Reconstruction talk:Proto-Indo-European/káput

This root seems bogus.

• Sanskrit कपुच्छल kapúcchala "tuft of hair on the head" is explained without recourse to etymology as क ka- "head" + पुच्छ puccha "tail" + diminutive ल -la (i.e. "little head-tail").
• Celtic cuach/cawg is the same metaphorical extension we see in the Romance family deriving words for "head" from Latin testa; the primary meaning is not "head" but "cup, goblet, bowl".

That just leaves Germanic and Latin, which are easy pickings for a loanword spread along trade routes crossing the Alps. 90.202.1.172 05:01, 27 January 2019 (UTC)

RFD discussion: October 2019

The following discussion has been moved from Wiktionary:Requests for deletion (permalink).

This discussion is no longer live and is left here as an archive. Please do not modify this conversation, but feel free to discuss its conclusions.

---

Reconstruction:Proto-Indo-European/káput-

A bunch of strewn together unrelated words. See talk page. Please delete. --{{victar|talk}} 04:12, 20 October 2019 (UTC)

de Vaan gives Proto-Italic *kaput, Old Irish cúäch and Proto-Germanic *haubud- as IE cognates of Latin caput, but does not give a concrete PIE reconstruction and does not mention Sanskrit. It looks to me the entry may need some cleaning up but is not totally kaputt. — This unsigned comment was added by Lambiam (talk • contribs) at 11:05, 20 October 2019 (UTC).

@Lambiam: Sounds like your missed the talk page. --{{victar|talk}} 02:40, 22 October 2019 (UTC)

I would like to note that the borrowing (Latin-Germanic or vice versa) supposed by the IP on that talk page seems very fishy: it does not at all fit in the general picture of loanword contact between Latin and PGmc., where loans of simple anatomical terms like this one are not otherwise found. Such basic words are in general rarely borrowed even in situations of widespread and prolonged bilingualism (which is not the case with Latin/Proto-Germanic, in which case contact was relatively sporadic and indeed mostly limited to trade routes and occasional military contact), rendering the idea that this word was borrowed through contact on a trade route nonsensical. On the other hand, the similarity between the alternation caput/capit- in the Latin paradigm and the realization variously as *haubud- or *haubid- in the Germanic reflexes is odd, and I don't know enough about Indo-European linguistics to know whether or not this could be due to shared inheritance or has to be due to some other factor. I guess it could also be a substrate term borrowed into both pre-Germanic (must've entered the lexicon pre-Grimm's and Verner's laws) and Italic separately (if the Sanskrit is indeed not cognate) although that still leaves the problem of why a word as basic as "head" would be borrowed in the first place. — Mnemosientje (t · c) 13:26, 20 October 2019 (UTC)

@Mnemosientje: 1. Sanskrit: garbage, 2. The Celtic forms are most likely borrowed from Latin caucus, caucum (“cup”), Ancient Greek καῦκος (kaûkos, “cup”), or directly some substrate word, 3. Also probably from some substrate. I don't think PIE borrowing the word is no more unusual than Latin borrowing the word from Germanic, or vice versa. --{{victar|talk}} 02:40, 22 October 2019 (UTC)

I am really not sure how a Latin to pre-Germanic (or vice-versa) borrowing (as supposed by the IP on the talk page) would not be remarkable. Lexical exchanges don't happen in a vacuum and a borrowing of a word such as this one for 'head' absolutely does not fit in the general picture of early Latin-Germanic contacts. (This is pretty much a restatement of my previous post I guess: the point still stands.) Anyway I won't vote to keep as it definitely doesn't look like a PIE word (I'll take the arguments re:Sanskrit and Celtic), but I'm not sure what it does look like. Probably substrate, yes, but the details of its borrowing (via what route? why only Italic and Germanic? why borrow a word for "head" in the first place?) are unclear to me and the theory presented for Latin and Gmc. on the talk page was and remains unsatisfying imo. — Mnemosientje (t · c) 09:30, 22 October 2019 (UTC)

As to the “cup” > “head” issue, de Vaan writes, “Since words meaning ‘cup’ frequently become ‘head’, it is not unlikely that caput and its Germanic counterparts are derived from the root of capiō ‘to seize’, which would point to a substratum form *kap-ut- ‘cup’ > ‘head’.” The Celtic terms are presented as descendants of a different but cognate form *kapuko- , which “may continue an earlier form *kap-u-, a u-extension to a substratum root *kap-”. The uses of “substratum” and the absence of a concrete PIE reconstruction indicate that de Vaan does not subscribe to the theory of a common IE ancestor. So while cognacy may hold, there appears to be insufficient basis for this specific PIE reconstruction. In conclusion, all considered: Delete.  --Lambiam 11:20, 22 October 2019 (UTC)

@Mnemosientje: I'm totally with you on it being all kinds of "remarkable", and the vice versa is a little more palatable, in my opinion, but for an example of "head" words being borrowings, look no further than skull.

I guess that example doesn't really apply, since head is still the primary word for head and skull can be used to mean head only metonymically, and since English-Norse contact was a couple orders of magnitude more intense than early (pre-Grimm+Verner's law) Latin-Germanic contact. Again though, I am pretty much in agreement with you here (as is Lambiam, it seems; in any case to clear up any confusion I vote delete on the PIE), it definitely doesn't seem PIE. As for its age, I would again point to the effects of Grimm's and Verner's law that are visible here: do they not point to a quite early date? I am not sure how those sound laws compare chronologically to the last stages of Proto-Italic, but perhaps even so early that the Proto-Italic term may also have existed? (Btw: I just realized we have another example of the cup-to-head development right in my own language, kop!) — Mnemosientje (t · c) 08:47, 23 October 2019 (UTC)

@Lambiam: We have PIE roots that are borrowed from substrates, but when you just have two forms, and one could be conceivably be borrowed from the other, that's where I cry delete.

I have to say, I'm also a little annoyed when linguists feel they need to reconstruct PIE roots for every word, especially in Germanic, where every word must obviously be PIE in origin. Greek and Indo-Iranian have gotten on the substrate train, why can't Germanic? </rant> --{{victar|talk}} 20:08, 22 October 2019 (UTC)

With no other Italic cognates -- another reason to question its age -- Reconstruction:Proto-Italic/kaput should also be deleted. --{{victar|talk}} 22:17, 22 October 2019 (UTC)

I realize I'm playing devil's advocate a lot in this thread but this word really interests me, so I have to ask: couldn't the absence of this term in other Italic languages not be explained by the fact that other Italic languages are very poorly attested in general? Is a different (non-cognate to caput) word for head at all attested in a non-Latin Italic language? — Mnemosientje (t · c) 08:52, 23 October 2019 (UTC)

Sure, but if you have no interfamilial cognates, why bother reconstructing an intermediary ancestor to a PIE form that didn't exist?. Also, another good reason to theorize that Germanic borrowed the word directly from the substrate language is the unusual free variation we see. --{{victar|talk}} 18:39, 23 October 2019 (UTC)

@Mnemosientje, Lambiam I had a change of heart and cleaned up the entry and its possible parent entry. --{{victar|talk}} 06:16, 28 October 2019 (UTC)

What about the claim of cognacy with Sanskrit कपुच्छल (kapúcchala), denounced by the IP at Reconstruction talk:Proto-Indo-European/káput and now removed from the descendants of *káput but still stated at the entry *haubudą?  --Lambiam 08:15, 28 October 2019 (UTC)

*kap-, kap-

I also do not find it likely that the alleged root *kap- (or with others: *keh₂p-) may be actually reconstructed. This /kap/ is, what few language people are conscious of, a kind of imitative or ideophonic element. I have observed babies saying with much resolution kap, kap, kap when they gripped or trained to grip, as is in the human infant’s nature. That is, not having heard languages where there is a similar word meaning “to grip”. That's why you find the word kapmak (“to grab, to snatch”), kapamak (“to close dff”) reconstructed for Proto-Turkic. Now what do ye think knowing that?

This I can tell you from own knowledge but there must studies on the kinds of words that babies utter but aren’t words in their target languages (not the kind of words that one finds searching for baby’s “first words” or the like – the babbling). What’s even the linguistic terms for such utterances? There must be some since there are whole chairs dealing with language acquisition by infants. If Indo-Europeanists dealt with this discipline more often they would avoid many chance correspondences. Also to note: what is possible for infants also happens with adults, so one does not need to assume that babies wreak language change, it is rather child-imitation, child-directed language (or non-language) that spawns such terms across languages independently. Fay Freak (talk) 12:29, 28 October 2019 (UTC)

My understanding (from a child language acquisition class I took in the '80s) is that children babble in order to develop the coordination necessary to make speech sounds. They start out with generalized labial sounds, which require the least motor coordination, and progressively add other places and manners of articulation as they gain control of their vocal apparatus. That's why there are so many unrelated languages with words for mother and father that start with labial sounds. The combination of a labial and a velar in the same root seems like it would require more coordination than one would expect from early-childhood speech- at least the kind that would be so natural as to be universal like this. Of course, a lot has happened in the field in three decades, so understandings may have changed. Chuck Entz (talk) 14:09, 28 October 2019 (UTC)

What do I think? I think that's nothing new -- we reconstruct plenty of onomatopoeic PIE roots. I do think though that the page should be fashioned as *gʰebʰ- ~ *kap-. What really convinced me of that is we have *gʰébʰ-ōl ~ *káp-ōl (“head”), whence Sanskrit कपाल (kapāla), Ancient Greek κεφαλή (kephalḗ), Tocharian B śpāl, Old High German gebal.

However, as per my {{rfd}} above on Reconstruction:Proto-Indo-European/bew-, I think some are alleged onomatopoeic roots are junk piles. Something needs to be done with Reconstruction:Proto-Indo-European/leb- as well. --{{victar|talk}} 15:05, 28 October 2019 (UTC)