Re: [tied] Demonstratives

>> With such good argumentation for reconstructing *-ih2, we may

compare IE possessive adjectives (and particioles) ending in m. *-to-s vs. f. *-to-ih2 > *-ta: with Old Mongolic m. -tu vs. f. -tai (e.g. mori-tu "(man) who has a horse", mori-tai "(woman) who has a horse") !!!

> Greenberg reconstructs a diminutive K (Eurasiatic, p.
>

164-166), which I would analyze as *-iq (> PIE *-ih2/*-ik-).

> He also

gives examples from Mongolian, where this

> diminutive, as in IE, has

developed into a feminine. "An

> example of feminine use is noya(n)

"prince", noy-ika(n)

> "princess".

The thing Grinberg is speaking about is very different. Noyan vs. noyiqan is nothing but a lexical derivation that exists everywhere without any necessary connection with the existence of grammatical gender. But in Mongolian (only in its oldest texts) we have sometimes gender inflection of adjectives and verbs. This feature is something very uncommon for Altaic; in fact, it exists nowhere in Altaic except for Old Mongolic. But in my opinion, there is a possibility that the gender (inclueding the feminine) can be traced up to proto-Nostratic, having its reflexes in IE, Mongolic, and maybe also in Afrasian and Dravidian (although now I can find traces of common marking only in IE and Mongolic.).

> Hitt. we:s and Gothic we:s are from *wey-es.

Before accepting, it needs to be determinated whether intervocaloic *-y- was lost in Hitt. and Goth. I don't know about Gothic, but Hittite doesn't seem to have such a soundlaw.

>> Really, initial *mW- is very unprobable for Uralic (and

also Altaic), because these languages do not allow initial clusters at all. It is still possible to believe in *tW, understanding it as a single phoneme: such labialized <t> exists, e.g., in Abkhaz. But <m>, being labial itself, nowhere can have a additional feature of labialization!

>

Ladefoged & Maddieson, "The sounds of the world's

>

languages", p. 356:

> Labialization is the most widely found

secondary consonantal

> articulation, both with respect to the number of

different

> types of segments with which it co-occurs, and the number

> lanmguages in which it is found. It is especially

common

> with velar obstruents and, relative to their frequency,

with

> uvulars. Many languages, including such varied ones

> Amharic, Wantoat and Guaraní, permit labialization only

> such back consonants [] Other languages, including certain
>

Australian and Caucasian languages, permit labialization of

> a much wider

range of consonants, including those whose

> primary place of articulation

is labial. Examples from

> Arrernte are given in table

10.10

> Table 10.10 shows the labialized phonemes of

Arrernte:

>

labial dental alv. postalv. retrofl. velar

>

stop pW t_W tW t^W t.W kW

>

nasal mW nW n^W n.W NW

> prestop.

pmW tn_W tnW tn.W kNW

> prenas.

mpW nt_W ntW nt^W nt.W NkW

>

lat. l_W lW l^W l.W

>

rhotic rW jW r.W

Your example is from an Australian language. Also languages of the Caucasus are mentianed. Which of them do possesse this? It is important, because it is more probable that proto-IE shared certain features with some near languages, and not with lying very far...

>>> Hittite zi:k comes from *ti:-g (Palaic ti), and the

acc. tuk

>>> (Palaic tu), as well as all the other oblique forms

(tuel,

>>> tuedaz etc.) show that the pronominal root was *tu(:), as

>>> is everwywhere in IE. Hittite/Palaic -i: in the

nominative

>>> is a special development of *-u: > *-y: > -i:

in the Auslaut

>>> (there is an exact duplicate of that soundlaw in

Albanian).

>>> Hitt. zi- comes from *ti- not from

*si-.

>>
>>As long as I know, the development *-u: > *-y: > *-i: does not exist in Hittite. Do you have more examples on

it?

> The rule is confirmed by H. Craig Melchert,

"Anatolian

> Hitorical Phonology", p. 84. There are

no other cases of

> -u:, so the rule is, as Melchert says,

"unverifiable". It

> is, however, the only logical

explanation of Hitt. zi:- and

> Palaic ti. Hitt. zi comes from

*ti-. Hittite and Palaic ti

> (acc. tu) must be connected with PIE

*tu(:).

So, the hypothesis is a week one. Note that in Hittite verbal 2nd person singular really both -si and -ti exist (and the latter's consonant does not change into /z/). It seems very probable that both of these endings could have their source in two different stems of the corresponding pronoun.

In fact, the question on Hittite palatalization *ti > zi is not so clear as it is commonly assumed. Sometimes it no doubt exists, but there are examples where the palatalization is not found. Note also that verbal 2Sg. -t and 3Sg. -s exist in Tocharian. Perhaps even Germanic sometimes demonstrates the same: cf. English 3Sg. -s vs. archaic -th (and German -t). Also Germanis 2Sg. -st can be a result of 2Sg. *-s and 2Sg. *-t contamination. These Anatolian, Tocharian and Germanic facts allow to assume that in PIE verbs both 2Sg. -s vs. 3Sg. -t and 2Sg. -t vs. 3Sg. -s were possible.

> The 1st. person marker *m(a)

is unproblematical from a

> Nostratic viewpoint. It is used in the

independent pronoun

> and in the verbal object marker. The verbal

subject is

> marked by *xw- (< *kw-?), which is perhaps relatable to

the

> 1sg. form of the PNostratic copula/stative ending "I

am"

> (*?a-ku, Sem. -(a:)ku, PIE *-h2(a), Uralic -k, Esk-Al.
>

-k-/-ng, Turk. 1pl. -k?). The possessive and the plural are

> then

also based on *kw: *kw-i-m "my" > *c^wim, *kwin- "our"

> (originally possessive only) > *c^win-.

>

If we assume the 2nd. person markers are structured

> similarly, the

independent 2sg. pronoun and the 2nd. person

> object marker must have a

common origin, which can only be

> PN *ki "you", giving *ki >

*si in the independent pronoun,

> *-kV- > *-g- as the verbal object

marker.

> The 2nd. person form of the copula was *ta-kV "you

are"

>(Sem. -ka/-ki ~ -ta/-ti, PIE *-th2(a), Uralic -n (<

*-ng),

> Altaic -ng (< *-nk < *-tk)). This gives the

Kartvelian 2nd.

> person subject marker *x-, as well as the possessive

forms

> *tkw-i-n > *s^win- (OGeo. s^en-, Zan skan-/ckan-, Svan

sg.

> poss. isgwi, pl. sgäj, isgwe:j) and plural

*tkw-an.

The problem of a guttural stop and a dental/guttural nasal as Nostratic personal markers is a difficult one. On the one hand, we have:

1pp. guttural:

IE *e-g^(h)o(m) "I",; *-oH/*-Hi (verbal 1Sg.); *k^i-/*k^e-/*k^o- "this near me"; Lat. hic < *ghei-ke;

Uralic *-k; ? *-kwV (verbal 1Pl., attested only in Ob-Ugric);

Turkic *-uk (verbal 1Pl.);

Kartvelian *gw- (verbal 1Pl.);

Korean ky "I".

2pp. nasal:

Uralic *nV "you", *-n/*-ntV (both attested only in Permic, Ob-Ugric and Samoyedic);

Turkic *-ng;

Japanese, Korean *nV "you".

On the other hand:

1pp. nasal:

IE *nos "we" (in Latin and Slavic);

Uralic: Mordvin -n (kunda-ta-n "I katch you "; elsewhere in Mordvin 1pp. -m);

Yukaghir -ng;

Turkic: Turkmen -ng (verbal 1Sg.);

Mongolic na- "me";

? Old Japanese a < *nga

2pp. guttural:

Uralic: Mordvin -k (kunda-sa-ma-k "you katch me"; elsewhere in Mordvin 2pp. -t/-d);

Yukaghir -k;

Turkic -g (attested only in Old Turkic, Yakut and Chuvash);

Kartvelian *g-.

So, I think, there were also some deictic stems with uncertain personal meaning based on guttural stop and dental/guttural nasal in proto-Nostratic. But what was their exact meaning? Were they also personal?

> The phoneme *tW must have been rare. I see the same in

PIE,

> where *tW > *sW is mostly found in morphological items
>

(plural, 2nd. person, demonstrative), and in only a handful

> of lexical

items.

It is a very strange situation when certain sounds are more used in suffixes than in roots. When more used in roots, it is common, but more in grammatical items? Just unprobable.

===========

Vadim Ponaryadov