Greek question & the pre-history of *nekwt

[Patrick C. Ryan]

[ moderator re-formatted ]

Dear Rich and IEists:

-----Original Message-----
From: Patrick C. Ryan <proto-language at email.msn.com>
Date: Friday, March 05, 1999 10:14 PM

>I know of no reflex that suggests a palatal (^) + labial combination. Are you
>suggesting that I am considering g[^]h-w-? I am not. I think -gh-w- is still
>the likeliest termination.

>[ Moderator's comment:
>  Perhaps I am confused about what you mean when you write <gh-w->:  To me,
>  this suggests a segment *gh followed by a segment *w, especially when you
>  write about the latter being "carried over into the first syllable".

Yes, this is exactly what I meant: *negh(V)w-.

>  My point is that the symbol used in all Indo-Europeanist literature which is
>  not limited to ASCII has a superscript <w>, which in my TeX-influenced way
>  I would write as *g{^w}h (or less preferably *gh{^w}).

Rich, I never write anything as <X> because I frequently am switching from
email to .html and whatever is between the < and > is treated as an
instruction rather than copy in .html.

[ Moderator's comment:
  Old habit:  If square brackets indicate phonetic values, and slashes indicate
  phonemic values, then angle brackets indicate *graphemic* values, that is,
  how something is spelled in the orthography under discussion.  Much older
  than HTML, so I'm going to claim priority for it. ;-]

I have tried to indicated the labiovelars as g[w], k[w], and the palatalized
versions g[^][w], k[^][w].

[ Moderator continues:
  See above:  That is extremely confusing--either mixing phonetics in, or if
  taken as in syntax, indicating something optional.
  --rma ]

>  I'm afraid that the sloppy manner in which labiovelars get written has
>  misled you into thinking that the -u- of the Greek word _nuks, nuktos_ is
>  metathesized from after a palatal (or simple velar, if you allow three
>  series of dorsals).

Well, I might not have it right, but that is what I favor at the moment.
*negh(V)w- -> *neugh- + s/t -? neuk(h)s/t-.

[ Moderator's response:
  It may be what you favor, but it flies in the face of the data:  Sanskrit
  *requires* a labiovelar, or the accusative would be **nas.t.am rather than
  the attested _naktam_.  And what is represented by that "*s/t -?" ?
  --rma ]

>But, let me ask a question: are you saying that Hittite does *not* suggest
>that the final element before the [w], glide or extension, was voiced? That
>is a perfectly legtimate position but I was not aware it was very
>well-represented these days.

>[ Moderator's response:
>  I've not addressed this issue before.  Sturtvant himself noted a *tendency*
>  for single vs. double writing of (mostly voiceless) stops to correspond to
>  a voiced vs. voiceless distinction in the rest of Indo-European (or, as he
>  would have it, in Indo-European proper).  However, as I remember what he
>  said about Hittite _nekuz_, he considered the spelling <ku> to represent a
>  labiovelar which could not otherwise be written in cuneiform--and since it
>  thus appears before another consonant, the single/double writing tendency
>  would not be germane.

[ Moderator's comment on previous response:
  I have since looked in Sturtevant's _Comparative Grammar of Hittite_ (2nd,
  1951) and his _Indo-Hittite Laryngeals_, and find that I have mis-stated his
  views:  He clearly reads the syllable as such.  I cannot for the life of me
  remember where I learned the other interpretation.
  --rma ]

Yes, I believe that must have been Sturtevant's view with a modification. On
pg. 43 of "A Comparative Grammar of the Hittite Langauge), he indicates
Indo-Hittite *nekwts. But then on p. 59, contradictorily, he indicates
*neg'wty; and goes on to say: ". . . such forms as nukha . nuktor (Hesych.),
ennukhos 'of night', pannukhios, autonnnukhi 'in the same night', whose
aspirate proves that the second consonant of the IH word was g' ".

On the previous page, he defines IH g' as "IE gh, g[^]h, or the velar part
of ghw".

Finally, on page 181, Sturtevant lists "ne-ku-uz . .  .  .  .  .  neguts",
which is the "Suggested phonetic interpretation".

Now, this all suggests to me that Sturtevant believed (at least as of the
writing of this book) that the IE stem was *negh(V)w-.

[ Moderator's response:
  You are correct, this does appear to be what he believed.  However, I think
  he was wrong.  The argument I first cited was not Sturtevant's, but it was
  nevertheless the right interpretation, based on readings of other lexical
  items such as the interrogatives.
  --rma ]

Pat