phonetic resemblances etc.

Sally Thomason sally at
Sun Jan 31 19:52:21 UTC 1999

----------------------------Original message----------------------------

   Alexis Manaster Ramer refers to Goddard's Comecrudan proposal
as "this little bit of classificatory work" and says that this is,
as far as he [MR] knows, Goddard's "sole substantive contribution
to this area of research".  If by "this area of research" MR means
"classificatory work" in general, his claim is startling.  Goddard's
1975 article "Algonquian, Wiyot, and Yurok: Proving a Distant
Genetic Relationship" (in a Festschrift for C.F. Voegelin), though
of course not the first or only proposal of that grouping, is the
most substantive and convincing, and is certainly a well-known
contribution to the literature on long-distance relationships.

   It's worth remembering that the Comecrudan example first came up
because MR suggested it as an example of a genetic
relationship posited by a highly respected historical
linguist on the basis of phonetic resemblances alone.  My
point in mentioning some of the details of Goddard's proposal
was not to endorse it particularly -- I agree with Larry
Trask that there isn't enough data for serious hypothesis
testing (though we should also remember that Goddard
*postulated* the relationship; he did not say that he had
established one) -- but to point out that Goddard emphasizes
the recurring correspondence as a crucial bit of evidence.
That is quite different from mere phonetic resemblance.

   For the record, in his 1979 article on languages of South
Texas and the Lower Rio Grande, Goddard doesn't give an exact
indication of the amount of data available for Comecrudo --
he cites an early "vocabulary of 148 entries" (1829) and
refers to a "more extensive body of material" collected
later.  But he does give figures for the other two languages
in his postulated family: a 21-word vocabulary for Garza (1828),
whose speakers were already "largely acculturated and all spoke
Spanish" (p. 371) and a 22-entry vocabulary for Mamulique.  So
the database is sparse indeed, as Goddard notes.

   Sally Thomason
   sally at

More information about the Histling mailing list