22 Sept 2010

From whence Sanskrit kapúcchala?


As I've probably mentioned before, I strongly suspect Julius Pokorny and followers have lazily lumped Sanskrit kapúcchala 'tuft of hair from the back of the head' in with other evidence supposedly supporting Proto-Indo-European (PIE) *kaput 'head', all just to give the illusion that the evidence is more robust and geographically dispersed than it honestly is. It's also a lot easier in any bureaucracy, including in academia, to simply go with the flow and ne'er question the status quo. However in this case, I'm fortunately not the only one out there that thinks this smells fishy. I insist that this PIE root never existed and that there are only Western European reflexes of this 'head' word, all attributable to loans from the Aegean family during the 2nd millennium BCE and later, ie. from either Minoan *kaupada (> Greek κεφαλή) or Etruscan *kaupaθ (> Latin caput; indirectly into Germanic as *haubidaz prior to Grimm's Law, perhaps through Venetic).

Though I found one lead online stating that Mayrhofer once dared to analyse kapucchala into a pejorative prefix ka- plus puccha- 'tail' (Mayrhofer, Kurzgefaßtes etymologisches Wörterbuch des Altindischen [1956], p.157), I've just come across a curious entry in both Cologne Digital Sanskrit Dictionary and Capeller's Sanskrit-English Dictionary that identifies the syllable ka alone as 'head'. This tickles me. Since I knew already that पुच्छ puccha meant 'tail', this implies that कपुच्छ ka-púccha-la- with diminutive -la- just means 'little head-tail', perfectly fitting for a tuft at the back of the head.

If the word can be explained purely in Sanskrit terms, a PIE origin would be woefully extravagant by comparison and then easily dismissed as bunk. The other spelling kaputsala would be just an alternative phonetically-faithful rendering and certainly adds nothing to the arguments of the **kaput camp until they can substantiate both **kaput and **śala-. Even the justification for this unmotivated segmentation of the word is lacking. It seems to be based on wishful thinking.[1]

That being said, now I'm having trouble confirming the source of the equation ka = 'head'. Is it attested somewhere directly? Or is this purely assumed by 19th-century Indicists attempting to etymologize Sanskrit vocabulary (in which case, an asterisked *ka is in order)? Oddly enough, there a few other words that strongly seem prefixed with this morpheme ka-: क-स्तम्भी kastambhī 'prop of a carriage-pole' (cf. स्तम्भ stambha 'post, pillar') and कं-धर kaṃdhara 'neck' (lit. 'head-bearer', cf. धर dhara 'supporting').

Rejecting PIE **kaput, what then is the etymology of Sanskrit क ka 'head'?


NOTES
[1] After posting this, I managed to discover one tantalizing lead that may help settle this issue (see Brugmann/Streitberg, Indogermanische Forschungen, v. 3 [1894], p.236) which I subsequently posted in my commentbox. If I'm reading the German correctly, it seems like the authors are admitting that kaputsala was caused by a more modern modification of kapucchala, based on an etymological whim.

8 comments:

  1. As far as I can see from the source abbreviations in the Peterburg dictionary, the assignment of the meaning "head" etc. for ka(m.)- is based on Sanscrit lexicographers and glossaries. All examples I can find in a quick search of the Cologne dictionaries are compounds, i.e. we're talking about assigning meanings to a compound element, not an independent word. FWIW, Witzel (p. 37) lists kandhara as a candidate for a Para-Munda substrate word. As both stambh and dhar have good IE pedigrees, we would have either to dismiss a Para-Munda link here or we could assume the loaning of a prefix (in itself nothing strange, as the use of Latin and Greek prefixes with native words in modern European lnguages shows).

    ReplyDelete
  2. Great blog even for the non-expert. Thanks!

    ReplyDelete
  3. The Monier-Williams dictionary, which often gives one (arbitrary) source for meanings, attributes the "head; hair, a head of hair" meaning to "L.", which stands for "Lexicographers, esp. such as अमरसिंह , हलायुध , हेमचन्द्र , &c." So the meaning is present in Sanskrit, although MW doesn't exactly specify where or what the etymology is.

    BTW, irrespective of whether "ka" is "head" or not, I suspect that "kapucchala" does come from "puccha" (tail) in some way.

    ReplyDelete
  4. Hans,

    Yes, puccha is also given a PIE etymology (< *puk- 'tail', hence also Germanic *fuhsaz/*fuhōn 'fox' and Tocharian pako 'tail'). I'd love to know a reasonable answer to all this. I guess time will tell because I've run out of ideas for now.

    ReplyDelete
  5. Although the analysis ka-puccha-la is interesting, it doesn't explain why there is a variant form kaputsala with the same meaning.

    The variance needs to be accounted for. It's not a phonetically faithful variant. ccha is a result of *ske, or *voiceless stop+śe/o and never for ts

    puccha is never found as putsa.

    Also the meaning head as ka- is a word purely found in dictionaries, and never in an actual text. It could very well be that this word was invented to account for the word kapucchala with the analysis you gave., but not necessarily by a 19th century scholar. There is a very long and fruitful tradition of coming up with folk etymologies for words ever since Sanskrit has been used as a religious language.

    An example would be the root ind- which according to the lexicographers means 'to be powerful'. But the root was simply invented to account for the name indra-.

    Same could be true for ka- to account for kapucchala.

    Especially when you get to Buddhist literature you realise that any form of etymology thought up by classical Indian etymologists is per definition extremely unreliable.

    Take the name the concept of 'Worthy-One' in Buddhism Arhān: clearly a present participle of root arh- `to be worthy'.

    Yet the old Indian scholars reinterpreted this as ari+han or: enemy-kill. Because he is a killer/defeater/destroyer etc. of enemies.

    Nice thing about this is, that many of these folk etymologies snuck into Tibetan translations of these words. So an Arhān in Tibetan is known as a dgra-bcom-pa: enemy-destroy-Nounsuffix.

    So if there really is no attestation of the word ka- as 'head' on its own, how can we be sure it is not a old folk etymology that they are trying to explain by coming up with this word?

    Only way to be sure is to find many more words that convincingly have an element ka- in its compound where it has the meaning 'head'.

    It has been suggested a long time ago, though, by Kuiper that there was a Substrate of Munda languages which apparently have a ka- prefix. I’m not too well read on the subject matter.

    Later Kuiper gave up on the Munda theory, but did write a book about non-Indo-Iranian words in the Rig Veda. Here’s a link to the book: http://books.google.com/books?id=aelomzRcHrUC&printsec=frontcover&dq=Aryans+in+the+Rig+Veda&hl=en&ei=tfmcTOfUJeKiOJW6nJ8L&sa=X&oi=book_result&ct=result&resnum=1&ved=0CCsQ6AEwAA#v=onepage&q=Aryans%20in%20the%20Rig%20Veda&f=false
    I haven’t read it myself, but I’m sure you’ll find it an interesting read. It seems like a good part of the textis accessible. (At least the part about prefixes is so: yay)

    ReplyDelete
  6. Phoenix: "Although the analysis ka-puccha-la is interesting, it doesn't explain why there is a variant form kaputsala with the same meaning."

    And yet what exactly is the source of kaputsala? I only see it described as a varia lectio in Sanskrit dictionaries with no other explanation or reference.

    "It's not a phonetically faithful variant. ccha is a result of *ske, or *voiceless stop+śe/o and never for ts"

    I wrote phonetically faithful not etymologically faithful and I meant it in the sense that the affricate cch was pronounced similar enough to /ts/ that this variant is perfectly understandable without added assumptions.

    "Especially when you get to Buddhist literature you realise that any form of etymology thought up by classical Indian etymologists is per definition extremely unreliable."

    Let's not place blame completely on the Buddhists. In fact, these sort of brilliantly imaginative, metaphorically driven alterations of religious names and concepts are common among all sorts of faiths. Notice the many Greek, Hittite or Egyptian religious names that have been reinterpreted and modified in new ways. For example, is Apollo the 'Destroyer' or is he 'Father Lion'?

    Such etymologies demand much lateral rather than linear thinking. That's also what makes them endless fun to think about.

    "It has been suggested a long time ago [...] that there was a Substrate of Munda languages which apparently have a ka- prefix."

    Yes, the only problem I'm having with that is there doesn't seem to be much in-depth info about this subject. I'm having trouble understanding why a foreign noun formant would attach to these words if it were so yet, being very satisfied that PIE *kaput is a nonsense root, I suppose my only recourse is to let the subject go until more relevant info surfaces.

    ReplyDelete
  7. Let's not place blame completely on the Buddhists. In fact, these sort of brilliantly imaginative, metaphorically driven alterations of religious names and concepts are common among all sorts of faiths. Notice the many Greek, Hittite or Egyptian religious names that have been reinterpreted and modified in new ways. For example, is Apollo the 'Destroyer' or is he 'Father Lion'?

    Yeah, religious folk etymologies are great.

    What's fascinating is that we find some commentaries of texts that will actually give SEVERAL etymologies for one word. Just different ways you could interpret these names.

    This isn't considered contradictory but rather supplementary, all these different meanings are considered to add up to the greater meaning of the name.

    I find that really interesting. I wish I could think up a good way to find one of these commentaries. It'd be fun to write a blog post about them.

    ReplyDelete
  8. I just uncovered an interesting snippet from Brugmann/Streitberg, Indogermanische Forschungen, v. 3 (1894), p.236 (see link): "Die v. l. kaput-sala- ist wohl nur eine durch Missverständniss entstandene etymologische Schreibung, indem man für die prakritisch aussehende Form mit ch ts einsetzte." (v. l. = varia lectio)

    ReplyDelete