On Sat, 10 Mar 2001 22:19:00 +0100, "Piotr Gasiorowski"
<
gpiotr@...> wrote:
>>> The form *pe:r is often thought to reflect *per-r (thus in the EIEC) or *per-n with compensatory length, though is could equally well represent the bare root *per-.
>
>> But why shouldn't the -n be part of the root?
>
>Oblique forms with a nasal extension to n-less Nom.sgs are not unknown, e.g. OInd. drunás (beside droh. < *drous), Gk. dóratos ~ doúratos, as if from *dórw-n-tos (beside dourós < *dorwós), Hit. sius, Gen. siunas.
Also Hitt. haras, haranas "eagle"; memiyas, memiyanas "word, thing",
etc., and other examples like the OInd. i/n-stems. I think that, at
least in most of these cases, the -n- was originally part of the root,
and was lost in the nominative in various ways (**-ni > *-n^ > *-i;
maybe **-nu > *-{nw} > *-u in the "tree"-words above; possibly regular
loss of -n- in the cluster -ns for the Hitt. examples, cf. acc.pl. -us
< *-ns).
>The reconstruction *per-r is a little suspect as far as I'm concerned (I'd expect a different vowel pattern in a Root-r/-(V)n- heterocliton)
Something like **pé:r(r), **prén(o)s (like *yé:kwr, *y[e]kwén[o]s) or
**pó:r, **pérn(o)s (like *wódr, *wédn[o]s)? The pattern *pe:rr,
*prnós looks indeed hysterodynamic, which is strange for a neuter and
a heteroclitic.
>and as for *per-n, I'm not convinced such a form could underlie Hit. pir (any similar examples?).
kir < *kerd is somewhat similar.
>Anyway, I'm not saying that the *-n can't have been part of the original stem; I just admit the possibility that it wasn't.
OK.
>> The change l > r (in this case, pVl > pVr) in Egyptian is to be dated to the beginning of the Middle Kingdom, ca. 2000 BC., a bit too late.
>
>I am aware that according to current revisions Egyptian <3> was originally a rhotic (uvular [R] or emphatically trilled [rr]) according to one camp of Egyptologists, or some kind of lateral according to the other camp; but the interpretation of <r> as [l] is new to me. What is your authority for that (and for the date)? What is the evidence for a lateral rather than, say, a tapped [r]?
The date is from Kammerzell "Zur Umschreibung und Lautung", in the
introduction to Hannig's "Grosses Handwörterbuch". That the phoneme
transcribed as <r> was originally /l/ (c.q. comes from PAA *l) is the
unanimous opinion of Kammerzell, Loprieno ("Ancient Egyptian") and
Schenkel ("Einführung in die altägyptische Sprachwissenschaft"). To
be sure, some *l's surface in Egyptian as <n>, <3> or <j>, but <r> is
always from *l.
What I gather from Kammerzell, Schenkel and Loprieno is that the
relation between Egyptian and PAA consonants is the following:
Eg. PAA
<3> *r (occasionally *l)
<j> *y, *?, palatalized *l, *g and *G
<y> *w
<`> *d, *dz^, *dz, *dL; also *3 [?]
<w> *w
<b> *b
<p> *p (occasionally *t, by coronal dissimilation)
<f> *p. (occasionally *h, *s^)
<m> *m
<n> *n (occasionally *l)
<h> possibly *h (no good etymologies available)
<h.> *x.
<x> *G
<h_> *x
<z> *s, *c (occasionally *c^)
<s> *(t)L, *s^, *c^
<s^> palatalized *x (occasionally *s^)
<q> *k.
<k> *k
<g> *g
<t> *t
<d> *t., *c., *c^., *tL. (occasionally *d, near *r)
<t_> palatalized *k
<d_> palatalized *k., *x.
From PAA -> Eg.:
*b b
*p p
*p. f
*d `
*t t
*t. d
*dz `
*c z
*c. d
*s z
*dz^ `
*c^ z/s
*c^. d
*s^ s
*dL `
*tL s
*tL. d
*L s
*g g, j
*k k, t_
*k. q, d_
*G x, j
*x h_,t_
*x. h.,d_
*3 `
*h h
*? j
*m m
*n n
*l r/n/3, j
*r 3
*w w, y
*y j
=======================
Miguel Carrasquer Vidal
mcv@...