is to be expected.

Larry continued:

> In an intransitive sentence, the absolutive subject cannot be reflexive
> or reciprocal.  In a transitive sentence, the ergative subject cannot be
> reflexive or reciprocal, but the absolutive object *can* be.

Pat responds:

Answered above.

Larry continued:

> All tests for subjecthood in Basque give the same result: it is the
> ergative NP in a transitive sentence which shares the subject properties
> of the absolutive NP in an intransitive sentence, and therefore it is
> the ergative NP, and not the absolutive NP, which is the subject of a
> transitive sentence.

Pat responds:

If you are referring to the "tests" above, you have proved nothing.

Larry continued:

> There are unusual languages, of course.  In Dyirbal, and in
> Nass-Gitksan, the absolutive NP in a transitive sentence shows at least
> some subject properties, while the ergative NP does not show the same
> properties -- though it does show other subject properties.  This is the
> phenomenon we call `syntactic ergativity'.  But I know of no language
> which is syntactically ergative without exception, and I know of few
> languages which are syntactically ergative at all.

Pat responds:

Of all the languages I have ever seen, Basque is, by a mile, far the most
"unusual" language.

Larry continued:

> The "passive" view of ergative languages in general is indefensible.

Pat responds:

Obviously, I do not think so. And Estival and Myhill (and probably
Shibatani) do not either --- not to mention the majority of linguists of the
past.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From proto-language at email.msn.com  Mon Jul 19 06:26:43 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Mon, 19 Jul 1999 01:26:43 -0500
Subject: accusative and ergative languages
Message-ID: <MON.19.JUL.1999.012643.0500.>

Dear Ralf-Stefan and IEists:

 ----- Original Message -----
From: Ralf-Stefan Georg <Georg at home.ivm.de>
Sent: Friday, July 16, 1999 4:54 PM

R-S wrote:

> How do you live with the fact that some "ergative languages" have
> independent passives ?

Pat asks:

Generalities are somewhat interesting but specifics would be even more
interesting.

<snip>

R-S wrote:

> And, as far as I remember, this whole brouhaha started with me asserting
> that all known "ergative languages" have some split, followed by you
> prompting we to show such a thing in Sumerian. I'm not a mind-reader, but
> this could be interpreted as a challenge aiming at the gist of my
> assertion, n'est-ce pas ? I managed to show those splits,

Pat interjects:

To my knowledge, you did not.

R-S continued:

> although the
> discussion got a bit swamped under some hassle over maru:- and
> HamTu-conjugations, but I managed to do it. At least I got you to accept,
> late in coming, though, that my initial assertian still stands up.

Pat responds:

I did *not* accept your initial assertion. Re-read my concession.

Pat wrote previously:

>> You might review your own procedures for employing quotation marks. Your use
>> of them on "there are ergative languages without any splits" strongly and
>> falsely implies that I wrote this in the context of a judgment on the
>> question.

R-S replied:

> I accept the reproach of not being a mind-reader ...

Pat answers:

Not 'mind-reading', just simple 'reading' will do nicely.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From colkitto at sprint.ca  Tue Jul 20 10:53:36 1999
From: colkitto at sprint.ca (Robert Orr)
Date: Tue, 20 Jul 1999 06:53:36 -0400
Subject: accusative and ergative languages
Message-ID: <TUE.20.JUL.1999.065336.0400.>

There's also "sneak - snuck" in English and "schreiben -schrieb-geschrieben"
in German, both of which are analogical formations  on originally borrowed
words..

Robert Orr


From larryt at cogs.susx.ac.uk  Mon Jul 19 10:49:05 1999
From: larryt at cogs.susx.ac.uk (Larry Trask)
Date: Mon, 19 Jul 1999 11:49:05 +0100
Subject: accusative and ergative languages
In-Reply-To: <002201bed014$3c87ab00$8f9ffad0@patrickcryan>
Message-ID: <MON.19.JUL.1999.114905.0100.>

On Sat, 17 Jul 1999, Patrick C. Ryan wrote:

> As for the far-reaching conclusions of Dixon based on discourse
> conhesion strategies, in my opinion they are flawed because these
> strategies are purely conventional.

Not so.  Read on.

> In English, I can say: "John hit me, and he went away" or "I hit
> John and he went away". What is the discourse cohesion strategy for
> English?

You've overlooked the crucial null-subject cases:

`John hit me and went away' *must* mean `John went away'.

But `I hit John and went away' *must* mean `I went away'.

Control of null NPs is one of the syntactic properties which crucially
distinguish subjects from non-subjects in English and in some other
languages.  And there is nothing "conventional" about it: this is a rule
of English syntax.

Let me change both NPs to third-person, to avoid any complications with
agreement; suppose I say this:

`John hit Bill and went away'.

Now, in English, it is John who went away, not Bill.  However, according
to my understanding of Dixon, if you say what looks like the literal
equivalent of this in Dyirbal, it is *Bill* who went away, not John.
This is one of the ways in which syntactic ergativity manifests itself
in Dyirbal.

But not all ergative languages are the same here, and probably not even
most.  If you translate this sentence as literally as possible into
Basque, it is once again John who went away, and not Bill, just as in
English.  This is so even though the ergative morphology of Basque is
more thoroughgoing than that of Dyirbal, which is split.

That is, Basque, like English, allows subjects to be coordinated with
subjects, but not with non-subjects.  This is so even when one of the
coordinated subject NPs is ergative and the other absolutive.  Basque
does not allow the absolutive subject of an intransitive sentence to be
coordinated with the absolutive object of a transitive sentence.

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk


From Georg at home.ivm.de  Mon Jul 19 11:42:32 1999
From: Georg at home.ivm.de (Ralf-Stefan Georg)
Date: Mon, 19 Jul 1999 13:42:32 +0200
Subject: accusative and ergative languages
In-Reply-To: <002201bed014$3c87ab00$8f9ffad0@patrickcryan>
Message-ID: <MON.19.JUL.1999.134232.0200.>

>In English, I can say: "John hit me, and he went away" or "I hit John and he
>went away". What is the discourse cohesion strategy for English?

Anaphora.

Stefan Georg
Heerstrasse 7
D-53111 Bonn
FRG
+49-228-69-13-32


From proto-language at email.msn.com  Mon Jul 19 12:51:38 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Mon, 19 Jul 1999 07:51:38 -0500
Subject: accusative and ergative languages
Message-ID: <MON.19.JUL.1999.075138.0500.>

Dear Miguel and IEists:

 ----- Original Message -----
From: Miguel Carrasquer Vidal <mcv at wxs.nl>
Sent: Friday, July 16, 1999 5:50 PM

> "Patrick C. Ryan" <proto-language at email.msn.com> wrote:

>> And to answer your -- I hope not purposeful -- distortion of what I wrote,
>> let me say explicitly that I did not assert "there are ergative languages
>> without (any) splits". I asserted that Thomsen did not, at least in her
>> grammar, identify splits in Sumerian, which you seemed to think she had.

Miguel wrote:

> Of course she has, unless your edition differs from mine: p. 51,
> paragraph 42:

> On the morphological level Sumerian has thus an ergative
>   system in the nouns and the intransitive vs. the
> transitive hamTu conjugation [...]. In the pronouns and
> the transitive maru^ conjugation vs. the intransitive
> verb, on the other hand, the system is nominative-
> accusative [...].
> This `split ergativity' is no uncommon phenomenon, in
> fact no ergative language is entirely ergative in both
> syntax and morphology.

Pat responds:

Yes, Miguel, she did write that. In the foregoing paragraph or so, she gives
a concocted example of a transitive maru: sentence:

za-e sag{~}-0 mu-zi.zi-en    you (sing.) raise the head

In this so-called "nominative-accusative" sentence, -e is the normal
termination of the ergative case, so, by form, za-e is 'you (sing.) ERG'; -0
is the normal termination of the absolutive case, so, by form, sag{~}-0 is
'head ABS'.

No, I suppose, it is also possible but fruitless to argue that -e is both
ERGATIVE and NOMINATIVE and -0 is both ABSOLUTIVE and ACCUSATIVE but what
real purpose does that serve except to play games?

I am also sure that you know that the termination -en of the verb above,
which is supposed to cross-reference the transitive maru: subject does *not*
exist in Old Sumerian (presumably before Akkadian scribes messed it up too
much).

One could always argue that the termination -e, which does exist in Old
Sumerian, was a cross-reference to the nominative subject. Is that what you
believe?

So, in conclusion, though Thomsen does mention them, she certainly did not
identify "splits" in Sumerian.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From proto-language at email.msn.com  Mon Jul 19 13:17:00 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Mon, 19 Jul 1999 08:17:00 -0500
Subject: accusative and ergative languages
Message-ID: <MON.19.JUL.1999.081700.0500.>

Dear Peter and IEists:

 ----- Original Message -----
From: petegray <petegray at btinternet.com>
Sent: Friday, July 16, 1999 1:30 PM

>> Pat said:
>> .... Why not list two or three additional elements that function in
>> Chinese equivalent to inflectional affixes?

Peter wrote:

> The most obvious omission from your list is the "directional" affixes,
> simple and complex, e.g.: lai (come); qi3-lai; etc
> And you can add zhe, guo, zheng4zai4, yao4, ...etc etc all showing different
> limitations to the action.

> Whether you count these as "inflexions" or as "affixes" or as common second
> elements in compound verbs, is probably a matter of taste.

> There are books of Chinese verbs that list all these things for you, but
> alas, I have none handy.

Pat responds:

I can only refer you to the definition of 'inflection' in Larry's _A
Dictionary of Grammatical Terms in Linguistics_.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From jer at cphling.dk  Mon Jul 19 19:23:11 1999
From: jer at cphling.dk (Jens Elmegaard Rasmussen)
Date: Mon, 19 Jul 1999 21:23:11 +0200
Subject: accusative and ergative languages
In-Reply-To: <Pine.GSO.4.03.9907161610401.19800-100000@rsunx.crn.cogs.susx.ac.uk>
Message-ID: <MON.19.JUL.1999.212311.0200.>

On Fri, 16 Jul 1999, [in continuation of a discussion with Par Ryan]
Larry Trask wrote:

> [...]
> But briefly: if a transitive sentence in an ergative language were
> "really" a passive, then its absolutive NP would be the subject and
> would exhibit subject properties.  But this [is] not the case in most of
> the ergative languages I have heard about.
[...]
> The "passive" view of ergative languages in general is indefensible.

I guess some key words are "really" and "split". If ergative syntax is
allowed to be limited in a language and may still be called ergative, why
can't the passive interpetation of it be allowed a similar latitude and
still be called passive? Is it not possible that some syntactic structures
that were originally passive (with all the trimmings that word entails)
_lost_ their passive value and simply became the normal way of saying
things (like a joke that gets repeated and isn't even meant to be funny
anymore) - and that, subsequent to this change, reflexive reference and
all other subject properties were transferred to the new pragmatic
subject? It does not make the analysis pretty, but if the linguistic
history consists of a period in which people changed their minds about the
basic roles of ordinary sentences, should an accout of the truth not be as
ugly as what really happened?

I do not know of many ergative-structured languages; but all I have seen
can be safely or highly probably analyzed as being passive circumlocutions
in origin, but none of them is completely so in terms of synchronic syntax
anymore. But that does not change their prehistory as passives. Is A
asking about origins, and B answering about descriptive synchrony?

Jens


From petegray at btinternet.com  Mon Jul 19 19:41:40 1999
From: petegray at btinternet.com (petegray)
Date: Mon, 19 Jul 1999 20:41:40 +0100
Subject: accusative and ergative languages
Message-ID: <MON.19.JUL.1999.204140.0100.>

Pat said:

>... the pattern "e-a-o" *...There was never a time
> during which *any* verb in [e] followed this "pattern", or even *any* verb
> in /eR/ followed it (weNden, waNdte, gewaNdt).

Perhaps I misunderstand you.   This pattern is one of the standard patterns
of strong verbs in modern German, most prevalent among verbs in -eR-, but
not restricted to them, e.g.:

befehlen, befahl, befohlen             to command
bergen, barg, geborgen                to salvage
nehmen, nahm, genommen          to take
sprechen, sprach, gesprochen     to speak
stechen, stach, gestochen             to stick
stehlen, stahl, gestohlen                 to steal
treffen, traf, getroffen                       to meet
verderben, verdarb, verdoben       to spoil
werben, warb, geworben                to recruit
and others

Peter


From W.Schulze at lrz.uni-muenchen.de  Tue Jul 20 10:15:16 1999
From: W.Schulze at lrz.uni-muenchen.de (Wolfgang Schulze)
Date: Tue, 20 Jul 1999 12:15:16 +0200
Subject: accusative and ergative languages
Message-ID: <TUE.20.JUL.1999.121516.0200.>

Dear Pat and IEists,

before turning to some points addressed by Pat, let me first say
something more general: Having read most of the contributions to this
list on the subject 'accusativity- ergativity' I got /and still have)
the impression that some of these contributions exhibit a considerable
ammount of redundancy regarding the arguments already addressed. This
may be due to the fact that semtimes new 'sub-threads' are opened or
special data discussed that provoke arguments already discussed in
another context. This fact renders the contributions sometimes rather
idiosyncratic and makes it difficult to react on ALL of them in a
non-redundant way. Sometimes, important arguments are hidden in such
'sub-threads' and hence do not receive the attendance they deserve. In
order to render our discussion a bit more straight-forward I would
propose that all contributors distribute a short (general) statement
about what they think is crucial in our discussion. Perhaps it is
helpfull if people (among others) refer to the following questions:

1. What is YOUR over-all definition of 'accusativity'?
2. What is YOU over-all definition of 'ergativity'?
3. Do YOU think of ACC vs. ERG as a parallelly weighted, polar
opposition? Or do YOU propose that one of the poles is more 'natural'
than the other? If yes: What is YOUR notion of 'naturalness', and how do
YOU substantiate your claim?
4. What is the distinction between 'morphological' and 'syntactic'
accusativity/ergativity according to YOUR paradigm?
5. Which role does 'markedness' play in YOUR arguments?
6. How do YOU define 'subject'?
7. What is - in YOUR mind - the correlation between case marking and
agreement with respect to ACC/ERG?
8. How do YOU classify a language with respect to ACC/ERG in case a) AGR
not present, b) CASE is not present, c) neither AGR nor CASE are
present?
9. Wich role des word order play with respect to a) core arguments, b)
peripheric (backgrounded) arguments in diathesis?
10. Describe YOUR scenario(s) of how a) ERG paradigms emerge, b) ACC
paradigms can emerge (cf. 3)!
11. What are - in YOUR opinion - the most crucial aspects of ACC/ERG: a)
semantic (conceptual), b) syntactic (sentence organization), c)
pragmatic (discourse organization, information flow etc.)?
12. Do YOU claim that one (or more) of the parameters mentioned in (11)
are specifically relevant for either ACC or ERG (e.g., ERG _> semantic,
ACC -> pragmatic etc.)?
13. Which role do split systems play in your arguments (Split-S,
Fluid-S, Split-O, Fluid-O)?
14. How do you define ACC/ERG in polypersonal / poly-congruent languages
without case marking? (cf. (7))?
15. Which are the languages that YOU use in your arguments? Do you use
a) reference books, b) texts, c) informants?
16. Do YOU use a special theoretical frame work in order to substantiate
your claims? If yes, which one?

*********************

Now, let me finally turn to some claims by Pat:

"Patrick C. Ryan" schrieb:

Wolfgang wrote:

>> Hence, there are NO "ergative (or "accusative") languages" or
>> only, if you use this term in a very informal sense.

> Pat writes:

> In the sense you are using these, they seem to be of little value in
> describing anything.

I can only repeat what I have said in my earlier postings: ACC and ERG
are nothing by the structural (and structured) reaction to more general
principles of language based information processing. The polycentric
architecture of language systems that encode these principles allows
that the individual centers of this polycentric cluster react
differently on these principles. Here, I cannot elaborate the underlying
frame work which is labeled "Grammar of Scenes and Sceanrios" (GSS) and
documented in Schulze 1998, chapter IV, but let me briefly say that I
propose a (more or less) universal cause-effect 'vector' (C->E) itself
metaphorized from underlying figure-ground relations (F->G) to be one of
the most dominant principles of information processing. This vector can
be weighted which leads to a continnum (here shortened) C->0(e) C->e >
C->E > c->E > 0(c)->E (capitals represent heavy domains, small letters
represent light domains). An ACC strategy would be to behave in a C->e
sense, an ERG strategy would infer c->E. Note that these vector
representations are NOT (in themselves) language specific or sensitive
for specuial (linguistic) categories! Different categories may BEHAVE
differently with respect to this continuum, regardless their own
architectural make-up. ANYTHING in a language system may be sensitive
for the AEC (accusative ergative continuum) as long as it is relevant
for encoding the cause-effect vector (and its derivations). Hence, CASE
may play a role just as AGR, word order, subject assignment,
topicalization, discourse cohesion, co/subordination, paradigmatization
of speech act participants and much more. But they may play their roles
DIFFERENTLY! The description of their roles heavily depends from the
diachrony of the paradigm in question, its formal architecture as well
as its integration in aa co-paradigmatic context ('structural coupling'
in a broader sense). From this it follows that the individual centers of
ALL language systems have to react upon the universal demands of the
C->E vector, disregarding their internal architecture. Only IF ALL
relevant centers (and they are many, I grant!) behave in one direction,
THEN we are allowed to call the language system (or better, its
Operating System) ACC or ERG. However, it is a much more difficult task
to describe intermediate states that allow ACC in some parts of the
Operating System, and ERG in some others. Here, we have to establish a
(motivated!) hierarchy first of co-paradigmtic structures (such as CASE
and AGR, CASE and word order, AGR and Personality, AGR and Noun Classes,
to name only some). In such structures, one parts sometimes is more
dominant than the other with respect to its behavior on the AEC. If we
can describe such dominant behavior we can refer to the whole structure
as either more ACC or ERG. In a second step we have to go on describing
the higher levels of this hierarchy (which itself should find an
adequate linguistic explanation based on an appropriate language
theory). Finally (and idealiter) we would arrive at a term that would
describe the functional dominance of one of the poles on the AEC with
respect to an Operating System (not a language system) in toto. Only
the, and I stress, only THEN we are allowed to use the term ergative or
accusative with respect to an Operating System (for which 'ergative' or
'accusative language' would be an informal label).

> Wolfgang wrote further:

> Pat writes:

> I am sorry that I do not agree with the validity of this distinction (actant
> vs. agent). For me, 'actant' is 'agent'. Perhaps you can explain the
> difference.

In terms of Functinal Grammars (as well as in GSS) 'actants' refer to
ALL such linguistic expressions that encode a referential entity in a
clause. Hence, in a senetcne such as 'I met John several times in
Chicago', 'I', 'John', 'times', and 'Chicago' are (abstract) actants
that play different roles in the scene. But only 'I' is a linguistic
agent, whereas John plays the role of a patient etc. Note that 'agent'
and 'patient' are labels for semantic hyperroles (or macroroles in the
sense of Foley/VanValin). I think that such a distinction is very
helpfull. It is based on strong theoretical arguments and helps to avoid
many false or at least problematic generalizations. A much more
controverse (and much more difficult question is to define the labels
'subjective' (S), 'agentive' (A), and 'objective' (O) which should not
be immediately equated to neither 'subject'/'object' nor to
'agent'/'patient'. S, A, and O are highly abstract terms that describe
more structural than semantic or syntatic properties.

> Wolfgang wrote:

>> The fact, however, is that many 'ergative' languages lack an
>> antipassive. For instance, there are nearly 30 East Caucasian languages
>> all of them using some ergative strategies in at least parts of their
>> operating systems. But only a handfull of them (five or six, to be
>> precise) have true antipassives (only one has some kind of
>> "pseudo-passive").

> Pat writes:

> Are you asserting that the majority of ergative languages do not have
> anti-passives?

I do not assert anything in the sense of 'ALL language have...'. Even
the claim 'the majority of ergative languages (sic!) have...' is rather
suspect to me. What I said is that in those 'ERG systems' I looked at
(about 200) antipassives are ratgher the exception than the norm.

> Pat writes:

> As for the far-reaching conclusions of Dixon based on discourse conhesion
> strategies, in my opinion they are flawed because these strategies are
> purely conventional.

Language systems ARE conventional! This is one of the major points of
language tradition and L1 acquisition (despite of minimalism etc.).
Discourse probably is one of the most important factors in the
emergence, organization, and dynamics of language systems. We should not
refer to the abstract notion of context-free 'sentences' that would be
responsible for for grammatical 'events'. Such a view stems from the
tradition of Classical Philosophy which is an INTERPRETATION of what
goes on language. Today, we have become used to think of language in
single sentences, to brak them up the way we do etc. But this is an
analytic tradition, not part of the ontology of language itself, which
is much more synthetic in ature than we are used to think. - A sentence
does not function but in its co-text (as well as in its con-text). All
sentence internal strategies used to be embedded in the techniques of
co(n)textualization. No wonder, that ACC and ERG also work in this
direction (though they may appear as more 'autonomous',
sentence-internal mechanisms secondarily, especially if a language
system as developed separate means to indicate discourse cohesion).

> In English, I can say: "John hit me, and he went away" or "I hit John and he
> went away". What is the discourse cohesion strategy for English?

See the standard literature on this (its counts legion!).

> Pat writes (on Lak, East Caucasian):

> I have to plead ignorance of Lak however, I find an analysis of an otherwise
> completely ergative sentence as having an ACC word-order impossible. I
> believe this opinion rests on a false analysis of word-order significance.

I tried to show that in Lak only some sentence patterns show 'complete
ergativity'. Most of them sho a mixed paradigmatic organization with
respect to moprhology (CASE, AGR). WOrd order is another VERY imporant
aspect of the AEC. Consider e.g. a language that ahs canonical

#SV
#AOV

(# = sentence boundary). Here, S behave like A with respect to #, hence
it is ACC. In #SV vs. #OAV it is S=O which indicates ERG behavior. The
problem of actance serialization is also addressed in polypersonal
systems (without CASE, e.g, Abxaz, Lakotha etc.). It IS a very important
indicator for ACC, ERG, no question. Anything else would refer to the
linguistic tradition of say 60 years ago (Bloomfieldian tradition).

> Wolfgang continued (on Lak):

>> But if you say "I am surely hitting you (plural)", you get:

>> na b-at-la-ti-s:a-ra zu
>> I:ABS I:PL-hit1-DUR-hit2-ASS-SAP:SG you:PL:ABS

>> Here we have:

>> ACC with respect to word order
>> ERG with respect to class agreement
>> ACC (or neutral) with respect to case marking)
>> ACC with respect to SAP agreement (-ra is triggered by 'I:ABS').

> Pat writes:

> So, agent marked ABS and patient marked ABS is, according to you, ACC with
> respect to case marking? Sorry, not convincing at all! And I think the SAP
> difference can be explained as indicating seriality of the patients involved
> as recipients of the action by *one* agent.

I said ACC (or neutral!) with respect to case marking! That means that A
lacks ERG marking (which is OK with Silverstein). Semantically, it means
that S is encoded like A (which is ACC). That fact that O is NOT in an
accusative-like case does not argue against this assumption, because
CASE is here structrually coupled with ADGR which allows us to classify
AGR as ACC *here*. The SAP difference in Lak cannot be explained in the
sense Pat proposes, Lak totally lacks such strategies.

> Wolfgang wrote further:

>> Now, please tell me: Is Lak an 'ergative' or an 'accusative' language?
>> [Please note that I did not include (among others) strategies of
>> discourse cohesion, reflexivization and logophization].

> Pat writes:

> By these examples, I would tell you that Lak is 'ergative'; and that
> accusativity is not demonstrable from these examples.

If you refer to the standard (morphological), but rather obsolete
interpretation of ERG you may be right (but not for SAP NPs). But,
fortunately, the ACC/ERG typology has freed itself from such a narrow
interpretation of ACC/ERG which is nothing but a very small excerpt from
the over-all typology.

> Pat wrote previously:

>>> I think it is likelier that, because of perceived greater animacy (or
>>> definiteness), pronouns have a different method of marking that can still
>>> be interpreted within an ergative context.

> Wolfgang wrote:

>> This a (very simplified) 'on-dit' that stems from the earlier version of
>> the Silverstsein hierachy. Again, we have to deal with the question,
>> whether a 'pronoun' (I guess you mean some kind of 'personal pronouns')
>> can behave 'ergatively' or 'accusatively'. The list below gives you a
>> selection of SAP case marking in East Caucasian languages with respect
>> to ABS/ERG:

>> ABS vs. ERG ABS = ERG
>> ALL ---
>> Singular Plural
>> Plural Singular
>> 1.Incl. Rest
>> 1:SG Rest
>> 2:SG Rest
>> 1:SG/PL Rest
>> --- ALL

>> This list (aspects of personal agreement NOT included!) shows that SAP
>> pronouns may behave different within the same paradigm. Any
>> generalization like that  one quoted above does not help to convey for
>> these data...

> Pat writes:

> I do not have a reference book for Lak so that my hands are somewhat tied.
> But, I have found that paradigms are often inconsistent in ways that reflect
> earlier lost phonological changes, or other lost schemata. The locative
> plural terminations of IE are certainly not, in origin, terminations of the
> locative plural. Etc.

Nothing the like! The list I gave refers to what can be described for
East Caucasian languages in toto! Most of these paradigms are
functionally motivated, I grant (schulze in PKK 2 (Schulze 1999) will
tell you the whole story).

Wolfgang

[Please note new phone number (office) :+89-2180 5343]
___________________________________
| Prof. Dr. Wolfgang Schulze
| Institut fuer Allgemeine und Indogermanische Sprachwissenschaft
| Ludwig-Maximilians-Universitaet Muenchen
| Geschwister-Scholl-Platz 1
| D-80539 Muenchen
| Tel:	+89-21802486 (secr.)
|      	+89-21805343 (office) NEW ! NEW !
| Fax:	+89-21805345
| Email: W.Schulze at lrz.uni-muenchen.de
| http://www.lrz-muenchen.de/~wschulze/
_____________________________________________________


From vidynath at math.ohio-state.edu  Tue Jul 20 11:29:04 1999
From: vidynath at math.ohio-state.edu (Vidhyanath Rao)
Date: Tue, 20 Jul 1999 07:29:04 -0400
Subject: accusative and ergative languages
Message-ID: <TUE.20.JUL.1999.072904.0400.>

Jens Elmegaard Rasmussen <jer at cphling.dk> wrote:

> I'd say the cases I have seen of ergativity in Indic and Iranian languages
> so clearly reflect underlying/earlier passive circumlocutions that
> controversy is absurd.

No!

I have already pointed out the problem of distinguishing resultatives and
passives. The ta-adjective is resultative in Vedic (Jamison, IIJ 198?) and
Early Pali (Hendrikson, Infinite verb forms of Pali). And it patterns
ergatively as resultative participles often do. The last part has been known
forever. In English, see Speyer's ``Sanskrit Syntax'', will find the
following: ``Of the participles in -ta the great majority have a passive
meaning, hence it is customary to call the whole class the passive
participle of the past. But some others are not passives, but intransitives,
as gata (gone), m.rta (died) [rather dead, VKR] bhinna (split). Some again
may even be transitive actives, as pi:ta (having drunk) [better drunk, but
without the restriction to a special meaning in English] ...'' [para 360, p.
280].  Speyer goes on to note a:ru:d.ha has active meaning more commonly.
This is not how we expect the passive to behave. But for resultative, it is
understandable. The ergative patterning is based on pragmatics. But in a
culture that considers it more worthy of note whether a man is mounted on a
horse or a vehicle than whether a horse or a vehicle is carrying someone, it
makes sense to use a:ru:d.ha in the active sense.

The difference is unmistakable in the following:

(1) ra:mo 's'vam a:ru:d.hah.
    R is mounted on [a] horse.

(2) ra:men.a+as'vam a:ruhyate
   [A] horse is being mounted by R.

(3) *ra:mo 's'vam a:ruhyate

(1) and (2) are quite grammatical and examples of easy to come by. (3) does
not occur and is underivable in traditional grammar. As I explained above,
this is understandable if a:ru:d.ha is resultative. If we take the passive
view, how do we explain the fact that (1) is acceptable but (3) is not?
Without such an explanation, it is far from absurd to contest the passive
interpreation.


From adahyl at cphling.dk  Wed Jul 21 16:59:02 1999
From: adahyl at cphling.dk (Adam Hyllested)
Date: Wed, 21 Jul 1999 18:59:02 +0200
Subject: accusative and ergative languages
In-Reply-To: <Pine.GSO.3.95.990716160742.15501B-100000@unagi.cis.upenn.edu>
Message-ID: <WED.21.JUL.1999.185902.0200.>

On Fri, 16 Jul 1999, Sean Crist wrote:

> Somebody (I've lost track of who) wrote:

>>> Yes, but if a new verb were formed, e.g. webben (from English 'web'), there
>>> is no possibility that it would be conjugated webben, wabb, gewobben.

> This sort of thing actually _does_ happen sometimes; it just isn't that
> common.  The English verb "strive" is a loan from Old French; but it is
> often declined as if it were an old Class I ablauting strong verb (strove,
> striven).  So I'd correct "no possibility" to "much less likely".

Another example is the Danish verb <skrive> "to write",
ultimately from latin <scribere>, but always conjugated as an
ablauting strong verb: pres. <skriver>, pret. <skrev> (and not
<**skrivede>. However, this way of conjugating is no longer productive.

Adam Hyllested


From larryt at cogs.susx.ac.uk  Mon Jul 19 09:31:41 1999
From: larryt at cogs.susx.ac.uk (Larry Trask)
Date: Mon, 19 Jul 1999 10:31:41 +0100
Subject: Interpreting ergative sentences
In-Reply-To: <007201becdef$67235840$6701703e@edsel>
Message-ID: <MON.19.JUL.1999.103141.0100.>

On Wed, 14 Jul 1999, Eduard Selleslagh wrote:

> Speaking about Castilian, I would like to have your views on a peculiarity,
> or rather a tendency, that is still productive at least in popular speech
> ('le-ismo'), namely the tendency to use the indirect subject form where all
> other Latin (and other West-European) languages use the direct object form,
> and almost exclusively with persons (animate), e.g. "le ví" ("le vi' "), "I
> saw him".

> It has sometimes be suggested that this was a substrate influence from
> Basque, an ergative language (that lacks an accusative, of course), but
> opinions are extremely divided on this subject. 

Indeed.  I have seen this suggestion in print, and I've replied to it in
print.  The observation is that Castilian fails to distinguish direct
and indirect objects as sharply as other Romance languages.  The
suggestion is that this results from Basque substrate influence.  But
this suggestion would only make sense if Basque were a language that
also failed to distinguish direct and indirect objects sharply.  And
it's not.

In Basque, direct and indirect objects are sharply and absolutely
distinguished in all respects.  The two differ in case-marking, in
verbal agreement and in ability to be passivized.  This is true in all
circumstances without exception.  In fact, I would say that Basque
distinguishes direct and indirect objects more sharply than any Romance
language I know of, and more sharply than English.  (English has
sentences like `My wife gave me this book', in which it is a moot point
what kind of object`me' might be.)

Accordingly, I can't for the life of me see how Basque substrate
influence could be invoked to account for the Castilian facts.  True,
Basque has no accusative case, but it does have a very well-defined
class of direct objects and a very well-defined class of indirect
objects, and these two classes are entirely distinct in every
morphological and syntactic respect I can think of.  Given what I've
read about objects in the literature, I would say that, if anything,
Basque is rather unusual in its absolute distinction between the two
kinds of object.

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk


From larryt at cogs.susx.ac.uk  Mon Jul 19 11:29:14 1999
From: larryt at cogs.susx.ac.uk (Larry Trask)
Date: Mon, 19 Jul 1999 12:29:14 +0100
Subject: Ergative & Basque
In-Reply-To: <Pine.SO4.4.05.9907161834110.6415-100000@rask>
Message-ID: <MON.19.JUL.1999.122914.0100.>

On Sat, 17 Jul 1999, Jens Elmegaard Rasmussen wrote:

[on my Basque example with `have' as the transitive auxiliary]

> If there is no passive expressed as belonging to the agent in this,
> what then is the "have" verb doing here?

The Basque construction is parallel in form to the `have' perfects and
past-tenses found in a number of Romance and Germanic languages (at
least).  In English, `I have drunk the wine' is still strictly a
perfect, with no past-tense reading possible.  In French, the similarly
constructed <J'ai bu le vin> is primarily an ordinary past tense, though
a perfect reading is possible.  Other languages exhibit various
intermediate stages here.  Castilian Spanish <(Yo) he bebido el vino> is
interesting in that it functions both as a perfect and as a hodiernal
past (earlier today).  Basque <Ardoa edan dut> works like Castilian.

All these languages also allow a past-tense auxiliary, as in English
`I had drunk the wine', but again the precise value of this varies from
language to language.  In Basque it functions both as the ordinary past
tense (before today) and as a pluperfect.

> Can't the underlying construction be analyzed in a sensible way at
> all?

Well, I'm not sure what you mean by "a sensible way".  What would you
regard as a sensible analysis of English `I have drunk the wine'?

> A silly question from a complete outsider: is the translation "have"
> internally motivated? Can it express the "having" of anything other
> than a "participled" object?

Yes.  In all varieties of Basque, the auxiliary used in constructing
periphrastic forms of transitive verbs is the defective verb *<edun>,
which has no non-finite forms and whose non-finite forms are supplied
suppletively by other verbs.  In the east, this *<edun> is the ordinary
lexical verb `have', as in `I have blue eyes' or `I have a new car'.

Eastern varieties have another verb, <eduki>, which means `hold', `hold
on to', `have in one's hand', `grasp', `clutch'.  In western varieties,
however, *<edun> is generally specialized as the transitive auxiliary,
and <eduki> is the ordinary lexical verb for `have'.

Now, since *<edun> is also recorded for `have' in early texts in the
west, and since it is still possible for `have' in elevated styles in
the west today, we may reasonably surmise that *<edun> was once the
ordinary lexical verb for `have' everywhere, and that its replacement by
<eduki> in the west is an innovation there.

We may further surmise that this western innovation is a calque on
Castilian.  In Castilian, the inherited lexical verb for `have' is
<haber>, but this is now entirely specialized as an auxiliary.  The verb
<tener>, which originally meant `hold, hold on to, grasp', has become
the ordinary lexical verb for `have'.

This proposal of a calque on Castilian can be supported by another case.
In all varieties of Basque, the verb used as an auxiliary for
constructing periphrastic forms of intransitive verbs is <izan>.
Now, in eastern varieties, this <izan> is also the ordinary copular verb
`be', as in `I am a teacher' or `Your books are on the table'.

However, eastern varieties have a second verb, <egon>, whose earlier
sense is `wait', `stay', `remain'.  At least in the imperative, it still
has this sense in all varieties.  In the east, though, it has acquired a
second sense, that of copular `be' *in the following circumstances
only*: the subject is animate, and the predicate is locative or
comitative (the Basque comitative itself derives from a locative
phrase).  So, in the east, `Mary is in the kitchen', `The men are in the
fields' and `Mary is with John' all require <egon>, not <izan>.

In western varieties, however, the use of <egon> has been generalized:
it is used for `be' in *all* locative expressions, and also with all
predicates denoting states or conditions.  So, in western varieties,
`Your books are on the table', `The Guggenheim Museum is in Bilbao',
`Mary is happy', `Mary is asleep', `Mary is drunk', and so on, require
<egon>, whereas eastern varieties use <izan>.

This western usage is clearly calqued on the famous Castilian
distinction between <ser> `be' (unmarked) and <estar> `be' (in a place
or in a state).  Wherever Castilian uses <estar>, western Basque uses
<egon>.  This is true even in idiosyncratic cases.  For example,
Castilian expresses `He's dead' as <Esta' muerto>, with <estar>, and
western Basque likewise has <Hil dago>, with <egon>, while eastern
Basque has <Hila da>, with <izan>.

And this, I think, is as far back as we can go in tracing the prehistory
of the Basque periphrastic verb-forms.  As I remarked earlier, we have
no way of knowing whether these things originated as calques on Romance
or whether they are independent creations in Basque, from an unknown
source.  All I can tell you is that the modern periphrastic forms were
clearly already established in Basque by the tenth century, when we find
the first recorded verb-forms.

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk


From connolly at memphis.edu  Mon Jul 19 20:35:22 1999
From: connolly at memphis.edu (Leo A. Connolly)
Date: Mon, 19 Jul 1999 15:35:22 -0500
Subject: accusative and ergative languages
Message-ID: <MON.19.JUL.1999.153522.0500.>

Larry Trask wrote, replying to Pat:

> But briefly: if a transitive sentence in an ergative language were
> "really" a passive, then its absolutive NP would be the subject and
> would exhibit subject properties.  But this not the case in most of the
> ergative languages I have heard about.  In spite of the ergative
> case-marking and/or verbal agreement, it is usually the *ergative* NP
> which exhibits subject properties, not the absolutive NP.

Agreed -- and yet not quite agreed.  Are these really "subject
properties"?  I freely acknowledge that that's what they're called, but
for mainly
for lack of any better name in normal linguistic terminology.

It might be better to take the tack that I did in my recent post "On
interpreting ergative languages": that one must distinguist between a
"morphological subject" (the absolutive or nominative, or positional
equivalent in languages such as English), and the "highest-ranking NP"
in a hierarchy of the Case Grammar sort.  Then it turns out that even in
accusative languages, some properties adhere to the morphological
subject, others to the HRNP, the split varying greatly from one language
to another.  In most ergative languages (which can then be defined
simply as languages with extensive ergative casemarking), it is the HRNP
that has all or most of the syntactic "subject properties".  But there
is no particular reason to call it a "subject".

True enough: to some extent, this is a matter of terminology.  But I
think it matters in this sense: unlike the morphological subject, the
HRNP is not constituted, but is given by the Case Hierarchy; it may,
however, be demoted by use of a passive construction.  "Active"
languages are then explained naturally: they have no morphological
subject, only the HRNP, whose realization notoriously varies.

	[stuff omitted]

> There are unusual languages, of course.  In Dyirbal, and in
> Nass-Gitksan, the absolutive NP in a transitive sentence shows at least
> some subject properties, while the ergative NP does not show the same
> properties -- though it does show other subject properties.  This is the
> phenomenon we call `syntactic ergativity'.  But I know of no language
> which is syntactically ergative without exception, and I know of few
> languages which are syntactically ergative at all.

I have gone through Dixon's book on Dyirbal pretty thoroughly and don't
recall any accusative or active syntax, only a few accusative pronouns
(which exhibit ergative syntax).

> The "passive" view of ergative languages in general is indefensible.

Absolutely agreed!

Leo


From proto-language at email.msn.com  Mon Jul 19 04:17:28 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Sun, 18 Jul 1999 23:17:28 -0500
Subject: Recoverability
Message-ID: <SUN.18.JUL.1999.231728.0500.>

Dear Sean and IEists:

 ----- Original Message -----
From: Sean Crist <kurisuto at unagi.cis.upenn.edu>
Sent: Thursday, July 15, 1999 11:37 PM

> You need multiple examples of sound correspondences to be able to conduct
> the Comparative Method at all; and when the the cognates become as
> rarefied as they are this time-depth, the likelihood of having access to
> an adequate number of examples to work out the relevant sound changes
> becomes proportionately smaller, eventually reaching what for practical
> purposes is an impossibility.

Pat interjects:

"Rarefied" is a characterization only possible when one refuses to look at
the actual facts. An adequate number of examples to establish cognation is
furnished in the studies on Uralic at my website:

http://www.geocities.com/Athens/Forum/2803

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From jer at cphling.dk  Tue Jul 20 01:38:14 1999
From: jer at cphling.dk (Jens Elmegaard Rasmussen)
Date: Tue, 20 Jul 1999 03:38:14 +0200
Subject: Recoverability
In-Reply-To: <Pine.GSO.3.95.990715232942.17263C-100000@unagi.cis.upenn.edu>
Message-ID: <TUE.20.JUL.1999.033814.0200.>

On Fri, 16 Jul 1999, Sean Crist wrote:

> [...]
> I think the point was something more like this.  Suppose for the sake of
> argument that there is in fact a genetic relationship between
> Proto-Indo-European and Proto-Uralic (to take one possible example). Even
> if this were true, and given the evidence which we actually do have at
> hand, would we expect to be able to _show_ that there is such a relation?

> Here, the problem _is_ like trying to reconstruct PIE on the basis of only
> modern English and Albanian.

Yes it is - and one would not get very far that way. One just might
discover the relationship, though: Alb. jam 'am', e"shte" 'is', nate"
'night', dere" 'door', tre, tri 'three', me" 'me', ju 'you'. If English
had had some proper morphology, one would have been able to go further.
Collinder and Cop (the latter with c-hachek) did that for "Indo-Uralisch".
Now of course large-scale comparison is not restricted to IE and Uralic,
but includes quite a few extra branches into Nostratic - and the more
there is to draw on, the greater the chance of finding links. I think it's
looking good.

(Sorry if I shifted the focus of the question in my posting.)

Jens


From connolly at memphis.edu  Mon Jul 19 20:39:35 1999
From: connolly at memphis.edu (Leo A. Connolly)
Date: Mon, 19 Jul 1999 15:39:35 -0500
Subject: accusative and ergative languages
Message-ID: <MON.19.JUL.1999.153935.0500.>

Sean Crist wrote:

> Somebody (I've lost track of who) wrote:

>>> Yes, but if a new verb were formed, e.g. webben (from English 'web'), there
>>> is no possibility that it would be conjugated webben, wabb, gewobben.

> This sort of thing actually _does_ happen sometimes; it just isn't that
> common.  The English verb "strive" is a loan from Old French; but it is
> often declined as if it were an old Class I ablauting strong verb (strove,
> striven).  So I'd correct "no possibility" to "much less likely".

Well, maybe.  Trouble is, the word is of Germanic origin (cf. German
_streben_), so the strong inflexion *might* actually be original but
unattested in OE.

Leo


From proto-language at email.msn.com  Mon Jul 19 04:31:05 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Sun, 18 Jul 1999 23:31:05 -0500
Subject: PIE vs. Proto-World (Proto-Language)
Message-ID: <SUN.18.JUL.1999.233105.0500.>

Dear Joat and IEists:

 ----- Original Message -----
From: <JoatSimeon at aol.com>
Sent: Friday, July 16, 1999 11:53 AM

 Joat wrote:

> -- with minor exceptions (such as "kuku"), words _are_ arbitrary sound
> assemblages.

Pat writes:

Why do you not explain to us all why that is true? I am firmly convinced
that it is unequivocally incorrect.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From proto-language at email.msn.com  Mon Jul 19 05:21:18 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Mon, 19 Jul 1999 00:21:18 -0500
Subject: PIE vs. Proto-World (Proto-Language)
Message-ID: <MON.19.JUL.1999.002118.0500.>

Dear Ralf-Stefan and IEists:

 ----- Original Message -----
From: Ralf-Stefan Georg <Georg at home.ivm.de>
Sent: Friday, July 16, 1999 4:28 AM

Ralf-Stefan wrote:

> And, if I may insert this much to Pat's chagrin, the notion that human
> language is a monogenetic phenomenon is aprioristic ideology. Even the
> notion that every known language, as Basque, Burushaski or whatnot, has to
> be related to some other language is ideology. "Having consonants" may be a
> universal feature of human language, "Being related to some other lg."
> simply is not.

Pat responds:

There is no *tangible* way for us to ever know whether languages arose
monogenetically or polygenetically however most linguists, even when they
deny its recoverability, have correctly weighed the odds of mono- vs.
polygenesis, and subscribe to monogenesis. If a probablistically calculated
hypothesis is "ideology", then everything done in historical linguistics is
"ideology".

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From proto-language at email.msn.com  Mon Jul 19 05:48:42 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Mon, 19 Jul 1999 00:48:42 -0500
Subject: PIE vs. Proto-World (Proto-Language)
Message-ID: <MON.19.JUL.1999.004842.0500.>

Dear Larry and IEists:

 ----- Original Message -----
From: Larry Trask <larryt at cogs.susx.ac.uk>
Sent: Friday, July 16, 1999 10:06 AM

Larry wrote:

> And the same goes for "Proto-World", or whatever.  Until a persuasive
> case has been made that *all* of the world's 6000 or so known languages
> are genuinely related, there is no point in attempting a
> "reconstruction" of "Proto-World".  The result of such a rash attempt
> can be no more than legerdemain.  Of course you can show that this, that
> and the other *might* have a common ancestor, but you can do this in
> countless entirely different ways, none of them superior to any other,
> and you cannot show that these things really *are* related.

Pat writes:

The "persuasive case" has already been made. Monogenesis is much likelier
than polygenesis. And you have written so yourself!

What mitigates against it is incorrect ideas of vocabulary (*CVC-root) loss,
mistaken applications of Kinderlallsprache, loan specialists that would make
the Rothschilds wild with envy, and fuzzy concepts like your favorite
"expressive" --- all of which seeming to foredoom its recoverability.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From Georg at home.ivm.de  Mon Jul 19 09:36:29 1999
From: Georg at home.ivm.de (Ralf-Stefan Georg)
Date: Mon, 19 Jul 1999 11:36:29 +0200
Subject: PIE vs. Proto-World (Proto-Language)
In-Reply-To: <001801becf8b$dc828e60$869ffad0@patrickcryan>
Message-ID: <MON.19.JUL.1999.113629.0200.>

>Pat interjects:

>Here I think you are dangerously introducing the mistaken terminology of the
>opposing argument. I do *not* look for "look-alikes"; I only am interested
>in "cognates" the phonological forms of which can be supported through
>multiple comparisons. For example, one set of interesting IE and Sumerian
>"cognates" shows IE *-wey- = Sumerian -g{~}-; interestingly, this
>development *is* found in *some* IE languages, like nearby Armenian.

"interestingly" ? Why interestingly ? The only way this could be
"interesting" would, imho, be to base a claim of Leskienian solidaric
innovation on it, leading to a Sumero-Armenian subgroup of Indo-European.
Very interesting.
Or some kind of areal phenomenon, potentially interesting (this time
without irony). But: supposing for a split-second that IE and Sumerian are
somehow related, the /w/ --> /g/ shift would be something very old, right ?
Now, the Armenian /w/ ---> /g/ shift is young, since known loanwords
participate in it (so young actually, that Kartvelian managed to preserve
the intermediary stage /gw/ in some of those LWs). And of course, even if
chronology were no obstacle, the whole thing still begs the question of
Sumero-IE, but I understand that this is a) not a topic for this list and
b) that I don't stand the slightest shred of a chance to talk you out of
it; so I don't try.

St.G.

Stefan Georg
Heerstrasse 7
D-53111 Bonn
FRG
+49-228-69-13-32


From rmccalli at sunmuw1.MUW.Edu  Mon Jul 19 21:27:49 1999
From: rmccalli at sunmuw1.MUW.Edu (Rick Mc Callister)
Date: Mon, 19 Jul 1999 16:27:49 -0500
Subject: PIE vs. Proto-World (Proto-Language)
In-Reply-To: <Pine.GSO.4.03.9907161556400.19800-100000@rsunx.crn.cogs.susx.ac.uk>
Message-ID: <MON.19.JUL.1999.162749.0500.>

	You raise the questions of what is relationship and what is the
difference between language and speech.
	All languages are related in the same sense that all humans are
related. All humans settled on oral speech rather than sign language or
flatulence for communication. Given that all human brains are wired for
speech in the same way, this suggests that there is a degree of
relationship.
	The question is whether the time span between speech and language
allowed languages to develop separately or whether quickly developed from
that single event.
	I don't think this question can be answered in absolute terms but
it may possibly be answered in practical terms. Given that modern humans
left Africa about 100,000 years ago, there is a good chance that all of
these are ultimately related. This presupposes that language had developed
before humans left Africa. DNA studies obtensibly show that non-African
humans seem to go back a single population distinct from Africans. While
one cannot prove that the members of this group spoke related languages,
it's more likely than not. If one establishes that non-African languages
are related, only Nilo-Saharan, Niger-Kordofanian, Khoisan languages [and
possibly Afro-Asiatic] are left.
	The next step is to prove this, and I'll leave that all of you :>

[snip]

>And the same goes for "Proto-World", or whatever.  Until a persuasive
>case has been made that *all* of the world's 6000 or so known languages
>are genuinely related, there is no point in attempting a
>"reconstruction" of "Proto-World".  The result of such a rash attempt
>can be no more than legerdemain.  Of course you can show that this, that
>and the other *might* have a common ancestor, but you can do this in
>countless entirely different ways, none of them superior to any other,
>and you cannot show that these things really *are* related.

>Larry Trask

[ moderator snip ]

Rick Mc Callister
W-1634
Mississippi University for Women
Columbus MS 39701


From adahyl at cphling.dk  Wed Jul 21 15:57:04 1999
From: adahyl at cphling.dk (Adam Hyllested)
Date: Wed, 21 Jul 1999 17:57:04 +0200
Subject: PIE vs. Proto-World (Proto-Language)
In-Reply-To: <l03102800b3b4aa9358c0@[195.247.226.14]>
Message-ID: <WED.21.JUL.1999.175704.0200.>

On Fri, 16 Jul 1999, Ralf-Stefan Georg wrote:

> "Having consonants" may be a
> universal feature of human language, "Being related to some other lg."
> simply is not.

A fascinating thought:
Imagine "Being related to some other lg." WAS actually a universal feature
of (spoken) human language, wouldn't it then be the only true universal?
In other words, if all spoken languages should prove to be derived from
one common source, would we then be able to distinguish true language
universals from other features shared by all languages ONLY because of the
fact that these features were present in "proto-world"? Language typology
would still be a great help in the work of reconstructing proto-languages,
but would it prove anything about the nature of spoken human language?

The above are only questions, not statements...

Adam Hyllested


From adahyl at cphling.dk  Wed Jul 21 16:17:15 1999
From: adahyl at cphling.dk (Adam Hyllested)
Date: Wed, 21 Jul 1999 18:17:15 +0200
Subject: PIE vs. Proto-World (Proto-Language)
In-Reply-To: <Pine.GSO.4.03.9907161556400.19800-100000@rsunx.crn.cogs.susx.ac.uk>
Message-ID: <WED.21.JUL.1999.181715.0200.>

On Fri, 16 Jul 1999, Larry Trask wrote:

> But you can't just pick some languages that catch your eye and then
> "reconstruct" a common ancestor for them.  That is, you can't pick, say,
> Zulu, Sumerian and Korean and "reconstruct" Proto-ZSK.  You must first
> make a good case that a common ancestor for these languages is
> preferable to any other scenario.

Exactly. This is why people shouldn't always neglect distant language
relationship using the argument that no sound correspondences have been
set up. Every scientific project has a primitive beginning.

Adam Hyllested


From adahyl at cphling.dk  Wed Jul 21 17:14:09 1999
From: adahyl at cphling.dk (Adam Hyllested)
Date: Wed, 21 Jul 1999 19:14:09 +0200
Subject: PIE vs. Proto-World (Proto-Language)
In-Reply-To: <1e1b93d4.24c0bd9b@aol.com>
Message-ID: <WED.21.JUL.1999.191409.0200.>

On Fri, 16 Jul 1999 JoatSimeon at aol.com wrote:

> Our picture of PIE would be very different, and much less
> detailed, if we had no written records of extinct languages.

> Hell, the original insight that the IE languages were derived from a common
> ancestor was made by comparting Sanskrit with Classical Greek and Latin --
> all extinct languages!

> The link between Sanskrit and Latin or Classical Greek is so close that any
> layman can see it; whole phrases are nearly identical.
>
> This is, to put it mildly, not the case between, say, English and Bengali.

X
But if we only knew the modern IE languages, we would still be able to
reconstruct words like:

*p at ter 'father'

(on the basis of, say, English <father>, Italian <padre>, and Hindi
<pita>)

*newos 'new'

(on the basis of, say, Modern Greek <neos>, Portuguese <novo>, and Polish
<nowy>)

The same is true for morphological paradigms etc.

Adam Hyllested


From adahyl at cphling.dk  Fri Jul 23 12:49:35 1999
From: adahyl at cphling.dk (Adam Hyllested)
Date: Fri, 23 Jul 1999 14:49:35 +0200
Subject: PIE vs. Proto-World (Proto-Language)
In-Reply-To: <l03102800b3b4aa9358c0@[195.247.226.14]>
Message-ID: <FRI.23.JUL.1999.144935.0200.>

On Fri, 16 Jul 1999, Ralf-Stefan Georg wrote:

> Even the
> notion that every known language, as Basque, Burushaski or whatnot, has to
> be related to some other language is ideology.

I suppose it is common sense to believe that "isolates" like Basque or
Burushaski at least have *dead* relatives. A language in colloquial
use never stays ONE language, but will inevitably split up into several
new languages. Which is exactly why "isolate" is merely a term
used for languages that have not yet been proven to be related to other
languages. Every single language has a mother and sisters. This is also
true of PIE, of course.
I could accept your statement above, if you inserted the word "living"
between "other" and "language".

Best regards,
Adam Hyllested


From jer at cphling.dk  Tue Jul 20 01:06:11 1999
From: jer at cphling.dk (Jens Elmegaard Rasmussen)
Date: Tue, 20 Jul 1999 03:06:11 +0200
Subject: Momentary-Durative
In-Reply-To: <005201bed144$a5098b40$6b058cd4@xpoxkjlf>
Message-ID: <TUE.20.JUL.1999.030611.0200.>

On Sun, 18 Jul 1999, petegray wrote:

>>>> [Jens:] Strunk has shown that nasal presents go with root aorists,

> Can you give us a reference please, Jens?   I would like to question this,
> for the following reasons:
> (a) In Greek it is largely true, but there are exceptions;   so a bald
> statement would need qualification.

It was never claimed for the individual languages, only for the common
reconstructed protolanguage (or even relatively recent prestage of it), so
"qualifications" are not a problem out of the ordinary.

> (b) In Latin it is largely untrue, since the aorists are either sigmatic or
> the rare reduplicative  aorist (tango tetigi, claimed by some as an aorist
> on the basis of Homer tetago:n, or the lengthened vowel: pango pe:gi (~
> perfect pepigi).   I can only find cumbo cubui which supports the claim in
> Latin.

You might also have thought of cerno:/cre:vi:, fundo:/fu:di:,
linquo:/li:qui:, rumpo:/ru:pi:, sino:/si:vi:, sperno:/spre:vi:,
sterno:/stra:vi:, vinco:/vi:ci:.

> (c) In Sanskrit there is a variety of possible presents and possible
> aorists.   There are 29 roots listed by Whitney which take nasal presents
> (ignoring the class nine presents).  Of these the first retains its nasal in
> the aorist, the second has no aorist, the third shows non-nasal present
> forms, and has either reduplicated or sigmatic aorist, and so on .....
> through the list.

What's wrong with yunakti/ayok, s'rnoti/as'rot, bhinadmi/bhet,
vrnakti/vark, and why ignore those of class nine (too good?) - ?

> (d) As stated earlier, a root may show nasal and non-nasal presents in
> different languages and even in the same language.

Yes, and where there is evidence about what is old and what is young, it
is a very general picture that the nasal present went with a root aorist.

> So as a bald statement I find it unbelievable.   The evidence doesn't seem
> to be there.

Much depends on how the statement is phrased - and about what language.

>    I would like to see what Strunk actually says.

Klaus Strunk wrote the book Nasalpra"sentien und Aoriste which appeared in
1967 at Winter, Heidelberg: Discounting refs. and index etc., it's just
about 100 pages of text that makes for quite easy reading. Strunk makes a
very strong case for the PIE paradigmatic companionship between nasal
presents and root aorists. It would be wrong to report this as a definite
rule, but Strunk does conclude at ''die paradigmatische Kombination beider
Sta"mme als ein uraltes vorhistorisches Pha"nom'' (p. 128). The book
treats verbs of the Indic classes V and IX (why did you exclude the
latter?), adding in a special chapter that VII seems to have worked the
same, but has not left such a clear picture. - From looking all through
IE, I have got the impression that an inherited nasal present is as the
general rule accompanied by a root aorist (or whatever became of root
aorists in a given branch - in Greek they often added -s-). The reverse is
not the case - and nobody ever claimed that: the root aorist is also the
normal aorist type for reduplicated presents, and very often also for
y-presents (which may have been back-formed to replace nasal or
reduplicated presents that fell out of use, or simply to create a present
stem where none existed before).

> [Jens:]

>> If my observation that there is an alliance between the sk^-present type
>> and the s-aorist is correct,

> Both forms may be used for some other reason, for example phonetic.   In
> both Greek and Latin the sk- ending occurs only after a vowel, which then
> takes a sigmatic aorist for purely phonetic reasons.

How do you mean? Do you consider the s-aorist morpheme a reduction of sk^?
Or even of *-sk^e/o-?? If so, by what rule?

>>> This brings me to a general question. There seem to be two camps about the
>>> category system of the PIE verb. One believes that the aorist-imperfect
>>> distinction, to be equated to perfective-imperfective distinction,
>>> ``always'' existed in PIE and Hittie lost this distinction, while Vedic
>>> changed things around. The other considers the aspectual distinction to
>>> postdate the separation of Anatolian.

>> [Jens] There are these two camps, yes, and I am in no doubt that camp one is
>> right. There is no way the specific forms of the aspect stems could have
>> been formed secondarily in "the rest of IE" left after the exodus of (or
>> from) the Anatolians. At the very least, all the _forms_ must be assigned
>> to a protolanguage from which also Anatolian is descended.

> I thoroughly agree with you here, Jens.   The forms are scattered over most
> of the IE languages.

>> [Jens:] And what would
>> the forms be there for, if the functions that go with them only developed
>> later?

> Here I disagree.   There can be variety of form in a language without a
> difference in meaning.   Look at German plurals (or English for that
> matter).   Sometimes a difference of meaning will emerge, fixing one meaning
> on one form and the other on the other (German Woerter and Worte, Dinger and
> Dinge), but this does not mean that all variations must necessarily
> correspond to a difference in function.    Language simply doesn't work that
> way, and there are far too many counter-examples for that claim to stand.

> Counter examples in English:
> (a) strong versus weak past tenses, e.g. dived ~ dove (origin historical /
> analogical)
> (b) plural /s/ versus plural /z/ (origin phonetic)
> (c) gentive versus preposition e.g. Tom's ~ of Tom (basis stylistic)
> (d) Time expressions (believe it or not, 10:45 really is the same as a
> quarter to eleven)
> and so on.

> I see the development of a functional difference as a later phenomenon, and
> I see it emerging differently in Sansksrit and Greek and Latin.

These are not counterexamples: All of the items you quote (except d) will
have to be very old - as types. And those of the types that are not
productive will have to be very old for the items quoted (goes for dove,
not for dived). In so far as the types can be shown not to be simple
variants of each other (as the two English sibilants), they must
originally belong to different categories. In fact, Eng. _of_ did once
mean something different from the relation expressed by the genitive. The
German plural variants used to belong to different stem classes - and
different stem classes used to have different derivative suffixes that
used to express different semantic shadings.

The last part completely escapes me: Are you anticipating that 10:45 and
"a quarter to eleven" will one day come to mean two different things? Is
THAT to you the only normal way language change works?

>   I suspect
> that some scholars are temporarily misled by the use of Greek names for the
> Sanskrit verbal system.

I plead not guilty, and I'll defend most of my colleagues who also know
that the terminology is misleading and only traditional anyway.

Are you in effect saying that IE morphology cannot be reconstructed? If
so, do you mean not at all, or just not beyond some specific limit which
you feel I have passed?

Jens


From X99Lynx at aol.com  Tue Jul 20 04:22:55 1999
From: X99Lynx at aol.com (X99Lynx at aol.com)
Date: Tue, 20 Jul 1999 00:22:55 EDT
Subject: Momentary-Durative (functionlessness)
Message-ID: <TUE.20.JUL.1999.002255.EDT.>

[ moderator re-formatted ]

In a message dated 7/19/99 7:04:48 AM, you wrote:

<<Sometimes a difference of meaning will emerge, fixing one meaning
<<on one form and the other on the other (German Woerter and Worte, Dinger and
<<Dinge), but this does not mean that all variations must necessarily
<<correspond to a difference in function.    Language simply doesn't work that
<<way, and there are far too many counter-examples for that claim to stand.

<<Counter examples in English:
<<(a) strong versus weak past tenses, e.g. dived ~ dove (origin historical /
<<analogical)
<<(b) plural /s/ versus plural /z/ (origin phonetic)
<<(c) gentive versus preposition e.g. Tom's ~ of Tom (basis stylistic)
<<(d) Time expressions (believe it or not, 10:45 really is the same as a
<<quarter to eleven)
<<and so on.

<<I see the development of a functional difference as a later phenomenon,>>

Just want to point out that every single one of the examples given IS
functional:
<<(a) strong versus weak past tenses, e.g. dived ~ dove (origin historical /
<<analogical)>>

 Historical continuity promotes predictability, and expectations
in listeners.  Analogy generalizes rules, promoting structural symmetry,
retention and again predictability for the sake of listeners.

<<(b) plural /s/ versus plural /z/ (origin phonetic)>>

  If by phonetic, you
mean dependent on the preceding phonemes, then that is functional. (/boys/
versus /boyz/) I don't think you mean that clearly articulated /s/ and /z/
are randomly interchangeable without consequence to the listener.

<<(c) gentive versus preposition e.g. Tom's ~ of Tom (basis stylistic)>>

  The
use of one or the other can materially affect sense and communicative effect.
 "Tom's dog bit the mailman." v. "The dog of Tom bit the mailman."  In
American English, the 'style' makes one version common, the other very rare
indeed.  They are not interchangeable and there is an obvious functional
difference between them.

<<(d) Time expressions (believe it or not, 10:45 really is the same as a
quarter to eleven)>>

  Numerics obviously have a functional advantage in many
situations, particularly tv listings and flight schedules, etc..  At the
airport, you will not find departures listed under "three quarters to
eleven."  Or even "quarter to eleven," for that matter.  Precison is one
functional difference.

Granted that the
aorist-imperfective/perfective-imperfective/momentary-durative-narrative,
etc., distinctions may not be the only way to explain the structural
differences that arose at any point pre- and post- Anatolian.  But to say
that those structural features had no functions at all is perhaps going too
far - especially because one can only guess what those functions were in a
real world language.

How would you explain, for example, why in American English, "Tom's dog bit
the man" is expected, normal and sounds right.  But "The dog of Tom bit the
man" is clearly odd and rarely used?  There is a functional difference
between using one form or the other.  If you don't think so, try talking the
odd way and see what kind of reaction you get from your listeners

Regards,
Steve Long


From vidynath at math.ohio-state.edu  Tue Jul 20 11:38:00 1999
From: vidynath at math.ohio-state.edu (Vidhyanath Rao)
Date: Tue, 20 Jul 1999 07:38:00 -0400
Subject: Momentary-Durative
Message-ID: <TUE.20.JUL.1999.073800.0400.>

Rick Mc Callister <rmccalli at sunmuw1.MUW.Edu> wrote:

> Spanish distinguishes them [resultative and passive]
> by resorting to

> estar --a "momentary" verb--
> [actually a verb indicating condition] for resultant conditions--

> and ser--a "durative" verb
> [actually a verb indicating  characteristics]
> for passive constructions.

I don't know much Spanish, but I ought to have remembered this for I have
been told about this before.

I find something curious about your glosses. Resultatives, as indicating a
state of indefinite duration, ought to be duratives, though examples of
languages which use `go', `come', `finish' to form resultatives are
mentioned in ``The evolution of grammar''. How old is the use of `estar'?.


From vidynath at math.ohio-state.edu  Tue Jul 20 15:26:26 1999
From: vidynath at math.ohio-state.edu (Vidhyanath Rao)
Date: Tue, 20 Jul 1999 11:26:26 -0400
Subject: Momentary-Durative
Message-ID: <TUE.20.JUL.1999.112626.0400.>

 petegray <petegray at btinternet.com> wrote:
To: <Indo-European at xkl.com>
> >>> I don't see any parts of Sans lit in which aorist has resultative
> >>> meaning.

I have access only to the e-texts of the critical editions, which I will
use. Given the difficulties in disentangling resultatives from hot-news
forms, we need to look at the context. To really convince me, you need to
translate into a living language that distinguishes the two and explain why
the resultative is preferable.

> or Mahabharata 2:60:7 (CE 2.60.8ef)
>   kim nu pu:rvam para:jais.i:t a:tma:nam
>  whom you have lost first, yourself....

This is the easiest to dispose of. Here is the whole stanza:
kasyes'o nah paraajaiSi:r iti tvaam aaha draupadii/
kim nu puurvam paraajaiSiir aatmaanam atha vaaapi maam.//
The context is Draupadi's question of how Yudhisthira can bet her if he had
already lost himself. Translating the second line as ``whom >have< you lost
first, yourself or else me?'' misses the point that D. was bet legally.

>   Ramayana 2:64:52 (CE 2.58.44cd)
>   yah s'aren.aikaputram tvam aka:rsi:r aputrakam
>  with one arrow you have rendered me childless, who had but one son

The first half is
 adyaiva jahi ma:m. raajan maran.e na:sti me vyathaa.

Verse b is a paranthetical remark, placed there for metrical reasons. [It
would be quite odd to claim that the father feared eath when the son was
still living.] I would translate it as ``Slay me, whom you made childless
from being one-sonned, now itself, King; I do not fear death.'' The point is
that Dasaratha, by killing the son, has metaphorically slain the whole
family. It is the event of killing, not the resulting death, that is being
highlighted in the second half.

> or Mahabharata 5:3:10
>   tad akars.it prajagaram
>   that has produced sleepiness

A semantic point. praja:garam is ``awakefulness'', not sleepiness. Anyway,
it is not so obvious to me that the result is being highlighted here. In
other words, it is not clear to me that I should translate this into Tamil
as ``atu muzhipp- un.t.u pan.n.i-y- ul_l_atu'' rather than ``...
vit.t.atu.'' [I don't know enough English to render this difference
effectively.] It is the immediciacy of the event that comes to the fore.


From petegray at btinternet.com  Tue Jul 20 19:45:44 1999
From: petegray at btinternet.com (petegray)
Date: Tue, 20 Jul 1999 20:45:44 +0100
Subject: Momentary-Durative
Message-ID: <TUE.20.JUL.1999.204544.0100.>

Thank you for your patience, Jens.    I'm replying to you points in reverse
order - it seems better that way, since it puts the more important things
first.

You said:

> ... an increasingly well-established structure, with archaisms and
> productive patterns, .... I react in
> defense of the field when I see somebody turning the clock back ...

Rightly so.   I am not a linguistic Luddite.  But I am an enquirer, and I
believe I am right to ask for evidence rather than assertion.    You have
insisted several times that nasal presents and root aorists go together.
Apart from mentioning Strunk, you have not given evidence.  I don't dispute
that the evidence is there - and perhaps it's so well known that you haven't
felt it necessary to give any - but I'd like to see it, or a reference to
Strunk so that I can chase it down for myself.   You do allow that PIE can
have variant presents, so that seems to run against your main claim.

Your examples of reduplicated presents developing into nasal presents is a
well chosen answer to my point - thank you.

Thank you, too, for the references to Rix, and Renou.   I was not aware of
either.

You said:

> The impression that [in the IE verbal stem] ...
> no common system is
> recoverable is not compatible with current knowledge

There is little real "Current knowledge" with PIE.   It is in many cases
really rather "current opinion which someone happens to accept".   So how
much support is there for Strunk?  How much debate is there?   I read as
widely as I can in PIE, and have not met a mention of him - but I'm sure the
failing is mine.   In any case, I do not dispute that patterns are
perceivable.   I do question, however, how much weight we should give these
patterns, and in particular whether we over-prioritise the patterns based on
Greek & Indo-Aryan.   It has been shown again and again since the 70's that
the "south-eastern" group of Greek, I-A, and Armenian is highly innovative.

I said:

>> your argument (that a root aorist implies a nasal present) is
>> really rather weak.

Here I must apologise - as you rightly point out, I had your argument back
to front.

You said:

> No, thematic presents and root-presents typically take the s-aorist (or
> suppletion).

Root presents are rare outside I-I, and in Sanskrit -s- aorists do not seem
to be typical for root presents, since the both sigmatic and asigmatic
aorists occur.   E.g. am has asigmatic, i:d has both, i:r shows both, u
shows none, U:h asigmatic, and so on.   Where, again, is your evidence that
makes it "typical" for a root present to take an -s- aorist?
    Thematic presents appear to be a later formation, in any case, and you
seem to suggest an origin for them in root aorist subjunctives.

Incidentally, I read today in JIES vol12, 1984, the argument that even
the -s- aorists are an innovation within Greek, I-A and closely related
groups.   The article alleged that evidence outside this area is weak, and
tried to dispose of Latin -s- perfects by suggesting they were either
limited to verbs ending in a velar, or they were back formations from the
supine.
    Don't hear me supporting this idea.  I mention it because it raises the
kind of question I want to ask about Strunk.   Here an article attacks
something that is well established for PIE.   So I cannot believe that there
is no argument at all about Strunk's ideas, or that they are "current
knowledge".

Now I hate to be a pain, Jens, and I freely acknowledge the limitation of my
understanding.   But I hope the above shows the kind of questions I have to
grapple with before I can lie down and accept what you say.

Peter


From vidynath at math.ohio-state.edu  Mon Jul 26 13:03:47 1999
From: vidynath at math.ohio-state.edu (Vidhyanath Rao)
Date: Mon, 26 Jul 1999 09:03:47 -0400
Subject: Momentary-Durative
Message-ID: <MON.26.JUL.1999.090347.0400.>

Jens Elmegaard Rasmussen <jer at cphling.dk> wrote:

> On Mon, 12 Jul 1999, Vidhyanath Rao wrote:

>> The tuda'ti type, with primary endings, is somewhat rare in RV. Perhaps,
>> tuda'ti type was a replacement for injunctives, i.e., tuda'ti was a
>> general  present and not progressive present. Just a thought.

> Hardly so, for the injunctive is not a stem-formation, but an inflectional
> category that can be formed from all verbal stems.

Present, with primary endings, is also an inflectional category.

The roots of the tudati type, excepting transfers from root and nasal
classes, tend to have punctual meanings. Punctual verbs cannot be used in
the progressive (unless repetition is indicated) but can be used in generic
sense etc. But when the injunctive was lost, only the present was available
for this. This is what I meant.

>> [Jens:]

> If my observation that there is an alliance between the sk^-present type
> and the s-aorist is correct ... then the s-aorist was originally
> inchoative in function. I'd say that makes very good sense, for the
> s-aor. is also widely used with verbs that form radical or thematic
> presents, as *weg^h-e/o- 'drive', whose aor. *we:g^h-s- will then
> originally have meant 'start driving, set out (by carriage)'.

This makes it harder for me to understand how the aorist became the
perfective. `Started driving', in contrast to `drove', suggests incomplete
action.

>> This brings me to a general question. There seem to be two camps about
>> the category system of the PIE verb. One believes that the
>> aorist-imperfect distinction, to be equated to perfective-imperfective
>> distinction, ``always'' existed in PIE and Hittie lost this distinction,
>> while Vedic changed things around. The other considers the aspectual
>> distinction to postdate the separation of Anatolian.

> There are these two camps, yes, and I am in no doubt that camp one is
> right. There is no way the specific forms of the aspect stems could have
> been formed secondarily in "the rest of IE" left after the exodus of (or
> from) the Anatolians. At the very least, all the _forms_ must be assigned
> to a protolanguage from which also Anatolian is descended. And what would
> the forms be there for, if the functions that go with them only developed
> later? We find practically all the IE verbal stems in Anatolian, many only
> in a few lexicalized remains; are we to assume they have had totally
> enigmatic earlier functions, and that they were later dug up by "the rest
> of IE" and given totally new functions there? Clearly the only unforced
> interpretation is that their functions in the common protolanguage were
> the ones with which they are found where they do survive with a palpable
> functional identity.

The stem formants might have been derivational in nature. In particular,
they might change the meaning, valence and/or aktionsart. It is the
assumption that they changed only the aspect and nothing else that is in
question. Your view, expressed elsewhere, that different present stems might
be because there are different kinds of durativity is halfway to this.

Incidentally the assumption that all kinds of repetition must be represented
by imperfectives is as mistaken as the assumption that atelics cannot be
represented by perfectives. Languages differ in how they convey repeated
complete events. Czech, for example, is said to (permit the use?) use the
perfective in this case [Dahl, TAS]. Gonda (somewhere in ``The aspectual
function of Rgvedic ...'') gives examples from Greek where the aorist is
used with adverbs denoting repetition. And, just because we can think of
`walk' as a repetitive`step', we cannot conclude that `walk' cannot be used
in a telic manner.

Grammatization of derivational affixes is found often enough that its
occurance in Pre-IE -> PIE or PIE -> dialects cannot be rejected out of
hand. As it seems that very different languages underwent similar evolution
of grammatical categories (see Bybee et al, ``The evolution of grammar), we
cannot assume the similar functions must go back to the proto-language. That
might have had a precursor function.

>> There are certain nagging questions about the first thesis: The change
>> in Vedic is not explained and how it came about without the prior
>> loss of  aspect has, AFAIK, not been explained. Those who adhere
>> to this also feel the need to explain away as much as possible of
>> root presents. But there are enough of them remain in Hittite and
>> Vedic to raise doubts.

> I don't follow - what change in Vedic are you talking about?

In Vedic, the so-called imperfect is the tense of narration. Aorist has a
recent past meaning. In living languages with aspect, the perfective is the
tense of narration, unless historical present is being used. Languages with
remoteness distinction, either it applies only in the perfective or without
regard to aspect. It is this that needs to be explained.

People have tried to explain this. The attempts I know of are by Gonda and
Hoffman. Gonda's explanation is that this is due to the ``national character
of ancient Indians'' that led them to use in lento description in place of
narration. I won't comment on this idea of explaining language change by
reference to ``national characters'', except to ask about reactions to a
claim that the use of passe compose in French is due the Gallic
characteristic of exaggeration that sees everything as having present
relevance.

Hoffman's answer is to posit an intermediate stage in which aspect was
limited to non-recent past. He does not give any contemporary examples, nor
does he explain how the aorist, which in such a stage must have been even
more common, was lost in reference to remote past.

The point is that an alternate explanation is possible: Completives have a
``hot news'' value, which makes it plausible to see them develop into recent
past. They also can develop into perfectives, Slavic being a usable example.
There are gaps in the examples, but I find this more plausible than
deriving the Vedic usage out of perfecitve-imperfective opposition.

> Why would anyone want to explain away root presents where
> they are securely reconstructible? One would do that only to avoid
> having a language combining a root-present with a root-aorist, for in that
> case the two aspect stems are identical. That is why I am so sceptical
> about the authenticity of the Vedic root presents lehmi and dehmi,
> because for these verbs we have nasal presents in some other IE
> languages pointing to the existence of a root aorist; thus leh- deh- look
> like displaced aorists. But not so for eti 'goes' or asti 'is': these are
> durative verbs, and so their unmarked form could function as a
> durative (socalled "present") stem.

This argument is valid only if you have an independent reason to assume
obligatory perfective-imperfective contrast for PIE. But firstly, we see
stems transferred from one class to another within a single group. This is
most obvious in Indic, where we have a long recorded history. Secondly,
there are too many root presents that remain: In Vedic, amiti (injures),
da:ti (cuts, divides), yauti (joins, unites) and of course, hanti confirmed
by Hittite kuen-. None of these is eligible for a nasal present (expecpt
perhaps amiti, if you believe in nasal presents for roots of the shape CeNH,
with N standing for a nasal). Especially *g'henti is unavoidable, unless you
are going to argue that the two oldest recorded dialects innovated in
precisely the same way. Such an argument needs more compelling evidence
than a just so story.

If you add potentialy telic verbs, we get some more: ta:s.t.i (taks,
fashion), ya:ti etc. These can and are used in the so-called imperfect with
definite objects without any indication of non-completion. etc. How can they
be called imperfetive?

>> [part of my (Nath's) post deleted]

> Again, I do not think there is any problem in accepting root present as
> original for inherently durative verbs, and root aorist as equally
> original for inherently punctual verbs.

What about telic verbs? These are inherently durative, but with a determined
object, their preterite (in languages without perfective-imperfective
distinction) carries the implicature of completion, while their imperfective
past would do the opposite. How, without recorded narratives, can you decide
how a given language operated?

> But the IE aorist is not restricted to any special kind of verbs - it is
> only _unmarked_ (better: apparently originally unmarked) for inherently
> punctual verbs; for other verbs the aorist need a morphological marking,
> and the meaning is then some nuance that can be regarded as punctual
> ("started to -") or it just reports that the action got done (Meillet's
> action pure et simple).

Could the imperfect report `action pure et simple', or was that reserved for
the aorist? How did PIE speakers report a durative action that was done,
like ``I walked home''? How did they say ``I made pots yesterday''?

> One important functional point with the aorist,
> however, is that it marks a turn of events which creates a new situation,
> whereas the "present aspect" stays in the situation already given and
> reports another action contributing to that situation.

How do you explain that it is the so-called imperfect that is the tense of
narration in Vedic?

> This is seen remarkably well in the prohibitive use of the prs. vs. aor.
> injunctive, as propounded so clearly by Hoffmann.

Hoffmann's claim requires morphological gymnastics, such as taking i:'s'ata
as an aorist and still has some holes (eg, jivi:t).

> Still, even verbs generally signifying completed action could form
> duratives, indicating e.g. a repetition of the action (give one thing, and
> then another) or an as yet unsuccessful attempt (I'm opening the window).

How do you classify ``I learned that chapter in one month?''

> We know that this kind of change was small enough in the languages here
> concerned to lead to a number of misplaced aspect stems. E.g., the
> Armenian aor. eber is an old ipf. Is the jump from "narrative past" to
> "recent past" so great? If it is, even great changes happen.

But if the PIE ``imperfect'' was really just a preterite, what we see in
Armenian eber is old preterite becoming an aorist due to the rise of an
imperfective past. That is attested. For example, Kui, a Dravidian language,
has generalized an old progressive into an imperfective and this limited the
preexisting past into an aorist.

It is not the size of the change, but the direction and manner of the
change, that must be credible. All I am asking for is a single
uncontrovertable evidence of this change. It won't do to say, as I was once
told by an Indologist, that this change is possible because it must have
happened in Vedic. You need to give an example where the perfective past
category can be established from >preserved texts<.

What I find difficult to swollow with the argument for aspect in PIE is that
the association ``imperfect'' = imperfective is limited to Greek. The two
oldest recorded dialects, Hittite and Vedic, do not work that way. We are
simply supposed to believe that they innovated, but what do we get in
return? What can you explain this way that you cannot explain otherwise?

[quoted from a different post]
> The most rewarding experience during the time I have been
> watching Indo-European Studies has been to see the protolanguage
> come alive and assume an increasingly well-established structure, [...]

Do we really understand the variety of syntactic structures and their
diachrony that well? I have mentioned the Tamil -vidu construction a few
times. You will find some linguists call that a perfective and the Tamil
simple past an imperfective. This is simply wrong as the simple past is and
has been the tense of narration for the 2000+ year recorded history, and
this distinction is nothing like the perfective-imperfective distintion in
Russian or Arabic, as can be seen by comparing the translations of Dahl's
questionairre [or simply use his summaries]. Use of the inappropriate lables
is due to either trying to fit every non-tense non-mood apposition into the
Procrustean bed of perfective-imperfective distinction or the lack of
adequate terminology. If anyone thinks that they know how Tamil syntax works
based on these labels, the mildest thing one can say is that (s)he is
woefully misinformed.

It can be even worse: Similar constructions exist in some NIA languages
which also have a prefective-imperfective distinction that is different
(imperfective out of old progressive). I have seen at least one linguist
apply the label of perfective to the construction with auxillary, and
imperfective to the one without the auxillary. This completely misrepresents
the syntax.

What reason is there to think that we are not making the same kind of
mistake with PIE? I am not asking for certainity here, but evidence for
greater probability than the opposite.


From jbisso at jps.net  Mon Jul 19 04:39:57 1999
From: jbisso at jps.net (James F. Bisso)
Date: Sun, 18 Jul 1999 21:39:57 -0700
Subject: connected text in PIE; Proto-World language
Message-ID: <SUN.18.JUL.1999.213957.0700.>

Anthony Appleyard wrote:

>  (1) When neural-net computers get advanced enough for people to
>simulate better than now human brain processes including language, it
>would be interesting to simulate language evolution and see over how
>many centuries and by what steps a language changes so much that all
>trace of common ancestry vanishes behind the `noise' of accidental
>resemblances.

Some folks are using artificial life (see http://alife.santafe.edu/)
techniques to model language change: (e.g., Luc Steels at the Vrije
Universiteit Brussel (http://arti.vub.ac.be/origin/origins.html).

jim


From proto-language at email.msn.com  Mon Jul 19 05:41:44 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Mon, 19 Jul 1999 00:41:44 -0500
Subject: Lexical Retention
Message-ID: <MON.19.JUL.1999.004144.0500.>

Dear Joat and IEists:

 ----- Original Message -----
From: <JoatSimeon at aol.com>
Sent: Friday, July 16, 1999 11:43 AM

Joat wrote:

> As time goes on, any given language gives us less and less information about
> previous forms.  Information is _lost_.

> Eg., no amount of analysis of Germanic will give you a PIE word for "bear"
> because that word dropped out of that family.  All you'll get is variations
> on "the brown one".

Pat responds:

You have chosen an example about which we are singularly poorly informed ---
not to mention the influence of possible taboo deformation.

If I assumed that, regardless of its highly unusual form (*R:k{^}tho-s), the
basis for 'bear'-words is the IE root *rek-, 'tower above', it would not be
true that the ultimate root was not attested in Germanic.

In any case, you have it rather backwards. If the root and all manifestions
of it (like, perhaps, in 'bear') were absent in Germanic, what would, as it
has, make possible our reconstructing an IE root would be its presence in,
at least, three (nominally) other branches.

I thought I made clear that I was talking about *CVC-roots. Your example
does not really address my assertion.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From stevegus at aye.net  Mon Jul 19 11:51:19 1999
From: stevegus at aye.net (Steve Gustafson)
Date: Mon, 19 Jul 1999 07:51:19 -0400
Subject: indoeuropean/hand
Message-ID: <MON.19.JUL.1999.075119.0400.>

Ed Selleslagh writes:

<<Actually it's 'cornu' in the nominative.>>

-Cornu- is "horn."  -Cornus- is the dogwood tree.

<<All tree names, also those with -u stems like 'quercus' (oak) (and -o stems
like po:pulus) are feminine.>>

I had ever understood that all trees were feminine for essentially
mythological reasons; they were thought to be the dwelling places of female
spirits.

<<Could there be some relationship here?  I mean
'extremities, branches...'.  That would also account for manus, acus, cornu
and
tribus being feminine, but not for domus nor idus.  'Foot, pes...' are
probably
not viewed as extremities, but as 'base' to stand on.>>

Now, my understanding is that the root idea of -idus- is of a division by
halves.  When the months used to be lunar, the kalends marked the first
appearance of the slip of the new moon, and the ides were the date of the
full moon.  Perhaps this could be worked into your hypothesis as well.

--
L'an mil neuf sens nonante neuf sept mois
Du ciel viendra grand Roy deffraieur
Resusciter le grand Roy d'Angolmois
Avant apres Mars regner par bonheur.
                                --- M. de Notre-Dame


From ECOLING at aol.com  Mon Jul 19 15:08:45 1999
From: ECOLING at aol.com (ECOLING at aol.com)
Date: Mon, 19 Jul 1999 11:08:45 EDT
Subject: Semantic change
Message-ID: <MON.19.JUL.1999.110845.EDT.>

In a message dated Patrick Ryan:

>Having worked extensively with language-family
>comparisons, my statistically unsupported "guess" is that there is amazingly
>little vocabulary loss if one allows reasonably semantically-shifted pairs
>to be counted as "retained".

That would be circular without further constraints,
because it would mean (not what Pat means,
but taken to extreme) that any meaning could change
into any other meaning, and so we are only then seeking
look-alikes, not cognates.

Rather, there needs to be a quantification of
degree of semantic change,
and of course a life-long learning of what meanings
ARE known to change into what other meanings
under what circumstances,
so we can even begin to try to measure
how far a purported related word is
from the meaning of another word we are comparing
it to.
Then, as a matter of degree, we can say that
as we allow the meanings to be more distant,
we get more *possible* cognates between any
two languages we wish to compare.
(for a fixed degree of phonetic resemblance we
might require, however measured).

All of these things are matters of degree.

A less plausible semantic relation means that two
look-alikes are less plausible as cognates.
But requiring "identity" of meanings or even
"near-identity" of meanings is an absurd
requirement also, when we are working at great
time depths.

So what to do?
Asserting simplistic extremes either for or against
is not particularly useful.

Lloyd Anderson
Ecological Linguistics


From ECOLING at aol.com  Mon Jul 19 15:08:41 1999
From: ECOLING at aol.com (ECOLING at aol.com)
Date: Mon, 19 Jul 1999 11:08:41 EDT
Subject: Principled Comparative Method
Message-ID: <MON.19.JUL.1999.110841.EDT.>

Here is an example of where the "Comparative Method" can be strengthened.
The matter of "recurring sound correspondences".
The standard presentation is pretty close to what Sean Crist just posted
(quoted at the end).

Now let's consider things as a matter of degrees, not a sharp cutoff in what
is and what is not reconstructible.

As the relation between two languages or language families becomes more
distant,
the number of examples of any sound CORRESPONDENCE become
fewer and fewer.  This is a matter of degree.  The reasoning must be
more sophisticated.  We then have more cases in which multiple sound changes
apply to the same word, and we might have cases in which
(to use the completely artificial and hypothetical example)

a)  there are many examples of PIE *d corresponding to PU *z
     (or even IE /d/ corresponding to Uralic /z/, using particular languages).
b)  there are 5 examples of PIE /d/ corresponding to /z/
c)  there are 2 examples of IE /d/ corresponding to Uralic /z/
d)  there is only 1 example of IE /d/ corresponding to Uralic /z/
e)  there are no examples of IE /d/ corresponding to Uralic /z/

EVEN IN CASE (E), IT STILL MIGHT BE CORRECT TO ARGUE
that IE /d/ "corresponds to" Uralic /z/.

The reason is that "recurring sound correspondences"
is actually NOT a cornerstone of the "Comparative Method",
if we mean by Comparative Method what it is in ideal practice,
that is the reconstruction of the series of changes by which
some ancestor language gave rise to descendent languages.
(Calvert Watkins is at least one prominent Indoeuropeanist
who has emphasized this principle.)

Rather, "recurring sound correspondences" is a short-cut
which works in the EASY cases.

What IS a cornerstone of the comparative method is
a PLAUSIBLE sequence of changes from ancestral language
to descendant languages or language families.

So if we had case (d)
and we also had case (d'),
one example of IE /t/ corresponding to Uralic /s/,
we might have more evidence for saying that /d/~/z/ was real.

Or if we had case (d) before a following /u/,
and we also had case (d'')
ONE example of IE /d/ corresponding to Uralic /z^/ before /i/,
(interpret the ^ as a hachek = wedge, upside down)
we would again have more evidence for saying that /d/~/z/ was real,
in this case because we would assume that /z/ became /z^/ before /i/.

And if lucky we might have case (d') above with following /u/,
and also case (d'''),
one example of IE /t/ corresponding to Uralic /s^/ before /i/.

With all four of these examples, we would have a plausible system,
and would in fact have four examples reflecting the same sound changes,
perhaps actually only a single sound change (namely
"affricate apical stops before high vowels, with the result being
the fricative of the same voicing and "reflecting" the vowel quality)
but we need not have more than one example of any particular correspondence
on the surface.

Of course multiple examples of the same surface results DO
make us more confident of our findings.  The point above is that
examples which are not identical on the surface can ALSO make
us regard the hypothesis of a particular sound change as more likely
valid than would be the case if we had only one of the examples
(d, d', d'', d''').  In extreme cases, I would think we could even
confidently hypothesize a sound change whose immediately results
remain NOWHERE visible on the surface, case (e) above.

The plausiblity of the highly specific changes hypothesized above
is sufficiently demonstrated by the occurrence of something like it
in modern Japanese, where we have in the syllabary

sa se s^i so su
ta te c^i to tsu
za ze j^i zo zu
da de j^i do dzu~zu.

Of course as the number of sound changes hypothesized
rises, we still do need an increasing number of examples,
since we don't want to be in the situation of having a new
hypothesis just to excuse each new example of words in
two languages we wish to claim decend from a common proto-form.
We just do not need to have "recurring sound correspondences"
recurring identically in their surface results.

This kind of situation can be seen in real cases
of language reconstruction, and it needs to be catalogued in standard
manuals of the comparative method, so that the short-cut of
"recurring sound correspondences" can be seen for just what it is,
a short-cut in the easy cases.

In fact, all well-trained specialists in comparative reconstruction
DO use more sophisticated tools than mere superficial recurring
sound correspondences.  Their task is easier with more recurrences,
but it is not limited to them, certainly not in principle.

What is considered "plausible" in the more difficult cases cannot be
simply the whim of the individual researcher about what is a "plausible"
vs. not a "plausible" sound change, on a priori grounds.  It must be
based on evidence of some kind.  We cannot assume our manuals
contain examples of all the possible sound changes,
we can only assume that known sound changes ARE plausible,
and that as-yet-unknown ones will always exist.
For beginning linguists, the substitution of /l/ and /s^/ comes as
a surprise, until they learn that these two segments have some strong
acoustic similarities.
Learning what constitutes a "plausible" sound change is a lifelong task.

Since the number of sound changes between a proto-language and its
descendants does increase with time depth,
the number of recurring sound correspondences will decrease,
and we must make increasing use of the more sophisticated tools.

The real task of extending the Comparative Method to deeper time
depths is to make explicit more of these sophisticated tools,
and to CREATE more such tools by discovering what ways of
handling the data are robust across what kinds of intervening changes.

One of them I am quite certain is to develop both articulatory and
acoustic "spaces", relative "distances" between different articulations
and different acoustic effects, so that when attempting to judge
likelihood of cognacy of pairs of words, we can judge similarity
by degrees, not by yes/no dichotomies.  (These will partly depend
on the general typology of the sound systems of the languages
concerned, they will not be completely universal, but they will
also not be completely idiosyncratic.)

So for example, the sound change w>k does not seem "obvious",
yet if we add y>c^ or even y>t, or even if we don't, it is not the
most distant pair of sounds /w/ and /k/ do share some things
(and we do have /erku/~/duo/~"two" with r~d~t and k~w~null before /u/).
We need a measure of DEGREE of similarity between such known
pairs of words, such that Albanian and English, or Armenian and English,
can be found to be related languages.
Throwing up our hands, as some do, and saying
"any sound can change into any other sound" past a certain time depth,
is not helpful.  At the extreme, in an absolute sense, it is quite possibly
true,
but that is the reaction of those who do not yet have the tools to handle
relative-degrees-of-plausibility in many dimensions simultaneously,
rather than simple yes/no decisions or even degrees of plausibility
in a tiny number of dimensions.

Instead of always using the easy method (recurring sound
correspondences), let's discover the more sophisticated tools that
work in the harder cases.

Who is doing that now?

I would like to see forums which can highlight and reward such
productive work.  How about testing methods which can work
to establish PIE based on English and Albanian (see below)?

Lloyd Anderson
Ecological Linguistics

Below are the quotations which quite adequately represent the
normal presentation of the Comparative Method.

>Suppose for the sake of
>argument that there is in fact a genetic relationship between
>Proto-Indo-European and Proto-Uralic (to take one possible example). Even
>if this were true, and given the evidence which we actually do have at
>hand, would we expect to be able to _show_ that there is such a relation?

>Here, the problem _is_ like trying to reconstruct PIE on the basis of only
>modern English and Albanian.

[Lloyd:  So to establish more empirical validities, we need to test
methods to find which highly general methods (not tailored to the specifics
of the English and Albanian case) would give a positive answer
to the question of relation of English and Albanian.  We might have
almost no recurring sound correspondences (though I suspect there are some),
and a large number of correspondences which result from successive
applications
of more than one sound change in different contexts, so not obviously
intuitive on the surface.]

>Even tho external evidence might lead us
>to
>guess that there could be a relation between PIE and PU, the cognations
>are so obscured by millenia of sound changes, loss of old lexical items,
>the noise of new lexical items, etc., that it might well be impossible
>to
>show a genetic relationship on the basis of the available information.

>Suppose, for example, that PIE *d corresponds to PU *z (I don't know that
>there is a *z in PU; I'm making this up).  Even if there were originally
>dozens of words showing this correspondence, it might well be that case
>that only one such pair exists between the small subsets of the original
>lexicons of PIE and PU which we can reconstruct.  Given the fundamental
>assumptions of the Comparative Method, you _can't_ show that the two words
>are cognate when you've only got one example of the correspondence.

>You need multiple examples of sound correspondences to be able to conduct
>the Comparative Method at all; and when the the cognates become as
>rarefied as they are this time-depth, the likelihood of having access to
>an adequate number of examples to work out the relevant sound changes
>becomes proportionately smaller, eventually reaching what for practical
>purposes is an impossibility.

[Lloyd:  The following clause, taken as an absolute, is false:

>You need multiple examples of sound correspondences to be able to conduct
>the Comparative Method at all; ...

It is true only when "sound changes" is substituted for "sound
correspondences",
yet then it cannot be as quickly seen whether we have that in a given
situation.]


From X99Lynx at aol.com  Tue Jul 20 03:15:38 1999
From: X99Lynx at aol.com (X99Lynx at aol.com)
Date: Mon, 19 Jul 1999 23:15:38 EDT
Subject: Chronology of the breakup of Common Romance [long]
Message-ID: <MON.19.JUL.1999.231538.EDT.>

In a message dated 7/19/99 12:15:24 AM, mcv at wxs.nl wrote:

<<That's merely because the Slavic languages have retained the
"(c)h" in the Germanic loanword which was dropped in the modern
Germanic languages (OE wealh, OHG walh).>>

That would explain a lot of the way the variances seem to go.  But what
happens in MHG?  I have 'Walch' (n), 'walsch', 'walche' (adj).   It seems
like there's some back and forth.  I guess that makes sense in that 'vlach'
itself is a form borrowed back from Slavic.

In Old Norse, where the reference to anything Rumanian is least likely,
'valskr' is typically reconstructed from '*valr' (which would put it before
800 AD.)  I wonder if that reconstruction isn't questionable.

Regards,
Steve Long


From X99Lynx at aol.com  Tue Jul 20 03:20:46 1999
From: X99Lynx at aol.com (X99Lynx at aol.com)
Date: Mon, 19 Jul 1999 23:20:46 EDT
Subject: 'Origins of the Rumanian Language' URL
Message-ID: <MON.19.JUL.1999.232046.EDT.>

Rather than answering more individual e-mails, I thought I might just post
the URL.

<<The site is at http://www.hungary.com/corvinus/dunay/

The Word file can be downloaded from http://www.hungary.com/corvinus.
>>

Steve Long


From adahyl at cphling.dk  Wed Jul 21 16:44:33 1999
From: adahyl at cphling.dk (Adam Hyllested)
Date: Wed, 21 Jul 1999 18:44:33 +0200
Subject: `cognate'
In-Reply-To: <Pine.GSO.3.95.990716161343.15501C-100000@unagi.cis.upenn.edu>
Message-ID: <WED.21.JUL.1999.184433.0200.>

On Fri, 16 Jul 1999, Sean Crist wrote:

>>> I'm sure it says PIE *<p at ter> in your manuscript (+ diacritics on the
>>> <e>).

>> Yes, it does, and my posting should have read `p at ter', but I slipped up.

> At least according to Don Ringe, this word should be reconstructed
> *pH2te'r.  I should ask him what the arguments are in favor of that
> reconstruction, since I don't know myself.

It is more or less the same reconstruction. <@> is (originally, at
least) the vocalic allophone of <H>, reconstructed on the basis of
Indo-Iranian, where the root vowel is <i> (an IE <**pater> would have
developed into Sanskrit <**pata:> and not the actual <pita:>)
Using <@2> or <H2> is simply a matter of phoneme-to-grapheme
correspondence.

Best regards,
Adam Hyllested


From pagos at bigfoot.com  Thu Jul 22 18:59:36 1999
From: pagos at bigfoot.com (Paolo Agostini)
Date: Thu, 22 Jul 1999 20:59:36 +0200
Subject: -n- adjectival suffix in Latin
Message-ID: <THU.22.JUL.1999.205936.0200.>

I was wondering about the origins of the Latin adjectival
suffix -n-us, -n-a, -n-um appearing in words like _mater-nus_ "maternal;
motherly; belonging/pertaining to the mother", _pater-nus_ "paternal;
fatherly; belonging/pertaining to the father", _feli-nus_ "feline;
belonging/pertaining to the cat; belonging/pertaining to  the family/genus
of the cats/felidae" and the like.  Can it be traced back to IE? Does the
morpheme exist in other IE or non-IE languages? Any idea abt its etymology
and/or development?

Thanks in advance for your answers.
Cheers

Paolo Agostini


From ECOLING at aol.com  Mon Jul 26 15:12:27 1999
From: ECOLING at aol.com (ECOLING at aol.com)
Date: Mon, 26 Jul 1999 11:12:27 EDT
Subject: Hittites ~ Phrygians ~ Balkan peoples?
Message-ID: <MON.26.JUL.1999.111227.EDT.>

There has been much discussion of the Hittites and the Phrygians,
and other names for these or other peoples of Anatolia
and the Balkans (Thracians, Illyrians, Dacians, Moesians, etc. etc.),
which may have been incorrectly taken as distinct peoples in the past,
instead of simply as different names used for the same peoples
at different times or by different authors or cultures.

Can someone produce a summary of the argument for a new
way of viewing things in TABULAR form?

Or, since this is email, at least in condensed summary
of some sort?  Given the slight differences of opinions
on Dacia and Moesia around the lower Danube river,
there will of course have to be slight difference in such
summaries.  That will be healthy, so I hope more than
one contributor will respond.  Something like this???:

"Hittite"  name for X people during time period S...T
     Who used this name?
"Phrygian"  name for X people during time period U...V
     Who used this name?

  (with whatever extra notes are needed on overlaps of time and place)
    "Who used this name"  is intended mainly for use when the source was
     Herodotus or another pre-modern source)

and so on for other peoples and sets of names?

I would like to have some handy reference for understanding
both the debates themselves and also the implications of the
claims being made, which are often buried in the text of the
discussions, or evident only to specialists.

What I am thinking of is the difficulties of the entire
Illyrian-Thracian- etc. areas, whether the discussants seem
to be implying that some of these peoples were also the descendants
of Hittites or Celts or close relatives of theirs, as distinct from
later Slavic peoples, or mostly-lost branches of IE.

It is by now taken as fact (?) that the
Celts were early branches off from the IE stem at about the same
time as the Tocharians and shortly after the Hittites,
or however that might be more exactly put these days.

Similar threads are proceding on the ANE and the IndoEuropean email lists,
and arose there about the same time on the two lists.

Lloyd Anderson
Ecological Linguistics


From ECOLING at aol.com  Mon Jul 26 15:12:25 1999
From: ECOLING at aol.com (ECOLING at aol.com)
Date: Mon, 26 Jul 1999 11:12:25 EDT
Subject: Hittite & Celtic dative in /k/ ?
Message-ID: <MON.26.JUL.1999.111225.EDT.>

In the light of the discussions on Hittites and other early
Indo-European peoples of Anatolia and the Balkans,
going on on IE and ANE lists,
it may be relevant to consider whether even Celts and Hittites
may have shared any things linguistically or culturally,
as areal manifestations.

The standard analyses I think would still say that anything
shared between them is a relic of the common IE stage,
because Hittite and Celtic did not share any innovations together.
I am asking also the IndoEuropean list (when it resumes activities),
if we can get some comments and guidance as to the current
state of knowledge.  Obviously the work of Ringe has been
prominent recently.  Any current views of what Ringe has succeeded
in pinning down re Proto-IE and Hittite, Celtic, Tocharian,
and what Ringe's methods have not pinned down conclusively,
either because the choice of data determines the conclusions
or for other reasons?

Here it is one small item for your consideration.

Here is a tidbit linking Celtic and Hittite
which I found many years ago, when compiling typological
comparisons of the semantic domains of "be" and "have".
I wondered at the time whether it is a relic Dative case
preposition / postposition which Hittite and Celtic shared,
presumably as a retention, but conceivably as an areal phenomenon,
or both of the above.  Does Tocharian have it too?
(I have no idea whether the later
Indic Dative postposition with /k/ is related or a chance lookalike.)

The forms being discussed, the dative case of the 1st singular pronoun,
are statistically likely to be highly conservative,
both because they are pronouns and also
as oblique cases rather than the more often innovating nominative etc.

**********

Hittite:

amm-el  es-tsi     "(it) is mine"
amm-uk  es-tsi     "(it) is to me", or rather in English, "I have it"
me-Postpos.  is-3sg.

amm-uk "me-to"

The above forms were cited by Calvert Watkins for the contrast of
Genitive with Dative.

**********

Celtic preposition ag- where Hittite had postposition -uk above:

Irish
ta' ... aige     "he has", with the preposition   /ag-/ (Watkins)
ni fhuil fear agam  "not husband to-me" or rather "I have no husband"

ag-am "to-me"

Gaelic

tha airgiod agam  "is money to-me", or rather "I have money"

***

Lloyd Anderson


From 114064.1241 at Compuserve.com  Wed Jul 28 19:14:00 1999
From: 114064.1241 at Compuserve.com (Damien Erwan Perrotin)
Date: Wed, 28 Jul 1999 21:14:00 +0200
Subject: Latin perfects and Fluent Etruscan in 30 days!
Message-ID: <WED.28.JUL.1999.211400.0200.>

-----Original Message-----
 From  Eduard Selleslagh <edsel at glo.be>
Date Sun, 11 Jul 1999 12:31:58 +0200

>Ed said

>There seems to be a slight misunderstanding here: with "later Lat. 'am-'" I
>meant the beginning of words like 'amare', 'amicus', etc. , i.e. without
>the -b-, which - according to the reasoning above - would have transited via
>Etruscan, as opposed to those that came straight from PIE and preserved the
>(a)m(.)b-.

>Apart from all that, I never suggested that 'ambo', ambi-', Grk. 'amphi'
>(and Cat. 'amb', which proved to be unrelated) etc. were of Etruscan origin,
>only that they should be added to the data pool when looking at possible
>relationships of Etr. 'am(e)-' and Lat. 'amicus' etc. It seems to me that
>this led to an interesting discussion that yielded the possibility of the
>existance of two parallel paths: one 'directly' from PIE to Latin and one
>via Etruscan.

Ed

[Damien Erwan Perrotin]

The idea is interesting, but probably inexact as am- is not present only
in Latin but also in Lydian (ama : to love) and in Breton (afan : to
kiss from an older Brythonnic *ama). It is still possible that the
Lydian form was a borrowing from an Etruscan-like tongue of the Aegean,
but that is quite unlikely for Breton. So there can be only three
explainations for the ressemblance you point out :
a: chance ressemblance (always possible)
b: Etruscan borrowed the word from Latin or from Celtic
c: Etruscan is remotely linked to IE and this root is a remnant of this
old relationship.

Personnally, I favor the third thesis, but there is still work to do
before proving it.

Damien Erwan Perrotin


From Georg at home.ivm.de  Thu Jul 29 18:14:56 1999
From: Georg at home.ivm.de (Ralf-Stefan Georg)
Date: Thu, 29 Jul 1999 20:14:56 +0200
Subject: PIE vs. Proto-World (Proto-Language)
In-Reply-To: <006c01bed1a6$a63637e0$319ffad0@patrickcryan>
Message-ID: <THU.29.JUL.1999.201456.0200.>

>Pat responds:

>There is no *tangible* way for us to ever know whether languages arose
>monogenetically or polygenetically however most linguists, even when they
>deny its recoverability, have correctly weighed the odds of mono- vs.
>polygenesis, and subscribe to monogenesis. If a probablistically calculated
>hypothesis is "ideology", then everything done in historical linguistics is
>"ideology".

"most linguists" ???????????????
I'm flabbergasted. I would like to know *one* of those, who did/does what
you claim "most linguists" do.

Stefan Georg
Heerstrasse 7
D-53111 Bonn
FRG
+49-228-69-13-32


From rmccalli at sunmuw1.MUW.Edu  Thu Jul 29 20:06:00 1999
From: rmccalli at sunmuw1.MUW.Edu (Rick Mc Callister)
Date: Thu, 29 Jul 1999 15:06:00 -0500
Subject: PIE vs. Proto-World (Proto-Language)
In-Reply-To: <006c01bed1a6$a63637e0$319ffad0@patrickcryan>
Message-ID: <THU.29.JUL.1999.150600.0500.>

	Unless you're saying that language arose in the last 100,000 years
[i.e. before humans left Africa], then I don't think you can make a serious
claim that there are any true isolates. Basque and Burushaski are surely
related to other languages but the lumpers are going to have to work a lot
harder to prove it.
	My understanding is that language arose before humans left Africa,
so any claims of polygenesis would have to be examined among African
languages. Given that the only existing language families in Africa are
Niger-Kordofanian, Nilo-Saharan, Afro-Asiatic and the Khoisan languages
[which may be between 1and 5 families], it seems that the onus of proof is
on the polygenesists.

[snip]

>Ralf-Stefan wrote:

>> And, if I may insert this much to Pat's chagrin, the notion that human
>> language is a monogenetic phenomenon is aprioristic ideology. Even the
>> notion that every known language, as Basque, Burushaski or whatnot, has to
>> be related to some other language is ideology. "Having consonants" may be a
>> universal feature of human language, "Being related to some other lg."
>> simply is not.

>Pat responds:

>There is no *tangible* way for us to ever know whether languages arose
>monogenetically or polygenetically however most linguists, even when they
>deny its recoverability, have correctly weighed the odds of mono- vs.
>polygenesis, and subscribe to monogenesis. If a probablistically calculated
>hypothesis is "ideology", then everything done in historical linguistics is
>"ideology".

[snip]

Rick Mc Callister
W-1634
Mississippi University for Women
Columbus MS 39701


From Georg at home.ivm.de  Thu Jul 29 19:42:59 1999
From: Georg at home.ivm.de (Ralf-Stefan Georg)
Date: Thu, 29 Jul 1999 21:42:59 +0200
Subject: PIE vs. Proto-World (Proto-Language)
In-Reply-To: <Pine.SO4.4.05.9907211722250.23558-100000@rask>
Message-ID: <THU.29.JUL.1999.214259.0200.>

>On Fri, 16 Jul 1999, Ralf-Stefan Georg wrote:

>> "Having consonants" may be a
>> universal feature of human language, "Being related to some other lg."
>> simply is not.

>A fascinating thought:
>Imagine "Being related to some other lg." WAS actually a universal feature
>of (spoken) human language, wouldn't it then be the only true universal?
>In other words, if all spoken languages should prove to be derived from
>one common source, would we then be able to distinguish true language
>universals from other features shared by all languages ONLY because of the
>fact that these features were present in "proto-world"?

Yes, of course we would, given a little progress in the field of typology
and universals (but in some subareas we are close). We could when and only
when we succeed in *explaining* language universals, i.e. explain them as
*necessary* ingredients of a communication system serving all the purposes
human language does serve. If it can be shown that, in order to be a
functional, viable means of intra-group communication, a sign system will
have to share in a (however small) set of features, without which it could
hardly work, let alone stabilize, these universals would be universals
independent of contiguous inheritance from some proto (where they could be
as arbitrary as meaning-to-form-relations generally are).
Admittedly, research into universals has not yet reached the state where we
can tell all this, but that's one of the direction it is heading in (at
least, imho, should be). Compare this to the "grand unified theory"
physicists are hunting for.
But, theory aside, "being related to some other lg." is not a language
universal. Polygenesis is equally likely as monogenesis, I can't see any
reasons why the opposite should hold.

St.G.

Stefan Georg
Heerstrasse 7
D-53111 Bonn
FRG
+49-228-69-13-32


From Georg at home.ivm.de  Thu Jul 29 20:17:39 1999
From: Georg at home.ivm.de (Ralf-Stefan Georg)
Date: Thu, 29 Jul 1999 22:17:39 +0200
Subject: PIE vs. Proto-World (Proto-Language)
In-Reply-To: <l03130305b3b941708923@[209.147.211.21]>
Message-ID: <THU.29.JUL.1999.221739.0200.>

I don't think I'm following.

>	You raise the questions of what is relationship and what is the
>difference between language and speech.
>	All languages are related in the same sense that all humans are
>related.

Why that ? Language is a cultural item, a world-three-phenomenon (Popper),
basically an artefact (though one constructed less consciously than hammers
and anvils, admittedly). Humans are related by material relationships we
are endowed with by birth. Language has to be learned, in each and every
case; every single bit of them (basically uncontroversial; I add a tiny bit
of possibly controversial stuff: even the traits of languages which are
"universal" are learned).

> All humans settled on oral speech rather than sign language or
>flatulence for communication.

There is no need to view this as an invention, which, by its specificity
and ingenuousity, can only have taken place once. It may have taken place
several times, because oral speech is the one means of communication which
is in functional terms vastly superior to signs or flatulence. This is
because of the possibility of having a doubly articulated system (a finite
set of meaningless elements are combined to form a [still ]finite set of
meaningful ones, which can be used to form an infinite set of meaningful
utterances; do this with modulations of flatulence and you have your
language).
Because it is so vastly, tremendously practical, oral speech won the day
among humans with brains capable to process all the neuronal activity
necessary to control and handle it.

> Given that all human brains are wired for
>speech in the same way, this suggests that there is a degree of
>relationship.

I contest this given. Human brains are *capable* to process language, i.e.
sign systems complicated and open enough to do the things language does.
That any specifics of these systems are hard-wired in human brains is
sometimes asserted (and sometimes makes its way into the popular press in
these days of reductionism) but not necessarily demonstrated. Some believe
this, some don't, guess what I do ;-)

>	I don't think this question can be answered in absolute terms but
>it may possibly be answered in practical terms. Given that modern humans
>left Africa about 100,000 years ago, there is a good chance that all of
>these are ultimately related.

What has this time-span to do with the good chance you mention ? There is
also a good chance that modern humans, when heading out of Africa had
several, if not many, languages, culturally stabilized to be
group-defining, yet materially independent from each other. We don't know,
whether the one scenario or the other is nearer to the "truth". But I, for
one, find the polygenetic one equally possible, if not slightly more
appealing. First you have to have the brains (and, yes, the vocal tract
will help as well) to manage such a thing as language (not with anything
pre-wired-in, just the capability, i.e. the neuronal complexitiy needed).
Then you will find out with tremendous speed just how *handy*  such a thing
as articulate vocal language is for any kind of group activity. It may well
be the most useful thing humans ever invented (I use "invented" without
irony, but please be aware of the "world-three-phenomenon"-caveat I
introduced above, or maybe in one of the other posts I have been firing
today ...). So, this very usefulness, this simply ir-re-sis-ti-ble
usefulness will lead intelligent beings to develop such a thing, resp. to
make the more intelligent and powerful ones bestow it on their
in-group-members (which are pre-human phenomena, of course). Being able to
do it and finding it so ri-di-cu-lous-ly useful is enough to actually do
it. Some attempts are successful, some less so (i.e. some languages make
the step from intra-group codes to means of inter-group communication
aso.). This *could* be the story.
Or do you think the trick of banging the rocks together was invented only
once ? Or that of throwing meat into the fire ?

We should overcome the romantic view of seeing in language something as
"natural" as having 32 teeth. Language is there because it serves a
fantastic range of *functions*, and it is in terms of these functions that
we will eventually understand it.

PS: I know this is controversial, I have at least the whole MIT community
against me (Or better, lest this sounds pretentious: I would have them
against me if they knew who I am, which fortunately they don't ;-), and,
though this scares me a bit, I take the freedom to express on this informal
medium: eppur si muove ;-)

But, on the other hand, I know I'm not entirely alone ...

St.G.

Stefan Georg
Heerstrasse 7
D-53111 Bonn
FRG
+49-228-69-13-32


From Georg at home.ivm.de  Thu Jul 29 20:31:41 1999
From: Georg at home.ivm.de (Ralf-Stefan Georg)
Date: Thu, 29 Jul 1999 22:31:41 +0200
Subject: PIE vs. Proto-World (Proto-Language)
In-Reply-To: <Pine.SO4.4.05.9907231417430.6558-100000@rask>
Message-ID: <THU.29.JUL.1999.223141.0200.>

>> Even the
>> notion that every known language, as Basque, Burushaski or whatnot, has to
>> be related to some other language is ideology.

>I suppose it is common sense to believe that "isolates" like Basque or
>Burushaski at least have *dead* relatives. A language in colloquial
>use never stays ONE language, but will inevitably split up into several
>new languages. Which is exactly why "isolate" is merely a term
>used for languages that have not yet been proven to be related to other
>languages. Every single language has a mother and sisters. This is also
>true of PIE, of course.
>I could accept your statement above, if you inserted the word "living"
>between "other" and "language".

Granted, but I'd prefer "known" to "living".
It is of course correct, given the overall presence of language change,
interaction between human groups, spread of human groups over territories
too big to maintain uniformity of language from border to border (and all
the other things which happen to human groups over millennia) that, at some
time and place in history, languages were spoken by some groups which we
would - had we access to them - identify as relatives of Basque (in this
case, Aquitanian is of course a partly known instance of this)  or
Burushaski (the fragmentary attested Bru-Zha of some Tibetan sources may as
well be Burushaski's Aquitanian, but noone really knows).
I was rather aiming at the pool/set of known and identifiable languages and
the question whether they of necessity are classifiable into one stemma,
or, slightly different, whether all members of this set are classifiable
into *some* stemma with others. Some are, some aren't. This does not mean
that, e.g. for Basque, this will of necessity never be the case. Perhaps it
will be the case. But it hasn't happened yet, and to force it, just because
Basque (or Kusunda, or Nivkh, or Yenisseyan) *has* to fit somewhere, just
because it *cannot* be allowed that they remain "orphans forever" is, imho,
misguided. They can, such is life.

St.G.

Stefan Georg
Heerstrasse 7
D-53111 Bonn
FRG
+49-228-69-13-32


From Georg at home.ivm.de  Thu Jul 29 20:38:29 1999
From: Georg at home.ivm.de (Ralf-Stefan Georg)
Date: Thu, 29 Jul 1999 22:38:29 +0200
Subject: PIE vs. Proto-World (Proto-Language)
In-Reply-To: <Pine.SO4.4.05.9907211902570.24271-100000@rask>
Message-ID: <THU.29.JUL.1999.223829.0200.>

>> The link between Sanskrit and Latin or Classical Greek is so close that any
>> layman can see it; whole phrases are nearly identical.

>> This is, to put it mildly, not the case between, say, English and Bengali.

>X
>But if we only knew the modern IE languages, we would still be able to
>reconstruct words like:
>*p at ter 'father'
>(on the basis of, say, English <father>, Italian <padre>, and Hindi
><pita>)
>*newos 'new'
>(on the basis of, say, Modern Greek <neos>, Portuguese <novo>, and Polish
><nowy>)

I doubt that we could *reconstruct* anything on the basis of these isolated
examples, but we could of course suspect that something with these
languages seems to be different than with, say, Burushaski and Quechua. At
least one thing is sure: we'd had a great deal of discussion about chance
resemblances and stuff at our hands as we have now. We would have a hard
time to tell the hows and whys of pairs like /father/: /padre/ but also e.
/paternal/ aso. aso.

>The same is true for morphological paradigms etc.

Well, I would love to see morphological paradigms reconstructable on the
basis of English, Italian and Hindi ...

Stefan Georg
Heerstrasse 7
D-53111 Bonn
FRG
+49-228-69-13-32


From X99Lynx at aol.com  Thu Jul 29 20:48:40 1999
From: X99Lynx at aol.com (X99Lynx at aol.com)
Date: Thu, 29 Jul 1999 16:48:40 EDT
Subject: PIE vs. Proto-World (Proto-Language)
Message-ID: <THU.29.JUL.1999.164840.EDT.>

In a message dated 7/29/99 2:27:35 PM, you wrote:

<<All humans settled on oral speech rather than sign language or
flatulence for communication.  Given that all human brains are wired for
speech in the same way...>>

Not a given some of us accept.  But given common human physiology in general,
the fact that sound travels farther and faster than hand signals
and that the vocal cords and jaw are more controllable than other body parts
that make sounds - oral speech is by far the most practicable way to
communicate before the invention of the written word (which makes use of
another part of our body that has more sensitive controls.)

<<Given that modern humans left Africa about 100,000 years ago,...>>

There's the rub.  You'll need to backtime that a bit.  And whether
genetically endowed by a miraculous quirk of nature or invented by us
ourselves, a universal first language would have to go back far enough to
include ancestors of the Chinese - and that may be going back a ways.

Was one day a child born with pre-wired language to non-lingual parents and
did she/he essentially sit around talking to her/himself until a similar
sibling showed up and the first conversation happened?
Or was language invented in one place at one date and spread from there?   Or
did it arise as human communities and cultures reached the point where
assorted communal grunts, groans, clicks and fricatives became customary and
structured and were passed on for the immense practical value of
understandable communication.  If the latter, then language could well have
developed in different places at different times.

Regards,
Steve Long


From larryt at cogs.susx.ac.uk  Fri Jul 30 07:54:17 1999
From: larryt at cogs.susx.ac.uk (Larry Trask)
Date: Fri, 30 Jul 1999 08:54:17 +0100
Subject: PIE vs. Proto-World (Proto-Language)
In-Reply-To: <008201bed1aa$79bc9480$319ffad0@patrickcryan>
Message-ID: <FRI.30.JUL.1999.085417.0100.>

On Mon, 19 Jul 1999, Patrick C. Ryan wrote:

[on monogenesis]

> The "persuasive case" has already been made. Monogenesis is much likelier
> than polygenesis. And you have written so yourself!

No, I haven't.

What I have said is that we have no evidence either way, but that the
majority of linguists probably lean toward monogenesis.

I myself am agnostic.  Monogenesis appeals to me because it is
economical, but there is no way on earth I could defend monogenesis
against someone who doubted it.

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk


From larryt at cogs.susx.ac.uk  Fri Jul 30 08:14:22 1999
From: larryt at cogs.susx.ac.uk (Larry Trask)
Date: Fri, 30 Jul 1999 09:14:22 +0100
Subject: PIE vs. Proto-World (Proto-Language)
In-Reply-To: <Pine.SO4.4.05.9907211722250.23558-100000@rask>
Message-ID: <FRI.30.JUL.1999.091422.0100.>

On Wed, 21 Jul 1999, Adam Hyllested wrote:

> On Fri, 16 Jul 1999, Ralf-Stefan Georg wrote:

>> "Having consonants" may be a
>> universal feature of human language, "Being related to some other lg."
>> simply is not.

> A fascinating thought: Imagine "Being related to some other lg." WAS
> actually a universal feature of (spoken) human language, wouldn't it
> then be the only true universal?  In other words, if all spoken
> languages should prove to be derived from one common source, would
> we then be able to distinguish true language universals from other
> features shared by all languages ONLY because of the fact that these
> features were present in "proto-world"? Language typology would
> still be a great help in the work of reconstructing proto-languages,
> but would it prove anything about the nature of spoken human
> language?

An interesting point.  This is the problem of `founder effects'.  A
founder effect is a feature of an ancestral language which is not
cognitively necessary -- that is, it is not a true language universal --
but which happens by chance to persist in all descendants of that
ancestral language.  Clearly founder effects have the potential to
distort our understanding of true linguistic universals.

This idea has received some discussion in the literature, though not, I
think, a great deal.  Johanna Nichols has appealed to founder effects to
account for the significantly non-random distribution of certain
structural and other features among the world's languages.  Nichols has
no interest in "Proto-World", but I can cite another case which is
relevant.

When I started my PhD some years ago, no example was known of a language
with object-initial basic word order (OVS or OSV), and it was widely
suspected that such languages were impossible.  Since then, a few
O-initial languages have turned up, practically all of them in Brazil.
If we hadn't managed to study those languages before they vanished, we
might have gone on believing forever that O-initial languages were
impossible.

Well, they are indeed possible, but they appear to be rare.  But why are
they rare?  Geoff Pullum once indulged in some musings on this point.
Maybe O-initial languages really are cognitively or functionally more
difficult than other types.  Then again, maybe they are not, and the
observed distribution is an accident, an artefact of human history:
speakers of non-O-initial languages have simply been lucky enough to
flourish at the expense of speakers of O-initial languages.  If the
second is the case, then we are looking at a kind of founder effect.
But who knows?

There exists a principle, called the Exit Principle, which says this:
any feature of a language which is not cognitively necessary can change.
This implies that every such feature *will* change eventually, but who
knows how long `eventually' might be?

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk


From larryt at cogs.susx.ac.uk  Fri Jul 30 08:17:43 1999
From: larryt at cogs.susx.ac.uk (Larry Trask)
Date: Fri, 30 Jul 1999 09:17:43 +0100
Subject: PIE vs. Proto-World (Proto-Language)
In-Reply-To: <Pine.SO4.4.05.9907231417430.6558-100000@rask>
Message-ID: <FRI.30.JUL.1999.091743.0100.>

On Fri, 23 Jul 1999, Adam Hyllested wrote:

> On Fri, 16 Jul 1999, Ralf-Stefan Georg wrote:

>> Even the
>> notion that every known language, as Basque, Burushaski or whatnot, has to
>> be related to some other language is ideology.

> I suppose it is common sense to believe that "isolates" like Basque
> or Burushaski at least have *dead* relatives. A language in
> colloquial use never stays ONE language, but will inevitably split
> up into several new languages. Which is exactly why "isolate" is
> merely a term used for languages that have not yet been proven to be
> related to other languages. Every single language has a mother and
> sisters. This is also true of PIE, of course.

> I could accept your statement above, if you inserted the word
> "living" between "other" and "language".

Well, `known' rather than `living', I would suggest.  I took Stefan's
statement to mean `known language', since I cannot conceive of any
program relating known languages to unknown languages.

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk


From JoatSimeon at aol.com  Fri Jul 30 20:30:28 1999
From: JoatSimeon at aol.com (JoatSimeon at aol.com)
Date: Fri, 30 Jul 1999 16:30:28 EDT
Subject: PIE vs. Proto-World (Proto-Language)
Message-ID: <FRI.30.JUL.1999.163028.EDT.>

>adahyl at cphling.dk writes:

>But if we only knew the modern IE languages, we would still be able to
>reconstruct words like:

-- if it had ever occurred to anyone that the languages were related.  Note
that the whole concept is only about 200 years old, and that comparative
linguistics only really got started in the 19th century.

In any case, the number of reconstructions possible with only extant
languages would be very much smaller.

Linguistic information is lost over time. The more time, the more lost.


From JoatSimeon at aol.com  Fri Jul 30 20:38:04 1999
From: JoatSimeon at aol.com (JoatSimeon at aol.com)
Date: Fri, 30 Jul 1999 16:38:04 EDT
Subject: PIE vs. Proto-World (Proto-Language)
Message-ID: <FRI.30.JUL.1999.163804.EDT.>

>(Patrick C. Ryan)

 > -- with minor exceptions (such as "kuku"), words _are_ arbitrary sound
 > assemblages.

>Pat writes:

>Why do you not explain to us all why that is true? I am firmly convinced
>that it is unequivocally incorrect.

-- because any sound within the human range will do as well as any other for
any given referent.  A rose may be a znfargle.

Humans have the capacity to learn a language, but not any particular
language; a language must have words, but not any particular words.


From kurisuto at unagi.cis.upenn.edu  Sat Jul 31 05:08:57 1999
From: kurisuto at unagi.cis.upenn.edu (Sean Crist)
Date: Sat, 31 Jul 1999 01:08:57 -0400
Subject: PIE vs. Proto-World (Proto-Language)
In-Reply-To: <006c01bed1a6$a63637e0$319ffad0@patrickcryan>
Message-ID: <SAT.31.JUL.1999.010857.0400.>

On Mon, 19 Jul 1999, Patrick C. Ryan wrote:

> There is no *tangible* way for us to ever know whether languages arose
> monogenetically or polygenetically however most linguists, even when they
> deny its recoverability, have correctly weighed the odds of mono- vs.
> polygenesis, and subscribe to monogenesis.

I don't think this correctly represents the field; I'd say that most
linguists are agnostic on this question.  The majority view is that there
isn't evidence to allow us to answer the question one way or the other. In
any case, it's not a question that most linguists spend a lot of time
pondering.  There was so much unverifiable silliness written on this topic
earlier in the history of linguistics that the whole topic has a rather
disreputable air to it.

> If a probablistically calculated
> hypothesis is "ideology", then everything done in historical linguistics is
> "ideology".

A probabilistic model is one in which each of the possible outcomes of an
event has a particular numeric probability assigned to it.  The
Comparative Method is not probabilistic; it is categorical.

I can't see any sense in which the monogenesis hypothesis is
"probabilistically calculated".  One can imagine that there could be a
probabilistic model of language genesis, but there are so many unknowns
that such a model would likely not be of much use.  It would probably be
worse off than (for example) the Drake formula for calculating the
probability that there is intelligent life elsewhere in the universe; the
fact is that we just don't know the answer, and there's nothing wrong with
saying that we don't know.

  \/ __ __    _\_     --Sean Crist  (kurisuto at unagi.cis.upenn.edu)
 ---  |  |    \ /     http://www.ling.upenn.edu/~kurisuto/
  _| ,| ,|   -----
  _| ,| ,|    [_]
   |  |  |    [_]


From petegray at btinternet.com  Sat Jul 31 10:00:11 1999
From: petegray at btinternet.com (petegray)
Date: Sat, 31 Jul 1999 11:00:11 +0100
Subject: PIE vs. Proto-World (Proto-Language)
Message-ID: <SAT.31.JUL.1999.110011.0100.>

Rick said:
>. DNA studies obtensibly show that non-African
> humans seem to go back a single population distinct from Africans.

"Distinct" from Africans?   Not in my reading of the texts.  I don't wish to
raise a non-linguistic topic, but your implication - that languages split
into African and non-African - is a language topic, and we need some
scientist out there to give us the DNA truth.

Peter


From proto-language at email.msn.com  Thu Jul 29 21:12:30 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Thu, 29 Jul 1999 16:12:30 -0500
Subject: PIE and Proto-Language
Message-ID: <THU.29.JUL.1999.161230.0500.>

Dear Ralf-Stefan and IEists:

 ----- Original Message -----
From: Ralf-Stefan Georg <Georg at home.ivm.de>
Sent: Monday, July 19, 1999 4:36 AM

>> Pat interjects:

>> Here I think you are dangerously introducing the mistaken terminology of the
>> opposing argument. I do *not* look for "look-alikes"; I only am interested
>> in "cognates" the phonological forms of which can be supported through
>> multiple comparisons. For example, one set of interesting IE and Sumerian
>> "cognates" shows IE *-wey- = Sumerian -g{~}-; interestingly, this
>> development *is* found in *some* IE languages, like nearby Armenian.

Ralf-Stefan responds:

> "interestingly" ? Why interestingly ? The only way this could be
> "interesting" would, imho, be to base a claim of Leskienian solidaric
> innovation on it, leading to a Sumero-Armenian subgroup of Indo-European.
> Very interesting.
> Or some kind of areal phenomenon, potentially interesting (this time
> without irony). But: supposing for a split-second that IE and Sumerian are
> somehow related, the /w/ --> /g/ shift would be something very old, right?
> Now, the Armenian /w/ ---> /g/ shift is young, since known loanwords
> participate in it (so young actually, that Kartvelian managed to preserve
> the intermediary stage /gw/ in some of those LWs). And of course, even if
> chronology were no obstacle, the whole thing still begs the question of
> Sumero-IE, but I understand that this is a) not a topic for this list and
> b) that I don't stand the slightest shred of a chance to talk you out of
> it; so I don't try.

Pat responds:

I do not know of any ready way tp gauge how old or young the phenomenon is
in Sumerian. But, I regard it probably as an areal phenomenon.

Also, as I perface every essay I have written with a caveat against assuming
that I regard IE and any compared language as particularly close, there is
no real question of "Sumero-IE" for me. On the other hand, Gordon Whittaker
in Goettinger Beitraege zur Sprachwissenschaft feels rather differently
about it. His idea is that Sumerian was imposed on an IE substratum. His
article is interesting for the many valid comparisons he makes (IMHO)
between some Sumerian words and IE ones.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From Georg at home.ivm.de  Thu Jul 29 09:52:49 1999
From: Georg at home.ivm.de (Ralf-Stefan Georg)
Date: Thu, 29 Jul 1999 11:52:49 +0200
Subject: accusative and ergative languages
In-Reply-To: <001e01bed1af$abebec80$629ffad0@patrickcryan>
Message-ID: <THU.29.JUL.1999.115249.0200.>

>S wrote:

>> How do you live with the fact that some "ergative languages" have
>> independent passives ?

>Pat asks:

>Generalities are somewhat interesting but specifics would be even more
>interesting.

Sorry, deeply sorry, but I'm not able to believe that we are discussing
ergativity here, the existence of splits of some sort in every language
conventionally dubbed "ERG", the question whether ERG-constructions are
really passive constructions and stuff like that, without you knowing that
the phenomenon I alluded to in passing exists. I don't believe it.
I don't think that anyone here will need examples for this, so I'll only
give them when you explicitly express that you doubt/know better/reject the
fact that ergative constructions and passives can coexist in a language.
One hint to spare you some energy: the name of the first language which
comes to my mind in this respect sounds pretty close to my last name ...

>> And, as far as I remember, this whole brouhaha started with me asserting
>> that all known "ergative languages" have some split, followed by you
>> prompting we to show such a thing in Sumerian. I'm not a mind-reader, but
>> this could be interpreted as a challenge aiming at the gist of my
>> assertion, n'est-ce pas ? I managed to show those splits,

>Pat interjects:

>To my knowledge, you did not.

I did, Larry did, Wolfgang did (put me last on this list). However, I never
actually expected you to concede this, nor will you ever, no matter what.
So why should I bother to try to make you change your mind ? Each time
something is shown which does not square with your notion of ergativity (or
inflection, or whatnot) you change the definition. Let's begin (all over
again, sigh) by answering Wolfgang's questionnaire. I think this will be
the only way to get somewhere. You start, of course.

>> although the
>> discussion got a bit swamped under some hassle over maru:- and
>> HamTu-conjugations, but I managed to do it. At least I got you to accept,
>> late in coming, though, that my initial assertian still stands up.

>Pat responds:

>I did *not* accept your initial assertion. Re-read my concession.

Well, you are right here. Actually, if you *had* accepted it, I would have
to look for a flaw in *my* argumentation ;-)

St.

Stefan Georg
Heerstrasse 7
D-53111 Bonn
FRG
+49-228-69-13-32


From Georg at home.ivm.de  Thu Jul 29 10:47:45 1999
From: Georg at home.ivm.de (Ralf-Stefan Georg)
Date: Thu, 29 Jul 1999 12:47:45 +0200
Subject: accusative and ergative languages
In-Reply-To: <006101bed1a5$98f8be00$319ffad0@patrickcryan>
Message-ID: <THU.29.JUL.1999.124745.0200.>

>Pat responds:

>I suggest you read it. "But we do still encounter scholars who insist that
>there is a necessary diachronic connection, e.g. Estival and Myhill
>(1988:445): 'we propose here the hypothesis that in fact all ergative
>constructions have developed from passives'."  Are you suggesting that
>Estival and Myhill are not "linguists", or that Shibatani, in whose book
>this essay appeared, is not a "linguist"?  The quotation above is from p.
>189 of Dixon's 1994 book. What are you playing at?

Linguists can do wrong.

>Pat responds (on L.T's catalogue of subject properties):

>Why do you not give us an example of these so-called subject properties in
>some ergative language besides Basque? And how were these properties
>selected?

Empirically. What Larry gave was a *selection*, as you correctly name it.
It is so easy, you can do it yourself. S.b.

>Pat responds:

>If you are referring to the "tests" above, you have proved nothing.

This is much in line with what I said in  today's other post of mine. You
see, you read, you understand (or don't), but you reject nevertheless.
Always.
The subject-property-test *is* sufficient to discriminate between ergative
and passive constructions. Again:
Take an intransitive sentence in an exclusively accusative language. S-V
S is, by convention, called "subject" (just a convention). It's semantic
role can be one of the following: Agent, Experiencer aso. (some people
insist on Agent being restricted to agents of transitive constructions, bog
s nimi).
Now, take a transitive one. A-V-O
What accusative languages do is to treat S *and* A alike under all
circumstances (there are such languages, i.e. split-free accusative-lgs.).
This leads to the notion that these languages know a macro-category
comprising S and A, which most conventionally is called "subject" also, in
compliance with western grammatical tradition.
Now subjects have properties of various sorts, part of which have been
enumerated by Larry Trask. The real list is longer, and you can expand it
yourself. It is nothing more than a list of things which are generally true
for "subjects" in these languages.
Now, part of the things a passive formation does to a sentence is that the
*subject* is now what was semantically the undergoer/patient (choose your
favourite term) of the corresponding active sentence, an example:

active: Stefan proves Pat wrong (active, A-V-O)
           Pat is proven wrong by Stefan (passive, S-V[pass]-Agent-periphrasis)
"Pat", in this passive sentence, has been moved to a position (not only in
terms of word order), where it assumes all the functions commonly found
with S. It has become the "subject" of the sentence, according to
traditional terminolgy. The syntactic properties (partly enumerated by
Larry, but the list is longer) found with "Stefan" in the active sentence
are now found with "Pat" in its passive equivalent.

Now let's look at ergative constructions:

Again, we find intransitive sentences there: S-V (gosh, I'm starting from
scratch, but I think I have to)
And transitive sentences as well: A-V-O. What makes the construction
ergative is, of course, nothing but the fact that S and O are treated alike
in this language (or in some subsystems of this language).
Now, in a morphologically ergative language, such a sentence may look like:

A-erg V O-abs
a passive sentence, we recall, looks like (word order irrelevant): S V
Agent-periphrasis (P. is proven wrong by S., to take a random example with
positive truth-value, however;-).
Of course, ergative constructions and passives bear some resemblance: in
both the (semantic !) Agent is morphologically marked, in both the semantic
patient may be morphologically unmarked (this does not work when there is
an overt accusative marker in the language, of course).  However, this is
where the resemblance ends. In an ergative sentence like:
Stefan-Erg proves Pat-Abs wrong we should expect, under the
ergative-as-passive scenario, that subject properties remain with Pat (as
in the "true" passive above); however, they rest with "Stefan".
This is the general picture only, and the test has to be carried out with
real "ergative languages" of course, but this is what you will find. It is
nothing less that the watershed betweed ergative constructions and passive
constructions. It does not matter whether you accept the full list of
"subject properties" enumerated. You may accept some and dislike others.
With those you do accept, though, you will find this picture. The more
subject properties you can bring yourself to accept, the more instances of
the difference between ergative and passive constructions you will find.
It's that easy. But you do have to look at complex constructions, not only
at the minimal sentence.
This does *not* mean that ergativity in some given language may not have
come about via a passive transformation. It may. In order to claim the
right of being accepted as a "real" ergative construction (and not the
passive it may have been in an earlier life), the shift of subject
properties to the Agent-phrase (the ERG-NP) from the Patient-phrase is
crucial.
For a succinct demonstration that, what looked to some people like an
ergative construction, but is rather to interpreted as passive, see my (and
A.P. Volodin's) "Die itelmenische Sprache", Wiesbaden 1999, though I admit
that we could have been a bit clearer on the issue. If anything, this
thread has prompted me to write a clearer exposition of the relevant
chapter in this book.

>Pat responds:

>Obviously, I do not think so. And Estival and Myhill (and probably
>Shibatani) do not either --- not to mention the majority of linguists of the
>past.

Estivall/Myhill/Shibatani should speak up for themselves. As for the
linguists of the past, well, they are linguists of the past, if they had
got everything straight already in 1890, what the hell are we doing here (I
assume, that's what you ask anyway reading all this ;-).

Stefan

Stefan Georg
Heerstrasse 7
D-53111 Bonn
FRG
+49-228-69-13-32


From vidynath at math.ohio-state.edu  Thu Jul 29 12:31:37 1999
From: vidynath at math.ohio-state.edu (Vidhyanath Rao)
Date: Thu, 29 Jul 1999 08:31:37 -0400
Subject: accusative and ergative languages
Message-ID: <THU.29.JUL.1999.083137.0400.>

I wrote:
> (2) ra:men.a+as'vam a:ruhyate
>    [A] horse is being mounted by R.

What a horrid mistake. It should be ra:men.a+as'va a:ruhyate, with as'va in
the nominative [loss of the ending s by sandhi].


From proto-language at email.msn.com  Thu Jul 29 13:41:17 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Thu, 29 Jul 1999 08:41:17 -0500
Subject: accusative and ergative languages
Message-ID: <THU.29.JUL.1999.084117.0500.>

Dear Stefan and IEists:

 ----- Original Message -----
From: Ralf-Stefan Georg <Georg at home.ivm.de>
Sent: Monday, July 19, 1999 6:42 AM

Pat wrote:

>> In English, I can say: "John hit me, and he went away" or "I hit John and he
>> went away". What is the discourse cohesion strategy for English?

R-S wrote:

> Anaphora.

Pat writes:

Well, 'anaphora' is primarily a rhetorical device; my dictionary, however,
does acknowledge the use of 'anaphora' in a grammatical sense although Larry
seems to prefer "anaphor" in the grammatical and, I presume (but do not
know), 'anaphora' in the rhetorical sense.

Since "discourse cohesion strategy" is not defined in Larry's dictionary, I
have no idea exactly how one will want to define it. Perhaps 'anaphors' are
excluded; perhaps not.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From proto-language at email.msn.com  Thu Jul 29 16:05:15 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Thu, 29 Jul 1999 11:05:15 -0500
Subject: accusative and ergative languages
Message-ID: <THU.29.JUL.1999.110515.0500.>

[ moderator re-formatted ]

Dear Peter and IEists:

 ----- Original Message -----
From: petegray <petegray at btinternet.com>
Sent: Monday, July 19, 1999 2:41 PM

> Pat said:

>> ... the pattern "e-a-o" *...There was never a time during which *any* verb
>> in [e] followed this "pattern", or even *any* verb in /eR/ followed it
>> (weNden, waNdte, gewaNdt).

Peter wrote:

> Perhaps I misunderstand you.   This pattern is one of the standard patterns
> of strong verbs in modern German, most prevalent among verbs in -eR-, but
> not restricted to them, e.g.:

> befehlen, befahl, befohlen             to command
> bergen, barg, geborgen                to salvage
> nehmen, nahm, genommen          to take

Pat responds:
Peter, I majored in German so I am familiar with these "patterns".

Like so much of what we seem to do on this list, it is a question of
definitions: in this case, 'pattern'.

I confess that the definition I was using when I made the statement above is
1) idiosyncratic (Platonic[?]), and 2) possibly unsustainable.

Mixed up in it was the idea of '*general* applicability', and I can see now
that it is a definition of extremely limited usefulness.

I withdraw my statement, and grant your point.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From proto-language at email.msn.com  Thu Jul 29 16:29:26 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Thu, 29 Jul 1999 11:29:26 -0500
Subject: accusative and ergative languages
Message-ID: <THU.29.JUL.1999.112926.0500.>

Dear Larry and IEists:

 ----- Original Message -----
From: Larry Trask <larryt at cogs.susx.ac.uk>
Sent: Monday, July 19, 1999 5:49 AM

> On Sat, 17 Jul 1999, Patrick C. Ryan wrote:

>> As for the far-reaching conclusions of Dixon based on discourse
>> cohesion strategies, in my opinion they are flawed because these
>> strategies are purely conventional.

Larry responded:

> Not so.  Read on.

Pat responds:

I am going to defer a specific comment pending reading your answer to:
Schulze (" Language systems ARE conventional! This is one of the major
points of
 language tradition and L1 acquisition (despite of minimalism etc.)." in
another posting to this list.

Pat wrote:

>> In English, I can say: "John hit me, and he went away" or "I hit
>> John and he went away". What is the discourse cohesion strategy for
>> English?

Larry responded:

> You've overlooked the crucial null-subject cases:

> `John hit me and went away' *must* mean `John went away'.

> But `I hit John and went away' *must* mean `I went away'.

> Control of null NPs is one of the syntactic properties which crucially
> distinguish subjects from non-subjects in English and in some other
> languages.  And there is nothing "conventional" about it: this is a rule
> of English syntax.

Pat responds:

Another word for a "rule" is a "convention". Again, we appear to be chasing
our tails (definitions).

And the point that Leo makes in another posting deserves to be considered:

"Agreed -- and yet not quite agreed.  Are these really "subject
 properties"?  I freely acknowledge that that's what they're called, but
 for mainly
 for lack of any better name in normal linguistic terminology."

Larry continued:

> Let me change both NPs to third-person, to avoid any complications with
> agreement; suppose I say this:

> `John hit Bill and went away'.

> Now, in English, it is John who went away, not Bill.  However, according
> to my understanding of Dixon, if you say what looks like the literal
> equivalent of this in Dyirbal, it is *Bill* who went away, not John.
> This is one of the ways in which syntactic ergativity manifests itself
> in Dyirbal.

Pat answers:

Is it really that simple? Would we not be coming close to describing both
'nominative' and 'ergative' phenomena if we said that "a null-NP frequently
refers to the foregoing NP with which it agrees in case"?

 Larry continued:

> But not all ergative languages are the same here, and probably not even
> most.  If you translate this sentence as literally as possible into
> Basque, it is once again John who went away, and not Bill, just as in
> English.  This is so even though the ergative morphology of Basque is
> more thoroughgoing than that of Dyirbal, which is split.

> That is, Basque, like English, allows subjects to be coordinated with
> subjects, but not with non-subjects.  This is so even when one of the
> coordinated subject NPs is ergative and the other absolutive.  Basque
> does not allow the absolutive subject of an intransitive sentence to be
> coordinated with the absolutive object of a transitive sentence.

Pat responds:

First, as you know, I have questioned whether Dyirbal is, in fact, "split",
and have proposed a different explanation for the data.

Second, the point that Leo is making about distinguishing syntactic from
morphological subjects needs to be addressed by you in terms of "subject"
properties.

 Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From proto-language at email.msn.com  Thu Jul 29 17:09:10 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Thu, 29 Jul 1999 12:09:10 -0500
Subject: accusative and ergative languages
Message-ID: <THU.29.JUL.1999.120910.0500.>

Dear Leo and IEists:

 ----- Original Message -----
From: Leo A. Connolly <connolly at memphis.edu>
Sent: Monday, July 19, 1999 3:35 PM

Pat writes:

I do not want to divert you both from the discussion you are having about
related points, which I find very interesting, but I would like to comment
on one point.

Larry wrote:

>> The "passive" view of ergative languages in general is indefensible.

Leo wrote:

> Absolutely agreed!

Pat writes:

Again, I feel that it is again a battle of definitions. If by 'passive' we
mean a two-element construction in which the non-ergatively marked NP is the
patient of the VP, then an 'ergative' language like Sumerian is 'passive'.

Part of the problem, I think, is utilizing concepts like Larry's
"intrinsically transitive". 'Transitive' can be defined either as a
three-element construction or a construction of two elements with an implied
or differently expressed third element ("passive-with-agent"). Since I do
not believe in phenomena ex machina, I believe that all VP's essentially
include a causal factor (one-element) and an effectual one (one-element) so
that expressed or unexpressed, every VP implies two additional elements.

So, to my way of thinking, an 'intransitive' sentence like 'I go to the
city' is, from the standpoint of analysis, better regarded as a paraphrase
for 'I make myself move to the city'.

And a 'stative' like 'The flower is red' is better analyzed as a paraphrase
for 'Somebody/thing has reddened the flower'.

This analysis makes it easy to understand why, in an ergative language, the
'subjects' of 'intransitives' and 'statives' are in the absolutive. It
suggests that in an 'intransitive' construction, the expressed NP is not
necessarily the agent but may be the patient of a reflexive construction in
which the agent, being identical with the patient, has been deleted for
redundancy.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From proto-language at email.msn.com  Thu Jul 29 18:18:56 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Thu, 29 Jul 1999 13:18:56 -0500
Subject: accusative and ergative languages
Message-ID: <THU.29.JUL.1999.131856.0500.>

Dear Wolfgang and IEists:

 ----- Original Message -----
From: Wolfgang Schulze <W.Schulze at lrz.uni-muenchen.de>
Sent: Tuesday, July 20, 1999 5:15 AM

Wolfgang wrote:

> 16. Do YOU use a special theoretical frame work in order to substantiate
> your claims? If yes, which one?

Pat responds:

This is an excellent suggestion --- theoretically --- but I do not think it
is going to happen on this list. Answering your pertinent questions as they
should be answered would involve writing a book at least the size of
Dixon's. But, good thought!

Wolfgang continued:

> Now, let me finally turn to some claims by Pat:

> Pat wrote:

> Wolfgang wrote:

>>> Hence, there are NO "ergative (or "accusative") languages" or
>>> only, if you use this term in a very informal sense.

>> Pat wrotes:

>> In the sense you are using these, they seem to be of little value in
>> describing anything.

Wolfgang responded:

> I can only repeat what I have said in my earlier postings: ACC and ERG
> are nothing but the structural (and structured) reaction to more general
> principles of language based information processing. The polycentric
> architecture of language systems that encode these principles allows
> that the individual centers of this polycentric cluster react
> differently on these principles. Here, I cannot elaborate the underlying
> frame work which is labeled "Grammar of Scenes and Sceanrios" (GSS) and
> documented in Schulze 1998, chapter IV, but let me briefly say that I
> propose a (more or less) universal cause-effect 'vector' (C->E) itself
> metaphorized from underlying figure-ground relations (F->G) to be one of
> the most dominant principles of information processing. This vector can
> be weighted which leads to a continnum (here shortened) C->0(e) C->e >
> C->E > c->E > 0(c)->E (capitals represent heavy domains, small letters
> represent light domains). An ACC strategy would be to behave in a C->e
> sense, an ERG strategy would infer c->E. Note that these vector
> representations are NOT (in themselves) language specific or sensitive
> for specuial (linguistic) categories! Different categories may BEHAVE
> differently with respect to this continuum, regardless their own
> architectural make-up. ANYTHING in a language system may be sensitive
> for the AEC (accusative ergative continuum) as long as it is relevant
> for encoding the cause-effect vector (and its derivations). Hence, CASE
> may play a role just as AGR, word order, subject assignment,
> topicalization, discourse cohesion, co/subordination, paradigmatization
> of speech act participants and much more. But they may play their roles
> DIFFERENTLY! The description of their roles heavily depends from the
> diachrony of the paradigm in question, its formal architecture as well
> as its integration in a co-paradigmatic context ('structural coupling'
> in a broader sense). From this it follows that the individual centers of
> ALL language systems have to react upon the universal demands of the
> C->E vector, disregarding their internal architecture. Only IF ALL
> relevant centers (and they are many, I grant!) behave in one direction,
> THEN we are allowed to call the language system (or better, its
> Operating System) ACC or ERG. However, it is a much more difficult task
> to describe intermediate states that allow ACC in some parts of the
> Operating System, and ERG in some others. Here, we have to establish a
> (motivated!) hierarchy first of co-paradigmatic structures (such as CASE
> and AGR, CASE and word order, AGR and Personality, AGR and Noun Classes,
> to name only some). In such structures, one part sometimes is more
> dominant than the other with respect to its behavior on the AEC. If we
> can describe such dominant behavior we can refer to the whole structure
> as either more ACC or ERG. In a second step we have to go on describing
> the higher levels of this hierarchy (which itself should find an
> adequate linguistic explanation based on an appropriate language
> theory). Finally (and ideally) we would arrive at a term that would
> describe the functional dominance of one of the poles on the AEC with
> respect to an Operating System (not a language system) in toto. Only
> then, and I stress, only THEN we are allowed to use the term ergative or
> accusative with respect to an Operating System (for which 'ergative' or
> 'accusative language' would be an informal label).

Pat responds:

It would take my background than I have to properly apperciate your argument
above but, from what I *can* understand of it, I concur. As I wrote in
another post, and I hope I understand the above well enough to validly
connect it, I believe the critical factor is how cause-effect is mapped but
that the underlying cause-effect is the reality behind the masks.

>> Wolfgang wrote further:

>> Pat wrote:

>> I am sorry that I do not agree with the validity of this distinction (actant
>> vs. agent). For me, 'actant' is 'agent'. Perhaps you can explain the
>> difference.

Wolfgang explained:

> In terms of Functinal Grammars (as well as in GSS) 'actants' refer to
> ALL such linguistic expressions that encode a referential entity in a
> clause. Hence, in a sentence such as 'I met John several times in
> Chicago', 'I', 'John', 'times', and 'Chicago' are (abstract) actants
> that play different roles in the scene. But only 'I' is a linguistic
> agent, whereas John plays the role of a patient etc. Note that 'agent'
> and 'patient' are labels for semantic hyperroles (or macroroles in the
> sense of Foley/VanValin). I think that such a distinction is very
> helpfull. It is based on strong theoretical arguments and helps to avoid
> many false or at least problematic generalizations. A much more
> controversial (and much more difficult question is to define the labels
> 'subjective' (S), 'agentive' (A), and 'objective' (O) which should not
> be immediately equated to neither 'subject'/'object' nor to
> 'agent'/'patient'. S, A, and O are highly abstract terms that describe
> more structural than semantic or syntatic properties.

Pat responds:

Although I do not care much for 'actant' and think it has little to
recommend it over, say, NP, it is, of course, the prerogative of theorists
to introduce and define new terms as they choose.

Otherwise, if I understand you properly, I concur again.

>> Wolfgang wrote:

>>> The fact, however, is that many 'ergative' languages lack an
>>> antipassive. For instance, there are nearly 30 East Caucasian languages
>>> all of them using some ergative strategies in at least parts of their
>>> operating systems. But only a handfull of them (five or six, to be
>>> precise) have true antipassives (only one has some kind of
>>> "pseudo-passive").

>> Pat wrote:

>> Are you asserting that the majority of ergative languages do not have
>> anti-passives?

Wolfgang answered:

> I do not assert anything in the sense of 'ALL language have...'. Even
> the claim 'the majority of ergative languages (sic!) have...' is rather
> suspect to me. What I said is that in those 'ERG systems' I looked at
> (about 200) antipassives are rather the exception than the norm.

Pat responds:

I am not surprised really. I suspect the (original [?]) real function of an
anti-passive is suggesting lessened effective agency. What do you think?

>> Pat wrote:

>> As for the far-reaching conclusions of Dixon based on discourse cohesion
>> strategies, in my opinion they are flawed because these strategies are
>> purely conventional.

Wolfgang answered:

> Language systems ARE conventional! This is one of the major points of
> language tradition and L1 acquisition (despite of minimalism etc.).
> Discourse probably is one of the most important factors in the
> emergence, organization, and dynamics of language systems. We should not
> refer to the abstract notion of context-free 'sentences' that would be
> responsible for for grammatical 'events'. Such a view stems from the
> tradition of Classical Philosophy which is an INTERPRETATION of what
> goes on language. Today, we have become used to think of language in
> single sentences, to brak them up the way we do etc. But this is an
> analytic tradition, not part of the ontology of language itself, which
> is much more synthetic in nature than we are used to think. - A sentence
> does not function but in its co-text (as well as in its con-text). All
> sentence internal strategies used to be embedded in the techniques of
> co(n)textualization. No wonder, that ACC and ERG also work in this
> direction (though they may appear as more 'autonomous',
> sentence-internal mechanisms secondarily, especially if a language
> system as developed separate means to indicate discourse cohesion).

Pat responds:

Again, I agree. A valuable caveat, and often neglected.

 <snip>

>> Pat wrote (on Lak, East Caucasian):

>> I have to plead ignorance of Lak however, I find an analysis of an otherwise
>> completely ergative sentence as having an ACC word-order impossible. I
>> believe this opinion rests on a false analysis of word-order significance.

Wolfgang wrote:

> I tried to show that in Lak only some sentence patterns show 'complete
> ergativity'. Most of them show a mixed paradigmatic organization with
> respect to morphology (CASE, AGR). WOrd order is another VERY imporant
> aspect of the AEC. Consider e.g. a language that has canonical

> #SV
> #AOV

> (# = sentence boundary). Here, S behaves like A with respect to #, hence
> it is ACC. In #SV vs. #OAV it is S=O which indicates ERG behavior. The
> problem of actance serialization is also addressed in polypersonal
> systems (without CASE, e.g, Abxaz, Lakotha etc.). It IS a very important
> indicator for ACC, ERG, no question. Anything else would refer to the
> linguistic tradition of say 60 years ago (Bloomfieldian tradition).

Pat responds:

Sorry, Wolfgang, after so much with which I can agree, I am unconvinced. I
do not believe the essence of accusativity is in word-order.

>> Wolfgang continued (on Lak):

>>> But if you say "I am surely hitting you (plural)", you get:

>>> na b-at-la-ti-s:a-ra zu
>>> I:ABS I:PL-hit1-DUR-hit2-ASS-SAP:SG you:PL:ABS

>>> Here we have:

>>> ACC with respect to word order
>>> ERG with respect to class agreement
>>> ACC (or neutral) with respect to case marking)
>>> ACC with respect to SAP agreement (-ra is triggered by 'I:ABS').

>> Pat wrote:

>> So, agent marked ABS and patient marked ABS is, according to you, ACC with
>> respect to case marking? Sorry, not convincing at all! And I think the SAP
>> difference can be explained as indicating seriality of the patients involved
>> as recipients of the action by *one* agent.

Wolfgang responded:

> I said ACC (or neutral!) with respect to case marking! That means that A
> lacks ERG marking (which is OK with Silverstein). Semantically, it means
> that S is encoded like A (which is ACC). That fact that O is NOT in an
> accusative-like case does not argue against this assumption, because
> CASE is here structrually coupled with ADGR which allows us to classify
> AGR as ACC *here*. The SAP difference in Lak cannot be explained in the
> sense Pat proposes, Lak totally lacks such strategies.

Pat responds:

Sorry again. Perhaps it is the term "accusative" that is preventing me from
grasping your argument.

>> Wolfgang wrote further:

>>> Now, please tell me: Is Lak an 'ergative' or an 'accusative' language?
>>> [Please note that I did not include (among others) strategies of
>>> discourse cohesion, reflexivization and logophization].

>> Pat wrote:

>> By these examples, I would tell you that Lak is 'ergative'; and that
>> accusativity is not demonstrable from these examples.

Wolfgang responded:

> If you refer to the standard (morphological), but rather obsolete
> interpretation of ERG you may be right (but not for SAP NPs). But,
> fortunately, the ACC/ERG typology has freed itself from such a narrow
> interpretation of ACC/ERG which is nothing but a very small excerpt from
> the over-all typology.

Pat writes:

"Unchain my heart, and let me be free, dum-dee-dum . . .  (:-{})

>> Pat wrote previously:

>>>> I think it is likelier that, because of perceived greater animacy (or
>>>> definiteness), pronouns have a different method of marking that can still
>>>> be interpreted within an ergative context.

>> Wolfgang wrote:

>>> This a (very simplified) 'on-dit' that stems from the earlier version of
>>> the Silverstsein hierachy. Again, we have to deal with the question,
>>> whether a 'pronoun' (I guess you mean some kind of 'personal pronouns')
>>> can behave 'ergatively' or 'accusatively'. The list below gives you a
>>> selection of SAP case marking in East Caucasian languages with respect
>>> to ABS/ERG:

>>> ABS vs. ERG ABS = ERG
>>> ALL ---
>>> Singular Plural
>>> Plural Singular
>>> 1.Incl. Rest
>>> 1:SG Rest
>>> 2:SG Rest
>>> 1:SG/PL Rest
>>> --- ALL

>>> This list (aspects of personal agreement NOT included!) shows that SAP
>>> pronouns may behave differently within the same paradigm. Any
>>> generalization like that  one quoted above does not help to convey for
>>> these data...

>> Pat wrote:

>> I do not have a reference book for Lak so that my hands are somewhat tied.
>> But, I have found that paradigms are often inconsistent in ways that reflect
>> earlier lost phonological changes, or other lost schemata. The locative
>> plural terminations of IE are certainly not, in origin, terminations of the
>> locative plural. Etc.

Wolfgang responded:

> Nothing the like! The list I gave refers to what can be described for
> East Caucasian languages in toto! Most of these paradigms are
> functionally motivated, I grant (schulze in PKK 2 (Schulze 1999) will
> tell you the whole story).

Pat concludes:

Still not convinced on this point . . . and, I am trying hard.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From connolly at memphis.edu  Thu Jul 29 22:30:10 1999
From: connolly at memphis.edu (Leo A. Connolly)
Date: Thu, 29 Jul 1999 17:30:10 -0500
Subject: accusative and ergative languages
Message-ID: <THU.29.JUL.1999.173010.0500.>

Robert Orr wrote:

> There's also "sneak - snuck" in English and "schreiben -schrieb-geschrieben"
> in German, both of which are analogical formations  on originally borrowed
> words..

> Robert Orr

_Snuck_ is unquestionably analogical.  But it is quite impossible to
tell whether the strong inflection of _schreiben_ is, since the expected
Germanic reflex of PIE *_skreibh-_, which must underly Latin _scri:bo:_,
is precisely PGmc. *_skri:b-_, which would be expected to produce a
German _schreiben_ with strong inflection.

Leo
-


From larryt at cogs.susx.ac.uk  Fri Jul 30 08:40:22 1999
From: larryt at cogs.susx.ac.uk (Larry Trask)
Date: Fri, 30 Jul 1999 09:40:22 +0100
Subject: accusative and ergative languages
In-Reply-To: <006101bed1a5$98f8be00$319ffad0@patrickcryan>
Message-ID: <FRI.30.JUL.1999.094022.0100.>

On Mon, 19 Jul 1999, Patrick C. Ryan wrote:

[PR]

>>> Now, when Larry recently quoted Dixon about the nature of the
>>> ergative, he conveniently neglected to mention that Dixon
>>> acknowledged that there were currently practising linguists --- not
>>> amateur linguists like myself --- still defending the passive
>>> interpretation of ergative constructions.

[LT]

>> No; this is not so.  Read p. 189 of Dixon's 1994 book.

[PR]

> I suggest you read it. "But we do still encounter scholars who
> insist that there is a necessary diachronic connection, e.g. Estival
> and Myhill (1988:445): 'we propose here the hypothesis that in fact
> all ergative constructions have developed from passives'."  Are you
> suggesting that Estival and Myhill are not "linguists", or that
> Shibatani, in whose book this essay appeared, is not a "linguist"?
> The quotation above is from p. 189 of Dixon's 1994 book. What are
> you playing at?

The question is not what I'm playing at, but what you're playing at.

Your original assertion was that the passive interpretation of ergative
languages was not only defensible but, in your view, correct.  By this
you clearly meant that ergative constructions, in general, *are*
passives.  This is the view which was once popular among European
linguists, which has been effectively demolished, and which is correctly
dismissed by Dixon elsewhere on that same page as without support among
linguists today.

The point in your cited passage is an entirely different one: do all
ergatives derive historically from passives?  Dixon notes that a few
people have argued that the answer is `yes', but then goes on to provide
what, in his view, is good evidence that the correct answer is `no'.

Whatever view one might adopt on this second point, it is clearly
distinct from the first point.  Claiming that "all ergatives *descend*
from passives" is the same proposition as "all ergatives *are* passives"
is rather like claiming that "all humans descend from their
grandparents" is the same proposition as "all humans *are* their
grandparents.

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk


From larryt at cogs.susx.ac.uk  Fri Jul 30 14:54:03 1999
From: larryt at cogs.susx.ac.uk (Larry Trask)
Date: Fri, 30 Jul 1999 15:54:03 +0100
Subject: accusative and ergative languages
In-Reply-To: <006101bed1a5$98f8be00$319ffad0@patrickcryan>
Message-ID: <FRI.30.JUL.1999.155403.0100.>

On Mon, 19 Jul 1999, Patrick C. Ryan wrote:

> Still no word on the scene of the great "shredding" which was
> *claimed* by Larry.

I suggest you read my article in the following forthcoming book:

K. Davidse (ed.), Case and Grammatical Relations Across Languages, vol.
5: Nominative and Accusative, Amsterdam: John Benjamins

The book should be out within a few months, but I don't have a date yet.

> And what are subject properties that an absolutive NP never displays?

As I've already explained: in Basque, and reportedly in many other
ergative languages, the absolutive NP in a transitive sentence --
entirely *unlike* the absolutive NP in an intransitive sentence --
cannot control empty NPs, cannot itself be an empty NP, cannot control
reflexive or reciprocal pronouns, can itself be a reflexive or
reciprocal pronoun, cannot be coordinated with an intransitive subject,
and (in varieties with genitivization only) can be genitivized in
suitable circumstances.

In Basque, there is *no* syntactic property shared by absolutive NPs in
transitive sentences and absolutive NPs in intransitive sentences, apart
from those that are shared also by ergative NPs and sometimes also by
dative NPs.  The only properties peculiar to all absolutive NPs and
shared by nothing else are purely morphological ones: case-marking and
verbal agreement.

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk


From jer at cphling.dk  Fri Jul 30 15:50:53 1999
From: jer at cphling.dk (Jens Elmegaard Rasmussen)
Date: Fri, 30 Jul 1999 17:50:53 +0200
Subject: accusative and ergative languages
In-Reply-To: <002401bed2a4$93faa8a0$3970fe8c@lucent.com>
Message-ID: <FRI.30.JUL.1999.175053.0200.>

On Tue, 20 Jul 1999, Vidhyanath Rao wrote:

> Jens Elmegaard Rasmussen <jer at cphling.dk> wrote:

>> I'd say the cases I have seen of ergativity in Indic and Iranian languages
>> so clearly reflect underlying/earlier passive circumlocutions that
>> controversy is absurd.

> No!

> I have already pointed out the problem of distinguishing resultatives and
> passives. The ta-adjective is resultative in Vedic (Jamison, IIJ 198?) and
> Early Pali (Hendrikson, Infinite verb forms of Pali). And it patterns
> ergatively as resultative participles often do. The last part has been known
> forever. In English, see Speyer's ``Sanskrit Syntax'', will find the
> following: ``Of the participles in -ta the great majority have a passive
> meaning, hence it is customary to call the whole class the passive
> participle of the past. But some others are not passives, but intransitives,
> as gata (gone), m.rta (died) [rather dead, VKR] bhinna (split). Some again
> may even be transitive actives, as pi:ta (having drunk) [better drunk, but
> without the restriction to a special meaning in English] ...'' [para 360, p.
> 280].  Speyer goes on to note a:ru:d.ha has active meaning more commonly.
> This is not how we expect the passive to behave. But for resultative, it is
> understandable. The ergative patterning is based on pragmatics. But in a
> culture that considers it more worthy of note whether a man is mounted on a
> horse or a vehicle than whether a horse or a vehicle is carrying someone, it
> makes sense to use a:ru:d.ha in the active sense.

> The difference is unmistakable in the following:

> (1) ra:mo 's'vam a:ru:d.hah.
>     R is mounted on [a] horse.

> (2) ra:men.a+as'vam a:ruhyate
>    [A] horse is being mounted by R.

> (3) *ra:mo 's'vam a:ruhyate

> (1) and (2) are quite grammatical and examples of easy to come by. (3) does
> not occur and is underivable in traditional grammar. As I explained above,
> this is understandable if a:ru:d.ha is resultative. If we take the passive
> view, how do we explain the fact that (1) is acceptable but (3) is not?
> Without such an explanation, it is far from absurd to contest the passive
> interpreation.

Is this anything other than the accusative of goal? You are right that
some ta-participles are not passive, namely those derived from
intransitive verbs (gata- like Eng. gone). I would take (1) to be
construed as to mean "Rama (is = Eng.) has climbed onto a horse". (2) must
mean "By Rama there is being climbed onto a horse". I do not know if you
can construe this particular verb so transitively that you can make its
passive have the patient in the nominative; trying my hand with the sandhi
rules, I make it come out as "ra:men.a+as'va a:ruhyate".  But is this not
the normal construction with the finite passive? - I cannot see the
relevance of this for a discussion of the question whether the
ta-participle is passive or resultative: with transitive verbs it plainly
is both, we are not that much is disagreement. - By the Modern Indic rules
of agreement it does seem to me to be the passive (of transitive verbs, of
course) that formed the pattern. Where am I wrong?

Jens


From inaki.agirre at si.unirioja.es  Fri Jul 30 10:56:45 1999
From: inaki.agirre at si.unirioja.es (Inaki Agirre Perez)
Date: Fri, 30 Jul 1999 11:56:45 +0100
Subject: Ergative & Basque
Message-ID: <FRI.30.JUL.1999.115645.0100.>

About <have> and auxiliary verbs:

> This western usage is clearly calqued on the famous Castilian
> distinction between <ser> `be' (unmarked) and <estar> `be' (in a place
> or in a state).  Wherever Castilian uses <estar>, western Basque uses
> <egon>.  This is true even in idiosyncratic cases.  For example,
> Castilian expresses `He's dead' as <Esta' muerto>, with <estar>, and
> western Basque likewise has <Hil dago>, with <egon>, while eastern
> Basque has <Hila da>, with <izan>

AFAIK, in English 'He's dead' is both Spanish <Esta' muerto> and <Ha
muerto>. The translation in Basque would be <Hil da>, with <izan> verb.
A momentary sense could be achieved by <Hilik dago>, as if you have just
discovered the fact that he's dead, but <Hil dago> is ungrammatical in
my Basque (western).

Inaki


From jer at cphling.dk  Sat Jul 31 19:04:06 1999
From: jer at cphling.dk (Jens Elmegaard Rasmussen)
Date: Sat, 31 Jul 1999 21:04:06 +0200
Subject: Ergative & Basque
In-Reply-To: <Pine.GSO.4.03.9907191150230.19800-100000@rsunx.crn.cogs.susx.ac.uk>
Message-ID: <SAT.31.JUL.1999.210406.0200.>

Thanks to Larry Trask for taking the trouble to answer my questions about
the ergatve in Basque (in mail of Mon, 19 Jul 1999). The message being
quite clear and presumaby helpful to others than me alone, there are just
a few things that prompt a comment:

In a general sense, I understand that there is very little difference
between the ergative and the "accusative" structure. If you say "I have
built a house" with a verb of "having" taking the accusative, then the
construction is not ergative. But if "have" is expressed by "be to/for/at"
or "be" + genitive, and you phrase it as if it meant "there is to me a
built house", or "a built house is mine" or the like with the logical
patient as the subject put in the nominative, then it is ergative.
Strangely, the main idea, the expression of the relation between te agent
and the patient as one of possession ("having") is the same in the two
cases. This seems like a lot of fuss over very little. Now, where "have"
is expressed as "be" + an adverbial case (or the genitive, as in
Lithuanian vyro yra butas "of the man is a house" = 'the man has a
house'), there is a good and straightforward reason to put the
logical agent in a non-nominative case, a reason that never arises with
intransitive verbs which can at best expressed with an _essive_ relation
between the actor and the predicate: "John is laughing", "those days are
gone". This scenario then offers an opportunity for a direct understanding
of the distinction between the subject of an intransitive verb and that of
a transitive one, for one "is", the other "has" (or, is the one to/by etc.
which/whom something is); and it also makes it possible to understand
why the object has the same form as the subject of an intransitive verb,
for both "are". The use of a "have" _verb_ with ergative case as opposed
to a verb "be" with inergative case offers no such understanding,
therefore I find it unattractive if the ergative construction in
general is simply taken for granted; I could much better understand
things if the search is continued back to an origin of the opposition in a
difference between "is" and "there is to" (= "has"). That would make the
ergative just a roundabout use of the same general case categories as
used by the "accusative" syntax. That the search cannot be pushed that far
back for many languages does not change the possibility of such a
prehistory: you can draw no conclusions from inconclusive evidence.

I asked Larry Trask:

>> Can't the underlying construction be analyzed in a sensible way at
>> all?

LT asked back:

> Well, I'm not sure what you mean by "a sensible way".  What would you
> regard as a sensible analysis of English `I have drunk the wine'?

I see the point: How can you have something you have consumed - or
annihilated or lost, to take some even worse examples? You cannot. Then,
how can you justify expressing such events as if you have things you don't
really have? By analogy, by simple extension of the syntactic pattern from
an original core in which it made good sense to a general use where this
is not always the case.

LT also said:

> And this, I think, is as far back as we can go in tracing the prehistory
> of the Basque periphrastic verb-forms.  As I remarked earlier, we have
> no way of knowing whether these things originated as calques on Romance
> or whether they are independent creations in Basque, from an unknown
> source.  All I can tell you is that the modern periphrastic forms were
> clearly already established in Basque by the tenth century, when we find
> the first recorded verb-forms.

I'd say this disqualifies Basque as a language to help us solve the riddle
of the ergative, if there still is one.

Jens


From X99Lynx at aol.com  Thu Jul 29 18:20:26 1999
From: X99Lynx at aol.com (X99Lynx at aol.com)
Date: Thu, 29 Jul 1999 14:20:26 EDT
Subject: Passivity as a transition
Message-ID: <THU.29.JUL.1999.142026.EDT.>

Jens Elmegaard Rasmussen <jer at cphling.dk> wrote:

<<I'd say the cases I have seen of ergativity in Indic and Iranian languages
so clearly reflect underlying/earlier passive circumlocutions that
controversy is absurd.>>

On 7/29/99 2:52:29 AM, vidynath at math.ohio-state.edu responded:

<<No! I have already pointed out the problem of distinguishing resultatives
and passives. The ta-adjective is resultative in Vedic...

Speyer goes on to note a:ru:d.ha has active meaning more commonly.
This is not how we expect the passive to behave. But for resultative, it is
understandable.  The ergative patterning is based on pragmatics. But in a
culture that considers it more worthy of note whether a man is mounted on a
horse or a vehicle than whether a horse or a vehicle is carrying someone, it
makes sense to use a:ru:d.ha in the active sense.>>

Here on the "accusative and ergative languages" forum, I believe these
exchanges started with someone taking issue with the idea that 'passivity'
was not (necessarily) the origin of 'ergativity'.    Having read Larry
Trask's handling of that issue in his Historical Linguistics, it seems that
"passivity" as the ONLY origin is quite out of the question.

It's interesting to compare W.P. Lehmann's approach to ergativity in his
Historical Linguistics textbook.  The accusative versus ergative is only
brought up once - as background for the Contentive Typology distinction
between 'govermental' (accusative OR ergative) and
'active'/'active-stative'/'class' languages.  This background is given to
illustrate the use of Typological frameworks for historical purposes (e.g.,
postprepositions in Hittite, Indic and English and the 'pur'/'ignis'
(animate/inanimate) distinction) and evidence of an Active residue in modern
and historical languages.

While dismissing any further discussion of ergativity, Lehmann brings up an
interesting point.  He says, "While the method of indicating subjects and
objects differ in accusative and ergative languages, both are comparable in
distinguishing subjects and objects when pertinent, that is, with transitive
verbs... viewed as essential in communicating....  The subject of an
intransitive verb does not require a special form since it does not need to
be distinguished from an object."

This brings up the issue of what a truly "passive" language would be like,
putting aside animate versus inanimate.  I think the presumption that the
intransitive cannot take a passive form (of sorts) may not be correct when
words are in transition.  Verbs turn into nouns.  In Homer, the players are
sometimes 'moved' by their actions.  It is clear that in "the killing was
made by Hector" it is the action that is governing the structure, even though
there is a helper verb in there.  (Something of the same happens in saying
"Hector killed", as in "Speed kills" - it can be interpreted as a transitive
(kills someone) disguished as an intransitive.  But the sense definitely
changes and there is a reason to think of it as intransitive.  cf., "X rules")

Going back to the qoute above: <<in a culture that considers it more worthy
of note whether a man is mounted on a horse or a vehicle than whether a horse
or a vehicle is carrying someone, it makes sense to use a:ru:d.ha in the
active sense.>>

The third approach of course is to focus on the riding or carrying.

In Greek, we can see a related development that demonstartes verbs becoming
nouns.  "Konis"/"konia" (dust) in Homer is the basis of the idea of raising
dust - "koniontes pedioio", galloping in a cloud of dust (in the Illiad
always relating to horses, later to armies.)  "Euru konisouson pedion",
running away in dusty (hasty) flight.  But also in the Illiad, the jump is
made to "raising dust" as a metaphor for haste, effort or service.  The maids
make Priam's bed, "epei storesan lechos enkoneousai" (pres part.) > in haste.
(Always with another verb, so that it is NOT the action itself, but a true,
separate element.)

So "enkoneo" will later come to mean in haste.  "Akoniti" will mean after
Homer, without effort.  And "diakonia" will become a common name for a
servant or service in general - literally "through dust" but actually
"through raising dust" making haste or effort.

Shouldn't we conclude that before "raising dust" > effort> servant could
acquire the indications of a noun and an accusative subject, it had to go
through a "passive" stage.  E.g. "raising dust was done by Mabel."  ("Mabel
was raising dust" inflected as a participle, would mean something different -
it is part of the action and not yet an "object.")

Before there was something called a "rider" there was something called
"riding".  The action may have prempted the object and became an object in
itself.  "John was riding."> "Riding was what John did."> "John was a rider."

As a language develops, it is clear that verbs (processes, regular actions)
become nouns.  E.g., raising dust becomes service or a servant.   While that
is happening, isn't passivity also a natural thing to happen somewhere along
the line?  The absence of an established morphology/inflection for the new
noun means there appears to be no subject.  Couldn't to some degree that
account for the absence of a designated subject in some ergative structures?

Regards,
Steve Long


From rmccalli at sunmuw1.MUW.Edu  Fri Jul 30 15:10:59 1999
From: rmccalli at sunmuw1.MUW.Edu (Rick Mc Callister)
Date: Fri, 30 Jul 1999 10:10:59 -0500
Subject: Momentary-Durative
In-Reply-To: <002501bed2a4$94f66dc0$3970fe8c@lucent.com>
Message-ID: <FRI.30.JUL.1999.101059.0500.>

As a non-linguist, I'm gonna have to ask for back-up

but --if I remember correctly--
the present usage of ser and estar goes backs to
about 1400 to 1500
before that --at least in my memory of medieval literature--
the usage of ser and estar seem to be more unstable
or function more like the dichotomy of
essere and stare in Italian
where some of the uses of Spanish estar
are performed by Italian essere

But given that the dichotomies are pretty close
this probably goes back to Common Romance

French does not have this dichotomy
and I don't know what the situation is in Occitan, Sardinian, Rheto-Romance
and Rumanian

[ Moderator's note:
  The following is quoted from Vidhyanath Rao's message of 20 July 1999.
  --rma ]

>Rick Mc Callister <rmccalli at sunmuw1.MUW.Edu> wrote:

>> Spanish distinguishes them [resultative and passive]
>> by resorting to

>> estar --a "momentary" verb--
>> [actually a verb indicating condition] for resultant conditions--

>> and ser--a "durative" verb
>> [actually a verb indicating  characteristics]
>> for passive constructions.

>I don't know much Spanish, but I ought to have remembered this for I have
>been told about this before.

>I find something curious about your glosses. Resultatives, as indicating a
>state of indefinite duration, ought to be duratives, though examples of
>languages which use `go', `come', `finish' to form resultatives are
>mentioned in ``The evolution of grammar''. How old is the use of `estar'?.

Rick Mc Callister
W-1634
Mississippi University for Women
Columbus MS 39701


From jer at cphling.dk  Fri Jul 30 15:12:19 1999
From: jer at cphling.dk (Jens Elmegaard Rasmussen)
Date: Fri, 30 Jul 1999 17:12:19 +0200
Subject: Momentary-Durative
In-Reply-To: <000f01bed2e8$ba591b00$fde3abc3@xpoxkjlf>
Message-ID: <FRI.30.JUL.1999.171219.0200.>

Dear Peter and List,

(- careful, this may be about Indo-European -)

I think we're getting some place. Specific and insisting questions based
in an open mind are the only way. Many could learn from this - and that
may be just the point, as we shall see:

On Tue, 20 Jul 1999, petegray wrote:

[...]

> You said:
> [Jens:]
>> The impression that [in the IE verbal stem] ...
>> no common system is
>> recoverable is not compatible with current knowledge

> [Peter:]
> There is little real "Current knowledge" with PIE.   It is in many cases
> really rather "current opinion which someone happens to accept".   So how
> much support is there for Strunk?  How much debate is there?

[Jens:] There is a lot of unqualified debate on everything in IE studies,
even on points where the material speaks with a very clear voice. I'm not
sure any statements about IE could carry a majority vote. The reason is
obvious: IE studies is where most comparativists are, the field is vast,
and the amount of previous scholarship staggering. It is common to begin
your career by making mistakes simply because you cannot know all the
pertinent facts; many (most?) stay that way, contributing only to the
advancement of confusion. Therefore opinions should be checked very
carefully before they are adopted; one can only go by the quality of the
arguments put forward. Now, that _is_ what you are doing:

[Peter:]
>    I do question, however, how much weight we should give these
> patterns, and in particular whether we over-prioritise the patterns 
> based on Greek & Indo-Aryan.   It has been shown again and again since
> the 70's that the "south-eastern" group of Greek, I-A, and Armenian is
> highly innovative.

[Jens:] I'm not sure "shown" is the proper word, rather "maintained",
"guessed" or "wishfully thought" (if you can say that). It's one of those
numerous things about which two opinions are logically possible (_before_
you look at the specifics, mind you): Either Balto-Slavic is a unit or it
isn't, either the limited morphology of Hittite is a fossil or it isn't,
either *H2 colored *o or it didn't, etc. Not unpredictably, opinions tend
to sway from one possibility to the other all the time; if the market is
for one solution, it is arguing for the other one that is interesting.
Much debate in the field can be written off as acts of would-be scholars
searching for a place to stand with their names in neon signs. The
correspondences uniting Greek and Indo-Iranian are remarkable in a respect
that is constantly ignored by one "side", namely by their ability to turn
up in relics in all the other branches also. Now, innovations would not do
that, so that's not what they are. One is naturally reluctant to believe
that it is as simple as that and that so many researchers can be so
pitifully wrong, but only until one has looked at the decisive facts,
then one side is patently wrong and the other one much closer to the
truth.

> [Peter:]
> You said:

> [Jens:]
>> No, thematic presents and root-presents typically take the s-aorist (or
>> suppletion).

> [Peter:]
> Root presents are rare outside I-I, and in Sanskrit -s- aorists do not seem
> to be typical for root presents, since the both sigmatic and asigmatic
> aorists occur.   E.g. am has asigmatic, i:d has both, i:r shows both, u
> shows none, U:h asigmatic, and so on.   Where, again, is your evidence that
> makes it "typical" for a root present to take an -s- aorist?
>     Thematic presents appear to be a later formation, in any case, and you
> seem to suggest an origin for them in root aorist subjunctives.

[Jens:]
The two questions are related. First, thematic presents indeed appear to
be old subjunctives (more often aorist than present, but both occur). This
means that we can use the thematic present on a par with the root present.
Incidentally, I see no signs of being young about the thematic formation
as long as it is a subjunctive - quite the contrary: the thematic vowel
alternates in its own way, I don't see how that can be accounted for
unless by rules that have later ceased to operate. Second, if you take
Sanskrit from a Sanskrit grammar like Whitney's or a large dictionary like
Monier-Williams' you get an unsorted mixture of old and new forms, some so
young that they belong to a language that was nobody's native language
anymore. You must discount all the young forms of later periods, which is
best done by sticking to Vedic alone. Narten has of course discussed all
Vedic forms that look superficially like s-aorists, but since her work
does not contain a catalogue of conclusions (which are often not
decisive), it is hard to use for our purpose. On a more modest level, one
can look in Macdonell's Vedic Grammar under s-aorist ad is-aorist
(discounting with the latter forms without an extra vowel mora, since they
are just reinterpreted set root aorists), and then check in Macdonell's
Vedic Grammar for students to see what kind of present the root concern
forms. The result is overwhelmingly positive if the search is limited to
s-aorist forms that are not marked as belonging to post-RV texts):

I may have missed a few, but my somewhat hasty inspection reveals a root
present or a thematic present beside an s-aor. in the following 37 cases:

akra:n (krand) : kra'ndati
a'ks.a:r (ks.ar 'flow') : ks.a'rati
ks.es.at (ks.i 'dwell') : ks.e'ti
ga:si (1sgM ga: 'sing') : ga:'y-a-ti (!)
ga:ri:t (gr.: 'devour') : gira'ti
aca:ris.am (car") : ca'rati
acait (cit 'perceive') : ce'tati; cite'
ajais.am (ji) : ja'yati
ta:ri:t (tr.:) : ta'rati
atsa:r (tsar 'approach') : tsa'rati
adha:k (dah) : da'hati
a'diks.i (dis'), Av. da:is^, do:is^.i: - : Lat. di:co:, Goth. teihan
dhuks.ata (duh) : do'gdhi
adyaut (dyut) : dyo'tate
nes.ati (nais.t.a, ni:) : na'yati
anu:s.i (nu:s.ata, nu:) : na'vati
a'bhutsi (budh) : b¢dhati
a'bhaks.i (bhaj) : bh jati
abha:rs.am (bhr.) : bha'rati
amatsur (mad) : ma'dati
aya:s (3sg aya:t., yaj) : y'ajati
aya:sam (ya: 'go') : ya:'ti
a'rabdha (rabh) : ra'bhate
a'ram.sta (ram) : ra'mate
ava:t (vas) : va'ste
ava:t. (vah) : va'hati
a'viks.ata (vis') : vis'a'ti
avr.tsata (vr.t) : va'rtati
asaks.ata (sac) : sa'cate (also root aor. with red.prs.)
asakta (saj) : s'ajati
asa:ks.i (sah) : sa'hate
asa:vi:t (su: 'impel') : suva'ti
asra:k (sr.j) : sr.ja'ti
astos.i (stu) : sta'uti
asya:n (syand) : sya'ndate
a'sva:r (svar) : sva'rati
a'ha:rs.am (hr.s.) : ha'rs.ate
(Forms with zero-grade + -ks.a- may be special innovations; that would
reduce the list by two.)

In 13 examples, commonplace renewal by the productive s-aorist can be
assumed. Root aorists are attested for the following for which, then, it
is no problem that they do not form root presents:

ata:n (tan): tano'ti, old aor. a'tan
da:si:t (das 'waste'): old aor. dasat (Narten)
naks.at (nas'): old root aor. a:'naks.am (prs. doubtful: length, red., y?)
apra:s (pra: 'fill'), old root aor. apra:t: n-prs. pr.n.a:'ti
muks.ata (muc), old root aor. a'mok : n-prs. mun'c'ati
a'yuks.a:ta:m (yuj), old root aor. yo'jam : n-prs. yuna'kti
a'ra:sata (ra:), also root aor. and red.prs.
a:'raik (ric), old root aor. riktha:'s, n-prs. rin.a'kti
avitsi (vid 'find') secondary: old root aor. a'vidat : n-prs. vinda'ti
asa:nis.am (san"), old root aor. a'sanat : n-prs. sano'ti
a'spa:rs.am (spr.), old root aor. a'spar : n-prs. spr.n.vate'
ahes.ata (hi), old root aor. a'hyan : n-prs. hino'ti
ahu:s.ata (hu:) : ha'vate (also root aor. + red.prs.)

A few combine s-aor. with sk-prs.:

a'pra:ks.am (pras') : pr.ccha'ti
yos.ati (yu) : yu'cchati (also root aor. : red.prs.)
a'ya:m.sam (yam) : ya'cchati
ava:t (vas 'shine') : uccha'ti

The residue is very small, perhaps as small as two examples:

mam.si (man): unclear relation to ma'nyate; manve' (with root aor.)
alipsata (lip) : n-prs. limpa'ti !!
aha:s (ha:) : red.pl. ja'ha:ti, ji'hi:te !!

So it's about two or three examples out of 57 that won't play ball! I do
not think that looks like random distribution?

> [Peter:]
> Incidentally, I read today in JIES vol12, 1984, the argument that even
> the -s- aorists are an innovation within Greek, I-A and closely related
> groups.   The article alleged that evidence outside this area is weak, and
> tried to dispose of Latin -s- perfects by suggesting they were either
> limited to verbs ending in a velar, or they were back formations from the
> supine.

[Jens:]
It's a popular view on the matter, and one of the favorite arguments for a
step-by-step model of development within the protolanguage, if I
understand it correctly (the whole idea is alien to my sense of logic).
For the s-aorist it is patently wrong. I have myself (in 1978) called
attention to the existence of Hitt. ganeszi 'recognizes' whose past
ganest(a) from *g^ne:H3-s-t is manifestly the s-aorist one expects beside
the sk-present (Lat. cogno:sco:, Gk. gigno:sko:, Lith. paz^iNstu, OPers.
xs^na:sa:tiy, Alb. njoh), and Jasanoff did the same independently (1987),
adding other branches (esp. the magnificent Armenian prs. c^anac^'em, aor.
caneay, the latter form from a stem *g^n.e:H3-s- with vocalization of the
/n/ pointing to an old monosyllable), and we presumably agree on the
prehistory of Germanic *kne:-(j)i/a- 'know' as based on the s-aorist, 3sg
*g^ne:H3-s-t, with simple replacement of the word-final cluster by
productive endings. The main point is here that the lengthened grade goes
back to a time when laryngeals had not yet colored adjacent e-vowels, for
the lengthened /e:/ turns up as /e:/ in the daughter languages, not as
/o:/, this proving the formation so old that there can be no talk of its
being a post-PIE innovation, no matter what weaknesses one may read into
specific parts of the material.

I hope this has been instructive.

Jens


From jer at cphling.dk  Sat Jul 31 01:11:52 1999
From: jer at cphling.dk (Jens Elmegaard Rasmussen)
Date: Sat, 31 Jul 1999 03:11:52 +0200
Subject: Momentary-Durative
In-Reply-To: <002401bed767$67c01f00$2e70fe8c@lucent.com>
Message-ID: <SAT.31.JUL.1999.031152.0200.>

On Mon, 26 Jul 1999, Vidhyanath Rao wrote:

>> [Jens:]
>> If my observation that there is an alliance between the sk^-present type
>> and the s-aorist is correct ... then the s-aorist was originally
>> inchoative in function. [...]

> [Nath:]
> This makes it harder for me to understand how the aorist became the
> perfective. `Started driving', in contrast to `drove', suggests incomplete
> action.

The aorist reports a turn of event that caused a new situation: is that
not the perfect thing to express a beginning?

[Nath:]
>>> This brings me to a general question. There seem to be two camps about
>>> the category system of the PIE verb. One believes that the
>>> aorist-imperfect distinction, to be equated to perfective-imperfective
>>> distinction, ``always'' existed in PIE and Hittie lost this distinction,
>>> while Vedic changed things around. The other considers the aspectual
>>> distinction to postdate the separation of Anatolian.

[Jens:]
>> There are these two camps, yes, and I am in no doubt that camp one is
>> right. There is no way the specific forms of the aspect stems could have
>> been formed secondarily in "the rest of IE" left after the exodus of (or
>> from) the Anatolians. At the very least, all the _forms_ must be assigned
>> to a protolanguage from which also Anatolian is descended. [...]

[Nath:]
> The stem formants might have been derivational in nature. In particular,
> they might change the meaning, valence and/or aktionsart. It is the
> assumption that they changed only the aspect and nothing else that is in
> question. Your view, expressed elsewhere, that different present stems might
> be because there are different kinds of durativity is halfway to this.

[Jens:] I am rather convinced this is all derivation in origin.

[..]
[Nath:]
> Grammatization of derivational affixes is found often enough that its
> occurance in Pre-IE -> PIE or PIE -> dialects cannot be rejected out of
> hand. As it seems that very different languages underwent similar evolution
> of grammatical categories (see Bybee et al, ``The evolution of grammar), we
> cannot assume the similar functions must go back to the proto-language. That
> might have had a precursor function.

[Jens:]
The PIE function of the different derivative categories must be at least
compatible with that of their later reflexes, and the simplest solution is
that wherever we find non-trivial correspondences between the daughter
languages we have a relatively direct reflexion of the protolanguage. Why
would a morpheme with lengthening and -s- turn up as the expression of
non-durative past in Italo-Celtic, Slavic, Armenian, Greek and Tocharian
if that were not its function in PIE already?

[Nath:]
>>> There are certain nagging questions about the first thesis [i.e. the
     thesis of Graeco-Aryan archaism and massive losses in Hittite - JER]:
>>> The change
>>> in Vedic is not explained and how it came about without the prior
>>> loss of aspect has, AFAIK, not been explained. Those who adhere
>>> to this also feel the need to explain away as much as possible of
>>> root presents. But there are enough of them remain in Hittite and
>>> Vedic to raise doubts.

>> [Jens:] I don't follow - what change in Vedic are you talking about?

> [Nath:]
> In Vedic, the so-called imperfect is the tense of narration. Aorist has a
> recent past meaning. In living languages with aspect, the perfective is the
> tense of narration, unless historical present is being used. Languages with
> remoteness distinction, either it applies only in the perfective or without
> regard to aspect. It is this that needs to be explained.
[...]
> Hoffman's answer is to posit an intermediate stage in which aspect was
> limited to non-recent past. He does not give any contemporary examples, nor
> does he explain how the aorist, which in such a stage must have been even
> more common, was lost in reference to remote past.

> The point is that an alternate explanation is possible: Completives have a
> ``hot news'' value, which makes it plausible to see them develop into recent
> past. They also can develop into perfectives, Slavic being a usable example.
> There are gaps in the examples, but I find this more plausible than
> deriving the Vedic usage out of perfecitve-imperfective opposition.

[Jens:] But Vedic _is_ an IE language.

[Jens, earlier:]
>> Why would anyone want to explain away root presents where
>> they are securely reconstructible? One would do that only to avoid
>> having a language combining a root-present with a root-aorist, for in that
>> case the two aspect stems are identical. That is why I am so sceptical
>> about the authenticity of the Vedic root presents lehmi and dehmi,
>> because for these verbs we have nasal presents in some other IE
>> languages pointing to the existence of a root aorist; thus leh- deh- look
>> like displaced aorists. But not so for eti 'goes' or asti 'is': these are
>> durative verbs, and so their unmarked form could function as a
>> durative (socalled "present") stem.

[Nath:]
> This argument is valid only if you have an independent reason to assume
> obligatory perfective-imperfective contrast for PIE. But firstly, we see
> stems transferred from one class to another within a single group. This is
> most obvious in Indic, where we have a long recorded history. Secondly,
> there are too many root presents that remain: In Vedic, amiti (injures),
> da:ti (cuts, divides), yauti (joins, unites) and of course, hanti confirmed
> by Hittite kuen-. None of these is eligible for a nasal present (expecpt
> perhaps amiti, if you believe in nasal presents for roots of the shape CeNH,
> with N standing for a nasal). Especially *g'henti is unavoidable, unless you
> are going to argue that the two oldest recorded dialects innovated in
> precisely the same way. Such an argument needs more compelling evidence
> than a just so story.

> If you add potentialy telic verbs, we get some more: ta:s.t.i (taks,
> fashion), ya:ti etc. These can and are used in the so-called imperfect with
> definite objects without any indication of non-completion. etc. How can they
> be called imperfetive?

[Jens:] With reference to PIE they must, in case such was the system - and
that it was is very well established.

>[...]
[Jens:]
>> Again, I do not think there is any problem in accepting root present as
>> original for inherently durative verbs, and root aorist as equally
>> original for inherently punctual verbs.

[Nath:]
> What about telic verbs? These are inherently durative, but with a determined
> object, their preterite (in languages without perfective-imperfective
> distinction) carries the implicature of completion, while their imperfective
> past would do the opposite. How, without recorded narratives, can you decide
> how a given language operated?

[Jens:]
Telic verbs are typically aoristic: *{gw}em- 'come' forms a root aorist,
*H1ey- 'walk' forms a root present.

[Jens earlier:]
>> But the IE aorist is not restricted to any special kind of verbs -
>> it is only _unmarked_ (better: apparently originally unmarked) for
>> inherently punctual verbs; for other verbs the aorist needs a
>> morphological marking, and the meaning is then some nuance that can be
>> regarded as punctual ("started to -") or it just reports that the
>> action got done (Meillet's > action pure et simple).

[Nath:]
> Could the imperfect report `action pure et simple', or was that reserved for
> the aorist? How did PIE speakers report a durative action that was done,
> like ``I walked home''? How did they say ``I made pots yesterday''?

[Jens:]
I guess they walked home in the aorist, made pots (generically) in the
ipf., but made a specific pot or set of pots in the aorist. But who am I
to know? Please don't demand that I write a fable.

[Jens:]
>> One important functional point with the aorist,
>> however, is that it marks a turn of events which creates a new situation,
>> whereas the "present aspect" stays in the situation already given and
>> reports another action contributing to that situation.

[Nath:]
> How do you explain that it is the so-called imperfect that is the tense
> of narration in Vedic?

[Jens:]
Sorry to take so long in coming to the point, this _is_ a very important
question which deserves more attention than is perhaps mostly accorded to
it. Now, under no circumstances can we completely divorce the imperfect
from the present, for they are formed from the same stem - that must mean
_something_. And I do not think we can disregard the situation- changing
effect of the aorist stem which turns up in all corners of IE. I repeat:

>> This is seen remarkably well in the prohibitive use of the prs. vs.
>> aor. injunctive, as propounded so clearly by Hoffmann.

And I maintain this against your comment:

> Hoffmann's claim requires morphological gymnastics, such as taking i:'s'ata
> as an aorist and still has some holes (eg, jivi:t).

I find no problem with the existence of individual verbs that act in
individual ways that have to be entered in the lexicon. That sound fairly
normal.

So, if the present stem is situation-preserving, and the aorist stem
situation-changing, how do we explain the Vedic facts? They do not look so
odd to me: The present is also a narrative form, namely to report what is
going on: "The horse is turning at the corner, it's emerging in full
sunlight, and is now approaching the finish line" - this would all be in
the present indicative in Vedic I guess. If the corresponding past
narrative is the imperfect, that could simply be due to the status of this
category as the past of the present stem. Other IE languages rather
clearly point to the one-time existence of a step-by-step turn-of-events
use of the aorist, opposed there to the background actions of longer
duration reported by the imperfect, but in Vedic the aorist has taken a
turn in the direction of exaggerating the situation-changing value of the
stem, while the imperfect has been generalized to all past narrative, be
it momentary or background-like. In Vedic the aorist is almost constantly
rendered by translators by means of the "have perfect" which states a
conclusion. The consensus on this matter is remarkable, especially since
there is little possibility of checking what is really meant in the text
of the Rigveda. The least one can say is that the use of tenses in
translations of the Rigveda does not strike one as unnatural (it's often
the only thing one does find natural). If I were to combine the Vedic
picture with that of the other IE branches, this would be my guess. It
appears to mean a lot to you that the IE aspect opposition is not supposed
to have been obliterated before the vedic development can take place -
well, in my account it isn't. And then there is nothing alarming (or even
interesting) about the number of preserved root presents.

[Jens earlier:]
>> Still, even verbs generally signifying completed action could form
>> duratives, indicating e.g. a repetition of the action (give one thing,
>> and then another) or an as yet unsuccessful attempt (I'm opening the
>> window).

[Nath:]
> How do you classify ``I learned that chapter in one month?''

[Jens:] I believe as a job for the aorist.

[Jens earlier:]
>> We know that this kind of change was small enough in the languages here
>> concerned to lead to a number of misplaced aspect stems. E.g., the
>> Armenian aor. eber is an old ipf. Is the jump from "narrative past" to
>> "recent past" so great? If it is, even great changes happen.

[Nath:]
> But if the PIE ``imperfect'' was really just a preterite, what we see in
> Armenian eber is old preterite becoming an aorist due to the rise of an
> imperfective past. That is attested. For example, Kui, a Dravidian language,
> has generalized an old progressive into an imperfective and this limited the
> preexisting past into an aorist.

[Jens:]
Not due to the creation of the ipf. for that was already there (eber was
one itself). I do not argue about Kui, I only say the Armenian (and IE)
story cannot be that way. We know too much be able to accept it.

> It is not the size of the change, but the direction and manner of the
> change, that must be credible. All I am asking for is a single
> uncontrovertable evidence of this change. It won't do to say, as I was once
> told by an Indologist, that this change is possible because it must have
> happened in Vedic. You need to give an example where the perfective past
> category can be established from >preserved texts<.

If instead of perfective past you read concluding past which is just a
natural further development, you find it supported several times in
practically every hymn of the whole Rigveda. And the distinction between
perfective (aorist) and imperfective past (imperfect) is a salient shared
feature of all other IE languages than Germanic, Indo-Iranian and
Anatolian in the system of synthetic verbal forms. There is no way that
can be an innovation. Specifically, it is not therefore not possible that
the isolated instances of misplaced aspect forms reflect an archaic state
of affairs where the whole aspect business had not even started.

[Nath:]
> What I find difficult to swollow with the argument for aspect in PIE is that
> the association ``imperfect'' = imperfective is limited to Greek. The two
> oldest recorded dialects, Hittite and Vedic, do not work that way. We are
> simply supposed to believe that they innovated, but what do we get in
> return? What can you explain this way that you cannot explain otherwise?

[Jens:]
The IE imperfect is not just Greek. The Slavic imperfect, which mostly
translates the Gk. ipf. in OCS, is an almost direct continuation of the IE
ipf. (in Baltic it has become a preterite pure and simple due to the loss
of the aorist). The Armenian ipf. has endings in -i- from *-e:- stemming
from the verb 'be' (all thematic verbs rhyme with 'be' in Arm.) which
formed *e:st from *e-H1es-t with the augment. The Toch. ipf. is basically
the optative, but there are some long-vowel imperfects which in my view
simply copy the old relation *es-/*e:s- of 'be'. Old Irish no-bered 'was
carrying' is from the middle-voice ipf. *bhereto, notably always
compounded (if only by the default preverb no-) and so rather obviously
continuing an augmented form. No matter what one thinks of Lat. ama:bam it
does contain the same span as Oscan fufans and so adds the same preterital
marker to the present stem as the latter had added to the perfect stem;
and ama:ver-a:-s has preteritalized the perfect stem just as er-a:-s has
the present stem, so here, too, the ipf. is the preterite of the present
stem. Even Albanian ish or ishte (the C,amian forms) may artlessly reflect
*est (in part with productive superimposed ending, probably borrowed from
the aor. qe), i.e. the present stem with secondary ending. Only Germanic
and Anatolian have obliterated the imperfect as an independent category,
using what's left of it simply as past tense (in Gmc., e.g. Eng. did is an
old ipf.). I cannot accept the statement that the imperfectivity of the
"imperfect" is restricted to Greek: the imperfect is opposed to a
non-durative preterite wherever it occurs, except for Indo-Iranian. That
I-Ir. and Anatol. have both innovated, should cause no concern, especially
since the two innovations have nothing in common.

> [Jens, quoted from a different post]
>> The most rewarding experience during the time I have been
>> watching Indo-European Studies has been to see the protolanguage
>> come alive and assume an increasingly well-established structure, [...]

[Nath:]
> Do we really understand the variety of syntactic structures and their
> diachrony that well? I have mentioned the Tamil -vidu construction a few
> times. You will find some linguists call that a perfective and the Tamil
> simple past an imperfective. This is simply wrong as the simple past is and
> has been the tense of narration for the 2000+ year recorded history, and
> this distinction is nothing like the perfective-imperfective distintion in
> Russian or Arabic [...].

[Jens:]
Maybe we have something as important as the explanation of the Indic
development here. Perhaps Anatolian and Iranian have been influenced by
some common source which could not distinguish different types of
synthetic preterites?

[Nath:]
> It can be even worse: Similar constructions exist in some NIA languages
> which also have a prefective-imperfective distinction that is different
> (imperfective out of old progressive). I have seen at least one linguist
> apply the label of perfective to the construction with auxillary, and
> imperfective to the one without the auxillary. This completely misrepresents
> the syntax.

> What reason is there to think that we are not making the same kind of
> mistake with PIE? I am not asking for certainity here, but evidence for
> greater probability than the opposite.

[Jens:]
As long as our mistakes are comparable to calling the expression of
imperfective action "imperfect" and the expresiion of perfective action
"perfect" in a language that really has a perfective-imperfective
distinction, I see little cause for alarm.

Thank you for your patience and most stimulating questions and
observations.

Jens


From proto-language at email.msn.com  Fri Jul 30 04:35:12 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Thu, 29 Jul 1999 23:35:12 -0500
Subject: Semantic change
Message-ID: <THU.29.JUL.1999.233512.0500.>

Dear Lloyd and IEists:

 ----- Original Message -----
From: <ECOLING at aol.com>
Sent: Monday, July 19, 1999 10:08 AM

> A less plausible semantic relation means that two
> look-alikes are less plausible as cognates.
> But requiring "identity" of meanings or even
> "near-identity" of meanings is an absurd
> requirement also, when we are working at great
> time depths.

> So what to do?
> Asserting simplistic extremes either for or against
> is not particularly useful.

Pat responds:

I agree completely. It is the subjectivity of the permissible limits that
militates against a consensus. It is the frustration of this situation that
has prompted well-intentioned researchers to insist on identity, which, as
you correctly identify, is also problematical. I sincerely wish there was
some mechanical or mathematical way to set these limits --- if only as I
check on some of my own wilder flights of fancy.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From 114064.1241 at compuserve.com  Fri Jul 30 10:51:00 1999
From: 114064.1241 at compuserve.com (Damien Perrotin)
Date: Fri, 30 Jul 1999 06:51:00 -0400
Subject: Hittite & Celtic dative in /k/ ?
Message-ID: <FRI.30.JUL.1999.065100.0400.>

[ moderator re-formatted ]

Initial message from
Lloyd Anderson

[ moderator snip ]

>Here is a tidbit linking Celtic and Hittite which I found many years ago, when
>compiling typological comparisons of the semantic domains of "be" and "have".
>I wondered at the time whether it is a relic Dative case preposition /
>postposition which Hittite and Celtic shared, presumably as a retention, but
>conceivably as an areal phenomenon, or both of the above.  Does Tocharian have
>it too?

>(I have no idea whether the later Indic Dative postposition with /k/ is
>related or a chance lookalike.)

>The forms being discussed, the dative case of the 1st singular pronoun, are
>statistically likely to be highly conservative, both because they are pronouns
>and also as oblique cases rather than the more often innovating nominative
>etc.

[ moderator snip ]

the comparison is unlikely, as the Hittite form is also used for the
accusative, an use for which it has cognates in Germanic (gotic mik) and
perhaps in Slavic (russian ko - to with only an allative meaning). The
primary meaning was probably allative, as for most accusative in IE.
the Goidelic form is assumed to derive from *angh "near" which is found in
Latin angustus. Its closest cognates, assuming to Stokes, is Breton hag
(and) and Welsh ac (same meaning)


From 114064.1241 at compuserve.com  Fri Jul 30 11:01:23 1999
From: 114064.1241 at compuserve.com (Damien Perrotin)
Date: Fri, 30 Jul 1999 07:01:23 -0400
Subject: -n- adjectival suffix in Latin
Message-ID: <FRI.30.JUL.1999.070123.0400.>

[ moderator re-formatted ]

Paolo Agostini<
wrote

>I was wondering about the origins of the Latin adjectival
>suffix -n-us, -n-a, -n-um appearing in words like _mater-nus_ "maternal;
>motherly; belonging/pertaining to the mother", _pater-nus_ "paternal;
>fatherly; belonging/pertaining to the father", _feli-nus_ "feline;
>belonging/pertaining to the cat; belonging/pertaining to the family/genus
>of the cats/felidae" and the like.  Can it be traced back to IE? Does the
>morpheme exist in other IE or non-IE languages? Any idea abt its etymology
>and/or development?

This ending is first found in Etruscan, where it has the same use as in
Latin, but as the affiliation of Etruscan is far from being sure, it is not
of great utility.
A similar ending is used in Russian to form past passive participles
(-nny). A suffix -ina- is also used in Germanic to form adjectives (Gotic
quineins : feminine)

Damien Erwan Perrotin


From kurisuto at unagi.cis.upenn.edu  Fri Jul 30 14:23:36 1999
From: kurisuto at unagi.cis.upenn.edu (Sean Crist)
Date: Fri, 30 Jul 1999 10:23:36 -0400
Subject: Principled Comparative Method
In-Reply-To: <ed0acce.24c49979@aol.com>
Message-ID: <FRI.30.JUL.1999.102336.0400.>

On Mon, 19 Jul 1999 ECOLING at aol.com wrote:

> So if we had case (d)
> and we also had case (d'),
> one example of IE /t/ corresponding to Uralic /s/,
> we might have more evidence for saying that /d/~/z/ was real.

> Or if we had case (d) before a following /u/,
> and we also had case (d'')
> ONE example of IE /d/ corresponding to Uralic /z^/ before /i/,
> (interpret the ^ as a hachek = wedge, upside down)
> we would again have more evidence for saying that /d/~/z/ was real,
> in this case because we would assume that /z/ became /z^/ before /i/.

> And if lucky we might have case (d') above with following /u/,
> and also case (d'''),
> one example of IE /t/ corresponding to Uralic /s^/ before /i/.

> With all four of these examples, we would have a plausible system,
> and would in fact have four examples reflecting the same sound changes,
> perhaps actually only a single sound change (namely
> "affricate apical stops before high vowels, with the result being
> the fricative of the same voicing and "reflecting" the vowel quality)
> but we need not have more than one example of any particular correspondence
> on the surface.

	As a matter of terminology, the kind of argumentation you propose
here falls outside the Comparative Method.  It's true that we do sometimes
fall back on this kind of external guesswork, but it's the sort of special
pleading you make to bolster some assumption which you need for your
argument.

	In any case, I'd be really careful in assuming that phonological
rules always apply across entire classes of categories.  It's true that
things often do work this way, but they don't always.
	Around 12 years ago, I was interviewing a speaker of the South
Midlands dialect area of American English.  In this dialect, /I/ > /i/ for
some speakers, and /U/ > /u/ for some speakers.  I assumed that if the
speaker had one rule, she'd have the other as well; I figured that the
general rule was "high lax vowels become tense".  To my great surprise,
she had the second rule but not the first.  There was no denying what I
was hearing; my assumptions about the expected symmetry within the system
were just plain wrong.
	I've fallen into this same kind of trap plenty of other times.
There are so many cases of beautiful symmetry and parallelism in
phonological systems that it's an ongoing challenge to remember that
things don't always work out so neatly.  Even in the Japanese case that
you give, there are recent loan words (e.g. tiishatsu "T-shirt") to mess
up the nice symmetry of the system.

> One of them I am quite certain is to develop both articulatory and
> acoustic "spaces", relative "distances" between different articulations
> and different acoustic effects, so that when attempting to judge
> likelihood of cognacy of pairs of words, we can judge similarity
> by degrees, not by yes/no dichotomies.  (These will partly depend
> on the general typology of the sound systems of the languages
> concerned, they will not be completely universal, but they will
> also not be completely idiosyncratic.)

The Comparative Method is concerned with the reconstruction of categories,
_not_ the phonetic values which might have been the realization of those
categories.  This is a very important point.  When we talk about
Proto-Indo-European */a/, we don't mean "the phonological category in
Proto-Indo-European which had the phonetic realization [a]"; we mean "the
hypothetical PIE category which gave rise to the Sanskrit category /a/,
the Latin category /a/, etc."

As far as the Comparative Method is concerned, we could designate that PIE
category with an integer, e.g. "Category 27".  */a/ is just a convenient
label or nickname for it.  Along with that label comes a built-in guess
about what the prehistoric phonetic value for that category might have
been, but this is _just_ a guess; the actual phonetic value is beyond the
reach of the Comparative Method.

To look at it another way, imagine that we have two languages, A and B.
Suppose that in every single case where A has dental [d], B has alveolar
[d], and vice versa.  From the standpoint of the Comparative Method, A and
B are the same language, because there have been no mergers in categories.
Strictly speaking, the phonetic realization of the categories is of no
consequence to the Comparative Method.

It is true that we do sometimes point out that if the phonetic values for
such-and-such categories were such-and-such, then such-and-such a rule is
phonetically plausible.  But this is guesswork on top of guesswork, and
it's really shaky ground to build an argument on.

  \/ __ __    _\_     --Sean Crist  (kurisuto at unagi.cis.upenn.edu)
 ---  |  |    \ /     http://www.ling.upenn.edu/~kurisuto/
  _| ,| ,|   -----
  _| ,| ,|    [_]
   |  |  |    [_]


From proto-language at email.msn.com  Fri Jul 30 14:59:15 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Fri, 30 Jul 1999 09:59:15 -0500
Subject: Latin perfects and Fluent Etruscan in 30 days!
Message-ID: <FRI.30.JUL.1999.095915.0500.>

Dear Damien and IEists:

 ----- Original Message -----
From: Damien Erwan Perrotin <114064.1241 at Compuserve.com>
Sent: Wednesday, July 28, 1999 2:14 PM

[ moderator snip ]

> [Damien Erwan Perrotin]

> The idea is interesting, but probably inexact as am- is not present only
> in Latin but also in Lydian (ama : to love) and in Breton (afan : to
> kiss from an older Brythonnic *ama). It is still possible that the
> Lydian form was a borrowing from an Etruscan-like tongue of the Aegean,
> but that is quite unlikely for Breton. So there can be only three
> explainations for the ressemblance you point out :
> a: chance ressemblance (always possible)
> b: Etruscan borrowed the word from Latin or from Celtic
> c: Etruscan is remotely linked to IE and this root is a remnant of this
> old relationship.

> Personnally, I favor the third thesis, but there is still work to do
> before proving it.

Pat comments:

Some might be interested in Egyptian jm-3, 'kind, gentle, well-disposed,
pleasining, gracious, be delighted, charmed', which I believe is likely to
represent an AA example of the same root.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From JoatSimeon at aol.com  Fri Jul 30 20:44:32 1999
From: JoatSimeon at aol.com (JoatSimeon at aol.com)
Date: Fri, 30 Jul 1999 16:44:32 EDT
Subject: Lexical Retention
Message-ID: <FRI.30.JUL.1999.164432.EDT.>

>Patrick C. Ryan

>In any case, you have it rather backwards. If the root and all manifestions
>of it (like, perhaps, in 'bear') were absent in Germanic, what would, as it
>has, make possible our reconstructing an IE root would be its presence in,
>at least, three (nominally) other branches.

-- quite true, if it _exists_  in at least three branches.  Some roots,
obviously, do.  Some, equally obviously, don't -- we don't have a complete
vocabulary of PIE, to put it mildly.  And in the nature of things we never
can.

Some parts of PIE have been completely and irrevocably lost because they
didn't survive into extant or recorded languages.

Obviously, this portion will be larger the further back we go in time and the
less we have in the way of writtten records of forms no longer extant -- a
reconstructed proto-Romance, even if we had no records of Latin, would be
much more complete than PIE.

The greater the temporal distance, the more information lost to entropy.


From JoatSimeon at aol.com  Fri Jul 30 20:53:35 1999
From: JoatSimeon at aol.com (JoatSimeon at aol.com)
Date: Fri, 30 Jul 1999 16:53:35 EDT
Subject: Hittites ~ Phrygians ~ Balkan peoples?
Message-ID: <FRI.30.JUL.1999.165335.EDT.>

ECOLING at aol.com

>There has been much discussion of the Hittites and the Phrygians, and other
>names for these or other peoples of Anatolia and the Balkans (Thracians,
>Illyrians, Dacians, Moesians, etc. etc.),  which may have been incorrectly
>taken as distinct peoples in the past, instead of simply as different names
>used for the same peoples at different times or by different authors or
>cultures.

-- Hittite and Phrygian are distinct.  They're both IE, of course, but
Phrygian is not a member of the Anatolian subgroup of IE.  Like Armenian,
it's intrusive in Anatolia and is post-Hittite.


From proto-language at email.msn.com  Fri Jul 30 23:52:56 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Fri, 30 Jul 1999 18:52:56 -0500
Subject: `cognate'
Message-ID: <FRI.30.JUL.1999.185256.0500.>

Dear Adam and IEists:

 ----- Original Message -----
From: Adam Hyllested <adahyl at cphling.dk>
Sent: Wednesday, July 21, 1999 11:44 AM

Adam wrote:

> It is more or less the same reconstruction. <@> is (originally, at
> least) the vocalic allophone of <H>, reconstructed on the basis of
> Indo-Iranian, where the root vowel is <i> (an IE <**pater> would have
> developed into Sanskrit <**pata:> and not the actual <pita:>)
> Using <@2> or <H2> is simply a matter of phoneme-to-grapheme
> correspondence.

Pat comments:

A few of you may be interested in reading the short essay:

http://www.geocities.com/Athens/Forum/2803/comparison-AFRASIAN-3_schwa.htm

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From X99Lynx at aol.com  Thu Jul  1 07:31:10 1999
From: X99Lynx at aol.com (X99Lynx at aol.com)
Date: Thu, 1 Jul 1999 03:31:10 EDT
Subject: indoeuropean
Message-ID: <THU.1.JUL.1999.033110.EDT.>

In a message dated 6/29/1999 12:17:30 AM, Sunnet at worldnet.att.net writes:

<<Correction: Slavic *renka/ronka means "arm", not "hand". The word for "hand"
in Russian is /kist'/ - looks like it wasn't lost there after all.

Gene>>

On the road but checking the list and I couldn't let this one go by.

'Ruka/ruce' (Czech) and 'reka/rece' (Pol) definitely are hand/hands and have
been for many centuries.  Wherever <<*renka = arm>> comes from, it
demonstrates again that Russian and 'Slavic' are not equivalents.

In Czech, 'Naruci' will translate as 'arms'.  But this is no real surprise.
I recall that 'Cheir' itself can mean either or both hand and arm in the
Illiad.

In a message dated 6/28/1999 6:16:14 PM, georg at letmail.let.LeidenUniv.nl
writes:

<</manus/ seems to be able to claim an IE pedigree as well, with as yet
unknown semantic differences between it and *ghes-r-.... The other
possibility is that the more abstract meanings (circling around "power,
ability to protect athl.")
are older for this etymon, *ghes-r- being the anatomical term from the
beginning....

Other languages have replaced this apparently oldest word for "hand", to wit
Latin manus, Gothic handus,Baltic and Slavic *renka/ronka (from a verb
meaning "to grasp", cf.Lithuanian /rinkti/.>>

There is possibly another explanation for 'manus.'  'Mane^s' appears in Greek
as early as Aristophanes as a common word for a slave.  It is explained by
Strabo later on as being a name borrowed from the Phrygian.  Strabo explains
that the self-names of foreign captives became the nick-names used by the
Greeks - 'mane^s' being one of them given as an example.  'Mancips' of course
became the Latin word for a slave, particularly a captive.  The obvious
connection between hands and manual labor - be they farmhands or otherwise -
might explain the multiple sense.  And might even be the source of the
Germanic 'mann'.  The relation between 'androo'/'aner' and 'hand'/'manus'
might also be something to look into.

[ Moderator's note:
  These words obviously cannot be related.
  --rma ]

'Cheir' itself in C. Greek suggests the same relation between status and
'hand' - 'cheiron' in Homer refers to a person of 'meaner rank', inferior.

'*ghes-r-' as the original form for hand seems to be associated in Homer
(aside from cheir?) with such things as handcrafts - 'kestoros',(embroidery),
'keiro' (carving) cf. theros (harvest), ge^s (of the earth), keranos
(pottery, anything made of earth), kiste, ches/chernips/chersos (cups and
such), ar[i]/isteros (left handed).  Perhaps this was not THE original word,
but only one tradition among a craftsperson class.

[ Moderator's note:
  These words obviously cannot be related.
  --rma ]

There is also an odd thing that happens in early Greek.  'Dexios' refering
mainly only to the right hand, but with an obvious (to me at least) relation
to 'deka' (ten) suggesting the name for the count of fingers on both hands
was shifted to describe one hand but not the other.  'Endexios' in Homer
however refers to 'from one hand to the other', literally between hands,
possibly reflecting an earlier usage.

Finally there is again the Greek 'keiro' (to cut, carve) that would explain
churgeon/surgeon much more adequately than a direct connection to /cheir/.
In Aristophanes The Birds, 'keirulos' describes a barber.

[ Moderator's note:
  These words obviously cannot be related.
  --rma ]

Regards,
Steve Long


From edsel at glo.be  Thu Jul  1 14:18:51 1999
From: edsel at glo.be (Eduard Selleslagh)
Date: Thu, 1 Jul 1999 16:18:51 +0200
Subject: Latin perfects and Fluent Etruscan in 30 days!
Message-ID: <THU.1.JUL.1999.161851.0200.>

-----Original Message-----
From: Macia Riutort Riutort <mrr at astor.urv.es>
Date: Tuesday, June 29, 1999 3:39 AM

>Es gibt eine kleine Anzahl von W?rtern im Spanischen, die im Kastilischen
>-rd- aufweisen, im Katalanischen jedoch -rr-. Zum Beispiel:
>Katalanisch: esquerre - esquerra - Spanisch izquierdo - izquierda
>Katalanisch: cerra - Spanisch cerda ("Borste")
>Katalanisch: marr? - Spanisch mardano ("Schafbock") usw.
>Da dies haupts?chlich in W?rtern aus dem Substrat vorkommt, wird
>angenommen, dass -rd- bzw. -rr- die Anpassung an das vorkastilische bzw.
>vorkatalanische Lautsystem eines ihm unbekannten Lautes -oder Lautgruppe- der
>gebenden Substratsprache.

>M.R.

>>Yes, though there is a phonological problem here, since Basque <ezker>
>>`left', definite form <ezkerra>, should not have yielded Castilian
>><izquierdo> (m.), <izquierda> (f.).  Since there is evidence that Basque
>>once had a word-forming suffix *<-do>, meaning something like `bad
>>thing', it is possible that an unrecorded Basque derivative *<ezkerdo>
>>was borrowed into Castilian before being lost from Basque itself.
>>Nobody knows.

>>Larry Trask

[Ed Selleslagh]

I just re-discovered an article you're probably interested in:

Mary Carmen Iribarren-Argaiz*: "Los vocablos en -rr- de la lengua sarda.
Conexiones con la pen?nsula ib?rica", Fontes Linguae Vasconum
(Pamplona/Iru?ea, Navarra/Nafarroa), N? 76, pp. 335-354, Sept.-Dic. 1997.
It has a huge bibliography.
*University of New Mexico (in 1997).

It is based upon her doctoral dissertation at the University of Florida,
Gainesville, 1995: "Origen y desarrollo de la sufijaci?n ibero-romance
en -rr-: vinculaciones y contrastes con las otras lenguas".

Resumen:
El presente estudio examina c?mo el sardo, especialmente en sus dialectos
centrales, contiene elementos relacionables con las lenguas ib?ricas.  Entre
tales elementos, para los estudiosos del euskera resulta de particular
inter?s la existencia de antiguos vocablos sardos con formaciones en -rr- en
sus s?labas finales. En sardo estas formaciones han quedado fosilizadas y
resulta m?s evidente su car?cter de elemento prerromano, como hemos dicho
relativamente fosilizado, de poca evoluci?n y de muy escasa o inexistente
productividad sufijal rom?nica.
Ciertas circunstancias hist?ricas como colonizaciones, expediciones
militares u otros contactos de naturaleza m?s espor?dica hacen veros?mil la
hip?tesis de contactos de lenguas en los que hablantes de vasco pueden haber
estado implicados.  La exposici?n hist?rica y los argumentos ling??sticos
ofrecen interesantes sugerencias en la l?nea de la expansi?n mediterr?nea de
vocablos vascos.

Ed.

[ moderator re-encoded ]

-----Original Message-----
From: Macia Riutort Riutort <mrr at astor.urv.es>
Date: Tuesday, June 29, 1999 3:39 AM

>Es gibt eine kleine Anzahl von W{\"o}rtern im Spanischen, die im Kastilischen
>-rd- aufweisen, im Katalanischen jedoch -rr-. Zum Beispiel:
>Katalanisch: esquerre - esquerra - Spanisch izquierdo - izquierda
>Katalanisch: cerra - Spanisch cerda ("Borste")
>Katalanisch: marr{\`a} - Spanisch mardano ("Schafbock") usw.
>Da dies haupts{\"a}chlich in W{\"o}rtern aus dem Substrat vorkommt, wird
>angenommen, dass -rd- bzw. -rr- die Anpassung an das vorkastilische bzw.
>vorkatalanische Lautsystem eines ihm unbekannten Lautes -oder Lautgruppe- der
>gebenden Substratsprache.

>M.R.

>>Yes, though there is a phonological problem here, since Basque <ezker>
>>`left', definite form <ezkerra>, should not have yielded Castilian
>><izquierdo> (m.), <izquierda> (f.).  Since there is evidence that Basque
>>once had a word-forming suffix *<-do>, meaning something like `bad
>>thing', it is possible that an unrecorded Basque derivative *<ezkerdo>
>>was borrowed into Castilian before being lost from Basque itself.
>>Nobody knows.

>>Larry Trask

[Ed Selleslagh]

I just re-discovered an article you're probably interested in:

Mary Carmen Iribarren-Argaiz*: "Los vocablos en -rr- de la lengua sarda.
Conexiones con la pen{\'i}nsula ib{\'e}rica", Fontes Linguae Vasconum
(Pamplona/Iru{\~n}ea, Navarra/Nafarroa), N{^o} 76, pp. 335-354, Sept.-Dic.
1997.
It has a huge bibliography.
*University of New Mexico (in 1997).

It is based upon her doctoral dissertation at the University of Florida,
Gainesville, 1995: "Origen y desarrollo de la sufijaci{\'o}n ibero-romance
en -rr-: vinculaciones y contrastes con las otras lenguas".

Resumen:
El presente estudio examina c{\'o}mo el sardo, especialmente en sus dialectos
centrales, contiene elementos relacionables con las lenguas ib{\'e}ricas.
Entre tales elementos, para los estudiosos del euskera resulta de particular
inter{\'e}s la existencia de antiguos vocablos sardos con formaciones en -rr-
en sus s{\'i}labas finales. En sardo estas formaciones han quedado fosilizadas
y resulta m{\'a}s evidente su car{\'a}cter de elemento prerromano, como hemos
dicho relativamente fosilizado, de poca evoluci{\'o}n y de muy escasa o
inexistente productividad sufijal rom{\'a}nica.  Ciertas circunstancias
hist{\'o}ricas como colonizaciones, expediciones militares u otros contactos de
naturaleza m{\'a}s espor{\'a}dica hacen veros{\'i}mil la hip{\'o}tesis de
contactos de lenguas en los que hablantes de vasco pueden haber estado
implicados.  La exposici{\'o}n hist{\'o}rica y los argumentos
ling{\"u\'i}sticos ofrecen interesantes sugerencias en la l{\'i}nea de la
expansi{\'o}n mediterr{\'a}nea de vocablos vascos.

Ed.


From X99Lynx at aol.com  Tue Jul  6 06:48:05 1999
From: X99Lynx at aol.com (X99Lynx at aol.com)
Date: Tue, 6 Jul 1999 02:48:05 EDT
Subject: Chronology of the breakup of Common Romance/CELTS
Message-ID: <TUE.6.JUL.1999.024805.EDT.>

In a message dated 6/18/99 10:36:23 PM, mclssaa2 at fs2.mt.umist.ac.uk wrote:

<<I have seen a suggestion that:-
  (1) The Romanians are not descended from the Dacians, but from Vlachs
(scattered nomadic mountain shepherds who are found over much of the Balkans)
who found what is now Romania....>>
<<(2) The Vlachs are descended from Latinized Illyrians who fled into the
mountains when the Avars marauded in;
  in which case the ancestors of the Romanians started learning Latin when
Rome invaded Illyria, not when Rome invaded Dacia, and that alters the
linguistic timetable a bit.>>

There's another element in this piece of history that could skew everything.

"Vlach" or "Wallach" seems to be descended from a name that was commonly
applied to Celts.  Appearing as "Walh" or "Walah" in OHG, it has interpreted
as meaning "foreigner", sometimes Roman, but in usage it is closely
associated with Celts and regions of Celtic habitation - e.g., "Wales",
"Walloon".  I seem to recall an suggestion that there may be a tie here to L.
vallum (fortified wall, the earliest meaning of "wall") and refer to the
Celtic or Romano-Celtic oppidum or walled town.

The strong Celtic presence along the Danube is attested by Classical writers
well before the common era.  E.g., Alexander fights them before turning
against the Persians and meets Celts who are from Illyria.  Galicia in
southern Poland is a region whose name remembers a Celtic presence even
farther north.  I remember an old passage where Vlachs are identified as "the
shepherds of the Romans" and in this role they may also have been imported
help as they were in northern Italy.

<<in which case the ancestors of the Romanians started learning Latin when
Rome invaded Illyria, not when Rome invaded Dacia, and that alters the
linguistic timetable a bit.>>

The ancient ethnic designations in that part of the world are a little
difficult to follow,  but it seems clear that the whole region from Illyria
to present day Romania was under Rome by 250 ace.  And there is some
possibility that Vlachs represented Romanized Celts across those regions.  In
any case, the small difference in time between Rome's entry into Illyria and
into Dacia would be a de minimis factor.

In a message dated 6/20/99 5:55:33 PM, rmccalli at sunmuw1.MUW.Edu wrote:

<<Rumanian does share a common pre-Romance substrate with Albanian
which presumibly should help pinpoint the origin of Rumanian.>>

The suggested Thraco-Illyrian-Albanian substrate has always been called
tentative, especially in light of potential borrowing across short distances
where there would have been much opportunity for contact, particularly prior
to the Slavic invasions of the Balkans.

This brings up that familiar problem of finding a Celtic substrate, one that
might be fruitful.

E.g., one example of the Albanian substrate often given (per G. Mallinson)
has been Rom. "abure"/ Alb. "avull" (steam.)  But in Gaelic I see "co-bur"
(foam), "to-bar" (well) and "bruich" (boil) - add the Gaelic "a-" (out of) to
get /a-bruich (out of boiling- steam?) and the Celtic seems as possible as
the Albanian as a source.  Another substrate example has been Rom. "vatra"/
Alb. "vatre" (hearth), but perhaps forms like the Gaelic "fadadh" (kindling)
and "bradhadair" (blazing fire) offers evidence of a common origin and
original meaning for these later similarities.

I haven't found any recent consideration of the Celtic remnants in this area,
but I would think it might offer some real possibilities.

In a message dated 6/20/99 5:55:33 PM, rmccalli at sunmuw1.MUW.Edu wrote:

<<BUT there is also an argument that Albanian is descended from Thracian
speakers who fled to the west>>

An important question would be when this migration would have occured.

<<I seem to remember reading that while Albanian has a strong substrate, that
it is significantly lacking in Ancient or Classical Greek substrate. If
this is true, then it tells us where not to look...>>

But the problem is that Rumanian does have a substantial Greek substrate (and
had a larger one centuries ago before Rumanian was "purified" in recent
times.)  Much of this is attributed to Old Church Slavonic carrying Greek
lexicon and to Greek administration under the Turks. (The highly-related
Arumanian particularly has morphological and syntactic similarities to Greek,
e.g, the use of compound pluperfect.)  But as one scholar pointed out, "The
desire to improve and purify Rumanian... has complicated the already
difficult task of carrying out research on the origins of Rumanian
vocabulary."

Furthermore, given the extreme amount of intercourse between Greek and Thrace
during those many early centuries, wouldn't one expect Thracian to have a
strong and ancient Greek substrate, especially with regard to trade items and
such?  If Albanian doesn't have such substrate, it may not be Thracian.

Regards,
S. Long


From JoatSimeon at aol.com  Wed Jul  7 04:46:51 1999
From: JoatSimeon at aol.com (JoatSimeon at aol.com)
Date: Wed, 7 Jul 1999 00:46:51 EDT
Subject: accusative and ergative languages
Message-ID: <WED.7.JUL.1999.004651.EDT.>

>proto-language at email.msn.com writes:

>So what? It only requires us to expand our universe of applicable data. The
>idea that vocabulary is irretrievably lost is jejeune.

-- actually, it's obvious.  Information vanishes as entropy increases.  If we
didn't have written or other artificially preserved examples of now-extinct
languages, much of the PIE vocabulary we have would be completely
unrecoverable.

Try reconstructing PIE using _nothing_ but contemporary Albanian and English.
You'd be hard-put to prove that such a language even existed.


From proto-language at email.msn.com  Wed Jul  7 05:18:47 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Wed, 7 Jul 1999 00:18:47 -0500
Subject: PIE vs. Proto-World (Proto-Language)
Message-ID: <WED.7.JUL.1999.001847.0500.>

Dear Steven and IEists:

 ----- Original Message -----
From: Steven Schaufele <fcosw5 at mail.scu.edu.tw>
Sent: Tuesday, June 29, 1999 1:08 PM

> I would, however, reject any suggestion that our knowledge of PIE is
> ipso facto on a par with our knowledge (or lack thereof) of
> `Proto-World'.  Though i admit frankly to being little more than an
> amateur at this game, i am quite confident of *most* of what we claim to
> know about PIE.  Although i'm not prepared to go as far as Calvert
> Watkins (i think it was?) who composed a fable in PIE, i certainly do
> not doubt that, in principle, it could be done with our current state of
> knowledge.  Whereas i regard `Proto-World' as little more than an
> entertaining fantasy.

There are others who are working on reconstructing "Proto-World" from a
totally different methodology.

What I have been working on, I call "Proto-Language".

I simply do not understand why some find it difficult to understand that
reconstructing the Proto-Language is only primarily different from
reconstructing Indo-European in the wider selection of source languages for
data. Of course, secondarily I have labored to reconstruct the underlying
monosyllables by analysis of attested compounds.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From edsel at glo.be  Wed Jul  7 10:35:36 1999
From: edsel at glo.be (Eduard Selleslagh)
Date: Wed, 7 Jul 1999 12:35:36 +0200
Subject: 50% Spanish or German, 50% Chinese
Message-ID: <WED.7.JUL.1999.123536.0200.>

[snip]

>At the risk of beating an undead horse, i would very much like it
>reiterated that not all of us have software capable of recognizing the
>means by which some lucky subscribers' software can encode diacritics
>and other expanded character sets.  And worse, some of us have software
>that recognizes such codes, but interprets them quite differently from
>the way they are intended.

[Ed Selleslagh]

I know of no software (e.g. Windows, text processors, browsers, e-mail
programs, etc.) that are incapable of handling the full 256 characters
(using 8 bits) of code page 437 or 850, unless you are using a '70's IBM
mainframe.

[ Moderator's comment:
  Try a 1990's XKL mainframe.  One that uses 7-bit ASCII natively, and barfs
  on anything else.  Like the one running this mailing list.
  --rma ]

It is simply a matter of settings which you can change easily. Unless... See
below.

>For instance, the software we have here at SCU automatically converts
>all such codes into representations of various Chinese characters.  The
>result is that i have recently received many postings including extended
>and richly-exemplified discussions of Spanish vocabulary, and most of
>the words in question have been printed half in Spanish and half in
>Chinese, to the point that i have been unable to make any sense out of
>the posting at all, and have regretfully decided i must automatically
>dump & ignore the whole discussion.

[Ed]

In the English speaking countries a number of servers, but far from all,
use only 7-bit encoding, i.e. only 128 characters.

In the case of 8-bit transmission, local (on your own PC) switching to the
right character set (code page) will solve all the problems (e.g. in
Windows: 'international settings'). You clearly use another code page in
which only the first 128 characters are recognized as classic ASCII. BUT:
languages like Chinese usually use 16-bit characters; I guess your PC is set
for a 16-bit (256 x 256 = 65,536 characters) code page, the first 128 of
which are reserved for the classic ASCII characters like in all non-Latin
code pages (e.g. Greek, Cyrillic, etc.).

Apparently, the *.xkl.com server correctly transmits 8-bit (256) codes,
which can locally (your PC) be recognized correctly using the right
settings, since I receive the diacritics correctly, using code page 850 (437
works equally well).

[ Moderator's comment:
  No, the server here transmits only 7-bit ASCII.  MIME-quoted-printable
  messages, which use only 7-bit ASCII, are translated at the *receiving* end
  into 8-bit (which I cannot read with any mail program available to me, to
  check on this).

  Further, you are assuming a Windows system, with your discussion of "code
  pages" and the like.  Neither Macintosh nor Unix systems agree with Windows
  on 8-bit conventions.  Therefore, it is unfriendly to use 8-bit characters
  in mailing list messages.
  --rma ]

Of course, there may be some people on this list that depend on a local
(somewhat old-fashioned) server that only transmits 7 bits, in which case
there is no solution to the problem.

[ Moderator's comment:
  Your moderator, for instance.
  --rma ]

It is not obvious that setting the right code page would eliminate your
'Chinese problem' because of the 16-bit setting, which creates a problem on
a deeper level than that among European-type PC's (actually their software).

>I've recently begun noticing similar problems with postings in German.

[Ed]

The solution is the same for all Western European languages (code page 437
or 850).

[ moderator snip ]

>[ Moderator's comment:
>  I have pointed this out in the past, and been roundly excoriated for my
>  point of view, taken somehow to be "English-only".  I will once again
>  suggest that we adopt a modified TeX-like accent-writing system, in which
>  the accent (in the typographical sense, which includes umlaut/diaeresis/
>  trema and the like) is written next to the character affected.  (In TeX
>  systems, it must precede, but I think that context can disambiguate for
>  human readers.)  Should I send out a list of the TeX conventions, for those
>  unused to them?
>  --rma ]

[Ed]

The list would be very welcome.
However, there is a problem: on most non-US keyboards, either the diacritics
are on dead keys (How do you write a diacritic without a letter under it? I
tried the dead key plus Spacebar: it works) or the letters with diacritics
are normal keys (Maybe you have to use ALT+number codes?).

[ Moderator's response:
  The diacritics used in TeX are 7-bit ASCII characters.  I will post a list of
  them shortly.  Let's get caught up on the backlog first.
  --rma ]


From X99Lynx at aol.com  Wed Jul  7 13:02:14 1999
From: X99Lynx at aol.com (X99Lynx at aol.com)
Date: Wed, 7 Jul 1999 09:02:14 EDT
Subject: PIE vs. Proto-World
Message-ID: <WED.7.JUL.1999.090214.EDT.>

Pat Ryan <proto-language at email.msn.com> writes:

<<Of course we have no data on the earliest human language in the same
way we have no data on IE>>

There's a slight problem here in our use of the word "data."  Data is
strictly speaking raw and awaiting interpretation.  In that sense, there is
data on both PIE and earlier languages.  The level of certainty about our
conclusions based on this data however is the real question.

Because neither PIE or early human languages are directly observable, we are
always dealing with inferences from data.  Much as with radiation or the
atom, we draw conclusions from secondary events.  These secondary events
would be the consequences of there having been a PIE or an early universal
human language (if there was one.)

The data pool in which we look for evidence of PIE is a large one - starting
with the attributes of every known IE language over thousands of years of
time.  But as large as that potential mass of data is, it is nothing compared
to the data we need to examine in pursuit of a universal human language.  How
many times as large?  Hundreds, thousands?  Maybe more?

And there is also the fact that the comparative is not available.  A
universal language excludes no past or present language, so there is no way
to differentiate x from y.  This makes the null hypothesis (there was never a
universal language) difficult to test.

Two hundred years ago it was not apparent that there had been such a thing as
PIE.  The pool of data was there, but the hypothesis of a single parent
tongue had not been made.

There is nothing inappropriate about searching for evidence of a proto-world
language.  But it should be obvious that the task is many, many times the
size of tracking PIE.

In a message dated 7/6/99 10:42:34 PM, fcosw5 at mail.scu.edu.tw responded:

<<Although i'm not prepared to go as far as Calvert Watkins (i think it was?)
who composed a fable in PIE, i certainly do not doubt that, in principle, it
could be done with our current state of knowledge.>>

I'm beginning to suspect this isn't quite true.  There are two factors of
interference with our certainty about PIE.  One is endemic to any historical
science.  As in geology, we are dealing with strata that can overlap,
disappear or appear out of chronological context.  The relationship between
Latin and Romance languages is a good example.  A second factor is semantic
shift.  This appears to be the weak link in identifying cognition and
relationships between forms in time.

regards,
Steve Long


From thorinn at diku.dk  Wed Jul  7 13:37:12 1999
From: thorinn at diku.dk (Lars Henrik Mathiesen)
Date: Wed, 7 Jul 1999 15:37:12 +0200
Subject: 50% Spanish or German, 50% Chinese
In-Reply-To: <377909D2.1335@mail.scu.edu.tw> (message from Steven Schaufele on
 Tue, 29 Jun 1999 11:00:50 -0700)
Message-ID: <WED.7.JUL.1999.153712.0200.>

> Date: Tue, 29 Jun 1999 11:00:50 -0700
> From: Steven Schaufele <fcosw5 at mail.scu.edu.tw>

> [... I] have recently received many postings including extended and
> richly-exemplified discussions of Spanish vocabulary, and most of
> the words in question have been printed half in Spanish and half in
> Chinese [...]

Check to see if the mail program and/or the terminal program you are
using has a character set setting --- and set it to Latin-1 to read
list mail. If it doesn't, enter a bug report.

(Technically, for this to happen, the software has to recognize the
MIME format, and convert from a pure-ASCII representation to 8-bit
data. The MIME format clearly and unambiguously indicates which
character set the data are in (e.g., Latin-1) --- but the software
(another bit of it, perhaps) then turns around and pretends that the
data are really in another character set, presumably the standard
Taiwanese multi-byte character set.)

> I've recently begun noticing similar problems with postings in German.

It's unavoidable that when the technology exists to use national
characters, they will be used. The dominant mail platform in the
Internet today is a browser on Windows, where users don't even know
which characters are ASCII and which aren't --- as witnessed by many
posts using the double left and right quote characters (not even part
of Latin-1).

Unicode is on its way in too --- if Microsoft Internet Explorer 5 does
not already by default send mail that is unreadable in a non-MIME
reader, I'm sure that will happen in the Windows 2000 version.

So, it may be possible to make this list into a backwater where people
learn TeX codes and refrain from using accented characters out of
politeness. But it's only a question of time before people have to
upgrade anyway, if only to be able to read mail from aunt Agatha.

Lars Mathiesen (U of Copenhagen CS Dep) <thorinn at diku.dk> (Humour NOT marked)

[ Moderator's comment:
  The mail system on which this list is operated is 7-bit ASCII only, and will
  remain so for the foreseeable future.  I'm sorry, but the lowest common
  denominator is preferable to Windows-only capabilities.  As I have noted
  elsewhere, not even other 8-bit-capable systems agree with Windows encodings.
  --rma ]


From cjustus at mail.utexas.edu  Wed Jul  7 15:32:06 1999
From: cjustus at mail.utexas.edu (Carol F. Justus)
Date: Wed, 7 Jul 1999 10:32:06 -0500
Subject: PIE vs. Proto-World
Message-ID: <WED.7.JUL.1999.103206.0500.>

August Schleicher composed the PIE Fable based on the hypothesis that
Sanskrit was closer to PIE. Hermann Hirt revised it updating with the
knowledge that Sanskrit 'a' came from PIE 'e' and 'o' among other things,
and Lehmann & Zgusta (in the Szemerenyi Fs. ca. 1980) further revised it
based on late 20th century views. I reviewed it in Language some time
thereafter suggesting that our even newer knowledge of the fact that *kwi-
was a focus marker might be integrated into yet another revision.

I'm not sure that these Fable writings reflect a confidence in our
knowledge of what PIE was like, but more likely a synthesis of the
hypothesis so far about the kind of details people usually feel more
comfortable commenting on.

I realize that opinions differ as to what one conceives the goal of
reconstruction to be. Statements like 'The reconstruction X has not been
proven', of course, fly in the face of the face of the assumption that
reconstruction is intended to reflect the hypothesis so far. Continued
revisions of the Fable, however, would seem to be a good example of the
fact that at least some IEists view reconstruction as the most plausibly
account of the known facts so far.

Carol Justus


From alderson at netcom.com  Wed Jul  7 16:26:40 1999
From: alderson at netcom.com (Richard M. Alderson III)
Date: Wed, 7 Jul 1999 09:26:40 -0700
Subject: PIE vs. Proto-World
In-Reply-To: <37790BA6.252E@mail.scu.edu.tw> (message from Steven Schaufele on
 Tue, 29 Jun 1999 11:08:38 -0700)
Message-ID: <WED.7.JUL.1999.092640.0700.>

Steven Schaufele wrote [inter alia]:

>Although i'm not prepared to go as far as Calvert Watkins (i think it was?)
>who composed a fable in PIE,

I'm not familiar with a Watkins fable, though it's possible.  Perhaps you're
thinking of Schleicher's fable (in his very Sanskrit-like PIE), which was
re-worked by Hirt (his version is filled with reduced vowels, but otherwise
looks very Neogrammarian), and most recently to my knowledge by Lehmann and
Zgusta (with Lehmann's versions of the laryngeals and of PIE syntax).

								Rich


From petegray at btinternet.com  Wed Jul  7 19:28:14 1999
From: petegray at btinternet.com (petegray)
Date: Wed, 7 Jul 1999 20:28:14 +0100
Subject: PIE vs. Proto-World
Message-ID: <WED.7.JUL.1999.202814.0100.>

Steven said:
>  Although i'm not prepared to [compose] a fable in PIE, i certainly do
> not doubt that, .. it could be done with our current state of
> knowledge.

Yes it has, quite recently.   A number of linguists were asked to give their
versions, and they were published together in a special volume.   I'm afraid
they prove you wrong - the diversity was enormous, with some linguists using
a more-or-less Brugmannian approach, and others being much more glottalic or
laryngeal.   The variety is great fun to see, and very educative!

Peter

[ Moderator's request:
  Would you mind providing bibliographical information for the volume in
  question?  I for one would be fascinated.
  --rma ]


From petegray at btinternet.com  Wed Jul  7 19:41:28 1999
From: petegray at btinternet.com (petegray)
Date: Wed, 7 Jul 1999 20:41:28 +0100
Subject: accusative and ergative languages
Message-ID: <WED.7.JUL.1999.204128.0100.>

Pat said:
> The
> idea that vocabulary is irretrievably lost is jejeune.

The idea that it is not lost is equally jejune (if I may use the English
spelling - I mean no offence).   One of the problems in PIE is identifying
which isolates are PIE and which are not,  or, where there are similar words
in two languages, which are inherited cognates, and which are loans (yes, I
know there are techniques for this - but there are still big problems).

Proto-proto- languages simply cannot be recovered with the same degree of
certainty as languages nearer to our attested texts.

Peter


From evenstar at mail.utexas.edu  Thu Jul  8 00:05:37 1999
From: evenstar at mail.utexas.edu (Shilpi Misty Bhadra)
Date: Wed, 7 Jul 1999 19:05:37 -0500
Subject: TeX conventions
In-Reply-To: <377909D2.1335@mail.scu.edu.tw>
Message-ID: <WED.7.JUL.1999.190537.0500.>

[ moderator snip ]

I am unfamilar to the TeX conventions. Would you please send me a list of
it. Thank you.

Shilpi Misty Bhadra
University of Texas at Austin
Classics/Ancient History/Humanities major
evenstar at mail.utexas.edu
512-474-2368

[ Moderator response:
  I will post a suggested style in a couple of days.  I'm trying to get the
  recent backlog under control.


From ECOLING at aol.com  Thu Jul  8 23:03:18 1999
From: ECOLING at aol.com (ECOLING at aol.com)
Date: Thu, 8 Jul 1999 19:03:18 EDT
Subject: 50% Span. or German, 50% Chinese
Message-ID: <THU.8.JUL.1999.190318.EDT.>

Steven Schaufele: English Department
Soochow University, Waishuanghsi Campus,
Taipei 11102, Taiwan, ROC

That is a very interesting problem you describe.
Unique I think, so the problem may lie in some quirk of
the software or mailers you use, or which your university uses.
It would be best I think if you directed your request to
the managers of those mailers or of that software.

I was accustomed to receiving unintelligible nonsense,
or at least material readable with difficulty,
but that seems to be decreasing, and I believe that
precisely the continued USE of the accented European vowels
will cause the makers of software to make the necessary
corrections to their software.
(Probably, in the Taiwan case, providing a mode to
select a segment of text, and interpret it as from a
different code standard than that containing Chinese characters
using the "upper ASCII" codes (128 to 255 more or less).

My recommendation is therefore just the opposite of yours.

Lloyd Anderson
Ecological Linguistics


From Anthony.Appleyard at umist.ac.uk  Fri Jul  9 07:13:49 1999
From: Anthony.Appleyard at umist.ac.uk (Anthony Appleyard)
Date: Fri, 9 Jul 1999 08:13:49 +0100
Subject: connected text in PIE; Proto-World language
Message-ID: <FRI.9.JUL.1999.081349.0100.>

[ moderator re-formatted ]

  Steven Schaufele <fcosw5 at mail.scu.edu.tw> wrote (Subject:      PIE vs.
Proto-World):-

> ... everything we have on PIE has to be starred because it's reconstructed,
> ergo hypothetical. ... Calvert Watkins (i think it was?) who composed a fable
> in PIE, i certainly do not doubt that, in principle, it could be done with
> our current state of knowledge.  Whereas I regard `Proto-World' as little
> more than an entertaining fantasy.

  (1) When neural-net computers get advanced enough for people to
simulate better than now human brain processes including language, it
would be interesting to simulate language evolution and see over how
many centuries and by what steps a language changes so much that all
trace of common ancestry vanishes behind the `noise' of accidental
resemblances.

  (2) What is the source of that fable in PIE? Has anyone made a good
collection of PIE grammar and accidence and vocabulary, enough to learn
the language from, to write connected text in it? I once heard of
someone writing a fable not in PIE but in `Aryan', i.e. the common
ancestor of Iranian and Sanskrit, with such special regional features as
turning PIE `e' and `o' into `a'. Who else has had a go at writing
connected text in PIE?

[ Moderator's response:
  August Schleicher.  Details have been noted in other answers to the original
  posting.
  --rma ]


From Anthony.Appleyard at umist.ac.uk  Fri Jul  9 07:28:38 1999
From: Anthony.Appleyard at umist.ac.uk (Anthony Appleyard)
Date: Fri, 9 Jul 1999 08:28:38 +0100
Subject: hand (was: Re: indoeuropean)
Message-ID: <FRI.9.JUL.1999.082838.0100.>

[ moderator re-formatted ]

Stefan Georg wrote:
> ...Other languages have replaced this apparently oldest word for "hand", to
> wit Latin manus, Gothic handus, Baltic and Slavic *renka/ronka (from a verb
> meaning "to grasp", cf. Lithuanian /rinkti/.  The limited distribution of
> this core-vocabulary item in IE has once given rise to the bromide that the
> early Indo-Europeans did >have feet but no hands (mocking at linguistic
> palaeontology, of course).

Compare the amount of modern English slang words for people's hands:
`grabs', `paws', `dooks', etc. Likely one relevant factor here is how
much a particular object or process is likely to accumulate slang names
for itself.


From pagos at bigfoot.com  Fri Jul  9 08:16:06 1999
From: pagos at bigfoot.com (pagos at bigfoot.com)
Date: Fri, 9 Jul 1999 10:16:06 +0200
Subject: Latin perfects and Fluent Etruscan in 30 days
In-Reply-To: <l03130309b39e90640b98@[209.147.211.55]>
Message-ID: <FRI.9.JUL.1999.101606.0200.>

At 10.13 29/06/99 -0500, Rick Mc Callister wrote:

>Mi suegro es agricultor en Costa Rica y llama "hijos" los reto?os
>V?stago, para m?, es la talla de cualquier planta en
>cualquier etapa de crecimiento.

Muy estimados Se?ores y Amigos!

With all respect, I have to inform you that the following dictionary may
prove helpful in your postings:

Guenther Haensch - Gisela Haberkamp de Anto'n: Woerterbuch der
Landwirtschaft (Dictionary of Agriculture; Dictionnaire Agricole;
Diccionario de Agricultura; Dizionario di Agricoltura;
Sel6skoxoziajstvennyij slovar6). BLV Verlagsgesellschaft. Muenchen - Wien -
Zuerich, 1987.

It contains all the specific agricultural technical terms you might ever
need in German, English, French, Spanish, Italian and Russian. No need to
cite more knowledgeable informants, i.e. your father-in-law, the very
father of your own sons aut similia.

In Spanish you talk of _planta madre_, exactly as in English you say
"parent plant", "mother plant", "stock plant" (German "Mutterpflanze",
French "plante m`ere" or "pied m`ere", etc.). Yet I never ran across a
_[planta] hija_ (f.) in Spanish.

If you really happened to hear the word "hijo", it might be due -- IMHO --
to an improper usage and/or a rare "modismo hispanoamericano".

Cheers

       Paolo Agostini


From pagos at bigfoot.com  Fri Jul  9 08:14:59 1999
From: pagos at bigfoot.com (pagos at bigfoot.com)
Date: Fri, 9 Jul 1999 10:14:59 +0200
Subject: TeX-encoding
In-Reply-To: <l03130309b39e90640b98@[209.147.211.55]>
Message-ID: <FRI.9.JUL.1999.101459.0200.>

>[ moderator re-encoded (experimental) ]
>hab{\'i}a, reto{\~n}os, v{\'a}stago, m{\'i}, ma{\'i}z

I use TeX, but I think that this way of encoding diacritics might not be
painless for those who don't. That's why I dare suggest to use a simpler
encoding as follows:

a' e' i' o' u'
a` e` i` o` u`
a^ e^ i^ o^ u^
a" e" i" o" u"
a~ e~ i~ o~ u~
n' n^ n~
s' s^

etc.

Best regards

            Paolo Agostini

[ Moderator's response:
  The problem with a simplified method is that it makes multiple diacritics
  difficult.  Bear with me; I will post a set of suggestions shortly.
  --rma ]


From jer at cphling.dk  Fri Jul  9 13:18:22 1999
From: jer at cphling.dk (Jens Elmegaard Rasmussen)
Date: Fri, 9 Jul 1999 15:18:22 +0200
Subject: accusative and ergative languages
In-Reply-To: <Pine.GSO.4.03.9906291119320.10157-100000@rsunx.crn.cogs.susx.ac.uk>
Message-ID: <FRI.9.JUL.1999.151822.0200.>

On Tue, 29 Jun 1999, Larry Trask wrote:

[Pat Ryan, quoting LT:]

>>> For Basque, and for other ergative languages, the "passive" view of
>>> transitive sentences can be shredded, point by devastating point.

>>> Mutila jo zuen.
>>> `He hit the boy.'

[LT:]

> In English, the utterance `He hit the boy' is *only* possible in a
> context in which `he' has already been identified: otherwise it's
> gibberish.

> And the same is true of Basque <Mutila jo zuen>: it is only possible in
> a context in which the identity of the hitter is already known, and
> otherwise it's gibberish.  In no context whatever could it be
> interpreted as `The boy was hit'.  There *must* be an identified hitter
> in the discourse.

> To express `The boy was hit', Basque uses other constructions.  One
> possibility is <Mutila jo zuten>.  This is literally `They hit the boy',
> and it can be used to mean this, when the identity of `they' is known.
> But equally it can mean `The boy was hit', in a context in which the
> identity of the hitters is unknown.  In this case, it is functionally,
> though not formally, identical to English `The boy was hit'.

> But Basque also has an overt passive: <Mutila jo zen>.  This means
> literally `The boy was hit', and it can be used with no hitter
> identified.  Moreover, this construction does not allow the addition of
> an overt agent: the Basque passive permits no agent.

Excuse my interruption, but I think your discussion is missing the only
point of any interest in this context: Can "Mutila jo zuen" not *come
from* something which *originally* meant, not 'the boy was hit' pure and
simple, but specifically 'the boy was hit by him (e.g., by the one we're
talking about)'? Nobody is claiming that the passive constructions out of
which the ergative grew in a number of languages were restricted to
impersonal use. Incidentally, if they were, there would not have been
anything to put in the case that subsequently got interpreted as a
"transitive-subject, i.e. ergative case".

I don't know a first thing about Basque, though I have been intrigued by
it on many occasions, especially since it offers such a good parallel to
Old Irish in the verb where you apparently have to memorize practically
all forms (which are many) to be able to say even the simplest of things -
and that of course was also what made me stop every time I got started.
>From the primitive and casual books at my disposal I do see that "zuen"
and "zuten" mean 'he had him' and 'they had him' resp. I also believe I
see that such auxiliaries are combined with a particularly short form of
the participles, referred to by Schuchardt as the root of the participle;
and "jo" is 'stick; beat' in its shortest form, says my little dictionary;
and "mutil-a" is 'boy' with the article "-a", but without case or number
marking. Therefore my persistent question: Why can't "mutila jo zuen" and
"mutila jo zuten" reflect a construction that was earlier meant to express
'the boy, he had him hit', 'the boy, they had him hit'? Schuchardt also
gives "zen" to mean 'he was', so that if you gloss "mutila jo zen" as 'the
boy was hit', it seems there is quite a bit of agreement that the verbal
root is a participle by itself. I do not see in what way this makes the
*diachronic* interpretation of "mutila jo zuen" any different from the
Hindi preterites that are based on Sanskrit constructions of the type
"A-Nominative + B-Genitive + PPP/nom." meaning earlier "A was (verb)-ed by
B", but now simply "B (verb)-ed A." Where am I wrong?

I am not saying you have not addressed that question properly elsewhere,
only we are some on this list who do not keep abreast of scholarly
discussion concerning Basque. And the question belongs here right now -
could we have your response to it?

Cheers,
Jens


From proto-language at email.msn.com  Fri Jul  9 13:34:46 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Fri, 9 Jul 1999 08:34:46 -0500
Subject: accusative and ergative languages
Message-ID: <FRI.9.JUL.1999.083446.0500.>

Dear Larry and IEists:

 ----- Original Message -----
From: Larry Trask <larryt at cogs.susx.ac.uk>
Sent: Tuesday, June 29, 1999 5:36 AM

[ moderator snip ]

[LT]

> OK, then -- try Japanese.  Japanese does not code subjects in the verb,
> and yet omission of the subject is perfectly normal in Japanese, but an
> object is still interpreted as an object.

[PR]

In my experience, many Japanese speaking English make "excessive" use of the
passive which reflects the fact that the subject is deleted in the Japanese
from which they are roughly mentally translating.

So, if it is the point about "Noun+acc. Verb" being "ungrammatical" in an
accusative language without an expressed Noun+nom. (including
cross-reference in the verb), I guess we should ask: "Is Japanese
accusative?"

Thirdly, an object in Japanese, is normally marked by [o] but. In the proper
context,

e{'}iga ga (subjective) suki{i} desu

will be translated as "I like movies" even though [ga] identifies a subject.

Now, with only a slight difference of nuance,

go{'}han o (objective) tabema{'}shita and

go{'}han wa (topical) tabema{'}shita

will be translated as: "I have eaten dinner".

I really do think that Japanese is not the best language to illustrate your
point.

<snip>

 [PR]

>> In the sentence mentioned above:  "noun(B)+abs. verb", which is
>> interpreted as an 'activity is performed by an unspecified agent on
>> B' --- this construction perfectly meets the definition: "a
>> construction in which an intrinsically transitive verb is construed
>> in such a way that its underlying object appears as its surface
>> subject"; accordingly, it is "passive".

[LT]

> No, not so.  See below.

[PR]
That definition is from your own dictionary.

> [LT]

>>> For Basque, and for other ergative languages, the "passive" view of
>>> transitive sentences can be shredded, point by devastating point.

> [PR]

>> Perhaps we should reopen the question of where you have "shredded,
>> point by devastating point" the view that "for Basque(, and for
>> other ergative languages,) the "passive" view of transitive
>> sentences". I saw nothing that I recognized as doing this in your
>> Basque grammar.

[LT]

> That's probably because I haven't written a Basque grammar.
> I have, however, written elsewhere on this point.

[PR]

Well, _The History of Basque_ will generally pass for a grammar in my
opinion. Well, where did you do the "shredding"? In a Basque cookbook
(:-{})?

[LT]

> [on my Basque example]

>>> Mutila jo zuen.
>>> `He hit the boy.'

> [PR]

>> Keine Endung ist auch eine Endung. Surely you must have run across
>> that someplace. And your translation of "mutila jo zuen" as '(he)
>> hit the boy' is not preferable to 'the boy was hit'.

[LT]

> Sorry, not so -- not so at all.

> In English, the utterance `He hit the boy' is *only* possible in a
> context in which `he' has already been identified: otherwise it's
> gibberish.

[PR]

Rather overly broad! I have heard conversations that go along the lines of
'He hit the boy' with the adressee responding, 'Who hit him, John or Phil?'
Unless you consider clarification a gibberish-process.

[LT]

> And the same is true of Basque <Mutila jo zuen>: it is only possible in
> a context in which the identity of the hitter is already known, and
> otherwise it's gibberish.  In no context whatever could it be
> interpreted as `The boy was hit'.  There *must* be an identified hitter
> in the discourse.

[PR]

Also not: 'The boy was hit (by someone known)'?

[LT]

> To express `The boy was hit', Basque uses other constructions.  One
> possibility is <Mutila jo zuten>.  This is literally `They hit the boy',
> and it can be used to mean this, when the identity of `they' is known.
> But equally it can mean `The boy was hit', in a context in which the
> identity of the hitters is unknown.  In this case, it is functionally,
> though not formally, identical to English `The boy was hit'.

[PR]

Well, obviously, according to you, Basque makes a fine distinction in
definiteness between singular and plural constructions.

[LT]

> But Basque also has an overt passive: <Mutila jo zen>.  This means
> literally `The boy was hit', and it can be used with no hitter
> identified.  Moreover, this construction does not allow the addition of
> an overt agent: the Basque passive permits no agent.

[PR]

I suspect but cannot prove that there is *also* a modal nunace here.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From petegray at btinternet.com  Fri Jul  9 19:39:08 1999
From: petegray at btinternet.com (petegray)
Date: Fri, 9 Jul 1999 20:39:08 +0100
Subject: indoeuropean/hand
Message-ID: <FRI.9.JUL.1999.203908.0100.>

> It's an interesting metaphor
> I was also interested in why the "masculine" form manus and not *mana?

The form is only apparent.   It is actually a -u stem, whereas the
masculines that you are thinking of are thematic (-o/e).   The -u stems can
be any gender, although masculines do predominate.   There are also a few -
a very few -  feminine thematic stems.

Peter


From jer at cphling.dk  Fri Jul  9 22:00:54 1999
From: jer at cphling.dk (Jens Elmegaard Rasmussen)
Date: Sat, 10 Jul 1999 00:00:54 +0200
Subject: accusative and ergative languages
In-Reply-To: <Pine.GSO.4.03.9906291609240.10157-100000@rsunx.crn.cogs.susx.ac.uk>
Message-ID: <SAT.10.JUL.1999.000054.0200.>

On Tue, 29 Jun 1999, Larry Trask wrote:

[...]

> There are two separate issues here, the synchronic one and the
> diachronic one.

> (1) Is a particular ergative construction "really" passive in nature?

That was not meant to be my question.

> (2) Does a particular ergative construction descend by reanalysis from
> an earlier passive?

That was what I was driving at, albeit only as a possibility, not as a
pervasive solution of all ergative.

> For the diachronic question (2), the answer "yes" has been defended in
> some particular cases, including Indo-Iranian.  But a passive origin for
> the Indo-Iranian ergative constructions has also been disputed, and I am
> not aware that there exists a consensus among specialists.

> Since Indo-Iranian is a rare case in which we have millennia of
> documentation of the intervening stages, and since the right answer to
> the question is still not obvious, then it must be very much harder to
> answer the question in respect of other cases for which little history
> is available.  At present, there appear to be few cases in which the
> origin of an ergative construction is fully understood and beyond
> controversy.
[...]

I'd say the cases I have seen of ergativity in Indic and Iranian languages
so clearly reflect underlying/earlier passive circumlocutions that
controversy is absurd. Yet, in that field controversy is to be expected
over anything.

I grant you (quoting Dixon) that Hittite is not this way: making neuters
animate when they are subjects of transitive verbs is indeed not the
reflex of a passive transformation. I have no knowledge of the Amazon
language Pari, let alone of its history. But note that I never claimed
that ergative always comes from passive, only that there are very clear
examples that it sometimes does - at least, that was what I meant.

Jens


From PIE at AN3039.spb.edu  Fri Jul  9 23:15:17 1999
From: PIE at AN3039.spb.edu (PIE at AN3039.spb.edu)
Date: Sat, 10 Jul 1999 02:15:17 +0300
Subject: accusative and ergative languages
In-Reply-To: <Pine.GSO.4.03.9906291609240.10157-100000@rsunx.crn.cogs.susx.ac>; from "Larry Trask" at Jun 29,
 99 6:49
Message-ID: <SAT.10.JUL.1999.021517.0300.>

Tue, 29 Jun 99 18:49 +0300 MSK Larry Trask wrote to Indo-European at xkl.com:

> But Dixon also cites examples of ergative constructions which, in his
> view, have very clearly *not* developed from passives, but from other
> constructions.  His examples are Hittite and Pari.

Pardon my ignorance, but i wonder what Hittite constructions can be
claimed as ergative? Could you clarify Dixon's opinion?

Does he mean the tendency to avoid using inanimates as 1st actants of
two-argument predicate and to transform them to animates in -nt-, like

    eshanants inan karapzi,
    witenants eshar parkunut,    etc?

_________
Alex


From jer at cphling.dk  Fri Jul  9 23:02:04 1999
From: jer at cphling.dk (Jens Elmegaard Rasmussen)
Date: Sat, 10 Jul 1999 01:02:04 +0200
Subject: Momentary-Durative
In-Reply-To: <006701bec26f$07f7eec0$b1fdabc3@xpoxkjlf>
Message-ID: <SAT.10.JUL.1999.010204.0200.>

On Tue, 29 Jun 1999, petegray wrote:

> Jens said:

>>    I wonder how else anybody would understand these data - except by
>> ignoring their being just that.

> Alas, Jens, I do not see data, but hypothesis and ideology.   The data is
> that some IE langs use nasal presents for a particular root where others do
> not, so your evidence from Latin cannot count for much in Sanskrit.    The
> data is that root aorists are associated with a variety of present
> formations, including nasal infixes.

I resent the use of the word "ideology"; is the common occurrence of a
nasal-infix structure from *tewd- 'thrust' in Skt. tundate and Lat. tundo
not part of our "data"? Do I have to add that the impression that the
Indo-European languages are related and descended from a common source is
not based on ideology, but on inspection and common-sense interpretation
of the data. And it takes outright rejection of the relationship between
the IE languages to make evidence from one IE language irrelevant in
another. There are so many instances of the _same_ verbal root turning up
with a nasal present in IE branches that have nothing else in common than
the mere fact that they are accepted as Indo-European. Is it a coincidence
that *k^lew- 'hear' forms a nasal present in Indo-Iranian and Celtic?
*telH2- 'endure', in Tocharian, Armenian, Italic and Celtic? *g^neH3-
'know', in Indo-Iranian and Germanic? This could go on - as Strunk's book
(and indeed already Brugmann's Grundriss) does. Does that not indicate
that the assigment of the nasal-infix structure as the present of certain
roots was fully lexicalized in the protolanguage?

> The hypothesis which has become an ideology, is that all root aorists must
> have had a nasal present.

Oh no, it works only in the direction that a nasal present has a root
aorist beside it in practically all cases. Also the reduplicated presents
and the y-presents generally form root aorists. I suppose there were some
old distinctions that only made sense with durative actions. However, by
PIE, each verb had mostly picked one form for its present. The massive
concord among the IE languages in this respect can be explained in no
other way that I can think of - while the even greater discord is easily
explained by continued normal language change.

For example, you said:

>> Strunk has shown that nasal
>> presents go with root aorists, thus we would like to derive tuda'ti from
>> an original root aorist if that is in any way possible.

> I see no reason to, since I do not share your ideology.   You may or may not
> be right - the important thing is that what you offer is not data, and so
> you should not insult those of us who do not agree with it.

Hey, hey, how can it be an insult to anybody that I try to account for the
greates amount of correspondences between the IE languages with the least
amount of force? I do not see you offering any understanding of the data
you apparently do not want me to talk about; I point out that they fit a
well-argued theory that is already on the market, and I add that the same
theory also accomodates the data you did choose to talk about; if that
is offensive, I'm afraid I may soon be doing it again.

> You said:

>> Likewise we would like to have a root aorist beside the nasal present
>> vinda'ti ... and so we
>> have a strong motivation to derive the thematic aorist a'vidat from a root
>> aorist.

> Traditionally, these are taken from different roots.

"Different roots"? But there is only one underived verbal paradigm, and
that is as trivial as nasal-infix present + root aorist if *(e-)wid-e/o-
is explained in exactly the same way as the Greek thematic reflexes of IE
root aorists (the e-lipon type). That there is a perfect beside it,
is normal; that it means 'know' rather than 'have caught sight of' is a
case-story of the most usual kind. The stem *wid-eH1- 'be looking' is a
derivative stem which can in principle be made from all verbs. There is no
need to project the Indic distinction between vid- 'know' and vid- 'find'
back into the protolangue, the functional range is delivered free of
charge by the large IE verbal system already.

Jens


From gordonselway at gn.apc.org  Fri Jul  9 23:57:40 1999
From: gordonselway at gn.apc.org (Gordon Selway)
Date: Sat, 10 Jul 1999 00:57:40 +0100
Subject: role of verb endings
Message-ID: <SAT.10.JUL.1999.005740.0100.>

At 16:25:19 -0500 on Tue, 29 Jun 1999, Patrick C. Ryan
<proto-language at email.msn.com>

>Larry indicated that verbal inflections should not be considered an
>expression of the subject in languages like Latin.

>I consider this position unjustifiable.

Why?

Gordon Selway
<gordonselway at gn.apc.org>


From gordonselway at gn.apc.org  Fri Jul  9 23:57:42 1999
From: gordonselway at gn.apc.org (Gordon Selway)
Date: Sat, 10 Jul 1999 00:57:42 +0100
Subject: indoeuropean/hand
Message-ID: <SAT.10.JUL.1999.005742.0100.>

At 10:15 am 29/6/1999, Rick Mc Callister wrote:

>It's an interesting metaphor

>I was also interested in why the "masculine" form manus and not *mana?

Various answers, or points of departure:

a.  it's a u- stem, not an o- stem
b.  there are several other grammatically f u- stems (domus, quercus &c)
c.  common enough with o- stems in Greek (hodos, nesos ktl)

Gordon
<gordonselway at gn.apc.org>


From proto-language at email.msn.com  Sat Jul 10 01:47:55 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Fri, 9 Jul 1999 20:47:55 -0500
Subject: accusative and ergative languages
Message-ID: <FRI.9.JUL.1999.204755.0500.>

Dear Peter and IEists:
 ----- Original Message -----
From: petegray <petegray at btinternet.com>
Sent: Tuesday, June 29, 1999 3:07 PM

<snip>

Peter:

> IE roots of the kind CRC had the ablauts CeRC, CoRC, CR.C.   The first
> appears in the Germanic present, the second in the past singular, the third
> in the past plural and the past participle (remember that PIE /o/ appears as
> /a/ in Germanic).  After standardisation of the vowel of the past, we find
> in modern German:
>     werfen    warf    geworfen

Pat:

Yes, but if a new verb were formed, e.g. webben (from English 'web'), there
is no possibility that it would be conjugated webben, wabb, gewobben.

What I consider 'internal inflection'  is Arabic yaktubu, kataba, katibun,
etc. which applies to any and all verbs, old, and those taken new into the
language.

 Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From proto-language at email.msn.com  Sat Jul 10 02:19:26 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Fri, 9 Jul 1999 21:19:26 -0500
Subject: accusative and ergative languages
Message-ID: <FRI.9.JUL.1999.211926.0500.>

Dear Wolfgang and IEists:

 ----- Original Message -----
From: Wolfgang Schulze <W.Schulze at lrz.uni-muenchen.de>
Sent: Wednesday, June 30, 1999 9:44 AM

W:

> Let me briefly comment upon some of the recent arguments:

> "Patrick C. Ryan" wrote:

>> A sequence like: "Me hurt. Tommy did it" is a virtual ergative so far as I
>> would judge. I have heard children speak in this way.

> This surely is not ERG-like! It's present tense, we ought to have "me hurts
> - Tommy does it" (or so) which shows that "me" still is in the ACC case
> (without AGR on the verb - "-s" has a "dummy agent" as a trigger. The
> same is true with structures like German "mich friert" etc...

Pat:

In spite of the grammar correction, what I have heard is "Me hurt". I assume
from this and other examples I have heard that 'me', regardless of its
function in standard English, is being used absolutively; and 'hurt', in
this phrase, I would think of more as a stative rather than transitive form.

Pat, previously:

>> (3)
>> However, in Language A,
>>         noun(B)+abs. verb
>> will be interpreted as an activity is performed by an unspecified agent--

Wolfgang:

> This is true for only some ERG systems. In such cases, the verb very
> often gets a plural morphology to install AGR with a "hidden" agent. In
> many other ERG systems, the omission of the "agent" leads to new
> intransitive structures with the inference ABS > AGENT (cf. I-ERG
> boy-ABS bring > boy-ABS come etc.).

Pat:

I have attempted to point out the salience of dividing these sentences into
two-element and three-element divisions rather than transitive and
intransitive categories. Obviously, you do not agree. To my analysis,
'boy-ABS. bring' is a transitive sentence with the object unspecified.
'boy-ABS come' is, unless we regard verbs of motion as reflexives (moves
himself to) as I would prefer, are, for most, intransitive --- perhaps
because many people regard a characteristic of transitivity as affecting in
some real way the properties of an object.

[ Moderator's comment:
  In an ergative system, a sentence "boy-ABS bring" would be transitive with an
  upspecified *subject*, as absolute marking on "boy" would require it to be
  the object.
  --rma ]

Wolfgang:

> Discussing the possible PASS background of ERG structures, Jens finally
> asks:

>> A truly innocent question for information: Are there other avenues that
>> are _known_ to have led to the creation of an ergative than the one
>> starting from an old passive?

> Sure, there are plenty such avenues! Most of them have to do with the
> Silverstein Hierarchy (or its expansion). The less a potential agent in
> (stronger) transitive scenarios is thought to bear inherent agentive
> features, the more it becomes likely that this "light agent" is marked
> by something that strengthens its agentive role. Some options are:
> (alienable) genitives (which can be extended to "heavy agents" on an
> inalienable basis), locatives (esp. with very light (or secondarily
> lightened) agents or weather phenomena etc...), instrumentals that are
> metaphorized to "agent" markers in the context of anthropomorphization),
> true "agentive" markers that are grammaticalized from e.g. a deicitic
> source, topic markers... Another possibility (though related to the
> strategies mentioned so far) is the reanalysis of 'active' structures in
> an ergative perspective (note that I do not want to suggest any 'active'
> typology for IE here, see my earlier postings): the S-split would then be
> harmonized on an S=0 level. All these strategies are based on the
> semantic or functional grading of the agentive (in terms of the
> manipulation of "lightness" and "heaviness"). Naturally, ERG techniques
> can also evolve from a special 'treatment' of the objective: One of the
> most prominent one is that of syntactic and/or pragmatic foregrounding
> which means that O is syntactically referred to as an intransitive S.
> Such a technique may be equivalent to passive strategies, however, this
> is only ONE of the many possible inferences.
> The syntactic/lexicical interface is touched upon when causatives of
> intransitives form the basis for newly established ERG features: Here,
> the morphosyntax of the causer can be introduced in the paradigm of
> other 'true' agentives via analogy.
> Finally, agreement strategies may play an important role in the game.
> If, for instance, agreement is coupled with some kind of person
> hierarchy, the presense of any SAP in a clause may condition agreement
> irresepctive their functional or semantic role. Hence, a scheme nSAP:A >
> SAP:O would necessariliy produce an erg-like AGR pattern (in case AGR
> becomes active), whereas SAP:A >nSAP:O would produce ACC-AGR. ERG AGR
> patterns may also result from the reinterpretion of clausal layers, e.g.
> the structure SV // AOV could be read as SV // A[OV] which means that O
> becomes some kind of closer attribute to the (participle-like) verb...

Pat:

All this theorization is undoubtedly interesting but, IMHO, does not answer
Jens' question. As for lessened degrees of animacy, most ergative languages
have antipassives to indicate this.

Although Dixon is certainly a man who has devoted much thought to
ergativity, I find something inherently problematical in combining ergative
and accusative features in one sentence (a little schzophrenic) which he is
forced to do by analyzing pronominal and nominal structures differently when
they occur in the same sentence.

I think it is likelier that, because of perceived greater animacy (or
definiteness), pronouns have a different method of marking that can still be
interpreted within an ergative context.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From JoatSimeon at aol.com  Sat Jul 10 03:12:19 1999
From: JoatSimeon at aol.com (JoatSimeon at aol.com)
Date: Fri, 9 Jul 1999 23:12:19 EDT
Subject: `cognate'
Message-ID: <FRI.9.JUL.1999.231219.EDT.>

>larryt at cogs.susx.ac.uk writes:

>This posting will not be news to the linguists on this list, who may prefer
>to stop reading at this point, unless they want to offer criticisms of the
>definitions below.  But I have noticed that a significant number of
>non-linguists on this list have apparently misunderstood the sense of our
>technical term `cognate': many of them
>appear to believe that `cognates' means something like `words of similar
>form

- that's odd; I always assumed it meant "derived from a common ancestral
word".  You know, like Tiwaz and dyaus.


From proto-language at email.msn.com  Sat Jul 10 14:24:11 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Sat, 10 Jul 1999 09:24:11 -0500
Subject: accusative and ergative languages
Message-ID: <SAT.10.JUL.1999.092411.0500.>

Dear Ralf-Stefan and IEists:

 ----- Original Message -----
From: Ralf-Stefan Georg <Georg at home.ivm.de>
Sent: Tuesday, June 29, 1999 7:37 AM

Pat wrote:

>> "Overt" means 'open and observable', and the overt subject of the Spanish
>> phrase is "he/she" indicated by -0 on [ha]; of the Latin phrase, 'he/she'
>> indicated by -(i)t. I would think you might have understood that I was
>> referring to languages which do not code the subject with affixes on the
>> verb.

R-S responded:

> No, it was not clear that you were referring to this kind of languages.
> Let's have a look at one of those, though:

> KhalkhaMongolian (no subject affixes on the verb):

> Nom chamd yavuulav.

>  (He, she, it, someone, nobody, I, we, you and whatnot) -"book (indefinite
> acc.)", "to-you", "sent". (Someone) sent you the book. "Someone" is not the
> translation, but the dummy for every agent you wish and which can be made
> clear by the context; it is overtly expressed by nuffin'. The
> extra-syntactic context, that is. A perfectly normal sentence in context.
> It is true that Khalkha does use the personal pronouns in examples like
> this to disambiguate, but it doesn't have to. The construction is certainly
> not ungrammatical.

Pat responds:

I do not readily have access to material on Khalkha Mongolian so I cannot
comment in detail on your example.

However, in general, I would hazard the guess that, regardless of its
apparently accusative construction, for all practical purposes, this
sentence could be just as easily translated: 'A book was sent to you'.  I am
also a little sceptical of an "accusative" in such a context which is marked
by -0.

But, having said that, I am willing to concede that for *some* accusative
languages, this construction, provided a nominative subject can be mentally
supplied from context, is grammatical.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From rmccalli at sunmuw1.MUW.Edu  Sat Jul 10 21:42:54 1999
From: rmccalli at sunmuw1.MUW.Edu (Rick Mc Callister)
Date: Sat, 10 Jul 1999 16:42:54 -0500
Subject: Chronology of the breakup of Common Romance [long]
In-Reply-To: <4835aed4.24b300a5@aol.com>
Message-ID: <SAT.10.JUL.1999.164254.0500.>

[snip]

>"Vlach" or "Wallach" seems to be descended from a name that was commonly
>applied to Celts.

	It has long been proposed that the name Volcae, a Gaulish tribe,
was the source of Anglo-Saxon Wealh, Balkans Vlakh, Polish WLochy, etc.
	The point I've generally seen is something to the extent that
supposedly the Volcae were the first Gaulish tribe to fall under Roman rule.
	Some popular writers have also proposed Volcae as the source of
English "folk" but I've never seen this proposal from professional
linguists.

>Appearing as "Walh" or "Walah" in OHG, it has interpreted
>as meaning "foreigner", sometimes Roman, but in usage it is closely
>associated with Celts and regions of Celtic habitation - e.g., "Wales",
>"Walloon".  I seem to recall an suggestion that there may be a tie here to L.
>vallum (fortified wall, the earliest meaning of "wall") and refer to the
>Celtic or Romano-Celtic oppidum or walled town.

	You'd have to account for the ending in Vlakh, WLochy, etc.

>The strong Celtic presence along the Danube is attested by Classical writers
>well before the common era.  E.g., Alexander fights them before turning
>against the Persians and meets Celts who are from Illyria.  Galicia in
>southern Poland is a region whose name remembers a Celtic presence even
>farther north.  I remember an old passage where Vlachs are identified as "the
>shepherds of the Romans" and in this role they may also have been imported
>help as they were in northern Italy.

	Isn't the term Vlakh anachronistic in this sense? Vlakh is first
documented in the middle ages, isn't it?

	[snip]

>The ancient ethnic designations in that part of the world are a little
>difficult to follow,  but it seems clear that the whole region from Illyria
>to present day Romania was under Rome by 250 ace.  And there is some
>possibility that Vlachs represented Romanized Celts across those regions.  In
>any case, the small difference in time between Rome's entry into Illyria and
>into Dacia would be a de minimis factor.

	My guess is that any Celtic element in the Balkan would have been
extremely thin.
	The Celts raided the Balkans and stormed through the region on the
way to Galatia but my understanding is that they lived north of the Balkans
in Galatia and Pannonia

>In a message dated 6/20/99 5:55:33 PM, rmccalli at sunmuw1.MUW.Edu wrote:

><<Rumanian does share a common pre-Romance substrate with Albanian
>which presumibly should help pinpoint the origin of Rumanian.>>

>The suggested Thraco-Illyrian-Albanian substrate has always been called
>tentative, especially in light of potential borrowing across short distances
>where there would have been much opportunity for contact, particularly prior
>to the Slavic invasions of the Balkans.

>This brings up that familiar problem of finding a Celtic substrate, one that
>might be fruitful.

>E.g., one example of the Albanian substrate often given (per G. Mallinson)
>has been Rom. "abure"/ Alb. "avull" (steam.)  But in Gaelic I see "co-bur"
>(foam), "to-bar" (well) and "bruich" (boil) - add the Gaelic "a-" (out of) to
>get /a-bruich (out of boiling- steam?) and the Celtic seems as possible as
>the Albanian as a source.

	Following Larry Trask's post that "boil" words tend to be
onomatopoeic --largely based on "b-l, f-r" [e.g. "boil, bubble", fervire",
etc.] I'd be real careful with this one

>Another substrate example has been Rom. "vatra"/
>Alb. "vatre" (hearth), but perhaps forms like the Gaelic "fadadh" (kindling)
>and "bradhadair" (blazing fire) offers evidence of a common origin and
>original meaning for these later similarities.

>I haven't found any recent consideration of the Celtic remnants in this area,
>but I would think it might offer some real possibilities.

>In a message dated 6/20/99 5:55:33 PM, rmccalli at sunmuw1.MUW.Edu wrote:

[snip]

><<I seem to remember reading that while Albanian has a strong substrate, that
>it is significantly lacking in Ancient or Classical Greek substrate. If
>this is true, then it tells us where not to look...>>

>But the problem is that Rumanian does have a substantial Greek substrate (and
>had a larger one centuries ago before Rumanian was "purified" in recent
>times.)

	As I understand, its substrate is from Byzantine and Modern Greek
but according to what I remember reading, it has little or no substrate
from Classical Greek.

[snip]

>Furthermore, given the extreme amount of intercourse between Greek and Thrace
>during those many early centuries, wouldn't one expect Thracian to have a
>strong and ancient Greek substrate, especially with regard to trade items and
>such?  If Albanian doesn't have such substrate, it may not be Thracian.

	Thracian was not necessarily a single entity. Some dialects may
have been influenced by Greek while others may have not. But I defer to
those who actually know something about Thracian.

	Here's what I have in my notes on the subject. My apologies for
spacing.

	Graham Mallinson, "Rumanian," 391-419. Martin Harris and Nigel
Vincent, eds. The Romance Languages. London: Croom Helm, 1988.

	There is very little substrate vocabulary in Rumanian, and its
origin is unknown. Since the grouping of Dacian, Thracian and Illyrian
within IE is uncertain, it is questionable whether Albanian and Rumanian
shared a substrate language, although they do have cognates:

[ Moderator's note:
  The following 8-bit entities are translated to TeX notation at the end of
  the message.
  --rma ]

	ALBANIAN			RUMANIAN
	avull "steam"			abure
	m?s "colt"			m?nz
	p?rrua "brook"		p?r?u
	shkrump "ash"			scrum
	vatr? "hearth"		vatra [413]

L. R. Palmer, The Latin Language London: Faber, 1954.

	The Illyrians left geonyms in Apulia and Calabria:
Brundisium < Brendou elaphon "Stag's horn," see Albanian bri,ni "horn,"
Salapia and Salapitani < Selepitani < sal "salt," + ap "water,"
Otranto < Odruntum contains Messapic odra "water." [39]
	Illyrian and/or Messapic words in Latin include
blatea < Illyrian balta "swamp,"
deda "nurse,"
par? "small ship,"
gandeia "gondola,"
h?reia "small fishing boat,"
mannus < Illyrian manda- "pony," see Messapic Menzana, to whom horses were
sacrificed. [40]
Lama "swamp," occurs as an element of Illyrian geonyms in Italy. [41]

A. Rosetti. Br?ve histoire de la langue roumaine des origines ? nos jours.
The Hague: Mouton, 1971.

	The Thracians, known as G?tes by Greek writers and Dacians by Latin
writers, left geonyms ending in -dava, -upa, -sara, -para, -bria. [51]

	Illyrian words included
Bardylis (proper name) > Albanian i bardh? "white,"
dranis (Greek elaphos) > Gheg dreni, Tosk dr?ri "stag,"
Dalmatia, Delminium (probably Bosnian Duvno) > Albanian dele, delme "schaf,"
Dardani > Albanian dardh? "pear, pear tree, compare Illyrian dard? "pear,"
Messapian Menzana "Jupiter," > Albanian m?s, m?z-i "mule," Daco-Romanian
m?nz "poulain." [52]

	Thracian words include
br?a "city," vs. Tokharian ri, riye "city,"
deva, dava "city, village," e.g. Pupuldeva > Bulgarian Plovdiv;
bur-, bour- > Albanian burr? "male"
Dr?nis (proper name) > Albanian dr?, dr?ni "stag," zeiz, zis, ziz >
Albanian i zi, e z?z? "black,"
mante?a, mantua, mantia "Brombeere, Brombeerebaum (blackberry) > Albanian
man, mand "Maulbeerbaum, Maulbeere,"
xarp?ies (Carpathians) < carpa > Albanian karp? "rock," xarpoi (Tribe from
north of Thrace) > Albanian karp? "Fels,"
xorp?loi (Tribe from south of Thrace),
Maluensis, Maluese, Maluntum (geonyms) > Albanian mal "mountain, mountain
range, Daco-Romanian mal "escarpment, mountain, riverbank, seashore,"
pinon, pinos > Albanian pi "drink," Albanian pir? "drunk,"
Dacian sxi?re "chardon" > Albanian shqer "reisse auseinander, zerreiser,
kratze." [52]

	The Daco-Romanian suffix -esc > Thracian -isk. [53]
	Certain words are the same or cognate in Albanian and Daco-Romanian:

		Albanian			Romanian
	buz? "lip, edge"		buz? "lip, edge"
	k?lbaz?, g?lbaz?		c?lbeaz?, g?lbeaz?
	(sheep illness)			(clavel?e)
	k?push? "tick"			c?pus,? "tick"
	avull "Dunst"			abur "steam"
	bres, brez "G?rtel"		br?u "waist, belt"
	ragal "hat"			argea "vo?te d'une cave"
	ba? "fromager"			baci "fromager"
	boll? "dragon, monster"	balaur "dragon, monster"
	baig? "Veternarian"		baleg? "fiente"
	balt?				balt? "lagoon, pond, lake"
	i bardh? "white"		barz? "stork"
	bask? "Fliess"			basc? "toison, laine"
	pellg "Weiher, Regenpf?tze"	b?lc "flaque, mare, bourbier"
	berr "Schaf, Weidevieh"		b?r (sheep call)
	bredh "Tanne"			brad "sapin"
	byk "p?glia"			buc "balle du bl?"
	bukur? "pretty"			bucurie "joy"
	bungu "Eiche"			bunget "thick part of forest"
	k?sul? cap, stocking cap	c?ciul? "bonnet de fourrure"
	katund "village"		c?tun "hamlet"
	qaf? "Kehle, Hals"		ceaf? "nuque"
	?ok "Fussfessel, Schnabel"	cioc "beak, point, prow"
	sorr? "Kr?he"			cioc? "summit"
	?ufk? "Quaste, Maishaar"	ciuf "tignasse, huppe"
	thump "Stachel"			ciump "bout, chicot, moignon"
	?upis "picke"			ciupi "picorer"
	shut "ohne h?rner"		ciut "?corn?"
	koq? "berry"			coac?z? "berry"
	kopil "young man"		copil "child"
	kulp?r? "cl?matite sauvage"	curpen "cl?matite des Alpes"
	kurth? "Falle, Fangeisen"	curs? "trappe, pi?ge, emb?che"
	droe "fruit"			droiae "band, troop, many"
	dru, drut? "wood, stick"	druete "short thick piece of wood"
	th?rrim? "crumb, debris"	f?r?m? "debris"
	gardh "Hecke, Zaun"		gard "enclosure"
	gat "bereit"			gata "near"
	gj?mle "thorn"			ghimpe "thorn, needle"
	gjon "Nacheteule"		ghionoaie "pivert"
	gjysh "grandfather"		ghuij "old"
	gog? "Gespenst, Wauwau"		gog? "mitten"
	grep "Haken"			grap? "herse"
	g?rres? "Schabholz, Schabeisen" gresie "gr?s, pierre a aiguiser"
	gurmaz "Kehle"			grumaz "neck, throat"
	grund? "Kleie"			grunz "grumeau, pelote"
	grop? "Grube, Graben, Grab"	graop? "tomb, ditch"
	gush? "Hals, cock's dewlap"	gus,? "throat, goiter"
	ham?s "scarfer"			hames, "avid"
	gjym?s "half"			jum?tate "half"
	i leht? "leicht" 		lete "slowly"
	hudh?r "Knoblauch"		leurd? "wild garlic"
	magul? "hill"			m?gur? "hill, height"
	mal "mountain (range)"		mal "escarpment, edge, shore"
	i math, i mall "great, big"	mare "great, big"
	maraj "fennel"			m?rar "fennel"
	modhull? "peas"			maz?re "peas"
	mosh? "Greis, Alter"		mos, "old (man), grandfather"
	mugull "Pfropfreis, Spross"	mugur "bourgeon"
	murg "dark, black, gray"	murg "brown"
	mushk "Maulesel"		mus,coi "mulet"
	n?p?rke "viper"			n?p?rc? "viper"
	uj?,"water" ujane "ocean"	noian "ocean, vastness"
	p?rroa "riverbed, brook"	p?r?u "torrent"
	pup?z? "bird" 			pup? "Wiedehopf", pupaz? "huppe"
	r?nd?s "Lab"			r?nz? "g?sier, cow's stomach"
	thab?t "sour"			sarb?t "sour, curdled"
	shkrep "schlage Funken, Feuer"	sc?p?ra "to spark, to kindle"
	shkrumb "alles Verkohlte"	scrum "ash, cinder, coal"
	thumb?z "Knopf"			s?mbure "nugget, nougat, seed"
	shpend?r "hellebore"		sp?nz "hellebore"
	shpuz? "glowing coal"		spuz? "hot coal"
	shtrep "worm"			strepede "worm"
	shtrung? Abteilung des Pferchs" "strung? "defile"
	shapi "eidechse"		s,op?rl? "lizard"
	cjap "Ziegenbock"		t,ap "buck, stag"
	thark "pen, corral"		t,arc "pen, menagerie"
	thep "point, sharp rock"	t,eap? "point"
	udhos "cheese"			urd? "cottage cheese"
	vot?r, vatr? "fireplace"	vatr? "fireplace"
	vjedull "burrowing mammal"	viezure "blaireau"
	dhall? "sour milk"		zar? "curdled milk"
	shkardh? "Hundekette"		zgard? "dog collar" [58-62]

John A. C. Greppin Grolier Electronic Encyclopedia

Phrygian
	Phrygian, a language of west central Anatolia, had two literary
periods, Old Phrygian (730-430 BC) & New Phrygian (AD 100-350). The later
stage used a Greek-like script; the earlier had an eclectic alphabet based
on Northwest Semitic models. Though of Indo-European origin, Phrygian is
poorly understood, especially in the writing of the earlier period. It
seems to be more closely related to Greek, but it also shows certain
affinities with Armenian. An older theory, no longer tenable, related
Phrygian to Thracian, & posited a Thraco-Phrygian language family.

Thracian
	Thracian was spoken in what is now Bulgaria & parts of Greece &
Turkey. No significant inscriptions exist but numerous words are known from
Greek & Roman texts. A large number of personal & place names have also
been recorded. Thracian is Indo-European, but its affinities to any
language other than Dacian & perhaps Phrygian are vague.

Dacian
	Also referred to as Getic, Dacian was spoken in what is now
Romania. Like Thracian, it is known both from words mentioned in Greek &
Latin texts, & also from proper names. It has recently been shown that
Dacian became distinct from Thracian, but this took place probably after
1500 BC. Some scholars believe that a Dacian layer underlies Albanian, &
that perhaps Dacian, rather than Illyrian, was the original form of that
language.

Illyrian
	Illyrian was spoken north & west of Greece during the Greco-Roman
period. Scholars are unsure whether the term refers to just one language or
many. Most evidence for Illyrian comes from proper names; the core of the
material is now called Messapic. Illyrian has been considered an ancient
form of Albanian, but this view is losing favor.

Rick Mc Callister
W-1634
Mississippi University for Women
Columbus MS 39701

[ Moderator re-encoding (experimental) ]

	ALBANIAN			RUMANIAN
	avull "steam"			abure
	m{\"e}s "colt"			m{\^i}nz
	p{\"e}rrua "brook"		p{\^i}r{\^i}u
	shkrump "ash"			scrum
	vatr{\"e} "hearth"		vatra [413]

L. R. Palmer, The Latin Language London: Faber, 1954.

	The Illyrians left geonyms in Apulia and Calabria:
Brundisium < Brendou elaphon "Stag's horn," see Albanian bri,ni "horn,"
Salapia and Salapitani < Selepitani < sal "salt," + ap "water,"
Otranto < Odruntum contains Messapic odra "water." [39]
	Illyrian and/or Messapic words in Latin include
blatea < Illyrian balta "swamp,"
deda "nurse,"
par{\'o} "small ship,"
gandeia "gondola,"
h{\'o}reia "small fishing boat,"
mannus < Illyrian manda- "pony," see Messapic Menzana, to whom horses were
sacrificed. [40]
Lama "swamp," occurs as an element of Illyrian geonyms in Italy. [41]

A. Rosetti. Br{\`e}ve histoire de la langue roumaine des origines {\`a} nos
jours.  The Hague: Mouton, 1971.

	The Thracians, known as G{\`e}tes by Greek writers and Dacians by Latin
writers, left geonyms ending in -dava, -upa, -sara, -para, -bria. [51]

	Illyrian words included
Bardylis (proper name) > Albanian i bardh{\"e} "white,"
dranis (Greek elaphos) > Gheg dreni, Tosk dr{\"e}ri "stag,"
Dalmatia, Delminium (probably Bosnian Duvno) > Albanian dele, delme "schaf,"
Dardani > Albanian dardh{\"e} "pear, pear tree, compare Illyrian dard{\'a}
"pear," Messapian Menzana "Jupiter," > Albanian m{\"e}s, m{\"e}z-i "mule,"
Daco-Romanian m{\^i}nz "poulain." [52]

	Thracian words include
br{\'i}a "city," vs. Tokharian ri, riye "city,"
deva, dava "city, village," e.g. Pupuldeva > Bulgarian Plovdiv;
bur-, bour- > Albanian burr{\"e} "male"
Dr{\'e}nis (proper name) > Albanian dr{\^e}, dr{\^e}ni "stag," zeiz, zis, ziz >
Albanian i zi, e z{\"e}z{\"e} "black,"
mante{\'i}a, mantua, mantia "Brombeere, Brombeerebaum (blackberry) > Albanian
man, mand "Maulbeerbaum, Maulbeere,"
xarp{\'a}ies (Carpathians) < carpa > Albanian karp{\"e} "rock," xarpoi (Tribe from
north of Thrace) > Albanian karp{\"e} "Fels,"
xorp{\'i}loi (Tribe from south of Thrace),
Maluensis, Maluese, Maluntum (geonyms) > Albanian mal "mountain, mountain
range, Daco-Romanian mal "escarpment, mountain, riverbank, seashore,"
pinon, pinos > Albanian pi "drink," Albanian pir{\"e} "drunk,"
Dacian sxi{\'a}re "chardon" > Albanian shqer "reisse auseinander, zerreiser,
kratze." [52]

	The Daco-Romanian suffix -esc > Thracian -isk. [53]
	Certain words are the same or cognate in Albanian and Daco-Romanian:

		Albanian			Romanian
	buz{\"e} "lip, edge"		  buz{\^a} "lip, edge"
	k{\"e}lbaz{\"e}, g{\"e}lbaz{\'e}  c{\^a}lbeaz{\^a}, g{\^a}lbeaz{\^a}
		(sheep illness)			(clavel{\'e}e)
	k{\"e}push{\"e} "tick"		  c{\^a}pus,{\^a} "tick"
	avull "Dunst"			  abur "steam"
	bres, brez "G{\"u}rtel"		  br{\^i}u "waist, belt"
	ragal "hat"			  argea "vo{\^u}te d'une cave"
	ba{\c c} "fromager"		  baci "fromager"
	boll{\"e} "dragon, monster"	  balaur "dragon, monster"
	baig{\"e} "Veternarian"		  baleg{\^a} "fiente"
	balt{\^e}			  balt{\^a} "lagoon, pond, lake"
	i bardh{\^e} "white"		  barz{\^a} "stork"
	bask{\"e} "Fliess"		  basc{\^a} "toison, laine"
	pellg "Weiher, Regenpf{\"u}tze"	  b{\^i}lc "flaque, mare, bourbier"
	berr "Schaf, Weidevieh"		  b{\^i}r (sheep call)
	bredh "Tanne"			  brad "sapin"
	byk "p{\`a}glia"		  buc "balle du bl{\'e}"
	bukur{\^e} "pretty"		  bucurie "joy"
	bungu "Eiche"			  bunget "thick part of forest"
	k{\^e}sul{\^e} cap, stocking cap  c{\^a}ciul{\^a} "bonnet de fourrure"
	katund "village"		  c{\^a}tun "hamlet"
	qaf{\"e} "Kehle, Hals"		  ceaf{\^a} "nuque"
	{\c c}ok "Fussfessel, Schnabel"	  cioc "beak, point, prow"
	sorr{\"e} "Kr{\"a}he"		  cioc{\^a} "summit"
	{\c c}ufk{\"e} "Quaste, Maishaar" ciuf "tignasse, huppe"
	thump "Stachel"			  ciump "bout, chicot, moignon"
	{\c c}upis "picke"		  ciupi "picorer"
	shut "ohne h{\"o}rner"		  ciut "{\'e}corn{\'e}"
	koq{\"e} "berry"		  coac{\^a}z{\^a} "berry"
	kopil "young man"		  copil "child"
	kulp{\"e}r{\"e} "cl{\'e}matite sauvage"
					  curpen "cl{\'e}matite des Alpes"
	kurth{\"e} "Falle, Fangeisen"	  curs{\^a} "trappe, pi{\`e}ge,
						emb{\^u}che"
	droe "fruit"			  droiae "band, troop, many"
	dru, drut{\"e} "wood, stick"	  druete "short thick piece of wood"
	th{\"e}rrim{\"e} "crumb, debris"  f{\^a}r{\^i}m{\^a} "debris"
	gardh "Hecke, Zaun"		  gard "enclosure"
	gat "bereit"			  gata "near"
	gj{\"e}mle "thorn"		  ghimpe "thorn, needle"
	gjon "Nacheteule"		  ghionoaie "pivert"
	gjysh "grandfather"		  ghuij "old"
	gog{\"e} "Gespenst, Wauwau"	  gog{\^a} "mitten"
	grep "Haken"			  grap{\^a} "herse"
	g{\"e}rres{\"e} "Schabholz, Schabeisen"
					  gresie "gr{\`e}s, pierre a aiguiser"
	gurmaz "Kehle"			  grumaz "neck, throat"
	grund{\"e} "Kleie"		  grunz "grumeau, pelote"
	grop{\"e} "Grube, Graben, Grab"	  graop{\^a} "tomb, ditch"
	gush{\"e} "Hals, cock's dewlap"	  gus,{\^a} "throat, goiter"
	ham{\"e}s "scarfer"		  hames, "avid"
	gjym{\"e}s "half"		  jum{\^a}tate "half"
	i leht{\"e} "leicht" 		  lete "slowly"
	hudh{\"e}r "Knoblauch"		  leurd{\^a} "wild garlic"
	magul{\"e} "hill"		  m{\^a}gur{\^a} "hill, height"
	mal "mountain (range)"		  mal "escarpment, edge, shore"
	i math, i mall "great, big"	  mare "great, big"
	maraj "fennel"			  m{\^a}rar "fennel"
	modhull{\"e} "peas"		  maz{\^a}re "peas"
	mosh{\"e} "Greis, Alter"	  mos, "old (man), grandfather"
	mugull "Pfropfreis, Spross"	  mugur "bourgeon"
	murg "dark, black, gray"	  murg "brown"
	mushk "Maulesel"		  mus,coi "mulet"
	n{\"e}p{\"e}rke "viper"		  n{\^a}p{\^i}rc{\^a} "viper"
	uj{\"u},"water" ujane "ocean"	  noian "ocean, vastness"
	p{\"e}rroa "riverbed, brook"	  p{\^a}r{\^a}u "torrent"
	pup{\"e}z{\"e} "bird" 		  pup{\"e} "Wiedehopf", pupaz{\^a}
						"huppe"
	r{\"e}nd{\"e}s "Lab"		  r{\^i}nz{\^a} "g{\'e}sier, cow's
						stomach"
	thab{\"e}t "sour"		  sarb{\^a}t "sour, curdled"
	shkrep "schlage Funken, Feuer"	  sc{\^a}p{\^a}ra "to spark, to kindle"
	shkrumb "alles Verkohlte"	  scrum "ash, cinder, coal"
	thumb{\"e}z "Knopf"		  s{\^i}mbure "nugget, nougat, seed"
	shpend{\"e}r "hellebore"	  sp{\^i}nz "hellebore"
	shpuz{\"e} "glowing coal"	  spuz{\^a} "hot coal"
	shtrep "worm"			  strepede "worm"
	shtrung{\"e} "Abteilung des Pferchs"
					  strung{\^a} "defile"
	shapi "eidechse"		  s,op{\^i}rl{\^a} "lizard"
	cjap "Ziegenbock"		  t,ap "buck, stag"
	thark "pen, corral"		  t,arc "pen, menagerie"
	thep "point, sharp rock"	  t,eap{\^a} "point"
	udhos "cheese"			  urd{\^a} "cottage cheese"
	vot{\"e}r, vatr{\"e} "fireplace"  vatr{\^a} "fireplace"
	vjedull "burrowing mammal"	  viezure "blaireau"
	dhall{\"e} "sour milk"		  zar{\^a} "curdled milk"
	shkardh{\"e} "Hundekette"	  zgard{\^a} "dog collar" [58-62]

[ End of moderator re-encoding (experimental) ]


From edsel at glo.be  Sun Jul 11 10:31:58 1999
From: edsel at glo.be (Eduard Selleslagh)
Date: Sun, 11 Jul 1999 12:31:58 +0200
Subject: Latin perfects and Fluent Etruscan in 30 days!
Message-ID: <SUN.11.JUL.1999.123158.0200.>

-----Original Message-----
From: petegray <petegray at btinternet.com>
Date: Friday, July 09, 1999 11:03 PM

>Ed connected Latin amicus and ambo, and suggested a derivation of them both
>from Etruscan.

>Max and I both pointed that ambo has a good PIE pedigree, so this can't
>work.

>Ed then said:

>>And I don't see why any relationship with 'amicus', 'amare', etc. should be
>>excluded a priori: ....  Couldn't it be a double
>>transfer: early Lat./IE 'amb(i)-' > Etr. 'am(e)-' > later Lat. 'am-'?

>Anything's possible!  But it's easier to take the word we know to be IE as
>IE, and leave open the possibility of an Etruscan origin for amicus, if you
>want to explore that.   Don't be mislead by the initial "am-".   In ambo it
>derives from syllabic /m./, which cannot be the case in amicus.

[Ed Selleslagh]

There seems to be a slight misunderstanding here: with "later Lat. 'am-'" I
meant the beginning of words like 'amare', 'amicus', etc. , i.e. without
the -b-, which - according to the reasoning above - would have transited via
Etruscan, as opposed to those that came straight from PIE and preserved the
(a)m(.)b-.

Apart from all that, I never suggested that 'ambo', ambi-', Grk. 'amphi'
(and Cat. 'amb', which proved to be unrelated) etc. were of Etruscan origin,
only that they should be added to the data pool when looking at possible
relationships of Etr. 'am(e)-' and Lat. 'amicus' etc. It seems to me that
this led to an interesting discussion that yielded the possibility of the
existance of two parallel paths: one 'directly' from PIE to Latin and one
via Etruscan.

Ed


From X99Lynx at aol.com  Sun Jul 11 17:58:46 1999
From: X99Lynx at aol.com (X99Lynx at aol.com)
Date: Sun, 11 Jul 1999 13:58:46 EDT
Subject: indoeuropean (correction)
Message-ID: <SUN.11.JUL.1999.135846.EDT.>

In a message dated 7/9/99 6:06:11 PM, I wrote:

<<Finally there is again the Greek 'keiro' (to cut, carve) that would explain
churgeon/surgeon much more adequately than a direct connection to /cheir/.
In Aristophanes The Birds, 'keirulos' describes a barber.

[ Moderator's note:
  These words obviously cannot be related.
  --rma ]>>

Inexcusibly, I suggested here that 'cheirourgia', 'cheirourgio^', etc., might
be directly from 'keiro' (cut) instead of 'cheir'/cheiroo' (hand).   Though
the ultimate connection between those two basic forms (in some way or other)
doesn't seem a bad idea to me, the source of 'cheirougia' in Classic Greek is
clearly 'cheir'.

The earliest appearances of 'cheirourgio^' apparently refer not to medicine
or surgery, but to handicraft and the skills of the artisan.  Plato uses it
in reference to musicians.  The later extension to a skilled surgeon would
only be natural.

Sorry for not going to the books first and I hope I'm correcting this piece
of misinformation here.

Regards,
Steve Long


From petegray at btinternet.com  Sun Jul 11 18:45:59 1999
From: petegray at btinternet.com (petegray)
Date: Sun, 11 Jul 1999 19:45:59 +0100
Subject: accusative and ergative languages
Message-ID: <SUN.11.JUL.1999.194559.0100.>

> Pat writes:

> Well, on page 40 of Chao's Mandarin Primer, are listed "Affixes":
> 11 are listed; ...

I simply cannot let this go.    Verbal compounds are rife in modern Chinese,
and the second elements ("affixes" if we are using terms inappropriate for
Chinese) are derived from a large number of concepts.   The fact that Chao
lists 11 only means that he has selected 11, and does not reflect the actual
nature of the Chinese language.   Things are not as simple as that grammar
makes them seem!

Peter


From CONNOLLY at LATTE.MEMPHIS.EDU  Sun Jul 11 20:49:29 1999
From: CONNOLLY at LATTE.MEMPHIS.EDU (CONNOLLY at LATTE.MEMPHIS.EDU)
Date: Sun, 11 Jul 1999 15:49:29 -0500
Subject: accusative and ergative languages
Message-ID: <SUN.11.JUL.1999.154929.0500.>

After some illuminating discussion of Germanic ablaut, petegray wrote:

>Some Germanic ablaut is indeed the result of "vowel harmony" with a vowel in
>an ending which has since been lost - but this example is not one.

Germanic ablaut *never* results from vowel harmony.  You're thinking of umlaut,
in which a stressed vowel is assimilated to the articulation of a following
segment: in the most common instance, back vowels before /i i: j/ are fronted
and become [-low]: PGmc. gasti:z > OHG gesti > German G"aste 'guests', PGmc.
*fo:dijan > OE f"odan > later OE fe:dan > modern feed.  Umlaut is a much later
process (it's an on-going process in Old High German) than ablaut, which was
conditioned in part by stress but not by vowel harmony.

Leo

Leo A. Connolly                         Foreign Languages & Literatures
connolly at latte.memphis.edu              University of Memphis


From CONNOLLY at LATTE.MEMPHIS.EDU  Sun Jul 11 21:00:54 1999
From: CONNOLLY at LATTE.MEMPHIS.EDU (CONNOLLY at LATTE.MEMPHIS.EDU)
Date: Sun, 11 Jul 1999 16:00:54 -0500
Subject: Ergative vs. accusative
Message-ID: <SUN.11.JUL.1999.160054.0500.>

Pat said, while discussing "ergative" languages:

>> [3]
>>> However, in Language A, noun(B)+abs. verb
>>> will be interpreted as an activity is performed by an unspecified
>>> agent on B

>> [4]
>>> whereas in Language B: noun(B)+acc. verb is *ungrammatical*.

Leo replied -- too obscurely, it would seem:

>> what you said about combining this verb with the patient alone simply isn't
>> true.  Some ergative languages happily omit the agent, others do not.  So
>> what I've numbered [3] above will be legal in some ergative languages but
>> not in others.

>> Conversely, [4] is perfectly grammatical in many accusative languages.
>> Couldn't think of an example good enough to convince you.  But look at
>> this post.  Must've seen stuff like this before, right?

Pat responded:

>Well, I agree with your first comment. But *most* ergative languages treat
>the agentive as a missible adverbial adjunct of the verb. Perhaps there may
>be a question of these languages being "truly" ergative?

>I am unaware of any accusative language in which this contsruction is
>grammatical. As you know by now, Larry indicated that verbal inflections
>should not be considered an expression of the subject in languages like
>Latin. I consider this position unjustifiable.

Should've looked a bit closer, Pat.  In discussing [4], I intentionally used
several sentences which lack overt subjects, as I just did again.  And the
subject is not marked on any of the verbs.  Are they grammatical?  Rather not
decide myself; let the rest of our merry group decide.

Leo

Leo A. Connolly                         Foreign Languages & Literatures
connolly at latte.memphis.edu              University of Memphis


From CONNOLLY at LATTE.MEMPHIS.EDU  Mon Jul 12 05:01:00 1999
From: CONNOLLY at LATTE.MEMPHIS.EDU (CONNOLLY at LATTE.MEMPHIS.EDU)
Date: Mon, 12 Jul 1999 00:01:00 -0500
Subject: Interpreting ergative sentences
Message-ID: <MON.12.JUL.1999.000100.0500.>

Dear Pat and everyone else,

I've been reading the recent (well, actually not so recent) discussion
of ergativity with more than a little interest, since I've been working
on the problem off and on for years -- strictly from the synchronic
point of view.  Please bear with me -- this will be long, since I have
to do some theoretical stuff to work through, some of which will be old
hat to many among us.

Much of the discussion has centered around the interpretation of
ergative sentences in Sumerian, of all things, which only Pat seems to
know at all.  (I sure don't.)  But I think it would be worth while to
consider various phenomena in modern languages, some of them ergative,
others not.

First, what is a "subject"?  Pat and the others have been talking past
each other on this, to a considerable extent.  The reason is simple:
there's great disagreement on what it means.

Dixon, in his work on ergativity, has consistently equated "subject"
with "morphological subject".  That is, if a language has a rule that
it must (usually) have one noun or pronoun which gets some specified
treatment not accorded to other items in the sentence, that item is the
morphological subject.  If there is case, it will be called
"nominative" in "accusative" languages, but (usually, but IMHO
unnecessarily) "absolutive" in "ergative" languages. -- Not all
languages have such a rule.  Those that do not are called "active"
languages, where casemarking (or equivalent) is not uniform for
one-argument sentences.

Pat definitely agrees with Dixon here, and so (but for very different
reasons) do I.  Most modern syntacticians do not: for them, the subject
is the noun phrase (NP) which, in an accusative language, would (a) be
the morphological subject and (b) have the *syntactic* privileges of a
subject.  That is, it would occupy "subject position", if the language
has such a thing, or it might be the only item to which a reflexive
pronoun might refer, or the only one which might be replaced by a
relative pronoun -- this is all highly language-specific, of course.

The catch is: for some verbs in (as far as I can tell) all languages,
and for many or most transitive verbs in ergative languages, the thing
that "ought" to be the subject might not be.  Then some funny things
can happen, which again are language specific.  But first: how can we
tell which NP "ought" to be the subject?

"Case Grammar" of the sort first proposed by Charles Fillmore is
helpful here.  It assumes that various NPs have semantico-syntactic
relations to verbs, which he foolishly called "deep cases".  There is a
hierarchy of these cases, which I believe is as follows:

	1.  Agent
	2.  Experiencer; beneficiary (including owner)
	3.  Instrument; effector
	4.  Patient (aka "theme")
	5.  Oblique cases (including place, time, goal, and a few
	    others)

In an accusative language, the NP which ranks highest in this hierarchy
is normally made the morphological subject.  In addition, it has the
syntactic privileges associated with it.  For convenience, I call this
one "HRNP", which stands for "highest-ranking NP".

But not all verbs in these languages actually "select" (to use
Fillmore's term) the normal subject.  Repeatedly we see that verbs with
experiencers or beneficiaries instead of agents may make the patient
the morphological subject.  It then often turns out that while the
morphogical subject (patient) has some syntactic privileges, the HRNP
has others.  Some examples:

English _own_ and _belong_ have Beneficiary and Patient as arguments.
_Own_ selects the Beneficiary as subject, but _belong_ has the Patient.
The result: sentences with _own_ can be made passive, but those with
_belong_ cannot.

	The Rockefellers own that railroad. ==>
	That railroad is owned by the Rockefellers.

	That railroad belongs to the Rockefellers.
	*The Rockefellers are belonged to by that railroad.

Why no passive for _belong_?  I suspect the reason is that
passivization serves to demote the "correct" subject (HRNP) and use the
"wrong" one instead -- but _belong_ already has the wrong subject!

German _helfen_ 'help' and _schmecken_ 'taste (good)' both have dative
objects.  Normally, German nominative nouns must precede dative nouns
when the NPs are contiguous:

	Will der Professor dem Studenten helfen?
	     Agent         Beneficiary
	'Does the professor wish to help the student?'

	*Will dem Studenten der Professor helfen?

But watch what happens with _schmecken_ and many other verbs that
select Patient as subject over the higher-ranking Experiencer:

	?Hat der Kaffee dem Professor geschmeckt?
	       Patient  Experiencer
	'Did the coffee taste good to the professor?'

	Hat dem Professor der Kaffee geschmeckt?
	'Did the professor like the coffee?'

The "normal" subject+object version of this sentence, while surely
acceptable, is simply not as good as the "abnormal" object+subject
version.  But why?   The reason is simply that the "abnormal" version
puts the HRNP ahead of the lower-ranking Patient-subject: the NP that
"ought" to be the subject is put where the subject "ought" to be!  And
it's no accident that the English gloss which puts _professor_ ahead of
_coffee_ sounds better than the other -- for exactly the same reason.

Accusative languages very greatly in which syntactic provileges are
attached to the morphological subject and which to the HRNP.  It may
well be true that in accusative languages, if any NP is marked on the
verb, the morphological subject must be; but otherwise it's a very
mixed bag.  In Icelandic, the HRNP seems to have *all* the syntactic
"subject" properties, regardless of what case it's in.  In English, the
morphological subject is king (except re. passivization).

A similar situation obtains in ergative constructions.  In Dyirbal, the
HRNP is has no syntactic privileges whatsoever; everything depends on
the morphological subject.  But most ergative languages do exactly the
opposite: the HRNP has all the syntactic properties, and in some it is
even the only one which controls verb agreement.  The Patient is then
still the "morphological subject" in that it's in the absolutive case,
but that's all.

That being said, we must ask what is the most appropriate way to
translate a "normal" transitive sentence in a "typical" ergative
language?  Larry Trask has already pointed out that Basque (like many
other ergative languages) has a passive formation.  But since the
Patient is already the morphological subject, what does the passive
accomplish?  It simply demotes the HRNP, so that it no longer has any
syntactic "subject" properties and must (in many other ergative
languages: may) be omitted.

Many also have an "antipassive" formation.  This makes the HRNP the
morphological subject, demoting the Patient.  The HRNP now has all the
syntactic subject properties, as it typically does in typical sentences
in accusative languages.  Different languages have different reasons
for doing this; Dyirbal uses it mainly as a technique for stringing
sentences together that can share a single absolutive.

A "normal" ergative sentence -- no passive or antipassive, and agent NP
present in the ergative case -- corresponds most closely to the English
active voice; the agent should then be the English subject.  A passive
translation would be inappropriate, since passivization downplays the
agent and makes it syntactically irrelevant, making the translation
quite different in tone (and content) than the original.  But a passive
sentence in an ergative language might best be translated with an
English passive.

This is actually quite like what often happens with experiencer and
beneficiary verbs in accusative languages.  The fact that e.g. Spanish
makes the patient the morphological and syntactic subject of a verb
does not mean that an English translator must do the same.  _No me
gusta la mu'sica_ does not mean 'the music doesn't please me' -- that
translation suggests that someone was trying to please (or perhaps
annoy) me with the music.  Rather, it means, very precisely, 'I don't
like the music': the Spanish doesn't insinuate that anyone was thinking
of me, so neither should the translation.

Similarly, I had no hesitation above in glossing _Hat dem Professor der
Kaffee geschmeckt?_ as 'Did the professor like the coffee?', even
though (a) the two languages have different morphological subjects, and
(b) _like_ is not an exact translation of _schmecken_, which always
means 'taste'. I also changed the tenses: present perfect in German,
past in English, since each is the normal tense for past time in the
respective language.  And this is legitimate, since in tone and meaning
it most closely approximates the German original.  Another example:

	Der Wagen ist nicht mehr zu reparieren.
	the car   is  not   more to repair

A good translation would be 'The car can't be fixed any more' -- or
'The car is beyond repair' -- or 'They can't fix the car this time'
etc.  They're all pretty good, but they're not literal renderings.

I could have said instead: 'The car is no longer to be repaired.'  That
seems closer -- except that I will have used a *passive* infinitive
instead of the German active.  And let's face it: it sounds stilted,
while the original doesn't.

I can take these "liberties" because German is a modern language that I
know well.  There are very few who could do this for a dead language,
even one as well known and thoroughly studied as Latin or Greek.  And
that's the trap: because we don't know the nuances, we want to be as
literal as possible, in the hope that a literal interpretation might be
correct.  Sumerian is even worse, since we don't have all that much of
it (no poetry, no menus, nothing that seems to be finely nuanced
speech.  And on top of that, it's ergative, and most of us (including
me) are *not* used to dealing with such languages.

So how should we render Sumerian sentences of the type

	verb = 'distributed'
	NP absolutive = 'camels'
	NP dative = 'heirs'

(Sorry I didn't cite the actual sentence, but this time it really is
the thought that counts!)  Not knowing Sumerian nuances, we can only
draw on seemingly similar sentences in other ergative languages.
Despite what Pat says, it is far from clear that the English must have
a passive of the type 'the camels were distributed to the heirs by some
unknown agent'.  That is a plausible attempt at a rendering -- but
since we do not know the effect of omitting the agent, and there seems
to be no hint of a passive transformation being done, there's no reason
to think it's "better" or "more precise" than any of the following:

	The heirs shared the camels.
	The heirs divided the camels aong them.
	The camels were shared by the heirs.
	"They" divided the camels among the heirs.
	All the heirs got some camels.
	The camels (jewels, books etc.) went to the heirs.

We don't know, because we don't know Sumerian the way we know English
and other modern languages.

So please, caution!  And take due note of what has been discovered
about *living* ergative languages.  You can't hope to understand the
dead ones without them.

Leo

Leo A. Connolly                         Foreign Languages & Literatures
connolly at latte.memphis.edu              University of Memphis


From Georg at home.ivm.de  Sat Jul 10 10:03:34 1999
From: Georg at home.ivm.de (Ralf-Stefan Georg)
Date: Sat, 10 Jul 1999 12:03:34 +0200
Subject: accusative and ergative languages
In-Reply-To: <002b01bec282$df10a6a0$dfd3fed0@patrickcryan>
Message-ID: <SAT.10.JUL.1999.120334.0200.>

>> Having to be rendered by the passive in English is not the same thing as
>> "being passive in nature".

>Pat responds:

>How about explaining "passive in nature"?  Is that a Platonic idea?

As you might have observed, I put double quotes around this expression, the
most widespread functions of which are:

- the expression is not mine, but used by others, and I use it only not to
complicate the discussion, assuming everyone knows what I'm talking about

or

- the expression is handy, but admittedly imprecise, I only use it since I
don't want to put too much effort in terminological precision at this
point, having different things to communicate (admittedly a vice, but
normally tolerable when speaking to attentive and knowledgable people)

I was reacting to some claim of yours, namely that ergatives are - now what
was your expression ? I don't recall, was it "really passives", "always
former. i.e. reanalyzed p.s", "p.s in nature", "p.s by nature". Hell, I
dunno. Something to this effect, at least. The important thing is that I
managed to point out that this, i.e. the ergative-as-passive-claim, however
phrased, is wrong, and Larry Trask, who has also a well-known publishing
record on the issue, has spoken the definite word about this here.
Pat, this is, alas, a familiar pattern: when one's theory gets into dire
straits, some aside discussion on irrelevant points is opened, obviously
with the intention to show weaknesses of whatever kind in the opponent's
standing, hoping that this will cloud the correct and justified remarks he
might have made on the relevant points before. This may work in some
election campaign, but not here, sorry.

To answer your question: "passives in nature", as used by me, is neither a
Platonic idea, nor a Wittgensteinean prototype, nor a correct description
of anything I may happen to think, it is just shorthand for a bundle of
concepts I don't share. And, above all, it is bad English. Should we open a
new thread about this, or shouldn't we better return to ergativity and your
misconceptions about it, such as "there are ergative languages without any
splits" athl. ?

Stefan

Stefan Georg
Heerstrasse 7
D-53111 Bonn
FRG
+49-228-69-13-32


From Georg at home.ivm.de  Sat Jul 10 10:27:00 1999
From: Georg at home.ivm.de (Ralf-Stefan Georg)
Date: Sat, 10 Jul 1999 12:27:00 +0200
Subject: accusative and ergative languages
In-Reply-To: <003f01bec285$ac207600$dfd3fed0@patrickcryan>
Message-ID: <SAT.10.JUL.1999.122700.0200.>

>Pat responds:

>Well, on page 40 of Chao's Mandarin Primer, are listed "Affixes":

>11 are listed; of these 6 qualify as related to "inflections" :
>modal -m(en), phrase marker -le; completed action -le; progressive
>action -j(y/e); possibility or ability -de; subordination -de.

>Undoubtedly, a historical grammar might provide a few more but I consider
>this a pretty simple system.

This is getting weary. I think no sane linguist will be unaware of the fact
that there are languages with fewer purely morphological means than others.
Also, "simpler" and "not-so simple" phonological systems have been heard of.
But, if I'm not completely mistaken, there was talk about sthl. like
general, overall simplicity of languages, which is a pre-scientific notion,
quite simply, if this pun is allowed.

If we assume, as of course we have to, that human languages are
problem-solving devices which face a set of possible communicative
problems, which is the same for all linguistic communities (don't come with
obvious cultural differences, which pertain to vocabulary only; I'm aware
of the fact that not all linguistic communities of the planet need to talk
meaningfully about the various kinds of fish in the Yenissey, or the
Sepik). Languages make different choices regarding the set of functions
they grammaticalize (tense vs. aspect or both, inferentiality, relative or
absolute tense, number, participant-identification, reference-tracking,
pragmatic categories, you name it), the hows and whys of which are the
object of linguistic typology. One language uses grammaticalized
bound-morphology for, say, inferentiality, others have to use different
means to convey the idea. But they are all able to convey the idea, however
elegant, or however clumsy. The idea is that they *use means* to do it, or
that speakers may find these means, even if the category is weakly or not
at all grammaticalized in their language (your inflection paradigm may be
my intonation pattern, your inferentiality affix may be my expletive
adverb; to convey the idea of the Dakota sentence-particle /yelo/ or Thai
/khrap/ I may even be forced to say each time "I am male"; clumsy, but
possible, if I face the necessity). The description of these means is the
task of the grammarian. Some grammarians are aware of this task, and write
a 600-page grammar of Chinese, some aren't (or simply want to produce a
primer to help you get along abroad) and write a 60-pp. treatment of
Russian. So what ?

I will not deny that notions of simplicity vs. complexity may be useful
distinctions when talking about subsystems. For the characterization of
whole languages they are definitely not useful. The grammar-school like
rote learning of paradigms, which your Chinese teacher will be able to
spare you, will be compensated by having to digest subtle and difficult
rules of syntax.

Stefan

Stefan Georg
Heerstrasse 7
D-53111 Bonn
FRG
+49-228-69-13-32


From adahyl at cphling.dk  Mon Jul 12 12:27:17 1999
From: adahyl at cphling.dk (Adam Hyllested)
Date: Mon, 12 Jul 1999 14:27:17 +0200
Subject: `cognate'
In-Reply-To: <Pine.GSO.4.03.9906301059430.22695-100000@rsunx.crn.cogs.susx.ac.uk>
Message-ID: <MON.12.JUL.1999.142717.0200.>

On Wed, 30 Jun 1999, Larry Trask wrote:

> I am still willing to accept suggestions for revisions, though I may
> be unable to consider large revisions.
> (...)
> `father', all of these being descended from PIE *<pater>.

I'm sure it says PIE *<p at ter> in your manuscript (+ diacritics on the
<e>).

Adam Hyllested


From vidynath at math.ohio-state.edu  Mon Jul 12 12:43:49 1999
From: vidynath at math.ohio-state.edu (Vidhyanath Rao)
Date: Mon, 12 Jul 1999 08:43:49 -0400
Subject: Momentary-Durative
Message-ID: <MON.12.JUL.1999.084349.0400.>

Jens Elmegaard Rasmussen <jer at cphling.dk> wrote:

> I guess the widespread iterative value of sk-formations has started with
> those that were reduplicated.

Just how widespread are the reduplicated forms with -sk-? It is found Greek.
Hittite has -sk- forms both with and without reduplication in many roots.
Others?


From vidynath at math.ohio-state.edu  Mon Jul 12 13:43:34 1999
From: vidynath at math.ohio-state.edu (Vidhyanath Rao)
Date: Mon, 12 Jul 1999 09:43:34 -0400
Subject: Momentary-Durative
Message-ID: <MON.12.JUL.1999.094334.0400.>

Due to the long break from reading this list, I am not sure if (and how
much) I am starting to repeat myself. Please forgive me if the two new posts
do contains little new information/question. It is just that I didn't see
them addressed.

Jens Elmegaard Rasmussen <jer at cphling.dk> wrote:

> I do not think the chain of reformulations is longer than many other
> stories generally accepted (or even as long as some known to be true). For
> the type _tuda'ti_, the key example itself has apparently replaced a nasal
> present still seen in tundate and Lat. tundo.

The tuda'ti type, with primary endings, is somewhat rare in RV. Perhaps,
tuda'ti type was a replacement for injunctives, i.e., tuda'ti was a general
present and not progressive present. Just a thought.

> Strunk has shown that nasal presents go with root aorists,

Reduplicated forms seem to go with root aorists as well. Anybody have info
on statistical comparisons between the two groups?

This also bears on the other proposal, going back to Kurylowicz at least,
for the origin of (asigmatic) aorist forms: namely, these are old preterits
that became limited to aorist function due to the rise of new presents.

Incidentally, what are your thoughts on the origin of nasal presents? The
rarity of infixes in PIE has led to the proposal that nasal presents
originated from a double affix, that is *wined- was really *wi-n-ed- (or
*wi-ne-d). This would mean that originally win(e)d- and w(e)id (or wide/o)
were not grammatically associated. In case of dehmi lehmi etc, it then
becomes a question of when the roots came into existence. I am not sure if
we have enough information to decide this.

This brings me to a general question. There seem to be two camps about the
category system of the PIE verb. One believes that the aorist-imperfect
distinction, to be equated to perfective-imperfective distinction,
``always'' existed in PIE and Hittie lost this distinction, while Vedic
changed things around. The other considers the aspectual distinction to
postdate the separation of Anatolian.

There are certain nagging questions about the first thesis: The change in
Vedic is not explained and how it came about without the prior loss of
aspect has, AFAIK, not been explained. Those who adhere to this also feel
the need to explain away as much as possible of root presents. But
there are enough of them remain in Hittite and Vedic to raise doubts.

Sigmatic aorists, as forming perfectives for root presents, present some
problems too. There are roots that look like older roots extended with an s.
This suggests that sigmatic aorists came from grammatization of a root
extention and fits in with the general opinion that sigmatic aorist as a
>grammatized< formation (as opposed to -s as derivational affix for forming
new verbs) is late PIE.

There is another question that typology suggests: There are basically two
forms of perfectives (Dahl, ``Tense and Aspect systems; Bybee and Dahl, in
Studies in Language, 13(1989) pp.51--103; Bybee et al ``The evolution of
grammar''): The first seems to come from old preterits that become limited
to perfectives due to the rise of new present/imperfective. The second is,
as in Russian, from ``bounders'', originally adverbs that denote the
attainment of a limit. In the first, aorist is unmarked and refers to action
considered as a whole, even if it is atelic. In the second, the perfective
does not fit that well with atelic events. The difference is said to be
evident in Bulgarian which has both the aorist-imperfect distinction (of the
first type) and perfective-imperfective distinction using prefixes (the
bounders).

Now, in PIE, some verbs seem to be like the first system (root aorist vs
marked presents) while we cannot escape the presence of the second type as
well (root present with sigmatic aorist). Sometimes Greek situation is used
to argue that aorist looks older due to the number of irregularities. But in
Sanskrit, as can be seen from looking at the lists in grammars, root
presents show the most irregularity while aorists (and class 3, 7
presents) are regular exemplars of internal sandhi however complicated the
rules may seem. Again we are faced with the question of who is more archaic
and who regularized.

One thing needs to be said about a different attempt to explain this, namely
the traditional (in IE studies) equation durative=atelic=imperfective,
momentary=perfective. Limiting perfectives to telic/momentary situations is
decidely the minority option among languages. In Dahl's survey it was found
only in Slavic, Finnish (based on acc vs partative), Japanese, Hindi,
Mandarin, Bandjalang and Cebuano. It is even absent in aorist-imperfect
distinction of Bulgarian. Positing this restriction for PIE requires
appropriately strong evidence, not just pointing to Russian.

The equation durative=telic is equally atypical. In such a language, one
could not just say ``John walked home'' or ``John ate one piece of bread''.
Instead one must say something like ``John walked, reached home'', or have
two verbs for every transitive durative, one used only for past with
definite objects, one used for past with indefinite/mass objects and for all
presents. Again, we cannot just assume such a property, especially given
that this does not seem all that common in contemporary languages.

>  for the functional change is quite small: it only takes the use
> of the aorist form as an imperfect, then the rest follows by itself.

The change of syntactic categories does not strike me as a small change. I
would like to know more about incontrovertable cases of such changes before
arriving at any conclusions.


From vidynath at math.ohio-state.edu  Mon Jul 12 13:40:14 1999
From: vidynath at math.ohio-state.edu (Vidhyanath Rao)
Date: Mon, 12 Jul 1999 09:40:14 -0400
Subject: Momentary-Durative
Message-ID: <MON.12.JUL.1999.094014.0400.>

petegray <petegray at btinternet.com> wrote:

> [Earlier, in the Vedic hymns,] The aorist can even
> be used where a present would be expected.

I suspect that augmented aorist was preferred for performatives. Another
thing which seems to occur a few times is that augmented aorist being used
for future perfect (where, in English, simple present can be used). [A
similar case occurs in Pali, where both pluperfect and future perfect are
expressed the same way.] But, an aorist for action in progress?
[Injunctives, being tenseless, do not count.]

> In the later language the perfect is simply a preterit or past tense
> equivalent with the imperfect and
> fully interchangeable, and sometimes co-ordinated with it.   Different
> authors appear to prefer different tenses.

This pays no attention to difference in genres and influence of Prakrit
syntax, as I pointed our earlier and which others have: for example Speyer,
``Sanskrit Syntax''.

> Older grammars (e.g. Whitney), modern grammars by Indian linguists (e.g.
> Misra) and modern lingistics books (e.g. Hewson & Bubenik "Tense and
> aspect in IE") all say the same kind of thing.

Unfortunately people sometimes just repeat what they have been told.

Incidentally, I would like to know what the list members think of Hewson and
Bubenik. I find the chapter on Sanskrit confused. Tables repeatedly list
`akarot' and `agacchat' as imperfective, but once admitting that it is the
regular tense of narration and is not really imperfective.

>> I don't see any parts of Sans lit in which aorist has resultative
>> meaning.

> An example from the RV (sorry I don't have the exact reference):
>   putrasya na:ma grhanti praja:m eva anu sam atanat.

> "He gives the son's name; and thus _he has extended_ his race."

There is no such sentence in RV. (it doesn't scan!) Perhaps you are thinking
of Maitra:yan.i Sam.hita (somewhere in bk 1) ``putrasya na:ma gr.hn.a:ti;
praja:m eva:nu sam atani:t''.

Let me first define what a resultative is to me: It is a form that conveys
that the result of the action indicated by the verb obtains at the time of
reference, irrespective of whether the action is recent or not, or if it can
be repeated or not, etc.

The trouble is that the differences between ``hot news'' forms and
resultatives (of recent events) are purely subjective and are not ameable to
derivation by decoding (that is the so-called scientific approach that
denying the understanding of native speakers a la Whitney). It is not easy
to come up with an example of hot news situation for which resultative is
out of the question. Only eamples I can come up with are momentary
light/noise production: for example, to say that ``the bell has rung'', I
can only say ``man.i at.t.ittu vi.t.tatu''. ``.. irukkiratu'' would be
interpreted with an inferential sense (due to a rule that if resultative and
experiential senses are ruled out, then present of iru- will be given an
inferential
meaning) [resultative sense is absent here because when I start talking, the
bell is not audible.] It is not surprising that such examples are hard to
come by in extant Vedic texts.

Even worse, in modern Western European languages, resultatives and passives
cannot be readily distinguished, and perfect and resultatives are also
similar. Thus we take ``John has eaten dinner'' to be resultative, but
``John has eaten ostrich meat'' as experiential perfect. ``The door is
closed'' can be resultative or passive, while ``The door is closed by John''
is passive; the resultative has be expressed as ``The door is closed because
of John'' or some such. We should not attribute labels to the constructions
of another language based on how we translate into the more familiar
languages. [I mention passives here because the question of possible passive
origin of ergatives discussed in a different thread.]  If we use
translations, we must use a language in which the distinction is obligatory.
So I would ask, for example, how to translate the above into Tamil:  Is it
``makanin peyaray upayokikra:n; (atana:l) kutumpat tod.arai ni:t.t.i
vid.ukira:n'' or ``... ni:t.t.iy iruppavana:y a:kira:n''? [But this seems to
disappear in post-Vedic times. It certainly cannot be made morphologically
in Pali or medieval Sanskrit.]  In Vedic the ta-adjective is resultative. It
is not clear that it is interchangeable with the aorist. This is often
claimed, but the evidence comes from much later form of Sanskrit and is not
applicable to Vedic. So the best we can say is that the evidence is murky.
But there are examples where the resultative sense is questionable if not
absent: Where aorist is used with jyok as in RV 10.124.1 ``jyog eva
di:rgham. tama
a:s'ayis.t.ha:h.'', or na + aorist, which denies occurance at any point in
time (``never have ...'') rather than just denying that the result obtains.

> I happily grant that there might be distinctions;  but if there are, they
> are subtle, and not present in all cases.    This does not weaken my
> argument that Sanskrit and Greek do not agree in the meaning and the
> function of these tenses, even if they do agree on the formation.

I agree with the second sentence and the first part of the first sentence.
But to check if a difference is present, we must look at the examples the
difference is palpable and not at the examples where pragmatics makes it
hard to decide. After all, we design experiments to remove the influence of
confounding variables. And the lack of agreement between different languages
should not stop us from looking for possbile common starting point(s) from
which the different systems may have evolved. Studies of grammaticization
that cover a wide spectrum (both areally and genetically) that are becoming
more popular should help here.


From ALDERSON at mathom.xkl.com  Thu Jul 15 01:36:36 1999
From: ALDERSON at mathom.xkl.com (Rich Alderson)
Date: Wed, 14 Jul 1999 18:36:36 -0700
Subject: [Meta-discussion] Backlogs and antique systems
Message-ID: <WED.14.JUL.1999.183636.0700.>

Dear Readers:

We are now caught up from the backlog caused by my failure to take into account
a limitation of the Tops-20 filesystem, the absolute maximum number of files in
a single directory.

There are subscribers on about 170 different systems worldwide; to make list
processing simpler, for each outgoing message I create a separate queue file
for each system in the database; if there are enough messages, I create more
queue files than the queue directory can hold.  This occurred late last week.

Because of the logjam thus created, I could not respond to some of the early
queries about the list; I apologize to those who may have felt slighted by the
lack of an answer.

And so it goes...

								Rich Alderson
								list owner and
								moderator

P. S.  If there are any responses to this message, I will post a summary or a
hand-created digest of any that seem to be of general interest.  --rma


From ALDERSON at mathom.xkl.com  Thu Jul 15 02:47:14 1999
From: ALDERSON at mathom.xkl.com (Rich Alderson)
Date: Wed, 14 Jul 1999 19:47:14 -0700
Subject: [Meta-discussion] TeX-style diacritics
Message-ID: <WED.14.JUL.1999.194714.0700.>

Dear Readers:

Recently, a reader noted that he was having trouble reading posts containing
8-bit (non-ASCII) characters in Spanish and German, because his mail system
interpreted them as the first part of a 16-bit Chinese encoding.  I have, as an
experiment in usability, re-coded the portions of some posts which contained
8-bit characters to use TeX-style diacritics; I hope those have been useful to
those who do not use Windows systems to read this list.

  (Aside:  I had to play games with a Windows system to create a cheat sheet
  for myself to do this.  As I have noted in the past, the Indo-European and
  Nostratic lists are maintained and processed on an XKL Toad-1 system running
  Tops-20, which is aggressively 7-bit-only, and it will be a long while before
  I can take the time to learn a different system to move the lists.  I'd like
  to spend that time on the lists themselves, of course.)

I have long advocated the use of TeX-style diacritics in e-mail and in Usenet
newsgroup postings, because both of these modes of information transfer are
significantly unfriendly to 8-bit character encodings in many systems; further,
even 8-bit-friendly systems disagree in the placement of accented characters
within the character set (Windows differs from Macintosh differs from various
flavours of Unix differ from Windows ...).

TeX is a text formatting system which, with its extensive facilities, allows
the creation of document formatting systems of great richness.  (I personally
think that all linguists should learn and use it, for its wonderful capacity to
use multiple character sets organized into easily-accessed fonts, but this is
not the forum for that discussion; I mention it only as background.)

In standard TeX, accents are created by the following sequences of characters:

	\'	acute accent
	\`	grave accent
	\^	"hat"-style circumflex accent
	\~	tilde or "swing"-style circumflex accent
	\"	umlaut/trema/diaeresis
	\=	macron/overline
	\.	superior dot
	\b	underline/bar-under
	\c	cedilla
	\d	inferior dot
	\u	breve accent
	\v	h{\'a}\v{c}ek
	\H	Hungarian long umlaut
	\t	superior tie

I usually enclose accented letters in braces {} to organize them.  In TeX,
spaces after any command (like \' or \v) are ignored, so {\v c} is one way to
write <c-with-hac^ek> (to use another mode of writing certain accents); this
can be hard to read, even when one is a TeX user, so it is preferable to use
the alternate I used above, \v{c}.  Thus, <c-with-cedilla> could be written
\c{c}.

NB:  This is standard TeX; I would actually prefer to use \, for the cedilla
accent, although it has a different meaning in standard TeX.  Since anything
in TeX is ultimately user-definable, we can adopt that as an alternate to \c .

Finally, in TeX superscripts and subscripts respectively are indicated by
^{super} and _{sub}, where braces are not needed for a single character in this
context:  *k{^w}is vs. *ekwos is unambiguous as to meaning--the former has a
labiovelar initial, the latter a medial cluster.  The laryngeals can be written
as *H_1, *H_2, *H_3, or \'x, x, x{^w} (as was Cowgill's wont), or @_1, @_2, @_3
(using the "ASCII IPA" symbol <@> for shwa), or some other mode if you prefer.

It was suggested, by a TeX user, that we use an abbreviated set of accents; in
many cases this is fine, but for full generality I would like to encourage the
readers to use the TeX notation when necessary.

I will collect the responses to this message and summarize them, or create a
digest by hand, rather than filling the list with further non-Indo-European
topics.  Simple replies will keep the "[Meta-discussion]" subject header and
make it easy for me to do so.

								Rich Alderson

References for those who would like to know more about TeX:

Lamport, L.  _LaTeX:  A Document Preparation System_, 2nd edition, (1994:
Addison-Wesley, ISBN 0-201-52983-1) describes the most commonly used variant
of TeX.

Snow, W.  _TeX for the Beginner_ (1992: Addison-Wesley, ISBN 0-201-54799-6) is
a hands-on tutorial which concentrates on "Plain TeX", with references to major
differences from LaTeX.

Kopka, H.  _LaTeX: Eine Einfuehrung_ (1994: Addison-Wesley) is a 3-volume work
on the basic LaTeX system and the large number of extension packages which have
been created for it.  I don't own a copy and can't get the ISBN.

Goossens, M., Mittelbach, F., and Samarin, A.  _The LaTeX Companion_ (1994:
Addison-Wesley, ISBN 0-201-54199-8) describes a large number of extensions to
the LaTeX system.

Knuth, D.  _The TeXbook_ (1986:  Addison-Wesley, ISBN 0-201-13447-0) is the
description of "plain TeX" by the gentleman who created it.  Difficult and
dense, but worth looking through for the erudite quotations at chapter ends,
and the subtle jokes throughout.
-------


From JoatSimeon at aol.com  Tue Jul 13 04:09:46 1999
From: JoatSimeon at aol.com (JoatSimeon at aol.com)
Date: Tue, 13 Jul 1999 00:09:46 EDT
Subject: PIE vs. Proto-World (Proto-Language)
Message-ID: <TUE.13.JUL.1999.000946.EDT.>

In a message dated 7/12/99 9:28:59 PM Mountain Daylight Time,
proto-language at email.msn.com writes:

>I simply do not understand why some find it difficult to understand that
>reconstructing the Proto-Language is only primarily different from
>reconstructing Indo-European in the wider selection of source languages for
>data.

-- Temporal distance.  Loss of information.  Entropy.

A distinction between 5,000 years of unrecoverable loss and 50,000 years of
unrecoverable loss.

Words _vanish_.  A certain percentage of vocabulary just ceases to exist in
every century.


From jer at cphling.dk  Tue Jul 13 11:39:25 1999
From: jer at cphling.dk (Jens Elmegaard Rasmussen)
Date: Tue, 13 Jul 1999 13:39:25 +0200
Subject: Recoverability
In-Reply-To: <d04dd9cf.24b435bb@aol.com>
Message-ID: <TUE.13.JUL.1999.133925.0200.>

On Wed, 7 Jul 1999 JoatSimeon at aol.com wrote:

> [...]
> If we
> didn't have written or other artificially preserved examples of now-extinct
> languages, much of the PIE vocabulary we have would be completely
> unrecoverable.

> Try reconstructing PIE using _nothing_ but contemporary Albanian and English.
> You'd be hard-put to prove that such a language even existed.

But that is not what you do in cases where only modern languages are
available: You do not pick two and ignore the rest, you take all there is.
In the case of IE, you would take _all_ modern Germanic languages, if need
be with their dialects; you take all Slavic languages, you don't forget
about Lithuanian and Latvian or Modern Greek which gives quite a lot; you
add four Celtic languages (Welsh, Breton, Irish and Scottish Gaelic with
due attention to their dialects); you get most of Latin by taking all of
the Romance languages and the many learned words of Latin origin preserved
in many languages (you get much of Ancient Greek the same way); you take
Albanian in at least two varieties; Armenian likewise; and you add the
total effort of comparative linguistics applied to the countless Modern
Indic and Modern Iranian (and Nuristani) languages. This is a LOT - and
NOBODY could even thgen be in doubt that this is indeed a family of
related languages sprung from a common source. True, we would still be in
doubt or indeed ignorant about many a finer point which is only added by
languages of old texts - and of course even they do not contain
everything, so doubt will remain - but the general utline and subgrouping
of the family would stand firm even on the sole basis of modern languages.

Jens


From proto-language at email.msn.com  Tue Jul 13 05:21:22 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Tue, 13 Jul 1999 00:21:22 -0500
Subject: accusative and ergative languages
Message-ID: <TUE.13.JUL.1999.002122.0500.>

[ moderator re-formatted ]

Dear Joat and IEists:

 ----- Original Message -----
From: <JoatSimeon at aol.com>
Sent: Tuesday, July 06, 1999 11:46 PM

>> proto-language at email.msn.com writes:

>> So what? It only requires us to expand our universe of applicable data.
>> The idea that vocabulary is irretrievably lost is jejeune.

> -- actually, it's obvious.  Information vanishes as entropy increases.  If we
> didn't have written or other artificially preserved examples of now-extinct
> languages, much of the PIE vocabulary we have would be completely
> unrecoverable.

> Try reconstructing PIE using _nothing_ but contemporary Albanian and
> English.  You'd be hard-put to prove that such a language even existed.

It is thuis type of illogic which stands in the way of real progress in
historical linguistics.

Larry Trask and others have done an excellent job, IMHO, of reconstructing
the language from which present Basque derives --- from only one language.

I also am certain that if we had reason to know that Albanian and English
were descendants of a common language, we would eventually discover a kind
of IE that would be inaccurate in some ways but essentially write.

Naturally, the more languages we added to the equation, the better
reconstruction of IE we would achieve.

Same principle.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From Georg at home.ivm.de  Tue Jul 13 08:55:38 1999
From: Georg at home.ivm.de (Ralf-Stefan Georg)
Date: Tue, 13 Jul 1999 10:55:38 +0200
Subject: connected text in PIE; Proto-World language
In-Reply-To: <3785A12D.9CF0DD1@umist.ac.uk>
Message-ID: <TUE.13.JUL.1999.105538.0200.>

>  (1) When neural-net computers get advanced enough for people to
>simulate better than now human brain processes including language, it
>would be interesting to simulate language evolution and see over how
>many centuries and by what steps a language changes so much that all
>trace of common ancestry vanishes behind the `noise' of accidental
>resemblances.

Sure we will learn a great deal from those neural-net computers. But, in
order to gain a better understanding of language change from them, we
should be able to bring them to interact socially with all that goes with
it.

Stefan Georg
Heerstrasse 7
D-53111 Bonn
FRG
+49-228-69-13-32


From larryt at cogs.susx.ac.uk  Tue Jul 13 10:24:45 1999
From: larryt at cogs.susx.ac.uk (Larry Trask)
Date: Tue, 13 Jul 1999 11:24:45 +0100
Subject: accusative and ergative languages
In-Reply-To: <Pine.SO4.4.05.9907091402440.15900-100000@rask>
Message-ID: <TUE.13.JUL.1999.112445.0100.>

On Fri, 9 Jul 1999, Jens Elmegaard Rasmussen wrote:

> Excuse my interruption, but I think your discussion is missing the
> only point of any interest in this context: Can "Mutila jo zuen" not
> *come from* something which *originally* meant, not 'the boy was
> hit' pure and simple, but specifically 'the boy was hit by him
> (e.g., by the one we're talking about)'? Nobody is claiming that the
> passive constructions out of which the ergative grew in a number of
> languages were restricted to impersonal use. Incidentally, if they
> were, there would not have been anything to put in the case that
> subsequently got interpreted as a "transitive-subject, i.e. ergative
> case".

Well, this suggestion is possible, I suppose, but I know of no evidence
to support it, while there is some evidence against it.  Let's look at
the structure of the Basque sentence.

Here <mutil> is `boy', and <-a> is the article.  This NP, being a direct
object, stands in the absolutive case, which has case-suffix zero.

The item <jo> is the perfective participle of the verb `hit'.

And <zuen> is the auxiliary verb-form.  Now, third person is usually
marked by zero in Basque verbs.  So, the usual absolutive agreement
slot, which is the first slot in the auxiliary, cannot be filled by an
agreement marker, and it is filled instead by <z->, a redundant marker
of past tense.  The next element, <-u->, is a reduced form of the verbal
root <-du->, from *<edun> `have'; this is the usual transitive
auxiliary.  Again, the ergative agreement slot, which follows the verbal
root, is empty, because the ergative (subject) NP is third-person.
Finally, <-en> marks past tense.

The whole thing is thus this:

	boy-the-Abs hit-Perf Past-have-0-Past

Or, roughly, `(She/he/it) had hit the boy.'

But it translates English `S/he hit the boy' (before today).  This is a
periphrastic form comparable to the ones we find in Romance and
Germanic.  Nobody knows if such periphrastic forms are calqued on
Romance or are an independent development in Basque.

Now, I can see little scope here either for an original passive
interpretation of the form or for any way of reading `by him' into the
auxiliary form.

> I don't know a first thing about Basque, though I have been
> intrigued by it on many occasions, especially since it offers such a
> good parallel to Old Irish in the verb where you apparently have to
> memorize practically all forms (which are many) to be able to say
> even the simplest of things - and that of course was also what made
> me stop every time I got started. >From the primitive and casual
> books at my disposal I do see that "zuen" and "zuten" mean 'he had
> him' and 'they had him' resp. I also believe I see that such
> auxiliaries are combined with a particularly short form of the
> participles, referred to by Schuchardt as the root of the
> participle;

It is simply the perfective participle of the verb.  Most participles in
Basque carry an overt suffix proclaiming them as such, but <jo> happens
to be one of the exceptions: it has no participial suffix, though I
suspect that it once did, and that the suffix has been lost by a
combination of phonological change and analogical readjustment.

> and "jo" is 'stick; beat' in its shortest form, says my little
> dictionary;

Actually, `hit', `strike', `beat' -- not `stick', which is a noun.

> and "mutil-a" is 'boy' with the article "-a", but without case or
> number marking.

Not quite.  The form <mutila> is marked as absolutive by its zero
suffix, and as singular by its singular article <-a>.  The plural
article is <-ak>, and `the boys' is <mutilak>, in the absolutive.

> Therefore my persistent question: Why can't "mutila jo zuen" and
> "mutila jo zuten" reflect a construction that was earlier meant to
> express 'the boy, he had him hit', 'the boy, they had him hit'?

Well, I can't rule that out, but I can't see any evidence to support it.
Note in particular that <mutila> exhibits *no* subject properties in
modern Basque, or in Basque of the historical period.  If it ever was a
subject, as this proposal requires, the reanalysis must have been
carried to completion a long time ago.

Note also that intransitive verbs are likewise conjugated
periphrastically but with the intransitive auxiliary <izan> `be'.

> Schuchardt also gives "zen" to mean 'he was', so that if you gloss
> "mutila jo zen" as 'the boy was hit', it seems there is quite a bit
> of agreement that the verbal root is a participle by itself.

The lexical verb stands in the form of its perfective participle in all
periphrastic past-tense forms, and also in all periphrastic perfects.
The perfect form corresponding to <Mutila jo zuen> is <Mutila jo du>,
which differs only in that the auxiliary is now present-tense.  This
form translates both English `He has hit him' and English `He hit him'
(earlier today) -- much as in European Spanish.

> I do not see in what way this makes the *diachronic* interpretation
> of "mutila jo zuen" any different from the Hindi preterites that are
> based on Sanskrit constructions of the type "A-Nominative +
> B-Genitive + PPP/nom." meaning earlier "A was (verb)-ed by B", but
> now simply "B (verb)-ed A." Where am I wrong?

Well, in the Indic case, we have several thousand years of texts to
consult, so that we can get an idea how the ergative construction arose.
With Basque, we are not so lucky.

Note also that, in Hindi, as in Indic generally (I think), the ergative
occurs only in the past tense, as is common with ergatives that have
arisen from perfective or passive constructions.  In Basque, however,
the ergative construction is used in all circumstances, without
exception.

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk


From stevegus at aye.net  Tue Jul 13 14:28:12 1999
From: stevegus at aye.net (Steven A. Gustafson)
Date: Tue, 13 Jul 1999 10:28:12 -0400
Subject: indoeuropean/hand
Message-ID: <TUE.13.JUL.1999.102812.0400.>

petegray wrote:

> The form is only apparent.   It is actually a -u stem, whereas the
> masculines that you are thinking of are thematic (-o/e).   The -u stems can
> be any gender, although masculines do predominate.

There are a fair number of Latin -u stems that are feminine, especially
considering how few -u stems there are to begin with: acus, domus,
cornus, idus, and tribus (as well as manus) are the ones that come to
mind.

--
Steven A. Gustafson, attorney at law
Fox & Cotner:  PHONE (812) 945 9600   FAX (812) 945 9615
http://www.foxcotner.com

Ecce domina quae fidet omnia micantia aurea esse, et scalam in
caelos emit.  Adveniente novit ipsa, etiamsi clausae sint portae
cauponum, propositum assequitur verbo.


From Georg at home.ivm.de  Tue Jul 13 12:20:10 1999
From: Georg at home.ivm.de (Ralf-Stefan Georg)
Date: Tue, 13 Jul 1999 14:20:10 +0200
Subject: accusative and ergative languages
In-Reply-To: <001801beca0f$f9125140$399ffad0@patrickcryan>
Message-ID: <TUE.13.JUL.1999.142010.0200.>

>[PR]

>Rather overly broad! I have heard conversations that go along the lines of
>'He hit the boy' with the adressee responding, 'Who hit him, John or Phil?'
>Unless you consider clarification a gibberish-process.

The explanation for this is, obviously, that speaker A had reasons to
believe (or was unattentive to reasons against this) that the reference was
clear enough for B from any kind of context (previous discourse, situation,
accompanying gesture, or whatnot). The gist is that, unless you want to be
deliberately opaque, you use anaphoric pronouns to refer to a referent
which is somehow identifyable by your addressee. You may, however, be wrong
about this, and then you may expect a clarifying question. The very fact
that this clarifying question makes up for a meaningful and
often-encountered micro-dialogue in your example bespeaks the
less-than-fully-acceptable nature of a communication strategy which opens a
conversation with anaphora, without making sure that everyone knows who you
refer to by "he".

And, by the way, this makes up for a crucial difference between person
marking on the verb and overt pronouns. Though both have comparable
functions (reference identification), the former is hardly ever explicitly
anaphoric. It would be interesting to watch out for a language with verbal
affixal person marking, which would only be used for anaphora, and
suppressed in other cases. I'm not aware of such a language, but I may be
missing the very obvious at the moment. Any ideas, someone ?

St.G.

Stefan Georg
Heerstrasse 7
D-53111 Bonn
FRG
+49-228-69-13-32


From ECOLING at aol.com  Tue Jul 13 15:51:00 1999
From: ECOLING at aol.com (ECOLING at aol.com)
Date: Tue, 13 Jul 1999 11:51:00 EDT
Subject: PIE vs. Proto-World (Proto-Language)
Message-ID: <TUE.13.JUL.1999.115100.EDT.>

Patrick Ryan writes:

>I simply do not understand why some find it difficult to understand that
>reconstructing the Proto-Language is only primarily different from
>reconstructing Indo-European in the wider selection of source languages
>for
>data. Of course, secondarily I have labored to reconstruct the underlying
>monosyllables by analysis of attested compounds.

>Pat

I strongly support work on earlier language families not part of the
standard doctrine today.
Against those who say (very nearly) that everything which
can be discovered has already been discovered.
That is of course a charicature, but with some truth to it.
(Or they say this while specifying "using the comparative method",
and defining it circularly to mean only the existing tools,
and only those ways of using those tools,
which are well known today.)

However, the above statement by Patrick Ryan I find highly surprising.

Of course it is going to get RELATIVELY more difficult
to reconstruct to greater time depths.
The mistake of those who reject all Nostratic and similar work
is in drawing a sharp fixed line,
saying that short of that time depth it can be done,
beyond that time depth it cannot be done.

The mistake of Patrick Ryan in the quotation above is to neglect
that it does get substantially more difficult as the time depths increase,
or in particular language families, because of the specific nature
of the sound-changes and grammatical changes which occurred,
or etc.

There is no sharp break.
There is a gradually increasing difficulty with time depth and
    with depth of changes (the two correlated but not the same concept).
There is ample room for expanding the tools used,
and for empirical studies of what sorts of changes each tool is
capable of penetrating beyond, and to what degree.

To give merely one example, one factor, sound symbolsim:

If there are sound-symbolic ideal forms for lexemes having
certain meanings,
then the more arbitrary ones (farther from those sound-symbolic
ideal forms)
have greater value as evidence for historical connections
of specific languages or language families,
simply because they are less likely to have resulted from
later pressure towards the sound-symbolic ideal.

This is of course a terribly difficult circularity,
because it means that look-alikes which are more widely attested
for a given meaning or set of related meanings
may be EITHER relics of an earlier historical unity
(whose changes were perhaps ALSO retarded by sound-symbolic forces),
OR the results of pressure towards some sound-symbolic ideal forms,
from diverse original and unrelated forms.
Much more subtle and difficult reasoning is therefore needed to
establish what are results of sound-symbolism and what are results
of historical common origins.
EVEN when we have a suprisingly widespread statistical sound-meaning
correlation.

The usual procedure is also circular.
Simply taking a sample of purportedly unrelated languages
and attempting to determine how many look-alikes word lists contain
is a bit naive,
because the languages may not be totally unrelated,
because chance resemblances may be more common in certain
meaning or sound ranges,
because the biases of different types of sound-system structures
are not yet well handled,
and for many other reasons.

Lloyd Anderson


From Georg at home.ivm.de  Tue Jul 13 12:24:17 1999
From: Georg at home.ivm.de (Ralf-Stefan Georg)
Date: Tue, 13 Jul 1999 14:24:17 +0200
Subject: accusative and ergative languages
In-Reply-To: <Pine.SO4.4.05.9907092339440.18724-100000@rask>
Message-ID: <TUE.13.JUL.1999.142417.0200.>

>I grant you (quoting Dixon) that Hittite is not this way: making neuters
>animate when they are subjects of transitive verbs is indeed not the
>reflex of a passive transformation. I have no knowledge of the Amazon
>language Pari, let alone of its history. But note that I never claimed
>that ergative always comes from passive, only that there are very clear
>examples that it sometimes does - at least, that was what I meant.

Sorry for playing the smartass, but P"ari is Nilotic, one of the celebrated
cases where ergativity has been found on the African continent. It is
particularly rare there.

St.G.

Stefan Georg
Heerstrasse 7
D-53111 Bonn
FRG
+49-228-69-13-32


From rmccalli at sunmuw1.MUW.Edu  Tue Jul 13 17:14:43 1999
From: rmccalli at sunmuw1.MUW.Edu (Rick Mc Callister)
Date: Tue, 13 Jul 1999 12:14:43 -0500
Subject: Latin perfects and Fluent Etruscan in 30 days
In-Reply-To: <3.0.6.32.19990709101606.00798950@box5.tin.it>
Message-ID: <TUE.13.JUL.1999.121443.0500.>

	hijo for "shoot from a dead plant, plant baby" is very common and
is a normal usage
	I've heard Spaniards as well as Latin Americans use it when talking
about "spider plants"
	Be care ful with dictionaries because so many of them use only
technical terms that agricultural engineers use while skipping many of the
actual terms farmers use

[snip]

>If you really happened to hear the word "hijo", it might be due -- IMHO --
>to an improper usage and/or a rare "modismo hispanoamericano".

>Cheers

>       Paolo Agostini

Rick Mc Callister
W-1634
Mississippi University for Women
Columbus MS 39701


From proto-language at email.msn.com  Tue Jul 13 13:16:18 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Tue, 13 Jul 1999 08:16:18 -0500
Subject: accusative and ergative languages
Message-ID: <TUE.13.JUL.1999.081618.0500.>

[ moderator re-formatted ]

Dear Peter and IEists:

 ----- Original Message -----
From: petegray <petegray at btinternet.com>
Sent: Wednesday, July 07, 1999 2:41 PM

> Pat wrote:

>> The idea that vocabulary is irretrievably lost is jejeune.

Pat writes:

Undoubtedly, a comment loses some pungency when the key concept is
misspelled.

Peter wrote:

> The idea that it is not lost is equally jejune (if I may use the English
> spelling - I mean no offence).

Pat writes:

How could I possibly be offended seeing that the language of the list is
English? I acknowledge the mistake.

Peter wrote further:

> One of the problems in PIE is identifying which isolates are PIE and which
> are not, or, where there are similar words in two languages, which are
> inherited cognates, and which are loans (yes, I know there are techniques for
> this - but there are still big problems).

> Proto-proto- languages simply cannot be recovered with the same degree of
> certainty as languages nearer to our attested texts.

Pat writes:

Cannot disagree with a word above. But, based on the study I have made of
supposedly unrelated languages, I am simply amazed that so many CVC(V) roots
seem to be preserved more or less semantically intact in so many of them.

When the languages in question developed CVCVC roots, the picture changes
considerably. CVCVC roots apparently were developed after the major language
families separated so that that level we can either 1) notice retention of
the CVC core; 2) say that CVCVC was not preserved. Is the glass half full or
half empty?

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From martinez at eucmos.sim.ucm.es  Tue Jul 13 17:15:16 1999
From: martinez at eucmos.sim.ucm.es (Javier Martinez)
Date: Tue, 13 Jul 1999 19:15:16 +0200
Subject: Schleicher's fable
Message-ID: <TUE.13.JUL.1999.191516.0200.>

> thinking of Schleicher's fable (in his very Sanskrit-like PIE), which was
> re-worked by Hirt (his version is filled with reduced vowels, but otherwise
> looks very Neogrammarian), and most recently to my knowledge by Lehmann and
> Zgusta (with Lehmann's versions of the laryngeals and of PIE syntax).

cf. also the trilarygalictic version of the fable by Martin Peters in H.
Birkhan, Etymologie des Deutschen, Bern, 1985 (Germanistische Lehrbuchsammlung,
15), p. 308.

j.m.
--
  Javier Mart?nez ~ Filolog?a Griega y Ling??stica Indoeuropea (A-327)
  Facultad de Filolog?a  ~  Universidad Complutense ~  E-28040 Madrid
  Fax: +34- 9131 49023~ Tlf. +34- 91314 4471~ (secret.)+34- 91394 5289
       http://www.ucm.es/info/griego/
  ~ TITUS-Projekt: Vergleichende Sprachwissenschaft ~ Postfach 111932
  Universit?t Frankfurt ~ D-60054 Frankfurt
       http://titus.uni-frankfurt.de

[ Moderator's re-encoding, with apologies.  Not TeX-style ]

  Javier Marti'nez ~ Filologi'a Griega y Lingu"i'stica Indoeuropea (A-327)
  Facultad de Filologi'a  ~  Universidad Complutense ~  E-28040 Madrid
  Fax: +34- 9131 49023~ Tlf. +34- 91314 4471~ (secret.)+34- 91394 5289
       http://www.ucm.es/info/griego/
  ~ TITUS-Projekt: Vergleichende Sprachwissenschaft ~ Postfach 111932
  Universita"t Frankfurt ~ D-60054 Frankfurt
       http://titus.uni-frankfurt.de


From mcv at wxs.nl  Tue Jul 13 18:44:13 1999
From: mcv at wxs.nl (Miguel Carrasquer Vidal)
Date: Tue, 13 Jul 1999 18:44:13 GMT
Subject: accusative and ergative languages
In-Reply-To: <Pine.SO4.4.05.9907091402440.15900-100000@rask>
Message-ID: <TUE.13.JUL.1999.184413.GMT.>

Jens Elmegaard Rasmussen <jer at cphling.dk> wrote:

>>From the primitive and casual books at my disposal I do see that "zuen"
>and "zuten" mean 'he had him' and 'they had him' resp. I also believe I
>see that such auxiliaries are combined with a particularly short form of
>the participles, referred to by Schuchardt as the root of the participle;
>and "jo" is 'stick; beat' in its shortest form, says my little dictionary;
>and "mutil-a" is 'boy' with the article "-a", but without case or number
>marking. Therefore my persistent question: Why can't "mutila jo zuen" and
>"mutila jo zuten" reflect a construction that was earlier meant to express
>'the boy, he had him hit', 'the boy, they had him hit'? Schuchardt also
>gives "zen" to mean 'he was', so that if you gloss "mutila jo zen" as 'the
>boy was hit', it seems there is quite a bit of agreement that the verbal
>root is a participle by itself. I do not see in what way this makes the
>*diachronic* interpretation of "mutila jo zuen" any different from the
>Hindi preterites that are based on Sanskrit constructions of the type
>"A-Nominative + B-Genitive + PPP/nom." meaning earlier "A was (verb)-ed by
>B", but now simply "B (verb)-ed A." Where am I wrong?

Not necessarily wrong, but a complicating factor is the
simultaneous existence in Basque of a set of synthetic verb forms
(first and foremost of the auxiliary verbs "to be" and "to
have"), which, with one exception, are also absolutive/ergative
in nature.  A diachronic interpretation in terms of ancient
passives of the periphrastic forms alone does not explain all of
the ergative verbal morphology of Basque.

We need to demonstrate that the synthetic forms too can be
explained as old passives, and I'm not sure the evidence quite
points that way.  But I think it is possible to derive the Basque
synthetic forms from older periphrastic ones.

Although not quite on topic, let me explain.

If we look at the more or less regular verbs first, we have:

ETORRI "to come"
(present)  (past)
na-torr    nen-torr-en  "I come, came"
ha-torr    hen-torr-en
da-torr    ze-torr-en
ga-toz     gen-toz-en
za-toz     zen-toz-en
da-toz     ze-toz-en

EKARRI "to bring"
na-karr-ERG    ninde-karr-ERG-(e)n     "ERG bring<s>/brought me"
ha-karr-ERG    hinde-karr-ERG-(e)n
da-karr-ERG    --
ga-kartza-ERG  ginde-kartza-ERG-(e)n
za-kartza-ERG  zinde-kartza-ERG-(e)n
da-kartza-ERG  --

where the ergative endings are:
1sg.   -t -da-
2sg.m  -k -ga-
2sg.f. -n -na-
3sg.   -0 -0-
1pl.   -gu -gu-
2pl.   -zu -zu-
3pl.   -te -te-

With 3rd. person abolutive object, however, we have in the
transitive past tense:

ne-karr-en      ne-kartza-n  "I brought it/them"
he-karr-en      he-kartza-n
(z)e-karr-en    (z)e-kartza-n
gene-karr-en    gene-kartza-n
zene-karr-en    zene-kartza-n
(z)e-karr-te-n  (z)e-kartza-te-n

So we have the following pronominal affix sets:

PRES  PAST (itr) PAST (tr1)   PAST (tr2)  ERG      DAT     [PP]
na-   nen-       ninde-       ne-         -t -da-  -t -da-  ni
ha-   hen-       hinde-       he-         -k -ga-  -k -ga-  hi
,,    ,,         ,,           ,,          -n -na-  -n- na-  ,,
da-   ze-        --           e-          -0       -o       --
ga-   gen-       ginde-       gene-       -gu      -gu      gu
za-   zen-       zinde-       zene-       -zu      -zu      zu
da-   ze-        --           e-          -te      -ote/-e  --

My proposal is that the first three sets contain an auxiliary
element *DA (present) / *DE(N) (past), as follows:

na- < *na-da-   nen-/ninde- < *nen-de-
ha- < *ha-da-   hen-/hinde- < *hen-de-
da- < *# -da-   ze-         < *    de-
ga- < *ga-da-   gen-/ginde- < *gen-de-
za- < *za-da-   zen-/zinde- < *zen-de-

As proof, I offer the paradigm of IZAN "to be", which in the 1st
and 2nd pp. sg. is composed regularly out of prefix + root
IZA(N), but in the other persons is composed of the auxiliary
DA/DEN alone (reduplicated in the plural):

naiz (naz, niz)   < *na-a-iz(a) < *na-da-iza
haiz (haz, hiz)   < *ha-a-iz(a) < *ha-da-iza
---
da                              < *#-da
gara (gera, gira) < *ga-ira     < *ga-di.da
zara (zera, zira) < *za-ira     < *za-di.da
dira                            < *di.da

nintzen (nintzan)      < *nen-de-izan-en
hintzen (hintzan)      < *hen-de-izan-en
---
zen (zan)              < *dan-en
ginen (ginan, gindan)  < *gen-di.dan-en
zinen (zinan, zindan)  < *zen-di.dan-en
ziren (ziran)          < *di.dan-en

We see that *d has suffered various phonological developments,
tentatively:

z-  in absolute initial
-d- in sandhi after an absolutive (pro)noun
-r- between vowels, before stressed?
-0- between vowels, before unstressed?

The -ai- in <naiz>, <haiz> (vs. -e- or -a- in <gara>, <zara>)
points to an earlier long vowel (*naaiz < *na-da-iza), just as in
the case of <nau->, <hau-> ("have me, have you") vs. <du-> "have
it" from *na-da-du (> *naau-), *ha-da-du (> *haau-) vs. *da-du (>
*dau-).

Another case where the "auxiliary" -d- resurfaces in the
transitive past is the verb JOAN "to go", whose past tense goes:
ninDoan, hinDoan, zihoan, ginDoazen, zinDoazen, zihoazen.

Reduplication (instead of the usual suffixation of *-(t)z(a)) to
denote plurality of the absolutive might also underlie the forms
of the verb "to have", if we allow for metathesis of the -i-:

nau-   < *naau-    < *na-da-du-
hau-   < *haau-    < *ha-da-du-
du-    < *dau-     < *#- da-du-
gaitu- < *gaaiddu- < *ga-da-di.du-
zaitu- < *zaaiddu- < *za-da-di.du-
ditu-  < *daiddu-  < *#- da-di.du-


The transitive past with 3rd. person object apparently contains
no auxilary.  I would reanalyze the forms as following:

*n(e) e-karr-en
*h(e) e-karr-en
*     e-karr-en
*gen  e-karr-en
*zen  e-karr-en
*     e-karr-en

The e- is possibly the same prefix as in the participle (e.g.
e-karr-i).  Assuming a word break between pronoun and verb avoids
the necessity of reconstructing 1 & 2 pl. as *genne-, *zenne- in
Pre-Basque.  That leaves the interesting question of why the 1st
and 2nd sg. forms have -n (nen-, hen-) in one set of past tense
prefixes, but not in the other (although in Gipuzkoan at least,
variants like <netorren>, <hetorren> do exist).  The 2sg.fem. in
-n (which we see in the ergative) might explain it for the 2nd.
person (cf. a similar phenomenon in Hausa, where the 2nd. person
feminine behaves morphologically like a plural in some personal
prefixes).

Finally, I notice that in this reconstruction, the auxiliary *DA
would have been used in both transitive and intransitive forms
(except 3sg. and pl. of the verb "to be", and the acc/nom.
transitive past).  I further suspect that the same *DA is present
in the causative formant -ra- (e.g. e-ra-man "to carry" <
*e-da-oan "to cause to go", etc.).  Originally "to do, make" [cf.
English "to do" as an auxiliary]?

=======================
Miguel Carrasquer Vidal
mcv at wxs.nl


From colkitto at sprint.ca  Thu Jul 15 01:47:14 1999
From: colkitto at sprint.ca (Robert Orr)
Date: Wed, 14 Jul 1999 21:47:14 -0400
Subject: PIE vs. Proto-World
Message-ID: <WED.14.JUL.1999.214714.0400.>

-----Original Message-----
From: petegray <petegray at btinternet.com>
Date: Tuesday, July 13, 1999 2:22 AM

>Steven said:

>> Although i'm not prepared to [compose] a fable in PIE, i certainly do not
>> doubt that, .. it could be done with our current state of knowledge.

>Yes it has, quite recently.  A number of linguists were asked to give their
>versions, and they were published together in a special volume.  I'm afraid
>they prove you wrong - the diversity was enormous, with some linguists using a
>more-or-less Brugmannian approach, and others being much more glottalic or
>laryngeal.  The variety is great fun to see, and very educative!

>Peter

>[ Moderator's request:
>  Would you mind providing bibliographical information for the volume in
>  question?  I for one would be fascinated.
>  --rma ]

So would I - Robert Orr


From Georg at home.ivm.de  Tue Jul 13 22:22:59 1999
From: Georg at home.ivm.de (Ralf-Stefan Georg)
Date: Wed, 14 Jul 1999 00:22:59 +0200
Subject: accusative and ergative languages
In-Reply-To: <006c01beca76$3bff6ea0$ec142399@patrickcryan>
Message-ID: <WED.14.JUL.1999.002259.0200.>

>Pat:

>Yes, but if a new verb were formed, e.g. webben (from English 'web'), there
>is no possibility that it would be conjugated webben, wabb, gewobben.

>What I consider 'internal inflection'  is Arabic yaktubu, kataba, katibun,
>etc. which applies to any and all verbs, old, and those taken new into the
>language.

The pattern lost productivity in German. So what ?
Also, I hardly understand your claim, which basically amounts to refusing
to speak of technique X in language L, unless it isn't present in each and
every item in which it theoretically could be present. So, you would refuse
to speak of, say, vowel harmony in, say, Turkish, because some borrowed
(and some native) words fail to obey this rule ? ("What I call VH is ...,
which applies to any and all ... Period. There is an exception, or the
phenomenon is gradually losing ground, therefore it doesn't extst at all."
?????).

Taking your example with the artificial verb "webben" (maybe meaning
something  like "to surf the web", for which we already adopted "surfen").
You are right that it could be expected to be inflected according to the
weak pattern. But, if some people choose to use the strong past participle
(leaving aside the simple preterite, which is rarely used in colloquial
speech nowadays), maybe out of some whim, maybe with humorous intentions,
and manage to place this strategically in the mass-media, say, in
online-service-ads, I see no stringent reason why this shouldn't fall on
fertile soil and get used gradually. First as a joke, later as a habit. The
advertising industry *is* able to do this and other things to your or my
language. The pattern is there, and any existent pattern can trigger
analogy.Maybe not too likely, I admit in this case, but certainly not
impossible. Analogy is not eternally confined to pre-modern times.

Stefan Georg
Heerstrasse 7
D-53111 Bonn
FRG
+49-228-69-13-32


From larryt at cogs.susx.ac.uk  Wed Jul 14 10:19:08 1999
From: larryt at cogs.susx.ac.uk (Larry Trask)
Date: Wed, 14 Jul 1999 11:19:08 +0100
Subject: `cognate'
In-Reply-To: <ef25a75f.24b81413@aol.com>
Message-ID: <WED.14.JUL.1999.111908.0100.>

On Fri, 9 Jul 1999 JoatSimeon at aol.com wrote:

[LT]

>> But I have noticed that a significant number of
>> non-linguists on this list have apparently misunderstood the sense of our
>> technical term `cognate': many of them
>> appear to believe that `cognates' means something like `words of similar
>> form

> - that's odd; I always assumed it meant "derived from a common
> ancestral word".  You know, like Tiwaz and dyaus.

Yes, but non-linguists often misunderstand the term, and I've noticed
several queries on this list in the last few weeks from people who had
apparently misunderstood it.

By the way, the following article presents a small but spectacular
sample of cognates which do not resemble one another at all:

John Lynch (1999), `Language change in southern Melanesia: linguistic
aberrancy and genetic distance'. In Roger Blench and Matthew Spriggs
(eds), Arachaeology and Language IV, pp. 149-159, London: Routledge.

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk


From jer at cphling.dk  Wed Jul 14 01:04:18 1999
From: jer at cphling.dk (Jens Elmegaard Rasmussen)
Date: Wed, 14 Jul 1999 03:04:18 +0200
Subject: accusative and ergative languages
In-Reply-To: <006801bec8b4$e04f1460$bdf9abc3@xpoxkjlf>
Message-ID: <WED.14.JUL.1999.030418.0200.>

On Wed, 7 Jul 1999, petegray wrote:

> [...]
> Proto-proto- languages simply cannot be recovered with the same degree of
> certainty as languages nearer to our attested texts.

At the risk of shocking some and destroying what little reputation I may
have, I'd say: That depends on where the lines meet.

Thus, on the basis of a handful of Modern Indo-European languages taken
from different branches you can hope to get little glimpses of the
language of the youngest period common to all of them, i.e. the IE
protolanguage. You cannot get Russian + English + Persian + Greek +
Italian + Albanian + Irish to tell you what 'name' was in Old Russian, or
in Proto-Balto-Slavic; your only result, if any, would apply to
Proto-Indo-European.

Therefore, in many respects we are better informed about very old periods
than about the intervening stages closer to the attested languages. But do
go easy with this truth: it is immediately misused to discredit our most
objective scholarship if told to the wrong people.

Jens


From larryt at cogs.susx.ac.uk  Wed Jul 14 10:20:17 1999
From: larryt at cogs.susx.ac.uk (Larry Trask)
Date: Wed, 14 Jul 1999 11:20:17 +0100
Subject: `cognate'
In-Reply-To: <Pine.SO4.4.05.9907121416130.5019-100000@rask>
Message-ID: <WED.14.JUL.1999.112017.0100.>

On Mon, 12 Jul 1999, Adam Hyllested wrote:

> On Wed, 30 Jun 1999, Larry Trask wrote:

[LT]

>> `father', all of these being descended from PIE *<pater>.

> I'm sure it says PIE *<p at ter> in your manuscript (+ diacritics on the
> <e>).

Yes, it does, and my posting should have read `p at ter', but I slipped up.

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk


From proto-language at email.msn.com  Wed Jul 14 01:20:09 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Tue, 13 Jul 1999 20:20:09 -0500
Subject: accusative and ergative languages
Message-ID: <TUE.13.JUL.1999.202009.0500.>

Dear Peter and IEists:

 ----- Original Message -----
From: petegray <petegray at btinternet.com>
Sent: Sunday, July 11, 1999 1:45 PM

>> Pat wrote:

>> Well, on page 40 of Chao's Mandarin Primer, are listed "Affixes":
>> 11 are listed; ...

Peter wrote:

> I simply cannot let this go.    Verbal compounds are rife in modern Chinese,
> and the second elements ("affixes" if we are using terms inappropriate for
> Chinese) are derived from a large number of concepts.   The fact that Chao
> lists 11 only means that he has selected 11, and does not reflect the actual
> nature of the Chinese language.   Things are not as simple as that grammar
> makes them seem!

Pat responds:

Fine, Peter. Why not list two or three additional elements that function in
Chinese equivalent to inflectional affixes?

 Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From edsel at glo.be  Wed Jul 14 10:21:19 1999
From: edsel at glo.be (Eduard Selleslagh)
Date: Wed, 14 Jul 1999 12:21:19 +0200
Subject: Chronology of the breakup of Common Romance [long]
Message-ID: <WED.14.JUL.1999.122119.0200.>

[ moderator re-formatted ]

-----Original Message-----
From: Rick Mc Callister

>> "Vlach" or "Wallach" seems to be descended from a name that was commonly
>> applied to Celts.

> It has long been proposed that the name Volcae, a Gaulish tribe, was the
> source of Anglo-Saxon Wealh, Balkans Vlakh, Polish WLochy, etc.  The point
> I've generally seen is something to the extent that supposedly the Volcae
> were the first Gaulish tribe to fall under Roman rule.

[Ed Selleslagh]

That seems highly unlikely since the term was originally used exclusively by
non-Romans, non-Celts, and applied to Celtic and Latin languages speaking
peoples.  That excludes a Roman origin of the word.
All this sounds like a typical 19th century invention by Latin-worshippers

> Some popular writers have also proposed Volcae as the source of
> English "folk" but I've never seen this proposal from professional
> linguists.

[Ed]

Even less likely : the peoples that used the word had probably never heard
of the Volcae, not even second-hand.

>> Appearing as "Walh" or "Walah" in OHG, it has interpreted as meaning
>> "foreigner", sometimes Roman, but in usage it is closely associated with
>> Celts and regions of Celtic habitation - e.g., "Wales", "Walloon".  I seem
>> to recall an suggestion that there may be a tie here to L. vallum (fortified
>> wall, the earliest meaning of "wall") and refer to the Celtic or
>> Romano-Celtic oppidum or walled town.

> You'd have to account for the ending in Vlakh, WLochy, etc.

[Ed]

Indeed, and a derivation from a Latin word is unlikely for the reasons given
above.

Note that the form with -ch is East-European, while the 'Wal-' form is
Western, mainly Germanic and in loanwords of Germanic origin, e.g. in Latin
languages (Walen-Wallons, Wallis-Valais, Wales...).

>> The strong Celtic presence along the Danube is attested by Classical writers
>> well before the common era.  E.g., Alexander fights them before turning
>> against the Persians and meets Celts who are from Illyria.  Galicia in
>> southern Poland is a region whose name remembers a Celtic presence even
>> farther north.  I remember an old passage where Vlachs are identified as
>> "the shepherds of the Romans" and in this role they may also have been
>> imported help as they were in northern Italy.

> Isn't the term Vlakh anachronistic in this sense? Vlakh is first
>documented in the middle ages, isn't it?

[Ed]

Indeed. The medieval use of 'Vlach' probably referred to the 'Latin'
Rumanians, cf. Wallachia.
In modern Greece, the word is also used to refer to the local nomads/gypsies
of unclear (i.e. to me) ethnic origin (maybe Rumanian Gypsies, 'Roma', or
from Pannonia? Maybe Albanians?).

>> The ancient ethnic designations in that part of the world are a little
>> difficult to follow, but it seems clear that the whole region from Illyria
>> to present day Romania was under Rome by 250 ace.  And there is some
>> possibility that Vlachs represented Romanized Celts across those regions.
>> In any case, the small difference in time between Rome's entry into Illyria
>> and into Dacia would be a de minimis factor.

> My guess is that any Celtic element in the Balkan would have been extremely
> thin.

> The Celts raided the Balkans and stormed through the region on the way to
> Galatia but my understanding is that they lived north of the Balkans in
> Galatia and Pannonia

[Ed]

See my last remark above.

[snip]

Ed. Selleslagh<edsel at glo.be>


From proto-language at email.msn.com  Wed Jul 14 01:52:32 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Tue, 13 Jul 1999 20:52:32 -0500
Subject: Interpreting ergative sentences
Message-ID: <TUE.13.JUL.1999.205232.0500.>

Dear Leo and IEists:

 ----- Original Message -----
From: <CONNOLLY at LATTE.MEMPHIS.EDU>
Sent: Monday, July 12, 1999 12:01 AM

<snip>

> Sumerian is even worse, since we don't have all that much of
> it (no poetry, no menus, nothing that seems to be finely nuanced
> speech.  And on top of that, it's ergative, and most of us (including
> me) are *not* used to dealing with such languages.

 <snip>

> So please, caution!  And take due note of what has been discovered
> about *living* ergative languages.  You can't hope to understand the
> dead ones without them.

Pat comments:

I think everyone will agree that this is an interesting posting. And the
cautions that Leo advises should be taken to heart by anyone dealing with
languages of the past.

But, Leo, there is quite a lot of Sumerian poetry --- some quite intriguing.

Pat


From edsel at glo.be  Wed Jul 14 11:34:17 1999
From: edsel at glo.be (Eduard Selleslagh)
Date: Wed, 14 Jul 1999 13:34:17 +0200
Subject: `cognate'
Message-ID: <WED.14.JUL.1999.133417.0200.>

-----Original Message-----
From: JoatSimeon at aol.com <JoatSimeon at aol.com>
Date: Wednesday, July 14, 1999 12:43 AM

>>larryt at cogs.susx.ac.uk writes:

>>This posting will not be news to the linguists on this list, who may prefer
>>to stop reading at this point, unless they want to offer criticisms of the
>>definitions below.  But I have noticed that a significant number of
>>non-linguists on this list have apparently misunderstood the sense of our
>>technical term `cognate': many of them
>>appear to believe that `cognates' means something like `words of similar
>>form

>- that's odd; I always assumed it meant "derived from a common ancestral
>word".  You know, like Tiwaz and dyaus.

[Ed Selleslagh]

That's what I meant when I caused all this by calling Grk. 'cheir' (N.Grk.
'cheri') and Georgian (Kartvelian) 'cheli' cognates. I should have said
'possible cognates' since I was asking the list members whether these
actually were or could be cognates.

Sorry for all the fuss, but at least things got clearer (1. they probably
aren't cognates, 2. we, the non-professionals,  have a better understanding
of what cognates are).

Ed.


From proto-language at email.msn.com  Wed Jul 14 07:49:56 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Wed, 14 Jul 1999 02:49:56 -0500
Subject: accusative and ergative languages
Message-ID: <WED.14.JUL.1999.024956.0500.>

Dear Ralf-Stefan and IEists:

 ----- Original Message -----
From: Ralf-Stefan Georg <Georg at home.ivm.de>
Sent: Saturday, July 10, 1999 5:03 AM

R-S wrote previously:

>>> Having to be rendered by the passive in English is not the same thing as
>>> "being passive in nature".

>> Pat responded:

>> How about explaining "passive in nature"?  Is that a Platonic idea?

R-S then responded:

> As you might have observed, I put double quotes around this expression, the
> most widespread functions of which are:

> - the expression is not mine, but used by others, and I use it only not to
> complicate the discussion, assuming everyone knows what I'm talking about

> or

> - the expression is handy, but admittedly imprecise, I only use it since I
> don't want to put too much effort in terminological precision at this
> point, having different things to communicate (admittedly a vice, but
> normally tolerable when speaking to attentive and knowledgable people)

Pat responds:

All very well and good. Having explained next to nothing, perhaps you will
now tell us how precisely *you* were using "passive in nature" and what in
God's name it is supposed to mean to you, to those you may be quoting, or
anyone else including me.

R-S continued:

> I was reacting to some claim of yours, namely that ergatives are - now what
> was your expression ? I don't recall, was it "really passives", "always
> former. i.e. reanalyzed p.s", "p.s in nature", "p.s by nature". Hell, I
> dunno. Something to this effect, at least. The important thing is that I
> managed to point out that this, i.e. the ergative-as-passive-claim, however
> phrased, is wrong, and Larry Trask, who has also a well-known publishing
> record on the issue, has spoken the definite word about this here.
> Pat, this is, alas, a familiar pattern: when one's theory gets into dire
> straits, some aside discussion on irrelevant points is opened, obviously
> with the intention to show weaknesses of whatever kind in the opponent's
> standing, hoping that this will cloud the correct and justified remarks he
> might have made on the relevant points before. This may work in some
> election campaign, but not here, sorry.

Pat responds:

Yes, you are absolutely correct. This is a familar pattern indeed. When
challenged for an argument, you sidestep the issues of the question by
conveying that someone you like (here, Larry Trask) "has spoken the definite
word about this here". If Larry could speak the definite word about
everything, then this list would be a waste of time. We could just subscribe
to his newsletter for the latest ex cathedra rulings on all our troublesome
questions.

Now, when Larry recently quoted Dixon about the nature of the ergative, he
conveniently neglected to mention that Dixon acknowledged that there were
currently practising linguists --- not amateur linguists like myself ---
still defending the passive interpretation of ergative constructions. I
asked Larry where he had "shredded" this interpretation, and to my
knowledge, got no answer. If I missed the "shredding", perhaps you will be
kind enough to rehearse his performance for us. I have seen nothing by
Larry's vehemence and your allegiance to support the idea that the ergative
should not be interpreted as a passive.

R-S then continued:

> To answer your question: "passives in nature", as used by me, is neither a
> Platonic idea, nor a Wittgensteinean prototype, nor a correct description
> of anything I may happen to think, it is just shorthand for a bundle of
> concepts I don't share. And, above all, it is bad English. Should we open a
> new thread about this, or shouldn't we better return to ergativity and your
> misconceptions about it, such as "there are ergative languages without any
> splits" athl. ?

Pat concludes:

That Ralf-Stefan is incapable of defining the term he introduced: "passive
in nature". By the way, being a native speaker of English, I can assure you
that in my dialect, "passive in nature" is *not* bad English.

And to answer your -- I hope not purposeful -- distortion of what I wrote,
let me say explicitly that I did not assert "there are ergative languages
without (any) splits". I asserted that Thomsen did not, at least in her
grammar, identify splits in Sumerian, which you seemed to think she had. I
then invited you to identify them in Sumerian if you *could*.

You might review your own procedures for employing quotation marks. Your use
of them on "there are ergative languages without any splits" strongly and
falsely implies that I wrote this in the context of a judgment on the
question.

 Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From rmccalli at sunmuw1.MUW.Edu  Wed Jul 14 19:25:03 1999
From: rmccalli at sunmuw1.MUW.Edu (Rick Mc Callister)
Date: Wed, 14 Jul 1999 14:25:03 -0500
Subject: Momentary-Durative
In-Reply-To: <001901becc6c$a3651e80$8471fe8c@lucent.com>
Message-ID: <WED.14.JUL.1999.142503.0500.>

Spanish distinguishes them
by resorting to

estar --a "momentary" verb--
[actually a verb indicating condition] for resultant conditions--

and ser--a "durative" verb
[actually a verb indicating  characteristics]
for passive constructions.

e.g. The store

Past-tense resultant conditions are usually descriptive and usually in the
imperfect while past-tense passive voices usually deals with completed
actions and is usually in the preterite

e.g.

The was closed. (resultant condition)
La tienda estaba cerrada.

The store was closed. (passive)
La tienda fue cerrada.

Curiously, in the present tense Spanish tends to use
a medio-passive type of construction,
instead of passive voice.

The store is closing/being closed. (medio-passive)
Se cierra la tiendas.

The store is closed. (resultant condition)
La tienda esta/ cerrada.

Does anyone know why?

[snip]

>Even worse, in modern Western European languages, resultatives and passives
>cannot be readily distinguished, and perfect and resultatives are also
>similar. Thus we take ``John has eaten dinner'' to be resultative, but
>``John has eaten ostrich meat'' as experiential perfect. ``The door is
>closed'' can be resultative or passive, while ``The door is closed by John''
>is passive; the resultative has be expressed as ``The door is closed because
>of John'' or some such.

[snip]

Rick Mc Callister
W-1634
Mississippi University for Women
Columbus MS 39701


From proto-language at email.msn.com  Wed Jul 14 07:58:06 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Wed, 14 Jul 1999 02:58:06 -0500
Subject: accusative and ergative languages
Message-ID: <WED.14.JUL.1999.025806.0500.>

Dear Ralf-Stefan and IEists:

 ----- Original Message -----
From: Ralf-Stefan Georg <Georg at home.ivm.de>
Sent: Saturday, July 10, 1999 5:27 AM

>> Pat wrote:

>> Well, on page 40 of Chao's Mandarin Primer, are listed "Affixes":

>> 11 are listed; of these 6 qualify as related to "inflections" :
>> modal -m(en), phrase marker -le; completed action -le; progressive
>> action -j(y/e); possibility or ability -de; subordination -de.

>> Undoubtedly, a historical grammar might provide a few more but I consider
>> this a pretty simple system.

R-S sighed:

> This is getting weary. I think no sane linguist will be unaware of the fact
> that there are languages with fewer purely morphological means than others.

Pat sympathizes:

Apparently not.

R-S continued:

> Also, "simpler" and "not-so simple" phonological systems have been heard of.
> But, if I'm not completely mistaken, there was talk about sthl. like
> general, overall simplicity of languages, which is a pre-scientific notion,
> quite simply, if this pun is allowed.

> If we assume, as of course we have to, that human languages are
> problem-solving devices which face a set of possible communicative
> problems, which is the same for all linguistic communities (don't come with
> obvious cultural differences, which pertain to vocabulary only; I'm aware
> of the fact that not all linguistic communities of the planet need to talk
> meaningfully about the various kinds of fish in the Yenissey, or the
> Sepik). Languages make different choices regarding the set of functions
> they grammaticalize (tense vs. aspect or both, inferentiality, relative or
> absolute tense, number, participant-identification, reference-tracking,
> pragmatic categories, you name it), the hows and whys of which are the
> object of linguistic typology. One language uses grammaticalized
> bound-morphology for, say, inferentiality, others have to use different
> means to convey the idea. But they are all able to convey the idea, however
> elegant, or however clumsy.

Pat interjects:

Yes, some people use a drill-stick and bow to make a fire, and some use
matches.

R-S continued explaining:

> The idea is that they *use means* to do it, or
> that speakers may find these means, even if the category is weakly or not
> at all grammaticalized in their language (your inflection paradigm may be
> my intonation pattern, your inferentiality affix may be my expletive
> adverb; to convey the idea of the Dakota sentence-particle /yelo/ or Thai
> /khrap/ I may even be forced to say each time "I am male"; clumsy, but
> possible, if I face the necessity). The description of these means is the
> task of the grammarian. Some grammarians are aware of this task, and write
> a 600-page grammar of Chinese, some aren't (or simply want to produce a
> primer to help you get along abroad) and write a 60-pp. treatment of
> Russian. So what ?

> I will not deny that notions of simplicity vs. complexity may be useful
> distinctions when talking about subsystems. For the characterization of
> whole languages they are definitely not useful. The grammar-school like
> rote learning of paradigms, which your Chinese teacher will be able to
> spare you, will be compensated by having to digest subtle and difficult
> rules of syntax.

Pat concludes:

In my opinion, dearth of morphological devices is validly characterised as
"simple".

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From petegray at btinternet.com  Wed Jul 14 19:21:14 1999
From: petegray at btinternet.com (petegray)
Date: Wed, 14 Jul 1999 20:21:14 +0100
Subject: Momentary-Durative
Message-ID: <WED.14.JUL.1999.202114.0100.>

Vidhyanath's two postings on aspect show clearly what the issue of aspect in
PIE may be - we simply don't have a clear definition of what we are talking
about.

Greek usage does not agree with Vedic, and where traces of aorists exist in
other languages (e.g. Armenian, OCS, Albanian, Latin, Baltic), we cannot
convincingly recover an original aspectual meaning.

The formations are clearly PIE, but I believe I can continue to claim that
the distinctions between them, if recoverable, are certainly not those we
find in Greek.   The Armenian-Greek- I-I sprachbund has certain known
innovations, such as vocabulary and the augment, and we should not be miseld
into accepting as original what may be another innovation, namely the
development of new distinctions based on an existing variety of forms.

Peter


From Georg at home.ivm.de  Wed Jul 14 08:52:08 1999
From: Georg at home.ivm.de (Ralf-Stefan Georg)
Date: Wed, 14 Jul 1999 10:52:08 +0200
Subject: accusative and ergative languages
In-Reply-To: <001001becadf$e28d9580$759ffad0@patrickcryan>
Message-ID: <WED.14.JUL.1999.105208.0200.>

>Pat responds:

>I do not readily have access to material on Khalkha Mongolian so I cannot
>comment in detail on your example.

>However, in general, I would hazard the guess that, regardless of its
>apparently accusative construction, for all practical purposes, this
>sentence could be just as easily translated: 'A book was sent to you'.

If I had to *translate* a Mongolian novel into English, I might choose the
passive as well, in *English*, since the "agentlessness" of this sentence
is something which goes well with English (and other) passives. But that's
*translation*, and not *analysis*. I chose Mongolian, because it has a
passive available. The active construction, as I gave it, will be most
naturally be found, when an agent is inferable from the
pragmatic/textual/situational context, passive being preferred when it is
unknown or information on it deliberately suppresses. there are other
parameters which govern the choice of diathesis, mostly of a pragmatic
nature and not entirely explored. The question was about grammaticality of
such a construction in an acc. language.

> I am
>also a little sceptical of an "accusative" in such a context which is marked
>by -0.

I made a slight mistake, thanks for your attentiveness. I should have
rendered the sentence in English with *a book*, not *the book*, the overt
accusative marker being used with definite patients only. With an
indefinite one, it is -0.

>But, having said that, I am willing to concede that for *some* accusative
>languages, this construction, provided a nominative subject can be mentally
>supplied from context, is grammatical.

OK, thanks, we are getting somewhere.

Stefan

Stefan Georg
Heerstrasse 7
D-53111 Bonn
FRG
+49-228-69-13-32


From petegray at btinternet.com  Wed Jul 14 19:29:06 1999
From: petegray at btinternet.com (petegray)
Date: Wed, 14 Jul 1999 20:29:06 +0100
Subject: Momentary-Durative
Message-ID: <WED.14.JUL.1999.202906.0100.>

>>> I don't see any parts of Sans lit in which aorist has resultative
>>> meaning.

If you didn't like my first example, how about:
  Ramayana 2:64:52
  yah s'aren.aikaputram tvam aka:rsi:r aputrakam
 with one arrow you have rendered me childless, who had but one son

(Sorry you don't like my transliteration - it's due to using ASCII)

or Mahabharata 5:3:10
  tad akars.it prajagaram
  that has produced sleepiness

or Mahabharata 2:60:7
  kim nu pu:rvam para:jais.i:t a:tma:nam
 whom you have lost first, yourself....

In all of these examples the result of the action is the point of the
statement - and it is an aorist which is used to describe it.

Peter


From W.Schulze at lrz.uni-muenchen.de  Wed Jul 14 10:18:26 1999
From: W.Schulze at lrz.uni-muenchen.de (Wolfgang Schulze)
Date: Wed, 14 Jul 1999 12:18:26 +0200
Subject: accusative and ergative languages
Message-ID: <WED.14.JUL.1999.121826.0200.>

"Patrick C. Ryan" schrieb:

>.... As for lessened degrees of animacy, most ergative languages
> have antipassives to indicate this.

Before you invite people to subscribe to this claim you should first
demonstrate that a) "most ergative languages have antipassives". This
claim suggests that "ergative" is a substantial attribute that can be
used with the referent "language". In earlier postings I have tried to
show that "ergativity" (as well as "accusativity") represents a label
for a structural BEHAVIOR of single paradigms WITHIN a language system.
For instance (as I have said) a language system may be ergative in its
agreement paradigm, accusative in owrd order, accusative in case
marking, ergative in discourse cohesion etc. (to give a fictive
example). Hence, there are NO "ergative (or "accusative") languages" or
only, if you use this term in a very informal sense. Now, if you talk
about antipassives, you should make clear to which morphosyntactic
domain you allude to. Moreover, your claim suggests that most 'ergative
languages' are reference dominated in the sense of Role and Reference
Grammar. Only if a given language system ("operating systems in terms of
the Grammar of Scenes and Scennarios" (GSS)) uses actant encoding
devices to indicate fore- and/or backgrouding (instead of - for instance
- smeanti coles such as agent/patient...) we can expect some kind of
diathesis be it passive or antipassive (note that again passives and
antipassives represent two poles on a much more complex scale that also
involes bi-absolutives, pseudo-passives and many more structures).
	The fact, however, is that many 'ergative' languages lack an
antipassive. For instance, there are nearly 30 East Caucasian languages
all of them using some ergative strategies in at least parts of their
operating systems. But only a handfull of them (five or six, to be
precise) have true antipassives (only one has some kind of
"pseudo-passive"). The same is true for accusative systems (as you
probably know). Hence, antipassives are a possible extension of ergative
stretagies, they cannot serve for any kind of typological
generalization.

b) Antipassives have rarely to do with the "lessening of animacy". The
most common inferences that allow antipassive structures are:

- Reduction of 'activity' (that is the degree to which an actant is
thought to be 'active' during a *specifc* (and single) event. From this
another inference is given:

- Habitual, durative action (-> imperfectiveness).....

- The event becomes less discrete, hence less transitive. Another
inference: The 'patient' looses its referentiality: It cannot be
subjected to wh-questions, it cannot be counted, very often such
referents are mass nouns or collectives...

- Antipassives are part of the discourse cohesion strategies (most
famous example is Dyirbal): Here, antipassives are neither semantically
nor syntactically motivated, but merely a pragmatic feature of topic
chaining.

There are much more functional options that are carried out by
antipassives. In fact, all these options NEVER allow such a claim as
quoted above (rather, they contradict it).

> Although Dixon is certainly a man who has devoted much thought to
> ergativity, I find something inherently problematical in combining ergative
> and accusative features in one sentence (a little schzophrenic) which he is
> forced to do by analyzing pronominal and nominal structures differently when
> they occur in the same sentence.

WHY? It all depends from how you interpret erg. and acc. features.
Confer for instance the following (one!) sentence from (informal) Lak
(East Caucasian):

t:ul b-at:-ay-s:a-ru zu
I:ERG I:PL-hit-PART:PRES-ASS-SAP:PL you:PL:ABS
'I surely hit you (plural)'

[For the expert: Standard (Literary) Lak would have 'na bat:ays:aru zu'].

This sentence is:

ACC with respect to word order [*zu t:ul bat:ays:ara would be ERG]
ERG with respect to case marking ['neutral with 'na' for "I" is also
possible]
ERG with respect to class agreement (b- = (here) class [+hum;+plural])
ERG with respect to personal (or, better, speech act participant)
agreement (-ru is SAP:PL and agrees with 'zu' "you:PL").

But if you say "I am surely hitting you (plural)", you get:

na b-at-la-ti-s:a-ra zu
I:ABS I:PL-hit1-DUR-hit2-ASS-SAP:SG you:PL:ABS

Here we have:

ACC with respect to word order
ERG with respect to calss agreement
ACC (or neutral) with respect to case marking)
ACC with respect to SAP agreement (-ra is triggered by 'I:ABS').

Now, please tell me: Is Lak an 'ergative' or an 'accusative' language?
[Please note that I did not include (among others) strategies of
discourse cohesion, reflexivization and logophization].

> I think it is likelier that, because of perceived greater animacy (or
> definiteness), pronouns have a different method of marking that can still be
> interpreted within an ergative context.

This a (very simplified) 'on-dit' that stems from the earlier version of
the Silverstsein hierachy. Again, we have to deal with the question,
whether a 'pronoun' (I guess you mean some kind of 'personal pronouns')
can behave 'ergatively' or 'accusatively'. The list below gives you a
selection of SAP case marking in East Caucasian languages with respect
to ABS/ERG:

	ABS vs. ERG	ABS = ERG
	ALL		---
	Singular	Plural
	Plural		Singular
	1.Incl.		Rest
	1:SG		Rest
	2:SG		Rest
	1:SG/PL		Rest
	---		ALL	

This list (aspects of personal agreement NOT included!) shows that SAP
pronouns may behave different within the same paradigm. Any
generalization like that  one quoted above does not help to convey for
these data...

[Please note new phone number (office) :+89-2180 5343]
___________________________________
| Prof. Dr. Wolfgang Schulze
| Institut fuer Allgemeine und Indogermanische Sprachwissenschaft
| Ludwig-Maximilians-Universitaet Muenchen
| Geschwister-Scholl-Platz 1
| D-80539 Muenchen
| Tel:	+89-21802486 (secr.)
|      	+89-21805343 (office) NEW ! NEW !
| Fax:	+89-21805345
| Email: W.Schulze at mail.lrz-muenchen.de
| http://www.lrz-muenchen.de/~wschulze/
_____________________________________________________


From petegray at btinternet.com  Wed Jul 14 20:19:23 1999
From: petegray at btinternet.com (petegray)
Date: Wed, 14 Jul 1999 21:19:23 +0100
Subject: Momentary-Durative
Message-ID: <WED.14.JUL.1999.211923.0100.>

Thank you for your long reply, Jens.   Firstly, I apologise for any offence
in the word "ideology".   I intended only to indicate what I see (rightly or
wrongly) as assumptions based on theories.

Now, going through your posting in detail:

>  is the common occurrence of a
> nasal-infix structure from *tewd- 'thrust' in Skt. tundate and Lat. tundo
> not part of our "data"?

Yes - but you have made the assumption that this root always shows the nasal
in the present.   This is not the case.
(a) Albanian shows two forms, one with, and one without the nasal.
(b) Old Irish shows no nasal.
(c) Greek a-tuzomai shows no nasal.
(d) If Pokorny is right to link the root with stud- then we have no nasal in
Latin studeo, nor in the various reflexes of it in Germanic.

>  There are so many instances of the _same_ verbal root turning up
> with a nasal present in IE branches that have nothing else in common ...

And there are very many instances of roots showing nasal presents in one IE
language, and not in another, or indeed appearing in both forms in the same
language, as Latin cumbo and cubo, or Greek leipo: and limpano:

>. Is it a coincidence
> that *k^lew- 'hear' forms a nasal present in Indo-Iranian and Celtic?

Yes.   It has no nasal in Greek or Latin or some others. (and so on with the
rest of your list)

I cannot agree with you that certain verbal rootts always had nasal presents
in all IE languages.   This is simply factually untrue.   But without that
assumption, your argument collapses.

>. Does that not indicate
> that the assigment of the nasal-infix structure as the present of certain
> roots was fully lexicalized in the protolanguage?

No, for the reasons just stated.   There are enough counter-examples to show
that whatever the conditions were for the "selection" of nasal, full grade
or suffixed present, they are now as unrecoverable as they are within Latin,
Greek, or Sanskrit, where very little difference of meaning can be
consistently shown.

> Oh no, it works only in the direction that a nasal present has a root
> aorist beside it in practically all cases.

Not a surprise, if you ignore all the langauges where this is not the case.
Likewise, not a surprise, since root aorists were an old form.   The same
could be said of full grade presents, or suffixed presents.   So the
statement really is without significant meaning.

> Also the reduplicated presents
> and the y-presents generally form root aorists.

Yes - so your argument (that a root aorist implies a nasal present) is
really rather weak.   Yet without this argument, you cannot derive tuda'ti
the way you do.

> [in] PIE, each verb had mostly picked one form for its present. The massive
> concord among the IE languages in this respect

There is no such concord.  Even a single language shows variety of
formations of the present, and the variety across the languages is
considerable.    But it would be good to have firm data on this - anyone got
three weeks to spare going through the text books?

> - while the even greater discord is easily
> explained by continued normal language change.

No.   Normal language change would not turn a nasal present into a
reduplicated one or vice versa, nor would it turn a -sk suffixed one into a
nasal present, etc, etc.

Peter


From edsel at glo.be  Wed Jul 14 10:54:06 1999
From: edsel at glo.be (Eduard Selleslagh)
Date: Wed, 14 Jul 1999 12:54:06 +0200
Subject: Interpreting ergative sentences
Message-ID: <WED.14.JUL.1999.125406.0200.>

-----Original Message-----
From: CONNOLLY at LATTE.MEMPHIS.EDU <CONNOLLY at LATTE.MEMPHIS.EDU>
Date: Wednesday, July 14, 1999 3:06 AM

>Dear Pat and everyone else,

[snip]

>This is actually quite like what often happens with experiencer and
>beneficiary verbs in accusative languages.  The fact that e.g. Spanish
>makes the patient the morphological and syntactic subject of a verb
>does not mean that an English translator must do the same.  _No me
>gusta la mu'sica_ does not mean 'the music doesn't please me' -- that
>translation suggests that someone was trying to please (or perhaps
>annoy) me with the music.  Rather, it means, very precisely, 'I don't
>like the music': the Spanish doesn't insinuate that anyone was thinking
>of me, so neither should the translation.

[Snip]

>Leo

[Ed Selleslagh]

First of all, thank you for this excellent piece of work.

Speaking about Castilian, I would like to have your views on a peculiarity,
or rather a tendency, that is still productive at least in popular speech
('le-ismo'), namely the tendency to use the indirect subject form where all
other Latin (and other West-European) languages use the direct object form,
and almost exclusively with persons (animate), e.g. "le v?" ("le vi' "), "I
saw him".
It has sometimes be suggested that this was a substrate influence from
Basque, an ergative language (that lacks an accusative, of course), but
opinions are extremely divided on this subject. On the other hand, it seems
to me this could (but I don't know) also be related to some of the arguments
presented in the discussion on ergative/accusative concerning the distinct
role of animates.

Ed.


From jer at cphling.dk  Thu Jul 15 02:20:50 1999
From: jer at cphling.dk (Jens Elmegaard Rasmussen)
Date: Thu, 15 Jul 1999 04:20:50 +0200
Subject: Momentary-Durative
In-Reply-To: <001a01becc6c$a4de02e0$8471fe8c@lucent.com>
Message-ID: <THU.15.JUL.1999.042050.0200.>

On Mon, 12 Jul 1999, Vidhyanath Rao wrote:

  [snip of my (Jens') message]

> The tuda'ti type, with primary endings, is somewhat rare in RV. Perhaps,
> tuda'ti type was a replacement for injunctives, i.e., tuda'ti was a general
> present and not progressive present. Just a thought.

Hardly so, for the injunctive is not a stem-formation, but an inflectional
category that can be formed from all verbal stems.

> [Jens:]

>> Strunk has shown that nasal presents go with root aorists,

> [VR:]

> Reduplicated forms seem to go with root aorists as well. [...]

Indeed they do, the percentage is _very_ high in both types. In other
words, the nasal and teh reduplicated present share the same aorist type,
and in fact they have another rival in the y-presents. I suppose
durativity could be of different kinds, e.g. duration, repetition, or a
mere attempt.

> This also bears on the other proposal, going back to Kurylowicz at least,
> for the origin of (asigmatic) aorist forms: namely, these are old preterits
> that became limited to aorist function due to the rise of new presents.

If my observation that there is an alliance between the sk^-present type
and the s-aorist is correct, and the morpheme *-sk^e/o- is a development
of the expected *-s-ye/o- (implementing the "lengthening s" also known
from the nom.sg., which must earlier have been a different phoneme from
other sources of IE /s/), then the s-aorist was originally inchoative in
function. I'd say that makes very good sense, for the s-aor. is also
widely used with verbs that form radical or thematic presents, as
*weg^h-e/o- 'drive', whose aor. *we:g^h-s- will then originally have meant
'start driving, set out (by carriage)'.

> Incidentally, what are your thoughts on the origin of nasal presents? The
> rarity of infixes in PIE has led to the proposal that nasal presents
> originated from a double affix, that is *wined- was really *wi-n-ed- (or
> *wi-ne-d). This would mean that originally win(e)d- and w(e)id (or wide/o)
> were not grammatically associated. In case of dehmi lehmi etc, it then
> becomes a question of when the roots came into existence. I am not sure if
> we have enough information to decide this.

The double suffix theory has little to recommend it, and for some root
types the nasal-infix structure shows surprises that can only be
reasonably understood as the result of metathesis. Thus roots ending in
*-yH and *-Hy (the later being the "long diphthongs" like *dheH1y- 'suck')
both form nasal presents with the -y- before the nasal infix and the -H-
after it. The natural origin of an infix is of course either a prefix or a
suffix which got displaced and moved into the interior af an adjacent
morpheme by simple metathesis. The nasal infix is located before the final
root consonant, therefore it must have moved in from behind which makes it
an old suffix. I take it that a present-stem structure like *li-ne-{k}-,
*{kw}ri-ne-H2- and *dhi-ne'-H1- from *ley{kw}- 'leave', *{kw}reyH2- 'buy',
*dheH1y- 'suck' proceed from old structures in which /n/ was simply added
to the root, i.e.

   *ley{kw}-n-      *{kw}reyH2-n-     *dheH1y-n-,

and that these structures were subsequently adjusted to some more widely
"preferred syllable types" by simple metathesis, which apparently gave

   *leyn{kw}-       *{kw}reynH2-      *dheynH1-.

Then, as generally in IE, stems ending in three consonants were alleviated
by insertion of an /e/ before the last consonant (my own observation);
that gave:

    *leyne{kw}-     *{kw}reyneH2-     *dheyneH1-.

The new vowel took the accent, so that the strong forms acquired a span
/-ne'-/; on that basis, the weak form which had syllabic endings, moved
the accent one slot further, leaving /-n-/. The product was then:

3sg *li-ne'-{kw}-t  *{kw}ri-ne'-H2-t  *dhi-ne'-H1-t
3pl *li-n-{kw}-e'nt *{kw}ri-n-H2-e'nt *dhi-n-H1-e'nt.

> This brings me to a general question. There seem to be two camps about the
> category system of the PIE verb. One believes that the aorist-imperfect
> distinction, to be equated to perfective-imperfective distinction,
> ``always'' existed in PIE and Hittie lost this distinction, while Vedic
> changed things around. The other considers the aspectual distinction to
> postdate the separation of Anatolian.

There are these two camps, yes, and I am in no doubt that camp one is
right. There is no way the specific forms of the aspect stems could have
been formed secondarily in "the rest of IE" left after the exodus of (or
from) the Anatolians. At the very least, all the _forms_ must be assigned
to a protolanguage from which also Anatolian is descended. And what would
the forms be there for, if the functions that go with them only developed
later? We find practically all the IE verbal stems in Anatolian, many only
in a few lexicalized remains; are we to assume they have had totally
enigmatic earlier functions, and that they were later dug up by "the rest
of IE" and given totally new functions there? Clearly the only unforced
interpretation is that their functions in the common protolanguage were
the ones with which they are found where they do survive with a palpable
functional identity.

> There are certain nagging questions about the first thesis: The change in
> Vedic is not explained and how it came about without the prior loss of
> aspect has, AFAIK, not been explained. Those who adhere to this also feel
> the need to explain away as much as possible of root presents. But
> there are enough of them remain in Hittite and Vedic to raise doubts.

I don't follow - what change in Vedic are you talking about? Why would
anyone want to explain away root presents where they are securely
reconstructible? One would do that only to avoid having a language
combining a root-present with a root-aorist, for in that case the two
aspect stems are identical. That is why I am so sceptical about the
authenticity of the Vedic root presents lehmi and dehmi, because for these
verbs we have nasal presents in some other IE languages pointing to the
existence of a root aorist; thus leh- deh- look like displaced aorists.
But not so for eti 'goes' or asti 'is': these are durative verbs, and so
their unmarked form could function as a durative (socalled "present")
stem.

> Sigmatic aorists, as forming perfectives for root presents, present some
> problems too. There are roots that look like older roots extended with an s.
> This suggests that sigmatic aorists came from grammatization of a root
> extention and fits in with the general opinion that sigmatic aorist as a
> >grammatized< formation (as opposed to -s as derivational affix for forming
> new verbs) is late PIE.

If my wild guess analyzing the inchoative present suffix *-sk^e/o- as
**-z-ye/o- (with a special sibilant, whence later IE /s/) is correct,
there is no way this could belong to any period we would call "late". I
don't expect everybody to accept this idea, however (though I don't see
why they shouldn't).

> There is another question that typology suggests: There are basically two
> forms of perfectives (Dahl, ``Tense and Aspect systems; Bybee and Dahl, in
> Studies in Language, 13(1989) pp.51--103; Bybee et al ``The evolution of
> grammar''): The first seems to come from old preterits that become limited
> to perfectives due to the rise of new present/imperfective. The second is,
> as in Russian, from ``bounders'', originally adverbs that denote the
> attainment of a limit. In the first, aorist is unmarked and refers to action
> considered as a whole, even if it is atelic. In the second, the perfective
> does not fit that well with atelic events. The difference is said to be
> evident in Bulgarian which has both the aorist-imperfect distinction (of the
> first type) and perfective-imperfective distinction using prefixes (the
> bounders).

> Now, in PIE, some verbs seem to be like the first system (root aorist vs
> marked presents) while we cannot escape the presence of the second type as
> well (root present with sigmatic aorist). Sometimes Greek situation is used
> to argue that aorist looks older due to the number of irregularities. But in
> Sanskrit, as can be seen from looking at the lists in grammars, root
> presents show the most irregularity while aorists (and class 3, 7
> presents) are regular exemplars of internal sandhi however complicated the
> rules may seem. Again we are faced with the question of who is more archaic
> and who regularized.

Again, I do not think there is any problem in accepting root present as
original for inherently durative verbs, and root aorist as equally
original for inherently punctual verbs.

> One thing needs to be said about a different attempt to explain this, namely
> the traditional (in IE studies) equation durative=atelic=imperfective,
> momentary=perfective. Limiting perfectives to telic/momentary situations is
> decidely the minority option among languages. In Dahl's survey it was found
> only in Slavic, Finnish (based on acc vs partative), Japanese, Hindi,
> Mandarin, Bandjalang and Cebuano. It is even absent in aorist-imperfect
> distinction of Bulgarian. Positing this restriction for PIE requires
> appropriately strong evidence, not just pointing to Russian.

But the IE aorist is not restricted to any special kind of verbs - it is
only _unmarked_ (better: apparently originally unmarked) for inherently
punctual verbs; for other verbs the aorist need a morphological marking,
and the meaning is then some nuance that can be regarded as punctual
("started to -") or it just reports that the action got done (Meillet's
action pure et simple). One important functional point with the aorist,
however, is that it marks a turn of events which creates a new situation,
whereas the "present aspect" stays in the situation already given and
reports another action contributing to that situation. This is seen
remarkably well in the prohibitive use of the prs. vs. aor. injunctive, as
propounded so clearly by Hoffmann.

> The equation durative=telic is equally atypical. In such a language, one
> could not just say ``John walked home'' or ``John ate one piece of bread''.
> Instead one must say something like ``John walked, reached home'', or have
> two verbs for every transitive durative, one used only for past with
> definite objects, one used for past with indefinite/mass objects and for all
> presents. Again, we cannot just assume such a property, especially given
> that this does not seem all that common in contemporary languages.

Still, even verbs generally signifying completed action could form
duratives, indicating e.g. a repetition of the action (give one thing, and
then another) or an as yet unsuccessful attempt (I'm opening the window).

[Jens:]

>> for the functional change is quite small: it only takes the use
>> of the aorist form as an imperfect, then the rest follows by itself.

> The change of syntactic categories does not strike me as a small change. I
> would like to know more about incontrovertable cases of such changes before
> arriving at any conclusions.

We know that this kind of change was small enough in the languages here
concerned to lead to a number of misplaced aspect stems. E.g., the
Armenian aor. eber is an old ipf. Is the jump from "narrative past" to
"recent past" so great? If it is, even great changes happen.

Jens


From larryt at cogs.susx.ac.uk  Wed Jul 14 14:39:07 1999
From: larryt at cogs.susx.ac.uk (Larry Trask)
Date: Wed, 14 Jul 1999 15:39:07 +0100
Subject: accusative and ergative languages
In-Reply-To: <1999Jul10.021517@AN3039.spb.edu>
Message-ID: <WED.14.JUL.1999.153907.0100.>

On Sat, 10 Jul 1999 PIE at AN3039.spb.edu wrote:

[LT]

>> But Dixon also cites examples of ergative constructions which, in his
>> view, have very clearly *not* developed from passives, but from other
>> constructions.  His examples are Hittite and Pari.

> Pardon my ignorance, but i wonder what Hittite constructions can be
> claimed as ergative? Could you clarify Dixon's opinion?

> Does he mean the tendency to avoid using inanimates as 1st actants of
> two-argument predicate and to transform them to animates in -nt-, like
>
>     eshanants inan karapzi,
>     witenants eshar parkunut,    etc?

Dixon appears to be relying entirely on the views of Garrett, as
presented in this article:

A. Garrett (1990), `The origin of NP split ergativity', Language 66:
261-296.

It would appear that Garrett argues for an instrumental origin for the
Hittite construction he (she?) interprets as an ergative.

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk


From obaumann at stud.uni-frankfurt.de  Thu Jul 15 15:51:48 1999
From: obaumann at stud.uni-frankfurt.de (Oliver Baumann)
Date: Thu, 15 Jul 1999 17:51:48 +0200
Subject: FYI: Cimbrish course
Message-ID: <THU.15.JUL.1999.175148.0200.>

Dear members,

We want to inform you that a course in Cimbrish, a very old westgermanic
language spoken in Norther Italy til today, will be held next week on
our mailing list. If there is interrest to submit or to take part,
please inform us with some words:
Cimbrian-List-request at em.uni-frankfurt.de

If there is further interrest in that seriously endangered language,
see:
http://www.diens.de/Zimberland

Gildo Bidese and Oliver Baumann

PS: Sorry for cross posting


From petegray at btinternet.com  Wed Jul 14 18:48:34 1999
From: petegray at btinternet.com (petegray)
Date: Wed, 14 Jul 1999 19:48:34 +0100
Subject: accusative and ergative languages
Message-ID: <WED.14.JUL.1999.194834.0100.>

Pat said:

> What I consider 'internal inflection'  is Arabic yaktubu, kataba, katibun,
> etc. which applies to any and all verbs, old, and those taken new into the
> language.

Like it or not, in the pattern "sing, sang, sung" there is inflection, and
it is internal.   If you mean "applying consistently to all verbs", that's a
different matter.   But "sing, sang, sung" woudl still be internal
inflection, even if it were the only surviving example.

I am quite happy if you want to say that this pattern is no longer
productive in the language.

Peter


From petegray at btinternet.com  Wed Jul 14 19:01:30 1999
From: petegray at btinternet.com (petegray)
Date: Wed, 14 Jul 1999 20:01:30 +0100
Subject: accusative and ergative languages
Message-ID: <WED.14.JUL.1999.200130.0100.>

Leo said:
> Germanic ablaut *never* results from vowel harmony.  You're thinking of
umlaut,

Of course.   Because of the nature of the discussion, I simply didn't bother
to add "but here it is more often called umlaut"

Peter


From proto-language at email.msn.com  Fri Jul 16 03:17:46 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Thu, 15 Jul 1999 22:17:46 -0500
Subject: PIE vs. Proto-World (Proto-Language)
Message-ID: <THU.15.JUL.1999.221746.0500.>

Dear Joat and IEists:

 ----- Original Message -----
From: <JoatSimeon at aol.com>
Sent: Monday, July 12, 1999 11:09 PM

> In a message dated 7/12/99 9:28:59 PM Mountain Daylight Time,
> proto-language at email.msn.com writes:

>> I simply do not understand why some find it difficult to understand that
>> reconstructing the Proto-Language is only primarily different from
>> reconstructing Indo-European in the wider selection of source languages
>> for data.

Joat responded:

> -- Temporal distance.  Loss of information.  Entropy.

> A distinction between 5,000 years of unrecoverable loss and 50,000 years
> of unrecoverable loss.

> Words _vanish_.  A certain percentage of vocabulary just ceases to exist
> in every century.

Pat responds:

Nice theory! But the rate of vocabulary replacement is so variable (not to
mention punctuated equilibrium) that it cannot be used to exclude possible
retention of words for periods longer than 5,000 years.

What is shameful is that when someone does show *regular* similarities, that
perforce exceed that date by far, bogus reasons for dismissing them are
employed: like Larry's favorite onomatopeia, or expressive, or fantastic
loan scenarios.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From kurisuto at unagi.cis.upenn.edu  Fri Jul 16 04:37:43 1999
From: kurisuto at unagi.cis.upenn.edu (Sean Crist)
Date: Fri, 16 Jul 1999 00:37:43 -0400
Subject: Recoverability
In-Reply-To: <Pine.SO4.4.05.9907131322380.11749-100000@rask>
Message-ID: <FRI.16.JUL.1999.003743.0400.>

On Tue, 13 Jul 1999, Jens Elmegaard Rasmussen wrote:

> But that is not what you do in cases where only modern languages are
> available: You do not pick two and ignore the rest, you take all there is.
> In the case of IE, you would take _all_ modern Germanic languages, if need
> be with their dialects; you take all Slavic languages, you don't forget
> about Lithuanian and Latvian or Modern Greek which gives quite a lot;

[...]

> This is a LOT - and
> NOBODY could even thgen be in doubt that this is indeed a family of
> related languages sprung from a common source. True, we would still be in
> doubt or indeed ignorant about many a finer point which is only added by
> languages of old texts - and of course even they do not contain
> everything, so doubt will remain - but the general utline and subgrouping
> of the family would stand firm even on the sole basis of modern languages.

As far as this goes, it is quite right.  However, I don't think the point
was that we ought to actually try to attempt to reconstruct PIE on the
basis of only modern English and Albanian, or otherwise arbitrarily limit
our selection of data.

I think the point was something more like this.  Suppose for the sake of
argument that there is in fact a genetic relationship between
Proto-Indo-European and Proto-Uralic (to take one possible example). Even
if this were true, and given the evidence which we actually do have at
hand, would we expect to be able to _show_ that there is such a relation?

Here, the problem _is_ like trying to reconstruct PIE on the basis of only
modern English and Albanian.  Even tho external evidence might lead us to
guess that there could be a relation between PIE and PU, the cognations
are so obscured by millenia of sound changes, loss of old lexical items,
the noise of new lexical items, etc., that it might well be impossible to
show a genetic relationship on the basis of the available information.

Suppose, for example, that PIE *d corresponds to PU *z (I don't know that
there is a *z in PU; I'm making this up).  Even if there were originally
dozens of words showing this correspondence, it might well be that case
that only one such pair exists between the small subsets of the original
lexicons of PIE and PU which we can reconstruct.  Given the fundamental
assumptions of the Comparative Method, you _can't_ show that the two words
are cognate when you've only got one example of the correspondence.

You need multiple examples of sound correspondences to be able to conduct
the Comparative Method at all; and when the the cognates become as
rarefied as they are this time-depth, the likelihood of having access to
an adequate number of examples to work out the relevant sound changes
becomes proportionately smaller, eventually reaching what for practical
purposes is an impossibility.

  \/ __ __    _\_     --Sean Crist  (kurisuto at unagi.cis.upenn.edu)
 ---  |  |    \ /     http://www.ling.upenn.edu/~kurisuto/
  _| ,| ,|   -----
  _| ,| ,|    [_]
   |  |  |    [_]


From cz at magnet.at  Fri Jul 16 04:42:44 1999
From: cz at magnet.at (Clemens Zeitlhofer)
Date: Fri, 16 Jul 1999 06:42:44 +0200
Subject: Re(2): indoeuropean/hand
Message-ID: <FRI.16.JUL.1999.064244.0200.>

Indo-European at xkl.com writes:
>acus, domus,
>cornus, idus, and tribus (as well as manus) are the ones that come to
>mind.

Don't forget porticus :-)

Just wanted to say hello to everybody (I'm a newbie).

Regards,

c.
********** ********** **********
Clemens J. Zeitlhofer
1096 Wien, Postfach 181
cz at magnet.at

Odi profanas litteras et arceo
********** ********** **********


From X99Lynx at aol.com  Fri Jul 16 06:10:28 1999
From: X99Lynx at aol.com (X99Lynx at aol.com)
Date: Fri, 16 Jul 1999 02:10:28 EDT
Subject: Chronology of the breakup of Common Romance/CELTS
Message-ID: <FRI.16.JUL.1999.021028.EDT.>

I wrote with regard to "Vlakhi", 'Vlachi', etc.:

<<Appearing as "Walh" or "Walah" in OHG, it has interpreted
as meaning "foreigner", sometimes Roman, but in usage it is closely
associated with Celts and regions of Celtic habitation - e.g., "Wales",
"Walloon".

In a message dated 7/13/99 6:16:38 PM, rmccalli at sunmuw1.MUW.Edu replied:

<<You'd have to account for the ending in Vlakh, WLochy, etc.>>

The -kh, -chy endings - as adj. affixes - appear to reflect variants anywhere
the Celt connection was made. E.g., 'Wilisc', 'Wylisc', 'Wielisc' (W. Saxon
7th Cent.) = Celt or Briton. Also, 'Waleis' (Anglo-French)> 'Wallace,' OHG
'walhisc,' 'walesc' = Roman, French, O.E. Chronicle 'wilsc', bryt-'wylsc'
(refering to the language).  By 1100, the forms 'Wylisca cing' and 'Wylsca
biscop' occur among the English. Also, 'Wallych', 'Walsche' and of course
'Welch'.  In MHG, 'walsch'.  Closer to the east, 'Valskr' (ON), (MSw) =
Gauls, Frenchmen.  The various Slavic endings I believe are in line with all
this.  (cf, B-T-W, 'Cornwall', 'Wellfleet', etc.)

The OED calls 'Vlach' a Slavic borrowing from the Germanic above.

I also have this note from the Hungarian scholar who annotated the Gesta
Hungarorum (1975 ed) re the Pannonia re a bad date in that chronicle:  "An
explanation of this erroneous record may have been the Slavic tradition which
held that the Hungarians ousted the Franks - the Volochs in the language of
the Slavs - from the area of the Danube."

Here is another indication that the term "vlach" was originally directed to
the west:  "The Slavic ethnic name of the 10th century, Vlach, was, in its
plural
form: vlasi, borrowed by the Hungarian language with the sense
'Neo-latin, Italian, French', in the form olasz(i); in the period of
Arp?d, this was the Hungarian name of the Neo-Latin peoples, thus also
of the French, i.e., the Franks (cf. Latin Frankavilla > Hungarian
Olaszi) and this is also at present the Hungarian name of the Italians."

It is interesting to note that the Franks themselves considered their
original homeland - according to their official historian, Gregory of Tours -
not Germania, but Pannonia.  And when Constantine II founds a military colony
in Moesia at present day Fenehpusta, it is called 'Valcum'.  It will later be
the site of a Frankish fortress destroyed by the Maygars.

I wrote:
<<I seem to recall an suggestion that there may be a tie here to L.
vallum (fortified wall, the earliest meaning of "wall") and refer to the
Celtic or Romano-Celtic oppidum or walled town.>>

This was based I believe on L. 'vallatii', entrenchers, cf 'valla^tio, onis',
an entrenchment, from 'vallo -a^vi, -atum -' (see Cicero, Tacitus and Pliny)
to surround with a rampart.  Nothing I'm sure of, in the least.

<<It has long been proposed that the name Volcae, a Gaulish tribe,
was the source of Anglo-Saxon Wealh, Balkans Vlakh, Polish WLochy, etc.
The point I've generally seen is something to the extent that
supposedly the Volcae were the first Gaulish tribe to fall under Roman rule.>>

The BIG problem with this is that the Romans NEVER call the Celts/Gauls in
general 'Volcae' or anything quite like that.  Caesar does tell us the Volcae
Tectosages invaded Germania, so the contact between this group of Celts and
Germanics was probably direct.  Which may account for why the Romans did not
use the word as a name for other Celts, but the Germans perhaps did.  (B-T-W,
the first time anything like the word 'Volc/-' appears in Latin, it is in
connection with the Etruscans, whose town 'Volci' or 'Vulci' falls to the
Romans in 280 BC.  Velitrae is the town first associated with the Celtic
Volscae.)

<<My guess is that any Celtic element in the Balkan would have been
extremely thin.  The Celts raided the Balkans and stormed through the region
on the way to Galatia but my understanding is that they lived north of the
Balkans
in Galatia and Pannonia.>>

If the Danube is the northern line of the Balkans, then the Celts were right
on it all the way to the Black Sea.  In fact, they seem to be the only ones
on it that founded towns - Gorsium (attested to having been founded by Celtic
Eraviscans), Sungidium (Belgrade) Siscia (at the mouth of the Odra), and many
other sites show that originally Celtic settlers preceded Roman occupation.
Konjic, not far from Sarajevo, decribed as "Illyrian-Celtic" yielded evidence
of centuries of extensive Celtic occupation before the Romans came and some
of these were still present in the 1st Century AD.  Recently I saw a
catalogue with pages of Greek coins found in Thrace (2/1 century BC) that
were described as Celtic imitations.  And around 80 BC, when the Dacians
under Burebista make their move towards Greece and the Black Sea, they are
constantly warring with the Celts.  Even on the north bank of the Danube in
Dacia before the Roman withdrawal, there is still a Celtic presence in the
archaeology.

It seems that in this area, as in France, a significant Celtic cultural
presence seems to disintegrate when the Romans shows up.  And as in France
the obvious conclusion is that the Celts became Romanized on a wholesale
basis.  And one would think therefore there would be some substrate in the
remaining languages in this area.  (Even given the difficulty of finding such
substrates in everything from English and French to Spanish and Bohemian.)

It is odd, isn't it, that Celtic culture left such an 'thin' linguistic trail
across so much territory where its presence was not just attested, but even
dominant?

Regards,
Steve Long


From JoatSimeon at aol.com  Fri Jul 16 16:53:47 1999
From: JoatSimeon at aol.com (JoatSimeon at aol.com)
Date: Fri, 16 Jul 1999 12:53:47 EDT
Subject: PIE vs. Proto-World (Proto-Language)
Message-ID: <FRI.16.JUL.1999.125347.EDT.>

>ECOLING at aol.com writes:

>Of course it is going to get RELATIVELY more difficult to reconstruct to
>greater time depths.

-- true.  And 50,000 years is a whole different ball game compared to 5000
years.

ALL of that extra time-distance is pre-literate, which means no records of
what was going on.  Our picture of PIE would be very different, and much less
detailed, if we had no written records of extinct languages.

Hell, the original insight that the IE languages were derived from a common
ancestor was made by comparting Sanskrit with Classical Greek and Latin --
all extinct languages!

The link between Sanskrit and Latin or Classical Greek is so close that any
layman can see it; whole phrases are nearly identical.

This is, to put it mildly, not the case between, say, English and Bengali.

>If there are sound-symbolic ideal forms for lexemes having certain meanings

-- with minor exceptions (such as "kuku"), words _are_ arbitrary sound
assemblages.


From larryt at cogs.susx.ac.uk  Fri Jul 16 15:35:10 1999
From: larryt at cogs.susx.ac.uk (Larry Trask)
Date: Fri, 16 Jul 1999 16:35:10 +0100
Subject: accusative and ergative languages
In-Reply-To: <001f01becdcd$786928e0$309ffad0@patrickcryan>
Message-ID: <FRI.16.JUL.1999.163510.0100.>

Pat Ryran writes:

> Now, when Larry recently quoted Dixon about the nature of the
> ergative, he conveniently neglected to mention that Dixon
> acknowledged that there were currently practising linguists --- not
> amateur linguists like myself --- still defending the passive
> interpretation of ergative constructions.

No; this is not so.  Read p. 189 of Dixon's 1994 book.

> I asked Larry where he had "shredded" this interpretation, and to my
> knowledge, got no answer. If I missed the "shredding", perhaps you
> will be kind enough to rehearse his performance for us. I have seen
> nothing by Larry's vehemence and your allegiance to support the idea
> that the ergative should not be interpreted as a passive.

Mr. Ryan, I'm afraid I lack the time to reply in detail to all, or even
most, of the 37 or so postings from you that greet me each morning.

But briefly: if a transitive sentence in an ergative language were
"really" a passive, then its absolutive NP would be the subject and
would exhibit subject properties.  But this not the case in most of the
ergative languages I have heard about.  In spite of the ergative
case-marking and/or verbal agreement, it is usually the *ergative* NP
which exhibits subject properties, not the absolutive NP.

Take any ergative language you like.  Assume that the subject of an
intransitive sentence is the sole (absolutive) NP in it.  Examine the
syntactic properties of that NP, and tabulate them.

Then look at transitive sentences.  Tabulate the syntactic properties of
the absolutive and ergative NPs, and compare these with the preceding.

In the great majority of ergative languages, it is the ergative NP which
shares the subject properties of the intransitive subject.

A few examples of typical subject properties:

The subject controls reflexive and reciprocal NPs.

The subject cannot itself be reflexive or reciprocal.

The subject controls the empty NP in an empty-NP complement.

The empty NP in an empty-NP complement is itself a subject.

A subject can be coordinated with another subject.

And so on.  There are also various language-specific tests.  For
example, in some varieties of Basque, the object of a gerund (but not
the subject of a gerund) goes, exceptionally, into the genitive case.
Neither the absolutive subject of an intransitive gerund nor the
ergative subject of a transitive gerund can be genitivized, but the
absolutive object of a transitive gerund can be.

One more test from Basque.  In Basque, the subject of an intransitive
sentence cannot be reflexive or reciprocal.  So, Basques can say,
literally, `Susie and Mike [ABSOLUTIVE] were talking to each other', but
they cannot say *`Each other [ABSOLUTIVE] was talking to Susie and
Mike'.

Now, in a transitive sentence, Basques can say `Susie and Mike
[ERGATIVE] slapped each other [ABSOLUTIVE]', but they cannot say *`Each
other [ERGATIVE] slapped Susie and Mike [ABSOLUTIVE]' -- or, in terms of
the discredited passive theory, they cannot say *`Susie and Mike
[ABSOLUTIVE] were slapped by each other [ERGATIVE]'.

In an intransitive sentence, the absolutive subject cannot be reflexive
or reciprocal.  In a transitive sentence, the ergative subject cannot be
reflexive or reciprocal, but the absolutive object *can* be.

All tests for subjecthood in Basque give the same result: it is the
ergative NP in a transitive sentence which shares the subject properties
of the absolutive NP in an intransitive sentence, and therefore it is
the ergative NP, and not the absolutive NP, which is the subject of a
transitive sentence.

There are unusual languages, of course.  In Dyirbal, and in
Nass-Gitksan, the absolutive NP in a transitive sentence shows at least
some subject properties, while the ergative NP does not show the same
properties -- though it does show other subject properties.  This is the
phenomenon we call `syntactic ergativity'.  But I know of no language
which is syntactically ergative without exception, and I know of few
languages which are syntactically ergative at all.

The "passive" view of ergative languages in general is indefensible.

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk


From Georg at home.ivm.de  Fri Jul 16 09:28:37 1999
From: Georg at home.ivm.de (Ralf-Stefan Georg)
Date: Fri, 16 Jul 1999 11:28:37 +0200
Subject: PIE vs. Proto-World (Proto-Language)
In-Reply-To: <6fa43e91.24bc160a@aol.com>
Message-ID: <FRI.16.JUL.1999.112837.0200.>

[ Moderator's note:
  The following quoted material is from a message by <JoatSimeon at aol.com>.
  Just keeping things straight.
  --rma ]

>In a message dated 7/12/99 9:28:59 PM Mountain Daylight Time,
>proto-language at email.msn.com writes:

>>I simply do not understand why some find it difficult to understand that
>>reconstructing the Proto-Language is only primarily different from
>>reconstructing Indo-European in the wider selection of source languages for
>>data.

>-- Temporal distance.  Loss of information.  Entropy.

>A distinction between 5,000 years of unrecoverable loss and 50,000 years of
>unrecoverable loss.

>Words _vanish_.  A certain percentage of vocabulary just ceases to exist in
>every century.

And, if I may insert this much to Pat's chagrin, the notion that human
language is a monogenetic phenomenon is aprioristic ideology. Even the
notion that every known language, as Basque, Burushaski or whatnot, has to
be related to some other language is ideology. "Having consonants" may be a
universal feature of human language, "Being related to some other lg."
simply is not.

Stefan Georg
Heerstrasse 7
D-53111 Bonn
FRG
+49-228-69-13-32


From JoatSimeon at aol.com  Fri Jul 16 16:43:23 1999
From: JoatSimeon at aol.com (JoatSimeon at aol.com)
Date: Fri, 16 Jul 1999 12:43:23 EDT
Subject: accusative and ergative languages
Message-ID: <FRI.16.JUL.1999.124323.EDT.>

proto-language at email.msn.com writes:

> I also am certain that if we had reason to know that Albanian and English
>were descendants of a common language, we would eventually discover a kind
>of IE that would be inaccurate in some ways but essentially write.

-- no we wouldn't.  Sigh.

As time goes on, any given language gives us less and less information about
previous forms.  Information is _lost_.

Eg., no amount of analysis of Germanic will give you a PIE word for "bear"
because that word dropped out of that family.  All you'll get is variations
on "the brown one".


From larryt at cogs.susx.ac.uk  Fri Jul 16 15:06:31 1999
From: larryt at cogs.susx.ac.uk (Larry Trask)
Date: Fri, 16 Jul 1999 16:06:31 +0100
Subject: PIE vs. Proto-World (Proto-Language)
In-Reply-To: <6fa43e91.24bc160a@aol.com>
Message-ID: <FRI.16.JUL.1999.160631.0100.>

Pat Ryan writes:

> I simply do not understand why some find it difficult to understand
> that reconstructing the Proto-Language is only primarily different
> from reconstructing Indo-European in the wider selection of source
> languages for data.

Others have already made substantial points in reply, but I'd like to
add one more.

Before reconstruction can be attempted, there must be a persuasive
*prima facie* case that the languages under consideration are genuinely
related.  Our predecessors did not begin reconstructing PIE until they
had persuaded themselves that the evidence for a genetic link among the
IE languages was too massive to be ignored.

But you can't just pick some languages that catch your eye and then
"reconstruct" a common ancestor for them.  That is, you can't pick, say,
Zulu, Sumerian and Korean and "reconstruct" Proto-ZSK.  You must first
make a good case that a common ancestor for these languages is
preferable to any other scenario.

And the same goes for "Proto-World", or whatever.  Until a persuasive
case has been made that *all* of the world's 6000 or so known languages
are genuinely related, there is no point in attempting a
"reconstruction" of "Proto-World".  The result of such a rash attempt
can be no more than legerdemain.  Of course you can show that this, that
and the other *might* have a common ancestor, but you can do this in
countless entirely different ways, none of them superior to any other,
and you cannot show that these things really *are* related.

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk


From DFOKeefe at aol.com  Fri Jul 16 09:39:54 1999
From: DFOKeefe at aol.com (DFOKeefe at aol.com)
Date: Fri, 16 Jul 1999 05:39:54 EDT
Subject: PIE vs. Proto-World (Proto-Language)
Message-ID: <FRI.16.JUL.1999.053954.EDT.>

Good Morning I-E-ists,
          I was under the impression that I.E. Roots are pretty much
time-invariant, They can't be created, change, or get lost.  Words derived
from I.E. roots frequently do change, as do languages.  As a matter fact,
entire languages and language families shift in organized manner particular
to each.  But I.E. roots do not. Thus, I.E. roots don't decay, degenerate, or
go out of existence.  In the time span of recorded human history, entropy
does seem to not apply to I.E. roots.  Maybe to words, though.

Best regards,
David O'Keefe
Houston, Texas

[ Moderator's response:
  Your impression is mistaken.  In the reconstructed morphology of PIE, we see
  a theoretical construct which we call the root; this is the morpheme which
  carries the base meaning assigned to the word(s) in which it is found.  It is
  no more time-invariant that the reconstructed stem formants or endings in the
  same words.

  A root can only survive in the history of a language as long as a word which
  contains it.  There is no reflex in modern English, for example, of the root
  *yebh- "fuck", although it is still alive and well in modern Russian (to the
  extent that the name of one IBM mainframe utility is a great joke).  There
  very likely were roots in IE of which we have no knowledge at all because
  they have not survived into *any* extant language in a manner which allows us
  to assume their existence.

  --rma ]


From X99Lynx at aol.com  Fri Jul 16 16:43:18 1999
From: X99Lynx at aol.com (X99Lynx at aol.com)
Date: Fri, 16 Jul 1999 12:43:18 EDT
Subject: The Rumanian Question (was: Chronology of the breakup of Common
 Romance)
Message-ID: <FRI.16.JUL.1999.124318.EDT.>

In a message dated 7/13/99 6:16:38 PM, rmccalli at sunmuw1.MUW.Edu wrote:

<<Graham Mallinson, "Rumanian," 391-419. Martin Harris and Nigel
Vincent, eds. The Romance Languages. London: Croom Helm, 1988.

	There is very little substrate vocabulary in Rumanian, and its
origin is unknown. Since the grouping of Dacian, Thracian and Illyrian
within IE is uncertain, it is questionable whether Albanian and Rumanian
shared a substrate language, although they do have cognates:>>

The question of the origins of the Rumanian language have involved historical
linguists and other historians in some pretty sensitive political issues in
southeast Europe.  The matter of the breakup of Hungary and Romanian national
identity have been debated on historical grounds for some time now.  (Whether
such things should really be relevant is another matter.)

One controversy is the question of "Daco-Roman Continuity."  Essentially,
this revolves around whether Romance was always spoken in Romania and
represents a continuation of what began as the Roman colonization of the
province Dacia in the first centuries of the current era.  At first glance,
this would seem logical enough and it had been the official line of most
historians and linguists for a long time.  But that conclusion has been put
into question from time to time and lately some substantial arguments have
been made to the contrary.

In 1996, Andre duNay published "The Origins of the Rumanians: the Early
History of the Rumanian Language." (Matthias Corvinus, Toronto).  It should
be of special interest to the members of this list for its methodology which
attempts to synchronize an apparently exhaustive mass of past linguistic
research with historical data.

Of particular interest is the way duNay goes beyond mere lexical comparisons
(such as those reproduced by rmccalli from Mallinson and Rosetti in the prior
post) and considers the chronology of syntactical developments, innovations
and changes in Latin, Southern Slavic, Greek, Albanian and Romanian - as well
as archaeology and primary historical records - to reach conclusions about
when linguistic events actually occurred.

duNay addresses a long list of linguistic theories regarding Rumanian
language and does a fairly compelling job of dissembling the Daco-Roman
continuity theory.  Particularly compelling is the linguistic and historical
evidence that Rumanian developed in close proximity to Albanian in the
southern Balkans.  The syntactical and other characteristics by which duNays
identifies Rumanian as a "southern Balkan" language joined with Bulgarian,
Albanian and Greek should again be of particular interest to members of this
list from the point of view of methodology.

The book BTW is reproduced in its entirety on the web, on the Hungarian web
site (for reasons that become more apparent as one gets into this.)  I will
get the URL and post it if there's interest.

I don't know if one must conclude that the Albanian or Proto-Albanian
substrate in Rumanian settles the issue of origins.  I do know that I
appreciate the historicity of this work very much - particularly because it
shows how silly statements like this one are:

<<A. Rosetti: "The Thracians, known as G?tes by Greek writers and Dacians by
Latin writers..." >> (This is the type of bad historical summary that must
pollute any understanding of the situation for linguists relying on it.)

There are forwards in the book by Robert A. Hall at Cornell and Adam Makke
that are also interesting, including the following statement:  "Lack of
knowledge of the Slavic languages,...hindered the Rumanian historians from
expressing the problem clearly and recognizing its significance in the study
of the beginnings of the Rumanian people."

This may explain a bit of the 'basic word' substrate mentioned in the prior
post.  E.g.,:

<<Albanian			Romanian
buz? "lip, edge"		buz? "lip, edge">>

See 'pusa' (Czech), 'buzia' (Pol.) which I believe follow the variances
expected between Romanian and Slavic found in other -a^ > -ia forms.

In fact, the following makes me think that maybe we are dealing with a
transfer of technology here, rather than proof of common ancestry:

<<The total number of words of significance for the everyday life of
shepherds among these  89 words common to Rumanian and Albanian is at least
60  - 66 % of all.  Of the rest, most denote basic human notions and
conditions, such as  parts of the human body and family relations. >>

Hope this is interesting or helpful.

Regards,
S. Long


From edsel at glo.be  Fri Jul 16 10:04:21 1999
From: edsel at glo.be (Eduard Selleslagh)
Date: Fri, 16 Jul 1999 12:04:21 +0200
Subject: indoeuropean/hand
Message-ID: <FRI.16.JUL.1999.120421.0200.>

[ moderator re-formatted ]

-----Original Message-----
From: Steven A. Gustafson <stevegus at aye.net>
Date: Friday, July 16, 1999 6:07 AM

[Snip]

>There are a fair number of Latin -u stems that are feminine, especially
>considering how few -u stems there are to begin with: acus, domus,
>cornus, idus, and tribus (as well as manus) are the ones that come to
>mind.

>Steven A. Gustafson

[Ed]

Actually it's 'cornu' in the nominative.

All tree names, also those with -u stems like 'quercus' (oak) (and -o stems
like po:pulus) are feminine.  Could there be some relationship here?  I mean
'extremities, branches...'.  That would also account for manus, acus, cornu and
tribus being feminine, but not for domus nor idus.  'Foot, pes...' are probably
not viewed as extremities, but as 'base' to stand on.

Ed.


From JoatSimeon at aol.com  Fri Jul 16 17:42:45 1999
From: JoatSimeon at aol.com (JoatSimeon at aol.com)
Date: Fri, 16 Jul 1999 13:42:45 EDT
Subject: Momentary-Durative
Message-ID: <FRI.16.JUL.1999.134245.EDT.>

>petegray at btinternet.com writes:

<< The Armenian-Greek- I-I sprachbund has certain known
>innovations, such as vocabulary and the augment, and we should not be miseld
>into accepting as original what may be another innovation, namely the
>development of new distinctions based on an existing variety of forms. >>

-- very good point.

A great pity the Linear B sources aren't more extensive.


From proto-language at email.msn.com  Fri Jul 16 13:25:32 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Fri, 16 Jul 1999 08:25:32 -0500
Subject: accusative and ergative languages
Message-ID: <FRI.16.JUL.1999.082532.0500.>

Dear Ralf-Stefan and IEists:

 ----- Original Message -----
From: Ralf-Stefan Georg <Georg at home.ivm.de>
Sent: Tuesday, July 13, 1999 5:22 PM

>> Pat:

>> Yes, but if a new verb were formed, e.g. webben (from English 'web'), there
>> is no possibility that it would be conjugated webben, wabb, gewobben.

>> What I consider 'internal inflection' is Arabic yaktubu, kataba, katibun,
>> etc. which applies to any and all verbs, old, and those taken new into the
>> language.

R-S:

> The pattern lost productivity in German. So what ?
> Also, I hardly understand your claim, which basically amounts to refusing
> to speak of technique X in language L, unless it isn't present in each and
> every item in which it theoretically could be present. So, you would refuse
> to speak of, say, vowel harmony in, say, Turkish, because some borrowed
> (and some native) words fail to obey this rule ? ("What I call VH is ...,
> which applies to any and all ... Period. There is an exception, or the
> phenomenon is gradually losing ground, therefore it doesn't extst at all."
> ?????).

Pat:

The distinction I am trying to make, and you are certainly not required to
make it also, is that the pattern "e-a-o" *never* was really "productive" in
German although one might term it "resultative". There was never a time
during which *any* verb in [e] followed this "pattern", or even *any* verb
in /eR/ followed it (weNden, waNdte, gewaNdt).

 Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From proto-language at email.msn.com  Fri Jul 16 13:05:19 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Fri, 16 Jul 1999 08:05:19 -0500
Subject: PIE vs. Proto-World (Proto-Language)
Message-ID: <FRI.16.JUL.1999.080519.0500.>

Dear Lloyd and IEists:

 ----- Original Message -----
From: <ECOLING at aol.com>
Sent: Tuesday, July 13, 1999 10:51 AM

> Patrick Ryan wrote:

>> I simply do not understand why some find it difficult to understand that
>> reconstructing the Proto-Language is only primarily different from
>> reconstructing Indo-European in the wider selection of source languages
>> for
>> data. Of course, secondarily I have labored to reconstruct the underlying
>> monosyllables by analysis of attested compounds.

Lloyd Anderson wrote:

> I strongly support work on earlier language families not part of the
> standard doctrine today.
> Against those who say (very nearly) that everything which
> can be discovered has already been discovered.
> That is of course a charicature, but with some truth to it.
> (Or they say this while specifying "using the comparative method",
> and defining it circularly to mean only the existing tools,
> and only those ways of using those tools,
> which are well known today.)

> However, the above statement by Patrick Ryan I find highly surprising.

> Of course it is going to get RELATIVELY more difficult
> to reconstruct to greater time depths.
> The mistake of those who reject all Nostratic and similar work
> is in drawing a sharp fixed line,
> saying that short of that time depth it can be done,
> beyond that time depth it cannot be done.

> The mistake of Patrick Ryan in the quotation above is to neglect
> that it does get substantially more difficult as the time depths increase,
> or in particular language families, because of the specific nature
> of the sound-changes and grammatical changes which occurred,
> or etc.

Pat responds:

First, let me say that I agree substantially with everything you have
written in this posting. But I really think the "difficult"  part is
correctly factoring in what may be sound-symbolic influences.

Lloyd Anderson continued:

> There is no sharp break.
> There is a gradually increasing difficulty with time depth and
>     with depth of changes (the two correlated but not the same concept).
> There is ample room for expanding the tools used,
> and for empirical studies of what sorts of changes each tool is
> capable of penetrating beyond, and to what degree.

> To give merely one example, one factor, sound symbolsim:

> If there are sound-symbolic ideal forms for lexemes having
> certain meanings,
> then the more arbitrary ones (farther from those sound-symbolic
> ideal forms)
> have greater value as evidence for historical connections
> of specific languages or language families,
> simply because they are less likely to have resulted from
> later pressure towards the sound-symbolic ideal.

> This is of course a terribly difficult circularity,
> because it means that look-alikes which are more widely attested

Pat interjects:

Here I think you are dangerously introducing the mistaken terminology of the
opposing argument. I do *not* look for "look-alikes"; I only am interested
in "cognates" the phonological forms of which can be supported through
multiple comparisons. For example, one set of interesting IE and Sumerian
"cognates" shows IE *-wey- = Sumerian -g{~}-; interestingly, this
development *is* found in *some* IE languages, like nearby Armenian.

Lloyd Anderson continued:

> for a given meaning or set of related meanings
> may be EITHER relics of an earlier historical unity
> (whose changes were perhaps ALSO retarded by sound-symbolic forces),
> OR the results of pressure towards some sound-symbolic ideal forms,
> from diverse original and unrelated forms.
> Much more subtle and difficult reasoning is therefore needed to
> establish what are results of sound-symbolism and what are results
> of historical common origins.
> EVEN when we have a suprisingly widespread statistical sound-meaning
> correlation.

> The usual procedure is also circular.
> Simply taking a sample of purportedly unrelated languages
> and attempting to determine how many look-alikes word lists contain
> is a bit naive,
> because the languages may not be totally unrelated,
> because chance resemblances may be more common in certain
> meaning or sound ranges,
> because the biases of different types of sound-system structures
> are not yet well handled,
> and for many other reasons.

Pat concludes:

Lloyd, I think the real hinge of this question is how one defines "loss";
and, after calculating "losses", what kind of percentages one arrives at for
"vocabulary loss". Having worked extensively with language-family
comparisons, my statistically unsupported "guess" is that there is amazingly
little vocabulary loss if one allows reasonably semantically-shifted pairs
to be counted as "retained".

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From Georg at home.ivm.de  Fri Jul 16 21:54:08 1999
From: Georg at home.ivm.de (Ralf-Stefan Georg)
Date: Fri, 16 Jul 1999 23:54:08 +0200
Subject: accusative and ergative languages
In-Reply-To: <001f01becdcd$786928e0$309ffad0@patrickcryan>
Message-ID: <FRI.16.JUL.1999.235408.0200.>

>All very well and good. Having explained next to nothing, perhaps you will
>now tell us how precisely *you* were using "passive in nature" and what in
>God's name it is supposed to mean to you, to those you may be quoting, or
>anyone else including me.

It means nothing to me, it was sloppy terminology and has nothing
whatsoever to do with the things we are talking about. I think I made that
clear. It was my incompetent rephrasing of "ergative constructions are, or
invariably come from, passive constructions", a position not endorsed by
me, but apparently by you.

>Pat responds:

>Yes, you are absolutely correct. This is a familar pattern indeed. When
>challenged for an argument, you sidestep the issues of the question by
>conveying that someone you like (here, Larry Trask) "has spoken the definite
>word about this here". If Larry could speak the definite word about
>everything, then this list would be a waste of time. We could just subscribe
>to his newsletter for the latest ex cathedra rulings on all our troublesome
>questions.

Sorry, I've not implied that Larry Trask is the pontifex maximus
ergativitatis, I only stressed that he, in my humble opinion, laid the
issue of "ergative is passive" to rest, with arguments sufficient enough to
show this, and nothing more. That my own position on ergativity happens to
coincide with this and, admittedly, informed my reading of these postings,
is of course instrumental in this. Not who says something matters, but how
well it is argued; believe me or not, I'd even take it from you, if you
only ever said such a thing.

>Now, when Larry recently quoted Dixon about the nature of the ergative, he
>conveniently neglected to mention that Dixon acknowledged that there were
>currently practising linguists --- not amateur linguists like myself ---
>still defending the passive interpretation of ergative constructions.

This is uninteresting. I know practising linguists I wouldn't buy a used
car from. From some I wouldn't buy a *new* car. Who is it and what are
their arguments ? Don't hide behind an anonymous bunch of practising
linguists.

>I asked Larry where he had "shredded" this interpretation, and to my
>knowledge, got no answer. If I missed the "shredding", perhaps you will be
>kind enough to rehearse his performance for us. I have seen nothing by
>Larry's vehemence and your allegiance to support the idea that the ergative
>should not be interpreted as a passive.

How do you live with the fact that some "ergative languages" have
independent passives ?

>That Ralf-Stefan is incapable of defining the term he introduced: "passive
>in nature". By the way, being a native speaker of English, I can assure you
>that in my dialect, "passive in nature" is *not* bad English.

Thanks a bundle for sparing me the one verdict I was really fearing ...

>And to answer your -- I hope not purposeful -- distortion of what I wrote,
>let me say explicitly that I did not assert "there are ergative languages
>without (any) splits". I asserted that Thomsen did not, at least in her
>grammar, identify splits in Sumerian, which you seemed to think she had. I
>then invited you to identify them in Sumerian if you *could*.

I did. I did, out of my die-hard habit, address every single point of your
objections to it, and I addressed them satisfactorily. If not, for heaven's
sake, we will have to go over it again, but I doubt that this time we will
have too many happy faces reading along ...

And, as far as I remember, this whole brouhaha started with me asserting
that all known "ergative languages" have some split, followed by you
prompting we to show such a thing in Sumerian. I'm not a mind-reader, but
this could be interpreted as a challenge aiming at the gist of my
assertion, n'est-ce pas ? I managed to show those splits, although the
discussion got a bit swamped under some hassle over maru:- and
HamTu-conjugations, but I managed to do it. At least I got you to accept,
late in coming, though, that my initial assertian still stands up.

>You might review your own procedures for employing quotation marks. Your use
>of them on "there are ergative languages without any splits" strongly and
>falsely implies that I wrote this in the context of a judgment on the
>question.

I accept the reproach of not being a mind-reader ...

St.G.

Stefan Georg
Heerstrasse 7
D-53111 Bonn
FRG
+49-228-69-13-32


From mcv at wxs.nl  Fri Jul 16 22:58:07 1999
From: mcv at wxs.nl (Miguel Carrasquer Vidal)
Date: Fri, 16 Jul 1999 22:58:07 GMT
Subject: Chronology of the breakup of Common Romance [long]
In-Reply-To: <007101becdef$66b57b40$6701703e@edsel>
Message-ID: <FRI.16.JUL.1999.225807.GMT.>

"Eduard Selleslagh" <edsel at glo.be> wrote:

>Note that the form with -ch is East-European, while the 'Wal-' form is
>Western, mainly Germanic and in loanwords of Germanic origin, e.g. in Latin
>languages (Walen-Wallons, Wallis-Valais, Wales...).

That's merely because the Slavic languages have retained the
"(c)h" in the Germanic loanword which was dropped in the modern
Germanic languages (OE wealh, OHG walh).

=======================
Miguel Carrasquer Vidal
mcv at wxs.nl


From jer at cphling.dk  Fri Jul 16 22:33:52 1999
From: jer at cphling.dk (Jens Elmegaard Rasmussen)
Date: Sat, 17 Jul 1999 00:33:52 +0200
Subject: Ergative & Basque
In-Reply-To: <Pine.GSO.4.03.9907131058110.17292-100000@rsunx.crn.cogs.susx.ac.uk>
Message-ID: <SAT.17.JUL.1999.003352.0200.>

On Tue, 13 Jul 1999, Larry Trask wrote:

> On Fri, 9 Jul 1999, Jens Elmegaard Rasmussen wrote:

>> [...] Can "Mutila jo zuen" not
>> *come from* something which *originally* meant, not 'the boy was
>> hit' pure and simple, but specifically 'the boy was hit by him
>> (e.g., by the one we're talking about)'? [...]

> [LT:]
> Well, this suggestion is possible, I suppose, but I know of no evidence
> to support it, while there is some evidence against it.  Let's look at
> the structure of the Basque sentence.

[JER:]
How can it be possible, if there is evidence against it? This is
getting intriguing now.

[LT:]
> Here <mutil> is `boy', and <-a> is the article.  This NP, being a direct
> object, stands in the absolutive case, which has case-suffix zero.

> The item <jo> is the perfective participle of the verb `hit'.

> And <zuen> is the auxiliary verb-form.  Now, third person is usually
> marked by zero in Basque verbs.  So, the usual absolutive agreement
> slot, which is the first slot in the auxiliary, cannot be filled by an
> agreement marker, and it is filled instead by <z->, a redundant marker
> of past tense.  The next element, <-u->, is a reduced form of the verbal
> root <-du->, from *<edun> `have'; this is the usual transitive
> auxiliary.  Again, the ergative agreement slot, which follows the verbal
> root, is empty, because the ergative (subject) NP is third-person.
> Finally, <-en> marks past tense.

> The whole thing is thus this:

> 	boy-the-Abs hit-Perf Past-have-0-Past

> Or, roughly, `(She/he/it) had hit the boy.'

> But it translates English `S/he hit the boy' (before today).  This is a
> periphrastic form comparable to the ones we find in Romance and
> Germanic.  Nobody knows if such periphrastic forms are calqued on
> Romance or are an independent development in Basque.

> Now, I can see little scope here either for an original passive
> interpretation of the form or for any way of reading `by him' into the
> auxiliary form.

[JER:]
If there is no passive expressed as belonging to the agent in this, what
then is the "have" verb doing here? Can't the underlying construction be
analyzed in a sensible way at all? A silly question from a complete
outsider: is the translation "have" internally motivated? Can it express
the "having" of anything other than a "participled" object?

>> [...] From the primitive and casual
>> books at my disposal I do see that "zuen" and "zuten" mean 'he had
>> him' and 'they had him' resp. I also believe I see that such
>> auxiliaries are combined with a particularly short form of the
>> participles, referred to by Schuchardt as the root of the
>> participle;

> [LT:]
> It is simply the perfective participle of the verb.  Most participles in
> Basque carry an overt suffix proclaiming them as such, but <jo> happens
> to be one of the exceptions: it has no participial suffix, though I
> suspect that it once did, and that the suffix has been lost by a
> combination of phonological change and analogical readjustment.

> [JER:]
>> and "jo" is 'stick; beat' in its shortest form, says my little
>> dictionary;

> [LT:]
> Actually, `hit', `strike', `beat' -- not `stick', which is a noun.

[JER:] In fact, I meant a different mistake, viz. the verb 'to stick,
adhere': I trusted the Spanish dictionary gloss "pegar" to express that
the Basque verb was as semantically broad as the Spanish which, looking in
the Sp.-B. part, I now see it isn't. Let's forget that.

>> and "mutil-a" is 'boy' with the article "-a", but without case or
>> number marking.

> [LT:]
> Not quite.  The form <mutila> is marked as absolutive by its zero
> suffix, and as singular by its singular article <-a>.  The plural
> article is <-ak>, and `the boys' is <mutilak>, in the absolutive.

[JER:] That's what I meant.

>> Therefore my persistent question: Why can't "mutila jo zuen" and
>> "mutila jo zuten" reflect a construction that was earlier meant to
>> express 'the boy, he had him hit', 'the boy, they had him hit'?

> [LT:]
> Well, I can't rule that out, but I can't see any evidence to support it.
> Note in particular that <mutila> exhibits *no* subject properties in
> modern Basque, or in Basque of the historical period.  If it ever was a
> subject, as this proposal requires, the reanalysis must have been
> carried to completion a long time ago.

[JER:] What do you mean mutila has no subject properties - is "the boy" as
the subject of an intransitive verb not precisely mutila? I'm not afraid
of "a long time ago"; I guess it's what we're trying to reach in diachrony
all the time.

[LT:]
> Note also that intransitive verbs are likewise conjugated
> periphrastically but with the intransitive auxiliary <izan> `be'.

> [JER:]
>> Schuchardt also gives "zen" to mean 'he was', so that if you gloss
>> "mutila jo zen" as 'the boy was hit', it seems there is quite a bit
>> of agreement that the verbal root is a participle by itself.

> [LT:]
> The lexical verb stands in the form of its perfective participle in all
> periphrastic past-tense forms, and also in all periphrastic perfects.
> The perfect form corresponding to <Mutila jo zuen> is <Mutila jo du>,
> which differs only in that the auxiliary is now present-tense.  This
> form translates both English `He has hit him' and English `He hit him'
> (earlier today) -- much as in European Spanish.

[JER:] This would fall into place if "have" itself contains "be": Is there
any possibility that the constructions "mutila jo zen" and "mutila jo du"
were originally designed to mean 'the-boy - beaten - he was' = 'the boy
was beaten', and 'the-boy - beaten - for/of-him-he-is' = 'the-boy beaten
he-has-him' = 'he has beaten the boy'/'he beat the boy'? Is there any
solid knowledge excluding such a prehistory?

>> I do not see in what way this makes the *diachronic* interpretation
>> of "mutila jo zuen" any different from the Hindi preterites that are
>> based on Sanskrit constructions of the type "A-Nominative +
>> B-Genitive + PPP/nom." meaning earlier "A was (verb)-ed by B", but
>> now simply "B (verb)-ed A." Where am I wrong?

> [LT:]
> Well, in the Indic case, we have several thousand years of texts to
> consult, so that we can get an idea how the ergative construction arose.
> With Basque, we are not so lucky.

[JER:] But the elements mean the same, one by one, and the semantic sum
total is the same, right? Does that not count for anything?

[LT:]
> Note also that, in Hindi, as in Indic generally (I think), the ergative
> occurs only in the past tense, as is common with ergatives that have
> arisen from perfective or passive constructions.  In Basque, however,
> the ergative construction is used in all circumstances, without
> exception.

[JER:] Looks like a very small difference to me. Can't languages stop at
different points in a process of generalizing a favoured structure? Who
knows that Post-Modern Indic won't introduce the ergative model in the
present tense some day?
   First and last, is translation of the "auxiliary 'have'" demanded by
the language itself? And even if it is, can "be" be contained in it?

Jens


From kurisuto at unagi.cis.upenn.edu  Fri Jul 16 20:15:03 1999
From: kurisuto at unagi.cis.upenn.edu (Sean Crist)
Date: Fri, 16 Jul 1999 16:15:03 -0400
Subject: `cognate'
In-Reply-To: <Pine.GSO.4.03.9907141119250.17570-100000@rsunx.crn.cogs.susx.ac.uk>
Message-ID: <FRI.16.JUL.1999.161503.0400.>

On Wed, 14 Jul 1999, Larry Trask wrote:

>> I'm sure it says PIE *<p at ter> in your manuscript (+ diacritics on the
>> <e>).

> Yes, it does, and my posting should have read `p at ter', but I slipped up.

At least according to Don Ringe, this word should be reconstructed
*pH2te'r.  I should ask him what the arguments are in favor of that
reconstruction, since I don't know myself.

  \/ __ __    _\_     --Sean Crist  (kurisuto at unagi.cis.upenn.edu)
 ---  |  |    \ /     http://www.ling.upenn.edu/~kurisuto/
  _| ,| ,|   -----
  _| ,| ,|    [_]
   |  |  |    [_]


From Georg at home.ivm.de  Fri Jul 16 09:32:01 1999
From: Georg at home.ivm.de (Ralf-Stefan Georg)
Date: Fri, 16 Jul 1999 11:32:01 +0200
Subject: accusative and ergative languages
In-Reply-To: <001001beccef$8ce328e0$40d3fed0@patrickcryan>
Message-ID: <FRI.16.JUL.1999.113201.0200.>

>> Try reconstructing PIE using _nothing_ but contemporary Albanian and
>> English.  You'd be hard-put to prove that such a language even existed.

>It is thuis type of illogic which stands in the way of real progress in
>historical linguistics.

>Larry Trask and others have done an excellent job, IMHO, of reconstructing
>the language from which present Basque derives --- from only one language.

That's because Basque has been for long known to be related to Basque. This
is called internal reconstruction.

>I also am certain that if we had reason to know that Albanian and English
>were descendants of a common language, we would eventually discover a kind
>of IE that would be inaccurate in some ways but essentially write.

The trick is about having reasons to know this, without resorting to some
ideology which tells you before, i.e. without looking at any data.

>Naturally, the more languages we added to the equation, the better
>reconstruction of IE we would achieve.

>Same principle.

>Pat
>
>PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.

[ moderator snip ]

Stefan Georg
Heerstrasse 7
D-53111 Bonn
FRG
+49-228-69-13-32


From mcv at wxs.nl  Fri Jul 16 22:50:18 1999
From: mcv at wxs.nl (Miguel Carrasquer Vidal)
Date: Fri, 16 Jul 1999 22:50:18 GMT
Subject: accusative and ergative languages
In-Reply-To: <001f01becdcd$786928e0$309ffad0@patrickcryan>
Message-ID: <FRI.16.JUL.1999.225018.GMT.>

"Patrick C. Ryan" <proto-language at email.msn.com> wrote:

>And to answer your -- I hope not purposeful -- distortion of what I wrote,
>let me say explicitly that I did not assert "there are ergative languages
>without (any) splits". I asserted that Thomsen did not, at least in her
>grammar, identify splits in Sumerian, which you seemed to think she had.

Of course she has, unless your edition differs from mine: p. 51,
paragraph 42:

	On the morphological level Sumerian has thus an ergative
 	system in the nouns and the intransitive vs. the
	transitive hamTu conjugation [...]. In the pronouns and
	the transitive maru^ conjugation vs. the intransitive
	verb, on the other hand, the system is nominative-
	accusative [...].
	This `split ergativity' is no uncommon phenomenon, in
	fact no ergative language is entirely ergative in both
	syntax and morphology.

=======================
Miguel Carrasquer Vidal
mcv at wxs.nl


From Georg at home.ivm.de  Fri Jul 16 22:16:05 1999
From: Georg at home.ivm.de (Ralf-Stefan Georg)
Date: Sat, 17 Jul 1999 00:16:05 +0200
Subject: accusative and ergative languages
In-Reply-To: <002601becdce$9c77b340$309ffad0@patrickcryan>
Message-ID: <SAT.17.JUL.1999.001605.0200.>

>R-S sighed:

I think I have the sympathy of most people for an occasional sigh during
this thread ...

>> This is getting weary. I think no sane linguist will be unaware of the fact
>> that there are languages with fewer purely morphological means than others.


>Pat concludes:

>In my opinion, dearth of morphological devices is validly characterised as
>"simple".

"Simple" as a characterization of the *morphological subsystem* of a
language. Any dearth of means in one subsystem has to be compensated in
another one to keep the language what it is meant to be, a proble-solving
device. A language is more than its morphology. Anything else is
balderdash, mystification, eighteenth-century stuff. Do you have an English
translation of the Iliad handy ? Do you think it is accurate enough to
convey what the unknown writer going by the name of Homer wanted to say,
its poetic beauties apart, of course ?? Yes ? See what I mean ?
I may have difficulties to come through to you, but, please, you have to
bear with me. I speak a language rich in morphological means, you cannot
expect me to be able to express the richness of my profound thoughts
accurately in a language as simple as English, the best parts are
irretrievably lost, because of the lack of a synthetic dative ;-)
["because-of-the-lack-of", good heavens, take Turkish: noksanlIGIN iCin,
two words instead of five, now *that's* a language !].
(This could also explain the strange ways Marxism went in China, ever
thought of this ? ;-))

Stefan

Stefan Georg
Heerstrasse 7
D-53111 Bonn
FRG
+49-228-69-13-32


From petegray at btinternet.com  Fri Jul 16 18:30:40 1999
From: petegray at btinternet.com (petegray)
Date: Fri, 16 Jul 1999 19:30:40 +0100
Subject: accusative and ergative languages
Message-ID: <FRI.16.JUL.1999.193040.0100.>

> Pat said:
> .... Why not list two or three additional elements that function in
> Chinese equivalent to inflectional affixes?

The most obvious omission from your list is the "directional" affixes,
simple and complex, e.g.: lai (come); qi3-lai; etc
And you can add zhe, guo, zheng4zai4, yao4, ...etc etc all showing different
limitations to the action.

Whether you count these as "inflexions" or as "affixes" or as common second
elements in compound verbs, is probably a matter of taste.

There are books of Chinese verbs that list all these things for you, but
alas, I have none handy.

Peter


From kurisuto at unagi.cis.upenn.edu  Fri Jul 16 20:12:28 1999
From: kurisuto at unagi.cis.upenn.edu (Sean Crist)
Date: Fri, 16 Jul 1999 16:12:28 -0400
Subject: accusative and ergative languages
In-Reply-To: <l03102801b3b1682958b3@[195.247.226.43]>
Message-ID: <FRI.16.JUL.1999.161228.0400.>

Somebody (I've lost track of who) wrote:

> >Yes, but if a new verb were formed, e.g. webben (from English 'web'), there
> >is no possibility that it would be conjugated webben, wabb, gewobben.

This sort of thing actually _does_ happen sometimes; it just isn't that
common.  The English verb "strive" is a loan from Old French; but it is
often declined as if it were an old Class I ablauting strong verb (strove,
striven).  So I'd correct "no possibility" to "much less likely".

  \/ __ __    _\_     --Sean Crist  (kurisuto at unagi.cis.upenn.edu)
 ---  |  |    \ /     http://www.ling.upenn.edu/~kurisuto/
  _| ,| ,|   -----
  _| ,| ,|    [_]
   |  |  |    [_]


From proto-language at email.msn.com  Sat Jul 17 04:14:58 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Fri, 16 Jul 1999 23:14:58 -0500
Subject: Momentary-Durative
Message-ID: <FRI.16.JUL.1999.231458.0500.>

Dear Peter, Vidhyanath, and IEists:

 ----- Original Message -----
From: petegray <petegray at btinternet.com>
Sent: Wednesday, July 14, 1999 2:21 PM

> Vidhyanath's two postings on aspect show clearly what the issue of aspect in
> PIE may be - we simply don't have a clear definition of what we are talking
> about.

> Greek usage does not agree with Vedic, and where traces of aorists exist in
> other languages (e.g. Armenian, OCS, Albanian, Latin, Baltic), we cannot
> convincingly recover an original aspectual meaning.

> The formations are clearly PIE, but I believe I can continue to claim that
> the distinctions between them, if recoverable, are certainly not those we
> find in Greek.   The Armenian-Greek- I-I sprachbund has certain known
> innovations, such as vocabulary and the augment, and we should not be miseld
> into accepting as original what may be another innovation, namely the
> development of new distinctions based on an existing variety of forms.

I do not wish to intrude on the interesting discussion between you in
progress. However, for whatever it may be worth, I have recently finished a
first draft of a small essay on IE aspects, and a few of you may find it of
passing interest:

http://www.geocities.com/Athens/Forum/2803/IE-VerbalInflection.htm

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From proto-language at email.msn.com  Sat Jul 17 04:27:07 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Fri, 16 Jul 1999 23:27:07 -0500
Subject: Momentary-Durative
Message-ID: <FRI.16.JUL.1999.232707.0500.>

Dear Peter and IEists:

 ----- Original Message -----
From: petegray <petegray at btinternet.com>
Sent: Wednesday, July 14, 1999 3:19 PM

<snip>

> And there are very many instances of roots showing nasal presents in one IE
> language, and not in another, or indeed appearing in both forms in the same
> language, as Latin cumbo and cubo, or Greek leipo: and limpano:

<snip>

> No, for the reasons just stated.   There are enough counter-examples to show
> that whatever the conditions were for the "selection" of nasal, full grade
> or suffixed present, they are now as unrecoverable as they are within Latin,
> Greek, or Sanskrit, where very little difference of meaning can be
> consistently shown.

I have found that postulating an ingressive meaning for the affix *-n works
very well. It is my opinion that this *-n was originally added to a root not
infixed.

The example you gave of Latin cubo:, 'lie', and cumbo, 'lie down', will
serve as an excellent example of the ingressive force of the affix.

 Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From proto-language at email.msn.com  Sat Jul 17 05:21:32 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Sat, 17 Jul 1999 00:21:32 -0500
Subject: accusative and ergative languages
Message-ID: <SAT.17.JUL.1999.002132.0500.>

Dear Wolfgang and IEists:

First let me thank you for your very interesting and thought-provoking
comments.

----- Original Message -----
From: Wolfgang Schulze <W.Schulze at lrz.uni-muenchen.de>
Sent: Wednesday, July 14, 1999 5:18 AM

> "Patrick C. Ryan" schrieb:

>> .... As for lessened degrees of animacy, most ergative languages
>>  have antipassives to indicate this.

Wolfgang wrote:

> Before you invite people to subscribe to this claim you should first
> demonstrate that a) "most ergative languages have antipassives".

Pat writes:

I have reviewed Dixon, and apologize for the claim. I inferred incorrectly
that beacuse Dixon so often mentioned antipassives in connection with
ergatives that the frequency of their co-occurence was high.

Wolfgang worte further:

> This
> claim suggests that "ergative" is a substantial attribute that can be
> used with the referent "language". In earlier postings I have tried to
> show that "ergativity" (as well as "accusativity") represents a label
> for a structural BEHAVIOR of single paradigms WITHIN a language system.
> For instance (as I have said) a language system may be ergative in its
> agreement paradigm, accusative in owrd order, accusative in case
> marking, ergative in discourse cohesion etc. (to give a fictive
> example). Hence, there are NO "ergative (or "accusative") languages" or
> only, if you use this term in a very informal sense.

Pat writes:

In the sense you are using these, they seem to be of little value in
describing anything.

Wolfgang wrote further:

> Now, if you talk
> about antipassives, you should make clear to which morphosyntactic
> domain you allude to. Moreover, your claim suggests that most 'ergative
> languages' are reference dominated in the sense of Role and Reference
> Grammar. Only if a given language system ("operating systems in terms of
> the Grammar of Scenes and Scennarios" (GSS)) uses actant encoding
> devices to indicate fore- and/or backgrouding (instead of - for instance
> - smeanti coles such as agent/patient...) we can expect some kind of
> diathesis be it passive or antipassive (note that again passives and
> antipassives represent two poles on a much more complex scale that also
> involes bi-absolutives, pseudo-passives and many more structures).

Pat writes:

I am sorry that I do not agree with the validity of this distinction. For
me, 'actant' is 'agent'. Perhaps you can explain the difference.

Wolfgang wrote:

> The fact, however, is that many 'ergative' languages lack an
> antipassive. For instance, there are nearly 30 East Caucasian languages
> all of them using some ergative strategies in at least parts of their
> operating systems. But only a handfull of them (five or six, to be
> precise) have true antipassives (only one has some kind of
> "pseudo-passive").

Pat writes:

Are you asserting that the majority of ergative languages do not have
anti-passives?

Wolfgang wrote:

> The same is true for accusative systems (as you
> probably know). Hence, antipassives are a possible extension of ergative
> stretagies, they cannot serve for any kind of typological
> generalization.

Pat writes:

I was not aware that I was using anti-passives as a "typological
generalization". I only suggested (I thought) that that was one method
ergative languages employed for degrees of animacy/actancy.

Wolfgang wrote further:

> b) Antipassives have rarely to do with the "lessening of animacy". The
> most common inferences that allow antipassive structures are:

> - Reduction of 'activity' (that is the degree to which an actant is
> thought to be 'active' during a *specifc* (and single) event. From this
> another inference is given:

> - Habitual, durative action (-> imperfectiveness).....

> - The event becomes less discrete, hence less transitive. Another
> inference: The 'patient' looses its referentiality: It cannot be
> subjected to wh-questions, it cannot be counted, very often such
> referents are mass nouns or collectives...

> - Antipassives are part of the discourse cohesion strategies (most
> famous example is Dyirbal): Here, antipassives are neither semantically
> nor syntactically motivated, but merely a pragmatic feature of topic
> chaining.

Pat writes:

After having read Dixon, I can hardly be unaware of these considerations;
however, I can see that degree of actantcy might be defined in this context
meaningfully. But actantcy and animacy are still closely related.

As for the far-reaching conclusions of Dixon based on discourse conhesion
strategies, in my opinion they are flawed because these strategies are
purely conventional.

In English, I can say: "John hit me, and he went away" or "I hit John and he
went away". What is the discourse cohesion strategy for English?

Wolfgang further wrote:

> There are much more functional options that are carried out by
> antipassives. In fact, all these options NEVER allow such a claim as
> quoted above (rather, they contradict it).

Pat wrote earlier:

>> Although Dixon is certainly a man who has devoted much thought to
>> ergativity, I find something inherently problematical in combining ergative
>> and accusative features in one sentence (a little schizophrenic) which he is
>> forced to do by analyzing pronominal and nominal structures differently when
>> they occur in the same sentence.

Wolfgang wrote further:

> WHY? It all depends from how you interpret erg. and acc. features.
> Confer for instance the following (one!) sentence from (informal) Lak
> (East Caucasian):

> t:ul b-at:-ay-s:a-ru zu
> I:ERG I:PL-hit-PART:PRES-ASS-SAP:PL you:PL:ABS
> 'I surely hit you (plural)'

> [For the expert: Standard (Literary) Lak would have 'na bat:ays:aru zu'].

> This sentence is:

> ACC with respect to word order [*zu t:ul bat:ays:ara would be ERG]

Pat writes:

I have to plead ignorance of Lak however, I find an analysis of an otherwise
completely ergative sentence as having an ACC word-order impossible. I
believe this opinion rests on a false analysis of word-order significance.

Wolfgang continued:

> ERG with respect to case marking ['neutral with 'na' for "I" is also
> possible]
> ERG with respect to class agreement (b- = (here) class [+hum;+plural])
> ERG with respect to personal (or, better, speech act participant)
> agreement (-ru is SAP:PL and agrees with 'zu' "you:PL").

> But if you say "I am surely hitting you (plural)", you get:

> na b-at-la-ti-s:a-ra zu
> I:ABS I:PL-hit1-DUR-hit2-ASS-SAP:SG you:PL:ABS

> Here we have:

> ACC with respect to word order
> ERG with respect to calss agreement
> ACC (or neutral) with respect to case marking)
> ACC with respect to SAP agreement (-ra is triggered by 'I:ABS').

Pat writes:

So, agent marked ABS and patient marked ABS is, according to you, ACC with
respect to case marking? Sorry, not convincing at all! And I think the SAP
difference can be explained as indicating seriality of the patients involved
as recipients of the action by *one* agent.

Wolfgang wrote further:

> Now, please tell me: Is Lak an 'ergative' or an 'accusative' language?
> [Please note that I did not include (among others) strategies of
> discourse cohesion, reflexivization and logophization].

Pat writes:

By these examples, I would tell you that Lak is 'ergative'; and that
accusativity is not demonstrable from these examples.

Pat wrote previously:

>> I think it is likelier that, because of perceived greater animacy (or
>> definiteness), pronouns have a different method of marking that can still be
>> interpreted within an ergative context.

Wolfgang wrote:

> This a (very simplified) 'on-dit' that stems from the earlier version of
> the Silverstsein hierachy. Again, we have to deal with the question,
> whether a 'pronoun' (I guess you mean some kind of 'personal pronouns')
> can behave 'ergatively' or 'accusatively'. The list below gives you a
> selection of SAP case marking in East Caucasian languages with respect
> to ABS/ERG:

> ABS vs. ERG ABS = ERG
> ALL ---
> Singular Plural
> Plural Singular
> 1.Incl. Rest
> 1:SG Rest
> 2:SG Rest
> 1:SG/PL Rest
> --- ALL

> This list (aspects of personal agreement NOT included!) shows that SAP
> pronouns may behave different within the same paradigm. Any
> generalization like that  one quoted above does not help to convey for
> these data...

Pat writes:

I do not have a reference book for Lak so that my hands are somewhat tied.
But, I have found that paradigms are often inconsistent in ways that reflect
earlier lost phonological changes, or other lost schemata. The locative
plural terminations of IE are certainly not, in origin, terminations of the
locative plural. Etc.

But, Wolfgang, I have to thank you because you are stretching my mind, and I
like the feeling.

Thank you.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From X99Lynx at aol.com  Sat Jul 17 07:39:22 1999
From: X99Lynx at aol.com (X99Lynx at aol.com)
Date: Sat, 17 Jul 1999 03:39:22 EDT
Subject: Chronology of the breakup of Common Romance [long]
Message-ID: <SAT.17.JUL.1999.033922.EDT.>

In a message dated 7/13/99 6:16:38 PM, rmccalli at sunmuw1.MUW.Edu wrote:
<<Some popular writers have also proposed Volcae as the source of English
"folk" but I've never seen this proposal from professional linguists.>>

In a message dated 7/16/99 2:53:27 AM, edsel at glo.be replied:
<<Even less likely : the peoples that used the word had probably never heard
of the Volcae, not even second-hand.>>

Actually, of the few things we hear of the Volcae Tectosages (versus the
Volcae Arecomici, a more southwestern federation of Celts) one of them is
that they moved into "Germania" and were becoming assimilated.   And we know
this from one of the most credible sources in antiquity (among others of the
same period):

<<And there was formerly a time when the Gauls excelled the Germans in
prowess, and waged war on them offensively, and, on account of the great
number of their people and the insufficiency of their land, sent colonies
over the Rhine.

Accordingly, the Volcae Tectosages, seized on those parts of Germany which
are the most fruitful [and lie] around the Hercynian forest, (which, I
perceive, was known by report to Eratosthenes and some other Greeks, and
which they call Orcynia), and settled there. Which nation to this time
retains its position in those settlements, and has a very high character for
justice and military merit; now also they continue in the same scarcity,
indigence, hardihood, as the Germans, and use the same food and dress; but
their proximity to the Province and knowledge of commodities from countries
beyond the sea supplies to the Gauls many things tending to luxury as well as
civilization.   Accustomed by degrees to be overmatched and worsted in many
engagements, they do not now even compare themselves to the Germans in
prowess.>>

- Caesar, Gallic War, 6.24 (Loeb)

It may be significant that about 150 years later, these Volcae have
completely disappeared in the descriptions of the same territories by Tacitus
and Ptolemy.  If the Volcae could have been assimilated, so could their
self-name.

Currently, the origins of the terms I use to describe my nationality and
other affilations include, first, the name of a 15th Century Italian sailor
(no, not 'Vespucian'). Then there's a Latin name compounding what looks to be
a Celtic one ('Pennsylvanian').  Others have included an anglicizing of a
Delaware name for an even earlier group of Native Americans, names from the
Greek alphabet, a sports team name borrowed from a rather unpleasant African
fauna and something in a Dutch place name that ended up being rendered as
"Brooklynite' - a term of no small pride and self-identification, as
mysterious as that may seem to the outsider.  If 'folk' moved around anything
like any of these other self-names, then I'm sure we should not be surprised
where it came from.

In a message dated 7/13/99 6:16:38 PM, rmccalli at sunmuw1.MUW.Edu wrote:
<<Isn't the term Vlakh anachronistic in this sense? Vlakh is first
documented in the middle ages, isn't it?>>

In a message dated 7/16/99 2:53:27 AM, edsel at glo.be replied:
<<Indeed. The medieval use of 'Vlach' probably referred to the 'Latin'
Rumanians, cf. Wallachia. In modern Greece, the word is also used to refer to
the local nomads/gypsies of unclear (i.e. to me) ethnic origin (maybe
Rumanian Gypsies, 'Roma', or from Pannonia? Maybe Albanians?).>>

But in fact the form 'vlach' appears very late in medieval times - one of the
arguments that the migration into Romania was relatively recent.  There are
even indications that at one point there was a diachronic presence with
'woloch' (Celt, Roman, Frank) still existing alongside of 'vlach' outside of
South and East Slavic.  In fact, before 'vlach' first enters English, it is
preceded by a century by 'Woloch" refering to Vlachs.  See also the Hungarian
form 'olasi' mentioned in my earlier post.  [And I see also that in Western
Slavic, "walska" (Old Bo. and something like it in Pol) is "war."]

I don't think anyone seriously argues with the idea that 'vlach' is derived
from the name applied to Romance-types, particularly Romanized Celts - and
not orginally a Rumanian-speaking group.  (BTW, what is 'balkan' from?)

With regard to the vallum/Volcae connection, edsel at glo.be replied:

<<That excludes a Roman origin of the word.  All this sounds like a typical
19th century invention by Latin-worshippers>>

As opposed to 19th Century Wagnerian Teutonists or Pan-Slavists, who weren't
above some invention of their own.  I got a post that corrects me about the
'wall' theory - apparently a British scholar named Lewis suggested in a RS
paper in the twenties that both vallum ('vallus stake') and waldh, etc.
derived from the  early proto-european [sic] "fala", which the Romans applied
to wall-building and seige scaffling, and created heights in general, fully
aware that it was a borrowed word "quod apud Etruscos significat caelum"; cf.
fulcio, fulsi, fultum: to strenghten, particularly by walls town-places ."

Regards,
S. Long


From jer at cphling.dk  Sat Jul 17 13:50:17 1999
From: jer at cphling.dk (Jens Elmegaard Rasmussen)
Date: Sat, 17 Jul 1999 15:50:17 +0200
Subject: Momentary-Durative
In-Reply-To: <007201bece36$602a66e0$a53bac3e@xpoxkjlf>
Message-ID: <SAT.17.JUL.1999.155017.0200.>

To "petegray" and List,

Thank you, Peter, for a challenging reply. I'll try and do it justice.

On Wed, 14 Jul 1999, petegray wrote:

> [...]
> Yes - but you have made the assumption that this root always shows the nasal
> in the present.   This is not the case.
> (a) Albanian shows two forms, one with, and one without the nasal.
> (b) Old Irish shows no nasal.
> (c) Greek a-tuzomai shows no nasal.
> (d) If Pokorny is right to link the root with stud- then we have no nasal in
> Latin studeo, nor in the various reflexes of it in Germanic.

Sure, verbs can form other derivatives too. Gmc. *staut-i/a- must be an
old intensive (*stu-sto'wd-); studeo is a stative in *-eH1-; atu'zomai 'am
frightened' has a velar in atukhthei's and does not look related; Alb.
shtyj, -yn is no evidence since the nj-present is productive (if the old
form is shtyenj it is hard to understand as related at all). But Old
Irish -tuit 'falls' is *tudeti and identical with tudati. So maybe for
this verb the stem restructuring should be pushed back in to the
prehistory of the protolanguage to give us PIE variant presents - though
palallel development of course must be very common.

>>  There are so many instances of the _same_ verbal root turning up
>> with a nasal present in IE branches that have nothing else in common ...

> And there are very many instances of roots showing nasal presents in one IE
> language, and not in another, or indeed appearing in both forms in the same
> language, as Latin cumbo and cubo, or Greek leipo: and limpano:

Right, and in such cases it is advisable to assume a set of nasal present
and root aorist, since they are so often combined, and are retained
together with so many verbs. Thus Gk. leip-/limp- are inded both presents,
but *ley{kw}-e/o- is a perfectly normal IE root-aorist subjunctive which,
due the closely allied function of that category, often came to compete
with the inherited present.

>>. Is it a coincidence
>> that *k^lew- 'hear' forms a nasal present in Indo-Iranian and Celtic?

> Yes.   It has no nasal in Greek or Latin or some others. (and so on with the
> rest of your list)

Why should it have, if the individual languages went crazy and levelled
almost all differences between the aspect stems?

> I cannot agree with you that certain verbal roots always had nasal presents
> in all IE languages.   This is simply factually untrue.   But without that
> assumption, your argument collapses.

Not "in all IE languages", only in the IE protolanguage, which apparently
had a general (unilateral) solidarity between nasal presents and root
aorists. From this common point of departure the languages developed all
sorts of differences (even closely related lgg. differ in this respect,
oh yes).

>>. Does that not indicate
>> that the assigment of the nasal-infix structure as the present of certain
>> roots was fully lexicalized in the protolanguage?

> No, for the reasons just stated.   There are enough counter-examples to show
> that whatever the conditions were for the "selection" of nasal, full grade
> or suffixed present, they are now as unrecoverable as they are within Latin,
> Greek, or Sanskrit, where very little difference of meaning can be
> consistently shown.

>> Oh no, it works only in the direction that a nasal present has a root
>> aorist beside it in practically all cases.

> Not a surprise, if you ignore all the langauges where this is not the case.
> Likewise, not a surprise, since root aorists were an old form.   The same
> could be said of full grade presents, or suffixed presents.   So the
> statement really is without significant meaning.

No, thematic presents and root-presents typically take the s-aorist (or
suppletion).

>> Also the reduplicated presents
>> and the y-presents generally form root aorists.

> Yes - so your argument (that a root aorist implies a nasal present) is
> really rather weak.   Yet without this argument, you cannot derive tuda'ti
> the way you do.

No again, a root aorist _is implied by_ a nasal present. And since that is
found, there is very good reason to believe that there was a root aorist
in IE. And then, surprise, that offers all the basis you need for a smooth
and natural derivation of the tudati type - smooth at least in comparison
with many stories of verbal stem-formation that can be followed in
attested languages. The story also explains what is perhaps the most
important point of all, which I did not mention in my posting: the
punctual aspect value observed by Renou in the use of the tudati present
type (Louis Renou: Le type ve'dique tuda'ti. Me'langes Vendryes, Paris
1925, 310-16, a classical and oft-quoted philological paper written
completely without comparative or even diachronic bias). - In fact, I am
only saying that tuda-, which is structured like vida-, may have been an
aorist stem just like the latter.

>> [in] PIE, each verb had mostly picked one form for its present. The massive
>> concord among the IE languages in this respect

> There is no such concord.  Even a single language shows variety of
> formations of the present, and the variety across the languages is
> considerable.

Yes, but easily explained as secondary, or not relevant, if the forms
adduced are not from the primary verb, or if they belong to types that got
productive in the separate lives of the individual languages. When that is
subtracted, it is quite strinking how well the old core of the different
branches get to resemble each other. The impression that IE verbal stem
formation just mutates erratically and that no common system is
recoverable is not compatible with current knowledge.

> But it would be good to have firm data on this - anyone got
> three weeks to spare going through the text books?

So you haven't done that? It's quite easy now: LIV (Lexikon
Indogermanischer Verben, by Helmut Rix & al., Wiesbaden 1998) offers all
you need, including reconstructions and developmental tracings.

>> - while the even greater discord is easily
>> explained by continued normal language change.

> No.   Normal language change would not turn a nasal present into a
> reduplicated one or vice versa -

Why not - if they were both equally expected as the companion of a root
aorist? Is that not the way analogy works? Let me give you two
illustrative case stories. The IE status of a reduplicated present from
the two verbs *sed- 'sit down' and *steH2- 'stand up' cannot be questioned
(there is *si-sd-e/o- in Ital., Gk., IIr.; and *sti-steH2- or rather its
old subj. *sti-stH2-e/o- in IIr., Gk., Ital. and Celt.), nor can a root
aorist be avoided for the latter; I accept it for both, of course, but
*sed- as aor. is not so well-preserved. Now, the paradigms *si'-sd-e-ti,
*se'd-t and *sti'-stH2-e-ti *sta'H2-t thus restored had a funny fate in
Balto-Slavic and Germanic: BSl. changed *sisd-e/o- into *sind-e/o-, and
Gmc. back-formed *st at 2-n-t-e/o- from the aor. *staH2-t (drawing the -t- to
the stem, i.e. apparently after an adjustment of *staH2-t to "*staH2t-e"
with the productive ending of the perfect). Why would they do such a
thing? There can be only one answer: because the reduplicated present type
was going out of fashion, while the nasal present was still productive and
still went well with root aorists.

>  -, nor would it turn a -sk suffixed one into a nasal present, etc, etc.

That's right, these do not belong together, and I never said they did.
However, in a few odd cases the whims of chance did end up combining them,
as when Hitt. nasal verbs form durative derivatives with -sk- like all
other verbs do, or Toch. makes itself a prs. formant -na"sk- (-na:sk-),
thereby giving inherited nasal presents a productive present-stem marking.
And if there are both a set "root-aor. + nas.-prs." and an inchoative
"s-aor. + sk-prs.", and the functional difference was lost, you may end up
finding a nasal present combined with an s-aorist, or an sk-prs. with a
root aorist. In some cases one may even suspect that this had happened
before the dissolution of the protolanguage; a prominent case is Gk.
gigno:'sko: with root-aor. e'gno:n, but Gothic kunnai{th} combining nasal
infix with the vocalism (*-n-e:-, not *-n-o:-) of the s-aor. that also
gave the stem of Gmc. *kne:-(j)i/a- 'know', retained with the -s- in Hitt.
ganest 'recognized' (*g^ne:'H3-s-t). In a principled analysis, this
reflects two IE verbs, (1) *g^n.-n?-H3-ti, aor. *g^no'H3-t; (2)
*g^n.H3-sk^e'-ti, aor. *g^ne:'H3-s-t. The latter set is directly continued
as a synchronic paradigm by Arm. {cv}ana{cv}'em, aor. caneay as seen by
Jasanoff.

The most rewarding experience during the time I have been watching
Indo-European Studies has been to see the protolanguage come alive and
assume an increasingly well-established structure, with archaisms and
productive patterns, just like the living languages we know. I react in
defense of the field when I see somebody turning the clock back to a stage
we have left behind.

Jens


From CONNOLLY at LATTE.MEMPHIS.EDU  Sat Jul 17 16:49:17 1999
From: CONNOLLY at LATTE.MEMPHIS.EDU (CONNOLLY at LATTE.MEMPHIS.EDU)
Date: Sat, 17 Jul 1999 11:49:17 -0500
Subject: Interpreting ergative sentences
Message-ID: <SAT.17.JUL.1999.114917.0500.>

>[Ed Selleslagh]

>Speaking about Castilian, I would like to have your views on a peculiarity,
>or rather a tendency, that is still productive at least in popular speech
>('le-ismo'), namely the tendency to use the indirect subject form where all
>other Latin (and other West-European) languages use the direct object form,
>and almost exclusively with persons (animate), e.g. "le v=ED" ("le vi' "), "I
>saw him".

I don't actually know Spanish; I just know a few examples of how such verbs
work in Spanish and many other languages.

>It has sometimes be suggested that this was a substrate influence from
>Basque, an ergative language (that lacks an accusative, of course), but
>opinions are extremely divided on this subject. On the other hand, it seems
>to me this could (but I don't know) also be related to some of the arguments
>presented in the discussion on ergative/accusative concerning the distinct
>role of animates.

I don't know any Basque either.  But I believe that the same tendency is found
in Sicilian dialects, so I'm a little leery of looking to Basque for an
explanation.  Especially since, as far as I know, personal absolutive NPs with
transitive verbs get the same treatment as impersonal NPs, while beneficiary
"indirect objects" with verbs such as 'give' get a different, "dative"
treatment.  If that's so, then Basque doesn't show the tendency to treat all
personal non-ergative NPs with bi- or trivalent verbs alike.  But better ask
Larry about that.

I should add further that I'm generally suspicious of explanations based on
substrate influence.  Yes, substrates are real, and sometimes they do seem to
provide a plausible explanation of one phenomenon or another.  But languages
are capable of changing strictly on their own, and I think in general we should
look first for language-internal solutions.  Since even American Spanish uses
_a_ before personal direct objects, as it does before indirect objects, it is
hardly surprising that Castillian would tend to use identical pronouns for all
personal objects, whether direct or indirect.

Leo

Leo A. Connolly                         Foreign Languages & Literatures
connolly at latte.memphis.edu              University of Memphis


From petegray at btinternet.com  Sun Jul 18 17:33:54 1999
From: petegray at btinternet.com (petegray)
Date: Sun, 18 Jul 1999 18:33:54 +0100
Subject: Momentary-Durative
Message-ID: <SUN.18.JUL.1999.183354.0100.>

>>> Strunk has shown that nasal presents go with root aorists,

Can you give us a reference please, Jens?   I would like to question this,
for the following reasons:
(a) In Greek it is largely true, but there are exceptions;   so a bald
statement would need qualification.
(b) In Latin it is largely untrue, since the aorists are either sigmatic or
the rare reduplicative  aorist (tango tetigi, claimed by some as an aorist
on the basis of Homer tetago:n, or the lengthened vowel: pango pe:gi (~
perfect pepigi).   I can only find cumbo cubui which supports the claim in
Latin.
(c) In Sanskrit there is a variety of possible presents and possible
aorists.   There are 29 roots listed by Whitney which take nasal presents
(ignoring the class nine presents).  Of these the first retains its nasal in
the aorist, the second has no aorist, the third shows non-nasal present
forms, and has either reduplicated or sigmatic aorist, and so on .....
through the list.
(d) As stated earlier, a root may show nasal and non-nasal presents in
different languages and even in the same language.

So as a bald statement I find it unbelievable.   The evidence doesn't seem
to be there.    I would like to see what Strunk actually says.

> If my observation that there is an alliance between the sk^-present type
> and the s-aorist is correct,

Both forms may be used for some other reason, for example phonetic.   In
both Greek and Latin the sk- ending occurs only after a vowel, which then
takes a sigmatic aorist for purely phonetic reasons.

>> This brings me to a general question. There seem to be two camps about the
>> category system of the PIE verb. One believes that the aorist-imperfect
>> distinction, to be equated to perfective-imperfective distinction,
>> ``always'' existed in PIE and Hittie lost this distinction, while Vedic
>> changed things around. The other considers the aspectual distinction to
>> postdate the separation of Anatolian.

> [Jens] There are these two camps, yes, and I am in no doubt that camp one is
> right. There is no way the specific forms of the aspect stems could have
> been formed secondarily in "the rest of IE" left after the exodus of (or
> from) the Anatolians. At the very least, all the _forms_ must be assigned
> to a protolanguage from which also Anatolian is descended.

I thoroughly agree with you here, Jens.   The forms are scattered over most
of the IE languages.

> And what would
> the forms be there for, if the functions that go with them only developed
> later?

Here I disagree.   There can be variety of form in a language without a
difference in meaning.   Look at German plurals (or English for that
matter).   Sometimes a difference of meaning will emerge, fixing one meaning
on one form and the other on the other (German Woerter and Worte, Dinger and
Dinge), but this does not mean that all variations must necessarily
correspond to a difference in function.    Language simply doesn't work that
way, and there are far too many counter-examples for that claim to stand.

Counter examples in English:
(a) strong versus weak past tenses, e.g. dived ~ dove (origin historical /
analogical)
(b) plural /s/ versus plural /z/ (origin phonetic)
(c) gentive versus preposition e.g. Tom's ~ of Tom (basis stylistic)
(d) Time expressions (believe it or not, 10:45 really is the same as a
quarter to eleven)
and so on.

I see the development of a functional difference as a later phenomenon, and
I see it emerging differently in Sansksrit and Greek and Latin.  I suspect
that some scholars are temporarily misled by the use of Greek names for the
Sanskrit verbal system.

Peter


From ALDERSON at toad.xkl.com  Tue Jul 20 17:01:33 1999
From: ALDERSON at toad.xkl.com (Rich Alderson)
Date: Tue, 20 Jul 1999 10:01:33 -0700
Subject: Hiatus
Message-ID: <TUE.20.JUL.1999.100133.0700.>

Dear Readers:

I will be away from these facilities until 25 July 1999.  There will be
no mailings from the Indo-European or Nostratic lists until I return.
I apologize for any inconvenience this causes.

								Rich Alderson
								list owner and
								moderator
-------


From proto-language at email.msn.com  Mon Jul 19 05:13:46 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Mon, 19 Jul 1999 00:13:46 -0500
Subject: accusative and ergative languages
Message-ID: <MON.19.JUL.1999.001346.0500.>

Dear Larry and IEists:

 ----- Original Message -----
From: Larry Trask <larryt at cogs.susx.ac.uk>
Sent: Friday, July 16, 1999 10:35 AM

> Pat Ryan wrote:

>> Now, when Larry recently quoted Dixon about the nature of the
>> ergative, he conveniently neglected to mention that Dixon
>> acknowledged that there were currently practising linguists --- not
>> amateur linguists like myself --- still defending the passive
>> interpretation of ergative constructions.

Larry responded:

> No; this is not so.  Read p. 189 of Dixon's 1994 book.

Pat responds:

I suggest you read it. "But we do still encounter scholars who insist that
there is a necessary diachronic connection, e.g. Estival and Myhill
(1988:445): 'we propose here the hypothesis that in fact all ergative
constructions have developed from passives'."  Are you suggesting that
Estival and Myhill are not "linguists", or that Shibatani, in whose book
this essay appeared, is not a "linguist"?  The quotation above is from p.
189 of Dixon's 1994 book. What are you playing at?

Pat asked:

>> I asked Larry where he had "shredded" this interpretation, and to my
>> knowledge, got no answer. If I missed the "shredding", perhaps you
>> will be kind enough to rehearse his performance for us. I have seen
>> nothing by Larry's vehemence and your allegiance to support the idea
>> that the ergative should not be interpreted as a passive.

Larry answered:

> Mr. Ryan, I'm afraid I lack the time to reply in detail to all, or even
> most, of the 37 or so postings from you that greet me each morning.

> But briefly: if a transitive sentence in an ergative language were
> "really" a passive, then its absolutive NP would be the subject and
> would exhibit subject properties.  But this not the case in most of the
> ergative languages I have heard about.  In spite of the ergative
> case-marking and/or verbal agreement, it is usually the *ergative* NP
> which exhibits subject properties, not the absolutive NP.

Pat interjects:

Still no word on the scene of the great "shredding" which was *claimed* by
Larry.
And what are subject properties that an absolutive NP never displays?

Larry continued:

> Take any ergative language you like.  Assume that the subject of an
> intransitive sentence is the sole (absolutive) NP in it.  Examine the
> syntactic properties of that NP, and tabulate them.

> Then look at transitive sentences.  Tabulate the syntactic properties of
> the absolutive and ergative NPs, and compare these with the preceding.

> In the great majority of ergative languages, it is the ergative NP which
> shares the subject properties of the intransitive subject.

> A few examples of typical subject properties:

> The subject controls reflexive and reciprocal NPs.

> The subject cannot itself be reflexive or reciprocal.

> The subject controls the empty NP in an empty-NP complement.

> The empty NP in an empty-NP complement is itself a subject.

> A subject can be coordinated with another subject.

Pat responds:

Why do you not give us an example of these so-called subject properties in
some ergative language besides Basque? And how were these properties
selected?

Larry continued:

> And so on.  There are also various language-specific tests.  For
> example, in some varieties of Basque, the object of a gerund (but not
> the subject of a gerund) goes, exceptionally, into the genitive case.
> Neither the absolutive subject of an intransitive gerund nor the
> ergative subject of a transitive gerund can be genitivized, but the
> absolutive object of a transitive gerund can be.

Pat responds:

Dialect-specific and totally unconvincing for ergative languages in general.

Larry continued:

> One more test from Basque.  In Basque, the subject of an intransitive
> sentence cannot be reflexive or reciprocal.  So, Basques can say,
> literally, `Susie and Mike [ABSOLUTIVE] were talking to each other', but
> they cannot say *`Each other [ABSOLUTIVE] was talking to Susie and
> Mike'.

Pat responds:

Since '*Each other was talking to Susie and Mike' is equally ridiculous, I
fail to see any valuable point made. Frankly, I am amazed. A reflexive
requires an agent and a patient, and a reciprocal requires two agents and
two patients. An intransitive verb, by definition, has only one NP element
so any two(or four)-NP-element construction obviously is a contradiction in
terms.

Larry continued:

> Now, in a transitive sentence, Basques can say `Susie and Mike
> [ERGATIVE] slapped each other [ABSOLUTIVE]', but they cannot say *`Each
> other [ERGATIVE] slapped Susie and Mike [ABSOLUTIVE]' -- or, in terms of
> the discredited passive theory, they cannot say *`Susie and Mike
> [ABSOLUTIVE] were slapped by each other [ERGATIVE]'.

Pat responds:

More nonsense! 'Each other' cannot function as an ergative or nominative
subject. So what? That is semantic not grammatical. That it can be used in
some languages in an oblique case (as in English above) is to be expected.

Larry continued:

> In an intransitive sentence, the absolutive subject cannot be reflexive
> or reciprocal.  In a transitive sentence, the ergative subject cannot be
> reflexive or reciprocal, but the absolutive object *can* be.

Pat responds:

Answered above.

Larry continued:

> All tests for subjecthood in Basque give the same result: it is the
> ergative NP in a transitive sentence which shares the subject properties
> of the absolutive NP in an intransitive sentence, and therefore it is
> the ergative NP, and not the absolutive NP, which is the subject of a
> transitive sentence.

Pat responds:

If you are referring to the "tests" above, you have proved nothing.

Larry continued:

> There are unusual languages, of course.  In Dyirbal, and in
> Nass-Gitksan, the absolutive NP in a transitive sentence shows at least
> some subject properties, while the ergative NP does not show the same
> properties -- though it does show other subject properties.  This is the
> phenomenon we call `syntactic ergativity'.  But I know of no language
> which is syntactically ergative without exception, and I know of few
> languages which are syntactically ergative at all.

Pat responds:

Of all the languages I have ever seen, Basque is, by a mile, far the most
"unusual" language.

Larry continued:

> The "passive" view of ergative languages in general is indefensible.

Pat responds:

Obviously, I do not think so. And Estival and Myhill (and probably
Shibatani) do not either --- not to mention the majority of linguists of the
past.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From proto-language at email.msn.com  Mon Jul 19 06:26:43 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Mon, 19 Jul 1999 01:26:43 -0500
Subject: accusative and ergative languages
Message-ID: <MON.19.JUL.1999.012643.0500.>

Dear Ralf-Stefan and IEists:

 ----- Original Message -----
From: Ralf-Stefan Georg <Georg at home.ivm.de>
Sent: Friday, July 16, 1999 4:54 PM

R-S wrote:

> How do you live with the fact that some "ergative languages" have
> independent passives ?

Pat asks:

Generalities are somewhat interesting but specifics would be even more
interesting.

<snip>

R-S wrote:

> And, as far as I remember, this whole brouhaha started with me asserting
> that all known "ergative languages" have some split, followed by you
> prompting we to show such a thing in Sumerian. I'm not a mind-reader, but
> this could be interpreted as a challenge aiming at the gist of my
> assertion, n'est-ce pas ? I managed to show those splits,

Pat interjects:

To my knowledge, you did not.

R-S continued:

> although the
> discussion got a bit swamped under some hassle over maru:- and
> HamTu-conjugations, but I managed to do it. At least I got you to accept,
> late in coming, though, that my initial assertian still stands up.

Pat responds:

I did *not* accept your initial assertion. Re-read my concession.

Pat wrote previously:

>> You might review your own procedures for employing quotation marks. Your use
>> of them on "there are ergative languages without any splits" strongly and
>> falsely implies that I wrote this in the context of a judgment on the
>> question.

R-S replied:

> I accept the reproach of not being a mind-reader ...

Pat answers:

Not 'mind-reading', just simple 'reading' will do nicely.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From colkitto at sprint.ca  Tue Jul 20 10:53:36 1999
From: colkitto at sprint.ca (Robert Orr)
Date: Tue, 20 Jul 1999 06:53:36 -0400
Subject: accusative and ergative languages
Message-ID: <TUE.20.JUL.1999.065336.0400.>

There's also "sneak - snuck" in English and "schreiben -schrieb-geschrieben"
in German, both of which are analogical formations  on originally borrowed
words..

Robert Orr


From larryt at cogs.susx.ac.uk  Mon Jul 19 10:49:05 1999
From: larryt at cogs.susx.ac.uk (Larry Trask)
Date: Mon, 19 Jul 1999 11:49:05 +0100
Subject: accusative and ergative languages
In-Reply-To: <002201bed014$3c87ab00$8f9ffad0@patrickcryan>
Message-ID: <MON.19.JUL.1999.114905.0100.>

On Sat, 17 Jul 1999, Patrick C. Ryan wrote:

> As for the far-reaching conclusions of Dixon based on discourse
> conhesion strategies, in my opinion they are flawed because these
> strategies are purely conventional.

Not so.  Read on.

> In English, I can say: "John hit me, and he went away" or "I hit
> John and he went away". What is the discourse cohesion strategy for
> English?

You've overlooked the crucial null-subject cases:

`John hit me and went away' *must* mean `John went away'.

But `I hit John and went away' *must* mean `I went away'.

Control of null NPs is one of the syntactic properties which crucially
distinguish subjects from non-subjects in English and in some other
languages.  And there is nothing "conventional" about it: this is a rule
of English syntax.

Let me change both NPs to third-person, to avoid any complications with
agreement; suppose I say this:

`John hit Bill and went away'.

Now, in English, it is John who went away, not Bill.  However, according
to my understanding of Dixon, if you say what looks like the literal
equivalent of this in Dyirbal, it is *Bill* who went away, not John.
This is one of the ways in which syntactic ergativity manifests itself
in Dyirbal.

But not all ergative languages are the same here, and probably not even
most.  If you translate this sentence as literally as possible into
Basque, it is once again John who went away, and not Bill, just as in
English.  This is so even though the ergative morphology of Basque is
more thoroughgoing than that of Dyirbal, which is split.

That is, Basque, like English, allows subjects to be coordinated with
subjects, but not with non-subjects.  This is so even when one of the
coordinated subject NPs is ergative and the other absolutive.  Basque
does not allow the absolutive subject of an intransitive sentence to be
coordinated with the absolutive object of a transitive sentence.

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk


From Georg at home.ivm.de  Mon Jul 19 11:42:32 1999
From: Georg at home.ivm.de (Ralf-Stefan Georg)
Date: Mon, 19 Jul 1999 13:42:32 +0200
Subject: accusative and ergative languages
In-Reply-To: <002201bed014$3c87ab00$8f9ffad0@patrickcryan>
Message-ID: <MON.19.JUL.1999.134232.0200.>

>In English, I can say: "John hit me, and he went away" or "I hit John and he
>went away". What is the discourse cohesion strategy for English?

Anaphora.

Stefan Georg
Heerstrasse 7
D-53111 Bonn
FRG
+49-228-69-13-32


From proto-language at email.msn.com  Mon Jul 19 12:51:38 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Mon, 19 Jul 1999 07:51:38 -0500
Subject: accusative and ergative languages
Message-ID: <MON.19.JUL.1999.075138.0500.>

Dear Miguel and IEists:

 ----- Original Message -----
From: Miguel Carrasquer Vidal <mcv at wxs.nl>
Sent: Friday, July 16, 1999 5:50 PM

> "Patrick C. Ryan" <proto-language at email.msn.com> wrote:

>> And to answer your -- I hope not purposeful -- distortion of what I wrote,
>> let me say explicitly that I did not assert "there are ergative languages
>> without (any) splits". I asserted that Thomsen did not, at least in her
>> grammar, identify splits in Sumerian, which you seemed to think she had.

Miguel wrote:

> Of course she has, unless your edition differs from mine: p. 51,
> paragraph 42:

> On the morphological level Sumerian has thus an ergative
>   system in the nouns and the intransitive vs. the
> transitive hamTu conjugation [...]. In the pronouns and
> the transitive maru^ conjugation vs. the intransitive
> verb, on the other hand, the system is nominative-
> accusative [...].
> This `split ergativity' is no uncommon phenomenon, in
> fact no ergative language is entirely ergative in both
> syntax and morphology.

Pat responds:

Yes, Miguel, she did write that. In the foregoing paragraph or so, she gives
a concocted example of a transitive maru: sentence:

za-e sag{~}-0 mu-zi.zi-en    you (sing.) raise the head

In this so-called "nominative-accusative" sentence, -e is the normal
termination of the ergative case, so, by form, za-e is 'you (sing.) ERG'; -0
is the normal termination of the absolutive case, so, by form, sag{~}-0 is
'head ABS'.

No, I suppose, it is also possible but fruitless to argue that -e is both
ERGATIVE and NOMINATIVE and -0 is both ABSOLUTIVE and ACCUSATIVE but what
real purpose does that serve except to play games?

I am also sure that you know that the termination -en of the verb above,
which is supposed to cross-reference the transitive maru: subject does *not*
exist in Old Sumerian (presumably before Akkadian scribes messed it up too
much).

One could always argue that the termination -e, which does exist in Old
Sumerian, was a cross-reference to the nominative subject. Is that what you
believe?

So, in conclusion, though Thomsen does mention them, she certainly did not
identify "splits" in Sumerian.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From proto-language at email.msn.com  Mon Jul 19 13:17:00 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Mon, 19 Jul 1999 08:17:00 -0500
Subject: accusative and ergative languages
Message-ID: <MON.19.JUL.1999.081700.0500.>

Dear Peter and IEists:

 ----- Original Message -----
From: petegray <petegray at btinternet.com>
Sent: Friday, July 16, 1999 1:30 PM

>> Pat said:
>> .... Why not list two or three additional elements that function in
>> Chinese equivalent to inflectional affixes?

Peter wrote:

> The most obvious omission from your list is the "directional" affixes,
> simple and complex, e.g.: lai (come); qi3-lai; etc
> And you can add zhe, guo, zheng4zai4, yao4, ...etc etc all showing different
> limitations to the action.

> Whether you count these as "inflexions" or as "affixes" or as common second
> elements in compound verbs, is probably a matter of taste.

> There are books of Chinese verbs that list all these things for you, but
> alas, I have none handy.

Pat responds:

I can only refer you to the definition of 'inflection' in Larry's _A
Dictionary of Grammatical Terms in Linguistics_.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From jer at cphling.dk  Mon Jul 19 19:23:11 1999
From: jer at cphling.dk (Jens Elmegaard Rasmussen)
Date: Mon, 19 Jul 1999 21:23:11 +0200
Subject: accusative and ergative languages
In-Reply-To: <Pine.GSO.4.03.9907161610401.19800-100000@rsunx.crn.cogs.susx.ac.uk>
Message-ID: <MON.19.JUL.1999.212311.0200.>

On Fri, 16 Jul 1999, [in continuation of a discussion with Par Ryan]
Larry Trask wrote:

> [...]
> But briefly: if a transitive sentence in an ergative language were
> "really" a passive, then its absolutive NP would be the subject and
> would exhibit subject properties.  But this [is] not the case in most of
> the ergative languages I have heard about.
[...]
> The "passive" view of ergative languages in general is indefensible.

I guess some key words are "really" and "split". If ergative syntax is
allowed to be limited in a language and may still be called ergative, why
can't the passive interpetation of it be allowed a similar latitude and
still be called passive? Is it not possible that some syntactic structures
that were originally passive (with all the trimmings that word entails)
_lost_ their passive value and simply became the normal way of saying
things (like a joke that gets repeated and isn't even meant to be funny
anymore) - and that, subsequent to this change, reflexive reference and
all other subject properties were transferred to the new pragmatic
subject? It does not make the analysis pretty, but if the linguistic
history consists of a period in which people changed their minds about the
basic roles of ordinary sentences, should an accout of the truth not be as
ugly as what really happened?

I do not know of many ergative-structured languages; but all I have seen
can be safely or highly probably analyzed as being passive circumlocutions
in origin, but none of them is completely so in terms of synchronic syntax
anymore. But that does not change their prehistory as passives. Is A
asking about origins, and B answering about descriptive synchrony?

Jens


From petegray at btinternet.com  Mon Jul 19 19:41:40 1999
From: petegray at btinternet.com (petegray)
Date: Mon, 19 Jul 1999 20:41:40 +0100
Subject: accusative and ergative languages
Message-ID: <MON.19.JUL.1999.204140.0100.>

Pat said:

>... the pattern "e-a-o" *...There was never a time
> during which *any* verb in [e] followed this "pattern", or even *any* verb
> in /eR/ followed it (weNden, waNdte, gewaNdt).

Perhaps I misunderstand you.   This pattern is one of the standard patterns
of strong verbs in modern German, most prevalent among verbs in -eR-, but
not restricted to them, e.g.:

befehlen, befahl, befohlen             to command
bergen, barg, geborgen                to salvage
nehmen, nahm, genommen          to take
sprechen, sprach, gesprochen     to speak
stechen, stach, gestochen             to stick
stehlen, stahl, gestohlen                 to steal
treffen, traf, getroffen                       to meet
verderben, verdarb, verdoben       to spoil
werben, warb, geworben                to recruit
and others

Peter


From W.Schulze at lrz.uni-muenchen.de  Tue Jul 20 10:15:16 1999
From: W.Schulze at lrz.uni-muenchen.de (Wolfgang Schulze)
Date: Tue, 20 Jul 1999 12:15:16 +0200
Subject: accusative and ergative languages
Message-ID: <TUE.20.JUL.1999.121516.0200.>

Dear Pat and IEists,

before turning to some points addressed by Pat, let me first say
something more general: Having read most of the contributions to this
list on the subject 'accusativity- ergativity' I got /and still have)
the impression that some of these contributions exhibit a considerable
ammount of redundancy regarding the arguments already addressed. This
may be due to the fact that semtimes new 'sub-threads' are opened or
special data discussed that provoke arguments already discussed in
another context. This fact renders the contributions sometimes rather
idiosyncratic and makes it difficult to react on ALL of them in a
non-redundant way. Sometimes, important arguments are hidden in such
'sub-threads' and hence do not receive the attendance they deserve. In
order to render our discussion a bit more straight-forward I would
propose that all contributors distribute a short (general) statement
about what they think is crucial in our discussion. Perhaps it is
helpfull if people (among others) refer to the following questions:

1. What is YOUR over-all definition of 'accusativity'?
2. What is YOU over-all definition of 'ergativity'?
3. Do YOU think of ACC vs. ERG as a parallelly weighted, polar
opposition? Or do YOU propose that one of the poles is more 'natural'
than the other? If yes: What is YOUR notion of 'naturalness', and how do
YOU substantiate your claim?
4. What is the distinction between 'morphological' and 'syntactic'
accusativity/ergativity according to YOUR paradigm?
5. Which role does 'markedness' play in YOUR arguments?
6. How do YOU define 'subject'?
7. What is - in YOUR mind - the correlation between case marking and
agreement with respect to ACC/ERG?
8. How do YOU classify a language with respect to ACC/ERG in case a) AGR
not present, b) CASE is not present, c) neither AGR nor CASE are
present?
9. Wich role des word order play with respect to a) core arguments, b)
peripheric (backgrounded) arguments in diathesis?
10. Describe YOUR scenario(s) of how a) ERG paradigms emerge, b) ACC
paradigms can emerge (cf. 3)!
11. What are - in YOUR opinion - the most crucial aspects of ACC/ERG: a)
semantic (conceptual), b) syntactic (sentence organization), c)
pragmatic (discourse organization, information flow etc.)?
12. Do YOU claim that one (or more) of the parameters mentioned in (11)
are specifically relevant for either ACC or ERG (e.g., ERG _> semantic,
ACC -> pragmatic etc.)?
13. Which role do split systems play in your arguments (Split-S,
Fluid-S, Split-O, Fluid-O)?
14. How do you define ACC/ERG in polypersonal / poly-congruent languages
without case marking? (cf. (7))?
15. Which are the languages that YOU use in your arguments? Do you use
a) reference books, b) texts, c) informants?
16. Do YOU use a special theoretical frame work in order to substantiate
your claims? If yes, which one?

*********************

Now, let me finally turn to some claims by Pat:

"Patrick C. Ryan" schrieb:

Wolfgang wrote:

>> Hence, there are NO "ergative (or "accusative") languages" or
>> only, if you use this term in a very informal sense.

> Pat writes:

> In the sense you are using these, they seem to be of little value in
> describing anything.

I can only repeat what I have said in my earlier postings: ACC and ERG
are nothing by the structural (and structured) reaction to more general
principles of language based information processing. The polycentric
architecture of language systems that encode these principles allows
that the individual centers of this polycentric cluster react
differently on these principles. Here, I cannot elaborate the underlying
frame work which is labeled "Grammar of Scenes and Sceanrios" (GSS) and
documented in Schulze 1998, chapter IV, but let me briefly say that I
propose a (more or less) universal cause-effect 'vector' (C->E) itself
metaphorized from underlying figure-ground relations (F->G) to be one of
the most dominant principles of information processing. This vector can
be weighted which leads to a continnum (here shortened) C->0(e) C->e >
C->E > c->E > 0(c)->E (capitals represent heavy domains, small letters
represent light domains). An ACC strategy would be to behave in a C->e
sense, an ERG strategy would infer c->E. Note that these vector
representations are NOT (in themselves) language specific or sensitive
for specuial (linguistic) categories! Different categories may BEHAVE
differently with respect to this continuum, regardless their own
architectural make-up. ANYTHING in a language system may be sensitive
for the AEC (accusative ergative continuum) as long as it is relevant
for encoding the cause-effect vector (and its derivations). Hence, CASE
may play a role just as AGR, word order, subject assignment,
topicalization, discourse cohesion, co/subordination, paradigmatization
of speech act participants and much more. But they may play their roles
DIFFERENTLY! The description of their roles heavily depends from the
diachrony of the paradigm in question, its formal architecture as well
as its integration in aa co-paradigmatic context ('structural coupling'
in a broader sense). From this it follows that the individual centers of
ALL language systems have to react upon the universal demands of the
C->E vector, disregarding their internal architecture. Only IF ALL
relevant centers (and they are many, I grant!) behave in one direction,
THEN we are allowed to call the language system (or better, its
Operating System) ACC or ERG. However, it is a much more difficult task
to describe intermediate states that allow ACC in some parts of the
Operating System, and ERG in some others. Here, we have to establish a
(motivated!) hierarchy first of co-paradigmtic structures (such as CASE
and AGR, CASE and word order, AGR and Personality, AGR and Noun Classes,
to name only some). In such structures, one parts sometimes is more
dominant than the other with respect to its behavior on the AEC. If we
can describe such dominant behavior we can refer to the whole structure
as either more ACC or ERG. In a second step we have to go on describing
the higher levels of this hierarchy (which itself should find an
adequate linguistic explanation based on an appropriate language
theory). Finally (and idealiter) we would arrive at a term that would
describe the functional dominance of one of the poles on the AEC with
respect to an Operating System (not a language system) in toto. Only
the, and I stress, only THEN we are allowed to use the term ergative or
accusative with respect to an Operating System (for which 'ergative' or
'accusative language' would be an informal label).

> Wolfgang wrote further:

> Pat writes:

> I am sorry that I do not agree with the validity of this distinction (actant
> vs. agent). For me, 'actant' is 'agent'. Perhaps you can explain the
> difference.

In terms of Functinal Grammars (as well as in GSS) 'actants' refer to
ALL such linguistic expressions that encode a referential entity in a
clause. Hence, in a senetcne such as 'I met John several times in
Chicago', 'I', 'John', 'times', and 'Chicago' are (abstract) actants
that play different roles in the scene. But only 'I' is a linguistic
agent, whereas John plays the role of a patient etc. Note that 'agent'
and 'patient' are labels for semantic hyperroles (or macroroles in the
sense of Foley/VanValin). I think that such a distinction is very
helpfull. It is based on strong theoretical arguments and helps to avoid
many false or at least problematic generalizations. A much more
controverse (and much more difficult question is to define the labels
'subjective' (S), 'agentive' (A), and 'objective' (O) which should not
be immediately equated to neither 'subject'/'object' nor to
'agent'/'patient'. S, A, and O are highly abstract terms that describe
more structural than semantic or syntatic properties.

> Wolfgang wrote:

>> The fact, however, is that many 'ergative' languages lack an
>> antipassive. For instance, there are nearly 30 East Caucasian languages
>> all of them using some ergative strategies in at least parts of their
>> operating systems. But only a handfull of them (five or six, to be
>> precise) have true antipassives (only one has some kind of
>> "pseudo-passive").

> Pat writes:

> Are you asserting that the majority of ergative languages do not have
> anti-passives?

I do not assert anything in the sense of 'ALL language have...'. Even
the claim 'the majority of ergative languages (sic!) have...' is rather
suspect to me. What I said is that in those 'ERG systems' I looked at
(about 200) antipassives are ratgher the exception than the norm.

> Pat writes:

> As for the far-reaching conclusions of Dixon based on discourse conhesion
> strategies, in my opinion they are flawed because these strategies are
> purely conventional.

Language systems ARE conventional! This is one of the major points of
language tradition and L1 acquisition (despite of minimalism etc.).
Discourse probably is one of the most important factors in the
emergence, organization, and dynamics of language systems. We should not
refer to the abstract notion of context-free 'sentences' that would be
responsible for for grammatical 'events'. Such a view stems from the
tradition of Classical Philosophy which is an INTERPRETATION of what
goes on language. Today, we have become used to think of language in
single sentences, to brak them up the way we do etc. But this is an
analytic tradition, not part of the ontology of language itself, which
is much more synthetic in ature than we are used to think. - A sentence
does not function but in its co-text (as well as in its con-text). All
sentence internal strategies used to be embedded in the techniques of
co(n)textualization. No wonder, that ACC and ERG also work in this
direction (though they may appear as more 'autonomous',
sentence-internal mechanisms secondarily, especially if a language
system as developed separate means to indicate discourse cohesion).

> In English, I can say: "John hit me, and he went away" or "I hit John and he
> went away". What is the discourse cohesion strategy for English?

See the standard literature on this (its counts legion!).

> Pat writes (on Lak, East Caucasian):

> I have to plead ignorance of Lak however, I find an analysis of an otherwise
> completely ergative sentence as having an ACC word-order impossible. I
> believe this opinion rests on a false analysis of word-order significance.

I tried to show that in Lak only some sentence patterns show 'complete
ergativity'. Most of them sho a mixed paradigmatic organization with
respect to moprhology (CASE, AGR). WOrd order is another VERY imporant
aspect of the AEC. Consider e.g. a language that ahs canonical

#SV
#AOV

(# = sentence boundary). Here, S behave like A with respect to #, hence
it is ACC. In #SV vs. #OAV it is S=O which indicates ERG behavior. The
problem of actance serialization is also addressed in polypersonal
systems (without CASE, e.g, Abxaz, Lakotha etc.). It IS a very important
indicator for ACC, ERG, no question. Anything else would refer to the
linguistic tradition of say 60 years ago (Bloomfieldian tradition).

> Wolfgang continued (on Lak):

>> But if you say "I am surely hitting you (plural)", you get:

>> na b-at-la-ti-s:a-ra zu
>> I:ABS I:PL-hit1-DUR-hit2-ASS-SAP:SG you:PL:ABS

>> Here we have:

>> ACC with respect to word order
>> ERG with respect to class agreement
>> ACC (or neutral) with respect to case marking)
>> ACC with respect to SAP agreement (-ra is triggered by 'I:ABS').

> Pat writes:

> So, agent marked ABS and patient marked ABS is, according to you, ACC with
> respect to case marking? Sorry, not convincing at all! And I think the SAP
> difference can be explained as indicating seriality of the patients involved
> as recipients of the action by *one* agent.

I said ACC (or neutral!) with respect to case marking! That means that A
lacks ERG marking (which is OK with Silverstein). Semantically, it means
that S is encoded like A (which is ACC). That fact that O is NOT in an
accusative-like case does not argue against this assumption, because
CASE is here structrually coupled with ADGR which allows us to classify
AGR as ACC *here*. The SAP difference in Lak cannot be explained in the
sense Pat proposes, Lak totally lacks such strategies.

> Wolfgang wrote further:

>> Now, please tell me: Is Lak an 'ergative' or an 'accusative' language?
>> [Please note that I did not include (among others) strategies of
>> discourse cohesion, reflexivization and logophization].

> Pat writes:

> By these examples, I would tell you that Lak is 'ergative'; and that
> accusativity is not demonstrable from these examples.

If you refer to the standard (morphological), but rather obsolete
interpretation of ERG you may be right (but not for SAP NPs). But,
fortunately, the ACC/ERG typology has freed itself from such a narrow
interpretation of ACC/ERG which is nothing but a very small excerpt from
the over-all typology.

> Pat wrote previously:

>>> I think it is likelier that, because of perceived greater animacy (or
>>> definiteness), pronouns have a different method of marking that can still
>>> be interpreted within an ergative context.

> Wolfgang wrote:

>> This a (very simplified) 'on-dit' that stems from the earlier version of
>> the Silverstsein hierachy. Again, we have to deal with the question,
>> whether a 'pronoun' (I guess you mean some kind of 'personal pronouns')
>> can behave 'ergatively' or 'accusatively'. The list below gives you a
>> selection of SAP case marking in East Caucasian languages with respect
>> to ABS/ERG:

>> ABS vs. ERG ABS = ERG
>> ALL ---
>> Singular Plural
>> Plural Singular
>> 1.Incl. Rest
>> 1:SG Rest
>> 2:SG Rest
>> 1:SG/PL Rest
>> --- ALL

>> This list (aspects of personal agreement NOT included!) shows that SAP
>> pronouns may behave different within the same paradigm. Any
>> generalization like that  one quoted above does not help to convey for
>> these data...

> Pat writes:

> I do not have a reference book for Lak so that my hands are somewhat tied.
> But, I have found that paradigms are often inconsistent in ways that reflect
> earlier lost phonological changes, or other lost schemata. The locative
> plural terminations of IE are certainly not, in origin, terminations of the
> locative plural. Etc.

Nothing the like! The list I gave refers to what can be described for
East Caucasian languages in toto! Most of these paradigms are
functionally motivated, I grant (schulze in PKK 2 (Schulze 1999) will
tell you the whole story).

Wolfgang

[Please note new phone number (office) :+89-2180 5343]
___________________________________
| Prof. Dr. Wolfgang Schulze
| Institut fuer Allgemeine und Indogermanische Sprachwissenschaft
| Ludwig-Maximilians-Universitaet Muenchen
| Geschwister-Scholl-Platz 1
| D-80539 Muenchen
| Tel:	+89-21802486 (secr.)
|      	+89-21805343 (office) NEW ! NEW !
| Fax:	+89-21805345
| Email: W.Schulze at lrz.uni-muenchen.de
| http://www.lrz-muenchen.de/~wschulze/
_____________________________________________________


From vidynath at math.ohio-state.edu  Tue Jul 20 11:29:04 1999
From: vidynath at math.ohio-state.edu (Vidhyanath Rao)
Date: Tue, 20 Jul 1999 07:29:04 -0400
Subject: accusative and ergative languages
Message-ID: <TUE.20.JUL.1999.072904.0400.>

Jens Elmegaard Rasmussen <jer at cphling.dk> wrote:

> I'd say the cases I have seen of ergativity in Indic and Iranian languages
> so clearly reflect underlying/earlier passive circumlocutions that
> controversy is absurd.

No!

I have already pointed out the problem of distinguishing resultatives and
passives. The ta-adjective is resultative in Vedic (Jamison, IIJ 198?) and
Early Pali (Hendrikson, Infinite verb forms of Pali). And it patterns
ergatively as resultative participles often do. The last part has been known
forever. In English, see Speyer's ``Sanskrit Syntax'', will find the
following: ``Of the participles in -ta the great majority have a passive
meaning, hence it is customary to call the whole class the passive
participle of the past. But some others are not passives, but intransitives,
as gata (gone), m.rta (died) [rather dead, VKR] bhinna (split). Some again
may even be transitive actives, as pi:ta (having drunk) [better drunk, but
without the restriction to a special meaning in English] ...'' [para 360, p.
280].  Speyer goes on to note a:ru:d.ha has active meaning more commonly.
This is not how we expect the passive to behave. But for resultative, it is
understandable. The ergative patterning is based on pragmatics. But in a
culture that considers it more worthy of note whether a man is mounted on a
horse or a vehicle than whether a horse or a vehicle is carrying someone, it
makes sense to use a:ru:d.ha in the active sense.

The difference is unmistakable in the following:

(1) ra:mo 's'vam a:ru:d.hah.
    R is mounted on [a] horse.

(2) ra:men.a+as'vam a:ruhyate
   [A] horse is being mounted by R.

(3) *ra:mo 's'vam a:ruhyate

(1) and (2) are quite grammatical and examples of easy to come by. (3) does
not occur and is underivable in traditional grammar. As I explained above,
this is understandable if a:ru:d.ha is resultative. If we take the passive
view, how do we explain the fact that (1) is acceptable but (3) is not?
Without such an explanation, it is far from absurd to contest the passive
interpreation.


From adahyl at cphling.dk  Wed Jul 21 16:59:02 1999
From: adahyl at cphling.dk (Adam Hyllested)
Date: Wed, 21 Jul 1999 18:59:02 +0200
Subject: accusative and ergative languages
In-Reply-To: <Pine.GSO.3.95.990716160742.15501B-100000@unagi.cis.upenn.edu>
Message-ID: <WED.21.JUL.1999.185902.0200.>

On Fri, 16 Jul 1999, Sean Crist wrote:

> Somebody (I've lost track of who) wrote:

>>> Yes, but if a new verb were formed, e.g. webben (from English 'web'), there
>>> is no possibility that it would be conjugated webben, wabb, gewobben.

> This sort of thing actually _does_ happen sometimes; it just isn't that
> common.  The English verb "strive" is a loan from Old French; but it is
> often declined as if it were an old Class I ablauting strong verb (strove,
> striven).  So I'd correct "no possibility" to "much less likely".

Another example is the Danish verb <skrive> "to write",
ultimately from latin <scribere>, but always conjugated as an
ablauting strong verb: pres. <skriver>, pret. <skrev> (and not
<**skrivede>. However, this way of conjugating is no longer productive.

Adam Hyllested


From larryt at cogs.susx.ac.uk  Mon Jul 19 09:31:41 1999
From: larryt at cogs.susx.ac.uk (Larry Trask)
Date: Mon, 19 Jul 1999 10:31:41 +0100
Subject: Interpreting ergative sentences
In-Reply-To: <007201becdef$67235840$6701703e@edsel>
Message-ID: <MON.19.JUL.1999.103141.0100.>

On Wed, 14 Jul 1999, Eduard Selleslagh wrote:

> Speaking about Castilian, I would like to have your views on a peculiarity,
> or rather a tendency, that is still productive at least in popular speech
> ('le-ismo'), namely the tendency to use the indirect subject form where all
> other Latin (and other West-European) languages use the direct object form,
> and almost exclusively with persons (animate), e.g. "le v?" ("le vi' "), "I
> saw him".

> It has sometimes be suggested that this was a substrate influence from
> Basque, an ergative language (that lacks an accusative, of course), but
> opinions are extremely divided on this subject. 

Indeed.  I have seen this suggestion in print, and I've replied to it in
print.  The observation is that Castilian fails to distinguish direct
and indirect objects as sharply as other Romance languages.  The
suggestion is that this results from Basque substrate influence.  But
this suggestion would only make sense if Basque were a language that
also failed to distinguish direct and indirect objects sharply.  And
it's not.

In Basque, direct and indirect objects are sharply and absolutely
distinguished in all respects.  The two differ in case-marking, in
verbal agreement and in ability to be passivized.  This is true in all
circumstances without exception.  In fact, I would say that Basque
distinguishes direct and indirect objects more sharply than any Romance
language I know of, and more sharply than English.  (English has
sentences like `My wife gave me this book', in which it is a moot point
what kind of object`me' might be.)

Accordingly, I can't for the life of me see how Basque substrate
influence could be invoked to account for the Castilian facts.  True,
Basque has no accusative case, but it does have a very well-defined
class of direct objects and a very well-defined class of indirect
objects, and these two classes are entirely distinct in every
morphological and syntactic respect I can think of.  Given what I've
read about objects in the literature, I would say that, if anything,
Basque is rather unusual in its absolute distinction between the two
kinds of object.

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk


From larryt at cogs.susx.ac.uk  Mon Jul 19 11:29:14 1999
From: larryt at cogs.susx.ac.uk (Larry Trask)
Date: Mon, 19 Jul 1999 12:29:14 +0100
Subject: Ergative & Basque
In-Reply-To: <Pine.SO4.4.05.9907161834110.6415-100000@rask>
Message-ID: <MON.19.JUL.1999.122914.0100.>

On Sat, 17 Jul 1999, Jens Elmegaard Rasmussen wrote:

[on my Basque example with `have' as the transitive auxiliary]

> If there is no passive expressed as belonging to the agent in this,
> what then is the "have" verb doing here?

The Basque construction is parallel in form to the `have' perfects and
past-tenses found in a number of Romance and Germanic languages (at
least).  In English, `I have drunk the wine' is still strictly a
perfect, with no past-tense reading possible.  In French, the similarly
constructed <J'ai bu le vin> is primarily an ordinary past tense, though
a perfect reading is possible.  Other languages exhibit various
intermediate stages here.  Castilian Spanish <(Yo) he bebido el vino> is
interesting in that it functions both as a perfect and as a hodiernal
past (earlier today).  Basque <Ardoa edan dut> works like Castilian.

All these languages also allow a past-tense auxiliary, as in English
`I had drunk the wine', but again the precise value of this varies from
language to language.  In Basque it functions both as the ordinary past
tense (before today) and as a pluperfect.

> Can't the underlying construction be analyzed in a sensible way at
> all?

Well, I'm not sure what you mean by "a sensible way".  What would you
regard as a sensible analysis of English `I have drunk the wine'?

> A silly question from a complete outsider: is the translation "have"
> internally motivated? Can it express the "having" of anything other
> than a "participled" object?

Yes.  In all varieties of Basque, the auxiliary used in constructing
periphrastic forms of transitive verbs is the defective verb *<edun>,
which has no non-finite forms and whose non-finite forms are supplied
suppletively by other verbs.  In the east, this *<edun> is the ordinary
lexical verb `have', as in `I have blue eyes' or `I have a new car'.

Eastern varieties have another verb, <eduki>, which means `hold', `hold
on to', `have in one's hand', `grasp', `clutch'.  In western varieties,
however, *<edun> is generally specialized as the transitive auxiliary,
and <eduki> is the ordinary lexical verb for `have'.

Now, since *<edun> is also recorded for `have' in early texts in the
west, and since it is still possible for `have' in elevated styles in
the west today, we may reasonably surmise that *<edun> was once the
ordinary lexical verb for `have' everywhere, and that its replacement by
<eduki> in the west is an innovation there.

We may further surmise that this western innovation is a calque on
Castilian.  In Castilian, the inherited lexical verb for `have' is
<haber>, but this is now entirely specialized as an auxiliary.  The verb
<tener>, which originally meant `hold, hold on to, grasp', has become
the ordinary lexical verb for `have'.

This proposal of a calque on Castilian can be supported by another case.
In all varieties of Basque, the verb used as an auxiliary for
constructing periphrastic forms of intransitive verbs is <izan>.
Now, in eastern varieties, this <izan> is also the ordinary copular verb
`be', as in `I am a teacher' or `Your books are on the table'.

However, eastern varieties have a second verb, <egon>, whose earlier
sense is `wait', `stay', `remain'.  At least in the imperative, it still
has this sense in all varieties.  In the east, though, it has acquired a
second sense, that of copular `be' *in the following circumstances
only*: the subject is animate, and the predicate is locative or
comitative (the Basque comitative itself derives from a locative
phrase).  So, in the east, `Mary is in the kitchen', `The men are in the
fields' and `Mary is with John' all require <egon>, not <izan>.

In western varieties, however, the use of <egon> has been generalized:
it is used for `be' in *all* locative expressions, and also with all
predicates denoting states or conditions.  So, in western varieties,
`Your books are on the table', `The Guggenheim Museum is in Bilbao',
`Mary is happy', `Mary is asleep', `Mary is drunk', and so on, require
<egon>, whereas eastern varieties use <izan>.

This western usage is clearly calqued on the famous Castilian
distinction between <ser> `be' (unmarked) and <estar> `be' (in a place
or in a state).  Wherever Castilian uses <estar>, western Basque uses
<egon>.  This is true even in idiosyncratic cases.  For example,
Castilian expresses `He's dead' as <Esta' muerto>, with <estar>, and
western Basque likewise has <Hil dago>, with <egon>, while eastern
Basque has <Hila da>, with <izan>.

And this, I think, is as far back as we can go in tracing the prehistory
of the Basque periphrastic verb-forms.  As I remarked earlier, we have
no way of knowing whether these things originated as calques on Romance
or whether they are independent creations in Basque, from an unknown
source.  All I can tell you is that the modern periphrastic forms were
clearly already established in Basque by the tenth century, when we find
the first recorded verb-forms.

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk


From connolly at memphis.edu  Mon Jul 19 20:35:22 1999
From: connolly at memphis.edu (Leo A. Connolly)
Date: Mon, 19 Jul 1999 15:35:22 -0500
Subject: accusative and ergative languages
Message-ID: <MON.19.JUL.1999.153522.0500.>

Larry Trask wrote, replying to Pat:

> But briefly: if a transitive sentence in an ergative language were
> "really" a passive, then its absolutive NP would be the subject and
> would exhibit subject properties.  But this not the case in most of the
> ergative languages I have heard about.  In spite of the ergative
> case-marking and/or verbal agreement, it is usually the *ergative* NP
> which exhibits subject properties, not the absolutive NP.

Agreed -- and yet not quite agreed.  Are these really "subject
properties"?  I freely acknowledge that that's what they're called, but
for mainly
for lack of any better name in normal linguistic terminology.

It might be better to take the tack that I did in my recent post "On
interpreting ergative languages": that one must distinguist between a
"morphological subject" (the absolutive or nominative, or positional
equivalent in languages such as English), and the "highest-ranking NP"
in a hierarchy of the Case Grammar sort.  Then it turns out that even in
accusative languages, some properties adhere to the morphological
subject, others to the HRNP, the split varying greatly from one language
to another.  In most ergative languages (which can then be defined
simply as languages with extensive ergative casemarking), it is the HRNP
that has all or most of the syntactic "subject properties".  But there
is no particular reason to call it a "subject".

True enough: to some extent, this is a matter of terminology.  But I
think it matters in this sense: unlike the morphological subject, the
HRNP is not constituted, but is given by the Case Hierarchy; it may,
however, be demoted by use of a passive construction.  "Active"
languages are then explained naturally: they have no morphological
subject, only the HRNP, whose realization notoriously varies.

	[stuff omitted]

> There are unusual languages, of course.  In Dyirbal, and in
> Nass-Gitksan, the absolutive NP in a transitive sentence shows at least
> some subject properties, while the ergative NP does not show the same
> properties -- though it does show other subject properties.  This is the
> phenomenon we call `syntactic ergativity'.  But I know of no language
> which is syntactically ergative without exception, and I know of few
> languages which are syntactically ergative at all.

I have gone through Dixon's book on Dyirbal pretty thoroughly and don't
recall any accusative or active syntax, only a few accusative pronouns
(which exhibit ergative syntax).

> The "passive" view of ergative languages in general is indefensible.

Absolutely agreed!

Leo


From proto-language at email.msn.com  Mon Jul 19 04:17:28 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Sun, 18 Jul 1999 23:17:28 -0500
Subject: Recoverability
Message-ID: <SUN.18.JUL.1999.231728.0500.>

Dear Sean and IEists:

 ----- Original Message -----
From: Sean Crist <kurisuto at unagi.cis.upenn.edu>
Sent: Thursday, July 15, 1999 11:37 PM

> You need multiple examples of sound correspondences to be able to conduct
> the Comparative Method at all; and when the the cognates become as
> rarefied as they are this time-depth, the likelihood of having access to
> an adequate number of examples to work out the relevant sound changes
> becomes proportionately smaller, eventually reaching what for practical
> purposes is an impossibility.

Pat interjects:

"Rarefied" is a characterization only possible when one refuses to look at
the actual facts. An adequate number of examples to establish cognation is
furnished in the studies on Uralic at my website:

http://www.geocities.com/Athens/Forum/2803

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From jer at cphling.dk  Tue Jul 20 01:38:14 1999
From: jer at cphling.dk (Jens Elmegaard Rasmussen)
Date: Tue, 20 Jul 1999 03:38:14 +0200
Subject: Recoverability
In-Reply-To: <Pine.GSO.3.95.990715232942.17263C-100000@unagi.cis.upenn.edu>
Message-ID: <TUE.20.JUL.1999.033814.0200.>

On Fri, 16 Jul 1999, Sean Crist wrote:

> [...]
> I think the point was something more like this.  Suppose for the sake of
> argument that there is in fact a genetic relationship between
> Proto-Indo-European and Proto-Uralic (to take one possible example). Even
> if this were true, and given the evidence which we actually do have at
> hand, would we expect to be able to _show_ that there is such a relation?

> Here, the problem _is_ like trying to reconstruct PIE on the basis of only
> modern English and Albanian.

Yes it is - and one would not get very far that way. One just might
discover the relationship, though: Alb. jam 'am', e"shte" 'is', nate"
'night', dere" 'door', tre, tri 'three', me" 'me', ju 'you'. If English
had had some proper morphology, one would have been able to go further.
Collinder and Cop (the latter with c-hachek) did that for "Indo-Uralisch".
Now of course large-scale comparison is not restricted to IE and Uralic,
but includes quite a few extra branches into Nostratic - and the more
there is to draw on, the greater the chance of finding links. I think it's
looking good.

(Sorry if I shifted the focus of the question in my posting.)

Jens


From connolly at memphis.edu  Mon Jul 19 20:39:35 1999
From: connolly at memphis.edu (Leo A. Connolly)
Date: Mon, 19 Jul 1999 15:39:35 -0500
Subject: accusative and ergative languages
Message-ID: <MON.19.JUL.1999.153935.0500.>

Sean Crist wrote:

> Somebody (I've lost track of who) wrote:

>>> Yes, but if a new verb were formed, e.g. webben (from English 'web'), there
>>> is no possibility that it would be conjugated webben, wabb, gewobben.

> This sort of thing actually _does_ happen sometimes; it just isn't that
> common.  The English verb "strive" is a loan from Old French; but it is
> often declined as if it were an old Class I ablauting strong verb (strove,
> striven).  So I'd correct "no possibility" to "much less likely".

Well, maybe.  Trouble is, the word is of Germanic origin (cf. German
_streben_), so the strong inflexion *might* actually be original but
unattested in OE.

Leo


From proto-language at email.msn.com  Mon Jul 19 04:31:05 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Sun, 18 Jul 1999 23:31:05 -0500
Subject: PIE vs. Proto-World (Proto-Language)
Message-ID: <SUN.18.JUL.1999.233105.0500.>

Dear Joat and IEists:

 ----- Original Message -----
From: <JoatSimeon at aol.com>
Sent: Friday, July 16, 1999 11:53 AM

 Joat wrote:

> -- with minor exceptions (such as "kuku"), words _are_ arbitrary sound
> assemblages.

Pat writes:

Why do you not explain to us all why that is true? I am firmly convinced
that it is unequivocally incorrect.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From proto-language at email.msn.com  Mon Jul 19 05:21:18 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Mon, 19 Jul 1999 00:21:18 -0500
Subject: PIE vs. Proto-World (Proto-Language)
Message-ID: <MON.19.JUL.1999.002118.0500.>

Dear Ralf-Stefan and IEists:

 ----- Original Message -----
From: Ralf-Stefan Georg <Georg at home.ivm.de>
Sent: Friday, July 16, 1999 4:28 AM

Ralf-Stefan wrote:

> And, if I may insert this much to Pat's chagrin, the notion that human
> language is a monogenetic phenomenon is aprioristic ideology. Even the
> notion that every known language, as Basque, Burushaski or whatnot, has to
> be related to some other language is ideology. "Having consonants" may be a
> universal feature of human language, "Being related to some other lg."
> simply is not.

Pat responds:

There is no *tangible* way for us to ever know whether languages arose
monogenetically or polygenetically however most linguists, even when they
deny its recoverability, have correctly weighed the odds of mono- vs.
polygenesis, and subscribe to monogenesis. If a probablistically calculated
hypothesis is "ideology", then everything done in historical linguistics is
"ideology".

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From proto-language at email.msn.com  Mon Jul 19 05:48:42 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Mon, 19 Jul 1999 00:48:42 -0500
Subject: PIE vs. Proto-World (Proto-Language)
Message-ID: <MON.19.JUL.1999.004842.0500.>

Dear Larry and IEists:

 ----- Original Message -----
From: Larry Trask <larryt at cogs.susx.ac.uk>
Sent: Friday, July 16, 1999 10:06 AM

Larry wrote:

> And the same goes for "Proto-World", or whatever.  Until a persuasive
> case has been made that *all* of the world's 6000 or so known languages
> are genuinely related, there is no point in attempting a
> "reconstruction" of "Proto-World".  The result of such a rash attempt
> can be no more than legerdemain.  Of course you can show that this, that
> and the other *might* have a common ancestor, but you can do this in
> countless entirely different ways, none of them superior to any other,
> and you cannot show that these things really *are* related.

Pat writes:

The "persuasive case" has already been made. Monogenesis is much likelier
than polygenesis. And you have written so yourself!

What mitigates against it is incorrect ideas of vocabulary (*CVC-root) loss,
mistaken applications of Kinderlallsprache, loan specialists that would make
the Rothschilds wild with envy, and fuzzy concepts like your favorite
"expressive" --- all of which seeming to foredoom its recoverability.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From Georg at home.ivm.de  Mon Jul 19 09:36:29 1999
From: Georg at home.ivm.de (Ralf-Stefan Georg)
Date: Mon, 19 Jul 1999 11:36:29 +0200
Subject: PIE vs. Proto-World (Proto-Language)
In-Reply-To: <001801becf8b$dc828e60$869ffad0@patrickcryan>
Message-ID: <MON.19.JUL.1999.113629.0200.>

>Pat interjects:

>Here I think you are dangerously introducing the mistaken terminology of the
>opposing argument. I do *not* look for "look-alikes"; I only am interested
>in "cognates" the phonological forms of which can be supported through
>multiple comparisons. For example, one set of interesting IE and Sumerian
>"cognates" shows IE *-wey- = Sumerian -g{~}-; interestingly, this
>development *is* found in *some* IE languages, like nearby Armenian.

"interestingly" ? Why interestingly ? The only way this could be
"interesting" would, imho, be to base a claim of Leskienian solidaric
innovation on it, leading to a Sumero-Armenian subgroup of Indo-European.
Very interesting.
Or some kind of areal phenomenon, potentially interesting (this time
without irony). But: supposing for a split-second that IE and Sumerian are
somehow related, the /w/ --> /g/ shift would be something very old, right ?
Now, the Armenian /w/ ---> /g/ shift is young, since known loanwords
participate in it (so young actually, that Kartvelian managed to preserve
the intermediary stage /gw/ in some of those LWs). And of course, even if
chronology were no obstacle, the whole thing still begs the question of
Sumero-IE, but I understand that this is a) not a topic for this list and
b) that I don't stand the slightest shred of a chance to talk you out of
it; so I don't try.

St.G.

Stefan Georg
Heerstrasse 7
D-53111 Bonn
FRG
+49-228-69-13-32


From rmccalli at sunmuw1.MUW.Edu  Mon Jul 19 21:27:49 1999
From: rmccalli at sunmuw1.MUW.Edu (Rick Mc Callister)
Date: Mon, 19 Jul 1999 16:27:49 -0500
Subject: PIE vs. Proto-World (Proto-Language)
In-Reply-To: <Pine.GSO.4.03.9907161556400.19800-100000@rsunx.crn.cogs.susx.ac.uk>
Message-ID: <MON.19.JUL.1999.162749.0500.>

	You raise the questions of what is relationship and what is the
difference between language and speech.
	All languages are related in the same sense that all humans are
related. All humans settled on oral speech rather than sign language or
flatulence for communication. Given that all human brains are wired for
speech in the same way, this suggests that there is a degree of
relationship.
	The question is whether the time span between speech and language
allowed languages to develop separately or whether quickly developed from
that single event.
	I don't think this question can be answered in absolute terms but
it may possibly be answered in practical terms. Given that modern humans
left Africa about 100,000 years ago, there is a good chance that all of
these are ultimately related. This presupposes that language had developed
before humans left Africa. DNA studies obtensibly show that non-African
humans seem to go back a single population distinct from Africans. While
one cannot prove that the members of this group spoke related languages,
it's more likely than not. If one establishes that non-African languages
are related, only Nilo-Saharan, Niger-Kordofanian, Khoisan languages [and
possibly Afro-Asiatic] are left.
	The next step is to prove this, and I'll leave that all of you :>

[snip]

>And the same goes for "Proto-World", or whatever.  Until a persuasive
>case has been made that *all* of the world's 6000 or so known languages
>are genuinely related, there is no point in attempting a
>"reconstruction" of "Proto-World".  The result of such a rash attempt
>can be no more than legerdemain.  Of course you can show that this, that
>and the other *might* have a common ancestor, but you can do this in
>countless entirely different ways, none of them superior to any other,
>and you cannot show that these things really *are* related.

>Larry Trask

[ moderator snip ]

Rick Mc Callister
W-1634
Mississippi University for Women
Columbus MS 39701


From adahyl at cphling.dk  Wed Jul 21 15:57:04 1999
From: adahyl at cphling.dk (Adam Hyllested)
Date: Wed, 21 Jul 1999 17:57:04 +0200
Subject: PIE vs. Proto-World (Proto-Language)
In-Reply-To: <l03102800b3b4aa9358c0@[195.247.226.14]>
Message-ID: <WED.21.JUL.1999.175704.0200.>

On Fri, 16 Jul 1999, Ralf-Stefan Georg wrote:

> "Having consonants" may be a
> universal feature of human language, "Being related to some other lg."
> simply is not.

A fascinating thought:
Imagine "Being related to some other lg." WAS actually a universal feature
of (spoken) human language, wouldn't it then be the only true universal?
In other words, if all spoken languages should prove to be derived from
one common source, would we then be able to distinguish true language
universals from other features shared by all languages ONLY because of the
fact that these features were present in "proto-world"? Language typology
would still be a great help in the work of reconstructing proto-languages,
but would it prove anything about the nature of spoken human language?

The above are only questions, not statements...

Adam Hyllested


From adahyl at cphling.dk  Wed Jul 21 16:17:15 1999
From: adahyl at cphling.dk (Adam Hyllested)
Date: Wed, 21 Jul 1999 18:17:15 +0200
Subject: PIE vs. Proto-World (Proto-Language)
In-Reply-To: <Pine.GSO.4.03.9907161556400.19800-100000@rsunx.crn.cogs.susx.ac.uk>
Message-ID: <WED.21.JUL.1999.181715.0200.>

On Fri, 16 Jul 1999, Larry Trask wrote:

> But you can't just pick some languages that catch your eye and then
> "reconstruct" a common ancestor for them.  That is, you can't pick, say,
> Zulu, Sumerian and Korean and "reconstruct" Proto-ZSK.  You must first
> make a good case that a common ancestor for these languages is
> preferable to any other scenario.

Exactly. This is why people shouldn't always neglect distant language
relationship using the argument that no sound correspondences have been
set up. Every scientific project has a primitive beginning.

Adam Hyllested


From adahyl at cphling.dk  Wed Jul 21 17:14:09 1999
From: adahyl at cphling.dk (Adam Hyllested)
Date: Wed, 21 Jul 1999 19:14:09 +0200
Subject: PIE vs. Proto-World (Proto-Language)
In-Reply-To: <1e1b93d4.24c0bd9b@aol.com>
Message-ID: <WED.21.JUL.1999.191409.0200.>

On Fri, 16 Jul 1999 JoatSimeon at aol.com wrote:

> Our picture of PIE would be very different, and much less
> detailed, if we had no written records of extinct languages.

> Hell, the original insight that the IE languages were derived from a common
> ancestor was made by comparting Sanskrit with Classical Greek and Latin --
> all extinct languages!

> The link between Sanskrit and Latin or Classical Greek is so close that any
> layman can see it; whole phrases are nearly identical.
>
> This is, to put it mildly, not the case between, say, English and Bengali.

X
But if we only knew the modern IE languages, we would still be able to
reconstruct words like:

*p at ter 'father'

(on the basis of, say, English <father>, Italian <padre>, and Hindi
<pita>)

*newos 'new'

(on the basis of, say, Modern Greek <neos>, Portuguese <novo>, and Polish
<nowy>)

The same is true for morphological paradigms etc.

Adam Hyllested


From adahyl at cphling.dk  Fri Jul 23 12:49:35 1999
From: adahyl at cphling.dk (Adam Hyllested)
Date: Fri, 23 Jul 1999 14:49:35 +0200
Subject: PIE vs. Proto-World (Proto-Language)
In-Reply-To: <l03102800b3b4aa9358c0@[195.247.226.14]>
Message-ID: <FRI.23.JUL.1999.144935.0200.>

On Fri, 16 Jul 1999, Ralf-Stefan Georg wrote:

> Even the
> notion that every known language, as Basque, Burushaski or whatnot, has to
> be related to some other language is ideology.

I suppose it is common sense to believe that "isolates" like Basque or
Burushaski at least have *dead* relatives. A language in colloquial
use never stays ONE language, but will inevitably split up into several
new languages. Which is exactly why "isolate" is merely a term
used for languages that have not yet been proven to be related to other
languages. Every single language has a mother and sisters. This is also
true of PIE, of course.
I could accept your statement above, if you inserted the word "living"
between "other" and "language".

Best regards,
Adam Hyllested


From jer at cphling.dk  Tue Jul 20 01:06:11 1999
From: jer at cphling.dk (Jens Elmegaard Rasmussen)
Date: Tue, 20 Jul 1999 03:06:11 +0200
Subject: Momentary-Durative
In-Reply-To: <005201bed144$a5098b40$6b058cd4@xpoxkjlf>
Message-ID: <TUE.20.JUL.1999.030611.0200.>

On Sun, 18 Jul 1999, petegray wrote:

>>>> [Jens:] Strunk has shown that nasal presents go with root aorists,

> Can you give us a reference please, Jens?   I would like to question this,
> for the following reasons:
> (a) In Greek it is largely true, but there are exceptions;   so a bald
> statement would need qualification.

It was never claimed for the individual languages, only for the common
reconstructed protolanguage (or even relatively recent prestage of it), so
"qualifications" are not a problem out of the ordinary.

> (b) In Latin it is largely untrue, since the aorists are either sigmatic or
> the rare reduplicative  aorist (tango tetigi, claimed by some as an aorist
> on the basis of Homer tetago:n, or the lengthened vowel: pango pe:gi (~
> perfect pepigi).   I can only find cumbo cubui which supports the claim in
> Latin.

You might also have thought of cerno:/cre:vi:, fundo:/fu:di:,
linquo:/li:qui:, rumpo:/ru:pi:, sino:/si:vi:, sperno:/spre:vi:,
sterno:/stra:vi:, vinco:/vi:ci:.

> (c) In Sanskrit there is a variety of possible presents and possible
> aorists.   There are 29 roots listed by Whitney which take nasal presents
> (ignoring the class nine presents).  Of these the first retains its nasal in
> the aorist, the second has no aorist, the third shows non-nasal present
> forms, and has either reduplicated or sigmatic aorist, and so on .....
> through the list.

What's wrong with yunakti/ayok, s'rnoti/as'rot, bhinadmi/bhet,
vrnakti/vark, and why ignore those of class nine (too good?) - ?

> (d) As stated earlier, a root may show nasal and non-nasal presents in
> different languages and even in the same language.

Yes, and where there is evidence about what is old and what is young, it
is a very general picture that the nasal present went with a root aorist.

> So as a bald statement I find it unbelievable.   The evidence doesn't seem
> to be there.

Much depends on how the statement is phrased - and about what language.

>    I would like to see what Strunk actually says.

Klaus Strunk wrote the book Nasalpra"sentien und Aoriste which appeared in
1967 at Winter, Heidelberg: Discounting refs. and index etc., it's just
about 100 pages of text that makes for quite easy reading. Strunk makes a
very strong case for the PIE paradigmatic companionship between nasal
presents and root aorists. It would be wrong to report this as a definite
rule, but Strunk does conclude at ''die paradigmatische Kombination beider
Sta"mme als ein uraltes vorhistorisches Pha"nom'' (p. 128). The book
treats verbs of the Indic classes V and IX (why did you exclude the
latter?), adding in a special chapter that VII seems to have worked the
same, but has not left such a clear picture. - From looking all through
IE, I have got the impression that an inherited nasal present is as the
general rule accompanied by a root aorist (or whatever became of root
aorists in a given branch - in Greek they often added -s-). The reverse is
not the case - and nobody ever claimed that: the root aorist is also the
normal aorist type for reduplicated presents, and very often also for
y-presents (which may have been back-formed to replace nasal or
reduplicated presents that fell out of use, or simply to create a present
stem where none existed before).

> [Jens:]

>> If my observation that there is an alliance between the sk^-present type
>> and the s-aorist is correct,

> Both forms may be used for some other reason, for example phonetic.   In
> both Greek and Latin the sk- ending occurs only after a vowel, which then
> takes a sigmatic aorist for purely phonetic reasons.

How do you mean? Do you consider the s-aorist morpheme a reduction of sk^?
Or even of *-sk^e/o-?? If so, by what rule?

>>> This brings me to a general question. There seem to be two camps about the
>>> category system of the PIE verb. One believes that the aorist-imperfect
>>> distinction, to be equated to perfective-imperfective distinction,
>>> ``always'' existed in PIE and Hittie lost this distinction, while Vedic
>>> changed things around. The other considers the aspectual distinction to
>>> postdate the separation of Anatolian.

>> [Jens] There are these two camps, yes, and I am in no doubt that camp one is
>> right. There is no way the specific forms of the aspect stems could have
>> been formed secondarily in "the rest of IE" left after the exodus of (or
>> from) the Anatolians. At the very least, all the _forms_ must be assigned
>> to a protolanguage from which also Anatolian is descended.

> I thoroughly agree with you here, Jens.   The forms are scattered over most
> of the IE languages.

>> [Jens:] And what would
>> the forms be there for, if the functions that go with them only developed
>> later?

> Here I disagree.   There can be variety of form in a language without a
> difference in meaning.   Look at German plurals (or English for that
> matter).   Sometimes a difference of meaning will emerge, fixing one meaning
> on one form and the other on the other (German Woerter and Worte, Dinger and
> Dinge), but this does not mean that all variations must necessarily
> correspond to a difference in function.    Language simply doesn't work that
> way, and there are far too many counter-examples for that claim to stand.

> Counter examples in English:
> (a) strong versus weak past tenses, e.g. dived ~ dove (origin historical /
> analogical)
> (b) plural /s/ versus plural /z/ (origin phonetic)
> (c) gentive versus preposition e.g. Tom's ~ of Tom (basis stylistic)
> (d) Time expressions (believe it or not, 10:45 really is the same as a
> quarter to eleven)
> and so on.

> I see the development of a functional difference as a later phenomenon, and
> I see it emerging differently in Sansksrit and Greek and Latin.

These are not counterexamples: All of the items you quote (except d) will
have to be very old - as types. And those of the types that are not
productive will have to be very old for the items quoted (goes for dove,
not for dived). In so far as the types can be shown not to be simple
variants of each other (as the two English sibilants), they must
originally belong to different categories. In fact, Eng. _of_ did once
mean something different from the relation expressed by the genitive. The
German plural variants used to belong to different stem classes - and
different stem classes used to have different derivative suffixes that
used to express different semantic shadings.

The last part completely escapes me: Are you anticipating that 10:45 and
"a quarter to eleven" will one day come to mean two different things? Is
THAT to you the only normal way language change works?

>   I suspect
> that some scholars are temporarily misled by the use of Greek names for the
> Sanskrit verbal system.

I plead not guilty, and I'll defend most of my colleagues who also know
that the terminology is misleading and only traditional anyway.

Are you in effect saying that IE morphology cannot be reconstructed? If
so, do you mean not at all, or just not beyond some specific limit which
you feel I have passed?

Jens


From X99Lynx at aol.com  Tue Jul 20 04:22:55 1999
From: X99Lynx at aol.com (X99Lynx at aol.com)
Date: Tue, 20 Jul 1999 00:22:55 EDT
Subject: Momentary-Durative (functionlessness)
Message-ID: <TUE.20.JUL.1999.002255.EDT.>

[ moderator re-formatted ]

In a message dated 7/19/99 7:04:48 AM, you wrote:

<<Sometimes a difference of meaning will emerge, fixing one meaning
<<on one form and the other on the other (German Woerter and Worte, Dinger and
<<Dinge), but this does not mean that all variations must necessarily
<<correspond to a difference in function.    Language simply doesn't work that
<<way, and there are far too many counter-examples for that claim to stand.

<<Counter examples in English:
<<(a) strong versus weak past tenses, e.g. dived ~ dove (origin historical /
<<analogical)
<<(b) plural /s/ versus plural /z/ (origin phonetic)
<<(c) gentive versus preposition e.g. Tom's ~ of Tom (basis stylistic)
<<(d) Time expressions (believe it or not, 10:45 really is the same as a
<<quarter to eleven)
<<and so on.

<<I see the development of a functional difference as a later phenomenon,>>

Just want to point out that every single one of the examples given IS
functional:
<<(a) strong versus weak past tenses, e.g. dived ~ dove (origin historical /
<<analogical)>>

 Historical continuity promotes predictability, and expectations
in listeners.  Analogy generalizes rules, promoting structural symmetry,
retention and again predictability for the sake of listeners.

<<(b) plural /s/ versus plural /z/ (origin phonetic)>>

  If by phonetic, you
mean dependent on the preceding phonemes, then that is functional. (/boys/
versus /boyz/) I don't think you mean that clearly articulated /s/ and /z/
are randomly interchangeable without consequence to the listener.

<<(c) gentive versus preposition e.g. Tom's ~ of Tom (basis stylistic)>>

  The
use of one or the other can materially affect sense and communicative effect.
 "Tom's dog bit the mailman." v. "The dog of Tom bit the mailman."  In
American English, the 'style' makes one version common, the other very rare
indeed.  They are not interchangeable and there is an obvious functional
difference between them.

<<(d) Time expressions (believe it or not, 10:45 really is the same as a
quarter to eleven)>>

  Numerics obviously have a functional advantage in many
situations, particularly tv listings and flight schedules, etc..  At the
airport, you will not find departures listed under "three quarters to
eleven."  Or even "quarter to eleven," for that matter.  Precison is one
functional difference.

Granted that the
aorist-imperfective/perfective-imperfective/momentary-durative-narrative,
etc., distinctions may not be the only way to explain the structural
differences that arose at any point pre- and post- Anatolian.  But to say
that those structural features had no functions at all is perhaps going too
far - especially because one can only guess what those functions were in a
real world language.

How would you explain, for example, why in American English, "Tom's dog bit
the man" is expected, normal and sounds right.  But "The dog of Tom bit the
man" is clearly odd and rarely used?  There is a functional difference
between using one form or the other.  If you don't think so, try talking the
odd way and see what kind of reaction you get from your listeners

Regards,
Steve Long


From vidynath at math.ohio-state.edu  Tue Jul 20 11:38:00 1999
From: vidynath at math.ohio-state.edu (Vidhyanath Rao)
Date: Tue, 20 Jul 1999 07:38:00 -0400
Subject: Momentary-Durative
Message-ID: <TUE.20.JUL.1999.073800.0400.>

Rick Mc Callister <rmccalli at sunmuw1.MUW.Edu> wrote:

> Spanish distinguishes them [resultative and passive]
> by resorting to

> estar --a "momentary" verb--
> [actually a verb indicating condition] for resultant conditions--

> and ser--a "durative" verb
> [actually a verb indicating  characteristics]
> for passive constructions.

I don't know much Spanish, but I ought to have remembered this for I have
been told about this before.

I find something curious about your glosses. Resultatives, as indicating a
state of indefinite duration, ought to be duratives, though examples of
languages which use `go', `come', `finish' to form resultatives are
mentioned in ``The evolution of grammar''. How old is the use of `estar'?.


From vidynath at math.ohio-state.edu  Tue Jul 20 15:26:26 1999
From: vidynath at math.ohio-state.edu (Vidhyanath Rao)
Date: Tue, 20 Jul 1999 11:26:26 -0400
Subject: Momentary-Durative
Message-ID: <TUE.20.JUL.1999.112626.0400.>

 petegray <petegray at btinternet.com> wrote:
To: <Indo-European at xkl.com>
> >>> I don't see any parts of Sans lit in which aorist has resultative
> >>> meaning.

I have access only to the e-texts of the critical editions, which I will
use. Given the difficulties in disentangling resultatives from hot-news
forms, we need to look at the context. To really convince me, you need to
translate into a living language that distinguishes the two and explain why
the resultative is preferable.

> or Mahabharata 2:60:7 (CE 2.60.8ef)
>   kim nu pu:rvam para:jais.i:t a:tma:nam
>  whom you have lost first, yourself....

This is the easiest to dispose of. Here is the whole stanza:
kasyes'o nah paraajaiSi:r iti tvaam aaha draupadii/
kim nu puurvam paraajaiSiir aatmaanam atha vaaapi maam.//
The context is Draupadi's question of how Yudhisthira can bet her if he had
already lost himself. Translating the second line as ``whom >have< you lost
first, yourself or else me?'' misses the point that D. was bet legally.

>   Ramayana 2:64:52 (CE 2.58.44cd)
>   yah s'aren.aikaputram tvam aka:rsi:r aputrakam
>  with one arrow you have rendered me childless, who had but one son

The first half is
 adyaiva jahi ma:m. raajan maran.e na:sti me vyathaa.

Verse b is a paranthetical remark, placed there for metrical reasons. [It
would be quite odd to claim that the father feared eath when the son was
still living.] I would translate it as ``Slay me, whom you made childless
from being one-sonned, now itself, King; I do not fear death.'' The point is
that Dasaratha, by killing the son, has metaphorically slain the whole
family. It is the event of killing, not the resulting death, that is being
highlighted in the second half.

> or Mahabharata 5:3:10
>   tad akars.it prajagaram
>   that has produced sleepiness

A semantic point. praja:garam is ``awakefulness'', not sleepiness. Anyway,
it is not so obvious to me that the result is being highlighted here. In
other words, it is not clear to me that I should translate this into Tamil
as ``atu muzhipp- un.t.u pan.n.i-y- ul_l_atu'' rather than ``...
vit.t.atu.'' [I don't know enough English to render this difference
effectively.] It is the immediciacy of the event that comes to the fore.


From petegray at btinternet.com  Tue Jul 20 19:45:44 1999
From: petegray at btinternet.com (petegray)
Date: Tue, 20 Jul 1999 20:45:44 +0100
Subject: Momentary-Durative
Message-ID: <TUE.20.JUL.1999.204544.0100.>

Thank you for your patience, Jens.    I'm replying to you points in reverse
order - it seems better that way, since it puts the more important things
first.

You said:

> ... an increasingly well-established structure, with archaisms and
> productive patterns, .... I react in
> defense of the field when I see somebody turning the clock back ...

Rightly so.   I am not a linguistic Luddite.  But I am an enquirer, and I
believe I am right to ask for evidence rather than assertion.    You have
insisted several times that nasal presents and root aorists go together.
Apart from mentioning Strunk, you have not given evidence.  I don't dispute
that the evidence is there - and perhaps it's so well known that you haven't
felt it necessary to give any - but I'd like to see it, or a reference to
Strunk so that I can chase it down for myself.   You do allow that PIE can
have variant presents, so that seems to run against your main claim.

Your examples of reduplicated presents developing into nasal presents is a
well chosen answer to my point - thank you.

Thank you, too, for the references to Rix, and Renou.   I was not aware of
either.

You said:

> The impression that [in the IE verbal stem] ...
> no common system is
> recoverable is not compatible with current knowledge

There is little real "Current knowledge" with PIE.   It is in many cases
really rather "current opinion which someone happens to accept".   So how
much support is there for Strunk?  How much debate is there?   I read as
widely as I can in PIE, and have not met a mention of him - but I'm sure the
failing is mine.   In any case, I do not dispute that patterns are
perceivable.   I do question, however, how much weight we should give these
patterns, and in particular whether we over-prioritise the patterns based on
Greek & Indo-Aryan.   It has been shown again and again since the 70's that
the "south-eastern" group of Greek, I-A, and Armenian is highly innovative.

I said:

>> your argument (that a root aorist implies a nasal present) is
>> really rather weak.

Here I must apologise - as you rightly point out, I had your argument back
to front.

You said:

> No, thematic presents and root-presents typically take the s-aorist (or
> suppletion).

Root presents are rare outside I-I, and in Sanskrit -s- aorists do not seem
to be typical for root presents, since the both sigmatic and asigmatic
aorists occur.   E.g. am has asigmatic, i:d has both, i:r shows both, u
shows none, U:h asigmatic, and so on.   Where, again, is your evidence that
makes it "typical" for a root present to take an -s- aorist?
    Thematic presents appear to be a later formation, in any case, and you
seem to suggest an origin for them in root aorist subjunctives.

Incidentally, I read today in JIES vol12, 1984, the argument that even
the -s- aorists are an innovation within Greek, I-A and closely related
groups.   The article alleged that evidence outside this area is weak, and
tried to dispose of Latin -s- perfects by suggesting they were either
limited to verbs ending in a velar, or they were back formations from the
supine.
    Don't hear me supporting this idea.  I mention it because it raises the
kind of question I want to ask about Strunk.   Here an article attacks
something that is well established for PIE.   So I cannot believe that there
is no argument at all about Strunk's ideas, or that they are "current
knowledge".

Now I hate to be a pain, Jens, and I freely acknowledge the limitation of my
understanding.   But I hope the above shows the kind of questions I have to
grapple with before I can lie down and accept what you say.

Peter


From vidynath at math.ohio-state.edu  Mon Jul 26 13:03:47 1999
From: vidynath at math.ohio-state.edu (Vidhyanath Rao)
Date: Mon, 26 Jul 1999 09:03:47 -0400
Subject: Momentary-Durative
Message-ID: <MON.26.JUL.1999.090347.0400.>

Jens Elmegaard Rasmussen <jer at cphling.dk> wrote:

> On Mon, 12 Jul 1999, Vidhyanath Rao wrote:

>> The tuda'ti type, with primary endings, is somewhat rare in RV. Perhaps,
>> tuda'ti type was a replacement for injunctives, i.e., tuda'ti was a
>> general  present and not progressive present. Just a thought.

> Hardly so, for the injunctive is not a stem-formation, but an inflectional
> category that can be formed from all verbal stems.

Present, with primary endings, is also an inflectional category.

The roots of the tudati type, excepting transfers from root and nasal
classes, tend to have punctual meanings. Punctual verbs cannot be used in
the progressive (unless repetition is indicated) but can be used in generic
sense etc. But when the injunctive was lost, only the present was available
for this. This is what I meant.

>> [Jens:]

> If my observation that there is an alliance between the sk^-present type
> and the s-aorist is correct ... then the s-aorist was originally
> inchoative in function. I'd say that makes very good sense, for the
> s-aor. is also widely used with verbs that form radical or thematic
> presents, as *weg^h-e/o- 'drive', whose aor. *we:g^h-s- will then
> originally have meant 'start driving, set out (by carriage)'.

This makes it harder for me to understand how the aorist became the
perfective. `Started driving', in contrast to `drove', suggests incomplete
action.

>> This brings me to a general question. There seem to be two camps about
>> the category system of the PIE verb. One believes that the
>> aorist-imperfect distinction, to be equated to perfective-imperfective
>> distinction, ``always'' existed in PIE and Hittie lost this distinction,
>> while Vedic changed things around. The other considers the aspectual
>> distinction to postdate the separation of Anatolian.

> There are these two camps, yes, and I am in no doubt that camp one is
> right. There is no way the specific forms of the aspect stems could have
> been formed secondarily in "the rest of IE" left after the exodus of (or
> from) the Anatolians. At the very least, all the _forms_ must be assigned
> to a protolanguage from which also Anatolian is descended. And what would
> the forms be there for, if the functions that go with them only developed
> later? We find practically all the IE verbal stems in Anatolian, many only
> in a few lexicalized remains; are we to assume they have had totally
> enigmatic earlier functions, and that they were later dug up by "the rest
> of IE" and given totally new functions there? Clearly the only unforced
> interpretation is that their functions in the common protolanguage were
> the ones with which they are found where they do survive with a palpable
> functional identity.

The stem formants might have been derivational in nature. In particular,
they might change the meaning, valence and/or aktionsart. It is the
assumption that they changed only the aspect and nothing else that is in
question. Your view, expressed elsewhere, that different present stems might
be because there are different kinds of durativity is halfway to this.

Incidentally the assumption that all kinds of repetition must be represented
by imperfectives is as mistaken as the assumption that atelics cannot be
represented by perfectives. Languages differ in how they convey repeated
complete events. Czech, for example, is said to (permit the use?) use the
perfective in this case [Dahl, TAS]. Gonda (somewhere in ``The aspectual
function of Rgvedic ...'') gives examples from Greek where the aorist is
used with adverbs denoting repetition. And, just because we can think of
`walk' as a repetitive`step', we cannot conclude that `walk' cannot be used
in a telic manner.

Grammatization of derivational affixes is found often enough that its
occurance in Pre-IE -> PIE or PIE -> dialects cannot be rejected out of
hand. As it seems that very different languages underwent similar evolution
of grammatical categories (see Bybee et al, ``The evolution of grammar), we
cannot assume the similar functions must go back to the proto-language. That
might have had a precursor function.

>> There are certain nagging questions about the first thesis: The change
>> in Vedic is not explained and how it came about without the prior
>> loss of  aspect has, AFAIK, not been explained. Those who adhere
>> to this also feel the need to explain away as much as possible of
>> root presents. But there are enough of them remain in Hittite and
>> Vedic to raise doubts.

> I don't follow - what change in Vedic are you talking about?

In Vedic, the so-called imperfect is the tense of narration. Aorist has a
recent past meaning. In living languages with aspect, the perfective is the
tense of narration, unless historical present is being used. Languages with
remoteness distinction, either it applies only in the perfective or without
regard to aspect. It is this that needs to be explained.

People have tried to explain this. The attempts I know of are by Gonda and
Hoffman. Gonda's explanation is that this is due to the ``national character
of ancient Indians'' that led them to use in lento description in place of
narration. I won't comment on this idea of explaining language change by
reference to ``national characters'', except to ask about reactions to a
claim that the use of passe compose in French is due the Gallic
characteristic of exaggeration that sees everything as having present
relevance.

Hoffman's answer is to posit an intermediate stage in which aspect was
limited to non-recent past. He does not give any contemporary examples, nor
does he explain how the aorist, which in such a stage must have been even
more common, was lost in reference to remote past.

The point is that an alternate explanation is possible: Completives have a
``hot news'' value, which makes it plausible to see them develop into recent
past. They also can develop into perfectives, Slavic being a usable example.
There are gaps in the examples, but I find this more plausible than
deriving the Vedic usage out of perfecitve-imperfective opposition.

> Why would anyone want to explain away root presents where
> they are securely reconstructible? One would do that only to avoid
> having a language combining a root-present with a root-aorist, for in that
> case the two aspect stems are identical. That is why I am so sceptical
> about the authenticity of the Vedic root presents lehmi and dehmi,
> because for these verbs we have nasal presents in some other IE
> languages pointing to the existence of a root aorist; thus leh- deh- look
> like displaced aorists. But not so for eti 'goes' or asti 'is': these are
> durative verbs, and so their unmarked form could function as a
> durative (socalled "present") stem.

This argument is valid only if you have an independent reason to assume
obligatory perfective-imperfective contrast for PIE. But firstly, we see
stems transferred from one class to another within a single group. This is
most obvious in Indic, where we have a long recorded history. Secondly,
there are too many root presents that remain: In Vedic, amiti (injures),
da:ti (cuts, divides), yauti (joins, unites) and of course, hanti confirmed
by Hittite kuen-. None of these is eligible for a nasal present (expecpt
perhaps amiti, if you believe in nasal presents for roots of the shape CeNH,
with N standing for a nasal). Especially *g'henti is unavoidable, unless you
are going to argue that the two oldest recorded dialects innovated in
precisely the same way. Such an argument needs more compelling evidence
than a just so story.

If you add potentialy telic verbs, we get some more: ta:s.t.i (taks,
fashion), ya:ti etc. These can and are used in the so-called imperfect with
definite objects without any indication of non-completion. etc. How can they
be called imperfetive?

>> [part of my (Nath's) post deleted]

> Again, I do not think there is any problem in accepting root present as
> original for inherently durative verbs, and root aorist as equally
> original for inherently punctual verbs.

What about telic verbs? These are inherently durative, but with a determined
object, their preterite (in languages without perfective-imperfective
distinction) carries the implicature of completion, while their imperfective
past would do the opposite. How, without recorded narratives, can you decide
how a given language operated?

> But the IE aorist is not restricted to any special kind of verbs - it is
> only _unmarked_ (better: apparently originally unmarked) for inherently
> punctual verbs; for other verbs the aorist need a morphological marking,
> and the meaning is then some nuance that can be regarded as punctual
> ("started to -") or it just reports that the action got done (Meillet's
> action pure et simple).

Could the imperfect report `action pure et simple', or was that reserved for
the aorist? How did PIE speakers report a durative action that was done,
like ``I walked home''? How did they say ``I made pots yesterday''?

> One important functional point with the aorist,
> however, is that it marks a turn of events which creates a new situation,
> whereas the "present aspect" stays in the situation already given and
> reports another action contributing to that situation.

How do you explain that it is the so-called imperfect that is the tense of
narration in Vedic?

> This is seen remarkably well in the prohibitive use of the prs. vs. aor.
> injunctive, as propounded so clearly by Hoffmann.

Hoffmann's claim requires morphological gymnastics, such as taking i:'s'ata
as an aorist and still has some holes (eg, jivi:t).

> Still, even verbs generally signifying completed action could form
> duratives, indicating e.g. a repetition of the action (give one thing, and
> then another) or an as yet unsuccessful attempt (I'm opening the window).

How do you classify ``I learned that chapter in one month?''

> We know that this kind of change was small enough in the languages here
> concerned to lead to a number of misplaced aspect stems. E.g., the
> Armenian aor. eber is an old ipf. Is the jump from "narrative past" to
> "recent past" so great? If it is, even great changes happen.

But if the PIE ``imperfect'' was really just a preterite, what we see in
Armenian eber is old preterite becoming an aorist due to the rise of an
imperfective past. That is attested. For example, Kui, a Dravidian language,
has generalized an old progressive into an imperfective and this limited the
preexisting past into an aorist.

It is not the size of the change, but the direction and manner of the
change, that must be credible. All I am asking for is a single
uncontrovertable evidence of this change. It won't do to say, as I was once
told by an Indologist, that this change is possible because it must have
happened in Vedic. You need to give an example where the perfective past
category can be established from >preserved texts<.

What I find difficult to swollow with the argument for aspect in PIE is that
the association ``imperfect'' = imperfective is limited to Greek. The two
oldest recorded dialects, Hittite and Vedic, do not work that way. We are
simply supposed to believe that they innovated, but what do we get in
return? What can you explain this way that you cannot explain otherwise?

[quoted from a different post]
> The most rewarding experience during the time I have been
> watching Indo-European Studies has been to see the protolanguage
> come alive and assume an increasingly well-established structure, [...]

Do we really understand the variety of syntactic structures and their
diachrony that well? I have mentioned the Tamil -vidu construction a few
times. You will find some linguists call that a perfective and the Tamil
simple past an imperfective. This is simply wrong as the simple past is and
has been the tense of narration for the 2000+ year recorded history, and
this distinction is nothing like the perfective-imperfective distintion in
Russian or Arabic, as can be seen by comparing the translations of Dahl's
questionairre [or simply use his summaries]. Use of the inappropriate lables
is due to either trying to fit every non-tense non-mood apposition into the
Procrustean bed of perfective-imperfective distinction or the lack of
adequate terminology. If anyone thinks that they know how Tamil syntax works
based on these labels, the mildest thing one can say is that (s)he is
woefully misinformed.

It can be even worse: Similar constructions exist in some NIA languages
which also have a prefective-imperfective distinction that is different
(imperfective out of old progressive). I have seen at least one linguist
apply the label of perfective to the construction with auxillary, and
imperfective to the one without the auxillary. This completely misrepresents
the syntax.

What reason is there to think that we are not making the same kind of
mistake with PIE? I am not asking for certainity here, but evidence for
greater probability than the opposite.


From jbisso at jps.net  Mon Jul 19 04:39:57 1999
From: jbisso at jps.net (James F. Bisso)
Date: Sun, 18 Jul 1999 21:39:57 -0700
Subject: connected text in PIE; Proto-World language
Message-ID: <SUN.18.JUL.1999.213957.0700.>

Anthony Appleyard wrote:

>  (1) When neural-net computers get advanced enough for people to
>simulate better than now human brain processes including language, it
>would be interesting to simulate language evolution and see over how
>many centuries and by what steps a language changes so much that all
>trace of common ancestry vanishes behind the `noise' of accidental
>resemblances.

Some folks are using artificial life (see http://alife.santafe.edu/)
techniques to model language change: (e.g., Luc Steels at the Vrije
Universiteit Brussel (http://arti.vub.ac.be/origin/origins.html).

jim


From proto-language at email.msn.com  Mon Jul 19 05:41:44 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Mon, 19 Jul 1999 00:41:44 -0500
Subject: Lexical Retention
Message-ID: <MON.19.JUL.1999.004144.0500.>

Dear Joat and IEists:

 ----- Original Message -----
From: <JoatSimeon at aol.com>
Sent: Friday, July 16, 1999 11:43 AM

Joat wrote:

> As time goes on, any given language gives us less and less information about
> previous forms.  Information is _lost_.

> Eg., no amount of analysis of Germanic will give you a PIE word for "bear"
> because that word dropped out of that family.  All you'll get is variations
> on "the brown one".

Pat responds:

You have chosen an example about which we are singularly poorly informed ---
not to mention the influence of possible taboo deformation.

If I assumed that, regardless of its highly unusual form (*R:k{^}tho-s), the
basis for 'bear'-words is the IE root *rek-, 'tower above', it would not be
true that the ultimate root was not attested in Germanic.

In any case, you have it rather backwards. If the root and all manifestions
of it (like, perhaps, in 'bear') were absent in Germanic, what would, as it
has, make possible our reconstructing an IE root would be its presence in,
at least, three (nominally) other branches.

I thought I made clear that I was talking about *CVC-roots. Your example
does not really address my assertion.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From stevegus at aye.net  Mon Jul 19 11:51:19 1999
From: stevegus at aye.net (Steve Gustafson)
Date: Mon, 19 Jul 1999 07:51:19 -0400
Subject: indoeuropean/hand
Message-ID: <MON.19.JUL.1999.075119.0400.>

Ed Selleslagh writes:

<<Actually it's 'cornu' in the nominative.>>

-Cornu- is "horn."  -Cornus- is the dogwood tree.

<<All tree names, also those with -u stems like 'quercus' (oak) (and -o stems
like po:pulus) are feminine.>>

I had ever understood that all trees were feminine for essentially
mythological reasons; they were thought to be the dwelling places of female
spirits.

<<Could there be some relationship here?  I mean
'extremities, branches...'.  That would also account for manus, acus, cornu
and
tribus being feminine, but not for domus nor idus.  'Foot, pes...' are
probably
not viewed as extremities, but as 'base' to stand on.>>

Now, my understanding is that the root idea of -idus- is of a division by
halves.  When the months used to be lunar, the kalends marked the first
appearance of the slip of the new moon, and the ides were the date of the
full moon.  Perhaps this could be worked into your hypothesis as well.

--
L'an mil neuf sens nonante neuf sept mois
Du ciel viendra grand Roy deffraieur
Resusciter le grand Roy d'Angolmois
Avant apres Mars regner par bonheur.
                                --- M. de Notre-Dame


From ECOLING at aol.com  Mon Jul 19 15:08:45 1999
From: ECOLING at aol.com (ECOLING at aol.com)
Date: Mon, 19 Jul 1999 11:08:45 EDT
Subject: Semantic change
Message-ID: <MON.19.JUL.1999.110845.EDT.>

In a message dated Patrick Ryan:

>Having worked extensively with language-family
>comparisons, my statistically unsupported "guess" is that there is amazingly
>little vocabulary loss if one allows reasonably semantically-shifted pairs
>to be counted as "retained".

That would be circular without further constraints,
because it would mean (not what Pat means,
but taken to extreme) that any meaning could change
into any other meaning, and so we are only then seeking
look-alikes, not cognates.

Rather, there needs to be a quantification of
degree of semantic change,
and of course a life-long learning of what meanings
ARE known to change into what other meanings
under what circumstances,
so we can even begin to try to measure
how far a purported related word is
from the meaning of another word we are comparing
it to.
Then, as a matter of degree, we can say that
as we allow the meanings to be more distant,
we get more *possible* cognates between any
two languages we wish to compare.
(for a fixed degree of phonetic resemblance we
might require, however measured).

All of these things are matters of degree.

A less plausible semantic relation means that two
look-alikes are less plausible as cognates.
But requiring "identity" of meanings or even
"near-identity" of meanings is an absurd
requirement also, when we are working at great
time depths.

So what to do?
Asserting simplistic extremes either for or against
is not particularly useful.

Lloyd Anderson
Ecological Linguistics


From ECOLING at aol.com  Mon Jul 19 15:08:41 1999
From: ECOLING at aol.com (ECOLING at aol.com)
Date: Mon, 19 Jul 1999 11:08:41 EDT
Subject: Principled Comparative Method
Message-ID: <MON.19.JUL.1999.110841.EDT.>

Here is an example of where the "Comparative Method" can be strengthened.
The matter of "recurring sound correspondences".
The standard presentation is pretty close to what Sean Crist just posted
(quoted at the end).

Now let's consider things as a matter of degrees, not a sharp cutoff in what
is and what is not reconstructible.

As the relation between two languages or language families becomes more
distant,
the number of examples of any sound CORRESPONDENCE become
fewer and fewer.  This is a matter of degree.  The reasoning must be
more sophisticated.  We then have more cases in which multiple sound changes
apply to the same word, and we might have cases in which
(to use the completely artificial and hypothetical example)

a)  there are many examples of PIE *d corresponding to PU *z
     (or even IE /d/ corresponding to Uralic /z/, using particular languages).
b)  there are 5 examples of PIE /d/ corresponding to /z/
c)  there are 2 examples of IE /d/ corresponding to Uralic /z/
d)  there is only 1 example of IE /d/ corresponding to Uralic /z/
e)  there are no examples of IE /d/ corresponding to Uralic /z/

EVEN IN CASE (E), IT STILL MIGHT BE CORRECT TO ARGUE
that IE /d/ "corresponds to" Uralic /z/.

The reason is that "recurring sound correspondences"
is actually NOT a cornerstone of the "Comparative Method",
if we mean by Comparative Method what it is in ideal practice,
that is the reconstruction of the series of changes by which
some ancestor language gave rise to descendent languages.
(Calvert Watkins is at least one prominent Indoeuropeanist
who has emphasized this principle.)

Rather, "recurring sound correspondences" is a short-cut
which works in the EASY cases.

What IS a cornerstone of the comparative method is
a PLAUSIBLE sequence of changes from ancestral language
to descendant languages or language families.

So if we had case (d)
and we also had case (d'),
one example of IE /t/ corresponding to Uralic /s/,
we might have more evidence for saying that /d/~/z/ was real.

Or if we had case (d) before a following /u/,
and we also had case (d'')
ONE example of IE /d/ corresponding to Uralic /z^/ before /i/,
(interpret the ^ as a hachek = wedge, upside down)
we would again have more evidence for saying that /d/~/z/ was real,
in this case because we would assume that /z/ became /z^/ before /i/.

And if lucky we might have case (d') above with following /u/,
and also case (d'''),
one example of IE /t/ corresponding to Uralic /s^/ before /i/.

With all four of these examples, we would have a plausible system,
and would in fact have four examples reflecting the same sound changes,
perhaps actually only a single sound change (namely
"affricate apical stops before high vowels, with the result being
the fricative of the same voicing and "reflecting" the vowel quality)
but we need not have more than one example of any particular correspondence
on the surface.

Of course multiple examples of the same surface results DO
make us more confident of our findings.  The point above is that
examples which are not identical on the surface can ALSO make
us regard the hypothesis of a particular sound change as more likely
valid than would be the case if we had only one of the examples
(d, d', d'', d''').  In extreme cases, I would think we could even
confidently hypothesize a sound change whose immediately results
remain NOWHERE visible on the surface, case (e) above.

The plausiblity of the highly specific changes hypothesized above
is sufficiently demonstrated by the occurrence of something like it
in modern Japanese, where we have in the syllabary

sa se s^i so su
ta te c^i to tsu
za ze j^i zo zu
da de j^i do dzu~zu.

Of course as the number of sound changes hypothesized
rises, we still do need an increasing number of examples,
since we don't want to be in the situation of having a new
hypothesis just to excuse each new example of words in
two languages we wish to claim decend from a common proto-form.
We just do not need to have "recurring sound correspondences"
recurring identically in their surface results.

This kind of situation can be seen in real cases
of language reconstruction, and it needs to be catalogued in standard
manuals of the comparative method, so that the short-cut of
"recurring sound correspondences" can be seen for just what it is,
a short-cut in the easy cases.

In fact, all well-trained specialists in comparative reconstruction
DO use more sophisticated tools than mere superficial recurring
sound correspondences.  Their task is easier with more recurrences,
but it is not limited to them, certainly not in principle.

What is considered "plausible" in the more difficult cases cannot be
simply the whim of the individual researcher about what is a "plausible"
vs. not a "plausible" sound change, on a priori grounds.  It must be
based on evidence of some kind.  We cannot assume our manuals
contain examples of all the possible sound changes,
we can only assume that known sound changes ARE plausible,
and that as-yet-unknown ones will always exist.
For beginning linguists, the substitution of /l/ and /s^/ comes as
a surprise, until they learn that these two segments have some strong
acoustic similarities.
Learning what constitutes a "plausible" sound change is a lifelong task.

Since the number of sound changes between a proto-language and its
descendants does increase with time depth,
the number of recurring sound correspondences will decrease,
and we must make increasing use of the more sophisticated tools.

The real task of extending the Comparative Method to deeper time
depths is to make explicit more of these sophisticated tools,
and to CREATE more such tools by discovering what ways of
handling the data are robust across what kinds of intervening changes.

One of them I am quite certain is to develop both articulatory and
acoustic "spaces", relative "distances" between different articulations
and different acoustic effects, so that when attempting to judge
likelihood of cognacy of pairs of words, we can judge similarity
by degrees, not by yes/no dichotomies.  (These will partly depend
on the general typology of the sound systems of the languages
concerned, they will not be completely universal, but they will
also not be completely idiosyncratic.)

So for example, the sound change w>k does not seem "obvious",
yet if we add y>c^ or even y>t, or even if we don't, it is not the
most distant pair of sounds /w/ and /k/ do share some things
(and we do have /erku/~/duo/~"two" with r~d~t and k~w~null before /u/).
We need a measure of DEGREE of similarity between such known
pairs of words, such that Albanian and English, or Armenian and English,
can be found to be related languages.
Throwing up our hands, as some do, and saying
"any sound can change into any other sound" past a certain time depth,
is not helpful.  At the extreme, in an absolute sense, it is quite possibly
true,
but that is the reaction of those who do not yet have the tools to handle
relative-degrees-of-plausibility in many dimensions simultaneously,
rather than simple yes/no decisions or even degrees of plausibility
in a tiny number of dimensions.

Instead of always using the easy method (recurring sound
correspondences), let's discover the more sophisticated tools that
work in the harder cases.

Who is doing that now?

I would like to see forums which can highlight and reward such
productive work.  How about testing methods which can work
to establish PIE based on English and Albanian (see below)?

Lloyd Anderson
Ecological Linguistics

Below are the quotations which quite adequately represent the
normal presentation of the Comparative Method.

>Suppose for the sake of
>argument that there is in fact a genetic relationship between
>Proto-Indo-European and Proto-Uralic (to take one possible example). Even
>if this were true, and given the evidence which we actually do have at
>hand, would we expect to be able to _show_ that there is such a relation?

>Here, the problem _is_ like trying to reconstruct PIE on the basis of only
>modern English and Albanian.

[Lloyd:  So to establish more empirical validities, we need to test
methods to find which highly general methods (not tailored to the specifics
of the English and Albanian case) would give a positive answer
to the question of relation of English and Albanian.  We might have
almost no recurring sound correspondences (though I suspect there are some),
and a large number of correspondences which result from successive
applications
of more than one sound change in different contexts, so not obviously
intuitive on the surface.]

>Even tho external evidence might lead us
>to
>guess that there could be a relation between PIE and PU, the cognations
>are so obscured by millenia of sound changes, loss of old lexical items,
>the noise of new lexical items, etc., that it might well be impossible
>to
>show a genetic relationship on the basis of the available information.

>Suppose, for example, that PIE *d corresponds to PU *z (I don't know that
>there is a *z in PU; I'm making this up).  Even if there were originally
>dozens of words showing this correspondence, it might well be that case
>that only one such pair exists between the small subsets of the original
>lexicons of PIE and PU which we can reconstruct.  Given the fundamental
>assumptions of the Comparative Method, you _can't_ show that the two words
>are cognate when you've only got one example of the correspondence.

>You need multiple examples of sound correspondences to be able to conduct
>the Comparative Method at all; and when the the cognates become as
>rarefied as they are this time-depth, the likelihood of having access to
>an adequate number of examples to work out the relevant sound changes
>becomes proportionately smaller, eventually reaching what for practical
>purposes is an impossibility.

[Lloyd:  The following clause, taken as an absolute, is false:

>You need multiple examples of sound correspondences to be able to conduct
>the Comparative Method at all; ...

It is true only when "sound changes" is substituted for "sound
correspondences",
yet then it cannot be as quickly seen whether we have that in a given
situation.]


From X99Lynx at aol.com  Tue Jul 20 03:15:38 1999
From: X99Lynx at aol.com (X99Lynx at aol.com)
Date: Mon, 19 Jul 1999 23:15:38 EDT
Subject: Chronology of the breakup of Common Romance [long]
Message-ID: <MON.19.JUL.1999.231538.EDT.>

In a message dated 7/19/99 12:15:24 AM, mcv at wxs.nl wrote:

<<That's merely because the Slavic languages have retained the
"(c)h" in the Germanic loanword which was dropped in the modern
Germanic languages (OE wealh, OHG walh).>>

That would explain a lot of the way the variances seem to go.  But what
happens in MHG?  I have 'Walch' (n), 'walsch', 'walche' (adj).   It seems
like there's some back and forth.  I guess that makes sense in that 'vlach'
itself is a form borrowed back from Slavic.

In Old Norse, where the reference to anything Rumanian is least likely,
'valskr' is typically reconstructed from '*valr' (which would put it before
800 AD.)  I wonder if that reconstruction isn't questionable.

Regards,
Steve Long


From X99Lynx at aol.com  Tue Jul 20 03:20:46 1999
From: X99Lynx at aol.com (X99Lynx at aol.com)
Date: Mon, 19 Jul 1999 23:20:46 EDT
Subject: 'Origins of the Rumanian Language' URL
Message-ID: <MON.19.JUL.1999.232046.EDT.>

Rather than answering more individual e-mails, I thought I might just post
the URL.

<<The site is at http://www.hungary.com/corvinus/dunay/

The Word file can be downloaded from http://www.hungary.com/corvinus.
>>

Steve Long


From adahyl at cphling.dk  Wed Jul 21 16:44:33 1999
From: adahyl at cphling.dk (Adam Hyllested)
Date: Wed, 21 Jul 1999 18:44:33 +0200
Subject: `cognate'
In-Reply-To: <Pine.GSO.3.95.990716161343.15501C-100000@unagi.cis.upenn.edu>
Message-ID: <WED.21.JUL.1999.184433.0200.>

On Fri, 16 Jul 1999, Sean Crist wrote:

>>> I'm sure it says PIE *<p at ter> in your manuscript (+ diacritics on the
>>> <e>).

>> Yes, it does, and my posting should have read `p at ter', but I slipped up.

> At least according to Don Ringe, this word should be reconstructed
> *pH2te'r.  I should ask him what the arguments are in favor of that
> reconstruction, since I don't know myself.

It is more or less the same reconstruction. <@> is (originally, at
least) the vocalic allophone of <H>, reconstructed on the basis of
Indo-Iranian, where the root vowel is <i> (an IE <**pater> would have
developed into Sanskrit <**pata:> and not the actual <pita:>)
Using <@2> or <H2> is simply a matter of phoneme-to-grapheme
correspondence.

Best regards,
Adam Hyllested


From pagos at bigfoot.com  Thu Jul 22 18:59:36 1999
From: pagos at bigfoot.com (Paolo Agostini)
Date: Thu, 22 Jul 1999 20:59:36 +0200
Subject: -n- adjectival suffix in Latin
Message-ID: <THU.22.JUL.1999.205936.0200.>

I was wondering about the origins of the Latin adjectival
suffix -n-us, -n-a, -n-um appearing in words like _mater-nus_ "maternal;
motherly; belonging/pertaining to the mother", _pater-nus_ "paternal;
fatherly; belonging/pertaining to the father", _feli-nus_ "feline;
belonging/pertaining to the cat; belonging/pertaining to  the family/genus
of the cats/felidae" and the like.  Can it be traced back to IE? Does the
morpheme exist in other IE or non-IE languages? Any idea abt its etymology
and/or development?

Thanks in advance for your answers.
Cheers

Paolo Agostini


From ECOLING at aol.com  Mon Jul 26 15:12:27 1999
From: ECOLING at aol.com (ECOLING at aol.com)
Date: Mon, 26 Jul 1999 11:12:27 EDT
Subject: Hittites ~ Phrygians ~ Balkan peoples?
Message-ID: <MON.26.JUL.1999.111227.EDT.>

There has been much discussion of the Hittites and the Phrygians,
and other names for these or other peoples of Anatolia
and the Balkans (Thracians, Illyrians, Dacians, Moesians, etc. etc.),
which may have been incorrectly taken as distinct peoples in the past,
instead of simply as different names used for the same peoples
at different times or by different authors or cultures.

Can someone produce a summary of the argument for a new
way of viewing things in TABULAR form?

Or, since this is email, at least in condensed summary
of some sort?  Given the slight differences of opinions
on Dacia and Moesia around the lower Danube river,
there will of course have to be slight difference in such
summaries.  That will be healthy, so I hope more than
one contributor will respond.  Something like this???:

"Hittite"  name for X people during time period S...T
     Who used this name?
"Phrygian"  name for X people during time period U...V
     Who used this name?

  (with whatever extra notes are needed on overlaps of time and place)
    "Who used this name"  is intended mainly for use when the source was
     Herodotus or another pre-modern source)

and so on for other peoples and sets of names?

I would like to have some handy reference for understanding
both the debates themselves and also the implications of the
claims being made, which are often buried in the text of the
discussions, or evident only to specialists.

What I am thinking of is the difficulties of the entire
Illyrian-Thracian- etc. areas, whether the discussants seem
to be implying that some of these peoples were also the descendants
of Hittites or Celts or close relatives of theirs, as distinct from
later Slavic peoples, or mostly-lost branches of IE.

It is by now taken as fact (?) that the
Celts were early branches off from the IE stem at about the same
time as the Tocharians and shortly after the Hittites,
or however that might be more exactly put these days.

Similar threads are proceding on the ANE and the IndoEuropean email lists,
and arose there about the same time on the two lists.

Lloyd Anderson
Ecological Linguistics


From ECOLING at aol.com  Mon Jul 26 15:12:25 1999
From: ECOLING at aol.com (ECOLING at aol.com)
Date: Mon, 26 Jul 1999 11:12:25 EDT
Subject: Hittite & Celtic dative in /k/ ?
Message-ID: <MON.26.JUL.1999.111225.EDT.>

In the light of the discussions on Hittites and other early
Indo-European peoples of Anatolia and the Balkans,
going on on IE and ANE lists,
it may be relevant to consider whether even Celts and Hittites
may have shared any things linguistically or culturally,
as areal manifestations.

The standard analyses I think would still say that anything
shared between them is a relic of the common IE stage,
because Hittite and Celtic did not share any innovations together.
I am asking also the IndoEuropean list (when it resumes activities),
if we can get some comments and guidance as to the current
state of knowledge.  Obviously the work of Ringe has been
prominent recently.  Any current views of what Ringe has succeeded
in pinning down re Proto-IE and Hittite, Celtic, Tocharian,
and what Ringe's methods have not pinned down conclusively,
either because the choice of data determines the conclusions
or for other reasons?

Here it is one small item for your consideration.

Here is a tidbit linking Celtic and Hittite
which I found many years ago, when compiling typological
comparisons of the semantic domains of "be" and "have".
I wondered at the time whether it is a relic Dative case
preposition / postposition which Hittite and Celtic shared,
presumably as a retention, but conceivably as an areal phenomenon,
or both of the above.  Does Tocharian have it too?
(I have no idea whether the later
Indic Dative postposition with /k/ is related or a chance lookalike.)

The forms being discussed, the dative case of the 1st singular pronoun,
are statistically likely to be highly conservative,
both because they are pronouns and also
as oblique cases rather than the more often innovating nominative etc.

**********

Hittite:

amm-el  es-tsi     "(it) is mine"
amm-uk  es-tsi     "(it) is to me", or rather in English, "I have it"
me-Postpos.  is-3sg.

amm-uk "me-to"

The above forms were cited by Calvert Watkins for the contrast of
Genitive with Dative.

**********

Celtic preposition ag- where Hittite had postposition -uk above:

Irish
ta' ... aige     "he has", with the preposition   /ag-/ (Watkins)
ni fhuil fear agam  "not husband to-me" or rather "I have no husband"

ag-am "to-me"

Gaelic

tha airgiod agam  "is money to-me", or rather "I have money"

***

Lloyd Anderson


From 114064.1241 at Compuserve.com  Wed Jul 28 19:14:00 1999
From: 114064.1241 at Compuserve.com (Damien Erwan Perrotin)
Date: Wed, 28 Jul 1999 21:14:00 +0200
Subject: Latin perfects and Fluent Etruscan in 30 days!
Message-ID: <WED.28.JUL.1999.211400.0200.>

-----Original Message-----
 From  Eduard Selleslagh <edsel at glo.be>
Date Sun, 11 Jul 1999 12:31:58 +0200

>Ed said

>There seems to be a slight misunderstanding here: with "later Lat. 'am-'" I
>meant the beginning of words like 'amare', 'amicus', etc. , i.e. without
>the -b-, which - according to the reasoning above - would have transited via
>Etruscan, as opposed to those that came straight from PIE and preserved the
>(a)m(.)b-.

>Apart from all that, I never suggested that 'ambo', ambi-', Grk. 'amphi'
>(and Cat. 'amb', which proved to be unrelated) etc. were of Etruscan origin,
>only that they should be added to the data pool when looking at possible
>relationships of Etr. 'am(e)-' and Lat. 'amicus' etc. It seems to me that
>this led to an interesting discussion that yielded the possibility of the
>existance of two parallel paths: one 'directly' from PIE to Latin and one
>via Etruscan.

Ed

[Damien Erwan Perrotin]

The idea is interesting, but probably inexact as am- is not present only
in Latin but also in Lydian (ama : to love) and in Breton (afan : to
kiss from an older Brythonnic *ama). It is still possible that the
Lydian form was a borrowing from an Etruscan-like tongue of the Aegean,
but that is quite unlikely for Breton. So there can be only three
explainations for the ressemblance you point out :
a: chance ressemblance (always possible)
b: Etruscan borrowed the word from Latin or from Celtic
c: Etruscan is remotely linked to IE and this root is a remnant of this
old relationship.

Personnally, I favor the third thesis, but there is still work to do
before proving it.

Damien Erwan Perrotin


From Georg at home.ivm.de  Thu Jul 29 18:14:56 1999
From: Georg at home.ivm.de (Ralf-Stefan Georg)
Date: Thu, 29 Jul 1999 20:14:56 +0200
Subject: PIE vs. Proto-World (Proto-Language)
In-Reply-To: <006c01bed1a6$a63637e0$319ffad0@patrickcryan>
Message-ID: <THU.29.JUL.1999.201456.0200.>

>Pat responds:

>There is no *tangible* way for us to ever know whether languages arose
>monogenetically or polygenetically however most linguists, even when they
>deny its recoverability, have correctly weighed the odds of mono- vs.
>polygenesis, and subscribe to monogenesis. If a probablistically calculated
>hypothesis is "ideology", then everything done in historical linguistics is
>"ideology".

"most linguists" ???????????????
I'm flabbergasted. I would like to know *one* of those, who did/does what
you claim "most linguists" do.

Stefan Georg
Heerstrasse 7
D-53111 Bonn
FRG
+49-228-69-13-32


From rmccalli at sunmuw1.MUW.Edu  Thu Jul 29 20:06:00 1999
From: rmccalli at sunmuw1.MUW.Edu (Rick Mc Callister)
Date: Thu, 29 Jul 1999 15:06:00 -0500
Subject: PIE vs. Proto-World (Proto-Language)
In-Reply-To: <006c01bed1a6$a63637e0$319ffad0@patrickcryan>
Message-ID: <THU.29.JUL.1999.150600.0500.>

	Unless you're saying that language arose in the last 100,000 years
[i.e. before humans left Africa], then I don't think you can make a serious
claim that there are any true isolates. Basque and Burushaski are surely
related to other languages but the lumpers are going to have to work a lot
harder to prove it.
	My understanding is that language arose before humans left Africa,
so any claims of polygenesis would have to be examined among African
languages. Given that the only existing language families in Africa are
Niger-Kordofanian, Nilo-Saharan, Afro-Asiatic and the Khoisan languages
[which may be between 1and 5 families], it seems that the onus of proof is
on the polygenesists.

[snip]

>Ralf-Stefan wrote:

>> And, if I may insert this much to Pat's chagrin, the notion that human
>> language is a monogenetic phenomenon is aprioristic ideology. Even the
>> notion that every known language, as Basque, Burushaski or whatnot, has to
>> be related to some other language is ideology. "Having consonants" may be a
>> universal feature of human language, "Being related to some other lg."
>> simply is not.

>Pat responds:

>There is no *tangible* way for us to ever know whether languages arose
>monogenetically or polygenetically however most linguists, even when they
>deny its recoverability, have correctly weighed the odds of mono- vs.
>polygenesis, and subscribe to monogenesis. If a probablistically calculated
>hypothesis is "ideology", then everything done in historical linguistics is
>"ideology".

[snip]

Rick Mc Callister
W-1634
Mississippi University for Women
Columbus MS 39701


From Georg at home.ivm.de  Thu Jul 29 19:42:59 1999
From: Georg at home.ivm.de (Ralf-Stefan Georg)
Date: Thu, 29 Jul 1999 21:42:59 +0200
Subject: PIE vs. Proto-World (Proto-Language)
In-Reply-To: <Pine.SO4.4.05.9907211722250.23558-100000@rask>
Message-ID: <THU.29.JUL.1999.214259.0200.>

>On Fri, 16 Jul 1999, Ralf-Stefan Georg wrote:

>> "Having consonants" may be a
>> universal feature of human language, "Being related to some other lg."
>> simply is not.

>A fascinating thought:
>Imagine "Being related to some other lg." WAS actually a universal feature
>of (spoken) human language, wouldn't it then be the only true universal?
>In other words, if all spoken languages should prove to be derived from
>one common source, would we then be able to distinguish true language
>universals from other features shared by all languages ONLY because of the
>fact that these features were present in "proto-world"?

Yes, of course we would, given a little progress in the field of typology
and universals (but in some subareas we are close). We could when and only
when we succeed in *explaining* language universals, i.e. explain them as
*necessary* ingredients of a communication system serving all the purposes
human language does serve. If it can be shown that, in order to be a
functional, viable means of intra-group communication, a sign system will
have to share in a (however small) set of features, without which it could
hardly work, let alone stabilize, these universals would be universals
independent of contiguous inheritance from some proto (where they could be
as arbitrary as meaning-to-form-relations generally are).
Admittedly, research into universals has not yet reached the state where we
can tell all this, but that's one of the direction it is heading in (at
least, imho, should be). Compare this to the "grand unified theory"
physicists are hunting for.
But, theory aside, "being related to some other lg." is not a language
universal. Polygenesis is equally likely as monogenesis, I can't see any
reasons why the opposite should hold.

St.G.

Stefan Georg
Heerstrasse 7
D-53111 Bonn
FRG
+49-228-69-13-32


From Georg at home.ivm.de  Thu Jul 29 20:17:39 1999
From: Georg at home.ivm.de (Ralf-Stefan Georg)
Date: Thu, 29 Jul 1999 22:17:39 +0200
Subject: PIE vs. Proto-World (Proto-Language)
In-Reply-To: <l03130305b3b941708923@[209.147.211.21]>
Message-ID: <THU.29.JUL.1999.221739.0200.>

I don't think I'm following.

>	You raise the questions of what is relationship and what is the
>difference between language and speech.
>	All languages are related in the same sense that all humans are
>related.

Why that ? Language is a cultural item, a world-three-phenomenon (Popper),
basically an artefact (though one constructed less consciously than hammers
and anvils, admittedly). Humans are related by material relationships we
are endowed with by birth. Language has to be learned, in each and every
case; every single bit of them (basically uncontroversial; I add a tiny bit
of possibly controversial stuff: even the traits of languages which are
"universal" are learned).

> All humans settled on oral speech rather than sign language or
>flatulence for communication.

There is no need to view this as an invention, which, by its specificity
and ingenuousity, can only have taken place once. It may have taken place
several times, because oral speech is the one means of communication which
is in functional terms vastly superior to signs or flatulence. This is
because of the possibility of having a doubly articulated system (a finite
set of meaningless elements are combined to form a [still ]finite set of
meaningful ones, which can be used to form an infinite set of meaningful
utterances; do this with modulations of flatulence and you have your
language).
Because it is so vastly, tremendously practical, oral speech won the day
among humans with brains capable to process all the neuronal activity
necessary to control and handle it.

> Given that all human brains are wired for
>speech in the same way, this suggests that there is a degree of
>relationship.

I contest this given. Human brains are *capable* to process language, i.e.
sign systems complicated and open enough to do the things language does.
That any specifics of these systems are hard-wired in human brains is
sometimes asserted (and sometimes makes its way into the popular press in
these days of reductionism) but not necessarily demonstrated. Some believe
this, some don't, guess what I do ;-)

>	I don't think this question can be answered in absolute terms but
>it may possibly be answered in practical terms. Given that modern humans
>left Africa about 100,000 years ago, there is a good chance that all of
>these are ultimately related.

What has this time-span to do with the good chance you mention ? There is
also a good chance that modern humans, when heading out of Africa had
several, if not many, languages, culturally stabilized to be
group-defining, yet materially independent from each other. We don't know,
whether the one scenario or the other is nearer to the "truth". But I, for
one, find the polygenetic one equally possible, if not slightly more
appealing. First you have to have the brains (and, yes, the vocal tract
will help as well) to manage such a thing as language (not with anything
pre-wired-in, just the capability, i.e. the neuronal complexitiy needed).
Then you will find out with tremendous speed just how *handy*  such a thing
as articulate vocal language is for any kind of group activity. It may well
be the most useful thing humans ever invented (I use "invented" without
irony, but please be aware of the "world-three-phenomenon"-caveat I
introduced above, or maybe in one of the other posts I have been firing
today ...). So, this very usefulness, this simply ir-re-sis-ti-ble
usefulness will lead intelligent beings to develop such a thing, resp. to
make the more intelligent and powerful ones bestow it on their
in-group-members (which are pre-human phenomena, of course). Being able to
do it and finding it so ri-di-cu-lous-ly useful is enough to actually do
it. Some attempts are successful, some less so (i.e. some languages make
the step from intra-group codes to means of inter-group communication
aso.). This *could* be the story.
Or do you think the trick of banging the rocks together was invented only
once ? Or that of throwing meat into the fire ?

We should overcome the romantic view of seeing in language something as
"natural" as having 32 teeth. Language is there because it serves a
fantastic range of *functions*, and it is in terms of these functions that
we will eventually understand it.

PS: I know this is controversial, I have at least the whole MIT community
against me (Or better, lest this sounds pretentious: I would have them
against me if they knew who I am, which fortunately they don't ;-), and,
though this scares me a bit, I take the freedom to express on this informal
medium: eppur si muove ;-)

But, on the other hand, I know I'm not entirely alone ...

St.G.

Stefan Georg
Heerstrasse 7
D-53111 Bonn
FRG
+49-228-69-13-32


From Georg at home.ivm.de  Thu Jul 29 20:31:41 1999
From: Georg at home.ivm.de (Ralf-Stefan Georg)
Date: Thu, 29 Jul 1999 22:31:41 +0200
Subject: PIE vs. Proto-World (Proto-Language)
In-Reply-To: <Pine.SO4.4.05.9907231417430.6558-100000@rask>
Message-ID: <THU.29.JUL.1999.223141.0200.>

>> Even the
>> notion that every known language, as Basque, Burushaski or whatnot, has to
>> be related to some other language is ideology.

>I suppose it is common sense to believe that "isolates" like Basque or
>Burushaski at least have *dead* relatives. A language in colloquial
>use never stays ONE language, but will inevitably split up into several
>new languages. Which is exactly why "isolate" is merely a term
>used for languages that have not yet been proven to be related to other
>languages. Every single language has a mother and sisters. This is also
>true of PIE, of course.
>I could accept your statement above, if you inserted the word "living"
>between "other" and "language".

Granted, but I'd prefer "known" to "living".
It is of course correct, given the overall presence of language change,
interaction between human groups, spread of human groups over territories
too big to maintain uniformity of language from border to border (and all
the other things which happen to human groups over millennia) that, at some
time and place in history, languages were spoken by some groups which we
would - had we access to them - identify as relatives of Basque (in this
case, Aquitanian is of course a partly known instance of this)  or
Burushaski (the fragmentary attested Bru-Zha of some Tibetan sources may as
well be Burushaski's Aquitanian, but noone really knows).
I was rather aiming at the pool/set of known and identifiable languages and
the question whether they of necessity are classifiable into one stemma,
or, slightly different, whether all members of this set are classifiable
into *some* stemma with others. Some are, some aren't. This does not mean
that, e.g. for Basque, this will of necessity never be the case. Perhaps it
will be the case. But it hasn't happened yet, and to force it, just because
Basque (or Kusunda, or Nivkh, or Yenisseyan) *has* to fit somewhere, just
because it *cannot* be allowed that they remain "orphans forever" is, imho,
misguided. They can, such is life.

St.G.

Stefan Georg
Heerstrasse 7
D-53111 Bonn
FRG
+49-228-69-13-32


From Georg at home.ivm.de  Thu Jul 29 20:38:29 1999
From: Georg at home.ivm.de (Ralf-Stefan Georg)
Date: Thu, 29 Jul 1999 22:38:29 +0200
Subject: PIE vs. Proto-World (Proto-Language)
In-Reply-To: <Pine.SO4.4.05.9907211902570.24271-100000@rask>
Message-ID: <THU.29.JUL.1999.223829.0200.>

>> The link between Sanskrit and Latin or Classical Greek is so close that any
>> layman can see it; whole phrases are nearly identical.

>> This is, to put it mildly, not the case between, say, English and Bengali.

>X
>But if we only knew the modern IE languages, we would still be able to
>reconstruct words like:
>*p at ter 'father'
>(on the basis of, say, English <father>, Italian <padre>, and Hindi
><pita>)
>*newos 'new'
>(on the basis of, say, Modern Greek <neos>, Portuguese <novo>, and Polish
><nowy>)

I doubt that we could *reconstruct* anything on the basis of these isolated
examples, but we could of course suspect that something with these
languages seems to be different than with, say, Burushaski and Quechua. At
least one thing is sure: we'd had a great deal of discussion about chance
resemblances and stuff at our hands as we have now. We would have a hard
time to tell the hows and whys of pairs like /father/: /padre/ but also e.
/paternal/ aso. aso.

>The same is true for morphological paradigms etc.

Well, I would love to see morphological paradigms reconstructable on the
basis of English, Italian and Hindi ...

Stefan Georg
Heerstrasse 7
D-53111 Bonn
FRG
+49-228-69-13-32


From X99Lynx at aol.com  Thu Jul 29 20:48:40 1999
From: X99Lynx at aol.com (X99Lynx at aol.com)
Date: Thu, 29 Jul 1999 16:48:40 EDT
Subject: PIE vs. Proto-World (Proto-Language)
Message-ID: <THU.29.JUL.1999.164840.EDT.>

In a message dated 7/29/99 2:27:35 PM, you wrote:

<<All humans settled on oral speech rather than sign language or
flatulence for communication.  Given that all human brains are wired for
speech in the same way...>>

Not a given some of us accept.  But given common human physiology in general,
the fact that sound travels farther and faster than hand signals
and that the vocal cords and jaw are more controllable than other body parts
that make sounds - oral speech is by far the most practicable way to
communicate before the invention of the written word (which makes use of
another part of our body that has more sensitive controls.)

<<Given that modern humans left Africa about 100,000 years ago,...>>

There's the rub.  You'll need to backtime that a bit.  And whether
genetically endowed by a miraculous quirk of nature or invented by us
ourselves, a universal first language would have to go back far enough to
include ancestors of the Chinese - and that may be going back a ways.

Was one day a child born with pre-wired language to non-lingual parents and
did she/he essentially sit around talking to her/himself until a similar
sibling showed up and the first conversation happened?
Or was language invented in one place at one date and spread from there?   Or
did it arise as human communities and cultures reached the point where
assorted communal grunts, groans, clicks and fricatives became customary and
structured and were passed on for the immense practical value of
understandable communication.  If the latter, then language could well have
developed in different places at different times.

Regards,
Steve Long


From larryt at cogs.susx.ac.uk  Fri Jul 30 07:54:17 1999
From: larryt at cogs.susx.ac.uk (Larry Trask)
Date: Fri, 30 Jul 1999 08:54:17 +0100
Subject: PIE vs. Proto-World (Proto-Language)
In-Reply-To: <008201bed1aa$79bc9480$319ffad0@patrickcryan>
Message-ID: <FRI.30.JUL.1999.085417.0100.>

On Mon, 19 Jul 1999, Patrick C. Ryan wrote:

[on monogenesis]

> The "persuasive case" has already been made. Monogenesis is much likelier
> than polygenesis. And you have written so yourself!

No, I haven't.

What I have said is that we have no evidence either way, but that the
majority of linguists probably lean toward monogenesis.

I myself am agnostic.  Monogenesis appeals to me because it is
economical, but there is no way on earth I could defend monogenesis
against someone who doubted it.

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk


From larryt at cogs.susx.ac.uk  Fri Jul 30 08:14:22 1999
From: larryt at cogs.susx.ac.uk (Larry Trask)
Date: Fri, 30 Jul 1999 09:14:22 +0100
Subject: PIE vs. Proto-World (Proto-Language)
In-Reply-To: <Pine.SO4.4.05.9907211722250.23558-100000@rask>
Message-ID: <FRI.30.JUL.1999.091422.0100.>

On Wed, 21 Jul 1999, Adam Hyllested wrote:

> On Fri, 16 Jul 1999, Ralf-Stefan Georg wrote:

>> "Having consonants" may be a
>> universal feature of human language, "Being related to some other lg."
>> simply is not.

> A fascinating thought: Imagine "Being related to some other lg." WAS
> actually a universal feature of (spoken) human language, wouldn't it
> then be the only true universal?  In other words, if all spoken
> languages should prove to be derived from one common source, would
> we then be able to distinguish true language universals from other
> features shared by all languages ONLY because of the fact that these
> features were present in "proto-world"? Language typology would
> still be a great help in the work of reconstructing proto-languages,
> but would it prove anything about the nature of spoken human
> language?

An interesting point.  This is the problem of `founder effects'.  A
founder effect is a feature of an ancestral language which is not
cognitively necessary -- that is, it is not a true language universal --
but which happens by chance to persist in all descendants of that
ancestral language.  Clearly founder effects have the potential to
distort our understanding of true linguistic universals.

This idea has received some discussion in the literature, though not, I
think, a great deal.  Johanna Nichols has appealed to founder effects to
account for the significantly non-random distribution of certain
structural and other features among the world's languages.  Nichols has
no interest in "Proto-World", but I can cite another case which is
relevant.

When I started my PhD some years ago, no example was known of a language
with object-initial basic word order (OVS or OSV), and it was widely
suspected that such languages were impossible.  Since then, a few
O-initial languages have turned up, practically all of them in Brazil.
If we hadn't managed to study those languages before they vanished, we
might have gone on believing forever that O-initial languages were
impossible.

Well, they are indeed possible, but they appear to be rare.  But why are
they rare?  Geoff Pullum once indulged in some musings on this point.
Maybe O-initial languages really are cognitively or functionally more
difficult than other types.  Then again, maybe they are not, and the
observed distribution is an accident, an artefact of human history:
speakers of non-O-initial languages have simply been lucky enough to
flourish at the expense of speakers of O-initial languages.  If the
second is the case, then we are looking at a kind of founder effect.
But who knows?

There exists a principle, called the Exit Principle, which says this:
any feature of a language which is not cognitively necessary can change.
This implies that every such feature *will* change eventually, but who
knows how long `eventually' might be?

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk


From larryt at cogs.susx.ac.uk  Fri Jul 30 08:17:43 1999
From: larryt at cogs.susx.ac.uk (Larry Trask)
Date: Fri, 30 Jul 1999 09:17:43 +0100
Subject: PIE vs. Proto-World (Proto-Language)
In-Reply-To: <Pine.SO4.4.05.9907231417430.6558-100000@rask>
Message-ID: <FRI.30.JUL.1999.091743.0100.>

On Fri, 23 Jul 1999, Adam Hyllested wrote:

> On Fri, 16 Jul 1999, Ralf-Stefan Georg wrote:

>> Even the
>> notion that every known language, as Basque, Burushaski or whatnot, has to
>> be related to some other language is ideology.

> I suppose it is common sense to believe that "isolates" like Basque
> or Burushaski at least have *dead* relatives. A language in
> colloquial use never stays ONE language, but will inevitably split
> up into several new languages. Which is exactly why "isolate" is
> merely a term used for languages that have not yet been proven to be
> related to other languages. Every single language has a mother and
> sisters. This is also true of PIE, of course.

> I could accept your statement above, if you inserted the word
> "living" between "other" and "language".

Well, `known' rather than `living', I would suggest.  I took Stefan's
statement to mean `known language', since I cannot conceive of any
program relating known languages to unknown languages.

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk


From JoatSimeon at aol.com  Fri Jul 30 20:30:28 1999
From: JoatSimeon at aol.com (JoatSimeon at aol.com)
Date: Fri, 30 Jul 1999 16:30:28 EDT
Subject: PIE vs. Proto-World (Proto-Language)
Message-ID: <FRI.30.JUL.1999.163028.EDT.>

>adahyl at cphling.dk writes:

>But if we only knew the modern IE languages, we would still be able to
>reconstruct words like:

-- if it had ever occurred to anyone that the languages were related.  Note
that the whole concept is only about 200 years old, and that comparative
linguistics only really got started in the 19th century.

In any case, the number of reconstructions possible with only extant
languages would be very much smaller.

Linguistic information is lost over time. The more time, the more lost.


From JoatSimeon at aol.com  Fri Jul 30 20:38:04 1999
From: JoatSimeon at aol.com (JoatSimeon at aol.com)
Date: Fri, 30 Jul 1999 16:38:04 EDT
Subject: PIE vs. Proto-World (Proto-Language)
Message-ID: <FRI.30.JUL.1999.163804.EDT.>

>(Patrick C. Ryan)

 > -- with minor exceptions (such as "kuku"), words _are_ arbitrary sound
 > assemblages.

>Pat writes:

>Why do you not explain to us all why that is true? I am firmly convinced
>that it is unequivocally incorrect.

-- because any sound within the human range will do as well as any other for
any given referent.  A rose may be a znfargle.

Humans have the capacity to learn a language, but not any particular
language; a language must have words, but not any particular words.


From kurisuto at unagi.cis.upenn.edu  Sat Jul 31 05:08:57 1999
From: kurisuto at unagi.cis.upenn.edu (Sean Crist)
Date: Sat, 31 Jul 1999 01:08:57 -0400
Subject: PIE vs. Proto-World (Proto-Language)
In-Reply-To: <006c01bed1a6$a63637e0$319ffad0@patrickcryan>
Message-ID: <SAT.31.JUL.1999.010857.0400.>

On Mon, 19 Jul 1999, Patrick C. Ryan wrote:

> There is no *tangible* way for us to ever know whether languages arose
> monogenetically or polygenetically however most linguists, even when they
> deny its recoverability, have correctly weighed the odds of mono- vs.
> polygenesis, and subscribe to monogenesis.

I don't think this correctly represents the field; I'd say that most
linguists are agnostic on this question.  The majority view is that there
isn't evidence to allow us to answer the question one way or the other. In
any case, it's not a question that most linguists spend a lot of time
pondering.  There was so much unverifiable silliness written on this topic
earlier in the history of linguistics that the whole topic has a rather
disreputable air to it.

> If a probablistically calculated
> hypothesis is "ideology", then everything done in historical linguistics is
> "ideology".

A probabilistic model is one in which each of the possible outcomes of an
event has a particular numeric probability assigned to it.  The
Comparative Method is not probabilistic; it is categorical.

I can't see any sense in which the monogenesis hypothesis is
"probabilistically calculated".  One can imagine that there could be a
probabilistic model of language genesis, but there are so many unknowns
that such a model would likely not be of much use.  It would probably be
worse off than (for example) the Drake formula for calculating the
probability that there is intelligent life elsewhere in the universe; the
fact is that we just don't know the answer, and there's nothing wrong with
saying that we don't know.

  \/ __ __    _\_     --Sean Crist  (kurisuto at unagi.cis.upenn.edu)
 ---  |  |    \ /     http://www.ling.upenn.edu/~kurisuto/
  _| ,| ,|   -----
  _| ,| ,|    [_]
   |  |  |    [_]


From petegray at btinternet.com  Sat Jul 31 10:00:11 1999
From: petegray at btinternet.com (petegray)
Date: Sat, 31 Jul 1999 11:00:11 +0100
Subject: PIE vs. Proto-World (Proto-Language)
Message-ID: <SAT.31.JUL.1999.110011.0100.>

Rick said:
>. DNA studies obtensibly show that non-African
> humans seem to go back a single population distinct from Africans.

"Distinct" from Africans?   Not in my reading of the texts.  I don't wish to
raise a non-linguistic topic, but your implication - that languages split
into African and non-African - is a language topic, and we need some
scientist out there to give us the DNA truth.

Peter


From proto-language at email.msn.com  Thu Jul 29 21:12:30 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Thu, 29 Jul 1999 16:12:30 -0500
Subject: PIE and Proto-Language
Message-ID: <THU.29.JUL.1999.161230.0500.>

Dear Ralf-Stefan and IEists:

 ----- Original Message -----
From: Ralf-Stefan Georg <Georg at home.ivm.de>
Sent: Monday, July 19, 1999 4:36 AM

>> Pat interjects:

>> Here I think you are dangerously introducing the mistaken terminology of the
>> opposing argument. I do *not* look for "look-alikes"; I only am interested
>> in "cognates" the phonological forms of which can be supported through
>> multiple comparisons. For example, one set of interesting IE and Sumerian
>> "cognates" shows IE *-wey- = Sumerian -g{~}-; interestingly, this
>> development *is* found in *some* IE languages, like nearby Armenian.

Ralf-Stefan responds:

> "interestingly" ? Why interestingly ? The only way this could be
> "interesting" would, imho, be to base a claim of Leskienian solidaric
> innovation on it, leading to a Sumero-Armenian subgroup of Indo-European.
> Very interesting.
> Or some kind of areal phenomenon, potentially interesting (this time
> without irony). But: supposing for a split-second that IE and Sumerian are
> somehow related, the /w/ --> /g/ shift would be something very old, right?
> Now, the Armenian /w/ ---> /g/ shift is young, since known loanwords
> participate in it (so young actually, that Kartvelian managed to preserve
> the intermediary stage /gw/ in some of those LWs). And of course, even if
> chronology were no obstacle, the whole thing still begs the question of
> Sumero-IE, but I understand that this is a) not a topic for this list and
> b) that I don't stand the slightest shred of a chance to talk you out of
> it; so I don't try.

Pat responds:

I do not know of any ready way tp gauge how old or young the phenomenon is
in Sumerian. But, I regard it probably as an areal phenomenon.

Also, as I perface every essay I have written with a caveat against assuming
that I regard IE and any compared language as particularly close, there is
no real question of "Sumero-IE" for me. On the other hand, Gordon Whittaker
in Goettinger Beitraege zur Sprachwissenschaft feels rather differently
about it. His idea is that Sumerian was imposed on an IE substratum. His
article is interesting for the many valid comparisons he makes (IMHO)
between some Sumerian words and IE ones.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From Georg at home.ivm.de  Thu Jul 29 09:52:49 1999
From: Georg at home.ivm.de (Ralf-Stefan Georg)
Date: Thu, 29 Jul 1999 11:52:49 +0200
Subject: accusative and ergative languages
In-Reply-To: <001e01bed1af$abebec80$629ffad0@patrickcryan>
Message-ID: <THU.29.JUL.1999.115249.0200.>

>S wrote:

>> How do you live with the fact that some "ergative languages" have
>> independent passives ?

>Pat asks:

>Generalities are somewhat interesting but specifics would be even more
>interesting.

Sorry, deeply sorry, but I'm not able to believe that we are discussing
ergativity here, the existence of splits of some sort in every language
conventionally dubbed "ERG", the question whether ERG-constructions are
really passive constructions and stuff like that, without you knowing that
the phenomenon I alluded to in passing exists. I don't believe it.
I don't think that anyone here will need examples for this, so I'll only
give them when you explicitly express that you doubt/know better/reject the
fact that ergative constructions and passives can coexist in a language.
One hint to spare you some energy: the name of the first language which
comes to my mind in this respect sounds pretty close to my last name ...

>> And, as far as I remember, this whole brouhaha started with me asserting
>> that all known "ergative languages" have some split, followed by you
>> prompting we to show such a thing in Sumerian. I'm not a mind-reader, but
>> this could be interpreted as a challenge aiming at the gist of my
>> assertion, n'est-ce pas ? I managed to show those splits,

>Pat interjects:

>To my knowledge, you did not.

I did, Larry did, Wolfgang did (put me last on this list). However, I never
actually expected you to concede this, nor will you ever, no matter what.
So why should I bother to try to make you change your mind ? Each time
something is shown which does not square with your notion of ergativity (or
inflection, or whatnot) you change the definition. Let's begin (all over
again, sigh) by answering Wolfgang's questionnaire. I think this will be
the only way to get somewhere. You start, of course.

>> although the
>> discussion got a bit swamped under some hassle over maru:- and
>> HamTu-conjugations, but I managed to do it. At least I got you to accept,
>> late in coming, though, that my initial assertian still stands up.

>Pat responds:

>I did *not* accept your initial assertion. Re-read my concession.

Well, you are right here. Actually, if you *had* accepted it, I would have
to look for a flaw in *my* argumentation ;-)

St.

Stefan Georg
Heerstrasse 7
D-53111 Bonn
FRG
+49-228-69-13-32


From Georg at home.ivm.de  Thu Jul 29 10:47:45 1999
From: Georg at home.ivm.de (Ralf-Stefan Georg)
Date: Thu, 29 Jul 1999 12:47:45 +0200
Subject: accusative and ergative languages
In-Reply-To: <006101bed1a5$98f8be00$319ffad0@patrickcryan>
Message-ID: <THU.29.JUL.1999.124745.0200.>

>Pat responds:

>I suggest you read it. "But we do still encounter scholars who insist that
>there is a necessary diachronic connection, e.g. Estival and Myhill
>(1988:445): 'we propose here the hypothesis that in fact all ergative
>constructions have developed from passives'."  Are you suggesting that
>Estival and Myhill are not "linguists", or that Shibatani, in whose book
>this essay appeared, is not a "linguist"?  The quotation above is from p.
>189 of Dixon's 1994 book. What are you playing at?

Linguists can do wrong.

>Pat responds (on L.T's catalogue of subject properties):

>Why do you not give us an example of these so-called subject properties in
>some ergative language besides Basque? And how were these properties
>selected?

Empirically. What Larry gave was a *selection*, as you correctly name it.
It is so easy, you can do it yourself. S.b.

>Pat responds:

>If you are referring to the "tests" above, you have proved nothing.

This is much in line with what I said in  today's other post of mine. You
see, you read, you understand (or don't), but you reject nevertheless.
Always.
The subject-property-test *is* sufficient to discriminate between ergative
and passive constructions. Again:
Take an intransitive sentence in an exclusively accusative language. S-V
S is, by convention, called "subject" (just a convention). It's semantic
role can be one of the following: Agent, Experiencer aso. (some people
insist on Agent being restricted to agents of transitive constructions, bog
s nimi).
Now, take a transitive one. A-V-O
What accusative languages do is to treat S *and* A alike under all
circumstances (there are such languages, i.e. split-free accusative-lgs.).
This leads to the notion that these languages know a macro-category
comprising S and A, which most conventionally is called "subject" also, in
compliance with western grammatical tradition.
Now subjects have properties of various sorts, part of which have been
enumerated by Larry Trask. The real list is longer, and you can expand it
yourself. It is nothing more than a list of things which are generally true
for "subjects" in these languages.
Now, part of the things a passive formation does to a sentence is that the
*subject* is now what was semantically the undergoer/patient (choose your
favourite term) of the corresponding active sentence, an example:

active: Stefan proves Pat wrong (active, A-V-O)
           Pat is proven wrong by Stefan (passive, S-V[pass]-Agent-periphrasis)
"Pat", in this passive sentence, has been moved to a position (not only in
terms of word order), where it assumes all the functions commonly found
with S. It has become the "subject" of the sentence, according to
traditional terminolgy. The syntactic properties (partly enumerated by
Larry, but the list is longer) found with "Stefan" in the active sentence
are now found with "Pat" in its passive equivalent.

Now let's look at ergative constructions:

Again, we find intransitive sentences there: S-V (gosh, I'm starting from
scratch, but I think I have to)
And transitive sentences as well: A-V-O. What makes the construction
ergative is, of course, nothing but the fact that S and O are treated alike
in this language (or in some subsystems of this language).
Now, in a morphologically ergative language, such a sentence may look like:

A-erg V O-abs
a passive sentence, we recall, looks like (word order irrelevant): S V
Agent-periphrasis (P. is proven wrong by S., to take a random example with
positive truth-value, however;-).
Of course, ergative constructions and passives bear some resemblance: in
both the (semantic !) Agent is morphologically marked, in both the semantic
patient may be morphologically unmarked (this does not work when there is
an overt accusative marker in the language, of course).  However, this is
where the resemblance ends. In an ergative sentence like:
Stefan-Erg proves Pat-Abs wrong we should expect, under the
ergative-as-passive scenario, that subject properties remain with Pat (as
in the "true" passive above); however, they rest with "Stefan".
This is the general picture only, and the test has to be carried out with
real "ergative languages" of course, but this is what you will find. It is
nothing less that the watershed betweed ergative constructions and passive
constructions. It does not matter whether you accept the full list of
"subject properties" enumerated. You may accept some and dislike others.
With those you do accept, though, you will find this picture. The more
subject properties you can bring yourself to accept, the more instances of
the difference between ergative and passive constructions you will find.
It's that easy. But you do have to look at complex constructions, not only
at the minimal sentence.
This does *not* mean that ergativity in some given language may not have
come about via a passive transformation. It may. In order to claim the
right of being accepted as a "real" ergative construction (and not the
passive it may have been in an earlier life), the shift of subject
properties to the Agent-phrase (the ERG-NP) from the Patient-phrase is
crucial.
For a succinct demonstration that, what looked to some people like an
ergative construction, but is rather to interpreted as passive, see my (and
A.P. Volodin's) "Die itelmenische Sprache", Wiesbaden 1999, though I admit
that we could have been a bit clearer on the issue. If anything, this
thread has prompted me to write a clearer exposition of the relevant
chapter in this book.

>Pat responds:

>Obviously, I do not think so. And Estival and Myhill (and probably
>Shibatani) do not either --- not to mention the majority of linguists of the
>past.

Estivall/Myhill/Shibatani should speak up for themselves. As for the
linguists of the past, well, they are linguists of the past, if they had
got everything straight already in 1890, what the hell are we doing here (I
assume, that's what you ask anyway reading all this ;-).

Stefan

Stefan Georg
Heerstrasse 7
D-53111 Bonn
FRG
+49-228-69-13-32


From vidynath at math.ohio-state.edu  Thu Jul 29 12:31:37 1999
From: vidynath at math.ohio-state.edu (Vidhyanath Rao)
Date: Thu, 29 Jul 1999 08:31:37 -0400
Subject: accusative and ergative languages
Message-ID: <THU.29.JUL.1999.083137.0400.>

I wrote:
> (2) ra:men.a+as'vam a:ruhyate
>    [A] horse is being mounted by R.

What a horrid mistake. It should be ra:men.a+as'va a:ruhyate, with as'va in
the nominative [loss of the ending s by sandhi].


From proto-language at email.msn.com  Thu Jul 29 13:41:17 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Thu, 29 Jul 1999 08:41:17 -0500
Subject: accusative and ergative languages
Message-ID: <THU.29.JUL.1999.084117.0500.>

Dear Stefan and IEists:

 ----- Original Message -----
From: Ralf-Stefan Georg <Georg at home.ivm.de>
Sent: Monday, July 19, 1999 6:42 AM

Pat wrote:

>> In English, I can say: "John hit me, and he went away" or "I hit John and he
>> went away". What is the discourse cohesion strategy for English?

R-S wrote:

> Anaphora.

Pat writes:

Well, 'anaphora' is primarily a rhetorical device; my dictionary, however,
does acknowledge the use of 'anaphora' in a grammatical sense although Larry
seems to prefer "anaphor" in the grammatical and, I presume (but do not
know), 'anaphora' in the rhetorical sense.

Since "discourse cohesion strategy" is not defined in Larry's dictionary, I
have no idea exactly how one will want to define it. Perhaps 'anaphors' are
excluded; perhaps not.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From proto-language at email.msn.com  Thu Jul 29 16:05:15 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Thu, 29 Jul 1999 11:05:15 -0500
Subject: accusative and ergative languages
Message-ID: <THU.29.JUL.1999.110515.0500.>

[ moderator re-formatted ]

Dear Peter and IEists:

 ----- Original Message -----
From: petegray <petegray at btinternet.com>
Sent: Monday, July 19, 1999 2:41 PM

> Pat said:

>> ... the pattern "e-a-o" *...There was never a time during which *any* verb
>> in [e] followed this "pattern", or even *any* verb in /eR/ followed it
>> (weNden, waNdte, gewaNdt).

Peter wrote:

> Perhaps I misunderstand you.   This pattern is one of the standard patterns
> of strong verbs in modern German, most prevalent among verbs in -eR-, but
> not restricted to them, e.g.:

> befehlen, befahl, befohlen             to command
> bergen, barg, geborgen                to salvage
> nehmen, nahm, genommen          to take

Pat responds:
Peter, I majored in German so I am familiar with these "patterns".

Like so much of what we seem to do on this list, it is a question of
definitions: in this case, 'pattern'.

I confess that the definition I was using when I made the statement above is
1) idiosyncratic (Platonic[?]), and 2) possibly unsustainable.

Mixed up in it was the idea of '*general* applicability', and I can see now
that it is a definition of extremely limited usefulness.

I withdraw my statement, and grant your point.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From proto-language at email.msn.com  Thu Jul 29 16:29:26 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Thu, 29 Jul 1999 11:29:26 -0500
Subject: accusative and ergative languages
Message-ID: <THU.29.JUL.1999.112926.0500.>

Dear Larry and IEists:

 ----- Original Message -----
From: Larry Trask <larryt at cogs.susx.ac.uk>
Sent: Monday, July 19, 1999 5:49 AM

> On Sat, 17 Jul 1999, Patrick C. Ryan wrote:

>> As for the far-reaching conclusions of Dixon based on discourse
>> cohesion strategies, in my opinion they are flawed because these
>> strategies are purely conventional.

Larry responded:

> Not so.  Read on.

Pat responds:

I am going to defer a specific comment pending reading your answer to:
Schulze (" Language systems ARE conventional! This is one of the major
points of
 language tradition and L1 acquisition (despite of minimalism etc.)." in
another posting to this list.

Pat wrote:

>> In English, I can say: "John hit me, and he went away" or "I hit
>> John and he went away". What is the discourse cohesion strategy for
>> English?

Larry responded:

> You've overlooked the crucial null-subject cases:

> `John hit me and went away' *must* mean `John went away'.

> But `I hit John and went away' *must* mean `I went away'.

> Control of null NPs is one of the syntactic properties which crucially
> distinguish subjects from non-subjects in English and in some other
> languages.  And there is nothing "conventional" about it: this is a rule
> of English syntax.

Pat responds:

Another word for a "rule" is a "convention". Again, we appear to be chasing
our tails (definitions).

And the point that Leo makes in another posting deserves to be considered:

"Agreed -- and yet not quite agreed.  Are these really "subject
 properties"?  I freely acknowledge that that's what they're called, but
 for mainly
 for lack of any better name in normal linguistic terminology."

Larry continued:

> Let me change both NPs to third-person, to avoid any complications with
> agreement; suppose I say this:

> `John hit Bill and went away'.

> Now, in English, it is John who went away, not Bill.  However, according
> to my understanding of Dixon, if you say what looks like the literal
> equivalent of this in Dyirbal, it is *Bill* who went away, not John.
> This is one of the ways in which syntactic ergativity manifests itself
> in Dyirbal.

Pat answers:

Is it really that simple? Would we not be coming close to describing both
'nominative' and 'ergative' phenomena if we said that "a null-NP frequently
refers to the foregoing NP with which it agrees in case"?

 Larry continued:

> But not all ergative languages are the same here, and probably not even
> most.  If you translate this sentence as literally as possible into
> Basque, it is once again John who went away, and not Bill, just as in
> English.  This is so even though the ergative morphology of Basque is
> more thoroughgoing than that of Dyirbal, which is split.

> That is, Basque, like English, allows subjects to be coordinated with
> subjects, but not with non-subjects.  This is so even when one of the
> coordinated subject NPs is ergative and the other absolutive.  Basque
> does not allow the absolutive subject of an intransitive sentence to be
> coordinated with the absolutive object of a transitive sentence.

Pat responds:

First, as you know, I have questioned whether Dyirbal is, in fact, "split",
and have proposed a different explanation for the data.

Second, the point that Leo is making about distinguishing syntactic from
morphological subjects needs to be addressed by you in terms of "subject"
properties.

 Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From proto-language at email.msn.com  Thu Jul 29 17:09:10 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Thu, 29 Jul 1999 12:09:10 -0500
Subject: accusative and ergative languages
Message-ID: <THU.29.JUL.1999.120910.0500.>

Dear Leo and IEists:

 ----- Original Message -----
From: Leo A. Connolly <connolly at memphis.edu>
Sent: Monday, July 19, 1999 3:35 PM

Pat writes:

I do not want to divert you both from the discussion you are having about
related points, which I find very interesting, but I would like to comment
on one point.

Larry wrote:

>> The "passive" view of ergative languages in general is indefensible.

Leo wrote:

> Absolutely agreed!

Pat writes:

Again, I feel that it is again a battle of definitions. If by 'passive' we
mean a two-element construction in which the non-ergatively marked NP is the
patient of the VP, then an 'ergative' language like Sumerian is 'passive'.

Part of the problem, I think, is utilizing concepts like Larry's
"intrinsically transitive". 'Transitive' can be defined either as a
three-element construction or a construction of two elements with an implied
or differently expressed third element ("passive-with-agent"). Since I do
not believe in phenomena ex machina, I believe that all VP's essentially
include a causal factor (one-element) and an effectual one (one-element) so
that expressed or unexpressed, every VP implies two additional elements.

So, to my way of thinking, an 'intransitive' sentence like 'I go to the
city' is, from the standpoint of analysis, better regarded as a paraphrase
for 'I make myself move to the city'.

And a 'stative' like 'The flower is red' is better analyzed as a paraphrase
for 'Somebody/thing has reddened the flower'.

This analysis makes it easy to understand why, in an ergative language, the
'subjects' of 'intransitives' and 'statives' are in the absolutive. It
suggests that in an 'intransitive' construction, the expressed NP is not
necessarily the agent but may be the patient of a reflexive construction in
which the agent, being identical with the patient, has been deleted for
redundancy.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From proto-language at email.msn.com  Thu Jul 29 18:18:56 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Thu, 29 Jul 1999 13:18:56 -0500
Subject: accusative and ergative languages
Message-ID: <THU.29.JUL.1999.131856.0500.>

Dear Wolfgang and IEists:

 ----- Original Message -----
From: Wolfgang Schulze <W.Schulze at lrz.uni-muenchen.de>
Sent: Tuesday, July 20, 1999 5:15 AM

Wolfgang wrote:

> 16. Do YOU use a special theoretical frame work in order to substantiate
> your claims? If yes, which one?

Pat responds:

This is an excellent suggestion --- theoretically --- but I do not think it
is going to happen on this list. Answering your pertinent questions as they
should be answered would involve writing a book at least the size of
Dixon's. But, good thought!

Wolfgang continued:

> Now, let me finally turn to some claims by Pat:

> Pat wrote:

> Wolfgang wrote:

>>> Hence, there are NO "ergative (or "accusative") languages" or
>>> only, if you use this term in a very informal sense.

>> Pat wrotes:

>> In the sense you are using these, they seem to be of little value in
>> describing anything.

Wolfgang responded:

> I can only repeat what I have said in my earlier postings: ACC and ERG
> are nothing but the structural (and structured) reaction to more general
> principles of language based information processing. The polycentric
> architecture of language systems that encode these principles allows
> that the individual centers of this polycentric cluster react
> differently on these principles. Here, I cannot elaborate the underlying
> frame work which is labeled "Grammar of Scenes and Sceanrios" (GSS) and
> documented in Schulze 1998, chapter IV, but let me briefly say that I
> propose a (more or less) universal cause-effect 'vector' (C->E) itself
> metaphorized from underlying figure-ground relations (F->G) to be one of
> the most dominant principles of information processing. This vector can
> be weighted which leads to a continnum (here shortened) C->0(e) C->e >
> C->E > c->E > 0(c)->E (capitals represent heavy domains, small letters
> represent light domains). An ACC strategy would be to behave in a C->e
> sense, an ERG strategy would infer c->E. Note that these vector
> representations are NOT (in themselves) language specific or sensitive
> for specuial (linguistic) categories! Different categories may BEHAVE
> differently with respect to this continuum, regardless their own
> architectural make-up. ANYTHING in a language system may be sensitive
> for the AEC (accusative ergative continuum) as long as it is relevant
> for encoding the cause-effect vector (and its derivations). Hence, CASE
> may play a role just as AGR, word order, subject assignment,
> topicalization, discourse cohesion, co/subordination, paradigmatization
> of speech act participants and much more. But they may play their roles
> DIFFERENTLY! The description of their roles heavily depends from the
> diachrony of the paradigm in question, its formal architecture as well
> as its integration in a co-paradigmatic context ('structural coupling'
> in a broader sense). From this it follows that the individual centers of
> ALL language systems have to react upon the universal demands of the
> C->E vector, disregarding their internal architecture. Only IF ALL
> relevant centers (and they are many, I grant!) behave in one direction,
> THEN we are allowed to call the language system (or better, its
> Operating System) ACC or ERG. However, it is a much more difficult task
> to describe intermediate states that allow ACC in some parts of the
> Operating System, and ERG in some others. Here, we have to establish a
> (motivated!) hierarchy first of co-paradigmatic structures (such as CASE
> and AGR, CASE and word order, AGR and Personality, AGR and Noun Classes,
> to name only some). In such structures, one part sometimes is more
> dominant than the other with respect to its behavior on the AEC. If we
> can describe such dominant behavior we can refer to the whole structure
> as either more ACC or ERG. In a second step we have to go on describing
> the higher levels of this hierarchy (which itself should find an
> adequate linguistic explanation based on an appropriate language
> theory). Finally (and ideally) we would arrive at a term that would
> describe the functional dominance of one of the poles on the AEC with
> respect to an Operating System (not a language system) in toto. Only
> then, and I stress, only THEN we are allowed to use the term ergative or
> accusative with respect to an Operating System (for which 'ergative' or
> 'accusative language' would be an informal label).

Pat responds:

It would take my background than I have to properly apperciate your argument
above but, from what I *can* understand of it, I concur. As I wrote in
another post, and I hope I understand the above well enough to validly
connect it, I believe the critical factor is how cause-effect is mapped but
that the underlying cause-effect is the reality behind the masks.

>> Wolfgang wrote further:

>> Pat wrote:

>> I am sorry that I do not agree with the validity of this distinction (actant
>> vs. agent). For me, 'actant' is 'agent'. Perhaps you can explain the
>> difference.

Wolfgang explained:

> In terms of Functinal Grammars (as well as in GSS) 'actants' refer to
> ALL such linguistic expressions that encode a referential entity in a
> clause. Hence, in a sentence such as 'I met John several times in
> Chicago', 'I', 'John', 'times', and 'Chicago' are (abstract) actants
> that play different roles in the scene. But only 'I' is a linguistic
> agent, whereas John plays the role of a patient etc. Note that 'agent'
> and 'patient' are labels for semantic hyperroles (or macroroles in the
> sense of Foley/VanValin). I think that such a distinction is very
> helpfull. It is based on strong theoretical arguments and helps to avoid
> many false or at least problematic generalizations. A much more
> controversial (and much more difficult question is to define the labels
> 'subjective' (S), 'agentive' (A), and 'objective' (O) which should not
> be immediately equated to neither 'subject'/'object' nor to
> 'agent'/'patient'. S, A, and O are highly abstract terms that describe
> more structural than semantic or syntatic properties.

Pat responds:

Although I do not care much for 'actant' and think it has little to
recommend it over, say, NP, it is, of course, the prerogative of theorists
to introduce and define new terms as they choose.

Otherwise, if I understand you properly, I concur again.

>> Wolfgang wrote:

>>> The fact, however, is that many 'ergative' languages lack an
>>> antipassive. For instance, there are nearly 30 East Caucasian languages
>>> all of them using some ergative strategies in at least parts of their
>>> operating systems. But only a handfull of them (five or six, to be
>>> precise) have true antipassives (only one has some kind of
>>> "pseudo-passive").

>> Pat wrote:

>> Are you asserting that the majority of ergative languages do not have
>> anti-passives?

Wolfgang answered:

> I do not assert anything in the sense of 'ALL language have...'. Even
> the claim 'the majority of ergative languages (sic!) have...' is rather
> suspect to me. What I said is that in those 'ERG systems' I looked at
> (about 200) antipassives are rather the exception than the norm.

Pat responds:

I am not surprised really. I suspect the (original [?]) real function of an
anti-passive is suggesting lessened effective agency. What do you think?

>> Pat wrote:

>> As for the far-reaching conclusions of Dixon based on discourse cohesion
>> strategies, in my opinion they are flawed because these strategies are
>> purely conventional.

Wolfgang answered:

> Language systems ARE conventional! This is one of the major points of
> language tradition and L1 acquisition (despite of minimalism etc.).
> Discourse probably is one of the most important factors in the
> emergence, organization, and dynamics of language systems. We should not
> refer to the abstract notion of context-free 'sentences' that would be
> responsible for for grammatical 'events'. Such a view stems from the
> tradition of Classical Philosophy which is an INTERPRETATION of what
> goes on language. Today, we have become used to think of language in
> single sentences, to brak them up the way we do etc. But this is an
> analytic tradition, not part of the ontology of language itself, which
> is much more synthetic in nature than we are used to think. - A sentence
> does not function but in its co-text (as well as in its con-text). All
> sentence internal strategies used to be embedded in the techniques of
> co(n)textualization. No wonder, that ACC and ERG also work in this
> direction (though they may appear as more 'autonomous',
> sentence-internal mechanisms secondarily, especially if a language
> system as developed separate means to indicate discourse cohesion).

Pat responds:

Again, I agree. A valuable caveat, and often neglected.

 <snip>

>> Pat wrote (on Lak, East Caucasian):

>> I have to plead ignorance of Lak however, I find an analysis of an otherwise
>> completely ergative sentence as having an ACC word-order impossible. I
>> believe this opinion rests on a false analysis of word-order significance.

Wolfgang wrote:

> I tried to show that in Lak only some sentence patterns show 'complete
> ergativity'. Most of them show a mixed paradigmatic organization with
> respect to morphology (CASE, AGR). WOrd order is another VERY imporant
> aspect of the AEC. Consider e.g. a language that has canonical

> #SV
> #AOV

> (# = sentence boundary). Here, S behaves like A with respect to #, hence
> it is ACC. In #SV vs. #OAV it is S=O which indicates ERG behavior. The
> problem of actance serialization is also addressed in polypersonal
> systems (without CASE, e.g, Abxaz, Lakotha etc.). It IS a very important
> indicator for ACC, ERG, no question. Anything else would refer to the
> linguistic tradition of say 60 years ago (Bloomfieldian tradition).

Pat responds:

Sorry, Wolfgang, after so much with which I can agree, I am unconvinced. I
do not believe the essence of accusativity is in word-order.

>> Wolfgang continued (on Lak):

>>> But if you say "I am surely hitting you (plural)", you get:

>>> na b-at-la-ti-s:a-ra zu
>>> I:ABS I:PL-hit1-DUR-hit2-ASS-SAP:SG you:PL:ABS

>>> Here we have:

>>> ACC with respect to word order
>>> ERG with respect to class agreement
>>> ACC (or neutral) with respect to case marking)
>>> ACC with respect to SAP agreement (-ra is triggered by 'I:ABS').

>> Pat wrote:

>> So, agent marked ABS and patient marked ABS is, according to you, ACC with
>> respect to case marking? Sorry, not convincing at all! And I think the SAP
>> difference can be explained as indicating seriality of the patients involved
>> as recipients of the action by *one* agent.

Wolfgang responded:

> I said ACC (or neutral!) with respect to case marking! That means that A
> lacks ERG marking (which is OK with Silverstein). Semantically, it means
> that S is encoded like A (which is ACC). That fact that O is NOT in an
> accusative-like case does not argue against this assumption, because
> CASE is here structrually coupled with ADGR which allows us to classify
> AGR as ACC *here*. The SAP difference in Lak cannot be explained in the
> sense Pat proposes, Lak totally lacks such strategies.

Pat responds:

Sorry again. Perhaps it is the term "accusative" that is preventing me from
grasping your argument.

>> Wolfgang wrote further:

>>> Now, please tell me: Is Lak an 'ergative' or an 'accusative' language?
>>> [Please note that I did not include (among others) strategies of
>>> discourse cohesion, reflexivization and logophization].

>> Pat wrote:

>> By these examples, I would tell you that Lak is 'ergative'; and that
>> accusativity is not demonstrable from these examples.

Wolfgang responded:

> If you refer to the standard (morphological), but rather obsolete
> interpretation of ERG you may be right (but not for SAP NPs). But,
> fortunately, the ACC/ERG typology has freed itself from such a narrow
> interpretation of ACC/ERG which is nothing but a very small excerpt from
> the over-all typology.

Pat writes:

"Unchain my heart, and let me be free, dum-dee-dum . . .  (:-{})

>> Pat wrote previously:

>>>> I think it is likelier that, because of perceived greater animacy (or
>>>> definiteness), pronouns have a different method of marking that can still
>>>> be interpreted within an ergative context.

>> Wolfgang wrote:

>>> This a (very simplified) 'on-dit' that stems from the earlier version of
>>> the Silverstsein hierachy. Again, we have to deal with the question,
>>> whether a 'pronoun' (I guess you mean some kind of 'personal pronouns')
>>> can behave 'ergatively' or 'accusatively'. The list below gives you a
>>> selection of SAP case marking in East Caucasian languages with respect
>>> to ABS/ERG:

>>> ABS vs. ERG ABS = ERG
>>> ALL ---
>>> Singular Plural
>>> Plural Singular
>>> 1.Incl. Rest
>>> 1:SG Rest
>>> 2:SG Rest
>>> 1:SG/PL Rest
>>> --- ALL

>>> This list (aspects of personal agreement NOT included!) shows that SAP
>>> pronouns may behave differently within the same paradigm. Any
>>> generalization like that  one quoted above does not help to convey for
>>> these data...

>> Pat wrote:

>> I do not have a reference book for Lak so that my hands are somewhat tied.
>> But, I have found that paradigms are often inconsistent in ways that reflect
>> earlier lost phonological changes, or other lost schemata. The locative
>> plural terminations of IE are certainly not, in origin, terminations of the
>> locative plural. Etc.

Wolfgang responded:

> Nothing the like! The list I gave refers to what can be described for
> East Caucasian languages in toto! Most of these paradigms are
> functionally motivated, I grant (schulze in PKK 2 (Schulze 1999) will
> tell you the whole story).

Pat concludes:

Still not convinced on this point . . . and, I am trying hard.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From connolly at memphis.edu  Thu Jul 29 22:30:10 1999
From: connolly at memphis.edu (Leo A. Connolly)
Date: Thu, 29 Jul 1999 17:30:10 -0500
Subject: accusative and ergative languages
Message-ID: <THU.29.JUL.1999.173010.0500.>

Robert Orr wrote:

> There's also "sneak - snuck" in English and "schreiben -schrieb-geschrieben"
> in German, both of which are analogical formations  on originally borrowed
> words..

> Robert Orr

_Snuck_ is unquestionably analogical.  But it is quite impossible to
tell whether the strong inflection of _schreiben_ is, since the expected
Germanic reflex of PIE *_skreibh-_, which must underly Latin _scri:bo:_,
is precisely PGmc. *_skri:b-_, which would be expected to produce a
German _schreiben_ with strong inflection.

Leo
-


From larryt at cogs.susx.ac.uk  Fri Jul 30 08:40:22 1999
From: larryt at cogs.susx.ac.uk (Larry Trask)
Date: Fri, 30 Jul 1999 09:40:22 +0100
Subject: accusative and ergative languages
In-Reply-To: <006101bed1a5$98f8be00$319ffad0@patrickcryan>
Message-ID: <FRI.30.JUL.1999.094022.0100.>

On Mon, 19 Jul 1999, Patrick C. Ryan wrote:

[PR]

>>> Now, when Larry recently quoted Dixon about the nature of the
>>> ergative, he conveniently neglected to mention that Dixon
>>> acknowledged that there were currently practising linguists --- not
>>> amateur linguists like myself --- still defending the passive
>>> interpretation of ergative constructions.

[LT]

>> No; this is not so.  Read p. 189 of Dixon's 1994 book.

[PR]

> I suggest you read it. "But we do still encounter scholars who
> insist that there is a necessary diachronic connection, e.g. Estival
> and Myhill (1988:445): 'we propose here the hypothesis that in fact
> all ergative constructions have developed from passives'."  Are you
> suggesting that Estival and Myhill are not "linguists", or that
> Shibatani, in whose book this essay appeared, is not a "linguist"?
> The quotation above is from p. 189 of Dixon's 1994 book. What are
> you playing at?

The question is not what I'm playing at, but what you're playing at.

Your original assertion was that the passive interpretation of ergative
languages was not only defensible but, in your view, correct.  By this
you clearly meant that ergative constructions, in general, *are*
passives.  This is the view which was once popular among European
linguists, which has been effectively demolished, and which is correctly
dismissed by Dixon elsewhere on that same page as without support among
linguists today.

The point in your cited passage is an entirely different one: do all
ergatives derive historically from passives?  Dixon notes that a few
people have argued that the answer is `yes', but then goes on to provide
what, in his view, is good evidence that the correct answer is `no'.

Whatever view one might adopt on this second point, it is clearly
distinct from the first point.  Claiming that "all ergatives *descend*
from passives" is the same proposition as "all ergatives *are* passives"
is rather like claiming that "all humans descend from their
grandparents" is the same proposition as "all humans *are* their
grandparents.

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk


From larryt at cogs.susx.ac.uk  Fri Jul 30 14:54:03 1999
From: larryt at cogs.susx.ac.uk (Larry Trask)
Date: Fri, 30 Jul 1999 15:54:03 +0100
Subject: accusative and ergative languages
In-Reply-To: <006101bed1a5$98f8be00$319ffad0@patrickcryan>
Message-ID: <FRI.30.JUL.1999.155403.0100.>

On Mon, 19 Jul 1999, Patrick C. Ryan wrote:

> Still no word on the scene of the great "shredding" which was
> *claimed* by Larry.

I suggest you read my article in the following forthcoming book:

K. Davidse (ed.), Case and Grammatical Relations Across Languages, vol.
5: Nominative and Accusative, Amsterdam: John Benjamins

The book should be out within a few months, but I don't have a date yet.

> And what are subject properties that an absolutive NP never displays?

As I've already explained: in Basque, and reportedly in many other
ergative languages, the absolutive NP in a transitive sentence --
entirely *unlike* the absolutive NP in an intransitive sentence --
cannot control empty NPs, cannot itself be an empty NP, cannot control
reflexive or reciprocal pronouns, can itself be a reflexive or
reciprocal pronoun, cannot be coordinated with an intransitive subject,
and (in varieties with genitivization only) can be genitivized in
suitable circumstances.

In Basque, there is *no* syntactic property shared by absolutive NPs in
transitive sentences and absolutive NPs in intransitive sentences, apart
from those that are shared also by ergative NPs and sometimes also by
dative NPs.  The only properties peculiar to all absolutive NPs and
shared by nothing else are purely morphological ones: case-marking and
verbal agreement.

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk


From jer at cphling.dk  Fri Jul 30 15:50:53 1999
From: jer at cphling.dk (Jens Elmegaard Rasmussen)
Date: Fri, 30 Jul 1999 17:50:53 +0200
Subject: accusative and ergative languages
In-Reply-To: <002401bed2a4$93faa8a0$3970fe8c@lucent.com>
Message-ID: <FRI.30.JUL.1999.175053.0200.>

On Tue, 20 Jul 1999, Vidhyanath Rao wrote:

> Jens Elmegaard Rasmussen <jer at cphling.dk> wrote:

>> I'd say the cases I have seen of ergativity in Indic and Iranian languages
>> so clearly reflect underlying/earlier passive circumlocutions that
>> controversy is absurd.

> No!

> I have already pointed out the problem of distinguishing resultatives and
> passives. The ta-adjective is resultative in Vedic (Jamison, IIJ 198?) and
> Early Pali (Hendrikson, Infinite verb forms of Pali). And it patterns
> ergatively as resultative participles often do. The last part has been known
> forever. In English, see Speyer's ``Sanskrit Syntax'', will find the
> following: ``Of the participles in -ta the great majority have a passive
> meaning, hence it is customary to call the whole class the passive
> participle of the past. But some others are not passives, but intransitives,
> as gata (gone), m.rta (died) [rather dead, VKR] bhinna (split). Some again
> may even be transitive actives, as pi:ta (having drunk) [better drunk, but
> without the restriction to a special meaning in English] ...'' [para 360, p.
> 280].  Speyer goes on to note a:ru:d.ha has active meaning more commonly.
> This is not how we expect the passive to behave. But for resultative, it is
> understandable. The ergative patterning is based on pragmatics. But in a
> culture that considers it more worthy of note whether a man is mounted on a
> horse or a vehicle than whether a horse or a vehicle is carrying someone, it
> makes sense to use a:ru:d.ha in the active sense.

> The difference is unmistakable in the following:

> (1) ra:mo 's'vam a:ru:d.hah.
>     R is mounted on [a] horse.

> (2) ra:men.a+as'vam a:ruhyate
>    [A] horse is being mounted by R.

> (3) *ra:mo 's'vam a:ruhyate

> (1) and (2) are quite grammatical and examples of easy to come by. (3) does
> not occur and is underivable in traditional grammar. As I explained above,
> this is understandable if a:ru:d.ha is resultative. If we take the passive
> view, how do we explain the fact that (1) is acceptable but (3) is not?
> Without such an explanation, it is far from absurd to contest the passive
> interpreation.

Is this anything other than the accusative of goal? You are right that
some ta-participles are not passive, namely those derived from
intransitive verbs (gata- like Eng. gone). I would take (1) to be
construed as to mean "Rama (is = Eng.) has climbed onto a horse". (2) must
mean "By Rama there is being climbed onto a horse". I do not know if you
can construe this particular verb so transitively that you can make its
passive have the patient in the nominative; trying my hand with the sandhi
rules, I make it come out as "ra:men.a+as'va a:ruhyate".  But is this not
the normal construction with the finite passive? - I cannot see the
relevance of this for a discussion of the question whether the
ta-participle is passive or resultative: with transitive verbs it plainly
is both, we are not that much is disagreement. - By the Modern Indic rules
of agreement it does seem to me to be the passive (of transitive verbs, of
course) that formed the pattern. Where am I wrong?

Jens


From inaki.agirre at si.unirioja.es  Fri Jul 30 10:56:45 1999
From: inaki.agirre at si.unirioja.es (Inaki Agirre Perez)
Date: Fri, 30 Jul 1999 11:56:45 +0100
Subject: Ergative & Basque
Message-ID: <FRI.30.JUL.1999.115645.0100.>

About <have> and auxiliary verbs:

> This western usage is clearly calqued on the famous Castilian
> distinction between <ser> `be' (unmarked) and <estar> `be' (in a place
> or in a state).  Wherever Castilian uses <estar>, western Basque uses
> <egon>.  This is true even in idiosyncratic cases.  For example,
> Castilian expresses `He's dead' as <Esta' muerto>, with <estar>, and
> western Basque likewise has <Hil dago>, with <egon>, while eastern
> Basque has <Hila da>, with <izan>

AFAIK, in English 'He's dead' is both Spanish <Esta' muerto> and <Ha
muerto>. The translation in Basque would be <Hil da>, with <izan> verb.
A momentary sense could be achieved by <Hilik dago>, as if you have just
discovered the fact that he's dead, but <Hil dago> is ungrammatical in
my Basque (western).

Inaki


From jer at cphling.dk  Sat Jul 31 19:04:06 1999
From: jer at cphling.dk (Jens Elmegaard Rasmussen)
Date: Sat, 31 Jul 1999 21:04:06 +0200
Subject: Ergative & Basque
In-Reply-To: <Pine.GSO.4.03.9907191150230.19800-100000@rsunx.crn.cogs.susx.ac.uk>
Message-ID: <SAT.31.JUL.1999.210406.0200.>

Thanks to Larry Trask for taking the trouble to answer my questions about
the ergatve in Basque (in mail of Mon, 19 Jul 1999). The message being
quite clear and presumaby helpful to others than me alone, there are just
a few things that prompt a comment:

In a general sense, I understand that there is very little difference
between the ergative and the "accusative" structure. If you say "I have
built a house" with a verb of "having" taking the accusative, then the
construction is not ergative. But if "have" is expressed by "be to/for/at"
or "be" + genitive, and you phrase it as if it meant "there is to me a
built house", or "a built house is mine" or the like with the logical
patient as the subject put in the nominative, then it is ergative.
Strangely, the main idea, the expression of the relation between te agent
and the patient as one of possession ("having") is the same in the two
cases. This seems like a lot of fuss over very little. Now, where "have"
is expressed as "be" + an adverbial case (or the genitive, as in
Lithuanian vyro yra butas "of the man is a house" = 'the man has a
house'), there is a good and straightforward reason to put the
logical agent in a non-nominative case, a reason that never arises with
intransitive verbs which can at best expressed with an _essive_ relation
between the actor and the predicate: "John is laughing", "those days are
gone". This scenario then offers an opportunity for a direct understanding
of the distinction between the subject of an intransitive verb and that of
a transitive one, for one "is", the other "has" (or, is the one to/by etc.
which/whom something is); and it also makes it possible to understand
why the object has the same form as the subject of an intransitive verb,
for both "are". The use of a "have" _verb_ with ergative case as opposed
to a verb "be" with inergative case offers no such understanding,
therefore I find it unattractive if the ergative construction in
general is simply taken for granted; I could much better understand
things if the search is continued back to an origin of the opposition in a
difference between "is" and "there is to" (= "has"). That would make the
ergative just a roundabout use of the same general case categories as
used by the "accusative" syntax. That the search cannot be pushed that far
back for many languages does not change the possibility of such a
prehistory: you can draw no conclusions from inconclusive evidence.

I asked Larry Trask:

>> Can't the underlying construction be analyzed in a sensible way at
>> all?

LT asked back:

> Well, I'm not sure what you mean by "a sensible way".  What would you
> regard as a sensible analysis of English `I have drunk the wine'?

I see the point: How can you have something you have consumed - or
annihilated or lost, to take some even worse examples? You cannot. Then,
how can you justify expressing such events as if you have things you don't
really have? By analogy, by simple extension of the syntactic pattern from
an original core in which it made good sense to a general use where this
is not always the case.

LT also said:

> And this, I think, is as far back as we can go in tracing the prehistory
> of the Basque periphrastic verb-forms.  As I remarked earlier, we have
> no way of knowing whether these things originated as calques on Romance
> or whether they are independent creations in Basque, from an unknown
> source.  All I can tell you is that the modern periphrastic forms were
> clearly already established in Basque by the tenth century, when we find
> the first recorded verb-forms.

I'd say this disqualifies Basque as a language to help us solve the riddle
of the ergative, if there still is one.

Jens


From X99Lynx at aol.com  Thu Jul 29 18:20:26 1999
From: X99Lynx at aol.com (X99Lynx at aol.com)
Date: Thu, 29 Jul 1999 14:20:26 EDT
Subject: Passivity as a transition
Message-ID: <THU.29.JUL.1999.142026.EDT.>

Jens Elmegaard Rasmussen <jer at cphling.dk> wrote:

<<I'd say the cases I have seen of ergativity in Indic and Iranian languages
so clearly reflect underlying/earlier passive circumlocutions that
controversy is absurd.>>

On 7/29/99 2:52:29 AM, vidynath at math.ohio-state.edu responded:

<<No! I have already pointed out the problem of distinguishing resultatives
and passives. The ta-adjective is resultative in Vedic...

Speyer goes on to note a:ru:d.ha has active meaning more commonly.
This is not how we expect the passive to behave. But for resultative, it is
understandable.  The ergative patterning is based on pragmatics. But in a
culture that considers it more worthy of note whether a man is mounted on a
horse or a vehicle than whether a horse or a vehicle is carrying someone, it
makes sense to use a:ru:d.ha in the active sense.>>

Here on the "accusative and ergative languages" forum, I believe these
exchanges started with someone taking issue with the idea that 'passivity'
was not (necessarily) the origin of 'ergativity'.    Having read Larry
Trask's handling of that issue in his Historical Linguistics, it seems that
"passivity" as the ONLY origin is quite out of the question.

It's interesting to compare W.P. Lehmann's approach to ergativity in his
Historical Linguistics textbook.  The accusative versus ergative is only
brought up once - as background for the Contentive Typology distinction
between 'govermental' (accusative OR ergative) and
'active'/'active-stative'/'class' languages.  This background is given to
illustrate the use of Typological frameworks for historical purposes (e.g.,
postprepositions in Hittite, Indic and English and the 'pur'/'ignis'
(animate/inanimate) distinction) and evidence of an Active residue in modern
and historical languages.

While dismissing any further discussion of ergativity, Lehmann brings up an
interesting point.  He says, "While the method of indicating subjects and
objects differ in accusative and ergative languages, both are comparable in
distinguishing subjects and objects when pertinent, that is, with transitive
verbs... viewed as essential in communicating....  The subject of an
intransitive verb does not require a special form since it does not need to
be distinguished from an object."

This brings up the issue of what a truly "passive" language would be like,
putting aside animate versus inanimate.  I think the presumption that the
intransitive cannot take a passive form (of sorts) may not be correct when
words are in transition.  Verbs turn into nouns.  In Homer, the players are
sometimes 'moved' by their actions.  It is clear that in "the killing was
made by Hector" it is the action that is governing the structure, even though
there is a helper verb in there.  (Something of the same happens in saying
"Hector killed", as in "Speed kills" - it can be interpreted as a transitive
(kills someone) disguished as an intransitive.  But the sense definitely
changes and there is a reason to think of it as intransitive.  cf., "X rules")

Going back to the qoute above: <<in a culture that considers it more worthy
of note whether a man is mounted on a horse or a vehicle than whether a horse
or a vehicle is carrying someone, it makes sense to use a:ru:d.ha in the
active sense.>>

The third approach of course is to focus on the riding or carrying.

In Greek, we can see a related development that demonstartes verbs becoming
nouns.  "Konis"/"konia" (dust) in Homer is the basis of the idea of raising
dust - "koniontes pedioio", galloping in a cloud of dust (in the Illiad
always relating to horses, later to armies.)  "Euru konisouson pedion",
running away in dusty (hasty) flight.  But also in the Illiad, the jump is
made to "raising dust" as a metaphor for haste, effort or service.  The maids
make Priam's bed, "epei storesan lechos enkoneousai" (pres part.) > in haste.
(Always with another verb, so that it is NOT the action itself, but a true,
separate element.)

So "enkoneo" will later come to mean in haste.  "Akoniti" will mean after
Homer, without effort.  And "diakonia" will become a common name for a
servant or service in general - literally "through dust" but actually
"through raising dust" making haste or effort.

Shouldn't we conclude that before "raising dust" > effort> servant could
acquire the indications of a noun and an accusative subject, it had to go
through a "passive" stage.  E.g. "raising dust was done by Mabel."  ("Mabel
was raising dust" inflected as a participle, would mean something different -
it is part of the action and not yet an "object.")

Before there was something called a "rider" there was something called
"riding".  The action may have prempted the object and became an object in
itself.  "John was riding."> "Riding was what John did."> "John was a rider."

As a language develops, it is clear that verbs (processes, regular actions)
become nouns.  E.g., raising dust becomes service or a servant.   While that
is happening, isn't passivity also a natural thing to happen somewhere along
the line?  The absence of an established morphology/inflection for the new
noun means there appears to be no subject.  Couldn't to some degree that
account for the absence of a designated subject in some ergative structures?

Regards,
Steve Long


From rmccalli at sunmuw1.MUW.Edu  Fri Jul 30 15:10:59 1999
From: rmccalli at sunmuw1.MUW.Edu (Rick Mc Callister)
Date: Fri, 30 Jul 1999 10:10:59 -0500
Subject: Momentary-Durative
In-Reply-To: <002501bed2a4$94f66dc0$3970fe8c@lucent.com>
Message-ID: <FRI.30.JUL.1999.101059.0500.>

As a non-linguist, I'm gonna have to ask for back-up

but --if I remember correctly--
the present usage of ser and estar goes backs to
about 1400 to 1500
before that --at least in my memory of medieval literature--
the usage of ser and estar seem to be more unstable
or function more like the dichotomy of
essere and stare in Italian
where some of the uses of Spanish estar
are performed by Italian essere

But given that the dichotomies are pretty close
this probably goes back to Common Romance

French does not have this dichotomy
and I don't know what the situation is in Occitan, Sardinian, Rheto-Romance
and Rumanian

[ Moderator's note:
  The following is quoted from Vidhyanath Rao's message of 20 July 1999.
  --rma ]

>Rick Mc Callister <rmccalli at sunmuw1.MUW.Edu> wrote:

>> Spanish distinguishes them [resultative and passive]
>> by resorting to

>> estar --a "momentary" verb--
>> [actually a verb indicating condition] for resultant conditions--

>> and ser--a "durative" verb
>> [actually a verb indicating  characteristics]
>> for passive constructions.

>I don't know much Spanish, but I ought to have remembered this for I have
>been told about this before.

>I find something curious about your glosses. Resultatives, as indicating a
>state of indefinite duration, ought to be duratives, though examples of
>languages which use `go', `come', `finish' to form resultatives are
>mentioned in ``The evolution of grammar''. How old is the use of `estar'?.

Rick Mc Callister
W-1634
Mississippi University for Women
Columbus MS 39701


From jer at cphling.dk  Fri Jul 30 15:12:19 1999
From: jer at cphling.dk (Jens Elmegaard Rasmussen)
Date: Fri, 30 Jul 1999 17:12:19 +0200
Subject: Momentary-Durative
In-Reply-To: <000f01bed2e8$ba591b00$fde3abc3@xpoxkjlf>
Message-ID: <FRI.30.JUL.1999.171219.0200.>

Dear Peter and List,

(- careful, this may be about Indo-European -)

I think we're getting some place. Specific and insisting questions based
in an open mind are the only way. Many could learn from this - and that
may be just the point, as we shall see:

On Tue, 20 Jul 1999, petegray wrote:

[...]

> You said:
> [Jens:]
>> The impression that [in the IE verbal stem] ...
>> no common system is
>> recoverable is not compatible with current knowledge

> [Peter:]
> There is little real "Current knowledge" with PIE.   It is in many cases
> really rather "current opinion which someone happens to accept".   So how
> much support is there for Strunk?  How much debate is there?

[Jens:] There is a lot of unqualified debate on everything in IE studies,
even on points where the material speaks with a very clear voice. I'm not
sure any statements about IE could carry a majority vote. The reason is
obvious: IE studies is where most comparativists are, the field is vast,
and the amount of previous scholarship staggering. It is common to begin
your career by making mistakes simply because you cannot know all the
pertinent facts; many (most?) stay that way, contributing only to the
advancement of confusion. Therefore opinions should be checked very
carefully before they are adopted; one can only go by the quality of the
arguments put forward. Now, that _is_ what you are doing:

[Peter:]
>    I do question, however, how much weight we should give these
> patterns, and in particular whether we over-prioritise the patterns 
> based on Greek & Indo-Aryan.   It has been shown again and again since
> the 70's that the "south-eastern" group of Greek, I-A, and Armenian is
> highly innovative.

[Jens:] I'm not sure "shown" is the proper word, rather "maintained",
"guessed" or "wishfully thought" (if you can say that). It's one of those
numerous things about which two opinions are logically possible (_before_
you look at the specifics, mind you): Either Balto-Slavic is a unit or it
isn't, either the limited morphology of Hittite is a fossil or it isn't,
either *H2 colored *o or it didn't, etc. Not unpredictably, opinions tend
to sway from one possibility to the other all the time; if the market is
for one solution, it is arguing for the other one that is interesting.
Much debate in the field can be written off as acts of would-be scholars
searching for a place to stand with their names in neon signs. The
correspondences uniting Greek and Indo-Iranian are remarkable in a respect
that is constantly ignored by one "side", namely by their ability to turn
up in relics in all the other branches also. Now, innovations would not do
that, so that's not what they are. One is naturally reluctant to believe
that it is as simple as that and that so many researchers can be so
pitifully wrong, but only until one has looked at the decisive facts,
then one side is patently wrong and the other one much closer to the
truth.

> [Peter:]
> You said:

> [Jens:]
>> No, thematic presents and root-presents typically take the s-aorist (or
>> suppletion).

> [Peter:]
> Root presents are rare outside I-I, and in Sanskrit -s- aorists do not seem
> to be typical for root presents, since the both sigmatic and asigmatic
> aorists occur.   E.g. am has asigmatic, i:d has both, i:r shows both, u
> shows none, U:h asigmatic, and so on.   Where, again, is your evidence that
> makes it "typical" for a root present to take an -s- aorist?
>     Thematic presents appear to be a later formation, in any case, and you
> seem to suggest an origin for them in root aorist subjunctives.

[Jens:]
The two questions are related. First, thematic presents indeed appear to
be old subjunctives (more often aorist than present, but both occur). This
means that we can use the thematic present on a par with the root present.
Incidentally, I see no signs of being young about the thematic formation
as long as it is a subjunctive - quite the contrary: the thematic vowel
alternates in its own way, I don't see how that can be accounted for
unless by rules that have later ceased to operate. Second, if you take
Sanskrit from a Sanskrit grammar like Whitney's or a large dictionary like
Monier-Williams' you get an unsorted mixture of old and new forms, some so
young that they belong to a language that was nobody's native language
anymore. You must discount all the young forms of later periods, which is
best done by sticking to Vedic alone. Narten has of course discussed all
Vedic forms that look superficially like s-aorists, but since her work
does not contain a catalogue of conclusions (which are often not
decisive), it is hard to use for our purpose. On a more modest level, one
can look in Macdonell's Vedic Grammar under s-aorist ad is-aorist
(discounting with the latter forms without an extra vowel mora, since they
are just reinterpreted set root aorists), and then check in Macdonell's
Vedic Grammar for students to see what kind of present the root concern
forms. The result is overwhelmingly positive if the search is limited to
s-aorist forms that are not marked as belonging to post-RV texts):

I may have missed a few, but my somewhat hasty inspection reveals a root
present or a thematic present beside an s-aor. in the following 37 cases:

akra:n (krand) : kra'ndati
a'ks.a:r (ks.ar 'flow') : ks.a'rati
ks.es.at (ks.i 'dwell') : ks.e'ti
ga:si (1sgM ga: 'sing') : ga:'y-a-ti (!)
ga:ri:t (gr.: 'devour') : gira'ti
aca:ris.am (car") : ca'rati
acait (cit 'perceive') : ce'tati; cite'
ajais.am (ji) : ja'yati
ta:ri:t (tr.:) : ta'rati
atsa:r (tsar 'approach') : tsa'rati
adha:k (dah) : da'hati
a'diks.i (dis'), Av. da:is^, do:is^.i: - : Lat. di:co:, Goth. teihan
dhuks.ata (duh) : do'gdhi
adyaut (dyut) : dyo'tate
nes.ati (nais.t.a, ni:) : na'yati
anu:s.i (nu:s.ata, nu:) : na'vati
a'bhutsi (budh) : b?dhati
a'bhaks.i (bhaj) : bh?jati
abha:rs.am (bhr.) : bha'rati
amatsur (mad) : ma'dati
aya:s (3sg aya:t., yaj) : y'ajati
aya:sam (ya: 'go') : ya:'ti
a'rabdha (rabh) : ra'bhate
a'ram.sta (ram) : ra'mate
ava:t (vas) : va'ste
ava:t. (vah) : va'hati
a'viks.ata (vis') : vis'a'ti
avr.tsata (vr.t) : va'rtati
asaks.ata (sac) : sa'cate (also root aor. with red.prs.)
asakta (saj) : s'ajati
asa:ks.i (sah) : sa'hate
asa:vi:t (su: 'impel') : suva'ti
asra:k (sr.j) : sr.ja'ti
astos.i (stu) : sta'uti
asya:n (syand) : sya'ndate
a'sva:r (svar) : sva'rati
a'ha:rs.am (hr.s.) : ha'rs.ate
(Forms with zero-grade + -ks.a- may be special innovations; that would
reduce the list by two.)

In 13 examples, commonplace renewal by the productive s-aorist can be
assumed. Root aorists are attested for the following for which, then, it
is no problem that they do not form root presents:

ata:n (tan): tano'ti, old aor. a'tan
da:si:t (das 'waste'): old aor. dasat (Narten)
naks.at (nas'): old root aor. a:'naks.am (prs. doubtful: length, red., y?)
apra:s (pra: 'fill'), old root aor. apra:t: n-prs. pr.n.a:'ti
muks.ata (muc), old root aor. a'mok : n-prs. mun'c'ati
a'yuks.a:ta:m (yuj), old root aor. yo'jam : n-prs. yuna'kti
a'ra:sata (ra:), also root aor. and red.prs.
a:'raik (ric), old root aor. riktha:'s, n-prs. rin.a'kti
avitsi (vid 'find') secondary: old root aor. a'vidat : n-prs. vinda'ti
asa:nis.am (san"), old root aor. a'sanat : n-prs. sano'ti
a'spa:rs.am (spr.), old root aor. a'spar : n-prs. spr.n.vate'
ahes.ata (hi), old root aor. a'hyan : n-prs. hino'ti
ahu:s.ata (hu:) : ha'vate (also root aor. + red.prs.)

A few combine s-aor. with sk-prs.:

a'pra:ks.am (pras') : pr.ccha'ti
yos.ati (yu) : yu'cchati (also root aor. : red.prs.)
a'ya:m.sam (yam) : ya'cchati
ava:t (vas 'shine') : uccha'ti

The residue is very small, perhaps as small as two examples:

mam.si (man): unclear relation to ma'nyate; manve' (with root aor.)
alipsata (lip) : n-prs. limpa'ti !!
aha:s (ha:) : red.pl. ja'ha:ti, ji'hi:te !!

So it's about two or three examples out of 57 that won't play ball! I do
not think that looks like random distribution?

> [Peter:]
> Incidentally, I read today in JIES vol12, 1984, the argument that even
> the -s- aorists are an innovation within Greek, I-A and closely related
> groups.   The article alleged that evidence outside this area is weak, and
> tried to dispose of Latin -s- perfects by suggesting they were either
> limited to verbs ending in a velar, or they were back formations from the
> supine.

[Jens:]
It's a popular view on the matter, and one of the favorite arguments for a
step-by-step model of development within the protolanguage, if I
understand it correctly (the whole idea is alien to my sense of logic).
For the s-aorist it is patently wrong. I have myself (in 1978) called
attention to the existence of Hitt. ganeszi 'recognizes' whose past
ganest(a) from *g^ne:H3-s-t is manifestly the s-aorist one expects beside
the sk-present (Lat. cogno:sco:, Gk. gigno:sko:, Lith. paz^iNstu, OPers.
xs^na:sa:tiy, Alb. njoh), and Jasanoff did the same independently (1987),
adding other branches (esp. the magnificent Armenian prs. c^anac^'em, aor.
caneay, the latter form from a stem *g^n.e:H3-s- with vocalization of the
/n/ pointing to an old monosyllable), and we presumably agree on the
prehistory of Germanic *kne:-(j)i/a- 'know' as based on the s-aorist, 3sg
*g^ne:H3-s-t, with simple replacement of the word-final cluster by
productive endings. The main point is here that the lengthened grade goes
back to a time when laryngeals had not yet colored adjacent e-vowels, for
the lengthened /e:/ turns up as /e:/ in the daughter languages, not as
/o:/, this proving the formation so old that there can be no talk of its
being a post-PIE innovation, no matter what weaknesses one may read into
specific parts of the material.

I hope this has been instructive.

Jens


From jer at cphling.dk  Sat Jul 31 01:11:52 1999
From: jer at cphling.dk (Jens Elmegaard Rasmussen)
Date: Sat, 31 Jul 1999 03:11:52 +0200
Subject: Momentary-Durative
In-Reply-To: <002401bed767$67c01f00$2e70fe8c@lucent.com>
Message-ID: <SAT.31.JUL.1999.031152.0200.>

On Mon, 26 Jul 1999, Vidhyanath Rao wrote:

>> [Jens:]
>> If my observation that there is an alliance between the sk^-present type
>> and the s-aorist is correct ... then the s-aorist was originally
>> inchoative in function. [...]

> [Nath:]
> This makes it harder for me to understand how the aorist became the
> perfective. `Started driving', in contrast to `drove', suggests incomplete
> action.

The aorist reports a turn of event that caused a new situation: is that
not the perfect thing to express a beginning?

[Nath:]
>>> This brings me to a general question. There seem to be two camps about
>>> the category system of the PIE verb. One believes that the
>>> aorist-imperfect distinction, to be equated to perfective-imperfective
>>> distinction, ``always'' existed in PIE and Hittie lost this distinction,
>>> while Vedic changed things around. The other considers the aspectual
>>> distinction to postdate the separation of Anatolian.

[Jens:]
>> There are these two camps, yes, and I am in no doubt that camp one is
>> right. There is no way the specific forms of the aspect stems could have
>> been formed secondarily in "the rest of IE" left after the exodus of (or
>> from) the Anatolians. At the very least, all the _forms_ must be assigned
>> to a protolanguage from which also Anatolian is descended. [...]

[Nath:]
> The stem formants might have been derivational in nature. In particular,
> they might change the meaning, valence and/or aktionsart. It is the
> assumption that they changed only the aspect and nothing else that is in
> question. Your view, expressed elsewhere, that different present stems might
> be because there are different kinds of durativity is halfway to this.

[Jens:] I am rather convinced this is all derivation in origin.

[..]
[Nath:]
> Grammatization of derivational affixes is found often enough that its
> occurance in Pre-IE -> PIE or PIE -> dialects cannot be rejected out of
> hand. As it seems that very different languages underwent similar evolution
> of grammatical categories (see Bybee et al, ``The evolution of grammar), we
> cannot assume the similar functions must go back to the proto-language. That
> might have had a precursor function.

[Jens:]
The PIE function of the different derivative categories must be at least
compatible with that of their later reflexes, and the simplest solution is
that wherever we find non-trivial correspondences between the daughter
languages we have a relatively direct reflexion of the protolanguage. Why
would a morpheme with lengthening and -s- turn up as the expression of
non-durative past in Italo-Celtic, Slavic, Armenian, Greek and Tocharian
if that were not its function in PIE already?

[Nath:]
>>> There are certain nagging questions about the first thesis [i.e. the
     thesis of Graeco-Aryan archaism and massive losses in Hittite - JER]:
>>> The change
>>> in Vedic is not explained and how it came about without the prior
>>> loss of aspect has, AFAIK, not been explained. Those who adhere
>>> to this also feel the need to explain away as much as possible of
>>> root presents. But there are enough of them remain in Hittite and
>>> Vedic to raise doubts.

>> [Jens:] I don't follow - what change in Vedic are you talking about?

> [Nath:]
> In Vedic, the so-called imperfect is the tense of narration. Aorist has a
> recent past meaning. In living languages with aspect, the perfective is the
> tense of narration, unless historical present is being used. Languages with
> remoteness distinction, either it applies only in the perfective or without
> regard to aspect. It is this that needs to be explained.
[...]
> Hoffman's answer is to posit an intermediate stage in which aspect was
> limited to non-recent past. He does not give any contemporary examples, nor
> does he explain how the aorist, which in such a stage must have been even
> more common, was lost in reference to remote past.

> The point is that an alternate explanation is possible: Completives have a
> ``hot news'' value, which makes it plausible to see them develop into recent
> past. They also can develop into perfectives, Slavic being a usable example.
> There are gaps in the examples, but I find this more plausible than
> deriving the Vedic usage out of perfecitve-imperfective opposition.

[Jens:] But Vedic _is_ an IE language.

[Jens, earlier:]
>> Why would anyone want to explain away root presents where
>> they are securely reconstructible? One would do that only to avoid
>> having a language combining a root-present with a root-aorist, for in that
>> case the two aspect stems are identical. That is why I am so sceptical
>> about the authenticity of the Vedic root presents lehmi and dehmi,
>> because for these verbs we have nasal presents in some other IE
>> languages pointing to the existence of a root aorist; thus leh- deh- look
>> like displaced aorists. But not so for eti 'goes' or asti 'is': these are
>> durative verbs, and so their unmarked form could function as a
>> durative (socalled "present") stem.

[Nath:]
> This argument is valid only if you have an independent reason to assume
> obligatory perfective-imperfective contrast for PIE. But firstly, we see
> stems transferred from one class to another within a single group. This is
> most obvious in Indic, where we have a long recorded history. Secondly,
> there are too many root presents that remain: In Vedic, amiti (injures),
> da:ti (cuts, divides), yauti (joins, unites) and of course, hanti confirmed
> by Hittite kuen-. None of these is eligible for a nasal present (expecpt
> perhaps amiti, if you believe in nasal presents for roots of the shape CeNH,
> with N standing for a nasal). Especially *g'henti is unavoidable, unless you
> are going to argue that the two oldest recorded dialects innovated in
> precisely the same way. Such an argument needs more compelling evidence
> than a just so story.

> If you add potentialy telic verbs, we get some more: ta:s.t.i (taks,
> fashion), ya:ti etc. These can and are used in the so-called imperfect with
> definite objects without any indication of non-completion. etc. How can they
> be called imperfetive?

[Jens:] With reference to PIE they must, in case such was the system - and
that it was is very well established.

>[...]
[Jens:]
>> Again, I do not think there is any problem in accepting root present as
>> original for inherently durative verbs, and root aorist as equally
>> original for inherently punctual verbs.

[Nath:]
> What about telic verbs? These are inherently durative, but with a determined
> object, their preterite (in languages without perfective-imperfective
> distinction) carries the implicature of completion, while their imperfective
> past would do the opposite. How, without recorded narratives, can you decide
> how a given language operated?

[Jens:]
Telic verbs are typically aoristic: *{gw}em- 'come' forms a root aorist,
*H1ey- 'walk' forms a root present.

[Jens earlier:]
>> But the IE aorist is not restricted to any special kind of verbs -
>> it is only _unmarked_ (better: apparently originally unmarked) for
>> inherently punctual verbs; for other verbs the aorist needs a
>> morphological marking, and the meaning is then some nuance that can be
>> regarded as punctual ("started to -") or it just reports that the
>> action got done (Meillet's > action pure et simple).

[Nath:]
> Could the imperfect report `action pure et simple', or was that reserved for
> the aorist? How did PIE speakers report a durative action that was done,
> like ``I walked home''? How did they say ``I made pots yesterday''?

[Jens:]
I guess they walked home in the aorist, made pots (generically) in the
ipf., but made a specific pot or set of pots in the aorist. But who am I
to know? Please don't demand that I write a fable.

[Jens:]
>> One important functional point with the aorist,
>> however, is that it marks a turn of events which creates a new situation,
>> whereas the "present aspect" stays in the situation already given and
>> reports another action contributing to that situation.

[Nath:]
> How do you explain that it is the so-called imperfect that is the tense
> of narration in Vedic?

[Jens:]
Sorry to take so long in coming to the point, this _is_ a very important
question which deserves more attention than is perhaps mostly accorded to
it. Now, under no circumstances can we completely divorce the imperfect
from the present, for they are formed from the same stem - that must mean
_something_. And I do not think we can disregard the situation- changing
effect of the aorist stem which turns up in all corners of IE. I repeat:

>> This is seen remarkably well in the prohibitive use of the prs. vs.
>> aor. injunctive, as propounded so clearly by Hoffmann.

And I maintain this against your comment:

> Hoffmann's claim requires morphological gymnastics, such as taking i:'s'ata
> as an aorist and still has some holes (eg, jivi:t).

I find no problem with the existence of individual verbs that act in
individual ways that have to be entered in the lexicon. That sound fairly
normal.

So, if the present stem is situation-preserving, and the aorist stem
situation-changing, how do we explain the Vedic facts? They do not look so
odd to me: The present is also a narrative form, namely to report what is
going on: "The horse is turning at the corner, it's emerging in full
sunlight, and is now approaching the finish line" - this would all be in
the present indicative in Vedic I guess. If the corresponding past
narrative is the imperfect, that could simply be due to the status of this
category as the past of the present stem. Other IE languages rather
clearly point to the one-time existence of a step-by-step turn-of-events
use of the aorist, opposed there to the background actions of longer
duration reported by the imperfect, but in Vedic the aorist has taken a
turn in the direction of exaggerating the situation-changing value of the
stem, while the imperfect has been generalized to all past narrative, be
it momentary or background-like. In Vedic the aorist is almost constantly
rendered by translators by means of the "have perfect" which states a
conclusion. The consensus on this matter is remarkable, especially since
there is little possibility of checking what is really meant in the text
of the Rigveda. The least one can say is that the use of tenses in
translations of the Rigveda does not strike one as unnatural (it's often
the only thing one does find natural). If I were to combine the Vedic
picture with that of the other IE branches, this would be my guess. It
appears to mean a lot to you that the IE aspect opposition is not supposed
to have been obliterated before the vedic development can take place -
well, in my account it isn't. And then there is nothing alarming (or even
interesting) about the number of preserved root presents.

[Jens earlier:]
>> Still, even verbs generally signifying completed action could form
>> duratives, indicating e.g. a repetition of the action (give one thing,
>> and then another) or an as yet unsuccessful attempt (I'm opening the
>> window).

[Nath:]
> How do you classify ``I learned that chapter in one month?''

[Jens:] I believe as a job for the aorist.

[Jens earlier:]
>> We know that this kind of change was small enough in the languages here
>> concerned to lead to a number of misplaced aspect stems. E.g., the
>> Armenian aor. eber is an old ipf. Is the jump from "narrative past" to
>> "recent past" so great? If it is, even great changes happen.

[Nath:]
> But if the PIE ``imperfect'' was really just a preterite, what we see in
> Armenian eber is old preterite becoming an aorist due to the rise of an
> imperfective past. That is attested. For example, Kui, a Dravidian language,
> has generalized an old progressive into an imperfective and this limited the
> preexisting past into an aorist.

[Jens:]
Not due to the creation of the ipf. for that was already there (eber was
one itself). I do not argue about Kui, I only say the Armenian (and IE)
story cannot be that way. We know too much be able to accept it.

> It is not the size of the change, but the direction and manner of the
> change, that must be credible. All I am asking for is a single
> uncontrovertable evidence of this change. It won't do to say, as I was once
> told by an Indologist, that this change is possible because it must have
> happened in Vedic. You need to give an example where the perfective past
> category can be established from >preserved texts<.

If instead of perfective past you read concluding past which is just a
natural further development, you find it supported several times in
practically every hymn of the whole Rigveda. And the distinction between
perfective (aorist) and imperfective past (imperfect) is a salient shared
feature of all other IE languages than Germanic, Indo-Iranian and
Anatolian in the system of synthetic verbal forms. There is no way that
can be an innovation. Specifically, it is not therefore not possible that
the isolated instances of misplaced aspect forms reflect an archaic state
of affairs where the whole aspect business had not even started.

[Nath:]
> What I find difficult to swollow with the argument for aspect in PIE is that
> the association ``imperfect'' = imperfective is limited to Greek. The two
> oldest recorded dialects, Hittite and Vedic, do not work that way. We are
> simply supposed to believe that they innovated, but what do we get in
> return? What can you explain this way that you cannot explain otherwise?

[Jens:]
The IE imperfect is not just Greek. The Slavic imperfect, which mostly
translates the Gk. ipf. in OCS, is an almost direct continuation of the IE
ipf. (in Baltic it has become a preterite pure and simple due to the loss
of the aorist). The Armenian ipf. has endings in -i- from *-e:- stemming
from the verb 'be' (all thematic verbs rhyme with 'be' in Arm.) which
formed *e:st from *e-H1es-t with the augment. The Toch. ipf. is basically
the optative, but there are some long-vowel imperfects which in my view
simply copy the old relation *es-/*e:s- of 'be'. Old Irish no-bered 'was
carrying' is from the middle-voice ipf. *bhereto, notably always
compounded (if only by the default preverb no-) and so rather obviously
continuing an augmented form. No matter what one thinks of Lat. ama:bam it
does contain the same span as Oscan fufans and so adds the same preterital
marker to the present stem as the latter had added to the perfect stem;
and ama:ver-a:-s has preteritalized the perfect stem just as er-a:-s has
the present stem, so here, too, the ipf. is the preterite of the present
stem. Even Albanian ish or ishte (the C,amian forms) may artlessly reflect
*est (in part with productive superimposed ending, probably borrowed from
the aor. qe), i.e. the present stem with secondary ending. Only Germanic
and Anatolian have obliterated the imperfect as an independent category,
using what's left of it simply as past tense (in Gmc., e.g. Eng. did is an
old ipf.). I cannot accept the statement that the imperfectivity of the
"imperfect" is restricted to Greek: the imperfect is opposed to a
non-durative preterite wherever it occurs, except for Indo-Iranian. That
I-Ir. and Anatol. have both innovated, should cause no concern, especially
since the two innovations have nothing in common.

> [Jens, quoted from a different post]
>> The most rewarding experience during the time I have been
>> watching Indo-European Studies has been to see the protolanguage
>> come alive and assume an increasingly well-established structure, [...]

[Nath:]
> Do we really understand the variety of syntactic structures and their
> diachrony that well? I have mentioned the Tamil -vidu construction a few
> times. You will find some linguists call that a perfective and the Tamil
> simple past an imperfective. This is simply wrong as the simple past is and
> has been the tense of narration for the 2000+ year recorded history, and
> this distinction is nothing like the perfective-imperfective distintion in
> Russian or Arabic [...].

[Jens:]
Maybe we have something as important as the explanation of the Indic
development here. Perhaps Anatolian and Iranian have been influenced by
some common source which could not distinguish different types of
synthetic preterites?

[Nath:]
> It can be even worse: Similar constructions exist in some NIA languages
> which also have a prefective-imperfective distinction that is different
> (imperfective out of old progressive). I have seen at least one linguist
> apply the label of perfective to the construction with auxillary, and
> imperfective to the one without the auxillary. This completely misrepresents
> the syntax.

> What reason is there to think that we are not making the same kind of
> mistake with PIE? I am not asking for certainity here, but evidence for
> greater probability than the opposite.

[Jens:]
As long as our mistakes are comparable to calling the expression of
imperfective action "imperfect" and the expresiion of perfective action
"perfect" in a language that really has a perfective-imperfective
distinction, I see little cause for alarm.

Thank you for your patience and most stimulating questions and
observations.

Jens


From proto-language at email.msn.com  Fri Jul 30 04:35:12 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Thu, 29 Jul 1999 23:35:12 -0500
Subject: Semantic change
Message-ID: <THU.29.JUL.1999.233512.0500.>

Dear Lloyd and IEists:

 ----- Original Message -----
From: <ECOLING at aol.com>
Sent: Monday, July 19, 1999 10:08 AM

> A less plausible semantic relation means that two
> look-alikes are less plausible as cognates.
> But requiring "identity" of meanings or even
> "near-identity" of meanings is an absurd
> requirement also, when we are working at great
> time depths.

> So what to do?
> Asserting simplistic extremes either for or against
> is not particularly useful.

Pat responds:

I agree completely. It is the subjectivity of the permissible limits that
militates against a consensus. It is the frustration of this situation that
has prompted well-intentioned researchers to insist on identity, which, as
you correctly identify, is also problematical. I sincerely wish there was
some mechanical or mathematical way to set these limits --- if only as I
check on some of my own wilder flights of fancy.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From 114064.1241 at compuserve.com  Fri Jul 30 10:51:00 1999
From: 114064.1241 at compuserve.com (Damien Perrotin)
Date: Fri, 30 Jul 1999 06:51:00 -0400
Subject: Hittite & Celtic dative in /k/ ?
Message-ID: <FRI.30.JUL.1999.065100.0400.>

[ moderator re-formatted ]

Initial message from
Lloyd Anderson

[ moderator snip ]

>Here is a tidbit linking Celtic and Hittite which I found many years ago, when
>compiling typological comparisons of the semantic domains of "be" and "have".
>I wondered at the time whether it is a relic Dative case preposition /
>postposition which Hittite and Celtic shared, presumably as a retention, but
>conceivably as an areal phenomenon, or both of the above.  Does Tocharian have
>it too?

>(I have no idea whether the later Indic Dative postposition with /k/ is
>related or a chance lookalike.)

>The forms being discussed, the dative case of the 1st singular pronoun, are
>statistically likely to be highly conservative, both because they are pronouns
>and also as oblique cases rather than the more often innovating nominative
>etc.

[ moderator snip ]

the comparison is unlikely, as the Hittite form is also used for the
accusative, an use for which it has cognates in Germanic (gotic mik) and
perhaps in Slavic (russian ko - to with only an allative meaning). The
primary meaning was probably allative, as for most accusative in IE.
the Goidelic form is assumed to derive from *angh "near" which is found in
Latin angustus. Its closest cognates, assuming to Stokes, is Breton hag
(and) and Welsh ac (same meaning)


From 114064.1241 at compuserve.com  Fri Jul 30 11:01:23 1999
From: 114064.1241 at compuserve.com (Damien Perrotin)
Date: Fri, 30 Jul 1999 07:01:23 -0400
Subject: -n- adjectival suffix in Latin
Message-ID: <FRI.30.JUL.1999.070123.0400.>

[ moderator re-formatted ]

Paolo Agostini<
wrote

>I was wondering about the origins of the Latin adjectival
>suffix -n-us, -n-a, -n-um appearing in words like _mater-nus_ "maternal;
>motherly; belonging/pertaining to the mother", _pater-nus_ "paternal;
>fatherly; belonging/pertaining to the father", _feli-nus_ "feline;
>belonging/pertaining to the cat; belonging/pertaining to the family/genus
>of the cats/felidae" and the like.  Can it be traced back to IE? Does the
>morpheme exist in other IE or non-IE languages? Any idea abt its etymology
>and/or development?

This ending is first found in Etruscan, where it has the same use as in
Latin, but as the affiliation of Etruscan is far from being sure, it is not
of great utility.
A similar ending is used in Russian to form past passive participles
(-nny). A suffix -ina- is also used in Germanic to form adjectives (Gotic
quineins : feminine)

Damien Erwan Perrotin


From kurisuto at unagi.cis.upenn.edu  Fri Jul 30 14:23:36 1999
From: kurisuto at unagi.cis.upenn.edu (Sean Crist)
Date: Fri, 30 Jul 1999 10:23:36 -0400
Subject: Principled Comparative Method
In-Reply-To: <ed0acce.24c49979@aol.com>
Message-ID: <FRI.30.JUL.1999.102336.0400.>

On Mon, 19 Jul 1999 ECOLING at aol.com wrote:

> So if we had case (d)
> and we also had case (d'),
> one example of IE /t/ corresponding to Uralic /s/,
> we might have more evidence for saying that /d/~/z/ was real.

> Or if we had case (d) before a following /u/,
> and we also had case (d'')
> ONE example of IE /d/ corresponding to Uralic /z^/ before /i/,
> (interpret the ^ as a hachek = wedge, upside down)
> we would again have more evidence for saying that /d/~/z/ was real,
> in this case because we would assume that /z/ became /z^/ before /i/.

> And if lucky we might have case (d') above with following /u/,
> and also case (d'''),
> one example of IE /t/ corresponding to Uralic /s^/ before /i/.

> With all four of these examples, we would have a plausible system,
> and would in fact have four examples reflecting the same sound changes,
> perhaps actually only a single sound change (namely
> "affricate apical stops before high vowels, with the result being
> the fricative of the same voicing and "reflecting" the vowel quality)
> but we need not have more than one example of any particular correspondence
> on the surface.

	As a matter of terminology, the kind of argumentation you propose
here falls outside the Comparative Method.  It's true that we do sometimes
fall back on this kind of external guesswork, but it's the sort of special
pleading you make to bolster some assumption which you need for your
argument.

	In any case, I'd be really careful in assuming that phonological
rules always apply across entire classes of categories.  It's true that
things often do work this way, but they don't always.
	Around 12 years ago, I was interviewing a speaker of the South
Midlands dialect area of American English.  In this dialect, /I/ > /i/ for
some speakers, and /U/ > /u/ for some speakers.  I assumed that if the
speaker had one rule, she'd have the other as well; I figured that the
general rule was "high lax vowels become tense".  To my great surprise,
she had the second rule but not the first.  There was no denying what I
was hearing; my assumptions about the expected symmetry within the system
were just plain wrong.
	I've fallen into this same kind of trap plenty of other times.
There are so many cases of beautiful symmetry and parallelism in
phonological systems that it's an ongoing challenge to remember that
things don't always work out so neatly.  Even in the Japanese case that
you give, there are recent loan words (e.g. tiishatsu "T-shirt") to mess
up the nice symmetry of the system.

> One of them I am quite certain is to develop both articulatory and
> acoustic "spaces", relative "distances" between different articulations
> and different acoustic effects, so that when attempting to judge
> likelihood of cognacy of pairs of words, we can judge similarity
> by degrees, not by yes/no dichotomies.  (These will partly depend
> on the general typology of the sound systems of the languages
> concerned, they will not be completely universal, but they will
> also not be completely idiosyncratic.)

The Comparative Method is concerned with the reconstruction of categories,
_not_ the phonetic values which might have been the realization of those
categories.  This is a very important point.  When we talk about
Proto-Indo-European */a/, we don't mean "the phonological category in
Proto-Indo-European which had the phonetic realization [a]"; we mean "the
hypothetical PIE category which gave rise to the Sanskrit category /a/,
the Latin category /a/, etc."

As far as the Comparative Method is concerned, we could designate that PIE
category with an integer, e.g. "Category 27".  */a/ is just a convenient
label or nickname for it.  Along with that label comes a built-in guess
about what the prehistoric phonetic value for that category might have
been, but this is _just_ a guess; the actual phonetic value is beyond the
reach of the Comparative Method.

To look at it another way, imagine that we have two languages, A and B.
Suppose that in every single case where A has dental [d], B has alveolar
[d], and vice versa.  From the standpoint of the Comparative Method, A and
B are the same language, because there have been no mergers in categories.
Strictly speaking, the phonetic realization of the categories is of no
consequence to the Comparative Method.

It is true that we do sometimes point out that if the phonetic values for
such-and-such categories were such-and-such, then such-and-such a rule is
phonetically plausible.  But this is guesswork on top of guesswork, and
it's really shaky ground to build an argument on.

  \/ __ __    _\_     --Sean Crist  (kurisuto at unagi.cis.upenn.edu)
 ---  |  |    \ /     http://www.ling.upenn.edu/~kurisuto/
  _| ,| ,|   -----
  _| ,| ,|    [_]
   |  |  |    [_]


From proto-language at email.msn.com  Fri Jul 30 14:59:15 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Fri, 30 Jul 1999 09:59:15 -0500
Subject: Latin perfects and Fluent Etruscan in 30 days!
Message-ID: <FRI.30.JUL.1999.095915.0500.>

Dear Damien and IEists:

 ----- Original Message -----
From: Damien Erwan Perrotin <114064.1241 at Compuserve.com>
Sent: Wednesday, July 28, 1999 2:14 PM

[ moderator snip ]

> [Damien Erwan Perrotin]

> The idea is interesting, but probably inexact as am- is not present only
> in Latin but also in Lydian (ama : to love) and in Breton (afan : to
> kiss from an older Brythonnic *ama). It is still possible that the
> Lydian form was a borrowing from an Etruscan-like tongue of the Aegean,
> but that is quite unlikely for Breton. So there can be only three
> explainations for the ressemblance you point out :
> a: chance ressemblance (always possible)
> b: Etruscan borrowed the word from Latin or from Celtic
> c: Etruscan is remotely linked to IE and this root is a remnant of this
> old relationship.

> Personnally, I favor the third thesis, but there is still work to do
> before proving it.

Pat comments:

Some might be interested in Egyptian jm-3, 'kind, gentle, well-disposed,
pleasining, gracious, be delighted, charmed', which I believe is likely to
represent an AA example of the same root.

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)


From JoatSimeon at aol.com  Fri Jul 30 20:44:32 1999
From: JoatSimeon at aol.com (JoatSimeon at aol.com)
Date: Fri, 30 Jul 1999 16:44:32 EDT
Subject: Lexical Retention
Message-ID: <FRI.30.JUL.1999.164432.EDT.>

>Patrick C. Ryan

>In any case, you have it rather backwards. If the root and all manifestions
>of it (like, perhaps, in 'bear') were absent in Germanic, what would, as it
>has, make possible our reconstructing an IE root would be its presence in,
>at least, three (nominally) other branches.

-- quite true, if it _exists_  in at least three branches.  Some roots,
obviously, do.  Some, equally obviously, don't -- we don't have a complete
vocabulary of PIE, to put it mildly.  And in the nature of things we never
can.

Some parts of PIE have been completely and irrevocably lost because they
didn't survive into extant or recorded languages.

Obviously, this portion will be larger the further back we go in time and the
less we have in the way of writtten records of forms no longer extant -- a
reconstructed proto-Romance, even if we had no records of Latin, would be
much more complete than PIE.

The greater the temporal distance, the more information lost to entropy.


From JoatSimeon at aol.com  Fri Jul 30 20:53:35 1999
From: JoatSimeon at aol.com (JoatSimeon at aol.com)
Date: Fri, 30 Jul 1999 16:53:35 EDT
Subject: Hittites ~ Phrygians ~ Balkan peoples?
Message-ID: <FRI.30.JUL.1999.165335.EDT.>

ECOLING at aol.com

>There has been much discussion of the Hittites and the Phrygians, and other
>names for these or other peoples of Anatolia and the Balkans (Thracians,
>Illyrians, Dacians, Moesians, etc. etc.),  which may have been incorrectly
>taken as distinct peoples in the past, instead of simply as different names
>used for the same peoples at different times or by different authors or
>cultures.

-- Hittite and Phrygian are distinct.  They're both IE, of course, but
Phrygian is not a member of the Anatolian subgroup of IE.  Like Armenian,
it's intrusive in Anatolia and is post-Hittite.


From proto-language at email.msn.com  Fri Jul 30 23:52:56 1999
From: proto-language at email.msn.com (Patrick C. Ryan)
Date: Fri, 30 Jul 1999 18:52:56 -0500
Subject: `cognate'
Message-ID: <FRI.30.JUL.1999.185256.0500.>

Dear Adam and IEists:

 ----- Original Message -----
From: Adam Hyllested <adahyl at cphling.dk>
Sent: Wednesday, July 21, 1999 11:44 AM

Adam wrote:

> It is more or less the same reconstruction. <@> is (originally, at
> least) the vocalic allophone of <H>, reconstructed on the basis of
> Indo-Iranian, where the root vowel is <i> (an IE <**pater> would have
> developed into Sanskrit <**pata:> and not the actual <pita:>)
> Using <@2> or <H2> is simply a matter of phoneme-to-grapheme
> correspondence.

Pat comments:

A few of you may be interested in reading the short essay:

http://www.geocities.com/Athens/Forum/2803/comparison-AFRASIAN-3_schwa.htm

Pat

PATRICK C. RYAN (501) 227-9947; FAX/DATA (501)312-9947 9115 W. 34th St.
Little Rock, AR 72204-4441 USA WEBPAGES:
http://www.geocities.com/Athens/Forum/2803 and PROTO-RELIGION:
http://www.geocities.com/Athens/Forum/2803/proto-religion/indexR.html "Veit
ek, at ek hekk, vindga meipi, nftr allar nmu, geiri undapr . . . a ~eim
meipi er mangi veit hvers hann af rstum renn." (Havamal 138)