Basque <ibili>

[roslyn frank]

[ moderator re-formatted ]

At 12:30 AM 2/6/00 -0600, Rick Mc Callister wrote:

[snip]
>  	you may wish to consider the following notes (keeping in mind that
> any errors in copying are my own)
> 	see Basque ekarri "to bring" Basque ekarri; < Pre-Basque *e-kaR-i
> [lt/B]
> PN263 *kar- "to twist, to turn, to wind" [b/k], see IE *(s)ker- "to jump,
> to move in circles"? [p/IE]
> maybe "to turn > to return s.t. > to bring s.t." [mcv]
> 	Celtic carru, carricare, see Basque ekarri "bring, carry"; [wje]
> 	carry English and Basque ekarri [rmcc]
> Eng. carry is of French-Romance origin: Latin carrus, carruca = cart,
> carriage. My Latin dictionary says both are of Gallic origin. see Fr.
> charrier. Original meaning: transport by cart. [es]
> 	carpentum "two-wheeled wagon" Gaulish > Spanish carpintero
> "carpenter" [abi 4], French charpentier "carpenter" [wde 188-89]; < ? carru
> [rmcc]
> 	carru Gaulish "cart" > Spanish carro "cart" [abi 4]; char artisan
> term French; from Gaulish [cb62: 13]
> 	carro "car, cart" Spanish/Portuguese; from Celtic [jng]; carrum
> Celtic > French char "cart" [mh 241]
> 	carruca Gaulish > French charrue [wje 188-89]; charrue agricultural
> word French; from Gaulish [cb62: 13]
> 	carrum Romance < Celtic [wje 183]
> 	carrus "cart" Latin; from Celtic [nv 75-76]
> 	carrus Italian < Celtic [bm66: 25]
> 	carrus "4-wheeled covered wagon" Latin; from Celtic [lrp 53]; carro
> Spanish; from Celtic [rks 12-13]

Rick, I'm glad to see that you brought up the case of <ekarri>. It's
certainly one that has puzzled investigators for some time. In arguing the
case for <ekarri>, i.e., how it fits into all of this, one might want to
point out what I mentioned earlier. If one looks in Buck (10.75), it turns
out that this root is far more wide-spread than any other single
root-stem/etynom  for 'wheel' -at least that's my impression from reading
what others have said on the list. In the case of Buck

In the case of the Basque item, using prototype semantics one would say
that its prototypic meaning is 'to bring by means of a wheeled vehicle' but
rather merely 'to bring', i.e., 'traer' Sp. or 'tirer, apporter' Fr.(cf.
Azkue I, 229). Hence, a compound such as <ekarrera> (with the common
suffixing element <-era> to added to the verbal stem) refers to '(the)
bringing, transporting' without explicitly stating the mode of transport.
If one were apply the work that has been done in diachronic prototype
semantics to this case, one could argue that a meaning such as 'to bring or
transport by means of a wheeled vehicle' could be derived from the broader
meaning of 'to bring, to transport (by any means)'. Given that the
invention of the wheel is a technological innovation (albeit not a terribly
recent one) the second definition, i.e., one that restricts the mode of
transport to that of wheeled vehicle, could only have come into existence
after the wheel itself was invented. Hence in this case a (hypothetical)
development from a generalized notion of 'to bring' or 'to transport' to a
narrowly defined one would also imply the following: that at some point in
that semasiological process the term came to be associated with a
particular mode of transport and that subsequently that meaning became the
dominant one, the prototypic one. Again I emphasize that there is no
evidence that in Euskera the word <ekarri> was ever used to refer
specifically to a mode of transport by a wheeled vehicle.

This in turn suggests that if one is going to relate the Celtic and Basque
items -if one is going to argue that they are cognates and they do give
every appearance of being so- there is a caveat. While it seems to me that
it would be relatively easy for the meaning to narrow itself and become
associated with a common type of transport, it would be more difficult for
a word that whose prototypic meaning was originally 'to bring' or 'to
transport by a wheeled vehicle' to end up meaning 'to bring' with no
connotation whatsoever of any wheeled vehicle which is the case in Basque.
But then there is the additional problem that the English word 'to carry'
which is traced back to French doesn't mean 'to bring by means of a wheeled
vehicle' but merely 'to carry' although I believe that a case could be made
that 'to carry' might refer to a particular 'way of bringing', but not
necessarily by means of a 'cart'. Someone with a better etymological
dictionary of French would have to judge whether originally terms like
<carriere> were restricted in this way semantically. Also, I'm fully aware
that the standard interpretation would derive the Romance items from the
Latin word for cart <carrus>, not from a substrate.

Now, it's clear that the Latin form was influential, i.e., in introducing a
term for a new technological marvel, but there seems to be a case that
could be made that we are looking at items that might be better categorized
as Basque/Pre-Celtic/Celtic (without trying to figure out exactly who was
first in line) data set that was picked up by Romance languages (including
Latin where it came to be associated with a particular object). Stated
differently, one explanation for its presence in these languages would be
that the diffusion of data set in question could be traced back in two
stages: 1) the first would situate the lexical feature as part of a
preexisting Basque/Pre-Celtic/Celtic substrate and 2) the second would
reflect the influence of the prototype meaning of the Latin lexical item
<carrus> and its diffusion throughout the Roman Empire as the name of a
particular kind of cart (chariot??). Hence in this simulation of linguistic
events, there would be at least two chronological layers, an earlier and
later one, that could be uncovered and two relatively different mechanisms
for the diffusion of members of the data set in question.

I don't know if the time-depth for Celtic/Gaullic would allow for that type
of pattern of areal diffusion for this root, but it might. Also, I've
noticed on the list several people mentioning that there are substantial
substrate lexical (?) items in Germanic languages. Wouldn't it be possible
that Celtic and Basque could share elements from a substrate that was found
in this zone earlier, a substrate about which we know little and whose of
geographical extension is unclear at this point?

Also, as I mentioned in an earlier email, since <-bil-> is packaged inside
<ibili> (a Class 1 verb), and the verbal prefix and suffixing elements date
back in all likelihood to Pre-Basque (or maybe even before), it is more
difficult to argue that Basque just lifted the Celtic word and turned it
into <ekarri>. Or if one argues that it happened this way, then we are
talking about a significant time depth for the contact period.

I've lost track of who said what in the exchange below so please forgive me
for not citing who is who:

Rick Mc Callister <rmccalli at sunmuw1.MUW.Edu wrote:

[snip]
>>>  What about being 'bil' a loan from Gaulish?
>>>      PIE *kwel-  > Cel *kwi:l-   > Gaul *pi:l  > ! >  bask. bil
>>>   cf PIE *penque > Cel *kwinkwe  > Gaul *pimpetos (ordinal)

>>1) Regarding the semantics:
>>    Remember that the Celts were famous for their cartwright-technique.

[snip]
>>   Of course the i-prefixed verbform is an argument /against/ borrowing from
>>   Celtic.

[rmcc]
>	Maybe, maybe not. The prefix could have been added after borrowing
>--but ask a specialist in Basque, don't take my word on it

So in response to your comment about the 'prefix'. This prefix shows no
sign of life whatsoever in Basque, whereas there are other elements, i.e.,
the <-eta> suffix now used to mark the plural of oblique cases (and for
other things) that is partly non-productive in one subroutine of the
system, but alive and well in other subroutines. This is not the case, to
my knowledge, of the verbal prefix <i->. And I think that Larry will agree
with me on this one.

...At 04:24 PM 2/2/00 +0000, Larry Trask wrote:
>Roz Frank writes:

[snip]

[RF]
>>  And as an aside, are there explict criteria set forth that determine
>>  which items are most representative. I'm speaking of crtieria along the
>>  lines of those that have been suggested by Larry Trask (and debated by
>>  many) concerning the selection of items in Pre-Basque.

[LT]
>But the two cases are very different.  We have lots of IE languages, and so
>our main tool is the comparative method here.  But Basque is isolated, and
>so the comparative method is of minimal use, and only internal reconstruction
>is available.  Hence different criteria are appropriate in the two cases.

[RF]
>>  I would be most
>>  interested in knowing if such criteria have been debated and/or laid out
>>  explicitly at some point in the past. For example, how many language groups
>>  must the item be attested in for it to quality? I assume, for example, that
>>  identifying cognates/reflexes of the same item in Sanskrit and Celtic would
>>  be sufficient for the item to qualify? Or is the bar set higher for these
>>  PIE items, e.g., that the item must be attested in Sanskrit, Germanic and
>>  Celtic or Hittite, Slavic and Romance, etc.

[LT]
>There is no unchallengeable answer to this question.  By Meillet's Principle,
>we require cognates in at least three branches of IE before we can reconstruct
>an etymon for PIE.  But this is only a rule of thumb, and skilful specialists
>need not adhere to it slavishly.

[RF]
>>  For example, just glancing over the entries in Buck, it would seem that
>>  there isn't as much uniformity for "wheel" across IE languages, as there is
>>  for, say, "cart" which shows up most IE languages (obviously with the help
>>  of Latin).

[LF]
>But loan words don't count for the purpose of reconstruction.  If an
>identifiable Celtic word is borrowed into Latin, from where it descends into
>the Romance languages and is borrowed into Germanic and elsewhere, it is still
>only the Celtic word which counts.

But if it could be shown that the Celtic word is cognate with a Basque word
and that it is unlikely that the Basque word could have been borrowed from
Celtic, e.g., if the Basque root-stem were to be encountered only as a
verbal root-stem in a Class I non-finite verb (e.g., such as <ekarri>),
wouldn't it follow that the Celtic word itself may represent not a
borrowing from Basque but rather from an earlier substrate that also was
the source of the Basque and Celtic data sets?

And on that note. Earlier we were discussing the semantic structure of the
non-finite verb <ibili>. I said that I believed that because of its
structure, it would qualify for admission to candidacy for Pre-Basque.
Would you agree that it is extremely difficult to date a word like <ibili>
or <ekarri> given that its structure reveals the presence of
morphosyntactic structures that are no longer productive in the language.
Let me try to explain what I'm getting at.

One of the topics that intrigues me is that of developing techniques for
discovering and/or elaborating criteria for judging "morphosyntactic
stratification" in the case of the Basque data. And I fully recognize that
I'm the neophyte on this list and, therefore, what I'm going to say may be
a bit like trying to reinvent the 'wheel' :). However, in the case of
Basque it is extremely important for one to be sensitive to the way in
which certain morphosyntactic data are situated within the overall system.
For instance, we have been speaking of the problem of developing criteria
that would allow us to determine the age of verbs such as <ibili> or
<ekarri>. Again keeping in mind that we are speaking of simulations, i.e.,
of the ways that we can go about modeling the data available, we could ask
what aspects of the morphosyntactic structure of Basque can give us clues
about the age of the linguistic artifact, the data set under study. The way
I see it is a bit like what happens in archaeology where researchers are
sensitive to 'context' in which an artifact is discovered, for example, to
the 'layer' in which it is encountered.

Another way of looking at the problem is to say that when one digs up an
artifact and discovers it wrapped in a given type of material, it is
assumed that the object inside is at least as old as the wrapping it is
found in. In the case of <ibili> and <ekarri> we are dealing with a
particular type of artifact, one that belongs to a class of similar
artifacts, i.e., non-finite Class I verbs equipped with a prefixing element
in *<e-> (which has a phonological variant in <i->) as well as a final
verbal suffix in <-i>.

As has been pointed out (Trask 1995) neither of these morphemes is
productive today in the language. Therefore, they appear to incorporate
morphosyntactic structures that may hearken back to a much earlier stage of
the linguistic system. This is because there is no evidence in the language
of even the slightest sign of life with respect to the verbal suffix <i->
nor that of the verbal prefix in *<e-> . This situation might be contrasted
with that of the suffxing element <-eta> where one finds it alive and
kicking in certain subroutines of the system but relatively moribund in
others, as I've mentioned above.

Moreover in the case of <ibili>, <ekarri> and other Class I non-finite
verbs, there is another factor that needs to be taken into consideration,
although what weight it should be given when modeling the time depth of the
artifact is not clear. I refer to the fact that Basque is a suffixing
language. There is no trace of prefixing in the language except in the case
of Class I verbs (I'm excluding a few lexical calques that have entered the
language relatively recently and which are obviously based on Romance
formations). Hence, the root-stem <-bil-> in <ibili> or that of <-karr-> in
<ekarri> is encountered wrapped up in material that has every sign of
belonging to the most archaic strata that can be detected in the
morphosyntactic structure of the linguistic system of Euskera. We are
talking about typological issues where the artifact's morphosyntactic
packaging provides the researcher with a certain type of information that
in turn permits a tentative assignment of the artifact to a given layer, to
a given morphosyntactic strata: the artifact ends up being situated at a
certain level because of the way that the morphosyntactic data. lends
itself to typological stratification.

Again I emphasize that all of the above should be considered a highly
tentative attempt to develop criteria that could be utilized in examining
and dating artifacts such as <ibili> and <ekarri>. As I mentioned in an
earlier email, if an attempt is made to relate the verbal stem of <ibili>
to artifacts found in IE languages, the time-depth that can be assigned to
the Basque artifact by means of the above argumentation, should serve as a
cautionary sign. This doesn't mean that I would reject such attempts to see
in the Basque data reflexes of the IE data. Rather, for such a proposal to
flourish, the investigator need to consider whether the purported
similarities between the Basque and IE might be best explained by alleging
that they result from an even earlier Sprachbund or other type of areal
phenomena that allowed reflexes of the same element to show up in two
different linguistic systems.

Hopefully the above discussion will shed some light on a few of the
difficulties involved in constructing an argument in which the verbal stems
found in Basque items such as <ibili> and <ekarri> would be viewed as
recent loans from an IE language(s).

As so often happens when I start a short response, it gets out of hand.
Thanks for your patience, that is, it you've reached this part of my
long-winded message :)).

Ondo ibili,
Roz Frank