Re: THEORY: A possible Proto-World phonology

> Date:         Thu, 29 Jun 2000 16:42:26 +0200
> From: BP Jonsson <bpj@...>
> (*As I said yesterday I adhere to the view that the whole
> [Afro-Asiatic] family originated in the Nile area, Semitic entering
> Asia from Ethiopia through Arabia. The sea-level in the Red Sea was
> lower in the Stone Age. I find it harder to believe that PS speakers
> passed through Egypt without leaving a trace before Egyptian
> established itself there, or that the family originated in the Near
> East and entered Africa through Egypt. AFAIK the recorded history of
> Mesopotamia and the archaeology of the Eastern Med seabord speak
> against that.
Well, it looks to me as if the whole Nostratic hypothesis basically
falls apart if IE and AA can't be related, since those are the two
families where the most derivations by far have been adduced. For
illustration, I'll quote a few etymologies from Illich-Svitych's "Opyt
sravnenija Nostraticheskix jazykov", Part I (b-K'), as typed in by
Miguel Carrasquer Vidal on the Nostratic list back in November 1997:

4. bal3/u/ "to swallow": AA bl3 "to swallow" ~ Alt. balgu-/bilga- "to
swallow, throat"

   AA ||  Arab. <bl3> (impf. <-bla3>) "to swallow", Geez <bl3> "to
eat", OHebr., Aram. <bl3> "to swallow" || OEgypt. <b`n.t> (with
metathesis) "bird's neck" || Cush.: Beja <bala> "throat", Chara
(S.Cush.) <borka:> "neck"  || Cf. Cohen 176, Ember 33, 45, Calice 144,
Ges. 101, Leslau Har. 41, Dolgopol'skij ASb. 54. Semantic development:
"swallow" --> "throat" --> "neck".

  Alt. || Lit.Mong. <balGu-> "to swallow"; Buryat <balga> "gulp";
Buryat <balga>, Kalmyk <balga~-> "to swallow" || Tung. *bilga: Manchu
<bilxa> "trachea"; Nanay <belga> "oesophagus"; Ulch. <bildz^a>, Orok.
<bilda>, Oroch. <bigga>, Negidal <belga>, Evenk. <bilga>, Even.
<belga> "throat" || Cf. Ramstedt KW 31, Cinc. 297, Vas. 54.  The
variant with -i- vocalism (Tungus) is likely secondary and connected
with a change in the ending of the root, cf. Alt. *siba < *sawe [*]
(Part II).

  [] Cf. Dolg. 12.  Original *-3- regularly gave Alt. *-g-.


[*] Subsequently, the author changed the reconstruction of this form,
taking *siwa as original; the author considers Ural. *s'awe secondary,
cf. #228 [Editor's note]

5. balq'a "to sparkle": AA brq "to sparkle, lightning" ~ Kartv.
berc'q'- "to sparkle, be bright" ~ IE bhelg-/bhleg- "to sparkle, be
bright" ~ Alt. [balkV- "to shine, be bright"].

   AA || Semit. *brq "to lighten": <brq> in all Sem. lgs. (Arab. impf.
<-bruq>/<-braq>, Akk. praet. <-briq>); "lightning": Arab. <barq>,
OSArab. <brq>, Shahri, Mehri <barq>, Tigre <ba"raq>, Tigrinya
<ba"rqi>, Syr. <barqo:>, Ugarit. <brq>, OHebr. <bA:rA:q>, Akk. <berqu>
|| Oegypt. <b3q> (< *brq>) "be bright" (Later Egypt. <brq> < Semit.,
see Ward JAOS 80, 323) || Cush. *m-brq "lightning" (with prefix m-;
Geez, Amh. <ma"-bra"q> likely formed under the influence of the
Cush. model): Bili <mirka:> (<mark> "to sparkle"), Hamir <mirqa:>,
Kemant <ma"rk>, Kuara <merk>, Kabenna, Kambatta <banqu-ta>, Hadiya,
Sidamo <banqo> (*m-barq-, with nasal metathesis; Sidamo <ba"la"qo>
"lightning", Mocha <p'ariqq(i)- "to sparkle" etc. < Semit., see Leslau
Moc^a 46) || Chad.: ? Hausa <wa`lk'i'ja:'> (prefixed *w-brq?) f.
"lightning"; Buduma <ba`rme'l> (Talbot), <baramil> "thunder,
lightning" (compound word); Musgu <bara> "to shine", <abera>
"thunder, lightning" || Cf. Behnk ZDMG7, 139, Ember 98, Greenb. 59,
Conti Rossini Kem. 231, Bergstr. 185, Aistl. 59-60, Leslau Soq. 97,
Leslau Har. 46, Moreno Sid. 207.  In AA -r- instead of expected *-l-
(*blq-) possibly under the influence of *br- "to shine" (Semit. *brr,
*brx, *brs., see Soden AW106).

   Kartv. *berc'q'-/*br.c'q'- || Geo. <brc'q'-in-> "to shine,
sparkle", OGeo. <na-berc'q'-al-> "spark" || Megrel. <rc'k'-in-> "to
shine" || See Kl. 50, Schmidt St. 99.  In Kartv. one can propose
secondary epenthesis of -c'- in the rare cluster *-rq'-, to turn it
into the more usual harmonic cluster -c'q'-.

   IE || OI <bha'rgas-> n. "blinding shine", <Bhr.'gavas> pl.
"mythical lightning priests" || Grk. <phle'go:> (< *bleg-, root-type
II) "I burn" || Lat. fulg- (< *bhl.g-) "to sparkle, be bright",
<fulgus> (gen. <fulgeris>; more often secondary <fulgur>) n.
"lightning" || OHG <blechazzan>, MHG <blecken> "to sparkle" || OLith.
<blingiti> (with infix) "to be bright" (against Fraenk. 48) || Toch.
AB <pa"lk-> "to burn, to light" || Cf. Pok. 124-125.  IE *bhelg-/
*bhleg- hardly connected with the root found in OI <bha:lam> "shine",
Latv. <ba~ls> "pale" (against Pok.): these forms derive from *bheh-l-,
extended from *bheh- "to shine" (see Vas. 1, 73).

   Alt. || Turk. *balky-: Gorno-Alt. (Teleut.) <malkyl> (m- instead of
b-) "clear, shiny"; Kazakh., Tatar., Karaim., OKipch. (Cum.) <balky->
"to shine"; OOghuz (Qis..), OTur. <balky-> "to be bright"
(Zaja,czkowski Kor. 79), Turkish (Edirne, see Eckmann ASAL 49)
<balkyz> "lightning" || ? Kor. <pa_lg-> "be clear, bright (we would
expect *palg-) || Cf. Ramstedt SKE 186, Radl. 4, 1499.

   ?Ural.  Interesting is Xanty (N.) <paGa~l> "lightning" (noted only
by Pa'pal-Beke 57), which might reflect Ural. *palkV.

   Cf. Tromb. 401 (AA ~ Kartv.).  Original *-l-, reconstructed on the
basis of the IE and Alt. data, regularly gave Kartv. -r- (in
compounds), and was possibly changed in AA by analogy.  In IE, the
structure of voiced aspirated with voiceless was, as usual, reshaped:
*bhelk- > *bhelg-.  Judging by IE velar *-g-, the ending of the root
had a central vowel *a; Turk. *-y- would then in all likelihood be
secondary.  The original semantics, as shown by the majority of
languages, denoted a brief outburst of bright light (lightning,
spark).

6. ? balV "blind": AA bll "blind" ~ Alt. balV "blind"

   AA || Egypt.: Copt. (Bocheir.) <belle>, (Said.) <b@lle> "blind". ||
E.Cush. "blind": Galla <ba"lla>, Sidamo <ball-ic^c^a>, <bal'-ic^c^a>,
Darasa, Burji <balla'a> (Cush. > Sem.: Harari <ba"lla>, E.Gurare
<balla>) || See Leslau JNES 21, 47, Leslau Har. 41, Moreno Sid. 207,
Moreno RStO 17, 380.

   Alt. || Mong. *bal-ai: MMong, Lit.Mong. <balai soqor> "blind"
(<soqor> "blind"), Lit.Mong. <balai> "dark, unknowing", <balai-ra->
"to become blind"; Dagur. <balie:>, Xalxa <bala~e:>, Buryat, Kalmyk
<bala":> "blind" (Mong. > Yakut <balai>) || Tung. "blind": Manchu
<balu>, Nany <bali>, Ulch. <ba:li> (possibly with metathesis of
vowel-length < *bali:), Oroch., Udei., Orok., Negidal <bali>, Evenk.
<bali:>, Even. <bali:kac^> || Cf. Ramstedt SKE 145, Ligeti AOH 14, 18,
Cinc. 296, Vas. 49.

   [] Cf. Dolgopol'skij ASb. 57-58.  Doubtful because of the rareness
of the AA forms.

7. bara "big, good": IE bher- "good, big" ~ Ural. para "good" ~ Drav.
[par- "big"] ~ Alt. [bara "much"].

   IE || Arm. <bari>, <barvok> (ar < *r.) "good" || Grk. (Homer.)
phe'r-ist-os (superl.) "best" || Alb. <mbare"> "good, happy" || Lat.
<fere:>, <ferme:> (< *ferime:, superl.) "almost; for the greatest
part" || OHG <bor->, <bora-> (<or> < *r.) "very" (pref., cf.
<bora-lang> "very long"); OSa. <bar> "very" (<bar-wirdig> "fully
worthy") || Cf. Pedersen KZ 38, 204, Muller Altit. 177.  Usually these
forms are seen as derived from *bher- "to bring, to bear" (<--
"fruit-bearing, harvest", cf. Bois. 1021, Pok. 128-131; differently
Persson Beitr. 1, 49: connects with *bher- "border, height"); In the
light of external comparison it seems more plausible to derive them
from "big" (cf. Gmc., Lat.) --> "good".

   Ural. || Finn. <paras> (superl.) "best", <parempi> (comp.) "better"
(forms based on lost *para "good") || Saam. (N.) <buore-> "good" ||
Mordv. (Moksh.) para~, (Erzy.) <paro> "good, well" || Mari (Gorn.)
<pur@^>, (Lugov.) <poro> "good, healthy" || Udmurt <bur> "right"; Komi
<bur> "good" || See SKES 490-491, Lytkin 205, Itkonen LChr. 85.

   Drav. || SDrav.: Tamil <paru> "to become big, to swell", <paruppu>
"fatness, highness"; Malayal. <paru> "big, voluminous; abscess, boil"
(<-- "swelling"); Tulu <parija> "very much".  See DED 267 ||.

   Alt. || ? Lit.Mong. <bar-da-gan>, Xalxa <barda:n> "wealth, wealthy"
|| Tung. *bara(n) "much": Manchu <baran> ("numerous"), Orok. <bara>,
Solon. <bara>, Evenk. <baran> (cf. <bara-l-> "to become big").  See
Vas. 50 ||

   The old meaning "big" is maintained in Drav., Alt. and partially in
IE.  In Ural. and IE we see a further semantic development towards
"good" (as for instance in SCr. <bo``lji:> "better" <-- "bigger").

8. bari "to take": AA br- "to seize, to catch" ~ IE bher- "to take, to
bring, to carry" ~ ? Drav. per_- "to choose, to collect" ~ Alt. bari-
"to take in the hand".

   AA || Semit.: Akk. (Babyl.) <b'r> (praet. <-ba:r>) "to catch
(fish)", <ba:'iru> "fisherman"; probably a secondary extension of
original *br, cf. Mehri, Shahri <btr> (< *t-br) "to angle for fish"
(it is likely that <3> is secondary in Sokotri <b3r> "to angle"; cf.
Leslau JAOS 82,2) || Berber: Tuareg <aber> (praet. <-uber>) "to take a
handfull"; -b- < *-bb- < *w-b-, cf. without prefixal *w, which
conditioned the doubling and retention of -b-, Tuareg <e-ha"re> (h <
*-b- with weakening as in the cases described by Beguinot RANL 33,
186-100) "possession, goods, cattle" || Cush, *brj; Beja <bari> "to
get, collect, to have, possess"; Saho (Irob, see Plaikowski-Wagner
ZDMG 103, 198) <bar.-> "to seize, hold", Afar (Tajurax, see Lucas
JSAfr 5, 198) <ber-> "to carry away" || Chad.: Ngla (Kotoko group)
<birre> "to seize" || Cf. Roessler Oriens 17, 215

   IE || OI <bha'arati>, Av. <baraiti> "carries"; OI <bha'ras>
"winnings" || Arm. <berem> "I carry, bring" || Phryg. <ab-beret>
"brings" || Grk. <phe'ro:> "I carry" (Myc. 3sg. praes. <pe-re>, see
Morpurgo 240) || Alb. <bie> (< *bhero:, cf. imper. <biere">) "I bring,
carry" || Lat. Osc. <fer-> "to carry" || OIr. <biru> "I carry" ||
Goth. <bai'ran> "to carry, to bring" || OCS <bero,> (inf. <bIrati> "I
take" || Toch. AB <pa"r-> "to brig, carry" || Cf. Pok. 128-132 (In
Pok. formations are considered from an originally different root, with
the meaning "to be born, offspring", see #32).  In the light of the
external comparisons it is seen that the meaning "to take" in Slavic,
usually considered an innovation, must be ancient.  From this meaning
developed the semantics that are more fully represented in IE: "to
bring" --> "to carry".

   ? Drav. "to choose, collect" || Sdrav. *per_ukk-: Tamil, Malayal.
<per_ukku>, Toda <per_k->, KOdagu <pori"k-> || Telugu <pedz^ipi>,
<pedz^dz^u> || C.Drav.: Kolamo <petk->, Naiki <pett->, Parji <ped->,
Gadaba (Salur) <pidz^->, Gondi (Adilabad) <per->, Konda <per_->, Kui
<pebg-> (< *peg-b-) || Kurux <pes-> || See DED 293.

   Alt || Tur. *bary-: OTur., OUigh. <barym> "possession"; Azer.
<baryn-> (Gazax) <barym-> "to make use of, to receive advantage";
OTur. (TS 2, 104) <baryn-> "to earn a living", Turkish (Edirne)
<bary-> "to care for, protect";  Cf. Hung. <barom>, OHung. <barum>
"cattle" (<-- "possessions"), borrowed from OBulg. (Gombocz BTL 40-41)
|| Mong. "to take with the hands, to seize" (and further "to offer
s.o. s.t."): MMong. <va:r->, Kalmyk <ba"r'->, Moghol <bari->; see
Poppe Mong. 26, Zirni 89 || Cf. Ramstedt KW 38 (where, as in Ram. 56,
a connection of the Mong. words is also suggested with Tur. *barNak
"finger").  In Turk., the semantic development was "to take" --> "to
get (possessions)"

   [] Cf. Ramstedt JSFOu 53(1), 23, Dolg. 12 (IE ~ Alt).  The basic
meaning "to take" is maintained in Alt., Drav. and partially IE.  In
Drav., *a > e umlaut apparently took place under influence of a front
vowel in the second syllable in the position before <r_> (as with
Drav. *e:r_- "to rise" < *Hora", #116).

9. berg/i/ "high": AA brg "high" ~ ? Kartv. br.g-e "high" ~ IE
bhergh-/bhregh- "high" ~ Ural. [p/e/r/-kV/ "high"] ~ ? Drav. pe:r_-
"high".

   AA || Berber: Tuareg <burg'@t> (aor. praet. <-bburg'@t>) "to stand
up" (suggests *w-brg-) || Cush.: Beja (Almkvist) <birga> "high", Galla
(Tutsehek) <borgi> "mountain top" || ? Chad.: Yegu <bu|lgi^t> m.
"high" ||

   ?Kartv || Geo. <brge> "grown, tall".  Formation with suff. -e.
Gam.-Mach. 99 suggest a link with Svan <b@gi> "hard".

   IE || OI. <br.h-a'nt-> "big, high", Av. <b@r@z-ant-> "high" || Arm.
<barjr> "high" || Hitt. <parkus^> "high" || OLat. <forctus> "strong"
(< *bhr.g^h-to-) || MIr. <bri:> (acc. <brig>) "hill", We. <bry> "high"
|| OIce. <bjarg>, OHG <berg> "mountain"; OE <brego> ("root type II"
*bhreg^h-) "mister" || Toch. AB <pa"rk-> "to rise" || Unclear are
forms with IE velar instead of expected palato-velar -- Alb. <Burg>
(in mountain names, see Jokl ZONF 10, 183-186), Slav. *ber~gU "bank,
height" (OCS <bre^gU>, SCr. bre~g etc.); possibly, these are
borrowings from "centum"-langugaes.  See Pok. 140-141, Vas. 1, 76.

   Ural. ||  Samoyed *pi:r (and derivations) "high": Nenets <pi"rc'a>
(<pi:r> "height"), Enets (Xantajka) <fid'e>, (Bajxa) <fize>, Nganasan
<fira>, <firaga>, Selqup (Taz) <perga"> (<pi:re> "height"), Kamas.
<pu.rz^e>, Koibal. <pric'e>, Motor. <hirge>, Taigi <hu"rke>; see
Lehtisalo MSFOu 56, 84, Castre'n Sam. 236.  The Samoyed form could
reflect Ural. *perkV or *pirkV.

   ? Drav. || S.Drav.: Kota <pe:r> "steep inclination", Toda <po":r_>
"rock, cliff" || See DED 294.  Here also, apparently, Kuvi <pe:r.h->
"to lift" (<-- "high"; less convincing the link with *pe:r_- "to load,
heap", as suggested in DED 294).

   [] Cf. Tromb. 399 (Cush ~ IE ~ Ural), Vogt NTS 9, 337 (Kartv ~ IE).
The e-vocalism in the first syllable is based on the Drav. data (cf.
also Uralic).  The length of Drav *e: probably compensates for the
simplification of the *-rk- cluster.  IE palatal *g^h suggests a front
vowel in the auslaut.