LINGUIST List 5.1489

Tue 20 Dec 1994

Disc: Comparative method

Editor for this issue: <>

Directory

1. "Valiquette, Philippe Luc", Comparative Syntax
2. Celso Alvarez Caccamo, Re: Two languages / One grammar?
3. ALICE FABER, Comparative Method
4. , Re: 5.1484 Comparative Syntax

On Mon, 12 Dec 94 21:04:31 EST (amrares.cs.wayne.edu) wrote:

)Subject: Comparative Syntax

)While I really like most of what Scott DeLancey had to say about
)syntactic reconstruction usually being based on clues buried in
)the morphology (or morphophonology), I don't think this is always
)the case. There is a rather famous example involving a rule of
)Ancient Greek and one variety of Old Iranian (the languages of
)the Gatha's, I seem to recall), whereby a neuter pl. subject
)triggers sg. agreement on a verb, a pattern which is often
)reconstructed for the proto-language because, as I understand it,
)of its apparent oddity. This reconstruction is not logically
)dependent, I don't believe, on the identity of the actual morphemes
)marking gender, number, and person in these languages.
)
)I would think that there are many such quirks of syntax which
)could be the basis of a reconstruction.

The phenomenon mentioned for Ancient Greek -- that can apply as well to
Latin -- doesn't appear to be a (<quirk of syntax)>. Rather than being
_apparently odd_ and motivated by a _rule_ of invariable agreement, it
should be regarded as a (<quirk of meaning)>. While neuter pl. subjects,
still showing in Ancient Greek and Latin evidence of an ancient collective
case, have *usually* triggered sg. agreement on a verb, numerous examples
show that this pseudo rule wasn't always observed, and that semantic
considerations, most of the time -- moreover, metrical reasons for poets
--, have governed the agreement (syllepsis).

Thus I do not believe (<quirks of syntax [at least this one in particular]
could be the basis of a reconstruction)>.

La plupart sont d'accord, n'est-ce pas ?(not literally: What about
French?)

Regards,
Philippe L. Valiquette
Universite Laval, Dep. Linguistique
(PHLCVALIVM1.ULAVAL.CA)
(PHILIPPE.VALIQUETTELLI.ULAVAL.CA)

Regarding John Cowan's (lojbabaccess.digex.net) message
(LINGUIST 5-1460. Sat 17 Dec 1994), about John Gumperz's
work on language convergence:

a) Two languages can't have the same grammar. If they do,
they are the same language. The two lexicons would be
considered sets of cooccurrent lexical variants.

b) If the two languages of (a) are spoken in different areas,
you may call them two (societal) languages, but they still
are just one language.

b) Two structurally unrelated languages may get to share a
great deal of grammar due to contact (Gumperz's case, also
in "Convergence & creolization: a case from the Indo-Aryan/
Dravidian border"). Gumperz's point is that only a morpho-
phonological features & lexicon work as markers of social
identity. Gumperz's (and my) point is, if a given population
of speakers use/understand these two varieties natively and
habitually, they are for all practical purposes the same language.

Celso Alvarez-Caccamo
Linguistica Geral e Teoria da Literatura, Univ. da Corunha, Galiza, Spain
lxalvarzudc.es

Prompted by his overview of attempts to establish a genetic relationship
between Basque and *any* other language, Larry Trask asks:

)> Is it possible to do ANY useful comparative work on languages you know
)> nothing about?

)> In other words, can you establish anything about the genetic
)> affiliations of a language merely by extracting data from secondary
)> sources, without yourself having any kind of specialist knowledge of
)> that language?

I may be reading too much into Larry's comments, but it appears that these are
(at least semi-) rhetorical questions, to which the answer self-evidently
*should be* no, from which it follows that long-range comparison *must* be
impossible, since how could any one investigator know enough about a large
number of not obviously similar languages to guard against falling into the
traps that Larry lists (e.g., treating obvious borrowings as native
vocabulary).

In this context it is, I believe, worth considering a distinction that I first
heard from John McCarthy (in a very different context): that between bad
analyses and bad theories. I think we would all agree that cross-linguistic
comparisons that do not take into account what is known about the histories of
the individual languages being compared would be bad analyses. But it doesn't
follow from this that it is impossible for me as a researcher to do enough
reading about Basque grammar and about the history of the language before I
attempt to establish that it is a variant of (to pick a geographically
plausible candidate) Berber. But I would hope that anyone wishing to establish
a Berber-Basque link would have more motivation than geographical
plausibility, perhaps in the form of surface similarities, which might prove,
on further investigation, to be misleading.

So, if Larry's point is that you need more than dictionaries of languages you
don't know in order to do responsible long-range comparison, I agree. But if
his point is that if you need the dictionary, you can't do long-range
comparison, I disagree.

Alice Faber
Faberhaskins.yale.edu

Message 4: Re: 5.1484 Comparative Syntax

In reply to Philippe Valiquette, I need to point out that the
fact that the rule of sg. verb with neuter pl. subject noun
was not always followed in Greek, say, would be taken to
mean that the language was precisely moving away from an
irregularity (which therefore is likely to reflect the
prehistory) to a more regular situation. And the standard
Indo-Europeanist account discussed by Meillet is that what
was synchronically irregular in Ancient Greek and Gathic, say,
had been regular in Proto-IE because the neuter plurals are
reconstructed as coming from collectives nouns, which would
have been singular in PIE. So it would be a classic example
of a synchronic irregularity shared by two (or more) languages
in a family, hence likely to be inherited from a proto-language
(in which we find that the apparent irregularity was really
regular after all). The same reasoning is used all the time
in reference to phonology, morphology, and lexicon. I was
just saying that, while not as commonly, it is also used
in relation to syntax (and it is not my idea, of course, but
a standard technique).