Reconstruction talk:Proto-Balto-Slavic/dṓˀtei

Old Prussian has infinitives:

• in -ton, -tun (from accusative singular of u-stems or a supine), e.g. daton
• in -wei (from dative singular u-stems), e.g. dātwei
• in -t, e.g. dāt which could be apocopic form but not necessarily. According to Matasović 2008:300, also referencing Christian Stang's Vergleichende Grammatik der baltischen Sprachen p. 446ff, this proves that Balto-Slavic infinitive cannot be reconstructed, especially because these infinitive formations in OP co-occur, in same verbs. Of course, it could easily be that the Balto-Slavic dialect ancestral to Slavic and Eastern Baltic had this infinitive, but it would be wrong to mark it as Proto-Balto-Slavic since it doesn't cover Old Prussian. --Ivan Štambuk (talk) 12:46, 18 August 2013 (UTC)

Still, isn't this the form that is most commonly reconstructed? Latvian and Lithuanian have a shortened form in -i, and that is probably the form that is ancestral to Old Prussian -t as well. The two u-stem forms are from *-tus. So perhaps the truth is that Balto-Slavic had two infinitive formations that were used primarily as verbal abstracts, and were in competition with each other, with one formation or the other eventually "winning". But even then, Slavic kept both as well. So it may not be correct to speak of "the" infinitive, but we can be pretty certain that this form was "an" infinitive, and that a parallel form *dōtawei likely existed too. This form (in the nominative, at least) then gave rise to the Slavic supine *datъ. —CodeCat 12:53, 18 August 2013 (UTC)

Lunt says in his OCS grammar that the Slavic supine is accusative in origin, and matches the Prussian -tun. So that clears up at least one side of things. —CodeCat 13:01, 18 August 2013 (UTC)

Stang says thet there was no syntactic difference in Old Prussian between supine and infinitive and that OPr. -t is from earlier -ti. So where do we list Old Prussian infinitives in -twei ? --Ivan Štambuk (talk) 14:15, 18 August 2013 (UTC)

I suppose this is where we have to make a distinction between word forms and paradigms. When we list descendants, we list them as whole paradigms even if the lemma forms don't match exactly. For example, Old English weak class 2 verbs have the infinitive ending -ian, but the Germanic ending is *-ōną which is not its direct ancestor. But we list those verbs there anyway.

In PBS, we can't list one single infinitive that is the ancestor of all the infinitive forms in all languages, but we do know that the -tei ending survived in all of them in at least some form. So I think we should just cheat a little and say that the verbs in -twei descended from the PBS form in *-tei, even though we know that the form -twei itself came from the u-stem dative. We should probably make a *-tawei entry (the original ending that I reconstruct based on PIE and PS) that explains this. We probably need *-tun as well. —CodeCat 14:23, 18 August 2013 (UTC)

Also, if you want to be this technical, what about Bulgarian, where the same thing happens? —CodeCat 14:25, 18 August 2013 (UTC)

Yes but for Bulgarian and Macedonian the loss of infinitive (as a characteristic of Balkan sprachbund) is something documented in the historical period, and mapping different ancestor lemma to daughter lemma is not an issue. For proto-languages it's all just speculation, and when you say that PBSl. *dōtei is reflected as Old Prussian dātwei you're implying that Old Prussian somehow lost the original form. Considering that Old Prussian is more archaic than Slavic or Eastern Baltic, if anything, it should weigh more in such decisions. The problem is in the whole Stammbaum model which is artificial and doesn't describe how the real-world languages change. There were probably other infinitives coexisting in unattested Balto-Slavic dialects.

Where do you get -a- in *-tawei, and why was it lost in Old Prussian? --Ivan Štambuk (talk) 14:37, 18 August 2013 (UTC)

It's based on the Slavic u-stem dative ending *-ovi, and also the PIE ending *-ewey with the regular change -ewV- > -owV- in Balto-Slavic. —CodeCat 15:17, 18 August 2013 (UTC)

According to Stang, Old Prussian -twei is from earlier -tuwei where this -u- was taken from case forms that had it (N, A, I), and PIE -o/ewey couldn't have produced Prussian -(t)wei. So PBSl. infinitive in -tawei probably didn't exist. --Ivan Štambuk (talk) 15:42, 18 August 2013 (UTC)