Kassian Hurro-Urart Lexicostat 2011

Sonderdruck aus
UGARIT-FORSCHUNGEN

Internationales Jahrbuch
fr die
Altertumskunde Syrien-Palstinas

Herausgegeben von
Manfried Dietrich Oswald Loretz

Band 42
2010

Ugarit-Verlag Mnster
2011

Herausgeber
Manfried Dietrich / Oswald Loretz, Ugarit-Verlag, Ricarda-Huch-Str. 6,
D-48161 Mnster
(Manfried Dietrich: ugarit@uni-muenster.de)

Redaktion
Ugarit-Verlag, c/o Institut fr Altorientalische Philologie und Vorderasiatische
Altertumskunde, Rosenstr. 9, D-48143 Mnster
(Kai A. Metzler : metzler@ugarit-verlag.de)

Fr unverlangt eingesandte Manuskripte kann keine Gewhr bernommen werden.
Die Herausgeber sind nicht verpflichtet,
unangeforderte Rezensionsexemplare zu besprechen.
Manuskripte fr die einzelnen Jahresbnde werden jeweils
bis zum 31. 12. des vorausgehenden Jahres erbeten.

2011 Ugarit-Verlag, Mnster
(www.ugarit-verlag.de)
Alle Rechte vorbehalten
All rights preserved. No part of this publication may be reproduced,
stored in a retrieval system, or transmitted, in any form or by any means,
electronic, mechanical, photo-copying, recording, or otherwise,
without the prior permission of the publisher.

Herstellung: Hubert & Co, Gttingen
Printed in Germany

ISBN 978-3-86835-053-1

Printed on acid-free paper

Inhalt

Artikel
Al-Shorman, Abdulla / Al-Bashaireh, Khaled /
Doomi, Mohammad Bani
The Paleoclimate of the Northwestern Jordan in Late Antiquity .................... 1
Best, Jan / Lia Rietveld
Structuring Byblos tablets c and d ................................................................. 15
Bloch, Yigal
Setting the Dates: Re-evaluation of the Chronology of Babylonia
in the 14
th
11
th
Centuries B.C.E. and Its Implications
for the Reigns of Ramesses II and attuili III ............................................. 41
Carbillet, Aurlie
Un chapiteau hathorique indit dAmathonte (Chypre) ................................ 97
Devecchi, Elena
RS 17.62 + RS 17.237 (CTH 64). Treaty, Edict or Verdict? ....................... 105
Dietrich, Manfried / Oswald, Loretz
Die Seevlkergruppe der rtnm ardana/ erdena in Ugarit.
Bemerkungen zum Brief KTU 2.61 und zur Liste KTU
3
4.497+. . . .......... 109
Bestallungsurkunde KTU 3.11 (RS 15.117) fr Bldq als
obersten Verwaltungsbeamten (skn) des Palastes .......................................... 125
Rhabdomantie im mykenischen Palast von Tiryns. Das Fragment eines
kurz-keilalphabetisch beschrifteten Elfenbeinstabs
(Ti 02 LXIII 34/91 VI d12.80 = KTU
3
6.104) ............................................. 141
el-Khouri, Lamia
Barsinia 1
st
Century B.C. 1
st
Century A.D. Pottery from the Cistern,
Area A ......................................................................................................... 161
Gerhards, Meik
Die Sonne lsst am Himmel erkennen Jahwe .
Text- und religionsgeschichtliche berlegungen zum Tempelweihspruch
aus I Reg 8,12f. (M) (III Reg 8,53a [LXX]) ................................................ 191
iv Inhalt [UF 42
Gestoso Singer, Graciela
Forms of payment in the Amarna Age and in the Uluburun and
Cape Gelidonya shipwrecks ........................................................................ 261
Gillmann, Nicolas
Un exemple de hilani Til Barsip? ............................................................. 279
Halayqa, Issam K. H.
The Demise of Ugarit in the Light of its Connections with atti ................ 297
Heide, Martin
The Domestication of the Camel: Biological, Archaeological and
Inscriptional Evidence from Mesopotamia, Egypt, Israel and Arabia,
and Traditional Evidence from the Hebrew Bible ....................................... 331
Kassian, Alexei
Hurro-Urartian from the lexicostatistical viewpoint .................................... 383
Lipschits, Oded / Koch, Ido / Shaus, Arie / Guil, Shlomo
The Enigma of the Biblical Bath and the System of Liquid Volume
Measurement during the First Temple Period ............................................. 453
Loretz, Oswald
Ugaritisch-altisraelitische Elemente des Neujahrsfestes
im nachexilischen Psalm 24 ............................................................................. 479
Naaman, Nadav
Khirbet Qeiyafa in Context ......................................................................... 497
Park, Sung Jin
Short Notes on the Etymology of Asherah .................................................. 527
Peterson, Jeremiah
Sumerian Literary Fragments in the University Museum,
Philadelphia II: Eduba Compositions, Debate Poems, Diatribes,
Elegies, Wisdom Literature, and Other Compositions ................................ 535
Peterson, Jeremiah
Sumerian Literary Fragments in the University Museum,
Philadelphia III: Hymns to Deities .............................................................. 573
Sazonov, Vladimir
Einige Bemerkungen zur altassyrischen Knigstitulatur.
Entwicklungsgeschichte und sdmesopotamische Einflsse ...................... 613
Snyder, Josey Bridges
Did Kemosh Have a Consort (or Any Other Friends)?
Re-assessing the Moabite Pantheon ............................................................ 645
Theis, Christoffer
Sollte Re sich schmen? Eine subliminale Bedeutung des Namens
in Jeremia 44,30 .................................................................................. 677
2010] Inhalt v
Tropper, Josef / Vita, Juan-Pablo
Die keilalphabetische Inschrift aus Tiryns .................................................. 693
Tugendhaft, Aaron
On ym and
d
A.AB.BA at Ugarit .................................................................. 697
Vernet, Eullia / Vernet, Mariona
Die groe Sphinx von Gizeh. Vergleichende und sprachwissenschaftliche
berlegungen zu einer afroasiatischen Etymologie .................................... 713
Vidal, Jordi
Ugarit at War (3): Prisoners of War ............................................................ 719
von der Osten-Sacken, Elisabeth
Aur, groer Berg, Knig von Himmel und Erde. Darstellungen des
assyrischen Hauptgottes im Wandel vom numen loci zum Gtterherr ........ 731
Watson, Wilfred G. E.
Getting to Grips with Ugaritic tdl .............................................................. 823
Watson, Wilfred G. E.
Non-Semitic Words in the Ugaritic Lexicon (8) ......................................... 831
Yogev, Johnathan
How wide should a Column be? .................................................................. 847
Yogev, Johnathan
The Strange Case of Diagonal Writing .................................................... 853
Zadok, Ran
Philistian Notes II ........................................................................................ 859
Zukerman, Alexander
On Aegean Involvement in Trade with the Near East During
the Late Bronze Age .................................................................................... 887
Replik
Pardee, D.
Illustrated Epigraphic Remarks to the First Tablet of the Aqhatu Text,
Lines 124 ................................................................................................... 903
Buchbesprechungen und Buchanzeigen
Yoram COHEN / Amir GILAN / Jared L. MILLER (Hrsg.): Pax Hethitica.
Studies on the Hittites and their Neighbours in Honour of Itamar
Singer (Manfred Hutter) .............................................................................. 919
Charles DOYEN: Posidon souverain. Contribution lhistoire religieuse
de la Grce mycnienne et archaique (Oswald Loretz) .............................. 923

vi Inhalt [UF 42
J.-M. DURAND / Th. RMER / M. LANGLOIS (Hrsg.): Le jeune hros.
Recherches sur la formation et la diffusion dun thme littraire au
Proche-Orient ancient. Actes du colloque organis par les chaires
dAssyriologie et des Milieux bibliques du Collge de France,
Paris, le 6 et 7 avril 2009 (Oswald Loretz) ................................................. 924
Giovanni GARBINI: Dio della Terra, Dio del Cielo. Dalle religioni
Semitiche al giudaismo e al cristianesimo (Oswald Loretz) ....................... 925
Brigitte GRONEBERG / Herrmann SPIECKERMANN (Hrsg.): Die Welt der
Gtterbilder (Michael Herles) ..................................................................... 926
Joel M. LEMON: Yahwehs Winged Form in the Psalms. Exploring
Congruent Iconography and Texts (Oswald Loretz) ................................... 932
Hartmut MATTHUS / Norbert OETTINGER / Stephan SCHRDER (Hrsg.):
Der Orient und die Anfnge Europas. Kulturelle Beziehungen von
der Spten Bronzezeit bis zur Frhen Eisenzeit (Oswald Loretz) ............... 933
Kevin M. MCGEOUGH, edited by Mark S. SMITH: Ugaritic Economic
Tablets:Text, Translation and Notes (Oswald Loretz) ................................ 934
Terence C. MITCHELL / Ann SEARIGHT: Catalogue of the Western Asiatic
Seals in the British Museum. Stamp Seals III. Impressions of Stamps
Seals on Cuneiform Tablets, Clay Bullae, and Jar Handles
(Ellen Rehm) ............................................................................................... 935
Ludwig D. MORENZ: Die Genese der Alphabetschrift. Ein Markstein
gyptisch-kanaanischer Kulturkontakte. Wahrnehmungen und
Spuren Altgyptens (Oswald Loretz) ........................................................... 936
Andreas SCHACHNER: Bilder eines Weltreichs. Kunst- und Kultur-
geschichtliche Untersuchungen zu den Verzierungen eines Tores aus
Balawat (IMGUR-ENLIL) aus der Zeit Salmanassar III,
Knig von Assyrien (Ellen Rehm) ............................................................... 937
Itamar SINGER (Hrsg.): ipamati kistamati pari tumatimis. Luwian and Hittite
Studied Presented to J. David Hawkins on the Occasion of his 70
th

Birthday (Manfred Hutter) .......................................................................... 940
Richard C. STEINER: Early Northwest Semitic Serpent Spells in the
Pyramid Texts (Oswald Loretz) ................................................................... 944
[Raymond WESTBROOK:] Law from the Tigris to the Tiber.
The Writings of Raymond Westbrook. Edited by Bruce Wells
and F. Rachel Magdalene (Kristin Kleber) ................................................. 945
Lorenz WINKLER-HORAEK (ed.): Wege der Sphinx. Monster zwischen
Orient und Okzident. Eine Ausstellung der Abguss-Sammlung Antiker
Plastik der Instituts fr Klassische Archologie der Freien Universitt
Berlin (Nadine Nys) .................................................................................... 947
2010] Inhalt vii
Abkrzungsverzeichnis ..................................................................... 951
Indizes
A Stellen ......................................................................................................... 967
B Wrter ......................................................................................................... 972
C Namen ......................................................................................................... 975
D Sachen ......................................................................................................... 986
Anschriften der Mitarbeiterinnen und Mitarbeiter ................................... 989

Hurro-Urartian from the lexicostatistical viewpoint

Alexei Kassian, Moscow
1

1. The Hurro-Urartian (hence: HU) linguistic family consists of two closely re-
lated languages: Hurrian (with several dialects)
2
and Urartian.
3
Despite the
chronological distance between the attested Hurrian and Urartian, it seems clear
that the latter is not a direct descendant of the former, but the two languages
represent two separate branches of a common proto-language (Proto-Hurro-
Urartian).
4
For the preliterate period, it is natural to associate the HU people
with the Kura-Araxes (Early Trans-Caucasian) archaeological culture (Kassian,
2010a, 423 ff. w. lit.).
External connections of the HU languages are not clear yet. The most natural
assumption, in view of the geographical distribution and typological similarity,
would be to include HU into the East Caucasian (Nakh-Dagestanian) stock of
the North Caucasian linguistic family. This was originally proposed as early as
the second half of the 19
th
century; more recently, this idea was further devel-
oped by some Soviet and Polish authors J. Braun, G. Klimov, Ju. Deeriev,
I. Diakonoff and others (see, e. g., / , 1954; , 1963, 41
51; Diakonoff, 1971, 157171; , 1978), after which the monograph

1
I am indebted to my colleagues in the Global Lexicostatistical Database project (Mos-
cow / Santa Fe) G. Starostin (head of the team) and M. Zhivlov: without their help,
etymological analysis of the Hurro-Urartian data proposed below would be much less
reliable. The Hurro-Urartian portions of the paper have been discussed with M. Khachi-
kyan (Erevan) and I. Yakubovich (Oxford/ Moscow) ; I express my gratitude to them for
a number of valuable additions and remarks. All possible factual errors or wrong inter-
pretations are the authors only.
2
Cuneiform and Ugaritic alphabetic sources from ca. the 23
rd
century to the late 2
nd

millennium BC (Salvini, 1998a, Wegner, 2007, 21 ff.).
3
Cuneiform (and apparently hieroglyphic) sources of the 1
st
half of the 2
nd
millennium
BC (see two recent editions of the Urartian corpus: and CdTU).
4
E. g., the ejective series of obstruents was lost in Hurrian, but retained in Urartian (Kha-
chikyan, 2009) ; for a certain syntactical archaism of Urartian, see , 2010a,
119 f. ; another Urartian archaism is the retention of the negative morpheme *mV with
the prohibitive semantics (see notes on 62. not
1
below). Cf. also the following case:
Old Hurr. peli ~ pili canal (Bo.Bil.) > New Hurr. pala id. (Ugar.C.) vs. Urart. pili id.
with the retention of the archaic vocalism.
384 A. Kassian [UF 42
Hurro-Urartian as an Eastern Caucasian Language by Diakonoff and S. Staros-
tin appeared (Diakonoff/ Starostin, 1986). However, in the last few decades, as
compared to the 1980s, there has been some substantial progress in North Cauca-
sian linguistics, on the one hand,
5
and in Hurritological studies on the other. As a
result, the East Caucasian HU hypothesis began to look much less attractive (cf.,
e. g., the criticism in Patri, 2009), currently it has no strong proponents.
6

An alternative theory discussed by some authors connects HU to the Indo-
European family, i. e., treats HU as a member of the Nostratic macro-family with
a specially close relationship between HU and IE. The IE-HU theory, which
originates from certain ideas of Holger Pedersen, has been developed by the Ar-
menian author G. Jahukyan (see, e. g., Jahukyan, 1961; , 1967) and
then followed by Fournet / Bomhard, 2010. For the criticism of this theory, see
Kassian, 2010b; Kassian, 2011a.
Because neither East Caucasian nor, a fortiori, Indo-European attribution of
HU appears to be likely at the current stage of research, it seems reasonable to
undertake a more formal analysis of HU linguistic data.

2. Over the last few decades, international linguistics continued to demonstrate
an ever-increasing interest in various statistical methods of language comparison
(see, e. g., overviews in Nichols/ Warnow, 2008; McMahon/ McMahon, 2005;
cf. esp. the projects of S. Wichmanns team, e. g., Wichmann el al., 2010, and
G. Starostins team, for which see below). It may be stated that such methods
have been increasingly becoming a basic and generally acknowledged tool for
formal classication of languages and dialects.
Linguistic data used in such calculations can be of several kinds:
1) Lexical data, i. e., wordlists of basic vocabulary (the so-called lexicosta-
tistics).
2) Auxiliary morphemes and grammatical features, which are either reten-
tions or innovations in relation to the reconstructed proto-language of the
language family in question (e. g., mutual phonetic changes, syncretism
of cases and so on).
3) General (universal) typological features, whose formulation does not de-
pend on the specic nature of particular languages (e. g., basic word order
VS/ SV or the presence of rounded front vowels).
The lexicostatistical method is currently adopted in the international Global
Lexicostatistical Database project (headed by G. Starostin).
7
As a template we
prefer to use a 110-item wordlist, which consists of 100 classical Swadesh

5
E.g., a fundamental reconstruction of the Proto-North Caucasian language by S. Niko-
laev and S. Starostin was completed and published as NCED (1994).
6
At the beginning of the 2000s, S. Starostin himself tended to reject the East Caucasian
HU hypothesis in favour of a Sino-Caucasian attribution of HU, for which see below.
7
See http://starling.rinet.ru/new100/main.htm
2010] Hurro-Urartian from the lexicostatistical viewpoint 385
words plus 10 additional words from S. Yakhontovs 100-wordlist, taken from
the second part of the Swadesh, 200-item wordlist (see / ,
2005, 1213 for detail). Among several important innovations of the GLD, a
strong semantic standard, which is currently used instead of the blind list of
English denitions proposed by M. Swadesh in the mid-20
th
century, must be
noted (see Kassian et al., 2010).
For the general principles of lexicostatistical procedure, the ways of its
application to specic languages and the difference between preliminary lexico-
statistics and classical lexicostatistics, see Starostin G., 2010a. As I have stated
elsewhere (Kassian, 2010b, 200 ff. ; Kassian, 2011a, 142 f.), a fact important for
future discussion is that any pair of languages conventionally assumed to be ge-
netically related at a reasonable time depth possesses a signicant number of
etymological matches with identical meanings between their basic and, most im-
portantly, core vocabularies summarized as the Swadesh wordlist.
8
That is, lexi-
costatistics is a reliable tool for language relationship tests and, what is more,
the presence of etymological matches with coinciding semantics between Swa-
desh wordlists of two languages (or proto-languages) is a necessary condition of
recognizing genetic relationship between languages.
9

Below I offer a 110-word list for Hurrian and Urartian languages as it can be
compiled on the basis of our current knowledge. Ca. 65 out of 110 slots can be
lled for Hurrian with reasonable certainty, and ca. 22 for Urartian. Despite such
scantiness, the available data seem to be sufcient for some conclusions.
Because the genetic afliation of the HU family is not established yet, only a
preliminary lexicostatistical (to use the terminology of Starostin G., 2010a)
verycation/ falsication can be undertaken. This means that, lacking the
knowledge of regular phonetic correspondences between HU and other linguis-
tic families, we are compelled to resort to the phonetic similarity between HU
forms and words of external (proto-)languages included into comparison.
Phonetic similarity is not a fully formalized concept. Below, as in the case of
Hattic etymologization (Kassian, 2010a), I prefer to use consonantal classes (cf.
Dolgopolsky, 1986; Turchin et al., 2010 w. lit. for detail),
10
on the one hand, and

8
More precisely: neither I myself nor any of my colleagues from the Moscow school are
aware of even a single reliable exception from this phenomenological rule.
9
It should be emphasized explicitly that a pair of etymological cognates whose meanings
are different between two compared languages constitutes positive evidence for etymo-
statistics a very different and, at such time depths, much less reliable procedure but
not for lexicostatistcs. For the latter, only direct phonetic and semantic correspondences
are taken into account. Thus, Lat. nox night ~ Grk. night or Soqotri ed hand ~
Arabic yad- hand are positive lexicostatistical pairs, whereas Lat. nox night ~ Hitt.
neguz evening, nightfall or Soqotri ed hand ~ Akkadian idu arm are not.
10
In fact, my classication is more strict. E.g., I treat r-like and t-like phonemes as two
different classes; the same concerns /-like, y-like and w-like sounds, which also should
represent different classes. Vowel quality is standardly ignored in such a procedure (i. e.,
all vowels are treated as one single class) because of the well-known instability of vo-
more complicated (albeit typologically common) phonetic shifts,
11
on the other.
As opposed to Baxter/ Manaster-Ramer, 2000, and some other similar studies,
where only an initial consonant of the root is taken into account, I follow
Turchin et al., 2010, in the present paper and adopt a more rigorous approach,
accepting that both consonants (initial and nal) of the roots compared must
comply with the principle of phonetic similarity.
Using data collected in the Global Lexicostatistical Database and the Tower
of Babel projects, I have compared the HU list to the forms that can be recon-
structed with the same Swadesh semantics for some linguistic families of the
Old World, namely: Indo-European, Kartvelian, Uralic, Altaic, Dravidian, Se-
mitic, Egyptian, Chadic, Berber, Cushitic and Omotic. The actual result nears
zero. Indeed, some HU roots show a certain phonetic similarity to forms of other
linguistic families, but these cases are too scanty and dispersed and look pro-
babilistically conditioned.
12

Cf. the following hypothetical isoglosses:
Hurr. ini 2 ~ Semitic *Vny 2 (a Semitic loanword in Hurrian? See
notes on 91. two below);
Hurr. imi(=)gi sun ~ Semitic *
x
Vm
x
(< redupl. *
x
Vm-
x
Vm?) sun.
Hurr. tari re ~ Uralic *tule re;
Urart. iwi(=)ni sun ~ IE *sw-el- / *sl- / *sw-en- sun;
Hurr. tali tree ~ IE *doru tree;
Hurr. iwe / iye water ~ Altaic *siuba water (one of two equal candi-
dates for the status of the Proto-Altaic term for water) ;
Hurr. uzi meat ~ Omotic *a- meat.
On the contrary, one can conclude in anticipation that comparison to linguis-
tic families, which constitute the Sino-Caucasian macro-family
13
(namely, North

calism in human languages (thus, there are, e. g., more than 220(!) rows of vowel corre-
spondences between such close languages as Modern English and Modern German, see
Dybo A. / Starostin G., 2008, 145 ff.). Note that, in accordance with this approach, roots
of the shape ta or at possess the normal CVC-structure, namely T- and -T respectively.
Therefore it is possible to compare a hypothetical root ta (from the language L
1
) to ta
(from the language L
2
), but comparison between roots ta (language L
1
) and at (language
L
2
) is much less justied.
11
Like, e. g., the consonantal cluster simplication ns > s, r > or the weakening > z.
12
Cf. Ringe, 1999, on the theoretical ease of nding isolated comparanda for any CVC-
root when many languages are compared simultaneously. Note also that the correspond-
ence r ~ l (as in a few hypothetical HU-IE and HU-Uralic matches below) does not fully
comply with the principle of phonetic similarity.
13
For a brief sketch of the history of Sino-Caucasian studies, see now Starostin G.,
2010b and esp. Bengtson/ Starostin G. forthc. The theory of a Sino-Caucasian linguistic
macro-family has partially been substantiated by the late S. Starostin. For the compara-
tive phonetics of the Sino-Caucasian macro-family, see S. Starostins SCC (this work
was not nished and therefore remains unpublished). The highly preliminary Sino-Cau-
casian etymological dictionary by S. Starostin is available as Sccet.dbf. Some other pa-
Caucasian,
14
Yeniseian,
15
Sino-Tibetan
16
together with Basque,
17
Burushaski
18

and apparently Na-Dene
19
), seems to yield some positive results. Below, I dis-
cuss possible Sino-Caucasian comparanda in the notes on HU items; for the
concluding discussion, see 5.

3. It is obvious that, during the millennium that attested Hurrian was spoken,
there were a number of Hurrian dialects that should differ in their basic
vocabularies.
20
Due to the scantiness of data, it is impossible, however, to adhere
to the principle of chronological and dialectal integrity (for which see Kassian,
2011b, 66 f.). The Hurrian list below is cumulative, that is, it includes all availa-
ble forms from any dialects. The following sigla are used for Hurrian sources.
The so-called Old Hurrian period: Tiatal (the royal inscription of Tiatal,
endan of Urke), OBab. (ca. 10 tablets with invocations, presumably all from
the site of Larsa in southern Babylonia), Mari (6 tablets from Mari), Bo.Bil. (the
Hurrian-Hittite bilingual series Song of Release).
The so-called New Hurrian period: Mit. (the Mittani Letter), Bogh. (texts
from the Hittite capitals archives, modern Boazky, other than Bo.Bil.),
Ugar.A. and Ugar.C. (texts from the Ugaritic archives written in the alphabetic
or syllabic cuneiform script respectively), Nuzi, Qatna, Msk., Ort., Tell Brak
(texts from Nuzi, Qana, Tell Meskene [Emar], Ortaky and Tell Brak respec-
tively), Subar. (Subarian glosses in Assyro-Babylonian sources).
Hurrian and Urartian forms are quoted in the traditional cuneiform transcrip-
tion, e. g., the main Hurrian and Urartian sibilant is rendered as , geminated

pers by the same author, dedicated to the Sino-Caucasian problem, can be found in -
, 2007 (in both Russian and English). A comparative grammar overview of the
Sino-Caucasian macrofamily can now be found in Bengtson/ Starostin G. forthc. A for-
mal (lexicostatistical) verication of the Sino-Caucasian theory is currently in prepa-
ration for publication as part of the Moscow-based Global Lexicostatistical Database
and Tower of Babel projects and the broader Evolution of Human Language (EHL) pro-
ject, centered around the Santa Fe Institute. For current major results and conclusions of
the EHL, see Gell-Mann et al., 2009.
14
See NCED and Caucet.dbf.
15
See , 1982/ 2007, and Yenet.dbf, based on , 1995, and Werner,
2002 with additions and corrections.
16
See Stibet.dbf, based on Peiros/ Starostin, 1996, but seriously emended.
17
See Basqet.dbf and corresponding sections in Bengtson, 2008.
18
See Buruet.dbf and some recent publications as, e. g., Bengtson, 2008a; Bengtson/
Blaek, 2011.
19
Proto-Na-Dene reconstruction is not done (or not published) yet. Cf. some rather
preliminary publications on the supposed Sino-Caucasian afliation of the Na-Dene
family: Nikolaev, 1991; Bengtson, 2008b. For the criticism of the so-called Dene-Ye-
niseian hypothesis, see Starostin G., 2010b.
20
Cf. notes on 34. good, 62. not, 92. to go below.
consonants as double letters (tt, kk, etc.),
21
velar fricatives as h/ hh, front affri-
cates as z/ zz, and so on. Normally, I do not discriminate between two Hurrian u-
phonemes (u & o), because this difference is irrelevant for my investigation. The
problem of inventory and phonetic realisation of Hurrian labial consonants is not
solved so far; p, f & w in the Hurrian forms below are used accordingly to the
Hurritological tradition. According to the same practice, the sign = denotes
morphemic boundaries in Hurrian and Urartian forms, whereas the hyphen - is
retained as a connector between signs in narrow transliteration.
All Sino-Caucasian forms are generally quoted from the Tower of Babel pro-
ject databases (Sccet.dbf, Caucet.dbf, Stibet.dbf, Yenet.dbf, Basqet.dbf, Bu-
ruet.dbf see the list of references).
22
Sumerian forms are quoted from the
ePSD.
In the list below, the titles of entries contain Hurrian Swadesh words without
special remarks, whereas Urartian forms (when known) are cited in separate sec-
tions and explicitly glossed as Urartian.
1. all (omnis): ua=lla
Sg. ue ~ ua=nna all (totus) [Mit., Bogh.] / pl. ua=lla all (omnis) [Mit.].
GLH, 240; Dietrich/ Mayer, 2010, 204, 261; Wegner, 2007, 84; Girbal, 1994,
175.
The pl. stem ua=lla is apparently unattested in Bogh., but there exists another
word for all in Bogh. texts: ummi all (totus/ omnis) (Girbal, 1994; Weg-
ner, 2007, 84), collocations: all deities, whole sin, whole body. The
difference between Bogh. ue and ummi is unclear, but the latter is less fre-
quent and seems more marginal. Cf. also the verbal stem umm-, which per-
haps means to assemble and therefore can be related to ummi, Campbell,
2007, 171. Both terms (ua=lla and u=m=mi) are apparently etymologically
related; double mm can originate from something like n-m.
A third word for all attested in Bogh. texts is heyari, pl. heyar=unna (GLH,
101), but this term is contextually bound and normally applied to deities only
(all gods).
Urartian: sg. ui=ni- totus / pl. uy=a- omnis (, 465; CdTU 2, 200;
Andr-Salvini / Salvini, 2002, 14 f.). An etymological cognate of the main
Hurrian term.
It could be possible that HU *u-i all contain the same root as the Hurrian
numeral ukki 1 (No. 63 below), if u-kki, but note that the derivation
all > 1 seems typologically problematic.
23

21
For a possible reconstruction of the HU obstruent series, see Khachikyan, 2009.
22
The system of transcription is the same as in Kassian, 2010a, 433 f.
23
Diakonoff / Starostin, 1986, 38, compare HU *u-i to several ECauc. forms for every,
everything, all, which are derived from the corresponding Lezgh., Andian and Tsezian
proto-forms for 1 (all originating from NCauc. *cH
1).
2. ashes: al=mi
Ugar.C.
Andr-Salvini / Salvini, 1998, 15, Huehnergard, 1987/ 2008, 378.
The nal =mi is a nominal sufx, for which see Wegner, 2007, 59. Cf. the verbal
root alm- (Msk.) to burn(?) (Andr-Salvini / Salvini, 1998, 15). Cf. also the
noun ulli (Ugar.C.) charcoal (Andr-Salvini / Salvini, 1998, 17).
Semantically and phonetically, it is possible to compare Hurr. al- to Sino-
Cauc. * Hu/ * Hu dirt, dust, earth, ground > NCauc. * Hu/ *Hu
dirt, dust, earth, ground, sand (> Nakh *()il (~ --) ashes, dust, Av.-And.
*VlV silt, slime, Lak aIlu/ aI- earth, ground, Lezgh. *il earth;
oor), Basque *olho meadow; eld; eld (prepared for sowing). See
Kassian, 2010a, 355 for more detail (adding Hattic ahhu/ tahhu ground,
bottom). No lexicostatistical matches, however.

3. bark: unattested.

4. belly: unattested.
The word kari is glossed as belly in GLH, 137, but further textual analysis
does not support this interpretation, see Girbal, 1994, 173 (as lip(s)? in
Wegner, 1995a, 118, 119). If the body part denoted by Hurr. kari is indeed
belly, this word is an Akkadian borrowing (< Akkad. karu belly).

5. big: tal=mi, tal=a=mi
Mit., Ugar.C., Bogh.
GLH, 253; Wegner, 2007, 282; Dietrich/ Mayer, 2010, 263; Huehnergard,
1987/ 2008, 381; Andr-Salvini / Salvini, 1998, 18.
For other formation from the root tal- with the supposed meaning big (vel
sim.), see Wegner, 1995b, 119; Catsanicos, 1996, 210, 214, 274, 256. For
the nominal sufx =mi, see Wegner, 2007, 59.
A second candidate for big is awoi attested once in Bo.Bil. (Wegner, 1995b,
118; Neu, 1996, 254; Catsanicos, 1996, 281), but this word is too marginal
and should rather be translated as great (great Teob), not as the generic
big.
Cf. the verb te- to be many (Mit.), see sub 52. many. Cf. also the verb al-
(Mit.) to be marvelous (vel sim. ?) (Dietrich/ Mayer, 2010, 234),
24
which
may correspond to Urart. al(a)sui great (, 433; Dietrich/ Mayer,
2010, 280).
No obvious Sino-Cauc. etymology. Cf. STib. *t[]r numerous (> Chin.
*tr numerous, in a ock, Tib. ther-bum 1,000,000,000, Lushai tuar in
great quantities, in large numbers), which is compared in Sccet.dbf with Bu-

24
Such an interpretation is textually possible, but not obligatory. Left without translation
in Wilhelm, 1992b, 66; Giorgieri, 1999, 382; Wilhelm, 2006, 184.
rush. *tVrap full.

6. bird: eradi
Ugar.C., Bogh.
Wegner, 2007, 255; Andr-Salvini / Salvini, 1998, 13; Haas/ Wegner, 2004, 341;
Grke, 2010, 336.
The word can be analyzed as er=adi with a nominal sufx (thus Wegner, 2007,
53).
No obvious etymology.

7. to bite: unattested.

8. black: time=ri ~ tima=ri
Bo.Bil., Bogh.
Neu, 1996, 246; Catsanicos, 1996, 227; Wegner, 2007, 53, 285.
Attested in the collocation timeri ee = Hitt. tankui tegan the Underworld, lit.
black earth, not dark earth, since the primary and basic meaning of Hitt.
tankui- (= Sum. GIGGI / GE
6
= Akkad. almu) is black, not merely dark.
Besides, the attribute black is more natural for a poetic phrase than the neu-
tral characteristics dark. The element -r- seems a synchronic sufx.
Hurr. tim- should be compared to the basic Yen. term *tum- black (> Ket
tm black, Yug tum black, Kott. thum black). Further, following Sccet.
dbf (sub Sino-Cauc. *rVmV black, dark), one can add here the poorly at-
tested STib. *rVm dark. In both Proto-Yen. and Proto-HU the initial r- was
indeed prohibited, the shift Sino-Cauc. *r- > Yen. *t- is regular (SCC, 26).
Lexicostatistical match: Yen.

9. blood: zur=gi
Bogh., Msk.
GLH, 309; Haas, 1998a, 254.
The nal =gi can be analysed as the sufx =ki, for which see Wegner, 1988,
149 f. ; Wegner, 2007, 5657, 59; Diakonoff/ Starostin, 1986, 69. Wegner
treats =ki as an instrumental and resultative sufx and =ka as honoricative
or diminutive, whereas Diakonoff & Starostin treat both =ki and =ka as
diminutive (a global diminutive in -k, attested in many language families
around the world).
Hurr. zur- should be compared to the basic Yen. term *sur blood (> Ket ,
Yug sur, Kott. ur, Arin sur, all meaning blood). Further Sccet.dbf (sub
Sino-Cauc. * wV) adds here NCauc. * wV vital force (vel sim.)
(> Nakh *gi blood, Av.-And. * agV- alive, Lezgh. * VV- alive;
lively, animated) and Basque *i-serdi sweat ; sap. Note that it is not a basic
word for blood in Proto-NCauc. as well as in ProtoSino-Cauc., although in
Yen. and in the Nakh subbranch of NCauc. the meaning blood occurs
secondarily and independently.
25

10. bone: unattested.

11. breast: neher=ni
Bo.Bil. (brisket), Ugar.C. (chest). The previous reading of this word is
aherni.
Catsanicos, 1996, 280; Huehnergard, 1987/ 2008, 382; Andr-Salvini / Salvini,
1998, 14; Wegner, 1995a, 121; Wegner, 2007, 269; Haas, 1993, 268.
Cf. also the nursery word zizzi female breast (GLH, 306). The anatomical term
irde means tongue (No. 88 below), not breast pace GLH, 125. The word
huri (Ugar.C.) denotes lap, crotch (vel sim.) (Andr-Salvini / Salvini, 1998,
12), not breast pace Patri, 2009, 347.
As noted in , 1998/ 2007, 749, Hurr. neher somewhat resembles
the basic NCauc. term *mnq (~ -) breast (> Nakh *nqa breast, Av-
And. *niqi (~ --) breast, Tsezian *mVrV (~ -i-) breast, Lak quru bris-
ket, Dargwa *miqiri breast, Lezgh. *moor breast, WCauc. *mVq
a
arm; bosom; apparently the basic Proto-NCauc. term for breast).
26

Phonetical details of the Hurrian-NCauc. comparison are obscure.

12. to burn (trans.): am-
Bo.Bil.
LHL 1, 70; Wegner, 2007, 246; Catsanicos, 1996, 232.
For various derivative stems such as am=b- to burn (intr.) or am=b=a=nni
rewood, see LHL 1, 70.
Cf. also the verb alm- (Msk.) to burn(?), see notes on 2. ashes.
Urartian: a good candidate is the verb am-, am=at- to burn off, down
(, 434; CdTU 2, 4344; Andr-Salvini / Salvini, 1999b, 272273). An
etymological cognate of the Hurrian term.
No obvious etymology for HU *am- to burn. It resembles a not very relia-
ble STib. root *Vm to blaze, glow (vel sim.) (> Chin. *wam(?) to
blaze, burn, Burm. um to glow, smoulder, Lushai vm (vam) to be in a
red-hot glowing condition (as re), Lepcha om to shine, Kiranti *[w]m
(/-p) to burn). , 1998/ 2007, 746, compares the HU *am- with a
rather questionable NCauc. root *mhlV- ~ *mhnV- warm that does not
seem apt.

13. ngernail : unattested.

25
Diakonoff / Starostin, 1986, 44, 69 compare Hurr. zur- directly to NCauc. * wV.
26
Further, in Sccet.dbf, the NCauc. stem *mnq is compared with STib. *ni (~ -i-)
(s-) heart (vel sim.) and Yen. *an[] mind, but it seems doubtful.
14. cloud: unattested.

15. cold: unattested.
Pace Diakonoff (Diakonoff/ Starostin, 1986, 60) and some other authors, Hurr.
egi means inside, interior, eg=ui means pure (see, e. g., Wegner, 2004,
19; Trmouille, 2005, 314). A phonetical coincidence with Hittite ega- cold
(n.), frost, ice, eguna- cold (adj.) is accidental.

16. come: un-
Bogh., Mit., Ugar.A.
GLH, 282; Wegner, 2007, 289 f. ; Dietrich/ Mayer, 1994, 87, 90; Dijkstra,
1993a, 160; Strau, 2001, 414; Grke, 2010, 342.
Polysemy: to come (intr.) / to bring (tr.). The verb is opposed to itt- to go
(away) (No. 92 below), GLH, 282. Hurr. un- may be related to Urart.
nun=a- to come (see below), but details are unclear (a reduplication n=un
< *un-un, as proposed in Diakonoff/ Starostin, 1986, 32?).
The second candidate is amm- (Bo.Bil., Bogh., Mit.) (LHL 1, 76; Neu, 1996,
118; Catsanicos, 1996, 228, 278; Wegner, 2007, 247), but its basic meaning
is rather to arrive (corresponds to Hittite ar- to arrive in Bo.Bil.); glossed
as werden zu in Dietrich/ Mayer, 2010, 234.
Urartian: it is not easy to make a substantiated choice between known Urartian
verbs of movement. Main candidates for the generic meanings to go and to
come are the following three verbs that occur most frequently:
1) nun=a-, , 456 (to come, arrive); CdTU 2, 163164 (venire);
2) ul-, ul=a-, , 471 (to go); CdTU 2, 227, 229 (uladi, uli an-
dare);
3) u-, u=t=a- (a cognate of Hurr. u- to go, No. 92), traditionally
translated as to depart, take the eld, , 1963, 92;
Haas/ Wegner, 2010, 101; , 474; CdTU 2, 239 ff. (partire (in
guerra)).
The verbs nun=a- and ul=a- correspond to the basic Assyrian verb for to go / to
come, alku, in the bilingual inscriptions. In its turn, u=t=a- is a close
synonym of ul=a-, both verbs may occur in one inscription in identical con-
texts.
In some passages each of these verbs can be equally well translated as either to
come or to go, but in a number of contexts one of the meanings is clearly
preferable.
27
Proceeding from this, I treat nun=a- as the basic Urartian verb
for to come, whereas ul-, ul=a- and u-, u=t=a- are provisionally regarded

27
Namely to come for nun=a- (e. g., 30 Ur. 17 = CdTU A 3-11 obv. 17) ; to go,
to go away for ul-, ul=a- (e. g., 173 VIII 8 = CdTU A 8-3 VIII 8) ; to go, to go
away for u-, u=t=a- (e. g., 56 6 = CdTU A 5-5 6; 446 10 = Haas/
Wegner, 2010, 101).
as synonyms for the general meaning to go (No. 92).
Cf. also some other Urartian verbs of movement, which are less frequent or pos-
sess more specic meanings:
i=a-, , 463 (to go, come), CdTU 2, 189 (andare, venire).
a=a-, , 437 (to go(?)), CdTU 2, 5152 (aadi ripartire, andare,
intraprendere una campagna militare), Diakonoff/ Starostin, 1986, 38
(to ee, leave).
ku=u-, , 452 (to reach), CdTU 2, 140 f. (pervenire).
Hurr. un- to come (and Urart. nun?) can be compared to Sino-Cauc.
*=VwV to go > NCauc. *=VwVn to go (vel sim.), STib. * (s-, -)
to go, Yen. *he-jV to go, Burush. *n- to go, ? Basque *e-oHa-n to
go.
28
This verb seems a default expression for to go in Yen. & Burush.,
probably in STib. and theoretically in NCauc., therefore it is a good candi-
date for the status of the ProtoSino-Cauc. basic term to go. The meaning
to come occasionally occurs for this root in daughter languages, but cannot
be projected onto the proto-level.
As for the Urart. verb a- (a=a-), it is compared by Diakonoff/ Starostin,
1986, 38 & , 1995/ 2007, 630 to NCauc. *=iA (a verb of move-
ment), which is theoretically possible.

17. to die: unattested.

18. dog: erwi ~ erbi
Bo.Bil., Bogh. Ugar.C.
GLH, 83; Wegner, 2007, 255; Catsanicos, 1996, 279; Andr-Salvini / Salvini,
1998, 11.
No obvious etymology. Resembles the main Sumerian term for dog: UR.
19. to drink: al-
Bogh.
LHL 1, 51.
Urartian: Diakonoff (Diakonoff/ Starostin, 1986, 60) singles out an Urartian
root *hur- to drink from the obscure substantive hurihi (, 447 one
who irrigates, moistens(?); CdTU 2, 103 abundance, fertility(?)), which
seems unjustied.
No obvious etymology for Hurr. al-.

20. dry: unattested.
Cf. the participle ib=a (Bo.Bil. ; said of the exhausted god Teob), which

28
In Diakonoff / Starostin, 1986, 32, Hurr. un- & Urart. n-un are compared directly to
NCauc. *=VwVn (> Nakh *o- to go (vel sim.), Av.-And. *=VVn- to go, Tsezian
*=
- to come, Lak na- to go (vel sim.)).

means something like ruined (thus CHD P, 62a) or exhausted, but could
be narrowly translated as dried up, dried out, see Neu, 1996, 316319;
Catsanicos, 1996, 275 fn. 220 (comparing to Akkad. abbu to burn (of
eyes, lips, etc.)); Hoffner, 1998, 7475 (dried out (? from the heat of the
day)).

21. ear: nui ~ nuhi
Bo.Bil. (nui), Bogh. (nui ~ nuhi), Tell Brak (nui).
Wegner, 1995a, 121 f. ; Catsanicos, 1996, 273; Grke, 2010, 71 f. ; Wilhelm,
1991a, 166.
Polysemy: ear/ understanding, mind. Perhaps /nuHi/, see notes on 25. eye for
the graphical variant nuhi.
Distinct from hazzizzi inner ear, hearing, intelligence (GLH, 100) < Akkad.
assu ear (anatomic); wisdom.
The meaning of Hurr. leli- (Bogh.) is uncertain; the collocation leli haari
apparently denotes a deied object (with haari oil or strength
29
?), but leli
can hardly mean ear here, pace Diakonoff/ Starostin, 1986, 53. This word
does not seem inherited because of the initial l-.
Should be compared to the basic STib. term *n
H ear > Chin. *nh,

Tib. r-na, Burm. nah, Kachin na
1
, Lushai (KC) *k-Na, Lepcha njor, Kiranti
*, all meaning ear.
30

Lexicostatistical match: STib.

22. earth (soil): ee
OBab., Bo.Bil., Ugar.A. (e), Ugar.C., Mit., Bogh., Ort.
Wegner, 2007, 256; Catsanicos, 1996, 230 f. ; Dietrich/ Mayer, 2010, 240; Die-
trich/ Mayer, 1994, 87; Prechel / Richter, 2001, 359; nal, 1998, 60; Wil-
helm, 1993, 111 ff. ; Grke, 2010, 336; GLH, 83 f. (erroneously glossed as
sky).
Polysemy: earth (world) / earth (soil) / position / place. The specic meaning
soil can be seen in the collocation timeri ee = Hitt. tankui tegan the Un-
derworld, literally black earth. A passage from an Old Babylonian incan-
tation (Prechel / Richter, 2001, 347 ff.) also seems to conrm the specic
meaning earth, soil. The Hurrian word corresponds to Urart. esi place
(, 443). Note that the phonetic correspondence Hurr. ~ Urart. s is
very infrequent, if not unique. Excepting the certain ee ~ esi, I am aware of

29
Thus Wegner, 1981, 107.
30
Sccet.dbf sub *wnV adds here Yen. *-gde/ *-qtV ear and NCauc. *wnV
ear (vel sim.). Such an etymologization of the Yen. term seems too factitious and should
be rejected, but the NCauc. *wnV could indeed be connected to STib. *nH (note
that *wnV is not the main Proto-NCauc. term for ear, the meaning of this proto-
form was rather inner ear / hearing). In such a case, a metathesis n~w should be as-
sumed for the NCauc. proto-form or for the STib. & Hurr. ones.
two potential instances: Hurr. ali daughter ~ Urart. sila wife
31
(a parallel
semantic development from some source like, e. g., *lady is not impossible,
but the vowel difference is suspicious; Salvini, 1998b, 9899, may be right
assuming two etymologically unreleated words here) and Hurr. al- to be
marvelous (vel sim. ?) ~ Urart. al(a)sui great (if the Hurr. verb indeed
means this, cf. notes on 5. big above). The set of Hurrian-Urartian lexical
correspondences is too small, however, for far-reaching conclusions.
32

On the other hand, M. Khachikyan (pers. comm.) suspects that there might exist
an Urartian word ei earth: cf. CdTU A 12-8 1722 = 421 1722
pili nikiduli A
ME
eia iuli when the canal overows
?
, when
the waters ow
?
upon the earth
?
, partly paralleled by CdTU A 11-1 rev. 25 =
407 rev. 25.
33

Cf. also umi=ni country, land (GLH, 281).
Urartian: iura- ~ ira- earth (world); earth (soil); land (, 459; CdTU
2, 178). See also notes on the hypothetical Urart. ei earth above.
The Proto-HU term for earth (soil) was apparently *eSi, because the mean-
ing shift earth (soil) > ground > place is natural, whereas place > earth
(soil) seems typologically odd.
No obvious etymology for HU *eSi. Cf. the following proto-roots:
Sino-Cauc. *jVmV soil (vel sim.) > NCauc. *jmV (> Nakh *jobs
(~ --) earth, Dargwa *ani earth, Lezgh. *jo(m)V (~ -) dirt,
Khin. ini earth, probably it is not the main Proto-NCauc. term for
earth), Yen. *e- (~ x-, --) damp sand, Basque *hau ashes.
NCauc. *en(V)sw soil, clay (vel sim.) (> Av.-And. *oni earth; mud,
Lak aIri earth, Dargwa *he() (~ -) dust (of rotten wood), WCauc.
*n clay, apparently it is not the main Proto-NCauc. term for
earth); proposed in , 1995/ 2007, 631.
Less probable is comparison with Lezghian *jis place; near (proposed by
Diakonoff/ Starostin, 1986, 39, for Urart. esi place), because this root is
isolated within the NCauc. family.
For Urart. iura- ~ ira-, Diakonoff/ Starostin, 1986, 58, propose the follow-
ing cognate NCauc. *qwIrV/ *rIqwV eld, arable land (> Nakh *qaw
arable land, eld, Av-And. *qurHV eld, Tsezian *u- (~ *-) eld,
Lak qu eld, Dargwa *qu eld, Lezgh. *uj eld, WCauc. *rq()a
(arable) eld),
34
which seems theoretically possible.

31
As has been demonstrated in Salvini, 1998b, 9899, Urart. sila must be translated
wife, not daughter (cf., however, Wilhelm, 2008b, 112 with the traditional translation
daughter).
32
See Wilhelm, 2008b, 121 f. for the most reliable HU cognate words.
33
On the traditional reading eiaiuli at the end of the passage cited, see , 443
w. lit. ; Salvini, 2002, 143 (to decrease
?
). In fact, the assumed verbal form eiaiuli
looks somewhat atypical, therefore it could be better analyzed as two separate words.
34
Sccet.dbf (sub *qwrV) adds some dubious STib. and Basque comparanda.
23. to eat: ul-
Bo.Bil.
Catsanicos, 1996, 216.
The Hurrian verb refers to a dog (ul=an=u=m (the dog) began to eat (the
bread)) and could therefore mean to devour rather than the neutral to eat.
But since it is translated by the basic Hitt. verb ed- to eat (of humans), not
by Hitt. karap- to eat (of animals), I tentatively consider ul- to be the de-
fault Hurrian verb to eat.
Cf. also the scarcely attested root *a- to eat (?), LHL 1, 117.
Following , 1998/ 2007, 751, one can compare Hurr. ul- to Sino-
Cauc. *=wVl- to feed on (vel sim.) > NCauc. *=iwVl to feed on, Yen.
*ir- (~ x-) to eat (vel sim.), Basque *aha to feed, graze, ? STib.
*q(h)ir present of food; appetite. Note that it is not the basic term for to
eat in the aforementioned families.
As for Hurr. a- to eat (?), , 1998/ 2007, 746 proposes to com-
pare it to NCauc. *=Vc V to drink; to gulp, to eat (with further Sino-Cauc.
cognates), which is theoretically possible.

24. egg: unattested.

25. eye: i ~ ihi
Bo.Bil. (i), Bogh. (i ~ ihe).
Catsanicos, 1996, 231, 270; Wegner, 1995a, 124 f. ; Wilhelm, 1998a, 174; Gr-
ke, 2010, 71, 76.
The nature of the graphical uctuation 0/ ~ h is unclear, cf. the same phenomenon
in the word for ear nui ~ nuhi (No. 21) and tae ~ tahe man (No. 51). Wil-
helm, 1998a, 174 may be right in suspecting here bisyllabic stems with a spe-
cial laryngeal phoneme: /siHi/ and, correspondingly, /nuHi/, /taHi/ (cf. also
discussions in Giorgieri, 2000, 188 fn. 50; Wegner, 2007, 49).
The previous candidate for eye, pai-, actually denotes mouth. The noun furi
( fur- to see, to look, see No. 72 below) more probably means something
like a look, not eye per se (Wegner, 2007, 274; Wilhelm, 1998a, 174;
Dietrich/ Mayer, 2010, 272), although furi is translated as eye in an Ugaritic
lexical list (Andr-Salvini / Salvini, 1998, 22). The word aye [Mit.] is glossed
as Auge; Gesicht in Dietrich/ Mayer, 2010, 233, but in fact Auge is a
loose translation of awi [Bo.Bil., Bogh., Msk.] ~ aye [Mit.] face; in front of
(Wegner, 2007, 248; GLH, 34, 40; Catsanicos, 1996, 200, 278).
No obvious etymology. , 1995/ 2007, 632 compares it to NCauc.
*=VrwEn (~ -\-) to look, although it is phonetically uncertain.
26. fat (noun): ae
Bo.Bil.
Catsanicos, 1996, 278; Neu, 1996, 115; Haas, 1998b, 143; Wegner, 2007, 250;
LHL 1, 117.
The fat of a killed deer is mentioned in the text. The Hurrian term corresponds to
the ideogram fat, oil in the Hittite version, therefore the literal translation
meat is impossible (cf. especially Haas, 1998b).
Cf. the verb ed=u- to fatten (trans.), make fat (Bo.Bil.), Catsanicos, 1996,
282; Neu, 1996, 112. Cf. also the word ha=ar=i (ne) oil (Bo.Bil.,
Bogh.), from the verbal root ha- to oil, smear (Wegner, 2007, 259; Catsa-
nicos, 1996, 279).
, 1998/ 2007, 746 compares Hurr. ae to the following NCauc.
root : Sino-Cauc. *=(rV) thick > NCauc. *=(rV) thick, fat (adj.),
Yen. *bV-sVr thick (of paper, skin, cloth etc.), Burush. *be-s fat (n.).
Such an etymologization seems theoretically possible (cf. also notes on 53.
meat below).

27. feather: unattested.

28. re: tari
Bo.Bil., Mit., Bogh., Ugar.C.
Catsanicos, 1996, 231; GLH, 257; Huehnergard, 1987/ 2008, 378; Andr-Sal-
vini / Salvini, 1998, 18.
35

29. sh: unattested.

30. to y: unattested.

31. foot: uri ~ ur=ni
OBab., Bo.Bil., Bogh., Ugar.C.
Wegner, 2007, 291; Wegner, 1995a, 125 f. ; Catsanicos, 1996, 204; Huehner-
gard, 1987/ 2008, 379; Andr-Salvini / Salvini, 1998, 21; Andr-Salvini / Sal-
vini, 1999, 272; Prechel / Richter, 2001, 356.
Polysemy: foot / leg; corresponds to Akkadian pnu ~ pmu thigh & Sume-
rian UR
2
thigh in lexical lists from Ugarit.
The similar noun ugri- (Bogh.) is attested in the meaning leg of the table
(GLH, 277). Cf. also the term ziyan=ni lap, hip, crotch (vel. sim) (Ugar.C.),
Andr-Salvini / Salvini, 1998, 24.
Urartian: kuri (, 452; CdTU 2, 137); attested several times in the stereo-
typical context he embraced rulers feet and prostrated himself. The re-
lationship between Hurr. uri, Hurr. ugri and Urart. kuri is unclear.
36
Wilhelm,

35
Ivanovs idea (Ivanov, 1998, 148 fn. 17) that Hurr. tari is an Iranian loanword (< Ir.
*tr ~ *tar re) is impossible for various reasons.
36
For Hurr. uri vs. ugri cf., e. g., Neu, 1996, 355.
2008b, 106 supposes that the Urartian spelling ku-ri- may cover phonetical
ukri, because the sign UK was not used in the Urartian cuneiform system.
37

No obvious etymology for Hurr. uri. Resembles Sumerian UR
2
thigh; loin,
lap; base, legs of a table.
Urart. kuri (and Hurr. ugri ?) can be compared with Sino-Cauc. *wrV a
part of the leg > NCauc. *wrV a part of the leg (> Nakh *urV-m bone
for playing dice, Lezgh. *ir(a) hoof; leg (of animal)), STib. *k(h)rej
(~ -j) a part of the leg, Burush. *kur- joint of nger or toe. Note that
these proto-roots are rather scantly attested in the corresponding families,
therefore the etymological nest looks dubious. The HU-NCauc. comparison
has been proposed in Diakonoff/ Starostin, 1986, 57.

32. full : unattested.
Cf. the verb kapp- (Bo.Bil.) to ll (Catsanicos, 1996, 215; Wegner, 2007, 261).

33. to give: ar-
Bo.Bil., Mit., Bogh., Ugar.C., Tell Brak.
LHL 1, 96; Wegner, 2007, 248; GLH, 52; Huehnergard, 1987/ 2008, 3435, 70;
Catsanicos, 1996, 245; Grke, 2010, 335.
Urartian: the best candidate is ar-, ar=u- and apparently ar=d=u- (,
435, 436 to give, grant, present; CdTU 2, 46, 50 dare, concedere), which
corresponds to the basic Assyrian and Sumerian terms for to give in bilin-
gual inscriptions (respectively nadnu and SUM). The second candidate is the
stem uhan=u-, , 473 f. (to grant (?)); CdTU 2, 235 (conferire).
Diakonoff/ Starostin, 1986, 51 compare HU *ar- to give to NCauc. *= V
to give (a good candidate for the status of the basic Proto-NCauc. term for
to give), which is phonetically uncertain.

34. good: wahri ~ wahr=ui
Bo.Bil., Mit., Bogh., Ugar.C., Ugar.A., Qatna, Msk.
GLH, 292; Wegner, 2007, 270; Andr-Salvini / Salvini, 1998, 22; Catsanicos,
1996, 281; Salvini / Wegner, 2004, 177; Richter, 2005, 27.
This frequent term seems the most basic and neutral for the meaning good / to
be good. Corresponds to the basic Akkad. term bu good in a lexical list
from Ugarit.
The second candidate is niri (Mit., Bogh., Ugar.C., Msk.) good / to be good
(GLH, 185; Wegner, 2007, 269), whose usage seems more abstract (e. g.,

37
Another instances of such a method could be the Urartian royal name Rusa (Urart. ru-
sa), which is spelled in Assyrian texts both as ru-sa-a or ur-sa-a (, 487; Salvini,
2007, 464a). Note that the HU phonotactics does not normally allow [r] in the initial po-
sition, but, as opposed to UK, the phonetical sign UR was known to Urartian scribes. An
unclear situation (Wilhelm, 2008b, 108, suspects that Rusa was in fact pronounced as
Ursa).
good price), but niri may be the main term for good in the Mittanni dia-
lect (Dietrich/ Mayer, 2010, 252; GLH, 185; Wegner, 2007, 269), cf. Mit.
good legate, good dowry.
, 1995/ 2007, 630 compares Hurr. wahri to the following NCauc.
root : Sino-Cauc. *HV[q]wV good (vel sim.) > NCauc. *HV
wV (~ -q
-)
good (vel sim.) (> Lak uj- good, Lezgh. *HVV good (vel sim.), a
poorly attested root), STib. *xH good (vel sim.) (may be the basic Proto-
STib. root for good), Yen. *haq- good (the main Proto-Yen. root for
good), Burush. *qha good (may be the basic Proto-Burush. root for
good). Such a comparison is very tempting semantically, but not self-evi-
dent phonetically.

35. green: unattested.

36. hair: unattested.

37. hand: u=ni ~ ummu=ni
Bo.Bil. (uni), Bogh. (uni, ummuni ~ umuni), Ort. (uni), Ugar.C. (umuni).
Wegner, 2007, 219, 281; Wegner, 1995a, 125; Girbal, 1994, 171173; Catsani-
cos, 1996, 200; GLH, 242, 243; Grke, 2010, 340; nal, 1998, 59, 64.
Probably with polysemy hand/ arm.
38
In the light of the Urartian cognate, the
Hurrian stem uni is to be analyzed as u=ni with the common nominal suf-
x -ni. The double mm in um(m)u=ni apparently points to an old cluster,
thus ummu- might be derived from uni with the nominal sufx -mi (on
which see Wegner, 2007, 59), if mm < nm, although the vowel shift *min >
mun looks strange.
Cf. the word herari (Bo.Bil., Bogh.), which denotes either sinew, biceps or
upper arm (Catsanicos, 1996, 201; Wegner, 1995, 118 f. ; Salvini / Wegner,
2004, 173; Grke, 2010, 74).
Urartian: u-, see Girbal, 2004, 26 f. (accepted in CdTU A 14-1 obv. 10) and
CdTU A 10-5 obv. 19 (--ki
?
-e
?
my
?
hand = Assyr. tu hand). But con-
tra such an interpretation, see Khachikyan, 2006, 145 f.
No obvious etymology for HU *u- hand; the root resembles Sumerian U
hand. Diakonoff/ Starostin, 1986, 23, compare Hurr. ummu- to NCauc.
*mr[] (~ -o) handful, armful, which seems unjustied for various rea-
sons.

38
It seems that there are no unequivocal contexts requiring the translation arm, but be-
cause of the polysemy ur(n)i foot / leg (No. 31 above), it is natural to assume the same
phenomenon for u=ni ~ ummu=ni (semantic typology suggests that commonly a
language possesses either one common term for hand/ arm and one for foot / leg, or
there are four basic terms, one for each body part).
38. head: pahi
Mari, Bogh., Ort., Ugar.C., Ugar.A.
GLH, 192; Wegner, 2007, 270; Andr-Salvini / Salvini, 1998, 14; Salvini / Weg-
ner, 2004, 177; nal, 1998, 64; Grke, 2010, 138 f.
The word also occurs several times in Mit., where it is traditionally translated as
head by Wilhelm, 1992b, Giorgieri, 1999, Wilhelm, 2006 (or left without
translation), but as news, message by Dietrich/ Mayer, 2010, 253.
A phonetically and semantically satisfactory comparandum could be Sino-
Cauc. *VHV top > STib. *pw (~ b-, -iw) top, upper part (> Tib. phu
the upper part, Burm. paw top, upper part, above, Kachin bo
1
the head,
Kiranti *p[]-jV head), Burush. *-ph top, tip.

39. to hear: ha-
Tiatal, Bo.Bil., Bogh., Mit., Tell Brak, Subar.
Wegner, 2007, 258; GLH, 95; Catsanicos, 1996, 267, 272; Dietrich/ Mayer,
2010, 242; Wilhelm, 1991a, 166; Wilhelm, 1998c, 140; Campbell, 2008,
283; Hra, 2010, 92 [l. 12], 239.
Another verb alh- (Mari, Bo.Bil., Bogh., Ugar.A. ; Wegner, 2007, 275; Catsa-
nicos, 1996, 267; Lam, 2007, 400) seems rather to have the meaning to
listen (to), hearken.
Urartian: ha=i-, ha=u- to hear; to listen(?) (, 446; CdTU 2, 97, 98).
Corresponds to the basic Assyrian term em to hear; to listen in bilingual
inscriptions.
No obvious etymology for HU *ha-. Diakonoff/ Starostin, 1986, 46 com-
pare it to Nakh *ac - to hear. This Proto-Nakh term is, however, isolated
within NCauc. and certainly does not reect the main Proto-NCauc. root for
to hear (rather a HU loan in Proto-Nakh, see 5.5 below and ,
2011).

40. heart: tia
Mari, Bo.Bil., Bogh., Mit., Ugar.C.
GLH, 266; Wegner, 2007, 286; Catsanicos, 1996, 199 f. ; Dietrich/ Mayer, 2010,
265.
Urartian: ti=ni- (, 469; CdTU 2, 217).
No obvious etymology for HU *ti-. Cf. Sino-Cauc. *rHV stomach,
intestines (vel sim.) > NCauc. *rHV (~ *rHV) stomach, intestine (vel
sim.) (> Nakh *VwV in, inside; belly, Av-And. *rVV-l intestine, a
poorly preserved root), Burush. *-ri kidney; testicle, Basque *erce/
*ete intestine. In this case one must assume the shift *r- > HU *t-.
39
Note

39
r-onset was prohibited in Proto-HU (for some recent loanwords in r- in Hurrian see,
e. g., Dietrich/ Mayer, 1994, 107). This restriction is a wide areal isogloss, which affected
many languages from certain East Caucasian ones (namely Proto-Nakh, Proto-Dargwa
and Early Proto-Lezghian) to Ancient Greek in the 3
rd
2
nd
millennia BC.
that there is no unequivocal Common Sino-Cauc. root for heart.
Diakonoff/ Starostin, 1986, 53, compare HU *ti- to NCauc. *risw throat (vel
sim.) (> Av.-And. *risV throat ; gullet, Tsez. *rs throat, Lezgh.
*sir(a) throat ; gullet, WCauc. *p-a (~ --) neck (vel sim.)) that does
not seem more convincing semantically.

41. horn: unattested.

42. I: i-/ u-
Paradigm: i- [abs. =te, erg.] (Mari, Bo.Bil., Mit., Bogh., Ugar.C.) / u- [obl.]
(Mit., Bogh.).
Wegner, 2007, 82 f. ; Wilhelm, 2008a, 95; Giorgieri, 2000, 219; GLH, 126, 127,
237; Dietrich/ Mayer, 2010, 193; Neu, 2003, 297.
Urartian: i-, ie- [absolutive subj. of an intr. verb itidi, ergative iei] / u=ke-
[oblique]. See , 2010c, 157158; Wilhelm, 2008b, 114; Girbal,
2001; , 448 (iei), also , 328, 450, 465 with itidi as here &
uki as as(?).
An exact comparandum seems the Sino-Cauc. suppletive pronoun of the
1
st
p. sg., which possessed two stems: z-like (direct & oblique) and -like
(oblique). The former is reconstructed on the basis of NCauc. *z [abs. ?] /
*ez(V) [erg. ?] / *iz(V) [gen. ?] (see Caucet.dbf sub *z for detail) I, Yen.
*a I, Burush. *ja I.
40

Lexicostatistical match: NCauc.-Yen.-Burush.

43. to kill : the generic term is unknown.
A possible candidates is the verb mar=al- (Bogh.), for which Grke, 2010, 89,
92, 338 tentatively proposes the translation to kill, cf. without translation
and with partially different grammatical analysis in Campbell, 2007, 438 f.
Cf. also uw- (Bo.Bil., Ugar.A.) to slaughter (ox etc.) and ur- (Bo.Bil.) to
slaughter (small cattle), Catsanicos, 1996, 234; Dietrich/ Mayer, 1994, 88,
93. Bo.Bil. uw- may be in some way related to a third verb: Hurr. urb- to
slaughter (cattle) (Nuzi ; GLH, 285) = Urart. urb=u- id. (, 473).
Urartian: cf. the verb zag=u- to kill (humans), annihilate (enemy troops),
etc., which is attested with the plural object only (, 475; CdTU 2,
248 ff.).
As noted in , 1995/ 2007, 632, Hurr. uw- resembles Nakh *-w-
to kill (it is a basic Proto-Nakh term for this meaning). Note that Nakh
*-w- is isolated within NCauc., lacking any external etymology; for such
HU-Nakh isoglosses, cf. 5.5 below and , 2011.

40
Diakonoff / Starostin, 1986, 81, with the HU-NCauc. comparison.
44. knee: unattested.
Cf. the terms ziyan=ni (Ugar.C.) lap, hip, crotch (vel. sim) (Andr-Sal-
vini / Salvini, 1998, 24) and huri (Ugar.C.) lap, crotch (vel sim.) (Andr-Sal-
vini / Salvini, 1998, 12).

45. to know: pal-
GLH, 194; Wegner, 2007, 270; Dietrich/ Mayer, 2010, 253; Dijkstra, 1993b,
167; Catsanicos, 1996, 281.
The occurence of this verb in an Ugaritic vocabulary glossing Sum. DUG
4
& Ak-
kad. ab to say is apparently an error (to be read kul- to speak? See notes
on 71. to say below).
Diakonoff/ Starostin, 1986, 17 compare it to NCauc. *m\V (~ -I-, -x-) ~
*bn\V to study, teach (> Av.-And. *mo- to teach, Tsez. *mL()-
(~ --, --) to teach, Lak max()- tame (adj.), Khin. mux- to know),
which can be theoretically possible.

46. leaf: unattested.

47. to lie: unattested.

48. liver: ur=mi
Ugar.C., Bogh.
Andr-Salvini / Salvini, 1998, 20; Andr-Salvini / Salvini, 1999a; Grke, 2010,
76 f.
Apparently ur=mi with the nominal sufx =mi, for which see Wegner, 2007, 59.
Urartian: zeldi apparently means small, little (No. 77 below), not liver (as in
, 476, following Diakonoffs proposal ;
41
without translation in CdTU
2, 251).
No obvious etymology for Hurr. ur=mi. Resembles the Sumerian term for
liver: UR
5
.
49. long: keri ~ ker=ai
Wegner, 2007, 263; GLH, 143; Catsanicos, 1996, 204; Dietrich/ Mayer, 2010,
242.
Polysemy: long (spatial) / long (temporal) (the meaning far proposed in Dia-
konoff/ Starostin, 1986, 59 is uncertain).
Diakonoff/ Starostin, 1986, 59, connect it to NCauc. *qVrV (/ *rVqV)
far (apparently the best candidate for this basic meaning in Proto-NCauc.),

41
Diakonoff / Starostin, 1986, 48 compare Urart. zeldi (zel=di) to NCauc. *Hl\V (one
of the two candidates for the status of the basic Proto-NCauc. term for liver) which in
any case should be dubious phonetically.
which seems probable.

50. louse: aphe
Ugar.C.
Andr-Salvini / Salvini, 1998, 9; Andr-Salvini / Salvini, 1999a; LHL 1, 93.
The Hurr. word is attested in the lexical list, glossing Sumerian E (i. e., U) and
Akkadian kalmatu. Both Sumerian and Akkadian terms may denote not only
louse, but also other parasites and vermin. In fact, however, lice are the
commonest human parasites, so it is natural to assume that this entry of the
trilingual lexical list (which covers basic and everyday vocabulary) denotes
just louse. Cf. the next entry in the same lexical list : Hurr. tame ea (An-
dr-Salvini / Salvini, 1999a).
It seems that an appropriate comparandum can be Sino-Cauc. *mkV
louse, small insect > NCauc. *mkV a k. of insect, vermin, worm
(> Av.-And. *unka name of a vermin, harmful insect ; beetle, Lezgh.
*kamk worm), STib. *m / *mk (~ --) midge, moth, Yen. *oke
(~ x-) louse (> Ket
4
, Yug. hk, Kott. iki ; a main Proto-Yen. term for
louse). The assumed Hurrian denasalization *mk > *pk requires additional
examples, however. Note that the cluster mK seems synchronically unat-
tested in proper Hurrian words.
42
An interesting parallel is the name of the
Sumerian goddess
d
damkina which can sometimes be spelled as
d
tap-ki-in-na
in Hurrian contexts (GLH, 70; van Gessel, 1998, 1: 438). It also seems natu-
ral to assume the further fricativization *pk > ph, since the cluster pk is
probably unattested in genuine Hurrian words.
43


51. man: tahe ~ tae
Bo.Bil. (tahe), Bogh. (tahe), Ugar.C. (tahe ~ tae).
GLH, 251; Wegner, 2007, 282; Catsanicos, 1996, 198; Huehnergard, 1987/
2008, 381; Andr-Salvini / Salvini, 1998, 17; Dijkstra, 1993b, 168; Grke,
2010, 80.
Perhaps /taHi/, see notes on 25. eye for the uctuation tahe ~ tae. Corresponds
to Hittite L-n- (i. e., pesn-) man (male human being) in Bo.Bil. & Bogh.
Cf. the noun tur=u=hhi male (adj.) (Bo.Bil., Bogh., Ugar.C., Ugar.A.), GLH,
274; Wegner, 2007, 288; Catsanicos, 1996, 198; Andr-Salvini / Salvini,
1998, 20.
Urartian: cf. ae-, i. e., phonetically wae-, man ( 439; CdTU 2,
64 f.), which is morphologically singular, but apparently attested only in the
plural meaning: men (the same case as 64. person and 99. woman).

42
Cf. Hurr. tamgari merchant < Akkad. tamkru id. > Sum. DAM.GR id. (GLH,
254).
43
But cf. Hurr. tupki ? (GLH, 272; Wegner, 2007, 22; Grke, 2010, 54 f.).
Without Hurrian cognates.
Hurr. ta(h)e is compared by Diakonoff/ Starostin, 1986, 26, to NCauc.
*dVrq
wV he-goat, which is unprovable (note that there is not enough data

to reconstruct *dVrq
wV as a Proto-NCauc. root, in fact it is an Av-And.

Tsez. isogloss only: Av.-And. *dVq()V-n he-goat, Tsez. *qV A kid
about 1 year old).
A probable comparandum for Urart. wae- has been proposed by Dia-
konoff/ Starostin, 1986, 20 NCauc. *wr bull-calf, male > Nakh *bor
bull-calf, Av.-And. *bia bull-calf, Lezgh. *wV(r)- man, husband.
As for Hurr. tur=u=hhi male (adj.), it is compared by Diakonoff/ Starostin,
1986, 52, to NCauc. *lw / *wlV man, male (vel sim.) (> Nakh *-law
a sufx denoting profession, Av.-And. *lV address to a man, Lak lak-
Laki (ethnonym), Dargwa *gal(i) son, boy, Lezgh. *ilV- male; man;
men, Khin. lgld man, WCauc. *A male; testiculus).
44
Phonetically
dubious.

52. many: te=u=na(i)
Mit., Bogh., Ugar.C.
Wegner, 2007, 285; GLH, 260, 264; Dietrich/ Mayer, 2010, 265; Dijkstra,
1993b, 169; Trmouille, 2005, 325.
The most frequent and generic expressions for many are derived from the ver-
bal root te- to be much, be many, be lot, be numerous (Dietrich/ Mayer,
2010, 265): adv. te=u=na(i) and adj. te=a numerous (both: Mit., Bogh.,
Ugar.C.). The adverb tian very (Mit.) is probably derived from the same
root.
Cf. an unclear akephalos []i many, much in an Ugaritic vocabulary (Hueh-
nergard, 1987/ 2008, 4041, 87, 381).
Urartian: cf. Diakonoffs reading
!
-a-e many (Diakonoff, 1971, 46 fn. 42;
, 469). An entirely different reading is in CdTU 1, 343.
An obvious comparandum seems Sino-Cauc. *dVHV to grow; big >
NCauc. (> WCauc. *dA big; most, at all ; much, very), STib. *tajH big,
many (vel sim.) (> Chin. *tj much, many, Burm. taj very, ti sign of
the plural, Kachin the
2
and, Lushai te much, very much, -te sufx de-
noting plurality, Lepcha t, t-m to be great, large, big, Kiranti *dV big;
one of two main candidates for the status of the Proto-STib. basic term
many), Yen. *tj- to grow. Also Hattic te-, ti- great, big (Kassian,
2010a, 363).
Lexicostatistical match: STib. ?

44
Further, as proposed in Sccet.dbf sub Sino-Cauc. *w, it is compared to STib. *lk
testicle; glans penis, Burush. *lei (< *leli), *lal addressing a man; free man, great
man, Basque *-le agent sufx.
53. meat: uzi
Bogh., Ugar.C.
GLH, 291; Huehnergard, 1987/ 2008, 2425, 46.
Urartian: cf. huu meat (?) in , 127, 447; without translation in CdTU
2, 103.
Hurr. uzi could be compared to Yen. *ise meat (> Ket meat, sh, Yug.
s meat, sh, Kott. i meat, Arin is meat; a basic Proto-Yen. term for
meat). The Yen. root must be further connected to STib. *a meat (the
main term for meat in the Tibeto-Burman branch, although superseded by
unclear nhuk meat in the Sinitic one) and maybe a similar form for
meat from the Na-Dene family that makes the corresponding Sino-Cauc.
proto-root the best candidate for the status of the ProtoSino-Cauc. term for
meat.
45

On the contrary, , 1995/ 2007, 631 unites Hurr. uzi with NCauc.
*=(rV) thick, fat (adj.). Despite the coincidence of the u-vocalism,
such an etymologization seems more problematic. First, it is possible to con-
nect NCauc. *=(rV) to Hurr. ae fat (n.) (No. 26 above). Next, the
meaning shift fat (adj.) > fat (n.) is possible, whereas fat (n.) > meat
seems atypical cross-linguistically.
Hurr. uzi resembles the main Sumerian word for meat: UZU.
Lexicostatistical match: Yen.-STib.

54. moon: kuuh
Mari, Nuzi, Ugar.C., Ugar.A., Bogh., Msk.
GLH, 156; Wegner, 2007, 265.
The Moon-gods name; can be analyzed morphologically as ku=u=h. Another
designation of the Moon-deity can be umbu (GLH, 280), although it is uncer-
tain.
Urartian: elardi Moon-god (, 490). Both and CdTU A 3-1 41
read the rst sign as I ; see , 51 for sometimes proposed reading ME
(i. e., melardi, cf. an enigmatic Armenian hapax meard, glossed as moon in
Aayan HAB 3: 299).
How we should reconstruct the Proto-HU term for moon is unclear, cf.
notes on 82. sun.
Sccet.dbf tentatively unites Hurr. kuuh, Yen. *q (~ -) skewer, spit ;
(sun-)beam and Burush. *uha (< *u?) new moon, time without moon;
time of waning moon as Sino-Cauc. *QVV(?), which does not seem justi-
ed.
Diakonoff/ Starostin, 1986, 37 compares Urart. elardi (i. e., el=ardi) to
NCauc. *VlV (~ -, --, --) light, ray (> Nakh *sa light ; eye-sight ;

45
In Sccet.dbf (sub *Hmc), Yen. *ise meat is connected to NCauc. *jomc bull,
ox (very well attested), STib. *chu (~ -, -n, -w) cow, bull (poorly attested), which is
a less probable solution.
dawn; to dawn, Tsez. *ri()V (~ l-, , u) sun ray, Dargwa *ala light,
WCauc. *P-sV to shine), which is semantically possible.
46

55. mountain: pab=ni ~ pab=a=ni
OBab., Mari, Bo.Bil., Bogh., Ugar.A.
GLH, 190 f. ; Wegner, 2007, 271; Catsanicos, 1996, 231; Grke, 2010, 85, 336;
Prechel / Richter, 2001, 357.
Pace Laroche (GLH, 190), the root is rather pab- (with the common nominal
sufx -ni) than the triple consonantal paban-, see Richter, 2007, 110.
Urartian: bab=a=ni-, 440; CdTU 2, 65 f.
No obvious etymology for HU *pab-a-ni ; the root resembles the Hattic term
pip stone. Diakonoff/ Starostin, 1986, 19, analyze the Proto-HU stem as
*fan-fan-, supposing a reduplication of the well preserved NCauc. root
*fnV shady slope of a mountain (vel sim.) (> Nakh *un forest, Av.-
And. *xan-dV (~ -o-) shady side, slope; hill ; meadow; haying place, Tsez.
*en A a part of mountain, Lak han shady slope of a mountain, Dargwa
*xVna northern slope, Lezgh. *[a]na northern slope of a mountain,
WCauc. *xA mountain), which does not seem apt from the morphological
viewpoint.

56. mouth: fai
Mari, Bogh., Ort., Ugar.C., Ugar.A.
Wegner, 2007, 272; Wegner, 1995a, 123; GLH, 197; Andr-Salvini / Salvini,
1998, 22; nal, 1998, 58; Lam, 2007, 401.
Previously interpreted as eye.
No obvious etymology. Cf. Sino-Cauc. *mw[]li (~ -e) face, cheek, jaw
> NCauc. *mwIli (~ -e) edge, tip, Yen. *bin- chin, cheek, Burush.
*mel jaw.

57. name: tiye
Bogh.
Giorgieri, 2001, 144 f.
Attested once with the meaning name, corresponding to UM-an from a parallel
Hittite text. In fact, the word is well-attested in two phonetical variants with
the following polysemy: tiwe [OBab., Bo.Bil., Bogh., Ort., Mit.] ~ tiye
[Bogh.] word / speech, word(s) / name / thing, see Wegner, 2007, 285;
GLH, 267 f., Giorgieri, 1998, 71, 77 ff. ; Giorgieri, 2001, 144. Giorgieri sug-
gests that tiwe and tiye differ in meaning, namely tiye speech vs. tiwe
word; thing, and treats w as a nominal sufx of unknown semantics, but I
suspect, following H. J. Thiel, that we are more likely to deal with a sporadic
phonetic process EwE > EyE here. The glide uctuation is actually the same

46
Sccet.dbf (sub *HVlV) unites NCauc. *VlV with STib. and Burush. terms for
moon, but I suppose that this is incorrect.
as in the pairs iwe ~ iye water (see No. 94 below) or awi [Bo.Bil., Bogh.,
Msk.] ~ aye [Mit.] face; in front of (Wegner, 2007, 248; GLH, 34, 40; Cat-
sanicos, 1996, 200, 278).
Urartian: ti=ni , 468; CdTU 2, 215 f. A cognate of the Hurrian term,
modied by the common nominal sufx =ni with the loss of intervocalic -w-.
HU *tiwi(=ni) word/ name is clearly related to the HU verb *tiw- to say,
for detail, see the etymological discussion sub 71. to say.

58. neck: kudu=ni
Bogh.
Wilhelm, 1998b, 180; Campbell, 2007, 418; Grke, 2010, 72. With reservations,
in Wegner, 1995a, 120.
No obvious etymology. Theoretically can go back to a metathetical variant of
Sino-Cauc. *dHq
w nape of the neck (vel sim.) > NCauc. *dHq
w
back of head (> Lezgh. *taqI back of head, WCauc. *tqA/ *qta
neck; back of neck), STib. *tk neck (a possible candidate for the status
of the main Proto-STib. term for neck), Yen. *tuGV (~ --) back, behind.

59. new: uhe
Ugar.C.
Andr-Salvini / Salvini, 1998, 17; Huehnergard, 1987/ 2008, 378.
Urartian: uhi (, 465; CdTU 2, 198 f.). An etymological cognate of the
Hurrian term.
Proto-HU *uhi new can probably be compared to the basic Sino-Cauc.
term *V new > NCauc. *nV new (> Nakh *in-, Av.-And.
*inhV-, Tsez. *-Vn- A, Lak u-, Dargwa *i-, Lezgh. *enj-/ *enw-,
WCauc. * A, all meaning new, a Proto-NCauc. term for new), STib.
*sn/ *s new (> Chin. *sin new, Tib. gin good, ne, Burmese sa
new, a Proto-STib. term for new), Yen. *-sa (~ --) to begin.
47
HU *h
in *uhi can be a reex of *. Another solution is to treat the nal =hi in
HU *uhi as a frequent HU adjectival sufx (see Wegner, 2007, 54, for the
Hurrian data). If so, the Proto-HU root is *u- with the loss of a nasal before
a laryngeal (* > * > 0/ ).
Lexicostatistical match: NCauc.-STib.

60. night: unattested.

61. nose: punhi ~ puhhi
Bogh. (punhi ~ wunhi), Ugar.C. (puhhi ~ wuhhi).
Wegner, 1995a, 123; Huehnergard, 1987/ 2008, 3839, 82; Andr-Salvini / Sal-
vini, 1998, 22; Grke, 2010, 72.

47
The Urartian-NCauc. comparison has been proposed in Diakonoff / Starostin, 1986, 34.
Apparently a dialectal assimilation nh > hh, cf. the same phenomenon in paban-
hi ~ pabahhi mountainous; east (GLH, 191) < pab=a=n=hi pab=a=n-
mountain.
No obvious etymology. , 1995/ 2007, 630, compares Hurr. puhhi
to Tsezic *poro part of face under the nose ~ WCauc. *pA nose (one of
the two main candidates for the status of the basic NCauc. term for nose),
whose proto-form is reconstructed in Caucet.dbf as NCauc. *prV, though
proto-forms like *plV, *plV, *prV are also allowed; the Hurr. -n-,
however, remains unexplained.

62. not
1
: =u= ~ =wa=
Campbell, 2007, 63 ff. ; Wegner, 2007, 94, 96, 136; Wilhelm, 2008a, 97;
, 2010b, 137; Giorgieri, 2000, 233; Dietrich/ Mayer, 2010, 216.
The basic Hurr. negation morpheme is the sufx =u= which is attested as =u= ~
=wa= in ergative indicative forms of the 1
st
& 2
nd
p. (Bo.Bil., Mit., Bogh.),
=u=d= in ergative indicative forms of the 3
rd
p. (Bo.Bil., probably Ugar.C.),
=wu=r= normally with the root mann- to be (Bo.Bil.), =wa= in non-indi-
cative (jussive) ergative/ non-ergative forms of the 1
st
& 3
rd
p. (Mit.).
Although in the 3
rd
p. u-based morphemes began to be superseded by the enclitic
=ma already in Bo.Bil., Bo.Bil. still retains negative ergative forms in
=u=d-, Wegner, 2007, 96 (some lexicalized negative forms with =ud= are
also observed in Bogh.).
Note that the enclitic =ma ~ =mma used in the negative forms of the 3
rd
p.
(Bo.Bil., Bogh., Mit.) can hardly be genetically related to the negative sufx
=u= discussed above (pace , 1985, 95, Giorgieri, 2000, 233, Weg-
ner, 2007, 96
48
). It seems more probable that =ma represents an old pro-
hibitive morpheme *mV,
49
the so-called global prohibitive in m, attested in
many language families around the world. It is strange, however, that the
synchronic Hurrian prohibitive (scil. negated jussive or vetitive), which is
attested in the 1
st
& 3
rd
p. only (Mit., also Tiatal, Bogh., probably Mari, Tell
Brak), is formed by u-sufxes, namely =wa=, =we=, =o= (Campbell, 2007,
241, 264 ff. ; Wegner, 2007, 108; , 2010b, 137; Giorgieri, 2000,
233; Dietrich/ Mayer, 2010, 220; Wilhelm, 1991a, 164). The proper pro-
hibitive of the 2
nd
p. seems, however, unattested in Hurrian texts. As for
Urartian, it retains this particle mi in the prohibitive function (Wilhelm,
2008b, 119).
Cf. the particle u=ya (Mit.) no! (Dietrich/ Mayer, 2010, 267).
Urartian: the antiquity of the Hurrian negative morpheme =u= is proven by
Urartian data, where the particle ui (a cognate of Hurr. u=ya) and the sufxal

48
Justied doubts in Campbell, 2007, 65 f.
49
The morphosyntactic status of =ma in the synchronic Hurrian (an enclitic occurring at
the end of verbal wordforms) should also suggest that its usage in the function of nega-
tion of an afrmative statement is a recent introduction.
=u=ri are default verbal exponents of negation of an afrmative statement,
see Yakubovich, 2010 [2011] w. lit.
No obvious etymology for the HU morpheme *u not. Diakonoff/ Starostin,
1986, 91, compare it to one of the several NCauc. negative morphemes:
*wV not (> Av.-And. *-hi, Tsez. *-(V), Dargwa *e, Khin. -i-). Such an
etymologization is theoretically possible, but not necessarily correct. Note
that NCauc. *wV is a rather marginal negative morpheme, although this in-
deed became a main exponent of negation of an afrmative statement in
ProtoDargwa-Kninalug.

62. not
2
: =kkV=
Campbell, 2007, 35 f., 47 f. ; Wegner, 2007, 97, 137; Wilhelm, 2008a, 97;
, 2010b, 137; Giorgieri, 2000, 233; Dietrich/ Mayer, 2010, 222;
Huehnergard, 1987/ 2008, 2425, 48.
=kkV is the basic Hurr. negative exponent in non-ergative (i. e., intrans. & anti-
pass.) indicative forms (Mari, Bo.Bil., Mit., Ugar.C., Bogh.). This morpheme
seems lost in the Urartian branch.
No obvious etymology. Diakonoff/ Starostin, 1986, 91, compare it to the
poorly attested NCauc. *-k-, a marginal morpheme expressing some kinds of
negation in several NCauc. subgroups. The comparison is formally possible,
but does not seem very reliable.

63. one: ukki ~ ukku
Bo.Bil., Mit., Bogh.
Wegner, 2007, 81, 280; Wilhelm, 2008a, 101; Giorgieri, 2000, 222; Catsanicos,
1996, 249; GLH, 241 (without translation); Dietrich/ Mayer, 2010, 261 (with
a different translation, namely weiter(hin); auerdem).
Urartian: apparently u=sini means 1, thus , 466; CdTU 2, 205, 331;
with some doubts in Wilhelm, 2008b, 119 and , 2010c, 159; see
Girbal, 2001, 142 for detail.
If morphologically u=kk- with a k-sufx (cf. Urart. usini, if its meaning is
indeed one), then the HU root *u- 1 is comparable with the main NCauc.
word for 1 *cH
(thus already Diakonoff/ Starostin, 1986, 38) > Nakh

*ca, Av.-And. *ci-, Tsez. *h
s, Lak ca, Dargwa *ca, Lezgh. *sa, Khin. sa,

WCauc. *zV, all meaning 1.
50

The Hurrian (Hurro-Urartian?) numeral may contain the same sufxed root as

50
In the current version of Sccet.dbf, NCauc. *cH 1 is united with the following
forms: STib. *t (~ -), Yen. *u-sa, Burush. *he-, all meaning 1, under the Sino-
Cauc. proto-form *HVc. In fact, however, the relationship between the NCauc., STib.
& Burush. forms is doubtful due to phonetical reasons. Next, besides *t, there is an-
other candidate for the main Proto-STib. term for 1 (namely *djik). As for Yen. *u-
sa 1, the main meaningful element here is *u, whereas *-sa is a singulative sufx (the
latter may indeed be cognate to NCauc. *cH 1).
ue/ ua=lla all (No. 1); note, however, that the development 1 > totus
(further > omnis) is typologically normal, but probably not vice versa.
Lexicostatistical match: NCauc.

64. person: taruwa=ni
Bo.Bil., Mit., Bogh., Ugar.C., Ugar.A.
Wegner, 2007, 284; GLH, 258; Catsanicos, 1996, 198; Huehnergard, 1987/
2008, 2425, 47, 379, 381; Dietrich/ Mayer, 2010, 264; Grke, 2010, 156.
Probably with occasional polysemy person, human being / man, male human
being, cf. Huehnergard, 1987/ 2008, 47, 388, but note that the basic word for
man (male) is tahe (No. 51).
The well-attested anatomic term edi body can occasionally be used in the
meaning person (GLH, 73; Wegner, 2007, 257). Cf. also an unclear form
[x]-lu-a-[x] people in an Ugaritic lexical list (Huehnergard, 1987/ 2008,
3839, 77, 386).
Urartian: cf. tarua=ni person (, 467; CdTU 2, 209; Wilhelm, 1981), a
clear cognate of the Hurr. term, but the Urart. word is apparently attested
only in the plural meaning persons, people (the same case as 51. man and
99. woman). Cf. also the word nara, whose meaning is unknown (without
translation in CdTU 2, 161 f. ; Wilhelm, 1981), although , 456 glosses
this as people.
Diakonoff/ Starostin, 1986, 39, analyze HU *tarua as a compound *tar-ua,
whose rst part is unclear (cf. the Hurrian root tur- in the adjective
tur=u=hhi male), whereas the second one is a cognate of NCauc. *wjo
person (> Nakh *psu bull, Av.-And. *VjV person; man, Tsez. *i- A
/ *zi- A person; man, Lak uw man, Dargwa *sub male (n.), Lezgh.
*ij man, WCauc. */ * person, the main Proto-NCauc. term for per-
son).
51
Theoretically possible.
Lexicostatistical match: NCauc. ?

65. rain (n.): iena
Ugar.C.
Andr-Salvini / Salvini, 1998, 13; Andr-Salvini / Salvini, 1999a.
Urartian: Diakonoff (Diakonoff, 1971, 167; Diakonoff/ Starostin, 1986, 44)
quotes an enigmatic verb i- to ow, to rain without references.
The Common Sino-Cauc. root for rain, to rain is *=GwV, but apparently
this is not the case. It seems possible to compare Hurr. iena to Sino-Cauc.
*HorwV to be cloudy, to rain (vel sim.) > NCauc. *HorwVn to be-
come cloudy (of weather) (> Nakh *j-, Av-And. *-ir-, Lak =aa-
/ =ai-, Lezgh. *raI()Vn- (~ --), all meaning to become cloudy), STib.

51
Cf. also Sino-Cauc. *wjo person > Yen. *e- people (- is the plural expo-
nent), Burush. *ses person(s), people.
*[j] shower, rain (> Kachin ca
2
to fall, as a light, cooling shower, to
squirt, Lushai h to rain continuously), Basque *orci / *oti sky;
storm; thunder; Thursday; rainbow; cloud. The loss of *r in such a cluster
can be regular for HU.
All these forms also resemble Sumerian E
3
to rain; rain (n.) (a second Sume-
rian term for rain (n.) is IM rain; wind; storm, perhaps with the semantic
development wind > storm > rain).

66. red: unattested.

67. road: hari
Mari, OBab., Bogh., Mit.
Wegner, 2007, 258; GLH, 94; Dietrich/ Mayer, 2010, 242; Prechel / Richter,
2001, 360, 362; Grke, 2010, 337. Occasionally borrowed into Hittite (HW
2

H, 275).
Urartian: the best candidate is hari, which probably means both road and
military campaign (, 445; CdTU 2, 97). Cf. the word arhe (attested
several times in an untranslatable formula), which is glossed in , 435
as path (uncritically following Diakonoffs interpretation), but prudently
left without translation in the main section of ; in CdTU also without
translation. Cf. also utipte military campaign; highway, road(?) (see Salvi-
ni in avuolu et al., 2010, 45; , 474; CdTU 2, 242 f.), derived from
the verb u-, u=t=a- to go (see No. 16. to come and 92. to go
1
).
HU *hari road is a member of a set of similar Near Eastern forms with the
semantics of road: Luwian (C) haruwa- road, Akkad. arrnu road;
journey,
52
Sumerian arran (AR-RA-AN) route, passage; path.
53
The Su-
merian form represents a transparent Akkadian loanword, but the origin of
Luw. haruwa- and Akkad. arrnu is obscure, because these words lack
obvious etymology within their families (Indo-European and Semitic respec-
tively).
Another phonetical shape is represented by NCauc. *ari (~ -) trip, march
(> Lak aIri trip, march, journey, Dargwa *arIa (~ -I-) road;
journey, Lezgh. *ja(r)I(a) foot-travelling) and Semitic *ur- road (vel
sim.) (> Akkad. [OA+] ur- road, way, path, Aramaic Bibl. ar, Pal.
wr, Syr. ur-, Mand. uhr-, Urm r-, all meaning road; one of the
candidates for the status of the main Proto-Sem. term for road).
An unclear situation.
54

52
This competes with other Akkadian terms for road : uru, girru.
53
This is not the main Sumerian word for road, the basic term is apparently KASKAL
way, road; journey, caravan.
54
Caucet.dbf connects HU *hari road to the aforementioned NCauc. *ari (suspect-
ing the dissimilation **harhi > *hari), but this is uncertain.
68. root: unattested.

69. round: unattested.

70. sand: unattested.

71. to say: hil- ~ hill-
Mari, Bo.Bil., Mit. Bogh., Ugar.C., Ugar.A., Tell Brak.
GLH, 105 f. ; Wegner, 2007, 259 f. ; Catsanicos, 1996, 256 ff. ; Dietrich/ Mayer,
1994, 74, 77, 87, 88, 89, 93, 96; Dietrich/ Mayer, 2010, 243; Grke, 2010,
107 f.
A difcult case, because there are several verba dicendi attested in the Hurrian
corpus and available contexts do not always permit to choose the basic verb
(it is also probable that the presumable difference of semantic nuances be-
tween various Hurrian verba dicendi has a dialectal origin). Provisionally I
choose the verb hil- ~ hill- as the basic word. It seems to be the default verb
to say (that) in Mit. In Mari, the verb is likewise used for to say (that)
(Wegner, 2007, 236237), the same apparently concerns the Bogh. usage
(e. g., Campbell, 2008, 270; also Grke, 2010, 107 f.). In Bo.Bil., hil- ~ hill-
is normally used in combination with al(u)- (another verbum decendi), with
the whole expression meaning to say (from smb.s heart) that. Cf. also
hill=u=i=t=an I has informed in Tell Brak, Wilhelm, 1991a, 163.
The second candidate is kad- (Bo.Bil., Mit., Bogh., Ugar.A.), GLH, 132;
Wegner, 2007, 262; Catsanicos, 1996, 256 ff. ; Wegner, 1994; Girbal, 1994,
173. It seems that, in Bo.Bil., Bogh., Mit., Ugar.A., the normal usage of kad-
is to speak the word, pronounce the word (with tiwe word), although some
contexts in Mit. demonstrates the usage to say that (translated as to report
that and so on in Dietrich/ Mayer, 2010). For the Ugar.A. attestation, see
Dijkstra, 1993a, 158, 161.
The verb kul- (Bo.Bil. Mit., Bogh. ; see GLH, 151; Wegner, 2007, 264; Catsani-
cos, 1996, 256 ff. ; Dietrich/ Mayer, 2010, 242) seems to be similar to kad-.
In Bo.Bil. and Mit., its meaning is to speak/ pronounce (the word/ curse/
etc.). This verb may also be attested in an Ugaritic vocabulary glossing
Sum. DUG
4
& Akkad. ab to say, speak, pronounce (GLH, 194 sub pal- to
know), if pa-li-u is to be emended to ku
!
-li-u, as has been plausibly
proposed in Wilhelm, 2003, 342 fn. 8.
The fourth verb is al=u- (Mari, Bo.Bil., Bogh. ; see LHL 1, 52; Wegner, 2007,
246; Catsanicos, 1996, 256 ff. ; Wegner, 1994; Salvini / Wegner, 2004, 167),
attested in the meaning to speak in Mari (Wegner, 2007, 236237). In
Bo.Bil. and Bogh., alu- is normally used in combination with hill- (see
above), with the whole expression meaning to say (from smb.s heart) that,
or in combination with kad-, with the whole expression meaning to speak
the word (with tiwe word).
Cf. also rare verbal stems tiw- to pronounce a word (vel sim.) (Dijkstra,
2006, 319 w. lit.) and maybe ti- to cry, exclaim (vel sim.) as per Die-
trich/ Mayer, 2010, 76, 265 (without translation in Wilhelm, 1992b, 69; Gior-
gieri, 1999, 388; Wilhelm, 2006, 188);
55
more frequent is the nominal stem
tiwe ~ tiye word/ speech, word(s)/ name/ thing (see 57. name above).
Note that, surprisingly, contexts with the simple verb to say (that) apparently
do not occur in Bo.Bil. In these texts, the most neutral expression for to say
(that) seems tiw(=ari) kad- to pronounce (the following) word(s) (= Hitt.
udar mema-).
Urartian: the best candidate is probably the verb ti=a-/ ti=u- (, 468;
CdTU 2, 215 dire), whose various forms are attested in two kinds of con-
texts: X says: (direct speech) (in monumental inscriptions) and you (the
messenger), do say (to the addressee so and so) (in letters). Urart. ti- corre-
sponds to the basic Assyrian verb ab to say in bilingual inscriptions.
The second candidate is the verb al- (, 432; CdTU 2, 16 ff. dire; Wil-
helm, 2008b, 117). The only known form is 3 sg. al=i (with numerous attes-
tations), which normally introduces direct speech of a ruler (e. g., stereo-
typical contexts like The ruler says: I did so and so). Perhaps it would be
better to translate al- as to enunciate vel sim.
Cf. also the rare verb pal-, for which the translation to speak, tell (?) is pro-
posed in , 1985, 48, 50 fn. 56; , 457.
How we should reconstruct the Proto-HU term to say is not entirely clear.
The HU root *tiw- has some advantage over Hurr. hil(l)-, because *tiw-
yielded not only the basic word for to say in Urartian, but also produced the
HU term *tiwi word/ name (see 57. name above) and a marginal verb of
saying in Hurrian. The semantic derivation verbum dicendi word is ty-
pologically normal ; on the contrary, the shift word name is not so com-
mon cross-linguistically, but cf., e. g., the Lithuanian-Lettish term *war
d-a-s
name; word, which goes back to the IE word for word. Thus, the Proto-
HU derivational chain could be *tiw- to say *tiwi word *tiwi
name. An alternative solution is also possible, though: the HU *tiwi word
acquired the meaning name (a Common Proto-HU process), on the other
hand, and developed into the basic Urartian verb to say and the Hurrian
verb to pronounce a word (vel sim.) (as independent Hurrian and Urartian
introductions?), on the other.
Phonetical details are not entirely clear as well. In seems that the discussed root
may safely be reconstructed as *tiw- > Hurr. tiwe ~ tiye word; name (for the
glide uctuation, see notes on 57. name above), Urart. ti=ni name with the
regular(?) loss of intervocalic *-w- and Urart. ti=a-/ ti=u- to say with the
same loss of -w-. On the other hand, if the Hurrian plain verb ti- to cry, ex-

55
The verbal root ti- is also proposed in Wegner, 1988, 152, for a certain Bogh. passage,
but the nominal interpretation of ti-i-e seems more likely (thus Giorgieri, 1998, 78;
Grke, 2010, 78, 152).
claim (vel sim.) indeed exists, one must follow Giorgieris proposal and an-
alyze HU *tiwi word; name as ti=w=i with an unclear nominal suf-
x -w- (cf. notes on 57. name above).
Diakonoff/ Starostin, 1986, 53, compare HU *tiw- to NCauc. *lHwV (~ -)
(> Tsez. *lje (~ r-) word, sound, voice, Dargwa *lui number). Further,
as proposed in Sccet.dbf sub Sino-Cauc. *lVhwV word, to Basque *ehe
word, story. The initial l- was indeed prohibited in Proto-HU, and the
whole etymology seems very tempting both phonetically and semantically.
The chief difculty is that the NCauc. root is extremely poorly attested and
this is not the Proto-NCauc. root for word;
56
the relationship between
NCauc. *lHwV and Basque *ehe is also not necessarily valid phonetically.
Hurr. hil(l)- can be plausibly compared to Sino-Cauc. *HV to say (vel
sim.) > NCauc. *HiLV (~ --) to say (vel sim.) (> Nakh *l-w-/ *- to
say, Av.-And. *hi- to say, Tsez. *i- to speak; to call, one of several
candidates for the status of the Proto-NCauc. term for to say), STib. *l to
speak, speech, Yen. *V()- (~ -r
1
-) to speak, Burush. *lte- to swear;
oath.
57

As for Hurr. kul- to speak, pronounce, this is compared by Diakonoff/ Sta-
rostin, 1986, 54 to NCauc. *=algwn (> Av.-And. *gVl- to speak, talk,
Lak =uki- to count ; to read, Dargwa *=[a]lgwVn to ask; to count, read,
Lezgh. *alga(n) to talk, speak, WCauc. *ga to call ; to be heard (of sound,
voice)), which is possible, because reduction of the medial vowel and
metathesis -CR- > -RC- is a regular morphological process in Proto-NCauc.
verbal roots of the original shape *CVRV, see SCC, 1 f.

72. to see: fur-
Mit., Bogh., Qatna.
GLH, 298; Wegner, 2007, 274; Dietrich/ Mayer, 2010, 272; Richter, 2005, 27.
The verb is also attested in Bo.Bil., but with the meaning to nd (Catsanicos,
1996, 271, 272).
The second candidate is the verb am- (Bo.Bil., Bogh. ?), but its meaning is rather
to notice; to look (LHL 1, 72; Wegner, 2007, 246; Catsanicos, 1996, 263,
272).
As proposed by Diakonoff/ Starostin, 1986, 64, Hurr. fur- could be compared
to NCauc. *werV look (n.), sight (> Nakh *bar look, sight ; to meet, to
look, to stare, Av.-And. *arV (~ -o-) to look, Lak =uru-ga- to look,
Dargwa *er look, sight, Lezgh. *wVr- to glance, peep; to visit).

73. seed: unattested.

56
Proto-NCauc. word is *rU, which is retained in this meaning in main NCauc. sub-
branches.
57
The Hurr.-NCauc. comparison has been proposed in Diakonoff / Starostin, 1986, 51.
74. to sit: nahh-
Bo.Bil., Bogh.
Wegner, 2007, 268; Catsanicos, 1996, 205 ff. ; Trmouille, 2005, 320; Sal-
vini / Wegner, 2004, 176; Campbell, 2007, 335.
The known contexts from Bo.Bil. point to the meaning to sit down (intr.) / to sit
down (tr.) / to appoint, commission / to occupy, reside, dwell (Catsanicos,
1996, 205 ff.), but the translation to sit, be sitting is more probable for some
Bogh. contexts (e. g., Campbell, 2007, 335). The underlying semantics to
sit, be sitting can al/so be proven by nominal derivatives with the meaning
seat, chair: nahh=idi (Bogh.), nahh=u=the (Bogh.), Wegner, 2007, 231,
268.
The verb a- means rather to wash than to be seated (LHL 1, 130; Haas/
Wegner, 2010, 102 ff.).
Urartian: no generic verb attested. Cf. the verb nah=a- to mount (the throne)
(, 455; CdTU 2, 159 salire sul trono). Such a cognate of the Hurrian
root apparently proves that *nah(h)- was the basic Proto-HU term for to sit.
No obvious etymology for HU *nah(h)- to sit.

75. skin: ahe
Bo.Bil., Ugar.C.
GLH, 59; LHL 1, 121; Catsanicos, 1996, 233; Huehnergard, 1987/ 2008, 2425,
47.
Provisionally, I ll the slot by the noun ahe. This is attested with the meaning
hide in Bo.Bil. (skin of the deer). In its turn, the attestation in the Ugaritic
lexical list can mean both (human) skin and hide (glossed by Sumerian
KU and Akkadian maku with the same polysemy). The general context of
the lexical list speaks, however, in favour of the human anatomical term.
No obvious etymology. Diakonoff/ Starostin, 1986, 47, compare Hurr. ahe
to NCauc. *wrw skin; colour, but such a comparison is not self-evi-
dent phonetically. Next, the NCauc. proto-form is highly dubious itself, be-
cause the comparison of Nakh *rcu anxiety, alarm, Av.-And. *i(r)i
anger, anxiety to Lezgh. *w(r) skin (of cattle), WCauc. *ca skin is
too problematic semantically.

76. to sleep: unattested.

77. small : unattested.
A possible candidate is zugi, attested in an Ugaritic lexical list with the meaning
low (said of rate of exchange), corresponding to Sumerian TUR small, lit-
tle (GLH, 306; Huehnergard, 1987/ 2008, 77). For the second possible at-
testation of zugi (with the meaning young?) in Ugaritic vocabularies, see
Huehnergard, 1987/ 2008, 3839, 77.
Cf. the word kalgi (signs: ga
?
-al-gi) a not quite reliable attestation in an Uga-
ritic vocabulary with the possible meaning small, little (Huehnergard,
1987/ 2008, 3839, 79).
The third candidate could be Subarian hayani = Akkad. eru small ; young
(Hra, 2010, 164 [l. 234], 286; Kilmer, 1963, 437 [No. 234], GLH, 89),
which in fact seems to represent the same term as hni child (Ugar.C.,
Bogh.) han- to give birth, Andr-Salvini / Salvini, 1998, 11; Giorgieri,
2001, 143; GLH, 93, although phonetic details are not entirely clear.
The fourth potential candidate is iu another gloss for Akkad. eru in the
same synonym list (Hra, 2010, 164 [l. 230B], 286; Kilmer, 1963, 437
[No. 230a]; AHw, 1253b; CAD / 3, 147a), whose attribution to Subarian
is possible, but by no means necessarily conclusive. Proceeding from this
form, Dietrich/ Mayer, 2010, 260, 280, postulate the Hurr. verb i- to be
small, to make small for some Mit. passages (applied to land and gift),
which is textually possible, but also not unquestionable.
58

Urartian: a good candidate is zeldi in the collocation a little piece of, for
which see Girbal, 2010 (previously interpreted by Diakonoff as liver, re-
peated in , 476; without translation in CdTU 2, 251). Cf. also the form
urdiali, , 353 fn. 31, 473, with a tentative translation small; an en-
tirely different and more probable interpretation is in CdTU 1, 572.
Diakonoff/ Starostin, 1986, 41, compare Hurr. zugi to NCauc. *
w/
*}w short (> Av.-And. * iV- small, short (in height), Lezgh.
*i[]V- short, WCauc. *a short), which is theoretically possible.
Diakonoff/ Starostin, 1986, 62, compare Hurr. hayani to NCauc. *=hV lit-
tle, few (vel sim.) (> Nakh *=awi-n (~ --) light (in weight), Av.-And.
*daha- (~ o) few (adv.), Tsez. *=ihV small, little), which does not look
reliable.
Urart. zeldi (if zel=di) can theoretically be compared to NCauc. *slV (~ --)
light (of weight) (> Lezgh. *sl- id., WCauc. *a-s id.).
59

78. smoke: hiuri
Ugar.C.
The well-attested hubr=uhi incense burner (GLH, 109; Wilhelm, 19721975;
Dietrich/ Mayer, 1997, 86) may be derived from the same root, although the
vowel alternation is not clear.
No obvious etymology. Cf. Sino-Cauc. *hVV breath; spirit > NCauc.
*h[]a (~ --, -e) steam (> Nakh *e steam, Av.-And. *halHV (~ -) to
boil ; boiling; steam, Tsez. *hel- (~ -) to boil, Lezgh. *hIela breath;

58
Translated as to amaze, be amazed in Wilhelm, 1992b, 64; Wilhelm, 2006, 183 and,
with reservations, in Giorgieri, 1999, 378.
59
Further, as tentatively proposed in Sccet.dbf sub Sino-Cauc. *slV (~ --), it can be
united with STib. *l (s-) easy, slow (> Chin. *sla slowly, leisurely, easy, relax,
Tib. sla easy).
steam, WCauc. *a smoke; god, spirit), STib. *l
spirit, Yen. *ir

1
-
(~ x-) breath, spirit.
60

Hurr. hiuri resembles Nakh *ur smoke (thus Diakonoff/ Starostin, 1986, 57)
the main Proto-Nakh term for smoke, which lacks, however, any NCauc.
etymology and certainly does not reect the main Proto-NCauc. root for
smoke (rather a HU loan in Proto-Nakh, see 5.5 below and ,
2011).

79. to stand: unattested.
Cf. the verb meh- (Bo.Bil., Bogh.) to go stand, place oneself, come to stand,
which corresponds to Hittite tiya- to go stand, place oneself, come to stand
and sara ar- to stand up in Bo.Bil. (Neu, 1996, 266, 405; Catsanicos, 1996,
209 f. ; Campbell, 2007, 87; Wegner, 2007, 267; Wilhelm, 1991b, 45; Wil-
helm, 1992a, 132).

80. star: unattested.
The rst candidate is Hurr. zuzuhe (Ugar.C.), but it seems more probable that the
Hurr. word denotes something like container, box, vessel, not star (see
discussion in Huehnergard, 1987/ 2008, 2627, 54, esp. 377), in which case it
can be a variant of Hurr. zizzuhi cruche, known as a loanword in Hittite text
(GLH, 306). In favour of the translation star speaks, however, the fact that
the Sumerogram for star is once attested with the complement -hi- in the
Hurrian context : MUL-hi- (Haas/ Wegner, 1996, 286).
The second candidate is Hurr. ilziri (Bogh.), i. e., morphologically ilz=i=ri (?),
for which see Haas/ Wegner, 1996, 285287. Such an interpretation is con-
textually likely, but does not seem fully reliable.
, 1995/ 2007, 630, tentatively compares Hurr. zuzuhe to the basic
Proto-NCauc. term for star: *
whr / *
whr (~ -),
61
but the proposed
explication of the Hurrian form (reduplication and *hr > *rh > h) looks
somewhat factitious.
81. stone: unattested.
Urartian: the word kauli may mean stone (in general), thus , 451 (fol-
lowing Diakonoff); cf. CdTU 2, 132.

82. sun: imigi
Mit., Ugar., Bogh., Msk.
GLH, 232; Wegner, 2007, 277; Dietrich/ Mayer, 2010, 276.
The Sun-gods name. Probably imi=gi with the common nominal sufx -k-. Cf.

60
Diakonoff / Starostin, 1986, 55, compare NCauc. *h[]a to Hurr. ahri incense (in
fact, the verbal root ahr- to burn incense, LHL 1, 41), which does not seem reliable.
61
Further, as proposed in Sccet.dbf sub Sino-Cauc. *whr, to Basque *isar star and
STib. *cer (~ -i-) shine.
also the term api (~ aapi ?) sun-disk(?), Haas, 1998a, 240.
Urartian: iwini, the name of the Sun-god (, 490; CdTU 2, 290).
, 1985, 48, 50 fn. 10, supposes that Hurr. imi(=gi) and Urart.
iwi(=ni) are etymologically related, although the correspondence w ~ m is
abnormal. In his turn, Diakonoff, 1971, 81, suspects that Urart. iwini might
be borrowed from Hittite siu(na)- god (in general). Both solutions do not
seem likely.
How we should reconstruct the Proto-HU term is unclear. In Anatolian and
Mesopotamian traditions, the words for sun and moon are identical to the
names of solar and lunar deities, therefore these terms are potentially subject
to replacement with loanwords, when the local cult changes.
62
It is rather
suspicious that, out of 5 discrepancies between Hurrian and Urartian 110-
item wordlists (see 4 below), two are designations of these heavenly bodies.
Hurr. imi- is compared by Diakonoff/ Starostin, 1986, 23, to NCauc.
*mw (~ --, -I-) day (vel sim.) (> Nakh *ml sun, Andian *mii
sun, WCauc. *mawV day; despite the attested meaning sun in two
ECauc. groups, it is certainly not the basic Proto-NCauc. term for sun
63
),
assuming a metathesis in the Hurr. root and further the shift *i > i, which
are not self-evident.
Cf. the basic Proto-Semitic term *
x
Vm
x
(< redupl. *
x
Vm-
x
Vm?) sun, SED
1, CI.
64

Urart. iwi=ni can be compared to NCauc. *wI sky (> Av.-And. *zibu
sky; day, Lak saw sky, Dargwa *cab sky; rain, Lezgh. *caw sky,
Khin. ca sky, WCauc. *zV sky). Also probably, as proposed in
Sccet.dbf sub Sino-Cauc. *w, to Burush. *sa sun; day.
Lexicostatistical match: Urart.-Burush. ?

83. to swim: unattested.

84. tail : unattested.

85. that: ani
Wilhelm, 2008a, 95; Wilhelm, 1984; Giorgieri, 2000, 221; Hazenbos, 2005,
143; , 2010b, 137; LHL 1, 88.
The system of Hurrian demonstrative pronouns is not entirely clear. The analysis
is based mostly on the data of Mit. According to Wilhelm, 2008a, 95; Gior-
gieri, 2000, 221; Hazenbos, 2005, 143; , 2010b, 137, the known

62
As, e. g., in Hittite, where sun was denoted by a Hattic loanword.
63
Cf. NCauc. *wiroq
sun, which retains its meaning in all groups excepting Nakh &
Andian.
64
The phonetic status of *
x
is in fact unclear (it may be a ghost phoneme in the Proto-
Semitic reconstruction).
system is as follows: an=ni this (demonstrative proximal) / an=di this
(anaphoric proximal) / ani that (anaphoric distal).
Differently in Wegner, 2007, 84: an=ni ~ ani this / an=di that.
Differently in Dietrich/ Mayer, 2010, 201: an=ni ~ ani this (article-like) /
an=di this (deictic).
The pronoun an=di (LHL 1, 87), however, is translated by Hittite aba- that
(esp. concerning the addressee) in Bo.Bil. (that man, that story, etc.),
see Neu, 2003, 304. This fact might incline us to prefer Wegners system.
Note, however, that the numerous passages from Mit., where an=di occurs,
speak rather in favour of the translation this, not that for an=di (Wilhelm,
1984).
Tentatively, I prefer to follow Wilhelms analysis, i. e., treat an=ni as Swadeshs
this and ani as Swadeshs that.
Hurr. ani that can be analyzed as an(=i), i. e., a plain deictic morpheme an,
which is also used in an=ni this (No. 86), an=di this (anaphoric proxi-
mal), etc. No obvious etymology for an in such a case.
If Hurr. ani that is to be analyzed as a=ni (for the second element, see notes on
this), then a- could be compared to the following pronominal deictic mor-
pheme: Sino-Cauc. *[a] > NCauc. * (> Nakh *o that, Lezgh. *a a
demonstrative stem of far deixis, WCauc. *a a demonstrative stem), STib.
*a/ *ha- (a demonstrative pronoun), Basque *ha (a demonstrative stem
used in expressions for this and that).
65

86. this: an=ni
Wilhelm, 2008a, 95; Wilhelm, 1984; Wegner, 2007, 84; Giorgieri, 2000, 221;
Hazenbos, 2005, 143; , 2010b, 137; Dietrich/ Mayer, 2010, 201;
LHL 1, 89. See notes on that.
Urartian: demonstrative this is ini, cf. also an=a=ni this (anaphoric), Wil-
helm, 2008b, 114; , 448.
It is natural to suppose that Hurr. an=ni this and Urart. ini possess an
identical structure, that is, the Urartian form should be analyzed as i=ni.
For Hurr. an in an=ni see notes on that above.
Urart. i in i=ni this could be compared to the following pronominal deictic
morpheme: Sino-Cauc. *i > NCauc. *i this (> Nakh *i this, Thsez.
*i-/ *ji- a demonstrative morpheme, Dargwa *i- this, Lezgh. *i a demon-
strative pronoun, WCauc. *jA this; the best candidate for the status of
Proto-NCauc. this), STib. * (Chin. *ij, Burm. i, Lushai i, Kiranti *,
normally a demonstrative stem of proximal deixis), Burush. *i-ne that.
66

As for the HU deictic morpheme *-nV (enclitic 3
rd
p. pronoun, also used as a se-
cond part of various pronouns, see some forms above), its obvious compa-

65
Diakonoff / Starostin, 1986, 8283 with a Hurr.-NCauc. comparison.
66
Diakonoff / Starostin, 1986, 83, with a Hurr.-NCauc. comparison.
randum is Sino-Cauc. *nV > NCauc. *nV (> Av.-And. *-nV- a deictic par-
ticle used only in conjunction with other deictic stems, Tsez. *nV a deictic
particle used only in conjunction with other deictic stems, WCauc. *nV a
demonstrative stem, primarily adverbial with far deixis), STib. *nV (> Chin.
*nj that (late), Kachin na
4
here, ni
4
thus, in this manner, Kiranti
*na- that, there), Burush. *-ne in *i-ne that, *khe-ne this.
67
Note the
morphosyntactical similarity between HU, NCauc. and Burush. morphemes.
Lexicostatistical match: Urart.-NCauc.

87. thou: fe-
Bo.Bil., Mit., Bogh.
Wegner, 2007, 82 f. ; Giorgieri, 2000, 219; GLH, 295; Dietrich/ Mayer, 2010,
193; Neu, 2003, 298.
The form iniam (or inibi ?) thou in an Ugaritic lexical list is enigmatic (Hueh-
nergard, 1987/ 2008, 2425, 47, esp. 376).
An obvious comparandum for Hurr. fe- seems Sino-Cauc. *wV thou (direct
stem?) > NCauc. *u thou (direct stem), Yen. *aw(/ *u) thou, Burush.
*u-n thou.
68

Lexicostatistical match: NCauc.-Yen.-Burush.

88. tongue: irde
Bogh., Ort.
Wegner, 1995a, 117 f. ; Girbal, 1994, 173; nal, 1998, 59, 64.

89. tooth: eri ~ ir=ni
Ugar.C., Bogh.
Andr-Salvini / Salvini, 1998, 16 f. ; Grke, 2010, 72 f.
The Ugaritic xation i-x-ni is unproblematic semantically, but not phonetically,
because the reading of the second sign is uncertain: IR or NI, see discussion
in Andr-Salvini / Salvini, 1998, 16 f. On the basis of Akkadian innu tooth
(< Common Semitic *inn- tooth), the aforementioned authors tend to read
the Hurrian form as ini=ni, suspecting a Semitic loanword here. It is not
clear, however, whether Akkadian inuence on the Hurrian language was so
profound as to make the borrowing of such a basic term possible (for Akka-
dian loanwords in Hurrian, see notes on 91. two below).
69

The second form eri comes from the Hurro-Hittite ritual of Atu, where this

67
68
69
Hurr. inniber=u=hhe ~ iniber=u=hhe of ivory (Mit.) is a regular adjective in
=u=hhe from a transparent Akkadian loanword (Akkad. inni pri ivory (lit. tooth of
elephant)). This form can hardly clarify the reading of the Hurrian term for tooth.
term occurs in a large, but partially damaged list of body parts. The transla-
tion tooth for eri has been cautiously proposed by Grke, 2010, 72 f., alt-
hough with many reservations. In fact, tooth ts the context very well, be-
cause eri is adjacent to the group of body parts associated with head (nor-
mally, such lists were organized in accordance with the anatomic sequence).
The list runs as follows (see edition in Grke, 2010, 60ff.): eye, [], ear, in-
ner ear, nose, saliva(?), neck (i. e. throat ?), tongue(?), eri tooth(?), fui
?, kerai ?, fayari ?, kelli ?, haripi ?, biceps / upper arm, kuri ?,
hand, tatti ?, puda-x ? [].
I suppose that these data should be sufcient to postulate the Hurrian term for
tooth as eri-.
An appropriate comparandum could be seen in the basic NCauc. term *cIV
tooth (Nakh *ca, Av.-And. *colu, Tsez. *sl A (~ s-), Dargwa *cula, Lezgh.
*sl, Khin. culoz, WCauc. *cA, all meaning tooth). As proposed in
Sccet.dbf (sub Sino-Cauc. *tIV), may correspond to STib. *Vj fang.
Note that this is apparently not the ProtoSino-Cauc. root for tooth (see
5.3 below).
70

Lexicostatistical match: NCauc.

90. tree: tali
Bo.Bil., Ugar.C., Bogh., Msk.
GLH, 253; Catsanicos, 1996, 282; Wegner, 2007, 282; Huehnergard, 1987/
2008, 2627, 54.
Polysemy: tree / wood (timber).
71
Another term for wood (timber) may be
here (Campbell, 2007, 158 ff. ; Campbell, 2008, 284).
No obvious etymology. Diakonoff/ Starostin, 1986, 25 compare Hurr. tali to
NCauc. *dwI stick (> Nakh *tl poplar; chock, wedge, Av.-And. *dalV
stick, Lak tala log, cudgel, club, Dargwa *tal post,
72
Lezgh. *tal
branch; long stick, pole).
73
Such a comparison is phonetically possible, but
not very apt semantically, because the meaning shift tree > wood (timber)
is cross-linguistically common, but not vice versa.
74

70
On the contrary, for the Hurrian shape **ini=ni, no good comparanda has been found
among Sino-Caucasian languages.
71
The meaning tree is apparently attested in RS h.2 obv. 12 (Laroche, 1955, 330;
Laroche, 1968, 458).
72
Caucet.dbf adds here Dargwa *tultV/ *taltV tree (reconstructed on the basis of
forms from Chirag, Kubachi and some other dialects), which is dubious. If Chirag-Kuba-
chi *tultV/ *taltV tree is indeed a reduplicated variant of the discussed NCauc. root,
this is certainly a very late and local innovation (the Proto-Dargwa word for tree is
*kalkV).
73
Further, as proposed in Sccet.dbf sub Sino-Cauc. *dwaI, to STib. *tul sharp wooden
stick, Burush. *tol awl, ? Basque *i-dul-ki block of wood; pedestal.
74
An idea that Hurr. tali reflects an old ProtoSino-Cauc. word for tree (which changed
91. two: ini
Nuzi, Mari, Bo.Bil., Bogh., Mit., Ugar.A., Msk.
GLH, 234; Wegner, 2007, 81; Wilhelm, 2008a, 101; Giorgieri, 2000, 222;
Catsanicos, 1996, 249; Dietrich/ Mayer, 2010, 191.
Urartian: Diakonoff (Diakonoff, 1971, 106; Diakonoff/ Starostin, 1986, 37)
quotes ii as an Urartian designation of 2, but this is not reliable. The
underlying Urartian form ii=ni was previously translated as 3 by -
, 1963, 76, 91, himself (repeated with reservations in , 464), cur-
rently left without translation in CdTU 2, 197.
75

No obvious etymology. The Hurrian stem resembles Yen. *si- four (which
lacks Sino-Cauc. etymology), but details are vague.
Diakonoff/ Starostin, 1986, 37 compare Hurr. ini 2 to the following NCauc.
root (as it is now reconstructed in Caucet.dbf): NCauc. *nwsi > Nakh *i,
obl. *ina- two, Lak nuwa ram (more than 2 years old), Lezgh. *nuaj
(~ -iw-) sheep (about 2 years old), WCauc. *n a ram slaughtered in
honor of the guest. The main NCauc. root for two is *qHw, however,
which retains its basic semantics in all groups excepting Proto-Nakh.
76
The
authors of NCED suspect that *nwi was a suppletive oblique stem of the
Proto-NCauc. designation of two (*qHw), although it is not supported by
any positive evidence. Phonetic correspondences within proposed NCauc.
*nwi are not regular, and I suppose that it is a ghost etymology; particu-
larly Nakh *i, obl. *ina- two should be kept apart. The Nakh numeral
represents one of a number of Proto-Nakh words which are isolated within
NCauc., but phonetically match Hurrian forms maybe old HU loanwords in
Proto-Nakh, see 5.5 below and , 2011.
The phonetic coincidence between Hurr. in(i) 2, Nakh *i / *ina- 2 and Se-
mitic *Vny 2 (Akkad. in, Ugar. n, Phoen. n etc., i. e., forms of the
shape in-, which were widespread among ancient Semitic languages of the
Near East) is, of course, very suspicious.
Such a similarity between the Hurrian and Semitic forms can be accidental, but
it is also possible to suppose that the Hurrian numeral was borrowed from
Akkadian, e. g., as a trade term. Cf. Hurr. ee 6, which apparently origi-
nates from Akkad. eu 6 (further to Ugar. 6 etc. < Sem. *d 6), and
other numerous Akkadian loanwords in Hurrian, for which see Diakonoff,
1971, 77 ff. ; Wilhelm, 2008a, 103; Dietrich/ Mayer, 1994, 109; Dietrich/
Mayer, 2010, 278 f.
77

its meaning in other Sino-Cauc. (proto-)languages) is possible only theoretically.
75
Cf. also the complemented ideogram III-e 3.
76
Further apparently corresponds to Proto-Yen. *x-na two.
77
Note that the borrowing of the term for 2 is a typological rarity. Normally, such a su-
perseding of the original term takes place in the situation of a complex borrowing of the
numeral set, cf. examples from languages of East and South-East Asia. But the wander-
ing nature of the numeral 2 is attested, e. g., in some African languages, namely Cush-
As for Proto-Nakh, it does not demonstrate any traces of contact with Semitic,
so, to explain Nakh *i / *ina- 2, an intermediary language should be
postulated, the only known candidate being Hurrian (or its close relative,
e. g., Urartian). Cf. also the Proto-Nakh numeral *qo [direct] / *qaa- [obl.]
3, which also lacks NCauc. cognates, but resembles Hurr. kig- 3 (Dia-
konoff/ Starostin, 1986, 59; Blaek, 2010, 118).

92. to go
1
: u-
Haas/ Wegner, 2010, 97 ff. ; Huehnergard, 1987/ 2008, 65; Prechel / Richter,
2001, 370.
A rare case, when the difference between dialects can be traced. Two main verbs
are attested in the meaning to go, walk.
1) u- (OBab., Mari, Bogh., Ugar.C.);
2) itt- (Mit.), see below.
Apparently, no verbs with generic semantics of going are attested in Bo.Bil. Cf.
various specic verbs of movement in Bo.Bil., such as far=i- (= Hittite iya-
to go, walk, but the contextual meaning is something like to take the road,
rather than general to go), Neu, 1996, 228; Catsanicos, 1996, 224, 228;
ehl- to go in (= Hittite anta iyanna-), Catsanicos, 1996, 228; hab- to go
forward (= Hittite para pai-), Catsanicos, 1996, 220, 228; and so on.
Urartian: u-, u=t=a-, apparently one of the two verbs for generic to go; see
notes on 16. to come for discussion.
The Proto-HU root for to go is therefore *u()-. Its phonetically best
comparandum is Sino-Cauc. *HuSV > Yen. *us- (~ x-) away, asunder,
Burush. *-us-/ *-was- to go out, away; formally this is not a Common
Proto-Sino-Cauc. root, but a local Yen.-Burush. isogloss. No lexicostatistical
matches.
For Hurr. hab-, , 1998/ 2007, 747 proposes NCauc. *=ixwV to go
(vel sim.); to ow, which in my opinion does not seem apt (for a hypothet-
ical development C > CVw, see notes on 110. year below).

92. to go
2
: itt-
Wegner, 2007, 257; GLH, 128; Haas/ Wegner, 2010, 97.
Reliably attested only in Mit., where, as noted already by A. Goetze in the
1930s, the verb itt- to go, go away is opposed to un- to come (No. 16).
78

Cf., e. g., And now I have given my brothers wife, and she has gone (itt-) to
my brother. When she comes (un-), my brother will see her (Wilhelm,
1992b, 66).

itic, Omotic and Surmic (G. Starosin, pers. com.).
78
It is a typologically normal situation, when the meanings to go (in general) and to go
away merge into one verbal root, which is lexically opposed to expressions for to
come.
In Dietrich/ Mayer, 2010, 247, itt- is glossed as to depart (losgehen), whereas
the generic meaning to go is ascribed to the verb i-. However, according to
Dietrich & Mayers translation proper, the mentioned Mit. passages do not
show the assumed direct meaning to go, walk (of humans) for i-. Alterna-
tively, the verb i- can be analyzed as to be worth, valuable in some Mit.
contexts (thus Girbal, 1992, 164 f., followed with reservations by Giorgieri,
1999 and Wilhelm, 2006).
Urartian: ul-, ul=a-, apparently one of the two verbs for generic to go; see
notes on 16. to come for discussion.
Hurr. itt- can be compared to Sino-Cauc. *=otV to move (vel sim.) >
NCauc. *=otV to move, go (vel sim.) (> Nakh *q-t- to get (somewhere),
Av.-And. *=ut- to reach, Tsezian *=ot- to turn back, return, Lezgh. *ata-
to arrive, to walk), STib. *ti to go (vel sim.) (> Chin. *t to go, pro-
ceed).
Diakonoff/ Starostin, 1986, 50 f., should be right, comparing Urat. ul- to
NCauc. *=or\ to go (vel sim.) (> Nakh *-- to walk, roam, Av.-And.
*=iu- to walk (vel sim.), Tsezian *o- (~ --, -L-) to walk, Lak
=u=ka- to go out, Lezgh. *are- to come, Khin. ka- to come, WCauc.
*V to enter). Note that this root can theoretically be the basic Proto-
NCauc. term for to come.

93. warm: unattested.

94. water: iwe ~ iye
Mari (iwe), OBab. (iye), Bo.Bil. (iye), Ugar.A. (y), Ugar.C. (iye), Bogh.
(iwe ~ iye), Ort. (iye), Subar. (eu).
GLH, 230; Wegner, 2007, 276; Catsanicos, 1996, 282; Dietrich/ Mayer, 1994,
75, 79, 82; Dietrich/ Mayer, 1997, 83; Grke, 2010, 340; Prechel / Richter,
2001, 358; nal, 1998, 62; Hra, 2010, 134 [l., 204], 547.
Polysemy: water/ river. For the glide uctuation, see notes on 57. name (tiwe
~ tiye).
No obvious etymology. Diakonoff/ Starostin, 1986, 49 compare Hurr. iwe ~
iye to NCauc. *=V(-)r to wet ; to be wet, soak (> Nakh *=il- to wash,
Av.-And. *=a- (~ -o-, --) to become liquid (after shaking); to rain,
Lezgh. *ar- to soak; to urinate; to melt ; liquid), assuming the shift * >
Hurr. before front vowels. Alternatively, Diakonoff/ Starostin, 1986, 49,
propose to compare the Hurr. stem to NCauc. *xnI water (> Nakh *i,
Av.-And. *ini, Tsez. * B, Lak in, Dargwa *xin, Lezgh. *n:, Khin.
xu, all meaning water), assuming the palatalization *x > and the simpli-
cation *n > * > in Hurrian. Both solution are not self-evident phoneti-
cally and in both cases Hurr. -w- remains unexplained.

95. we: a=tti(=lla) / ie=- / a=-
Bo.Bil., Bogh.
Wegner, 2007, 83, 276; Wegner, 1992; Wilhelm, 2008a, 95; Giorgieri, 2000,
219; Neu, 2003, 298.
Paradigm: a=tti(=lla) [abs.] / ie=- [erg.] / a=- [obl.].
Normally, the Hurrian personal plural pronouns we, you & they are
transparent derivatives from the corresponding singular pronouns I
(No. 42), thou (No. 87) & he with plural exponents.
79
In view of this, a
possible comparison with the NCauc. pronoun *i [abs.] / *a- [obl.] / *iV
[gen.] we (exclusive) seems unnecessary. I suppose that Hurrian matches
the Yen. model, cf. Yen. *a I ~ *a- we (- is a pluralizator), although
the plural exponents in Hurrian and Proto-Yenisseian are different.

96. what?: aw-
Bogh.
Wegner, 2010.
Polysemy: who (interrog.) / what (interrog.).
The Hurrian interrogative stem iya (Bo.Bil., Mit.) is attested in the meaning
what (relative) and why? (Neu, 2003, 305; Wegner, 2007, 84, 122; GLH,
118; Dietrich/ Mayer, 2010, 202; Campbell, 2007, 76 ff.), not in the meaning
what ? (interrog.) (pace , 2010b, 138).
No good etymology for Hurr. aw-. This resembles some dispersed Sino-
Cauc. forms for what ?, like Nakh *fu-(n) what ? (an isolated form within
NCauc. as well as Sino-Cauc.), Burush. *be what (interrog.) or STib. *Pa
(one of several candidates for the status of Proto-STib. term for what ?).

97. white: unattested.

98. who?: awi ~ abi
Bo.Bil., Mit.
LHL 1, 6; Wegner, 2007, 85; Neu, 2003, 305; Wegner, 2010, 406 ff. ; ,
2010b, 138; Dietrich/ Mayer, 2010, 238.
Polysemy: who (interrog.) / what (interrog.).
In Ugar.C., awi probably has an additional meaning who (relat.) / what (relat.),
Laroche, 1968, 462; Huehnergard, 1987/ 2008, 4041, 86. Note that the nor-
mal relative pronoun what is iya (Wegner, 2007, 84, 122; GLH, 118;
Campbell, 2007, 76 ff.).
See notes on 96. what ?.

79
Abs. a=tti(=lla) we abs. i=te I with the additional change of the root variant i
to a (a paradigmatic levelling after the oblique stem a=- ?).
99. woman: ati ~ ata
Mari, Bo.Bil., Mit., Bogh., Ugar.C., Ugar.A., Subar.
LHL 1, 136; de Martino/ Giorgieri, 2007; GLH, 62; Wegner, 2007, 250; Catsa-
nicos, 1996, 198; Andr-Salvini / Salvini, 1998, 10; Dietrich/ Mayer, 2010,
236; Hra, 2010, 152 (l. 75), 281; Krebernik, 2001, 158.
Polysemy: woman/ wife/ female (n.).
Urartian: as in the case of 51. man and 64. person, apparently no word for
woman (sg.) is attested, but two morphologically singular terms with the
meaning women are known: wediani (, 470; CdTU 2, 223 f.) and
lutu (, 453; CdTU 2, 142 f.). Both wediani and lutu seem full syno-
nyms and may even occur in one inscription in identical contexts. The differ-
ence is that attestations of wediani are earlier (these are known from inscrip-
tions of Ipuini and Minua on, whereas lutu is attested for the rst time in
Argiti Is inscriptions). Secondly, lutu should be considered a loanword be-
cause of the initial l- (the source of borrowing is unknown, however). Urar-
tian words in l- are few (, 453; CdTU 2, 141143) and only two of
them are attested in standard contexts and are translated unequivocally:
MUNUS
lutu
(ME)
women and
KUR
lulue foreign/ inimical land. The latter, to-
gether with Hurr. lulahhi a k. of barbarians (vel sim.) (GLH, 160), was ap-
parently borrowed from Proto-Nakh *llu- neighbour (n.); in the neigh-
bourhood (Chechen and Ingush only
80
); for HurrianProto-Nakh linguistic
interactions, cf. 5.5 below and , 2011.
81

No obvious etymology for Hurr. ati. Diakonoff/ Starostin, 1986, 39 compare
Hurr. ati to poorly attested Dargwa *cad- female (n.) and Nakh *stVd-
(~ -d-, -g-, -b-, -w-, -j-) female (n.) (sound correspondences between the
Dargwa & Nakh proto-forms are irregular), which does not seems reliable.
On the other hand, Caucet.dbf proposes to unite Hurr. ate woman with
Hurr. id=u=ri young girl, maiden (GLH, 229, Catsanicos, 1996, 275) and
connect both to Nakh *u bitch (the Nakh form lacks NCauc. cognates, a
hypothetical Proto-NCauc. could be *HVV),
82
which is not apt as well.

80
According to Caucet.dbf, the Nakh root has a reliable NCauc. etymology: *LlV
guest, neighbour (Av.-And. *irV person of the same age; parent-in-law ; Tsezic
*r (~ -e-, --, -l) guest ; Khin. ij guest). Diakonoff / Starostin, 1986, 48, treat the
Hurro-Urartian terms and NCauc. *LlV as etymological cognates, which is less likely.
81
The Hurrian term was further borrowed as Hittite
(L)
lulah(h)i- (generic designation
of uncivilized mountain dwellers), Luwian (C) lulahi(ya)- of the mountain-dwellers
(HEG LM, 70 w. lit.). In Kassian, 2010a, 428 fn. 74, I have assumed that there might be
some direct Proto-Nakh loans in Hittite, but currently I tend to treat such Hittite &
Luwian terms as Hurrian (i. e., Proto-Nakh > Hurrian > Hittite). Subsequently, the word
penetrated into southwest Anatolia, cf. Grk. one of the aboriginal tribes of
southwest Anatolia (Carians, according to Herodotus).
82
The comparison Hurr. id=u=ri ~ Nakh *u has been previously proposed in Diako-
noff / Starostin, 1986, 28.
Urart. wediani is likewise unetymologizable.
83

100. yellow: unattested.

101. far (adv.): unattested.
A possible candidate is awattu- (Mit.), Girbal, 1992, 166 f. ; Wegner, 2007, 248;
LHL 1, 152; quite differently translated as word in Dietrich/ Mayer, 2010,
237 (< Akkad. awtu word). The stem ukku must be analyzed as 1, one
(No. 63), not far.
No obvious etymology for Hurr. awattu.

102. heavy: unattested.

103. near (adv.): unattested.
A hypothtical candidate is the verbal stem ul=uh- to be near(?) (Mit.), Girbal,
1992, 166 f. ; Wegner, 2007, 289; Campbell, 2007, 304, 306; translated as
(ver)ndern in Dietrich/ Mayer, 2010, 268. Hurr. ul=uh- corresponds to Ak-
kad. an to become different, change (intr.) in Qatna (narrowly glossed as
Koalition wechseln in Richter, 2005, 29).
Urartian: cf. salmathi near(?), in the neighbourhood(?) (, 460); trans-
lated as on the left (?) or between(?) in CdTU 2, 181. Cf. also limeye
near(?), in the neighbourhood(?) (, 453), without translation in
CdTU 2, 141; note that limeye should be a loanword because of the initial l-.
No obvious etymology for Hurr. ul-.

104. salt: unattested.

105. short: unattested.
106. snake: api
OBab., Bogh., Ugar.C.
LHL 1, 94; Huehnergard, 1987/ 2008, 379; Salvini / Wegner, 2004, 168; Andr-
Salvini / Salvini, 1998, 9; Prechel / Richter, 2001, 359; Rseler, 1999, 399 f.
For the meaning (mythological) Serpent, (Sea)dragon see Dijkstra, 2006.

107. thin: nigale
Ugar.C.

83
I. Yakubovich (pers. com.) suspects that Urart. wediani could be in some way related
to Indo-Iranian *vadh wife (especially of the son, brother, etc.) as a loanword via
Mittanian. Also cf. Hurr. waduranni, if this term indeed means dowry (see Dietrich/
Mayer, 2010, 267 for an alternative translation plentiful (vel sim.) and discussion).
The usage is unknown. Cf. also harai thin(?) in the collocation anui harai
thin bread(?) (Bogh.), Campbell, 2007, 202; GLH, 94.
No obvious etymology for Hurr. nigale.
108. wind: unattested.

109. worm: unattested.
The form wali ~ weli (Mari) can hardly mean worm, cf. Giorgieri, 2002, 69 f. ;
Campbell, 2007, 390.

110. year: awali
Mit., Bogh., Ugar.C.
GLH, 221; Wegner, 2007, 275; Dietrich/ Mayer, 2010, 257. Maybe with a
syncopated variant awli (Haas, 1998a, 240).
Urartian: ali year (, 462; CdTU 2, 186). A match with the Hurr. term
(the loss of intervocalic -w- may be regular in Urartian).
No obvious etymology for HU *awal- year. Diakonoff/ Starostin, 1986,
30, compare it to NCauc. *swrho (~ --) old (vel sim.) (> Nakh *o year,
Av.-And. *-asara- (~ -o-) old, Dargwa *dus year, Lezgh. *jisV- old,
Khin. s year, WCauc. * old),
84
which does not seems apt (particularly
because the development C > CVw is typologically abnormal
85
).
Conclusions
4. In the above 110-item list, ca. 65 Hurrian and ca. 22 Urartian slots are lled.
86

The available data are too scant for formal lexicostatistical and/ or glottochrono-
logical calculations, but the lexicostatistical comparison of Hurrian and Urartian
meets our intuitive expectations. Out of the 22 Urartian items, 15 or 16 words
are cognate to the corresponding Hurrian terms:
01. all : Hurr. ua=lla ~ Urart. ui=ni-.
12. to burn (tr.): Hurr. am- ~ Urart. am-.
? 16. come: Hurr. un- ~ Urart. nun=a-.
33. to give: Hurr. ar- ~ Urart. ar-.
37. hand: Hurr. u=ni ~ Urart. u-.
39. to hear: Hurr. ha- ~ Urart. ha-.

84
An undoubted Proto-NCauc. root for year is *wnI (> Av.-And. *riin year, Lak
in year, Lezgh. *sn year, WCauc. *A year). The meaning year for some forms
listed above sub *swrho is the result of either the semantic shift old > year or a
contamination with the root *wnI year.
85
Clusters obstruent + w, conventionally reconstructed for Proto-NCauc., are in fact
labialized consonants (*s, *t and so on).
86
In one case, the Urartian term is attested, while the Hurrian one is still unknown:
77. small.
40. heart: Hurr. tia ~ Urart. ti=ni.
42. I: Hurr. i-/ u- ~ Urart. i-/ u-.
55. mountain: Hurr. pab=ni ~ Urart. baba=ni.
57. name: Hurr. tiye ~ Urart. ti=ni.
59. new: Hurr. uhe ~ Urart. uhi.
62. not: Hurr. =u ~ Urart. u=i, =u=ri.
63. one: Hurr. u=kki ~ Urart. u=sini.
67. road: Hurr. hari ~ Urart. hari.
92. to go: Hurr. u- ~ Urart. u-.
110. year: Hurr. awali ~ Urart. ali.
In 5 positions, Hurrian and Urartian diverge:
87

22. earth: Hurr. ee ~ Urart. i(u)ra-.
54. moon: Hurr. kuuh ~ Urart. elardi.
71. to say: Hurr. hil- ~ Urart. ti-.
82. sun: Hurr. imigi ~ Urart. iwini.
86. this: Hurr. an=ni ~ Urart. i=ni.
If we extrapolate these data to the full 110-item or 100-item wordlist, we shall
come up with ca. 2830 discrepancies in the basic lexicon between Hurrian and
Urartian, which suggests 2500 BC as an approximate date of the split of the HU
proto-language. Of course, due to its extrapolative origin, this result is not relia-
ble, although such a dating does not contradict our expectations either.

5. External relationship of HU is a more intricate question. In accordance with
the aforementioned principle of phonetic similarity, one can single out the
following 12 cases, when HU terms correspond to Swadeshs words of some
proto-languages within the Sino-Caucasian macro-family.

Group A: North Caucasian ~ Sino-Tibetan
59. new: HU *uhe ~ NCauc. *nV new ~ STib. *sn new [~ Yen.
*-sa to begin].

Group B: Yeniseian ~ Sino-Tibetan
53. meat: Hurr. uzi ~ Yen. *ise meat ~ STib. *a meat.

Group C: North Caucasian ~ Yeniseian (~ Burushaski)
42. I: HU *i-/ *u- ~ NCauc. *z/ *ez(V)/ *iz(V) I ~ Yen. *a I, Bu-
rush. *ja I.
87. thou: Hurr. fe- ~ NCauc. *u thou, Yen. *aw thou, Burush. *u-n
thou.

87
For the cases of sun and moon, cf. notes on 82. sun above, however.
Group D: Yeniseian
08. black: Hurr. time=ri ~ Yen. *tum- black [~ STib. *rVm dark?].
09. blood: Hurr. zur=gi ~ Yen. *sur blood [~ NCauc. * wV vital force
(vel sim.) ~ Basque *i-serdi sweat ; sap].
50. louse: Hurr. aphe ~ Yen. *oke louse [~ NCauc. *mkV a k. of in-
sect ~ STib. *m midge].
95. we: a=tti=lla / ie=- / a=- (I-PL) ~ Yen. *a- (I-PL) we.

Group E: North Caucasian
63. one: HU *u- ~ NCauc. *cH 1.
86. this: Urart. i=ni ~ NCauc. *i this [~ STib. * (proximal deixis) ~ Bu-
rush. *i-ne that].
89. tooth: Hurr. eri ~ NCauc. *cIV tooth.
Group F: Sino-Tibetan
21. ear: Hurr. nu(h)i ~ STib. *nH ear [~ NCauc. *wnV inner ear?].

5.1. The rst thing to be noted is that the number of the assumed HUYeniseian
isoglosses (7) or HUNorth Caucasian ones (6) stands in sharp contrast with a
couple of hypothetical HUIndo-European or HU-Semitic matches discussed in
2 above.

5.2. Secondly, the distribution of these twelve HU words within the lexico-
statistical wordlist is signicant.
The core of the basic vocabulary is the so-called Swadesh 100-item (or 110-
item, as in the present paper) wordlist. The Swadesh list is not homogeneous,
but its entries possess different degrees of stability. This factor was called the
relative index of stability by S. Starostin, who calculated it for each element of
the Swadesh list proceeding from typological data of various language families
of the Old World (see , 2007a, and Starostin G., 2010a, for detail).
E. g., the ten typologically stablest words are: we
1
, two
2
, I
3
, eye
4
, thou
5
, who
6
,
re
7
, tongue
8
, stone
9
, name
10
; the ten least stable elements from the 110-item
wordlist are: big
101
, bark
102
, sand
103
, short
104
, good
105
, many
106
, mountain
107
,
wind
108
, belly
109
, small
110
. As proposed in Starostin G., 2010a, for long-range
comparison, it is reasonable to divide the Swadesh list into two parts: 50 ultra-
stable items (see the list in Starostin G., 2010a, 113) and the remainder 60
less stable ones.
88

88
Another stability ranking is proposed in Holman et al., 2008. This chart partially dif-
fers from Starostins one. Because Holmans results seriously contradict my typological
knowledge and intuition (e. g., Holman et al., 2008, claim that the stablest word in human
languages is louse, sic!), I prefer to use Starostins data.
All twelve HU words with Sino-Caucasian lexicostatistical cognates occur in
this ultra-stable half of the Swadesh list.
89
Because the statistical probability of
such a distribution within the 110-item wordlist is less than 0.0003, this situation
should prove that we indeed deal with a genetic relationship between HU and
Sino-Caucasian proto-languages, i. e., that the HU proto-language retained some
stabler elements of the ProtoSino-Cauc. Swadesh list, whereas less stable items
were being naturally superseded during millennia of separate development of
Proto-HU.

5.3. Third, the distribution of cognates of the aforementioned 11 HU words
within the Sino-Caucasian macro-family is not uniform on the contrary, differ-
ing tendencies can be traced here.
The core lexicostatistical schema of Sino-Caucasian macro-family currently
looks as follows (Bengtson/ Starostin G., forthc. ; Kassian, 2010a, 323, 405):
90

Sino-Caucasian
/ \
Sino-TibetanNa-Dene North Caucasian-Yeniseian
/ \
North Caucasian-Basque Yeniseian-Burushaski

As I have noted elsewhere (Kassian, 2011a, 143), the conguration of the ge-
nealogical tree of a linguistic family is important for semantic reconstruction. In
situation when a proto-word has several different meanings in ancestral lan-
guages, one of the strongest criteria for its semantic reconstruction is the
distributional one. Let us envision the following genealogical tree where L is a
proto-language and A, B, C are its daughter languages.

L
X or Y?

A
X
B
Y
C
X

A certain word in A means X, its etymological cognate in B has a different
meaning Y, whereas its cognate in C also means X. In the absence of addi-

89
The demonstrative pronoun this possesses the index of stability 28 (,
2007a, 838), although this together with that are currently excluded from the 50-item
ultra-stable wordlist for some typological reasons (see Starostin G., 2010a, 92).
90
The tree has been compiled by G. Starostin (pers. comm.) as part of the ongoing re-
search on the Preliminary Lexicostatistical Tree of the worlds languages (as part of the
Evolution of Human Language project, supported by the Santa Fe Institute).
tional evidence, the likeliest solution is that, in L, this word denoted X, not
Y, since, according to general scientic principles, we should prefer the most
economic scenario (one semantic shift X > Y in B vs. two independent shifts
Y > X in A and C).
Let us now look at the specic etymologies quoted in 5 above.
Among the three main branches of the Sino-Cauc. macro-family (NCauc.,
Yen., STib.), it is Proto-Yen. that has the maximum number of lexicostatistical
matches with HU seven etymologies (groups B, C & D above).
Out of them, one etymology (group B) represents a basic ProtoSino-Cauc.
term for the corresponding Swadesh meaning:
53. meat: Hurr. uzi ~ Yen. *ise meat ~ STib. *a meat. According to the
distributional criterion described above, a match between Yen. and STib.
makes this word the best candidate for the status of the ProtoSino-Cauc.
term for meat.

Two etymologies (group C) represent common features of the NCauc.-Yen.
stock (in fact, I suspect that these pronouns go back to the ProtoSino-Cauc.
level, but they were lost in the STib. branch as a result of extensive rebuilding
and simplication of the Proto-STib. morphology):
91

42. I: HU *i-/ *u- ~ NCauc. *z/ *ez(V)/ *iz(V) I ~ Yen. *a I, Bu-
rush. *ja I.
87. thou: Hurr. fe- ~ NCauc. *u thou, Yen. *aw thou, Burush. *u-n
thou.

Two exclusive matches between HU and Yen. (part of group D) represent terms,
each of which is one of the several equally likely candidates for the status of the
corresponding Swadesh meaning in ProtoSino-Cauc. :
08. black: Hurr. time=ri ~ Yen. *tum- black [~ STib. *rVm dark?]. Ex-
pressions for black are very unstable in both NCauc. and STib.
branches, no unquestionable ProtoSino-Cauc. candidate for this mean-
ing can be proposed from the distributional viewpoint.
95. we: a=tti=lla / ie=- / a=- (I-PL) ~ Yen. *a- (I-PL) we. There
is no unquestionable ProtoSino-Cauc. candidate for we. Typologi-
cally, a derivation like we < I-PL and you < thou-PL is not so infre-
quent ; this pattern is attested in various languages and families around
the world (e. g., Turkic, Japanese, Dravidian, Old Armenian, Hokan and
so on; see WALS, 146, types 7 & 8; Siewierska, 2004, 79 ff.). On the
other hand, it seems that Proto-STib. did not discriminate between I and
we (both pronouns were expressed by *-forms), therefore the con-

91
For a typologo-etymological sketch of the personal pronouns in Sino-Cauc. languages
see now Starostin, 2010b.
cept of we as I-PL can be theoretically projected onto the ProtoSino-
Cauc. level.

Finally, two exclusive matches between HU and Yen. (the rest of group D)
should represent innovative Proto-Yen. terms, because there are other, more
valid, candidates for the status of the corresponding Swadesh terms in Proto
Sino-Cauc. :
09. blood: Hurr. zur=gi ~ Yen. *sur blood [~ NCauc. * wV vital force
(vel sim.) ~ Basque *i-serdi sweat ; sap]. A valuable candidate for
ProtoSino-Cauc. blood is *hwnV, which yields NCauc. *hwnV
blood (retaining its meaning in ECauc., but losing it in WCauc.) and
STib. *j (s-) blood (retaining its meaning in main STib. branches).
50. louse: Hurr. aphe ~ Yen. *oke louse [~ NCauc. *mkV a k. of in-
sect ~ STib. *m midge]. A valuable candidate for ProtoSino-Cauc.
louse is *rI (~ -), which yields Basque *sori louse (in NCauc. this
term shifted to worm) and STib. *ar ~ *rik (two variants occur
generally in complementary distribution within STib. languages).
Supposed matches between HU and NCauc. are less numerous six etymologies
(groups A, C, E). Out of them, one case (group A) represents a Common Sino-
Cauc. heritage:
59. new: HU *uhe ~ NCauc. *nV new ~ STib. *sn new [~ Yen.
*-sa to begin]. Consistency between NCauc. and STib. suggests a
Common Sino-Cauc. term *V new.
Two pronominal etymologies (group C) represent common isoglosses of the
NCauc.-Yen. stock, see notes on HU-Yen. matches above.

In two cases (group E), the NCauc. roots may theoretically represent Proto-Sino-
Cauc. terms:
63. one: HU *u- ~ NCauc. *cH 1. No valid ProtoSino-Cauc. candidate
for 1 can be proposed from the distributional viewpoint, but if the Yen.
singulative sufx *-s(a) was indeed grammaticalized from this numeral,
it makes this root the main candidate for at least NCauc.-Yen. 1.
86. this: Urart. i=ni ~ NCauc. *i this [~ STib. * (proximal deixis) ~ Bu-
rush. *i-ne that]. No unequivocal ProtoSino-Cauc. candidate for this
from the distributional viewpoint.

Finally, in one case one could suspect an innovative semantic in both HU and
NCauc. :
89. tooth: Hurr. eri ~ NCauc. *cIV tooth [~ STib. *Vj fang]. The re-
construction of the ProtoSino-Cauc. term for tooth is a nontrivial
problem, because there is a certain evidence that the Sino-Cauc. proto-
language lacked a generic term tooth, but possessed two separate
words, one for incisor (*HVdV) and one for molar (*HVwV) (this
opposition can be traced in Yen., STib. and Basque). If NCauc. *cIV
tooth indeed corresponds to STib. *Vj fang (thus Sccet.dbf sub
Sino-Cauc. *tIV), this may reect a specic Sino-Cauc. term for
fang, which totally superseded the old words for incisor and molar
in Proto-NCauc.

A number of HU-STib. matches is more limited: two cases can be noted (groups
A & F). Out of them, one etymology (59. new) represents a Common Sino-
Cauc. term, see notes on HU-NCauc. matches above. The second one may also
be a Sino-Cauc. heritage:
21. ear: Hurr. nu(h)i ~ STib. *nH ear [~ NCauc. *wnV inner ear?].
Expressions for ear differ in three main Sino-Cauc. branches, therefore
STib. *nH is one of the few equally acceptable candidates for the status
of ProtoSino-Cauc. ear.

Summing up, twelve aforementioned HUSino-Cauc. etymologies include:
2 cases of Common ProtoSino-Cauc. terms (53. meat, 59. new);
2 cases of Common NCauc.-Yen. and potentially Common ProtoSino-
Cauc. terms (42. I, 87. thou);
5 cases of potentially Common ProtoSino-Cauc. terms (8. black,
21. ear, 63. one, 86. this, 95. we);
2 cases of Proto-Yen. innovations (9. blood, 50. louse);
1 case of Proto-NCauc. innovation (89. tooth).
As one can see, these data could theoretically testify that HU is a third member
of the Yeniseian-Burushaski stock within the Sino-Caucasian macro-family.
92

On the one hand, such a result is intuitively surprising because of the geograph-
ical remoteness of modern Yeniseian languages (see the map in Kassian, 2010a,
418). On the other hand, a similar situation is observed with another extinct lan-
guage of Asia Minor Hattic, which also possesses some exclusive isoglosses
with Proto-Yeniseian (Kassian, 2010a), although there is no specic visible rela-
tionship between HU and Hattic.

5.4. A fact that could be interesting for future discussion is the number of pho-
netic coincidences between HU and Sumerian Swadesh items:

92
The main problem here is Hurr. eri tooth (No. 89), which is formally comparable to
the innovative NCauc. term *cIV tooth (see above). If this is not a ghost etymology,
forms in both languages could independently originate from the meaning fang (a typo-
logically normal semantic shift).
18. dog: Hurr. erwi ~ Sum. UR dog.
37. hand: HU *u- ~ Sum. U hand.
48. liver: Hurr. ur=mi ~ Sum. UR
5
liver.
53. meat: Hurr. uzi ~ Sum. UZU meat.
65. rain: Hurr. iena ~ Sum. E
3
to rain; rain (n.).
The genealogical afliation of Sumerian remains so far an open question and it
is not a goal of the present paper.
93
It should be emphasized explicitly, however,
that these words cannot be treated as borrowings between HU and Sumerian.
94

Typology of language contacts indicates that non-basic vocabulary is always
borrowed rst (see, e. g., Thomason/ Kaufman, 1988, 74 ff. ; Haspelmath, 2008,
50 ff.). It means that, if we assume that Sumerian loanwords are indeed present
in the HU Swadesh list (or vice versa: HU loanwords in the Sumerian list), one
should expect a great number of Sumerian cultural terms in HU (or HU cultural
words in Sumerian), which is certainly not the case: there are no recognizable
direct borrowings from Sumerian in the known HU lexicon
95
and, likewise, only
one HU cultural loanword in Sumerian has been revealed so far.
96

5.5. The last phenomenon I would like to state here is several phonetic matches
between HU and Proto-Nakh Swadesh lists. As I noted elsewhere (,
2011), three HU terms possess phonetic similarity with the corresponding Swa-
desh items of the Proto-Nakh language,
97
namely:

93
In fact, this topic is somewhat tabooed among orientalists, because some acceptable
working models are buried under a great number of amateur etymological theories (such
as Sumero-Hungarian, Sumero-Georgian, Sumero-Turkic and so on). For an example of
scientic approach, cf. the following discussion in Mother Tongue: Diakonoff, 1997;
Bengtson, 1997.
94
Pace, e. g., Diakonoff, 1971, 81 (Hurr. uzi < Sum. UZU meat).
95
It seems that all HU words of Sumerian origin (see, e. g., Diakonoff, 1971, 77 ff.) were
borrowed via Akkadian. Wilhelm and some other authors (e. g., Wilhelm, 2008a, 89, 103)
suppose that Hurr. en=dan ruler was borrowed directly from Sum. EN id., but Akkad.
enu id. as a source of the Hurrian root is an equivalent solution; the third solution is to
derive this term from the proper Hurrian word eni god, Wilhelm, 1989, 11; Wilhelm,
1998c, 121 f. (for the Hurrian sufx of professions =tann see Wilhelm, 1970; Wilhelm,
1998c, 122; Giorgieri, 2000, 204).
96
Hurr. tab=i=ri caster, (copper)smith > Sum. tibira, tabira sculptor, scil. metal
furniture-maker / craftsman working in metal and wood (= Akkad. ururru metal-
worker, esp. coppersmith), see Wilhelm, 1988, 5052 and, e. g., Wilcke, 2010, 10 (cf.
contra Waetzoldt, 1997 and P. Attinger apud Hazenbos, 2005, 135 fn. 6.).
97
The Nakh linguistic group consists of three modern languages: Chechen, Ingush and
Batsbi (Bats). It belongs to the East Caucasian stock of the NCauc. family. Its separation
from the ECauc. trunk is dated to the 1
st
half of the 3
rd
millennium BC; the split of the
Nakh group into Chechen-Ingush and Batsbi subdivisions are dated to the late 1
st
millen-
nium BC (see Kassian, 2010a, 315 for the NCauc. preliminary glottochronological tree).
39. to hear: HU *ha- ~ Nakh *a- to hear.
78. smoke: Hurr. hiuri ~ Nakh *ur smoke.
91. two: Hurr. ini ~ Nakh *i [dir. stem], *ina- [obl. stem] two.
98

The Nakh words listed above are isolated within the NCauc. family, i. e., have
no reliable comparanda in other NCauc. groups
99
and, secondly, there exist
valid candidates for the corresponding Swadesh terms in Proto-NCauc. other
than the Nakh roots: Common NCauc. *=q to hear, Common NCauc.
*}wInV smoke and Common NCauc. *qHw two.

In such a situation several scenarios can be discussed:
A null hypothesis: three HU-Nakh phonetical matches are accidental.
HU belongs to the East Caucasian stock of the NCauc. family and has a
special genealogical relationship with Proto-Nakh. Three aforementioned
isoglosses represent hereby innovations of the HU-Nakh proto-language
after its separation from the East Caucasian trunk. As was demonstrated
above, however, the North Caucasian attribution of HU contradicts both
traditional and formal (lexicostatistical) criteria.
HU and Nakh terms were separately borrowed from a certain, still un-
known, language. Such a hypothesis involves assumptions way beyond
reasonable. Firstly, for the 3
rd
2
nd
millennia BC, one should postulate a
culturally dominant ethnos, whose inuence on neighboring tribes was so
signicant as to make the borrowing of such basic terms possible, on the
one hand, though no actual historical traces of this hypothetical people
can be found, on the other. Secondly, the probability of separate borrow-
ing of the same set of terms into two different languages (HU and Proto-
Nakh) is obviously too low.
Three aforementioned isoglosses are HU loanwords in Proto-Nakh.
The last scenario seems the most productive one,
100
but note that, in this case,
we must presume a very strong inuence of the Hurro-Urartians on the Proto-
Nakh tribe(s). There are no, however, direct contradictions to such an assump-
tion so far, thoughaccording to the typology of linguistic contacts a great
number of HU cultural loanwords in Proto-Nakh is to be expected.
101
Hurro-
UrartianProto-Nakh linguistic interaction remains to be investigated.

98
All of these Nakh terms retain their meaning and basic status in modern Nakh lan-
guages and therefore can be assuredly reconstructed for Proto-Nakh.
99
Especially for Nakh *i / *ina- two, see etymological comments on HU two in 3
above.
100
Cf. similar suspicions also in Smeets, 1989, 269.
101
I think it is no coincidence that authors defending the ECauc. attribution of HU in
many cases suggest only isolated Nakh forms as comparanda for HU data (e. g.,
, 1963, 4151; , 1978).
List of the Hurrian words discussed in 3
al- to drink No. 19
al- to be marvelous(?) No. 5, 22
al=u- to speak No. 71
am-, am=b- to burn No. 12
am- to look No. 72
amm- to arrive No. 16
ani that No. 85, 86
an=ni this No. 85, 86
an=di this No. 85, 86
aw- what?, awi who? No. 96, 97
awattu- far(?) No. 101
aphe louse No. 50
awi, aye face; in front of No. 25, 57
api snake No. 106
ar- to give No. 33
a- to eat(?) No. 23
a- to wash No. 74
ahe skin No. 75
ae fat No. 26
ati, ata woman No. 99
hayani small No. 77
han- to give birth, hni child No. 77
hab- to go forward No. 92
hari road No. 67
ha- to hear No. 39
ha- to oil, ha=ar=i oil No. 26
hazzizzi inner ear No. 21
heyari all No. 1
hil-, hill- to say No. 71
herari biceps / upper arm No. 37
hiuri smoke No. 78
hubr=uhi incense burner No. 78
huri lap, crotch No. 11, 44
i- (to move?) / to be worth(?) No. 92
iya what No. 97
eg=ui pure No. 15
egi inside, interior No. 15
ilziri star(?) No. 80
eradi bird No. 6
erwi, erbi dog No. 18
irde tongue No. 88
i- / u- I No. 42
ee earth No. 22
iena rain No. 65
itt- to go No. 16, 92
edi body No. 64
=kkV= not No. 62
kalgi small(?) No. 77
kari (anatomic term) No. 4
kad- to pronounce No. 71
kig- three No. 91
keri, ker=ai long No. 49
kul- to pronounce No. 71
kuuh moon No. 54
kudu=ni neck No. 58
leli haari ? No. 21
lulahhi a k. of barbarians No. 99
=ma, =mma not No. 62
mar=al- to kill(?) No. 43
meh- to go stand No. 79
nahh- to sit No. 74
nahh=idi, nahh=u=the chair No. 74
neher=ni breast No. 11
nigale thin No. 107
niri good No. 34
nui, nuhi ear No. 21
pahi head No. 38
wahri, wahr=ui good No. 34
pal- to know No. 45
wali, weli ? No. 109
pabanhi, pabahhi mountainous No. 61
pab=ni, pab=a=ni mountain No. 55, 61
far=i- to take the road No. 92
fai mouth No. 56
waduranni dowry / plentiful(?) No. 99
fe- thou No. 87
punhi, puhhi nose No. 61
fur- to see No. 72
furi a look No. 25
alh- to listen No. 39
ali daughter No. 22
alm- to burn(?) No. 2
al=mi ashes No. 2
awali year No. 110
api sun-disk(?) No. 82
awoi great No. 5
a=tti(=lla), ie=-, a=- we No. 95
i- to be small(?), iu small(?) No. 77
i, ihi eye No. 25
ehl- to go in No. 92
imigi sun No. 82
ini two No. 91
iniam ~ inibi thou No. 87
ib=a exhausted No. 20
iwe, iye water No. 57, 94
eri, ir=ni tooth No. 89
ee six No. 91
ed=u- to fatten No. 26
id=u=ri young girl No. 99
uhe new No. 59
ue, ua=nna, ua=lla all No. 1
ukki, ukku one No. 63, 101
ulli charcoal No. 2
umm- to assemble(?) No. 1
ummi all No. 1
u=ni, ummu=ni hand No. 37
ur- to slaughter No. 43
tahe, tae man No. 51
tali tree No. 90
tal=mi, tal=a=mi big No. 5
tame flea No. 50
tamgari merchant No. 50
d
tap-ki-in-na No. 50
tari fire No. 28
taruwa=ni person No. 64
te- to be many, te=a numerous, te=u=na(i) many No. 52
ti- to cry(?) No. 71
time=ri, tima=ri black; dark No. 8
tiw- to pronounce No. 71
tiwe, tiye word; name No. 57, 71
tia heart No. 40
tian very No. 52
tupki ? No. 50
tur=u=hhi male No. 51
=u=, =wa=, =we=, =o=, =u=d= not No. 62
u=ya no! No. 62
ugri- leg of the table No. 31
ul- to eat No. 23
ul=uh- to be near(?) No. 103
umi=ni land No. 22
umbu moon(?) No. 54
un- to come No. 16, 92
uri, ur=ni foot No. 31
ur=mi liver No. 48
urb-, uw- to slaughter No. 43
u- to go No. 92
uzi meat No. 53
ziyan=ni lap, crotch No. 31, 44
zizzi female breast No. 11
zizzuhi cruche No. 80
zugi small No. 77
zur=gi blood No. 9
zuzuhe star(?) No. 80
[]i many No. 52
MUL-hi- star No. 80
List of the Urartian words discussed in 3
al- to enunciate No. 71
al(a)sui great No. 5, 22
am-, am=at- to burn No. 12
an=a=ni this No. 86
ar-, ar=u-, ar=d=u- to give No. 33
arhe ? No. 67
a=a- to go(?) No. 16
ae- (wae-) men No. 51
hari road No. 67
ha=i-, ha=u- to hear No. 39
hurihi ? No. 19
huu meat(?) No. 53
ini this No. 86
i-, ie- / u=ke- I No. 42
esi place No. 22
ei earth(?) No. 22
kauli stone(?) No. 81
iura-, ira- earth No. 22
kuri foot No. 31
ku=u- to reach No. 16
KUR
lulue foreign land No. 99
lutu women No. 99
mi not No. 62
nah=a- to mount (the throne) No. 74
nara ? No. 64
nun=a- to come No. 16
pal- to speak(?) No. 71
bab=a=ni- mountain No. 55
wediani women No. 99
Rusa No. 33
ali year No. 110
salmathi near(?) No. 103
i- (?) No. 65
i=a- to go(?) No. 16
sila wife No. 22
elardi moon No. 54
iwini sun No. 82
ii=ni three(?) No. 91
u- hand No. 37
uhi new No. 59
ui=ni-, uy=a- all No. 1
u=sini 1 No. 63
tarua=ni persons No. 64
ti=a-, ti=u- to say No. 71
ti=ni name No. 57, 71
ti=ni- heart No. 40
ui, =u=ri not No. 62
ul-, ul=a- to go No. 16, 92
urb=u- to slaughter No. 43
urdiali ? No. 77
u-, u=t=a- to go No. 16, 67, 92
uhan=u- to grant No. 33
utipte military campaign No. 67
zag=u- to kill No. 43
zeldi small No. 48, 77
III-e three No. 91
Language name abbreviations
Akkad. Akkadian
Av.-And. (Proto-)Avaro-Andian
Burm. Burmese
Burush. Burushaski
Chin. Chinese
ECauc. (Proto-)East Caucasian
Grk. Ancient Greek
Hitt. Hittite
HU (Proto-)Hurro-Urartian
Hurr. Hurrian
IE (Proto-)Indo-European
Khin. Khinalug
Kott. Kottish
Lezgh. (Proto-)Lezghian
Luw. Luwian (C = Cuneiform, H = Hieroglyphic)
NCauc. (Proto-)North Caucasian
Phoen. Phoenician
Sino-Cauc. (Proto-)Sino-Caucasian
Sem. (Proto-)Semitic
STib. (Proto-)Sino-Tibetan
Sum. Sumerian
Tib. Tibetan
Tsez. (Proto-)Tsezian
Ugar. Ugaritic
Urart. Urartian
WCauc. (Proto-)West Caucasian
Yen. (Proto-)Yenisseian
References
Aayan, H. H., HAB: Hayern armatakan baaran [Armenian Root Dictionary].
Reprinted in 4 vols. Yerevan: University Press, 197179. [The original edi-
tion: 7 vol. Yerevan: University Press, 19261935.]
Andr-Salvini, B. / Salvini, M., 1998: Un nouveau vocabulaire trilingue sum-
rien-akkadien-hourrite de Ras Shamra. SCCNH 9, 331.
1999a: La colonne I du vocabulaire S
a
trilingue RS 94-2939. SMEA
41/ 1, 145148.
1999b: A new trilingual vocabulary from Ras Shamra and the relationship
between Hurrian and Urartian. SCCNH 10, 267275.
2002: The bilingual stele of Rusa I from Movana (West-Azerbaijan, Iran).
SMEA 44/ 1, 566.
Basqet.dbf: Basque etymological database by John Bengtson. Available online
at the Tower of Babel project (http://starling.rinet.ru/cgi-bin/main.cgi?ags
=eygtnnl).
Baxter, W. / Manaster Ramer, A., 2000: Beyond lumping and splitting: Probabi-
listic issues in historical linguistics. In C. Renfrew et al. (eds.): Time Depth
in Historical Linguistics. McDonald Institute for Archaeological Research.
Oxford. Pp. 167188.
Bengtson, J. D., 1997: The riddle of Sumerian: a Dene-Caucasian language?
Mother Tongue 3, 6374.
2008: Linguistic Fossils: Studies in Historical Linguistics and Paleolinguis-
tics. Theophania Publishing.
2008a: The Problem of Isolates II: Burushaski. In Bengtson, 2008, pp. 55
70.
2008b: Materials for a Comparative Grammar of the Dene-Caucasian (Sino-
Caucasian) Languages. Aspects of Comparative Linguistics 3. Moscow:
RSUH Publishers. Pp. 45118.
Bengtson, J. D. / Blaek, V., 2011: On the BurushaskiIndo-European hypothe-
sis by I. aule. Journal of Language Relationship 6, 2563.
Bengtson, J. / Starostin, G., forthc. : The Sino-Caucasian (Dene-Caucasian) hy-
pothesis: state of the art and perspectives. To appear.
Blaek, V., 2010: Hurrian numerals. In
: . . . :
. . 117123.
Buruet.dbf: Burushaski etymological database by S. Starostin (based on H. Ber-
gers data). Available online at the Tower of Babel project (http://starling.
rinet.ru/cgi-bin/main.cgi?ags=eygtnnl).
Campbell, D., 2007: Mood and modality in Hurrian. Ph.D. diss., The University
of Chicago.
2008: Split ergativity in Hurrian. ZA 98, 262294.
Catsanicos, J. 1996: Lapport de la bilingue de Hattusa la lexicologie hourrite.
Amurru 1: Mari, bla et les hourrites. Paris. Pp. 197296.
Caucet.dbf: North Caucasian etymological database by S. Nikolayev and S. Sta-
rostin (edited as NCED). Available online at the Tower of Babel project
(http://starling.rinet.ru/cgi-bin/main.cgi?flags=eygtnnl).
avuolu, R. / Ik, K. / Salvini, M., 2010: New Urartian inscriptions from East
Turkey. Or. NS 79, 3654.
de Martino, S. / Giorgieri, M., 2007: Hurritisch ati (Ehe)frau. AoF 34/ 1
(Festschrift fr I. Wegner), 126148.
Diakonoff, I. M., 1971: Hurrisch und Urartisch. Mnchener Studien zur
Sprachwissenschaft Beiheft N. F. 6. Mnchen.
1997: External connections of the Sumerian language. Mother Tongue 3, 54
62.
Diakonoff, I. M. / Starostin, S. A., 1986: Hurro-Urartian as an Eastern Caucasian
Language. Mnchener Studien zur Sprachwissenschaft Beiheft N. F. 12.
Mnchen.
Dietrich, M. / Mayer, W., 1994: Hurritische Weihrauch-Beschwrungen in uga-
ritischer Alphabetschrift. UF 26, 73112.
1997: Ein hurritisches Totenritual fr Ammitamru III. (KTU 1.125). In B.
Pongratz-Leisten et al. (eds.): Ana ad Labnni l allik: Beitrge zu altorien-
talischen und mittelmeerischen Kulturen: Festschrift fr Wolfgang Rllig.
AOAT 247. Kevelaer/ Neukirchen-Vluyn. Pp. 7989.
2010: Der hurritische Brief des Duratta von Mttnni an Amenotep III.
Text Grammatik Kopie. AOAT 382. Mnster: Ugarit-Verlag.
Dijkstra, M., 1993a: The Ugaritic-Hurrian sacricial hymn to El (RS 24.278 =
KTU 1.128). UF 25, 157162.
1993b: The Akkado-Hurrian bilingual wisdom-text RS 15.010 reconsidered.
UF 25, 163171.
2006: The myth of api the (sea)dragon in the Hurrian tradition. UF
37, 315323.
Dolgopolsky, A. B., 1986: A probabilistic hypothesis concerning the oldest rela-
tionships among the language families of northern Eurasia. In V. V. Shevo-
roshkin / T. L. Markey (eds.): Typology, Relationship, and Time: A Collec-
tion of Papers on Language Change and Relationship by Soviet Linguists.
Ann Arbor (MI): Karoma. Pp. 2750.
Dybo, A. V. / Starostin G. S., 2008: In defense of the comparative method, or
the end of the Vovin controversy. Aspects of Comparative Linguistics 3.
Moscow: RSUH Publishers. Pp. 119258.
Fournet, A. / Bomhard, A. R., 2010: The Indo-European Elements in Hurrian.
La Garenne Colombes / Charleston.
Gell-Mann, M. / Peiros, I. / Starostin, G., 2009: Distant Language Relationship:
The Current Perspective. Journal of Language Relationship 1, 1330.
Giorgieri, M., 1998: Die erste Beschwrung der 8. Tafel des alau-Rituals.
SCCNH 9, 7186.
1999: La lettera in hurrita (EA 24). In M. Liverani : Le lettere di el-Amarna,
vol. 2. Brescia, Paideia Editrice. Pp. 374391.
2000: Schizzo grammaticale della lingua hurrica. In S. de Martino et al.
(eds.): La civilt dei Hurriti. La parola del Passato 55. Napoli, Macchiaroli
Editore. Pp. 171277.
2001: Die hurritische Fassung des Ullikummi-Lieds und ihre hethitische Pa-
rallele. In G. Wilhelm (ed.): Akten des IV. Internationalen Kongresses fr
Hethitologie, Wrzburg, 4.8. Oktober 1999. StBoT 45. Wiesbaden.
2002: Hurritisch tb/v- beschwren. SMEA 44/ 1, 6782.
Girbal, Chr., 1992: Zum hurritischen Vokabular. SMEA 29, 159169.
1994: ummi im Boazky-Hurritischen. AoF 21/1, 171175.
2001: Selbstndige Pronomina der ersten Person Singular im Urartischen.
In: Th. Richter et al. (eds.): Kulturgeschichten: Altorientalistische Studien
fr Volkert Haas zum 65. Geburtstag. Saarbrcken. Pp. 139144.
2004: Notizen zum Urartischen. SMEA 46/ 1, 2528.
2010: Urartisch barzani zelde. AOF 37/ 1, 153158.
Grke, S., 2010: Das Ritual der Atu (CTH 490). Rekonstruktion und Tradition
eines hurritisch-hethitischen Rituals aus Boazky-attua. Brill.
Haas, V., 1993: Hurritologische Miszellen. AoF, 20/ 2, 261268.
1998a: Die hurritischen Ritualtermini in hethitischem Kontext. 2 vols. ChS
1/ 9. Roma.
1998b: Der heilkundige Jger. SMEA 40/ 1, 143150.
Haas, V. / Wegner, I., 1996: Stern, Tag und Segen(?) im Hurritischen. SCCNH
8, 285290.
2004: Das Gegenwortpaar wahr und falsch im Hurritischen. In D. Grod-
dek / S. Rle (eds): arnikzel: Hethitologische Studien zum Gedenken an
Emil Orgetorix Forrer. Dresden. Pp. 339344.
2010: Beitrge zum hurritischen Lexikon: Die hurritischen Verben u- ge-
hen und a- abwaschen, abwischen. In J. Klinger et al. (eds.): Investiga-
tiones Anatolicae: Gedenkschrift fr Erich Neu. StBoT 52. Wiesbaden.
Pp. 97110.
Haspelmath, M., 2008: Loanword typology: Steps toward a systematic cross-
linguistic study of lexical borrowability. In Th. Stolz et al. (eds.): Aspects of
Language Contact. New Theoretical, Methodological and Empirical Findings
with Special Focus on Romancisation Processes. Mouton de Gruyter. Pp. 43
62.
Hazenbos, J., 2005: Hurritisch und Urartisch. In M. P. Streck (ed.): Sprachen
des Alten Orients. Darmstadt. Pp. 135158.
Hoffner, H. A., 1998: Hittite Myths. Ed. by G. M. Beckman. 2
nd
ed. SBLWAW
2. Atlanta.
Holman, E. W. / Wichmann, S. / Brown, C. H. / Velupillai, V. / Mller, A. /
Brown, P. / Bakker, D., 2008: Explorations in automated language classica-
tion. Folia Linguistica 42, 331354.
Hra, I., 2010: Die akkadische Synonymenliste malku = arru. Eine Textedi-
tion mit bersetzung und Kommentar. AOAT 50. Mnster.
Huehnergard, J., 1987/ 2008: Ugaritic Vocabulary in Syllabic Transcription.
HSS 32. 1
st
ed.: Atlanta, 1987; 2
nd
repr. ed.: Winona Lake, 2008, with ad-
denda on pp. 375406.
Ivanov, Vyach. Vs., 1998: Horse symbols and the name of the horse in Hurrian.
In G. Buccellati / M. Kelly-Buccellati (eds.): Urkesh and the Hurrians Stud-
ies in Honor of Lloyd Cotsen. Urkesh/ Mozan Studies 3. Malibu. Pp. 145
166.
Jahukyan, G. B, 1961: The Hayaa language and its relation to the Indo-Euro-
pean languages. ArOr 29, 353405.
Kassian, A., 2010a: Hattic as a Sino-Caucasian language. UF 41, 309447.
2010b: Review of: Fournet / Bomhard, 2010. Journal of Language Relation-
ship 4, 199206.
2011a: More about the theoretical foundations of lexicostatistics (An answer
to A. Fournet & A. Bomhard). Journal of Language Relationship 5, 142145.
2011b: Annotated 50-item wordlist of the basic lexicon of the Ancient Greek
language (the idiolect of Herodotus). Journal of Language Relationship 6,
65100.
Kassian, A. / Starostin, G. / Dybo, A. / Chernov, V., 2010: The Swadesh word-
list. An attempt at semantic specication. Journal of Language Relationship
4, 4689.
Khachikyan, M., 2006: Notes on the Urartian verb. SMEA 48, 143149.
2009: Notes on Hurro-Urartian Phonology and Morphology. AoF 36/ 2, 234
242.
Kilmer, A., 1963: The rst tablet of malku = arru together with its explicit ver-
sion. JAOS 83/ 4, 421446.
Krebernik, M., 2001: Tall Bia / Tuttul. Vol. 2: Die Altorientalischen Schrift-
funde. Saarbrcken.
Lam, J., 2007: The Hurrian Section of the Ugaritic Ritual Text RS 24.643 (KTU
1.148). UF 38, 399413.
Laroche, E., 1955: Textes hourrites. In C. F.-A. Schaeffer (ed.): Le palais royal
dUgarit. Vol. 3: J. Nougayrol. Textes accadiens et hourrites des archives est,
ouest et centrales. Paris. Pp. 327335, pl. CVIIICIX.
Laroche, E., 1968: Documents en langue hourrite provenant de Ras Shamra. In
J. Nougayrol et al. (eds.): Ugaritica 5. Paris. Pp. 447544.
McMahon, A. / McMahon, R., 2005: Language Classication by Numbers.
Oxford.
Neu, E., 1996: Das hurritische Epos der Freilassung I: Untersuchungen zu ei-
nem hurritisch-hethitischen Textensemble aus Hattua. StBoT 32. Wiesba-
den.
2003: Die hurritischen Pronomina der hurritisch-hethitischen Bilingue aus
Hattua. In G. Beckman et al. (eds.): Hittite Studies in Honor of Harry A.
Hoffner Jr. on the Occasion of His 65
th
Birthday. Eisenbrauns. Pp. 297307.
Nichols, J. / Warnow, T., 2008: Tutorial on Computational Linguistic Phylog-
eny. Language and Linguistics Compass 2/ 5, 760820.
Nikolaev, S., 1991: Sino-Caucasian Languages in America. Preliminary report.
In: Dene-Sino-Caucasian Languages: Materials from the First International
Interdisciplinary Symposium on Language and Prehistory, Ann Arbor, 812
November 1988. Bochum. P. 4266.
Patri, S., 2009: Aspects de la position gntique du hourrite. In G. Wilhlem
(ed.): General Studies and Excavations at Nuzi, 11/ 2: Studies in Honor of
David I. Owen. SCCNH 18. Bethesda: University Press of Maryland, CDL
Press. Pp. 341354.
Peiros, I. I. / Starostin, S. A., 1996: A Comparative Vocabulary of Five Si-
no-Tibetan Languages. 6 vols. Melbourne.
Prechel, D. / Richter, Th., 2001: Abrakadabra oder Althurritisch: Betrachtungen
zu einigen altbabylonischen Beschwrungstexten. In Th. Richter et al. (eds.):
Kulturgeschichten: Altorientalistische Studien fr Volkert Haas zum 65. Ge-
burtstag. Saarbrcken. Pp. 333371.
Richter, Th., 2005: Kleine Beitrge zum hurritischen Wrterbuch. AoF 32/ 1,
2344.
2007: Ergnzungen zum hurritischen Wrterbuch I. AoF 34/ 1, 78115.
Ringe, D., 1999: How hard is it to match CVC-roots? Transactions of the Philo-
logical Society 97/ 2, 213244.
Rseler, I., 1999: Hurritologische Miszellen. SCCNH 10, 393400.
Salvini, M., 1998a: The earliest evidences of the Hurrians before the formation
of the reign of Mittanni. In G. Buccellati / M. Kelly-Buccellati (eds.): Urkesh
and the Hurrians Studies in Honor of Lloyd Cotsen. Urkesh/ Mozan Studies
3. Malibu. P. 99115.
1998b: Eine urartische Felsinschrift in der Region Nachievan. ZA 88, 94
99.
2002: Una stele di Rusa III Erimenai dalla zona di Van. SMEA 44/ 1, 115
143.
2007: Rusa I. II. III. RlA 11, 20062008, 464466.
Salvini, M. / Wegner, I., 2004: Die mythologischen Texte. ChS 1/ 6. Roma.
Sccet.dbf: Sino-Caucasian etymological database by S. Starostin. Available on-
line at the Tower of Babel project (http://starling.rinet.ru/cgi-bin/main.cgi?
flags=eygtnnl).
Siewierska, A., 2004: Person. Cambridge.
Smeets, R., 1989: On Hurro-Urartian as an East Caucasian language [a review
of Diakonoff / Starostin, 1986]. BiOr 46/ 34, 259279.
Starostin, G. S., 2008: Making a Comparative Linguist out of your Computer:
Problems and Achievements. Presentation at the Santa Fe Institute, August
12, 2008. Available online at the Tower of Babel project (http://starling.rinet.
ru/Texts/computer.pdf).
2010a: Preliminary lexicostatistics as a basis for language classication: A
new approach. Journal of Language Relationship, No. 3. P. 79116.
2010b: Dene-Yeniseian and Dene-Caucasian: Pronouns and other thoughts.
In Working Papers in Athabaskan Languages, 2009: Alaska Native Lan-
guage Center Working Papers 8. Fairbanks, Alaska: ANLC. Pp. 107117.
Stibet.dbf: Sino-Tibetan etymological database by S. Starostin (= Peiros / Sta-
rostin, 1996, but with serious improvement). Available online at the Tower of
Babel project (http://starling.rinet.ru/cgi-bin/main.cgi?flags= eygtnnl).
Strau, R., 2001: Eine Rezeptur und Beschwrung fr die Zubereitung von
Weihwasser in dem Ritual CTH 471. In Th. Richter et al. (eds.): Kulturge-
schichten: Altorientalistische Studien fr Volkert Haas zum 65. Geburtstag.
Saarbrcken. Pp. 405416.
Thomason, S. G. / Kaufman T., 1988: Language Contact, Creolization, and Ge-
netic Linguistics. University of California Press.
Tower of Babel project Etymological database project The Tower of Babel,
headed by S. A. Starostin. In afliation with the Evolution of Human Lan-
guages (EHL) project at the Santa Fe Institute. Available online: http://
starling.rinet.ru/
Trmouille, M.-C., 2005: Texte verschiedenen Inhalts. ChS 1/ 8. Roma.
Turchin, P. / Peiros, I. / Gell-Mann, M., 2010: Analyzing genetic connections
between languages by matching consonant classes. Journal of Language
Relationship 3, 117126.
nal, A., 1998: Hittite and Hurrian cuneiform tablets from Ortaky (orum),
Central Turkey. Istanbul.
van Gessel, B., 1998: The Onomasticon of the Hittite Pantheon. HdO I/ 33. Lei-
den, 19982001.
Waetzoldt, H., 1997: Die Berufsbezeichnung tibira. NABU 1997, N
o
96.
Wegner, I., 1981: Gestalt und Kult der Itar-awuka in Kleinasien. AOAT 36.
Kevelaer/ Neukirchen-Vluyn.
1988: Grammatikalische und lexikalische Untersuchungen hurritischer Be-
schwrungsformeln aus Boazky. In V. Haas (ed.): Hurriter und Hurritisch.
Xenia 21. Konstanz. P. 145155.
1992: Die selbstndigen personalpronomina des Hurritischen. SMEA 29,
227237.
1994: Hurritische Verba dicendi mit einfacher und doppelter Absolutiv-Rek-
tion. AoF 21/ 1, 161170.
1995a: Die hurritischen Krperteilbezeichnungen. ZA 85, 116126.
1995b: Der Name der a(w)uka. SCCNH 7, 117120.
2004: Hurritische Opferlisten aus hethitischen Festbeschreibungen. T. 3: Das
Glossar. ChS 1/ 3-3. Roma.
2007: Hurritisch. Eine Einfhrung. 2
nd
rev. ed. Wiesbaden.
2010: Das hurritische Fragepronomen aunni (a--u(n)-ni) = hethitisch kuit.
In J. C. Fincke (ed.): Festschrift fr Gernot Wilhelm anllich seines 65. Ge-
burtstages am 28 Januar, 2010. ISLET. Pp. 403410.
Werner, H., 2002: Vergleichendes Wrterbuch der Jenissej-Sprachen. 3 vols.
Wiesbaden.
Wichmann, S. / Holman, E. W. / Mller, A. / Velupillai, V. / List, J.-M. / Bel-
yaev O. / Urban, M. / Bakker, D., 2010: Glottochronology as a heuristic for
genealogical language relationships. Journal of Quantitative Linguistics,
17/ 4, 303316.
Wilcke, C., 2010: Sumerian: What we know and what we want to know. In
L. Kogan et al. (eds.): Proceedings of the 53
e
Rencontre Assyriologique In-
ternationale, vol. 1, part 1. Babel und Bibel 4. Winona Lake. Pp. 576.
Wilhelm, G., 1970: Ta/erdennu, ta/urtannu, ta/urtnu. UF 2, 277282.
19721975: ubrui. RlA 4, 19721975, p. 478.
1981: Der Komitativ des Urartischen. SMEA 22, 133136.
1984: Die Inschrift auf der Statue der Tatu-Hepa und die hurritischen deikti-
schen Pronomina. SMEA 24, 215222.
1988: Gedanken zur Frhgeschichte der Hurriter und zum hurritisch-urarti-
schen Sprachvergleich. In V. Haas (ed.): Hurriter und Hurritisch. Xenia 21.
Konstanz. Pp. 4367.
1989: The Hurrians. Warminster.
1991a: A Hurrian Letter from Tell Brak. Iraq 53, 159168.
1991b: Zur hurritischen Gebetsliteratur. In D. R. Daniels et al. (eds.): Ernten,
was man st : Festschrift fr Klaus Koch zu seinem 65. Geburtstag. Neukir-
chener Verlag. Pp. 3747.
1992: Hurritische Lexikographie und Grammatik. Die hurritisch-hethitische
Bilingue aus Boazky. OrNS 61, 122141.
1992b: EA 24. A letter in Hurrian about marriage and friendship. In: The
Amarna Letters. Ed. and transl. by W. L. Moran. Baltimore/ London. P. 63
71.
1993: Zur Grammatik und zum Lexikon des Hurritischen. ZA 83, 99118.
1998a: Hurr. inussi Scheuklappe? SCCNH 9, 173176.
1998b: Zur Sufxaufnahme beim Instrumental. SCCNH 9, 177180.
1998c: Die Inschrift des Tiatal von Urke. In G. Buccellati / M. Kelly-Buc-
cellati (eds.): Urkesh and the Hurrians Studies in Honor of Lloyd Cotsen. Ur-
kesh/ Mozan Studies 3. Malibu. Pp. 117143.
2003: Bemerkungen zu der akkadisch-hurritischen Bilingue aus Ugarit. In
W. Sallaberger et al. (eds.): Literatur, Politik und Recht in Mesopotamien:
Festschrift fr Claus Wilcke. OBC 14. Wiesbaden. Pp. 341345.
2006: Der Brief Turattas von Mittani an Amenophis III. in hurritischer
Sprache (EA 24). In B. Janowski / G. Wilhelm (eds.): Texte der Umwelt des
Alten Testaments, Neue Folge 3: Briefe. Gtersloh. Pp. 180190.
2008a: Hurrian. In R. D. Woodard (ed.): The Ancient Languages of Asia Mi-
nor. Cambridge. Pp. 81104.
2008b: Urartian. In R. D. Woodard (ed.): The Ancient Languages of Asia
Minor. Cambridge. Pp. 105123.
Yakubovich, I., 2010 [2011]: Morphological Negation in Urartian. In A. Kosyan
et al. (eds.): Urartu and its Neighbors. Festschrift in honor of Nicolay Harut-
yunyan in occasion of his 90
th
birthday. Aramazd, Armenian Journal of Near
Eastern Studies 5. Yerevan. Pp. 141165.
Yenet.dbf: Yenisseian etymological database by S. Starostin (= ,
1995 and Werner, 2002, with additions and corrections). Available online at
the Tower of Babel project (http://starling.rinet.ru/cgi-bin/main.cgi?flags=
eygtnnl).
, . . / , . ., 1954:
- . In IV (XI)
. : .
.
, . . / , . ., 2005: - -
[Comparative Linguistics]. 2- . .
, . ., 1963: -

. .
, . ., 1967: , -
. .
, . ., 1963: . .
1978: - . In -
, . 3. . . 2538.
, ., 2011: -
[Some considerations on
Hurrian-North Caucasian lexical matches]. In . . . (.).
XV. -
. . 252258.
, . ., 1982/ 2007: -
. In , 2007. . 147246. [First publ.: -
. , 1982. . 144237]
1995: . In
(Studia Ketica), . 4. . . 176315.
1995/ 2007: . In: ,
2007. . 629632. [First publ.: , 2 (213), 1995.
. 133136]
1998/ 2007: Hurro-Caucasica. In: , 2007. . 745751. [First publ.:
: 70- . -
, 1998. . 9099]
2007: [Works in Linguistics]. .
2007a: [Dening the Stabil-
ity of Basic Lexicon]. In , 2007. . 827839.
, . ., 1985: . .
2010a: - . In :
. : , . . 118126.
2010b: . In :
. : , . . 126149.
2010c: . In : -
. : , . . 149168.
Abbreviations
AHw W. von Soden. Akkadisches Handwrterbuch. Wiesbaden, 1965
1981.
CAD The Assyrian Dictionary of the Oriental Institute of the University
of Chicago. Chicago, 1956.
CdTU M. Salvini. Corpus dei testi urartei. Vol. 13. Roma, 2008.
CHD The Hittite Dictionary of the Oriental Institute of the University of
Chicago. Chicago, 1980.
ePSD Electronic Pennsylvania Sumerian Dictionary Project (http://psd.
museum.upenn.edu/epsd/index.html).
GLH E. Laroche. Glossaire de la langue hourrite. Paris, 1980.
HEG J. Tischler. Hethitisches etymologisches Glossar. Innsbruck, 1977.
HW
2
J. Friedrich / A. Kammenhuber. Hethitisches Wrterbuch. 2. Aufl.
Heidelberg 1975.
LHL 1 S. de Martino, M. Giorgieri. Literatur zum hurritischen Lexikon.
Bd 1: A. LoGisma editore, 2008.
NCED S. L. Nikolayev, S. A. Starostin. A North Caucasian Etymological
Dictionary. Moscow, 1994 [reprinted: 3 vols. Ann Arbor: Caravan
Books, 2007]. Available online at the Tower of Babel project (http:
//starling.rinet.ru/cgi-bin/main.cgi?flags=eygtnnl) as Caucet.dbf.
SCC S. A. Starostin. Sino-Caucasian. Unnished MS, the middle of the
2000s. Available online at the Tower of Babel project (http://
starling.rinet.ru/Texts/scc.pdf).
WALS M. Haspelmath et al. The World Atlas of Language Structures. Ox-
ford, 2005.

. . . .
, 2003.

Kassian Hurro-Urart Lexicostat 2011

Transféré par

Informations du document

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

Kassian Hurro-Urart Lexicostat 2011

Transféré par

Droits d'auteur :

Formats disponibles

Sonderdruck aus

- to come, Lak na- to go (vel sim.)).

H ear > Chin. *nh,

wV he-goat, which is unprovable (note that there is not enough data

wV as a Proto-NCauc. root, in fact it is an Av-And.

w nape of the neck (vel sim.) > NCauc. *dHq

(thus already Diakonoff/ Starostin, 1986, 38) > Nakh

s, Lak ca, Dargwa ca, Lezgh. sa, Khin. sa,

spirit, Yen. *ir

Vous aimerez peut-être aussi