LINGUIST List 5.1405

Wed 07 Dec 1994

Disc: Comparative method in linguistics, continued

Editor for this issue: <>

Directory

1. ALICE FABER, Comparative Method
2. , accuracy of historical reconstruction
3. Jacques Guy, Re: Comparative linguistics

Alexis Manaster Ramer asks about the claim that relatedness and/or subgrouping
should be established only based on systematic morphological relationships of
the sort likely to be observed in paradigms or declensions. I'm familiar with
this claim only in a much weaker form, that morphological comparisons are more
reliable than phonological ones as a basis for establishing linguistic
relationships and subgroups. On this basis, for example, Robert Hetzron in
1976* proposed a rigorous internal subgrouping for the Semitic
languages based on affixes in the verb paradigms. Hetzron's proposal that
Hebrew, Arabic, and Aramaic constitute a Central Semitic group is, I believe,
correct; and it is supported by much more evidence than Hetzron adduces, much
of it morphological and morpho-syntactic (e.g., innovation of novel negative
markers, etc.).

I may be reading too much into these claims of morphological priority in
establishing subgrouping, but I have always interpreted them as a reaction to
the difficulty of distinguishing convergent from shared phonological
development on a principled basis. That is, because it CAN be difficult to
determine whether a particular recurrent sound change in a language group
represents shared innovation rather than convergent development, it might be
pragmatically safer to rely on morphological innovation. Thus, in the case
of the Semitic languages, such changes as *p to /f/ or *g to /jh/ (as in junk)
would, if treated as shared innovation, lead to subgroups that are
inconsistent with those deduced by other means. On the other hand, "unusual"
changes like the change of Proto-Semitic glottalic consonants to
pharyngealized consonants are much more likely to represent shared innovation,
given the typological rarity of pharyngealized consonants.

With regard to Fritz Newmeyer's questions about comparative syntactic
reconstruction, I don't know of any systematic published counters to Jeffers'
(and others') claims that it is *in principle* impossible. However, I think
that a good case can be made that this is an overly pessimistic assessment.
The problem, of course, is the appropriate context: we compare phonemes in
words and/or morphemes and morphemes in paradigms, but it's not clear what the
context might be for word orders. Presumably discourse context plays a role. I
would imagine that if all the languages in a family shared an unusual word
order (vis a vis their dominant types, whatever those might be) in
counterfactuals, we might want to attribute that order to their latest shared
ancestor. Pragmatically speaking, it's a lot easier to find information about
the morphological context of particular phonemes than it is to find reliable
information about the larger context for sentence and construction types.
Nonetheless, at least inchoately (and perhaps it is the inchoateness that
Jeffers objects to), *some* notion of syntactic reconstruction is surely
behind claims that Proto-Indo-European was SOV or Proto-Semitic was SVO, and
the like.

*"Two Principles of Genetic Reconstruction", Lingua 38: 89-104.

Alice Faber
Faberhaskins.yale.edu

 In writing in LINGUIST 5.1393 on another topic,
 martinhafub46.zedat.fu-berlin.de (Martin Haspelmath) says:

 ) It is true that Latin syntax could hardly be reconstructed from
 ) modern Romance languages, but neither could Latin morphology, and
 ) even the view of Latin phonology that we would get from Romance is
 ) very distorted. Our reconstruction of protolanguage grammar is
 ) always imperfect...

 In looking at historical reconstruction done for S. American languages
 (largely phonological), I've often wondered about this. Just how much
 could we trust the reconstructions that we did? One of my rules of
 thumb for those languages was that if an item was longer than one
 syllable, it was suspect as being polymorphemic, and if it was longer
 than two syllables it was almost certainly polymorphemic. The problem
 is that most attempts at reconstruction ignored this areal phenomenon
 (sometimes because the data was simply unavailable). From what I know
 of Romance languages, I would say polysyllabic morphemes are more
 common there. If anything, that should make it easier to reconstruct
 Latin, since you have more to work with. (Of course, the morphology
 of Romance languages is much better known than that of the languages
 of S. America, which also helps!) So if Latin would be very
 imperfectly reconstructed, what hope is there for Native American
 languages?

 Has anyone ever attempted, as an exercise in the comparative method,
 reconstruction of Latin from the Romance languages, then compared the
 results with the real thing? Or reconstruction of any other attested
 language from its descendents?

) From: amrares.cs.wayne.edu
)
) It seems
) to me that a lot of the people [...]
) also seem to hold another curious position, namely, that you
) cannot show the relatedness of a group of languages by exhibiting
) systematic correspondences of sounds in the vocabularies of said
) languages but only by exhibiting systematically related morphological
) paradigms. [...]
)
) Alexis Manaster Ramer

Good Lord! I have been watching this thread with a somewhat jaundiced
eye, thinking "I'm not going to get into this", but this...

)From my experience with languages of Vanuatu, morphological paradigms
are the *least* stable features, followed by phonology, then, most
stable, lexical. Yes, I remember having been taught that, I mean,
about morphology being most reliable, actually, the *only* reliable
criterion. I was also taught a lot of other stuff which experience
showed me to be false.

Think of it, why should this reliance on morphology have come about?

1. A hangover from the elaboration of the comparative method on
Indo-European, or, I should rather say, on the discovery of Sanskrit.
It may so happen that members of that language family have been
particularly retentive morphologically. Whence generalization to
all languages. The fallacy of extrapolating. All Irish barmaids
are redheads.

2. It is easy to measure and count lexical similarity. Claims
based on such measurements are therefore more easily open to
scrutiny, and to refutation. (An aside: there's been some
discussion on whether linguistics is a science, in the
meaning of Karl Popper. Well, here's one domain where it
could be). On the other hand, how do you measure morphological
similarity? And worse, systematic similarities of morphological
paradigms? This, then, is hardly open to refutation.

Speaking of the devil, I received a letter yesterday from Merritt Ruhlen
"I would appreciate it if you could send me a copy of your
forthcoming article in Anthropos concerning the probability
of chance resemblances". Which I did, commenting, meaning it
as a sort of salve -- but perhaps it will be felt as salt:
"In general I take a dim view of comparative linguistics.
Its various methodologies are mostly ad hoc and without a
sound basis, often relying on a complete misunderstanding
of the processes at work [here a few examples]. And the
same claims and methods crop up perennially." Which is
also my thoughts and feelings today. (* sigh *)

j.guytrl.oz.au