Search This Blog

Wednesday, 26 April 2017

Indo-European Connections

So lets proceed from where we left  :) . 


Kyriakos Samelis said...

Nirjhar, do you think the etymology of Skrt. nágara "city, town" is this one > (from न (na, “on”) +‎ गर (gara, “mountain, top”)?

Mayrhofer says that nágara means also "dwelling" or "temple" and gives also Tamil nakar >

As far I understand, nakar (as a verb) means also "crawl, creep, move around". Also, another word nakra in Sanskrit means "alligator" or "crocodile"; tjhat reminds the IE root for "crawl, snake", but is that so?

Nirjhar007 said...

Hi Kyriakos ,

I believed its from the sense 'not to go' i.e. settled . But I am not sure how to suggest it technically .

I don't know on mountain top can make sense :) , first towns were beside rivers not mountain tops! ;) .

'gara ' has two senses one 'swallowing' and another is to 'fluid,water' which can be connected IMO ( . Giacomo IIRC suggested from nRgara(नृगर)'human swallowing or swallowed by humans' in sense of seizing . But that word is unattested though makes sense.

Its not sure the etymology . Another can be with PIE *gher 'enclose,gird' :

So girded by humans (Surrounded ,Protected,Fenced)) can also make sense .

The Dravidian word should be a loan :

Giacomo Benedetti said...

Hi, just a correction: I do not propose 'human swallowing' but rather 'gathering of men', froṛm the same root as Greek agora, with Prakritic passage from resonant ṛ to a.
Now thanks to wiktionary I discover that also in Hebrew there is a similar root!

Nirjhar007 said...

Thanks Giacomo, thats why I wrote IIRC :) .

But since nR is not probable although etymology is a play of conjectures mostly ;) . Perhaps a Kentum form from this root :

Attested in Sanskrit as Carati{चरति} ( Root Car and also Cal):

So 'not moving' , Giacomo?.

Nirjhar007 said...

^Provable instead of probable.

Giacomo Benedetti said...

No, that is not possible, because of the voiced sound (and also semantically it is not strong). BTW, the etymology I have given is not mine, is cited by Mayrhofer and I find it convincing. It is also thinkable from nara-gara the second ra was lost for simplification, but since nagara is not a Vedic word a Prakritic form is likely.

Nirjhar007 said...

See also this :


agāra 52 agāra n. ʻ house ʼ ĀśvGr̥., āgāra -- Mn. [Prob. ← Drav., Mayrhofer EWA i 17 with lit.]
Pa. ā̆gāra -- , °aka -- n., Pk. āgāra -- , ag°, gāra -- n.
agnyagārá -- , *indrāgāra -- , *upavasathāgāra -- , *kapālāgāra -- , *kīṭikāgāra -- , *kulāgāra -- , kūṭāgāra -- , kōśāgāra -- , kōṣṭhāgāra -- , *kōlhuvagāra -- , *khaṇḍitāgāra -- , garbhāgāra -- , *ghōṭāgāra -- , *bhakṣāgāra -- , bhāṇḍāgāra -- , *bhūmiyagāra -- , *bhraṣṭrāgāra -- , *madhyāgāra -- , *mahāgāra -- , *mahānasāgāra -- , *rājakulāgāra -- , vāsāgāra -- , *sabhyāgāra -- ''

Nirjhar007 said...

Okay Giacomo , I sense that naragara -> nagara is a valid possibility , yes .

Here the PIE root :

Kyriakos Samelis said...

I was thinking about carati (< kwel/ *kwer) when we were talking about Sumerian ĝiri [FOOT] (10822x: Lagash II, Ur III) wr. ĝiri3; me-ri; ĝiri16 "via, by means of, under the authority of someone; foot; path" Akk. šēpu.

According to Pokorny: Old Indian cárati, calati `bewegt sich, wandert, weidet, treibt' etc. (heavy basis in cáritum, caritá-, cīrṇá-; carítra- n. `foot, leg'.

Kyriakos Samelis said...

There is also a Ge'ez word əgr 'foot' (plural: ā'gar).
I don't know if there is a Nostratic etymology for this.

About Nirjhar's proposal, I think that *kwel fits good for a word for "town", since we have wors like Lat. "incolo", "inquilinus" etc, words we have been talking about at Giacomo's posts. I think also that nagara's etymology according to Mayrhofer is not a solved issue.

I'm trying mostly to find out if Sumerian niĝir, wr. niĝir; li-bi-ir "herald" Akk. nagiru "(town) crier, herald" is related to Skt. nagara "town" (or Tamil nakar). If the etymology of niĝir has to do with a notion "cry" then they cannot be related; but if we think of a meaning like "messanger" (which includes notions like moving, feet etc) perhaps they do. The na- part is not easy to explain though.

Kyriakos Samelis said...

The semitic etymology of the Ge'ez word for foot is this one (from Starling):

Proto-Semitic: *ʔi(n)gi/ur-
Meaning: foot
Modern Arabic: SYR ʔižǝr [Maʕṣarānī-Segal 24], ʔǝǧǝr [Noeldeke ZA 20:414], ʔiǧr [Dozy 1 11]; DAT_ ʔiǧr do. [GD 63] (relationship to *riǧl- is doubted [ibid.])
Geʕez (Ethiopian): ʔǝgr, pl. ʔǝgar (also 'measure') [LGz 11]
Tigre: ʔǝgǝr [LH 386]
Tigrai (Tigriñña): ʔǝgri [Bass 568]
Amharic: ǝgǝr (also 'leg') [K 1325], ARG ingir [LGaf 173] do.
Gafat: ǝgʷrä [ibid. 173]
Harari: igir, ingir [LHar 28]
East Ethiopic: SEL WOL ZWY ǝngǝr (also 'leg') [LGur 26]

Nirjhar007 said...

Thanks Kyriakos :) . Its perhaps also possible that Indic word is a loan from Sumerian ! :) .

Aram said...

Was the case of thunder god Teshub/ Tushpa already discussed?
I mean comparing it with IE Deus Pater?

Giacomo Benedetti said...

About nagara, I had another idea, that I see was already present in Monier-Williams, when he writes: "prob. not fr. naga + ra". Now, naga means mountain, probably from na-ga, not going, immovable. -ra is a common suffix (see dhīra, vipra, mudra, medhira...)
In Mayrhofer, it is said that nagara in Pāli means 'fortress, fortified city', and in Khotanese noγor means 'fortress, castle'. So, it seems that originally it meant a fortress, maybe a hill fort like in Rajasthan or Madhya Pradesh, and we can compare with German Burg itself, that is related with the root of Berg 'mountain' and 'tower':
There the idea is highness, here would be stability, like in dṛḍha 'firm; stronghold' or the simple connection with mountains.

Nirjhar007 said...


I think yes, I don't think many topics survived from the etymology storm we create ;) .


I think , still don't like the mountain connection . But as I suspected 'not moving' 'not going' i.e. settled of course is a good possibility.

Kyriakos Samelis said...


I remember also an Akkadian word igāru = "wall", perhaps that fits also to the notion of "dwelling". Look also here:

Kyriakos Samelis said...

I think we have to deal again with many similar roots. Bear in mind that in case of a Semitic word, an initial ni (or ne- or- na-) can indicate an instruction which in Hebrew is called the Niph'al construction; for example, in the Hebrew word Giacomo has posted, the Niph'al construction is this one:

It's like the passive or middle voice in Greek (actually a bit more complicated than that); Here is the chapter from Gesenius' Hebrew Grammar:

Kyriakos Samelis said...

Similar roots, for example, are *kʷel / *kʷer (with incolo, carati etc); *gher for "encircle; *ger-1, *gere- "to gather, put together" (with agora, grex etc); *kʷel-3 "swarm, flock, shoal, school, clan, herd, crowd".

About Sum. niĝir "herald" (as "crier") there is also a comparable root *ger-2 "to scream"; Pokorny gives here Skt. járatē `it rushes, sounds, crackles, shouts', jarā `the rustling, murmuring'; or to *ĝā̆r- or *gʷer- according to Pokorny - modern reconstruction for this one gʷerH-:

Also there is Sk. gir, which Mayrhofer gives to this same root:

Akkadian nagiru could perhaps explain the initial na- as a semitic grammatic element. The labiovelar of this root *gʷer- could explain perhaps more easily the assimilated (Emesal) form libir of "niĝir".

I think Bomhard gives all the nostratic forms of these roots (usually with a "k' " instead of a "g", according to the glottalic theory).

Kyriakos Samelis said...

or perhaps ni- could be a Skt. grammatic element? For example like in here:

btw, the "nr-gara" reminded me Gr. μανδραγόρας mandragoras "mandrake", a human-like root, which is supposed to be of Iranian origin; could be from "man" and "swallow"? (not necessarily meaning "eaten" but something like "hidden in the ground").

Kyriakos Samelis said...

About the meaning "fortress, fortified city" for nagara, in Halloran's Lexicon there is a word ĝiri16, ĝir16 [GÌR×KÁR]: fortress, refuge.
Also ĝiri5, ĝir5[KAŠ4]: n., refugee, stranger. v., to run, trot; to seek refuge.
I haven't find them in ePSD though.

Nirjhar007 said...

Kyriakos ,

From Bomhard I have found another candidate but again not the explanation of na- :

Proto-Nostratic root *k’w¦ar- (~ *k’w¦ǝr-):
(vb.) *kw’¦ar- ‘to rest, to stay, to remain’;
(n.) *k’w¦ar-a ‘stillness, quietude, repose, rest, resting place’; (adj.) ‘still, quiet,
at rest’
A. Proto-Afrasian (?) *kw’¦ar- ‘to stay, to remain, to rest, to settle down’:
Proto-Semitic *kw’ar-ar- ‘to stay, to remain, to rest, to settle down’ >
Moabite ḳr ‘town’; Ugaritic ḳr ‘dweller (?), dwelling (?)’; Arabic ḳarra ‘to
settle down, to establish oneself, to become settled or sedentary, to take up one’s residence, to rest, to abide, to dwell, to reside, to remain, to stay, to linger’, maḳarr ‘abode, dwelling, habitation; residence; storage place; seat, center; site, place; station; position (at sea)’, ḳarār ‘fixedness, firmness, solidity; sedentariness, settledness, stationariness, sedentation; steadiness, constancy, continuance, permanency, stability; repose, rest, stillness, quietude; duration; abode, dwelling, habitation; residence, resting place’; Sabaean ḳrr ‘settlement’. Zammit 2002:337. Proto-Semitic *k’ar-ay- ‘(vb.)to stay, to remain, to settle down; (n.) town, village, settled area’ > Arabic
ḳarya ‘village, hamlet, small town, rural community’; Hebrew ḳiryāh
[hy*r+q]! ‘town, city’; Palmyrene ḳry ‘settled area’; Tigrinya ḳäräyä ‘to
remain’; Amharic ḳärrä ‘to be left, to remain, to be missing, to be absent;
to stay away, to absent oneself’; Gurage (Muher) ḳärrä, (Gogot) ḳerrä,
(Soddo) ḳirrä ‘to be absent, to stay away, to remain behind, to disappear,
to vanish, to be lost’. Murtonen 1989:385; Klein 1987:593—594; Leslau
1979:494. Berber: Tuareg aɣrəm ‘town, village’, taɣrəmt ‘a small village,
small castle’; Mzab aɣrəm ‘city, town, village, town surrounded by
ramparts’; Nefusa aɣrəm ‘town’; Zenaga irmi ‘village, settlement’;
Tamazight iɣrəm ‘village, fortified village, granary’, tiɣrəmt ‘fortified
house’. Cushitic: Highland East Cushitic: Hadiyya k’arar- ‘to settle (out)’;
Kambata k’arar- ‘to settle (out)’. Hudson 1989:288 and 328.
B. (?) Dravidian: Kannaḍa kūr ‘to sit down’, kūrisu ‘to cause to sit’; Telugu
kūr(u)cuṇḍu ‘to sit, to be seated’; Pengo kuc- ‘to sit’; Manḍa kuh- ‘to sit’.
Burrow—Emeneau 1984:173, no. 1900.
C. Proto-Indo-European *kw’¦er-/*kw’¦or-/*kw’¦r̥ - ‘gentle, mild, calm, at rest,
still’: Gothic qairrus ‘friendly, gentle’, qairrei ‘gentleness’; Old Icelandic
kvirr, kyrr ‘still, quiet, at rest’, kyrra (f.) ‘calmness, calm’, kyrra ‘to calm,
to still; to become calm’; Faroese kyrrur ‘still, quiet’, kyrra ‘to
domesticate, to tame’; Norwegian kyrr, kjørr, kvar, kver ‘still, quiet’,
kjørra ‘to domesticate, to tame’; Swedish kvar ‘still, quiet’; Danish kvKr
‘still, quiet’; Middle Low German querre ‘tame’; Middle High German
kürre ‘tame, docile, gentle, mild’ (New High German kirre). Mann 1984—
1987:357 *gu̯ersos (*gu̯ors-, *gu̯r̥s-) ‘sweet, soft, pleasant’; Orël 2003:229
Proto-Germanic *kwerruz; Kroonen 2013:318 Proto-Germanic *kwerru-
‘quiet, still’; Feist 1939:386; Lehmann 1986:275; De Vries 1977:341;
Falk—Torp 1903—1906.I:434 Germanic base *kwerru-; Kluge—Mitzka
1967:371 *gßersu-; Kluge—Seebold 1989:371 Proto-Germanic kwerru-
‘quiet, tame’.

The Afrasian elements are close.

Nirjhar007 said...

From Starling :

Proto-Afro-Asiatic: *ḳʷar-
Meaning: block of houses, settlement, town
Semitic: *ḳʷary- 'town'
Berber: *a-ɣaram 'town, settlement'
Western Chadic: *kwaru- 'hut' 1, 'foundation of a house' 2, 'abode, world, life' 3
Central Chadic: *kwakwar- 'world, region'
East Chadic: *kwVr- 'place' (?)
Low East Cushitic: *ḳor- 'block'
South Cushitic: *ḳor- 'brick house'
Omotic: *ḳer- 'house, dwelling'

There is also of course :

Proto-Afro-Asiatic: *ʔigar- ~ *ga/iʔur-
Meaning: wall, house, dwelling
Borean etymology: Borean etymology
Semitic: *ʔigār- 'wall' 1, 'roof' 2 ~ *gʷar- 'back of the house'
Berber: *gVrur- 'enclosure'1, 'wall'2, 'place, yard' 3
Egyptian: d_rw 'part of house' (MK), d_ry 'wall', d_ry.t 'dwelling' (gr)
Western Chadic: *gar- 'village, town' 1, 'stone wall' 2, 'town-wall' 3, 'low wall or mount' 4, 'corn bin' 5
Central Chadic: *gaHur- 'enclosure' 1, 'shed' 2, 'corral' 3
East Chadic: *giHVr- 'hut' 1, ' house' 2, 'compound' 3, 'village' 4, 'dwelling place' 5
Beḍauye (Beja): gaʔra, gaarʔa 'yard'
Low East Cushitic: *guʔur- 'house' 1, 'wall' 2
High East Cushitic: *goʔr- 'shed'
South Cushitic: *garVʔ- 'wall of the verandah'
Omotic: *gVHol- 'house' - cf.

Kyriakos Samelis said...

I think a root *kw’¦ar- ‘to stay, to remain, to rest, to settle down’ fits good also for Anatolian "pir" and "parna" for "house" (par- < kwar, p < kw like in Greek (Aeolic form) or mayber Celtic. Hittite parna "house" has also a final -na (instead of an initial one).
Maybe "na-" could mean just "in"? (like in-colo; kwar and kwal must share a common origin). If nagara has a meaning of fortress (or perhaps as refugee), maybe there was a notion of "move in" (f.e. to find shelter, to escape, to live in safety etc). Well, I'm not sure about that.

Giacomo Benedetti said...

I find very interesting Berber *a-ɣaram 'town, settlement': Tuareg aɣrəm ‘town, village’, taɣrəmt ‘a small village, small castle’; Mzab aɣrəm ‘city, town, village, town surrounded by ramparts’; Nefusa aɣrəm ‘town’; Zenaga irmi ‘village, settlement’; Tamazight iɣrəm ‘village, fortified village, granary’, tiɣrəmt ‘fortified
Because it is very close to Skt. grāma. But grāma means also a troop, a nomadic group, and originally indicates a group (also the gamut of notes!) rather than a settlement. In Middle Persian, grāmag means 'wealth, possession', in Baluchi grām is 'burden', in Russian gromada is a 'big heap', in Polish 'crowd, heap, village community'. So, the root is that of a collection, a group from *gr- 'to collect, gather'.
In Hebrew, we have already seen a root 'gr 'to hoard, store, gather' that corresponds very closely:

Then, we have Skt./Prakrit āgāra 'apartment, dwelling, house', that semantically seems quite close to the PAA root above.

About nagara, the connection is difficult, because if na- does not come from nṛ/nara 'man', it is not explained. However, now I think that a reference to men is not likely because every settlement concerns men first of all. The Turner entry is interesting:
I have already remarked the Pali and Khotanese meaning, here we can observe that in Oriya, na(h)ara means 'palace, king's harem'. In the compounds, there is antarnagara 'the palace of a king' (in Ramayana). We can add Tamil nakar 'house, abode, mansion, temple, palace, town, city'. So, it is clear that it meant also a palace or abode and not only town. It seems that the only IE etymology can be from naga 'not moving' (cp. also aga). The idea coud really be that of a particularly stable building like a palace or fort, or settlement like a fortified town.

Kyriakos Samelis said...

I think that the roots *kwel/*kwer and *ger are somehow connected (and this is not only inside the IE family), perhaps due to some kind of lenition (Giacomo has compared already Sumerian gigir "chariot" to the IE root of "circle", a reduplicated *kwel). For example, the meaning of "troop" could lead us to Gr. telos (< *kwel); in Sumerian, "tilla" (supposing it is from *kwil) means "street, town square, also market-place", according to J. Halloran (like in Gr. agora "marketplace" < *ger); even the root *pelH, like in Gr. "polis", for "town", could be connected somewow to "kwel and be a distant relative of agara (if this is from *ger, and further from *kwer / kwar); also about the pir / parna roots for house in Hititte and AfroAsiatic (Egyptian) could be connected, as I said. We assume of course some Nostratic here.

About Berber *a-ɣaram 'town, settlement', I think it looks like Sumerian agrun "cella; bedroom; a ritual building; the sanctuary of the goddess Ningal" Akk. agarunnu; kummu; šutukku (there is also a writing A.GAR.KA "part of temple"'; agarniĝin, "an official", agrig "steward, housekeeper", Akk. abarakku; also abrig "a cultic functionary; a type of priest", Akk.  abriqqu "steward, housekeeper").

The meaning "sanctuary, ritual building" could lead us also to Tamil nakar 'house, abode, mansion, temple, palace, town, city'. So, we are looking about an initial na-. I just though about Gr. ναός naos "temple, sanctuary"), from *nawos and now I am asking for any suggestion about this. Wikipedia gives the etymology:

For Proto-Hellenic *nahwós, perhaps from Proto-Indo-European *nes- (“to join with, to conceal oneself”). The verb is ναίω naíō "to dwell, abide.

proposed IE root here:

I'm not sure if a root *nahw- could be connected to Tamil nakar. Wikipedia lead me to a comparison to the root of "boat, ship" : *néh₂us. In Indo-Iranian *nāwiyas. This is not so absurd, in religion "ship" and "temple" are connected.

I have this idea: In Proto-Germanic there is a word *nakwô m "boat, ship", from this root:

I thougth about a possible *nakwar, Maybe it has to do with "turning". Because nakar as a verb means "crawl, creep, move around". And the proposed *nes root of naos has also a meaning of "re-turning".

Nirjhar007 said...

Very interesting :) .

Kyriakos Samelis said...

Ok, let's see: first, about ναίω naiō 'to dwell, inhabit', it has also a causal type "give (one) to dwell in, make habitable, build, let one dwell, settle him."

Another (possibly) related word in Greek is νάσσω nassō 'press, squeeze close, stamp down' (also 'pile up with, stuff quite full'), which stands close to ναίω; both verbs could be from *nas, but nassō has also forms with nag-, so it could be maybe from some *nag- (or *nak-). Maybe there is a kind of lenition here, as I 've said.

Now, in Sumerian, we find words like na, wr. na4na "pestle; a stone" Akk. na'u; there is also na, wr. na4; na; na4na "stone; stone weight" Akk. abnu.

Another word is naĝa, wr. ĝešnaĝa3; naĝa3ĝa2; ĝešnaĝa4 "mortar" Akk. esittu. That one, I think, could be compared to the above verb νάσσω nassō 'press, squeeze close, stamp down', since its possible root is *nag-.

In Hesychius there is a word · ἐρείσει. λιθάσει (naxei = ereisei, lithasei - νάξει is future, singular, 3rd person of νάσσω). The first verb 'ereisei' means 'he'll cause to lean, prop, he'll make things firm, he'll support'. The second verb ('lithasei') means "he'll stone someone, he'll throw stones" (from λίθος lithos 'stone'). So, I think there is a root *nag- having a meaning 'to build / to stone', connected probably with these Sum. words.

Of course there is also Sum. nagar, wr. nagar "carpenter" Akk. nagarum, which looks identical with Skt. nagara and Tamil nakar. Hebrew 'naggar' (craftsman) is translated in Greek as "tektōn", which means "carpenter, builder". So, I think that we have a notion 'build' also here.

Kyriakos Samelis said...

Another question I have about nagara / nakar is about its possible relationship (as an indic loanword perhaps) with the Gr. word μέγαρον megaron 'large room, hall; (in the plural) house, palace; sanctuary, shrine' etc:

According to Wikipedia: 'Of Semitic origin; compare Arabic مَغَارَة (maḡāra, “cave”), Hebrew מְעָרָה (məʿārâ, “cave”), Ugaritic 𐎎𐎙𐎗𐎚 (mġrt, “cave”). Influenced in form and meaning by μέγας (mégas, “big, large”)'.

Yet, even if it is influenced by μέγας "big, large", I don't think that a meaning "cave" fits well here. It seems to be like a lage habitation, not a cave, and the meaning of Tamil nakar as "palace" or "temnple" could fit better, I think.

Megara is also a name of a town in Greece, not far from Athens:

Kyriakos Samelis said...

The Semitic word for "cave" must be connected to the meaning "pits (for sacrifices)" (megara < magara - see IV below).*m%3Aentry+group%3D24%3Aentry%3Dme%2Fgaron

not to be confused with megaron as "large room, hall, palace, sanctuary, shrine, tomb", I believe.

Kyriakos Samelis said...

Wikipedia gives an unexpected etymology for Arabic maḡāra:
Persian غار (ğâr) 'cave, tavern': From Arabic غَار (ḡār), from Arabic مَغَارَة (maḡāra), from Parthian mgʾdg (maγādag, “cave”).

I think the comparison of arabic maḡāra to Hebrew məʿārâ, “cave” suggests a kind of lenition between the two words.

I also haven't found a root for "cave" giving Arabic maḡāra; btw, searching about Semitic roots for "pit, hole", I've found this one at Starling:

Proto-Semitic: *ḥapr-
Afroasiatic etymology: Afroasiatic etymology
Meaning: 'big well' 1, 'hole, cavity' 2, 'dig' 3
Arabic: ḥafr- 1, ḥfr i, 3
Jibbali: ḥfǝr-et 2

Looks like Sum. habruda "hole" Akk. hurru. Apparently no tracable IE here.

Another root is this one:
Proto-Afro-Asiatic: *bVʔr-
Meaning: pit, well
Borean etymology: Borean etymology
Semitic: *buʔr-
Notes: ND 225 adds some Berb. and Chad. forms

Number: 1918
Proto-Semitic: *buʔr-
Afroasiatic etymology: Afroasiatic etymology
Meaning: 'pit, well, hole'
Akkadian: būr-
Phoenician: bʔr
Hebrew: bōr
Arabic: buʔr-at-
Epigraphic South Arabian: bʔr
Gurage: bʷǝr
Notes: Moab br *-u- < HS *-a- after a labial

Perhaps compared to Sum. burud "breach, hole; depression, low-lying area, depth; to perforate; (to be) deep" Akk. palāšu; šapālu; pilšu; šupālu; šuplu
Also, no apparent IE connection here.

Kyriakos Samelis said...

Only this one *bhrew-er/n- meaning:spring (of water):

Kyriakos Samelis said...

Maybe proto Semitic *ḥapr- comes from Sum. habrud(a) and not the opposite. According to J. Halloran it's habrud: cesspit (2006 edition "animal burrow"); pit, hole; cave, cavern (hab, 'to stink', + bùru(-d), 'hole').

The other similar word nakra "alligator" or "crocodile" (also makara in Skt., magar in Hindi etc) is of uknown origin; perhaps connected to a notion of "moving"? (like "moving with feet?").

There is also Tamil nakar "crawl, creep, move around". Nirjhar, any suggestions?

Kyriakos Samelis said...

makara / magar looks like a (centum like) reflex of the root *māk̑- : mək̑- 'long and slender'.

English meaning: long, slender
German meaning: `lang and dönn, schlank'

Material: Av. mas- `long', compounds masyā̊ `the größere', Sup. masiṣta-, ap. maϑišta- `the höchste', av. masah- n. `length, greatness, bulk, extent' (*mas- for *mis- from idg. mǝs- through influence of mazyā̊ `größer': μέγας), gr. μά̆σσω, μά̆σσων (*μᾰκι̯ων) besides μᾶσσον (after ἆσσον `nöher') `lönger', μήκιστος `the löngste', μῆκος, dor. μᾶκος n. `length', Μάκετα `Hochland', Μακεδόνες hence `Hochlönder', μακεδνός `slim', nachhom. μηκεδανός ds., μακρός `long' (= lat. macer, dt. mager); perhaps μάκαρ n. `Glöckseligkeit';

lat. macer, -cra, -crum `lean', maceō, -ēre `lean sein', maciēs f. `Magerkeit';

perhaps here air. mēr m. `finger' (*makro-);

ahd. magar, ags. mæger, aisl. magr `lean'; in addition with l-suffix hitt. ma-ak-la-an-te-eš (maklantes) Nom. Pl. `lean'.

References: WP. II 223 f., WH. II 2, Benveniste BSL. 33, 140 f.
Page(s): 699

The question is if the word for crocodile is IE or not, I guess.

Kyriakos Samelis said...

another Sumerian word which look similar to Skt./Prakrit āgāra 'apartment, dwelling, house' is this one (from ePSD):

aĝarin [MATRIX] (18x: Old Babylonian, unknown) wr. aĝarin4; aĝarin3; aĝarin5; a-ĝa2-ri-in; aĝa3-ri2; aĝarin; aĝarin2; aĝarinx(|AB×HA|); aĝarinx(|LAGAB×HAL|); a-ĝa2-ri-im "matrix, mother-creator; beer mash, beer bread; crucible" Akk. agarinnu; bappiru; ummu

not that it means "dwelling, house" of course ( < *kwar/kwer/*kwel/); but, I would say it could be from the other *kwer root, meaning "to do, to make, to form".

The initial a- could be derived from some glottal (or laryngeal), or maybe from an *en- (= in) > a, as sometimes (rarely) does happen.

According to Halloran, it's agarin(2,3): father; mother; womb. agarin4,5: beer-wort; crucible, vat. At 2006 edition Halloran ellaborates: agarin(2,3): father; mother; womb; mold, molding, casting. agarin4,5: beer-wort; crucible, vat (zabar aga, 'a bronze beer vessel' + rin2, 'to be bright').

According to Foxvog, it's aĝarin(AMA.d INANNA), àga-rí(-n) mother(creatress); (fertile) soil; mold, crucible; mixing basin (Heimpel, CUSAS 5, 239).

Similar roots (to *kwer/ *kwel) for "womb"


*ku̯elp- 'to curve, vault' (especially Gr. kolpos)

A derivation from a *kwer could fit also, I think, for the other meaning 'beer mash, beer bread'
Not at all sure, but maybe even 'bappir' could be from *kwer (-pir < *kwer); the bap- part reminds the bap- of Gr. βάπτω

since "bappir" is supposed an ingredient for beer making:

Kyriakos Samelis said...

There is an initial a- in Skt. ákar (root aorist), which reminds of the initial a- in aĝarin.

About the affiliation of "do, form, mould" and "beget" in Sumerian, let me put also here this link about Sumerian "tud" = "hit" (as "to form / make statues"), having also a meaning "to beget" (the verb used for "make", in other occasions, is dim):

Nirjhar007 said...

Fascinating stuff Kyriakos :) .

Kyriakos Samelis said...

Hi Nirjhar, I'm glad you find it interesting ;)

The -in ending of aĝarin looks like the Skt. -in of karin "doing, making" - or kArin "producing"; (though, I think, akarin means "not doing" : D ) ; another word akArin means "produced in a mine".

I'not sure if karin "elephant" is connected, too; if from *kwer maybe it's like Gr. teras or pélōr, "supernatural monster":

Kyriakos Samelis said...

Also, about the meaning of aĝarin as 'vat' or 'basin' there is another *kwer:

I've made, meanwhile, some random thoughts about Sum. habruda 'pit, hole; cave etc' concerning its etymology as hab 'to stink' -according to ePSD also (to be) malodorous, fetid- + bùru(-d), 'hole'. 'habru-' reminded me Gr. sapro- 'putrid, rotten'

this ambiguous root sap- has been connected to Skrt. kyaku = 'mushroom, fungus'; yet, this doesn't seem convincing to Mayrhofer and Chantraine.

Since a "putrid hole" looks like a grave, I thought also about Gr. τάφος taphos "grave, tomb" (from θάπτω thapto "to bury") and a similar word τάφρος taphros "ditch, trench".

θάπτω thapto "to bury" has the same form with another verb σκάπτω skapto 'to dig'

Pokorny puts skapto in this well known root
there is another Sum. word hab meaning 'pot'.

Kyriakos Samelis said...

About aĝarin again, perhaps we don't have very much to bother about the initial a-; I've noticed that in Akkadian there is also a second type "garinnu"(other type: "agarinnu").


I've also the impression that *ku̯er-'to do, make, form' must be connected to k̑er-, k̑erə-, k̑rē- 'to grow, increase', from which there is Gr. κοῦρος kouros "son, young male", which, again in my opinion, is related to Sum. ĝuruš, guruš "young adult male; able-bodied male worker; semi-free male worker" (Akk. eţlu). This explains perhaps the meaning as "male worker" (doer).
Especially Skt. karin 'doer, doing, making' reminded me the female Gr. name Korinna (deminutive of kore / koure, fem. of kouros)

Kyriakos Samelis said...

About nágara and nakar again, I made a thought about the root *nes- of Gr. naos "temple" (Giacomo would say *nas-), that it looks like a satem version of an unattested root **nāk̑- (I though about such a possibility seeing Indic mas- of the *māk̑- root); and then I thought about a name of a Greek temple, Anakeion, a sanctuary of Dioskouroi (sons of Zeus) in Athens:
Dioskouroi were called anakes (lords) < from anaktes, pl. of anax = king, lord (we've been talking about this word at Giacomo's posts and its proposed etymology from *wen-aǵ-), also "anaktoron" means "palace":
(The aǵ- of this etymology is from ago < *aĝ- *heĝ-, like in agros "field", compared with Sum. agar "field", also with ak, meaning "to do, to make, to act, to perform" in Sumerian).

Then I thought that a root which could look like **nāk̑ could be this one:
enek̑-, nek̑-, enk̑-, n̥k̑- 'to reach, obtain'.

If the word "near" comes from this root

(having also in mind that Arabic maḡāra corresponds to Hebrew məʿārâ) (“cave”), we could imagine perhaps a kind of **na- ( < *nēhw "nigh")

connected to nágara; maybe the meaning is that in cities people come closer (the nakt- of the other similar root I've mentioned means also "pressed").

Kyriakos Samelis said...

About Arabic maḡāra and Hebrew məʿārâ (“cave”), I remembered that Nirjhar, when talked about Bomhard's Proto-Nostratic (n.) *k’¦war-b-a ‘the inside, the middle, interior, inward part’ (at January 2nd, 2017) made a reference to this Skt. word, coming from *kwar- /*gwar-.

Also Giacomo at his first post about the Sumerian - IE connection, wrote (it is stated on the list):
"One particular case is Sum. PA.TE.SI ‘lord of the city’, because it is the cuneiform spelling corresponding to the Sumerian ensi, and it recalls PIE *patis, Skt. patis, Avestan paiti- ‘lord, master, husband’, Latin potis sum 'I am master, able', hospes (*hosti-pets) 'lord of the guest, host', Greek despotes 'master of the house, absolute ruler'. Gordon Whittaker notices (see here) also the form GAR(A).PA.TE.SI, which recalls an IE compound like Sanskrit gṛhapatis 'master of the house', cp. Avestan gǝrǝδa-, Gothic gards 'house', Lithuanian gardas 'pen, enclosed area', Old Church Slav. gradь 'town, fort, garden', or -gara probably found in Skt. nagara- 'town, city' (from *nṛ-gara 'gathering of men'), or the already mentioned Greek agorà 'assembly, place of assembly' (see gar above). However, from the Pennsylvania Sumerian Dictionary GAR appears to be normally used in compounds as synonym of niĝ 'thing, possession'."

First, Sum. niĝ 'thing, posession' maybe is related to IE enek̑-, nek̑-, enk̑-, n̥k̑- 'to reach, obtain' (maybe labialized). Note that in Greek this same root enek̑-, nek̑-, enk̑-, n̥k̑ (maybe with a different initial laryngeal?) is related to the notion "to carry, to bring", in ἐνεγκεῖν enenkein (used in some tenses of the verb φέρω phero "to carry, to bear"), making perhaps comprehesible this use of niĝ as a synonym of Sum. gar "to heap up" (Akk. garānu "to stack up, pile on"). Of course, this root nek̑ becomes in Sanskrit also NAS (as the NAS from *nes).
btw, nagara reminded me of another town's name "Ancyra" in Phrygia (today's Ankara, the capital of Turkey), which sounds like "anchor" ("ankyra" in Greek), yet I doubt if this is the real meaning.

Kyriakos Samelis said...

Parenthesis: here, Sum. niĝ has been compared to Gr. νίκη nī́kē "victory, success"; as something you manage to obtain. The usual etymology is from *neik-; I wonder though if it's from the zero grade n̥k̑- of this same root enek̑- mentioned above.

Kyriakos Samelis said...

About niĝ "thing, possesion; something" (etymology from ni "self" and aĝ2 "to mete out to", according to J. Halloran) and the possible connection to νίκη nī́kē "victory", one could notice that in Sumerian "a2...ĝar" means "to defeat, to win, to conquer"; and since the writing of ĝar is the same with niĝ2, we could perhaps imagine a same meaning for niĝ... I think that both verbs are related to a semantic field of "establishment" or "acquisition". Also, the Emesal form of niĝ ( = aĝ2) reminds of the comparison nagara / agāra in Sanskrit.

Kyriakos Samelis said...

About Sum. ni "self" (wr. ni2), I think a comparable word could be Gr. μιν min or νιν nin meaning 'him(self), her(self), it(self)'.

μιν (min) ionic acc. sg. of the pron. of the 3rd pers. (v. ἵ) through all genders, for αὐτόν, αὐτήν, αὐτο always enclitic, Hom., Hdt.; doric and attic νιν (nin)

I.Hom. joins μὶν αὐτόν himself, as a stronger form; but αὐτόν μιν is reflexive, oneself, for ἑαυτόν, Od.
II.rarely as 3 pers. pl. for αὐτούς, αὐτάς, αὐτά.

Kyriakos Samelis said...

Another word which looks like 'nakt-' is Toch. B ñakte 'god', TchA ñkät, both from PTch *ñäk(ä)te According to the dictionary of Tocharian B by Douglas Q. Adams:

" ñakte (nm.) ‘god’ [voc. ñakta often used as respectful address to a king] [ñakteṃts ñakte is an epithet of the Buddha or of a maitreya]... extra-Tocharian connections are not altogether clear. VW (326-327) suggests a derivation from PIE *h1neḱ- ‘obtain, take’ [: Sanskrit aśnóti ‘attains,’ náśati ‘id.,’ Lith. nešù ‘carry,’ TchB enk- ‘take, seize,’ etc. (P:316-318; MA:35)], i.e. *h1neḱto- ‘he who brings, he who obtains.’ He notes the semantic similarity (which stops well short of a semantic identity) with Sanskrit bhága- ‘master,’ Avestan baɣa- ‘master, god,’ derivatives of a verb seen in Skt, bhájati ‘shares.’ Alternatively Watkins (1974:102) takes ‘god’ to be ‘the libated one’ (from PIE *ǵheu- ‘pour’ [P:447-448; MA:448]) with reference to Sanskrit āhuta- ‘begossen’ as an epithet of Agni. Normier (1980:267ff.), however, is probably right in taking PTch *ñäk(ä)te to reflect a virtual PIE *ní-ǵhuhx-to- ‘± the one called down’ (more particularly *ní-ǵhuhx-to- with the retracted accent characteristic of nouns derived from adjectives). The vowel of the root syllable has been shortened to *-u- (or the laryngeal was lost) in pre-Tocharian but after that the development is phonologically regular..."

Personally I prefer the first option. I. Mosenkis has connected Toch. nakt- 'god' (used also for kings) with Linear B (w)anakt- (which thinks it's Hattic in origin).

Kyriakos Samelis said...

Concerning the writing PATESI for Sum. ensi, there is an interesting etymology of IE *potis "lord, husband" (according to Giacomo *patis) at Wikipedia:

"Derksen: it's often assumed that the meaning "lord, husband" derived from an even older meaning "self", as found in Lithuanian pàts (“self”)."

I think thus that not only niĝ "possesion", but also Sum. nin, wr. nin; ga-ša-an; ga-ša2-an; ka-ša-an "lady; mistress, owner; lord" (Akk. bēltu; bēlu) could be derived from a (reduplicated?) ni, wr. ni2 = 'self' (Halloran says this is from a reduplicated ni = 'fear').

An interesting thing is that the ša2 sign of the Emesal "ga-ša2-an" (= nin) is the same with niĝ "thing, possesion" (and ĝar); now, according to I. Mosenkis, gašan or kašan "lady, lord" is derived from an Iranian xšain = 'queen'.

This must be derived from *tk(')ēy- "to possess, to acquire" > Old Indian kṣáyati `to possess, rule over'; kṣatrá- n. `dominion, power, might' etc. A kind of 'kSan' Skt. root.

We were talking about this IE root at Giacomo's posts, when we discussed about Sum. tuku = "to acquire, get", when I said it reminds this *tk(')ēy-.

So, I assume that we have also in Sumerian traces of a (kind of) "thorn clusters", as they were described in Kloekhorst's paper: TK > in tuku; KSH > in gašan, even the labialized PT > in PATESI.

Maybe this was a trait existing already in IE dialects, before the separation of the several branches; or maybe its origin is even pre-IE.

Another question if these TK clusters have some relationship to some possible NK / LP (let's say) ones (like in niĝ "possesion" / lib "rich", or niĝir/ libir 'herald').

Kyriakos Samelis said...

btw, since we're talking about *tk(')ēy- "to possess, to acquire", the main Gr. reflex is κτάομαι ktáomai̯ 'procure for oneself, get, acquire':

also, κτέαρ ktéar 'possessions, property':

Also (most probably) κτέρας kteras "possession", κτέρεα kterea (pl.) '(funeral) gifts'

ktear and kteras reminds, in my opinion, of this Sum. word:

kadra [GIFT] (44x: Old Akkadian, Lagash II, Ur III, Old Babylonian) wr. kadra, Akk. irbu "income; entry fee? to temple"; kadrû "present, greeting gift"; ţātu "bribe, baksheesh".

So, there is a possibility (also) of a (reversed) greek style KT form of this 'thorn cluster' about possession.

Kyriakos Samelis said...

Another example of such a 'TK' cluster in Sumerian I think is dehi [SUPPORT] (1x: Old Babylonian) wr. de-hi; dehi "support, stanchion; tax" Akk. imdu, which we have compared to the IE root *dʰeyǵʰ '-to knead, to form, to shape' and especially some cognates, meaning 'wall' or such things.

Like Sk. deha 'body, mass, bulk, person, appearance' etc and dehi 'surrounding wall, mound, rampart'

also Avestan daēza 'wall', Greek τεῖχος teîkhos 'mound, earth works, wall (especially one enclosing a town or city), fortified city, fortification, castle'.

Here Sanskrit has the TK form. But I have the impression that there is also a 'KS type' Sum. word of this same root, this one:

kisa [PLATFORM] (54x: Lagash II, Ur III, Early Old Babylonian, Old Babylonian) wr. ki-sa2; kissa "platform; supporting wall, abutment", also 'plinth', Akk. kisû "footing, plinth".
I still think this word is connected to Latin castrum.

For the KT form I think there is a word in Akkadian: kādu 'fortified outpost; (according to CAD), also its guards; 'fort' according to Porpola.

Finally, I think that, for the PT form of this cluster, we have this well known Sum. word:

bad [WALL] (2910x: ED IIIb, Ebla, Old Akkadian, Lagash II, Ur III, Early Old Babylonian, Old Babylonian, unknown) wr. bad3 "wall, fortification" Akk. dūru

which, I think, among all the IE languages, exists only in Armenian (as pat/bad).

Kyriakos Samelis said...

The PT forms, of course, are due to a kind of labialization.

BTW, I am thinking now of a kind of labialization in case of Sum. banda = 'child, son', maybe connected to this IE root (given by Bomhard in the glottalic form), which Nirjhar has already posted twice, I think:

Proto-Indo-European *kan *k’en-/*k’on-/*k’n̥- ‘to beget, to produce, to create,
to bring forth'

Sum. banda looks indeed like these Dravidian words compared by Bomhard - if labialized:

Kannaḍa kanda ‘young child’, kandu ‘calf' Telugu kandu‘, infant’, kanduvu ‘child’.

The labialization of "k'an" maybe has happened as a part of a de-palatalization process (I get this idea from a paper Nirjhar sent to me). The final -da could be a suffix.

On the other hand, another root which could be connected with banda, that is Semitic bin 'son' looks isoleted inside Afro-Asiatic. Could be also connected, due a similar labialization, to this same root "k'an"?

Nirjhar007 said...

On the other hand, another root which could be connected with banda, that is Semitic bin 'son' looks isoleted inside Afro-Asiatic. Could be also connected, due a similar labialization, to this same root "k'an"?.

Yes Kyriakos , I don't see why this can't be. I agree.

Kyriakos Samelis said...

Starling's etymology of kandu etc is a bit different: to Eurasiatic *ḲanV 'young, new' and the IE root *kan/*ken "young, new".

but the root meaning "beget", Eurasiatic: *kVnV, meaning 'to bear, kin' is very close:

So, maybe these two roots are connected, too.

Also, the Dravidian root as "*kanr" leads also, I think, to Aramaic b'rā 'son, child' (bar) Hebrew bará 'to create'.

An AA root close to'bin 'son' could be also this one:

Proto-Semitic: *bVn-
Afroasiatic etymology: Afroasiatic etymology
Meaning: 'build' 1, 'create' 2
Akkadian: banû 1, 2
Amorite: bny 1, MOAB bny 1
Ugaritic: bny, bnw 1
Phoenician: bny, PUN bnʔ, bny 1
Hebrew: bny 1
Aramaic: bny 1 OAram bny, Aram (Emp, Nab) bnh, bnʔ, (Palm) bnʔ 1
Mandaic Aramaic: bna 1
Arabic: bny [-i-] 1, 2
Epigraphic South Arabian: bny 1
Mehri: benû 1
Jibbali: bené 1
Soqotri: béne 1
Notes: Various triliteral formations based on *bVn-.

Kyriakos Samelis said...

I remember also, Nirjhar, the set of AA cognates of Bomhard's Proto-Nostratic root *k’an- (vb.) ‘to get, to acquire, to create, to produce, to beget’ (n.) *k’an-a ‘birth, offspring, child, young, produce’; (adj.) ‘born, begotten, produced’, that you've posted on Giacomo's New Indology (at November 2nd, 2016):

A. Proto-Afrasian *k’an- ‘to get, to acquire, to possess, to create, to produce’:
Proto-Semitic *k’an-aw/y- ‘to get, to acquire, to possess, to create, to
produce’ > Hebrew ḳānāh [hn*q*] ‘to get, to acquire, to create, to produce’;
Phoenician ḳny ‘to acquire’; Biblical Aramaic ḳǝnā ‘to acquire, to buy’;
Ugaritic ḳny ‘to create’; Akkadian ḳanū ‘to gain, to acquire’; Amorite ḳny
‘to create, to acquire’ (basic stem, Qal yaḳnī); Arabic ḳanā ‘to get, to
acquire, to create’; Sabaean ḳny ‘to possess, to acquire’; Geez / Ethiopic
ḳanaya [ቀነየ] ‘to acquire, to buy, to subjugate, to dominate, to rule, to
subdue, to tame, to train, to make serve, to make toil, to reduce to
servitude, to bring into bondage, to force to work, to create’. Murtonen
1989:380; Klein 1987:584; Leslau 1987:437; Zammit 2002:347. Egyptian
qn, qnÕ ‘to be strong, to make strong, to have power over, to possess, to
overcome’. Hannig 1995:858; Faulkner 1962:279; Gardiner 1957:596;
Erman—Grapow 1921:190 and 1926—1963.5:41—43. Berber: Tuareg
ǝ¦nu ‘to be created, to be started; to originate (from)’. Diakonoff
1992:23—24 *ḳn̥ (*ḳny/w) ‘begetting, giving birth’.

These are the 'non labialized forms', let's say, in AA. But these KN, let's say words, reminded me also a TKN (let's say) word, Gr. τέκνον teknon 'child' (also Skt. toka etc). Root teḱ, which leads perhaps to τέκτων tekton 'craftsman, creator', but also to *tḱey- 'to settle, to live':

This last one leads again to the similar (previous mentioned) root *tḱēy- "to possess, to acquire" of Skt. kṣáyati `to possess, rule over', also (as I think) Sumerian tuku etc.

All these similar roots I think are summarized in the TKN root of 'earth', which possess and creates / begets everything. As I've said to you, Gr. teknon reminds me Hittite tēkan for 'earth'.

Kyriakos Samelis said...

Perhaps we have a *teḱ- root in Sum. tešlug [HATCHLING] wr. tešlug "hatchling" Akk. atmu (Akkadian atmu means: '1) hatchling , young of a bird / snake / turtle , chick, chicken , fledgling ; 2) (by extension) human offspring , young child , toddler ')

(lug I think means "to live, dwell (of animals), pasture", so maybe teš- could be from *teḱ-).

If all these roots above are connected, maybe that explains also the vocalization of Sum. tuku 'acquire'; tuku reminds also of τόκος tokos 'interest' from tokos τόκος 'birth' in Greek (τοκεύς tokeus also is the parent), vocalization like Skt. toka < *teḱ.

Kyriakos Samelis said...

About *dʰeyǵʰ again ;to form, to mould, clay etc': Giacomo's view is that it is connected with the root of earth (and I strongly agrees with that); I was also thinking it is connected with the root *teḱ-'to beget', if you remember; you've posted Bomhard's glottalic version of this root on New Indology (at November 9th 2016): Proto-Nostratic root *diqº- (~ *deqº-): (vb.) *diqº- ‘to crush, to pound or tamp (earth), to mold or knead (clay)’; (n.) *diqº-a ‘earth, clay, mud’, and there were some Dravidian roots there: Konḍa tig- ‘to press down hard, to lay pressure on’; Pengo tig- (tikt-) ‘to push’; Manḍa tig- ‘to push’.

That Konḍa 'tikt-' reminded me Gr. τίκτω *tíktō 'to beget, to give birth':

But another (or the same?) tikt- means aparrently 'to die':

I think it is natural: we are formed from clay and we are deformed as clay. We are born with a push and we are pushed to death. That is life.

btw, Starling gives another Euroasiatic common root (yet not including IE).

Kyriakos Samelis said...

Another curious idea that I have about these 'diq, tek', let's say 'TK' words for 'pressuring the earth', is that, using the 'thorn cluster' pattern, and searching about a labialized 'PT' version, we could imagine a word like Sum. bad, 'hard ground', which also reminds Gr. πέδον pedon whith the same meaning:
maybe even the IE root for 'foot' could be connected here, in a very archaic IE stage, or maybe a pre-IE one... btw, another root *pid means 'to give birth':

Something else, all these metatheseis in 'thorn clusters' of the Greek roots from the IE 'TK' to 'KT' and to 'PT' are a bit annoying... At least in this early stage of the IE I'm in favour of Speirs' view about the labiovelars; permit me to put Dziebel's article
from Kinship Studiesalso here, for convenience. Maybe initially there was a kind of 'KK' reduplicated pattern, beyond all these roots...

Nirjhar007 said...

Yes many will not agree but as you say 'KK' reduplicated pattern can be the original and archaic form . I like the ideas of Dziebel, he thinks practically by putting data first , I consider its the attitude we all need !.

Kyriakos Samelis said...

On this, there is a Sumerian word kukku [LAND] (1x: Old Babylonian) wr. kukku3(|KI.KI|) "land", written as a reduplicated ki 'earth' (sign 'ki' is the same with gu14); maybe this "ku" came from kʷ (Speirs instead of *dheg'h-m– gives the IE word of 'earth' as *gʷhegʷh).

Kyriakos Samelis said...

Another Sum. KK word, which is given by J. Halloran (under question), is the following one:
KU-KU: ancestors (?) ('to found; to lie down').

That reminds me a Gr. word (given also by Hesychius), κοκύαι kokyai 'ancestors' (der. uncertain), Chantrain says: 'origin obscure'.

Halloran has also ugu4[KU]: to bear, procreate, produce (cf., ugu4-bi).
That reminds also Gr. κυέω kyeō / κύω kyō 'to bear in the womb, to be pregnant with a child'. I think we have talked about that already in Giacomo's Sumerian posts.

In Halloran also, ku: to base, found, build; to lie down (reduplication class) [KU archaic frequency: 64; concatenates 3 sign variants].
ePSD: ku [PLACE] (96x: ED IIIb, Old Akkadian, Ur III, Early Old Babylonian, Old Babylonian) wr. ku "to place, lay (down), lay eggs; to spread, discharge" Akk. nadû; uşşu

Note here that Dziebel's view about the root of "father" (pitar, pater, πατήρ etc), another PT word as I think, is just *kʷe. I am copying here from his article given above:

"kwe- ‘father, father’s brother; older male relative’: IE *pH2ter ‘father’ (Toch A paacar, Toch B paacer, Skrt pita, Gk pateer, Lat pater, Arm hayr, Goth fadar, OIr athir) ~ IE *te– ‘father’ (Lith tevas, tetis ‘father’, OPruss taaws, towis ‘father’, thewis ‘father’s brother’, Skrt tata ‘father’ (r.), ‘any male relative or acquaintance’ (a.) ...' As he also says 'It’s likely that Hitt atta, Slav *otici, Goth atta ‘father’ are also derived from *akwa-, and hence there was only one PIE term for ‘father’ (*kwe-, *akwe-) and not two (*pH2ter and *atta) as presently believed. Interestingly, in Gothic, the form fadar occurs only once, with atta being the main term for ‘father’. If the two forms are linked phonetically, the puzzle of why a PIE term for ‘father’ fell out of favor in Gothic simply disappears.' Also 'Unlike *pHter and *te-, which are isolated in a Eurasian perspective, PIE *kwe-, *akwe ‘father’ find plenty of potential cognates in the so-called “Nostratic” or “Eurasiatic” languages (comp. Nostr eka ‘older male relative’ such as ‘father’, ‘father’s brother’, ‘older brother’, ‘grandfather’, etc.)."

Starostin gives also long-range etymologies for the root of the father, but only with labials:

But I think a K (Kw) or KK (KwKw?) root could unite also other some other types and thus give bigger perspectives like KK > ()VK,()VT, PP, ()VP, PT or TT, and of course a connection to the TK roots above.

Kyriakos Samelis said...

About Sum. 'ku' < 'kwe', we can recall here Whittaker's comparison for Sum. tukum 'if' (From 'The Case for Euphratic'). He states:

"tukum ‘immediately, in a moment; as soon as; if’: *to-kom, lit. ‘with that’ (cf. Hitt. takkan‘(?),’ takku ‘if,when’ < *to-kom, *to-kwe; Kloekhorst 2008: 432-433,816). Such constructions occur widely in Indo-European(see esp. Wagner 1967; Eichner 1971). In English there is a parallel construction: with that ‘thereupon; (obs.:) provided that, if.’"

Though I have the impression that IE *kom and *kwe- are two different roots (yet, there is Latin quum (quom) > cum), the comparison of *to-kwe to Sum. tukum clearly shows a -kwe (-kwam?) > Sum. ku(m).

BTW, *to-kwe in Greek is τότε tote (Kloekhorst mentions also this at this Hittite dictionary) "at that time, when" (he has the meaning also 'as a result'). From τε < -kwe (In Doric is τόκα toka < to-kwe).

Otto Edzard, in his 'Sumerian Grammar' states that "tukumbi “if ” is newer than Old Sum. 'uda'. And I think in Giacomo's post I have compared 'uda' to Gr. ὅτε hote "when".

So, it is possible that in Sumerian there are many versions of the same root, as in Greek and other languages; we han have -da-" < kwa-, and ku- < kwu-, for example.

Kyriakos Samelis said...

I had suddenly this idea about Gr. ξύν "xyn" (ksyn) "plus, and", which became later in Greek just σύν "syn"; we have talked about it in Giacomo's first Sumerian post, Nirjhar, if you remember; Giacomo had mentioned the views of Dunkel about it, and then proposed a procedure k'om > k'on >k'un > xyn (inspired by Villar, as he said); the idea is that it could be derived from a TK "thorn cluster", like from a contracted "to-k'om" (from tk'on < tk'om < *to-k'om, "whith it"), but following a Sankrit-style development of these clusters from TK to KSh, sice in normal 'Greek-style it should be "ktyn").

I think that a "thorn cluster" could make things simpler. In any case, I think we have an Indic influence here (as you know, I think that this influence is to be found also in some other cases in Greek). Then, *som could have influenced ksyn, and this newer type σύν "syn" was created. It is the opposite of Dunkel's view, that Giacomo had mentioned, too.

This conservation was at September 2015 and you have also raised the question, if the ancient IE had a perticular "palatal" sound which evolved in kentum and satam; I think I agree now (if that matters :D).

Nirjhar007 said...

okay :) .

Kyriakos Samelis said...

An intersting thing is that there are not visible Indic reflexes of this root *ḱóm:

But maybe kshana "in a moment, instantly, instant etc" etc is derived from a thorn-cluster.

Maybe from a *tk'an <ta-k'an? (from ta-k'am?)