Horse argument revisited

Miguel Carrasquer Vidal mcv at WXS.NL
Sun May 10 15:32:54 EDT 1998

Vidhyanath Rao <vidynath at MATH.OHIO-STATE.EDU> wrote:

>Second: David Anthony is certain that Sredny Stog (sp?)

'Srednij' using the official Akademija Nauk transcription (where <j>
is English y).  But sredni or sredny are acceptable in English.

>culture used
>bits to control horses, based on evidence of tooth wear from a horse
>found there. However, Arrian and other Hellenistic sources clearly that
>the bit was unknown in India at the time of Alexander. Their description
>of Indian horse control mechanism indicates the use of a dropped nose
>band. This is supported by references to `nasor yama.h' [note the dual:
>nasor must mean `at the >nostrils<'] and horses bound at the nose [for
>true bits, the band(s) would be higher up.] This is known from 2nd
>millennium BCE Near East for horses and 3rd for asses/hemiones. If
>Indo-Aryans and their ancestors had unbroken association with
>domesticated horse, why did they give up the bit for the inferior
>nose band control? Without answering this question, the argument
>based on a"sva as put forth by M. Vidal  cannot be conclusive.

The argument indeed does not necessarily imply unbroken association
with _domesticated_ horses, just with horses.

>BTW, I remember Lehman saying somewhere in ``Theoretical Bases of
>Indo-European'' that `ek'uo' cannot be pre-IE and might be a loan! This
>is apparently a consequence of his views on the history of velars (or
>tectals) in IE. In particular, he gave the impression that he considers
>the sequence k'u strange. [What about the word for dog?]. I don't really
>understand `internal reconstruction', and would appreciate any comments
>of this `heresy' of Lehman. [Of course, the Greek hipppos is strange
>from the viewpoint of IE history, and the i is already there in Myc.]

The *k^u in *ek^uo- is, as you say, also found in *k^uon-, *k^un-
"dog", which is not likely to be a loanword in PIE.  The combination
of *k^ or *k with *w (as opposed to the single labio-velar *kw) is a
rare one, but it does occur in some other words as well (*kuapno-
"smoke" [Grk. kapnos, Lat. vapor, Lith kvapas], *kuatH- "to seethe,
boil, ferment" (Skt. kvath-, Lat. caseus > cheese, Russ kvas, etc.).

There is no need from "internal reconstruction" to think that PIE
*ek^wos is a borrowing, but if it is, a ready source can be found in
Proto-Yeniseian *ku?s- "horse", Proto-Dagestanian *kwaccwV "mare"
(unlike Altaic, Yeniseian and NE Caucasian *are* plausible candidates
for being descended from the Ur-inhabitants of the steppe, c.
8000-5000 BC, before the Indo-Europeans moved in).  In any case, the
word predates the domestication.

Greek <hippos> (Myc. i-qo) is curious, and must surely descend from a
form with preposed *y- (*yek^wos > *yik^wos, by vowel assimilation),
cf. Tocharian B yakwe, A yuk.

Miguel Carrasquer Vidal
mcv at

More information about the INDOLOGY mailing list