🚀 go-pugleaf

In article <1992Oct31.215441.15379@s1.gov>, lip@s1 (Loren I. Petrich) writes:
>OK, you guys, get ready to point your flamethrowers at Scientific American,
>because SciAm just published (Nov. issue) an article by Joseph Greenberg and
>Merritt Ruhlen describing their proposed Amerind family.
>
>Much of their article is devoted to discussing some of the correspondences
						^^^^
						*two*

>they have found; they focus on Amerind, though they do perform comparisons to
>Nostratic/Eurasiatic and Sino-Caucasian toward the end. They are silent about
>Proto-World (fortunately!).
>
>They do acknowledge their rejection by much of the linguistic community, but
>they state that their critics have no better ideas.

Matter of taste, doncha think?

>They also criticize many mainstream linguists for doing two-at-a-time
>comparisons, which they compare to working out Indo-European by comparing
>English and Albanian. Given the way Indo-European was actually worked out,
>with multiple comparisons, that criticism would seem justified.

You know, people on this group have on occasion thrown this charge at me when
I've complained about the work of various large-scale comparatists.  However, I
hadn't realized that the charge was being made *as raising a real issue*.

I realize that I've not met every historical and comparative linguist in the
world, but I would be surprised to learn that those I *have* met are in any way
unusual.  *NO ONE* of my acquaintance does strictly binary comparisons, or
suggests that that is the way to do comparative studies.

Leonard Bloomfield, in his work on the reconstruction of Proto-Central-
Algonquian, opined that 4 languages made a nice set with which to work, *in
order to simplify the task.* However, one should be careful in selecting them,
so as to obtain the most varied, rather than the most homogeneous, set of
correspondences, in order to allow one to capture the largest number of
generalizations.  (I think that he summarizes this as well in the chapters on
historical linguistics in _Language_--which chapters were published separately
as _Language History_ in an edition by Harry Hoijer.)

When I did a class on comparative linguistics with Dyen, we were required to
use 4 languages in our term project.  Of course, he studied with Bloomfield...

Callahan's work on Proto-Miwok used *all* the languages (8-10, as I recall), to
avoid the problems of small corpora.

Do people actually think that the usual practice of comparative linguists is to
do binary comparisons????  I thought I was being presented with an exaggeration
to make some point that I couldn't see.

>In his book _Language in the Americas_, JG makes an example of comparing Irish
>and Armenian; if one looked only at those two, would a relationship between
>them be very apparent? It is the knowlege of the other IE languages that
>helped establish their relationship, which is effectively constructed from:
>
>	Irish -- (other IE)
>	Armenian -- (other IE)
>
>Since the quality of relatedness is transitive, we find, though indirectly,
>
>	Irish -- Armenian
>
>However, they did not pay much attention to the question of the sort of
>precise sound correspondences that the more mainstream of comparative
>linguists, and even the Nostraticists, hope to discover.
>
>OK, now, fire away :-):-)

Minor irritation with their presentation of Jones' work:  His use of "Gothick"
[sic] would be better paraphrased in modern parlance by "Proto-Germanic."

Less minor irritations:  Their dismissive tone in discussing his work at all.
I'm convinced by Cannon's arguments that Jones used LARGE amounts of data from
his various languages, and did a truly comparative study, before publishing his
findings.  Further, they twist the reading of his statement so that he seems to
have made claims *only* about "roots," when in fact his statements made
reference to both inflections and roots, and *emphasized* that it was not
merely roots that were similar.

I find unconvincing their citation of the genetic data, which has come under
fire separately, as proof that their reconstruction of the number and course of
migrations is correct.

Admittedly, I don't have the time to keep up with the archaelogical journals
and have to take my information from popular media, but isn't it the case that
solid evidence for long-term human presence in the Americas is growing?  That
is, is their timeline long enough at this point?  And what route(s) do they
propose for the migrations?
-- 
Rich Alderson   'I wish life was not so short,' he thought.  'Languages take
                such a time, and so do all the things one wants to know about.'
                                                        --J. R. R. Tolkien,
alderson@leland.stanford.edu                              _The Lost Road_