From ALDERSON at mathom.xkl.com  Sun Aug  5 07:15:00 2001
From: ALDERSON at mathom.xkl.com (Rich Alderson)
Date: Sun, 5 Aug 2001 00:15:00 -0700
Subject: [ADMIN] Network problems
Message-ID: <SUN.5.AUG.2001.001500.0700.>

Dear Readers,

My apologies for the backlog of messages on the Indo-European list, which could
have the effect of chilling the discussion.  The host site, for which I am the
managing systems and networking adminstrator, has been disconnected from the
Internet since early Wednesday morning; all of my time has been spent in
getting services restored to normal.

I believe that we are once again stable, and so I can devote some of my time
once again to providing the Indo-European list.

								Rich Alderson
								IE list owner
								and moderator


From acnasvers at hotmail.com  Fri Aug  3 04:16:45 2001
From: acnasvers at hotmail.com (Douglas G Kilday)
Date: Fri, 3 Aug 2001 04:16:45 +0000
Subject: Return of the minimal pairs
Message-ID: <FRI.3.AUG.2001.041645.0000.>

Jim Rader (16 Jul 2001) wrote:

>Isn't <britches> for <breeches> standard English as the name for a
>garment?  A pronunciation with [i] in "too big for his breeches"
>would sound non-native to me....

I've never heard [i] in that expression. I suspect that <breeches> became
obsolete in non-rustic dialects of American English, and that "too big for
his britches" was borrowed from a rustic dialect in which [I] was the normal
stem-vowel. The Scottish form is given as <breeks>, and the word is a double
plural (from OE <bre:c>, pl. of <bro:c> 'leg-covering') in which one would
expect [i] in standard ModE, had there been no inter-dialect borrowing.

DGK


From acnasvers at hotmail.com  Fri Aug  3 09:10:08 2001
From: acnasvers at hotmail.com (Douglas G Kilday)
Date: Fri, 3 Aug 2001 09:10:08 +0000
Subject: Latin mecum, tecum, etc.
Message-ID: <FRI.3.AUG.2001.091008.0000.>

Lionel Bonnetier (5 Jul 2001) wrote:

>But the trickiest question was: Why did "cum" remain
>a postposition while all other adpositions became
>prepositions in Latin? (And why only with pronouns?)

Well, this is a real head-scratcher. All I can suggest is that <mecum>,
<tecum>, and <secum> were broadly similar in phonetic shape to other words
used adverbially in similar contexts, and continued to sound right despite
the general shift of adpositions to prepositions. I am thinking particularly
of accusative supines used with verbs of motion: "vadunt sessum" 'they go to
sit'; "vadunt mecum" 'they go with me'. Parallels like *mepro and *meab were
not close enough phonetically to adverbs which might fill similar syntactic
slots. If this is right, <nobiscum> and <vobiscum> were preserved by analogy
with the shorter forms. Neither analogy nor common phonetic shape could
preserve expressions like *"exercitu cum", however. We can of course say
"magno cum exercitu", emphasizing the adjective, but the noun must follow
<cum>.

>Thanks for this survey. The same question comes here
>with Attic "peri": Why did it keep the possibility of
>being a postposition while no other adpositions did?

I don't have a clue, since there is no phonetic argument: <amphi> and <anti>
can't follow their objects, even in poetry. It should be noted that
postpositional <peri> in Attic prose is somewhat archaizing. When Thucydides
says "seismon peri" he refers to the time of an earthquake which few of his
readers could have remembered, but most of them as children had likely heard
the graybeards talking about it. Thucydides achieves a similar effect with
the obsolescent dual number, so it is likely that postpositional <peri> was
nearly extinct in normal speech by his time.

DGK


From edsel at glo.be  Mon Aug  6 15:03:10 2001
From: edsel at glo.be (Eduard Selleslagh)
Date: Mon, 6 Aug 2001 11:03:10 -0400
Subject: bishop
In-Reply-To: <3B57E120.D18E78EE@swissonline.ch>
Message-ID: <MON.6.AUG.2001.110310.0400.>

At 15:43 20/07/01 +0800, you wrote:
>Leo Connoly wrote:

>> To which I say: The vowels are still all wrong, which renders the rest
>> of it moot.

>> Leo Connolly

>1. Not exactly true: the I- of Istambul corresponds perfectly to the i- in
>'is tin poli'.

>2. Cf. anc.-gr. propolis 'bee-glue' > mid-gr. ke'ropoli > turk. dial.
>girebullu:  The vowels are all wrong, but the word is Greek, believe it or
>not.

>R. Piva

[Ed Selleslagh]

I thought I showed (in an earlier contribution) that it is not just a
matter of vowels, but also of stressed syllables/prosody:

A derivation of [istim'bolin] would place the stress on the <o> (shifted to
[u]), whule it is on the <a>.The initial <i> is prothetic.

Even if you insist on the vowels, your argument, invoking 'girebullu', is
still weak, actually irrelevant, because it only alludes to the common
shift of e > i and o> u, which also means that the vowels are NOT 'wrong'
(the -ro- > -re- is an effect of the tendency in Turkish to make words of
not readily recognizable foreign origin conform to vocal harmony rules, i.e
with the first syllable). But THE vowel problem in 'Istambul' is the <a>.
It cannot possibly be derived fron 'is tin bolin', only from
'Konstantinopolis' where the <a> has a secondary stress.

Ed.


From acnasvers at hotmail.com  Fri Aug  3 07:50:42 2001
From: acnasvers at hotmail.com (Douglas G Kilday)
Date: Fri, 3 Aug 2001 07:50:42 +0000
Subject: Genetic Descent
Message-ID: <FRI.3.AUG.2001.075042.0000.>

>[DGK]

>> I'm not sure we need to invoke substrate for the feminine suffix -issa.  It
>> can plausibly be regarded as an extraction from feminine ethnonyms like
>> <Phoi'nissa> and <Ki'lissa> in which the final [-k-] of the ethnic stem was
>> palatalized by the feminine suffix [-ya].

[Peter Gray]

>There should be evidence of timing available here.   The change -ky- to -ss-
>occurs after Attic speakers are out of immediate touch with Ionic speakers,
>and in immediate touch with Boeotia (since Attic shares the Boeotian reflex
>as -tt-, not -ss- as Ionic). That's relatively late.  So are there feminine
>words in -issa attested significantly earlier than this, such as in
>Mycenaean?

[DGK]

I'm not aware of any. As for the timing, the -issa of Late Latin (< ModE
-ess) comes from Hellenistic Greek, in which the Attico-Boeotian fortition
of -ss- to -tt- is usually not followed. I've seen -issa referred to
Macedonian, which I don't think is necessary. The Hellenistic treatment of
-ss- (and the rough breathing, usually retained when resulting from *swV-,
but not *sV-) probably reflects some kind of consensual phonology in which
peculiarly Attic features, used by a minority of all Greek-speakers, were
not generally adopted.

Xenophon uses fortited velar-stem verbs (<ke:ru'tto:>, <pra'tto:>,
<ta'tto:>, etc.), <tha~tton>, <peritto's>, and <tha'latta>, but he has
<Ki'lissa> 'Cilician woman'. This suggests on its face that the change -ky-
 > -ss- occurred a second time in Attic (since, for example, earlier
<ke:ru'sso:> represents *ke:ruk-yo:). However, Xenophon does not fortite the
-ss- in foreign names (<Issoi'>, <Kolossai'>, <Tissaphe'rne:s>), and he may
have felt that the same rule applied to <Ki'lissa>. If this reflects general
Attic practice, it seems possible that the ethnonyms in -issa escaped the
fortition, being used only for foreign women (I can't think of any Greek
tribes in -kes). At any rate, the extraction of -issa as a general feminine
suffix is post-classical.


From X99Lynx at aol.com  Wed Aug  1 03:53:21 2001
From: X99Lynx at aol.com (X99Lynx at aol.com)
Date: Tue, 31 Jul 2001 23:53:21 EDT
Subject: Sound changes versus sound changes
Message-ID: <TUE.31.JUL.2001.235321.EDT.>

In a message dated 7/20/2001 4:48:10 PM, larryt at cogs.susx.ac.uk writes:

<<...what languages "sound like" is not a linguistic fact of any interest.>>

However, despite the statement above, Prof Trask goes on to prove something
with this fact that is not of any interest:

<<Spanish and French are rather closely related, but they sound very little
like each other.  The Spanish of northern Spain sounds very much like
modern Greek -- a speaker who knows neither can easily confuse these two --
yet Spanish and Greek are only very distantly related.  And Spanish sounds
quite a lot like Basque,... [etc.]>>

The problem here is that Prof Trask seems to be talking about accidental or
non-genetic "sound alikes" which wasn't what the prior posts were about.

My post was about related sound alikes.  (e.g., "How many sound changes
brought us from <*wixti> to /viisi/ and how long did they take?")

So, of course, Prof Trask's examples are irrelevant.

Just to be clear about this.  The French spoken by one French speaker "sounds
like" the French spoken by another French speaker.  They are using the same
or very similar sounds.  This is the most basic situation where words sound
alike.

Presumably, there was a time when the language that would become French more
or less "sounded like" Latin - the presumed ancestor of French.   Presumably,
the sound changes that occurred - between the time when Latin was spoken and
the time the daughter language French was spoken - caused those two languages
to "sound less alike."

Prof Trask also wrote:
<<But P-N languages have still undergone a very significant degree of change.
 Neighboring P-N languages are typically not mutually comprehensible at all.>>

And I'd presume that neither grammar or syntax was the main reason for that
incomprehensibility between previously mutually comprehensible languages or
dialects.  I'd presume that the main reason was a change in sound systems.
Am I incorrect?

Prof Trask also wrote:
<<Similarity in sound systems is not a good metric for degree of divergence.>>

But the sound systems of related languages are precisely how languages
diverge, isn't it?  Don't systematic correspondences show the diverging
changes in what were once unitary sounds in the ancestor?  It would seem that
the difference in sound that separates two related languages is precisely the
degree of divergence.

After all *p > f sounds more divergent than *p > p.  And p sounds less like f
than it sounds like p.  That's a pretty clear degree of divergence.  Isn't it?

Regards,
Steve Long


From ALDERSON at mathom.xkl.com  Sun Aug  5 07:15:00 2001
From: ALDERSON at mathom.xkl.com (Rich Alderson)
Date: Sun, 5 Aug 2001 00:15:00 -0700
Subject: [ADMIN] Network problems
Message-ID: <SUN.5.AUG.2001.001500.0700.>

Dear Readers,

My apologies for the backlog of messages on the Indo-European list, which could
have the effect of chilling the discussion.  The host site, for which I am the
managing systems and networking adminstrator, has been disconnected from the
Internet since early Wednesday morning; all of my time has been spent in
getting services restored to normal.

I believe that we are once again stable, and so I can devote some of my time
once again to providing the Indo-European list.

								Rich Alderson
								IE list owner
								and moderator


From acnasvers at hotmail.com  Fri Aug  3 04:16:45 2001
From: acnasvers at hotmail.com (Douglas G Kilday)
Date: Fri, 3 Aug 2001 04:16:45 +0000
Subject: Return of the minimal pairs
Message-ID: <FRI.3.AUG.2001.041645.0000.>

Jim Rader (16 Jul 2001) wrote:

>Isn't <britches> for <breeches> standard English as the name for a
>garment?  A pronunciation with [i] in "too big for his breeches"
>would sound non-native to me....

I've never heard [i] in that expression. I suspect that <breeches> became
obsolete in non-rustic dialects of American English, and that "too big for
his britches" was borrowed from a rustic dialect in which [I] was the normal
stem-vowel. The Scottish form is given as <breeks>, and the word is a double
plural (from OE <bre:c>, pl. of <bro:c> 'leg-covering') in which one would
expect [i] in standard ModE, had there been no inter-dialect borrowing.

DGK


From acnasvers at hotmail.com  Fri Aug  3 09:10:08 2001
From: acnasvers at hotmail.com (Douglas G Kilday)
Date: Fri, 3 Aug 2001 09:10:08 +0000
Subject: Latin mecum, tecum, etc.
Message-ID: <FRI.3.AUG.2001.091008.0000.>

Lionel Bonnetier (5 Jul 2001) wrote:

>But the trickiest question was: Why did "cum" remain
>a postposition while all other adpositions became
>prepositions in Latin? (And why only with pronouns?)

Well, this is a real head-scratcher. All I can suggest is that <mecum>,
<tecum>, and <secum> were broadly similar in phonetic shape to other words
used adverbially in similar contexts, and continued to sound right despite
the general shift of adpositions to prepositions. I am thinking particularly
of accusative supines used with verbs of motion: "vadunt sessum" 'they go to
sit'; "vadunt mecum" 'they go with me'. Parallels like *mepro and *meab were
not close enough phonetically to adverbs which might fill similar syntactic
slots. If this is right, <nobiscum> and <vobiscum> were preserved by analogy
with the shorter forms. Neither analogy nor common phonetic shape could
preserve expressions like *"exercitu cum", however. We can of course say
"magno cum exercitu", emphasizing the adjective, but the noun must follow
<cum>.

>Thanks for this survey. The same question comes here
>with Attic "peri": Why did it keep the possibility of
>being a postposition while no other adpositions did?

I don't have a clue, since there is no phonetic argument: <amphi> and <anti>
can't follow their objects, even in poetry. It should be noted that
postpositional <peri> in Attic prose is somewhat archaizing. When Thucydides
says "seismon peri" he refers to the time of an earthquake which few of his
readers could have remembered, but most of them as children had likely heard
the graybeards talking about it. Thucydides achieves a similar effect with
the obsolescent dual number, so it is likely that postpositional <peri> was
nearly extinct in normal speech by his time.

DGK


From edsel at glo.be  Mon Aug  6 15:03:10 2001
From: edsel at glo.be (Eduard Selleslagh)
Date: Mon, 6 Aug 2001 11:03:10 -0400
Subject: bishop
In-Reply-To: <3B57E120.D18E78EE@swissonline.ch>
Message-ID: <MON.6.AUG.2001.110310.0400.>

At 15:43 20/07/01 +0800, you wrote:
>Leo Connoly wrote:

>> To which I say: The vowels are still all wrong, which renders the rest
>> of it moot.

>> Leo Connolly

>1. Not exactly true: the I- of Istambul corresponds perfectly to the i- in
>'is tin poli'.

>2. Cf. anc.-gr. propolis 'bee-glue' > mid-gr. ke'ropoli > turk. dial.
>girebullu:  The vowels are all wrong, but the word is Greek, believe it or
>not.

>R. Piva

[Ed Selleslagh]

I thought I showed (in an earlier contribution) that it is not just a
matter of vowels, but also of stressed syllables/prosody:

A derivation of [istim'bolin] would place the stress on the <o> (shifted to
[u]), whule it is on the <a>.The initial <i> is prothetic.

Even if you insist on the vowels, your argument, invoking 'girebullu', is
still weak, actually irrelevant, because it only alludes to the common
shift of e > i and o> u, which also means that the vowels are NOT 'wrong'
(the -ro- > -re- is an effect of the tendency in Turkish to make words of
not readily recognizable foreign origin conform to vocal harmony rules, i.e
with the first syllable). But THE vowel problem in 'Istambul' is the <a>.
It cannot possibly be derived fron 'is tin bolin', only from
'Konstantinopolis' where the <a> has a secondary stress.

Ed.


From acnasvers at hotmail.com  Fri Aug  3 07:50:42 2001
From: acnasvers at hotmail.com (Douglas G Kilday)
Date: Fri, 3 Aug 2001 07:50:42 +0000
Subject: Genetic Descent
Message-ID: <FRI.3.AUG.2001.075042.0000.>

>[DGK]

>> I'm not sure we need to invoke substrate for the feminine suffix -issa.  It
>> can plausibly be regarded as an extraction from feminine ethnonyms like
>> <Phoi'nissa> and <Ki'lissa> in which the final [-k-] of the ethnic stem was
>> palatalized by the feminine suffix [-ya].

[Peter Gray]

>There should be evidence of timing available here.   The change -ky- to -ss-
>occurs after Attic speakers are out of immediate touch with Ionic speakers,
>and in immediate touch with Boeotia (since Attic shares the Boeotian reflex
>as -tt-, not -ss- as Ionic). That's relatively late.  So are there feminine
>words in -issa attested significantly earlier than this, such as in
>Mycenaean?

[DGK]

I'm not aware of any. As for the timing, the -issa of Late Latin (< ModE
-ess) comes from Hellenistic Greek, in which the Attico-Boeotian fortition
of -ss- to -tt- is usually not followed. I've seen -issa referred to
Macedonian, which I don't think is necessary. The Hellenistic treatment of
-ss- (and the rough breathing, usually retained when resulting from *swV-,
but not *sV-) probably reflects some kind of consensual phonology in which
peculiarly Attic features, used by a minority of all Greek-speakers, were
not generally adopted.

Xenophon uses fortited velar-stem verbs (<ke:ru'tto:>, <pra'tto:>,
<ta'tto:>, etc.), <tha~tton>, <peritto's>, and <tha'latta>, but he has
<Ki'lissa> 'Cilician woman'. This suggests on its face that the change -ky-
 > -ss- occurred a second time in Attic (since, for example, earlier
<ke:ru'sso:> represents *ke:ruk-yo:). However, Xenophon does not fortite the
-ss- in foreign names (<Issoi'>, <Kolossai'>, <Tissaphe'rne:s>), and he may
have felt that the same rule applied to <Ki'lissa>. If this reflects general
Attic practice, it seems possible that the ethnonyms in -issa escaped the
fortition, being used only for foreign women (I can't think of any Greek
tribes in -kes). At any rate, the extraction of -issa as a general feminine
suffix is post-classical.


From X99Lynx at aol.com  Wed Aug  1 03:53:21 2001
From: X99Lynx at aol.com (X99Lynx at aol.com)
Date: Tue, 31 Jul 2001 23:53:21 EDT
Subject: Sound changes versus sound changes
Message-ID: <TUE.31.JUL.2001.235321.EDT.>

In a message dated 7/20/2001 4:48:10 PM, larryt at cogs.susx.ac.uk writes:

<<...what languages "sound like" is not a linguistic fact of any interest.>>

However, despite the statement above, Prof Trask goes on to prove something
with this fact that is not of any interest:

<<Spanish and French are rather closely related, but they sound very little
like each other.  The Spanish of northern Spain sounds very much like
modern Greek -- a speaker who knows neither can easily confuse these two --
yet Spanish and Greek are only very distantly related.  And Spanish sounds
quite a lot like Basque,... [etc.]>>

The problem here is that Prof Trask seems to be talking about accidental or
non-genetic "sound alikes" which wasn't what the prior posts were about.

My post was about related sound alikes.  (e.g., "How many sound changes
brought us from <*wixti> to /viisi/ and how long did they take?")

So, of course, Prof Trask's examples are irrelevant.

Just to be clear about this.  The French spoken by one French speaker "sounds
like" the French spoken by another French speaker.  They are using the same
or very similar sounds.  This is the most basic situation where words sound
alike.

Presumably, there was a time when the language that would become French more
or less "sounded like" Latin - the presumed ancestor of French.   Presumably,
the sound changes that occurred - between the time when Latin was spoken and
the time the daughter language French was spoken - caused those two languages
to "sound less alike."

Prof Trask also wrote:
<<But P-N languages have still undergone a very significant degree of change.
 Neighboring P-N languages are typically not mutually comprehensible at all.>>

And I'd presume that neither grammar or syntax was the main reason for that
incomprehensibility between previously mutually comprehensible languages or
dialects.  I'd presume that the main reason was a change in sound systems.
Am I incorrect?

Prof Trask also wrote:
<<Similarity in sound systems is not a good metric for degree of divergence.>>

But the sound systems of related languages are precisely how languages
diverge, isn't it?  Don't systematic correspondences show the diverging
changes in what were once unitary sounds in the ancestor?  It would seem that
the difference in sound that separates two related languages is precisely the
degree of divergence.

After all *p > f sounds more divergent than *p > p.  And p sounds less like f
than it sounds like p.  That's a pretty clear degree of divergence.  Isn't it?

Regards,
Steve Long