[b-hebrew] Nostratic

Yitzhak Sapir yitzhaksapir at gmail.com
Thu Apr 19 01:02:17 EDT 2007

On 4/18/07, Peter Kirk wrote:

> Yitzhak, thank you for this. But note that in the post you are replying
> to I did not mention Nostratic, only Proto-Indo-European, whose
> reconstruction is much more solidly based. Indeed I mentioned Nostratic
> only in a PS to my earlier post. What I wrote did not at all depend on
> anyone's Nostratic theories, only on the correspondences between
> carefully reconstructed proto-forms of various language families.

It made indirect reference to the original post.  However, like you, the
substance of the above deals not specifically with Nostratic so much as
it deals with Afroasiatic vs. Indo-European similarities.  We cannot
reconstruct Afroasiatic to the point that we can reconstruct any
similarities with Indo-European.  That is partly what Don Ringe is getting
at.  The pronouns that you list are also problematic.  For example, the
second person accusative (enclitic) pronouns are not "te" or "wos" but
*ka/*ki and *km(/*kn) ("kum/"kin).  As if this is not enough, Ringe writes
earlier in the article, "Finally, an unpleasant fact of language change
imposes the most drastic limitation on what can be known.  All
languages gradually replace their inherited vocabulary with completely
different and unrelated vocabulary items, and also replace, lose, and
restructure the affixes with which full words are formed.  'Basic'
vocabulary is, of course, replaced at a relatively slow rate, and
inflectional affixes are also resistant to change; but in the long run every
word will be replaced, and inherited inflectional patterns will be
transformed beyond recognition.  When the vast majority of even the
most tenacious items have disappeared, the few remaining cognates
shared by genuinely related languages will be indistinguishable from
chance resemblances -- so that the relationship will be undiscoverable."
Your list of "morphological similarities" which did not include a list of
Afroasiatic counterparts are no different than chance resemblances.
Peculiar morphological patterns are useful to determine relatedness,
but such comparison of irregular or idiosyncratic developments are
useful only to fine-tune an already substantiated conclusion of the
language's relatedness.  Otherwise, they are no different than shared
occurrences or even that the two languages underwent similar sound
changes due to a period of bilingualism between the two languages.

> Such things are indeed possible. But, as I think is widely recognised,
> morphological suffixes of the kind I have been looking at are about the
> least likely language features to be borrowed between languages even in
> close contact.

It is just as widely recognized that Indo-European and Afroasiatic cannot
be reconstructed as derivatives of the same language and any similarities
between them are no different than chance occurrences.

Yitzhak Sapir

More information about the b-hebrew mailing list