cat < ?

[Hans-Werner Hatting]

Dear List members,

I think Kaystytis Beitas has made a valid point in suggesting a further
source for the name of the cat.
But there are problems with some of the links he proposes:

1. KB wrote:
>For example: 1. Latin  feles 'cat' reminds of Lithuanian pele 'mouse'. In
>Watkins Dictionary of Indo-European Roots pel- means "to thrust, to
>strike".

>>I suppose Hindi bhili "cat" would be too fortuitous to suggest a
>>connection based on *bhil-, *bhel-?

>2. Hindi bhili 'cat' <-> Russian bilo 'thing for beating', Eng. beat > etc.
>(maybe this bhili is cognate to Lithuanian pele)

I don't know the origin of Hindi *bhili. Based on what KB suggests, a link
to the root behind Engl. bite, Gm. beissen etc. might be possible.
Certainly it is not linked to Lith. pele, as Indian /bh/ corresponds to
Baltic /b/ (if we don't pull out the magic wand of taboo changes, of
course). The vocalism (Hindi /i/ - Latin /e/ should not correspond)makes it
also problematic to link bhili to feles. If we want to find a cognate for
"pele", we would have to look for something like *pel/pol- in Latin and like
*pal/par/p.r- in Indo-Aryan.

2. KB:
>English mouse and German Maus are similar to Russian musor "debris;
>litter". Chernykh states that musor is cognate with Russian musolit'
>"slabber, slaver" and both them are originated from IE *meu-, *mou- >"damp,
>moist" and "liquid dirt; mud". But this cognateness is between >musor and
>mud, moist is more distant in my opinion: musor is >originated as "litter,
>produced by chapping (or by mice?)"...

As KB himself notes, the Gmc. mouse words are related to Slavic mysh', going
back to PIE *muHs- or sim. As PIE /s/ became /sh/ in Slavic after /u/, /i/,
/r/, /k/, "musor" cannot go back to PIE *muHs-; the /s/ goes either back to
PIE /k'/, or the word is a compund of the root quoted by Chernykhov and the
root contained in Russian _ssorit'sja_ "quarrel". I don't have an
etymological dictionary here to check that question.

3. KB:
>So there is some basis to state that this all-Indo-European word cat >in
>all its variations may by descendant of some Indo-European root >with
>meaning "to hit, to strike, to make hole etc". Distant cognate of >this
>hypothetical (?) root may be Watkins's kat- "to fight" and kat- >"down".

<snip>

>Or cut in Chambers Dictionary of Etymology: ,,Probably before 1300 either
>as: cutten <...>, kitten <...>; of >uncertain orrigin (possibly borrowed
>from  Scandinavian source; >compare Swedish dialect kuta, kata "to cut",
>kuta "knife", and >Icelandic kuti "knife". ''

Well, "cut" cannot be a Gmc. cognate of a PIE *kat-, as PIE */k/ gives /h/
in Gmc. in Anlaut position, and /t/ should give /T/ or /D/. The same reason
speaks against deriving the Gmc. cat words from such a PIE root.

One aside - are there any data available for the first recordings of the
_katto- / gatto- _ - word? As already has been stated by Rick Mc Callister
at the start of this thread, the words look too much alike to be of PIE
origin, and I would add that the k/g variation in Greek, as well as between
individual Romance languages (e.g., It. gatto vs. French chat) also are
arguments against an inheritance from PIE. My preference is to see the word
as a loan from a non-IE source, maybe on such a route: (Source language) >
Greek > Romance, Celtic, Gmc, Slavic etc.

4. KB:
>So distant relativeof English cat or Lithuanian kate "male cat" or >Russian
>kot "male cat" may be English kettle, Lithuanian katilas >"kettle": In
>Chambers Dict.of Etym.: ,,kettle -- <...> borrowed directly from Latin
>catillus "small bowl, >dish or plate", diminutive of catinus "bowl, dish,
>pot"; perhaps >cognate with Greek kotyle "small vessel, cup" <...>. ''

Concerning the impossibility of Gmc. /k/ being from PIE /k/ (except if one
uses that other magic wand, a IE substrate language influence), see above.
Lith. "katilas" is most probably a loan from Latin, via Slavic (v. Russian
"kotel", Common Slavic *koti0lu0-) (i0, u0 denoting the front and back yers,
respectively).

5. KB:
>This excerpt from Chambers Dict.of Etym. reminds on my old posting (Lith.
>peilis "knife" <-> Lith. pele "mous"):

>>The similar case is with Lithuanian "peilis" 'knife'. It is similar >>to
>>Russian "pila" 'saw', Lat. "pilum" 'heavy javelin, pestle', OHG "pfil"
>>'arrow, stake'. In this context OE "pil" 'stake, shaft, spike' and Eng
>>"pile" 'arrow, >>dart' >may be not borrowings as it is stated in Chambers
>>Dict. of >>Etym. ( p.794)] but words of common Indo-European origin.

The vocalism speaks against pele and peilis being cognates - there is no PIE
ablaut pattern e-ei. The link peilis - pilum - pila looks possible. Because
of the initial p/pf the English and OHG words have to be loans from Latin.

I hope I did not spoil too much fun by insisting on the observance of sound
laws. But I think if one disregards them, one should give good reasons for
doing so, otherwise we'll very soon arrive again in a state where "in
etymology, vowels mean nothing, and consonants very little".

Best regards,
Hans-Werner Hatting