No Proto-Celtic?

[Eduard Selleslagh]

----- Original Message -----
From: "Stanley Friesen" <sarima at friesen.net>
Sent: Thursday, June 14, 2001 3:35 PM

> At 12:36 PM 6/8/01 +0000, Gabor Sandi wrote:

>> I think that it is odd as well, yet I can't help noticing that personal
>> markings of verbs in IE (and Uralic and Altaic as well, for all you
>> Nostraticists) consistently come after the verb stem and not before it.
>> What's more, some of these endings contain consonants identical or similar
>> to what is found in the corresponding personal pronoun. -m- in the first
>> person is the most obvious (in PIE and Uralic, plus some other proto
>> languages), as is the -t- in the 2nd person plural (also, 2nd person sing.
>> in Uralic), -t- in the third person less so, the -s- in the 2nd person sing.
>> not at all. There are three competing hypotheses:

>> 1. coincidence
>> 2. agglutination of the pronoun
>> 3. there is a relationship, but the ending is not the result of an
>> agglutination of the verbal root with a pronoun

>> If the answer is no.2 above, I would like to come up with a succession of
>> linguistic changes that look reasonable and result in the pattern we see.

> Well, it is certainly possible that some pre-PIE language was VSO without
> effecting the possibility that PIE itself was SOV.  Such transitions are
> attested.  Perhaps this means that Proto-Nostratic was VSO, perhaps
> not.  Either way, once the clitics have transitioned into inflections there
> is no bar against a change in the unmarked position of the subject.

> [Note, reconstruction of word order is difficult, there is good evidence
> for "early" VSO in both Germanic and Celtic - but that is still later than
> the classical IE languages].

[Ed Selleslagh]

Even as a non-specialist, I can only note that in agglutinating languages (e.g.
in some Uralic and Altaic, but also in Quechua, the Inca language) the person
marker for verbs and the possessive marker for nouns is virtually the same
suffix, or obviously derived from the same form. IMHO this pleads in favor of
'2. agglutination of the pronoun'. Maybe it also suggests a previous
agglutinative phase for PIE; if so, that would make the position of the old
suffixed pronoun irrelevant for the VSO/SOV... discussion.

It might also suggest that in the early stages of agglutinating languages the
distinction between verbs and nouns wasn't as clear-cut as we might want to
believe. That would suggest an even earlier stage when lexical words
represented essentially 'concepts', and were themselves invariable like in
isolating languages.

So, my first suggestion that the enclitic pronoun might be due to PIE having
been VSO was probably not so fortunate, but it yielded an interesting dicussion
nonetheless.

Ed.