Reconstruction talk:Proto-Indo-European/átta

Discussion

In czech, slovak, polak and maybe some other slavic languages (i dont know) exist also a word "táta" as a deminitivum of otec. It is probably from word "*átta" also and isnt written in this article. 93.91.144.178 15:43, 8 January 2010 (UTC)

Yes, but it's a later hypocoristic derivation of the inherited word and hence irrelevant. --Ivan Štambuk 19:17, 8 January 2010 (UTC)

It exists in Bulgarian as well, only in the vocative form - тате. The uſer hight Bogorm converſation 09:32, 13 January 2010 (UTC)

átta

This was originally a kind of baby-talk word, and so could be more accurately translated "daddy"... AnonMoos (talk) 01:04, 22 November 2012 (UTC)

Yes, it could well have been strictly a nursery word originally that, in view of its form (which doesn't suggest a specific inflection, and consequently we don't give an inflectional table) probably even existed only in the vocative as suggested at Ätti. Basically, it may have functioned as a familiar version of the more respectful and distanced vocative *ph₂ter. Its phonetic form, which diverges strongly from typical Proto-Indo-European nouns, supports this conclusion.

By the way, this also makes the reconstruction given for Proto-Indo-Iranian in the article dubious, and the Ossetic reflex presupposes *ata, in fact. I've changed it.

Ultimately, the foundation of these baby-talk words appears to be a syllable of the type /Ca/, with nasal stops typically for female relatives (ma, na) and stops typically used for male relatives (pa ~ ba, ta ~ da), very frequently reduplicated (yielding the typical CaCa forms), sometimes with gemination (CaCCa) and not infrequently with the first consonant dropped (aCCa, aCa), or the final vowel (CaC), or even both (aC); the final vowel may also be replaced by /i/ (CaCi, etc.). (See also: The mamas & the papas in babies' babbling on Sentence First).

This entry is actually a bit unusual because it suggests that the baby-talk word has been recreated in every branch anew, but inherited from Proto-Indo-European – but the reconstruction is reasonable since it is formally possible and some of the reflexes show the expected effects of sound laws.

Words like Latin tata, Proto-Slavic *tata or Proto-Brythonic *tad, however, were certainly formed independently and do not go back to this proto-lexeme; while Latin tata and Celtic *tatos could hypothetically be based on a separate Proto-Indo-European lexeme *táta, the Slavic lexeme cannot be directly descended from it (at least not without reformation), as the expected Proto-Slavic reflex would be *toto. Therefore the basis for such a reconstruction is probably too thin, and this kind of words is too easily recreated independently. --Florian Blaschke (talk) 12:45, 22 April 2019 (UTC)

Excellent exegesis, Florianie: tata in Pipil, Dakota, Quechua... and then these CaCa, aCa patterns.... Explains a lot.

Naj živi oče/otcovia/tatowie! (Mamy, baby, papy, opy, omy, nanie, ciocie też.) Zezen (talk) 17:34, 1 August 2021 (UTC)

Reconstruction

The current reconstruction éh₂tteh₂ does not make much sense. First, this form should have given long initial *ā-, which clearly is not the case; and second, the modern consensus is that no word in proto-Indo-European started with a plain vowel. A better reconstruction is h₂étteh₂, if we wish to avoid the baby-talk reconstruction *atta. I personally don't see anything wrong with it, though. It concurs with the Anatolian data, which in principle preserved *h₂ (albaeit not *h₄).

I suppose that's fair, but using that disputable law, how do we rationalize *átta starting with a vowel? -/ut͡ʃxʎørnɛja ☭/ ^{(탁ᷞ, кон-, ឯឌឹត្ស, 𐎛𐎓𐎄𐎛𐎚𐎒)}. 16:49, 25 February 2019 (UTC).

There is no need to rationalize baby-talk words. They have their own inner logic. Only "grammatical" roots must follow the established phonotactic rules. Bezimenen (talk) 00:45, 26 February 2019 (UTC)

Agree. There is no way to reconcile this reconstruction with the Leiden school principle "no phoneme /a/ in Proto-Indo-European", and even they make allowance for exceptions such as nursery words. --Florian Blaschke (talk) 13:02, 22 April 2019 (UTC)

Word-final *s

So, I've noticed that some reconstructions add a -s at the end of the word, or at least a (s), to reflect the Hittite & Slavic descendants. I'm wondering if we should put the parenthetical s at the end to accomodate this? -/ut͡ʃxʎørnɛja ☭/ ^{(탁ᷞ, кон-, ឯឌឹត្ស, 𐎛𐎓𐎄𐎛𐎚𐎒)}. 05:09, 26 February 2019 (UTC).

Scratch part of that, only the Hittite descendant. -/ut͡ʃxʎørnɛja ☭/ ^{(탁ᷞ, кон-, ឯឌឹត្ស, 𐎛𐎓𐎄𐎛𐎚𐎒)}. 05:11, 26 February 2019 (UTC).

Hittite remade this word to an a-stem. The original form did not conform to typical IE morphology, that's why it was remade into a different inflection class in practically every language. The entry is correct as it is. –Tom 144 (𒄩𒇻𒅗𒀸) 17:39, 3 March 2019 (UTC)

ata

The word exists in Assamese(An Indo Aryan Language) meaning Grandfather. This could be a decendant of PIE átta right? 47.29.205.53 08:35, 1 September 2020 (UTC)