saqarTvelos mecnierebaTa erovnuli akademiis moambe,,,,, 171717171755555,,,,, #22222, 200, 200, 200, 200, 20077777
BULLETIN OF THE GEORGIAN NATIONAL ACADEMY OF SCIENCES, 175, #22222, 200, 200, 200, 200, 20077777
© 2007 Bull. Georg. Natl. Acad. Sci.
Archaeology
Urkesh and the Question of the Hurrian Homeland
Giorgio Buccellati*
, Marilyn Kelly-Buccellati**
* Director, Mesopotamian Lab, Cotsen Institute of Archeology, UCLA, Los Angeles, Ca 90024, USA;
Professor Emeritus, Department of Near Eastern Languages and Cultures, UCLA;
Professor Emeritus, Department of History, UCLA
** Cotsen Institute of Archeology, UCLA, Los Angeles, Ca 90024, USA
(Presented by Academy Member Th.V. Gamkrelidze)
ARSTRACT. Sometimes, archaeological excavations at a given site open windows onto much wider views than
intra-site relationships might suggest. Such is very much the case with our work at Tell Mozan, a large tell in
northeastern Syria. When starting excavations there in1984, we suggested that this may be the site of ancient
Urkesh. This hypothesis was conclusively proven in1995 when the reading of the legends on a number of seal
impressions showed that we were digging the palace of a heretofore unknown “ruler (endan) of Urkesh,” who bore
the Hurrian name of Tupkish. Several other names of members of the royal court were also Hurrian.
Onomastics was but the first of many distinguishing features that convincingly gave witness to the nature of the
ethnic identity of this ancient city. Much of the material culture that came to light as a result of our excavations
provides a constellation of ethnic identifiers such as is not to be found at other sites. Two elements are of particular
significance in this perspective: the first is that the Urkesh hinterland extends effectively far into the northern
plateau, the second is that we are led to push further and further back in time the date for the urban beginnings of
Urkesh. These various factors allow us to propose what we believe to be a plausible hypothesis about the earliest
homeland of the Hurrians, a topic which is close to the interests of this journal and of its editor. This article will be
in the nature of an essay, in which we will summarize some of the points we have raised elsewhere,
1
with little space
for documentation or bibliographical references. © 2007 Bull. Georg. Natl. Acad. Sci.
Key words: Tell Mozan, Urkesh, homeland of the Hurrians.
1. Excavations at Tell Mozan,
ancient Urkesh
The period of major expansion of the ancient city, as
we know it today, is the third millennium: to this time
date the major public buildings excavated so far, and it is
also in this period that we can say the city reached its
maximum extension, between 130 and 150 hectares. In
the second millennium, the city retrenched to the central
high mound (about 30 hectares), and retained a modicum
of importance as a religious shrine. It was abandoned
about 1300 B.C., when the Assyrians finally took over
the region, but had no use for a religious center that had
so consistently retained it specific Hurrian character. It
was never re-occupied again: this meant that the original
coherence of the city, in its urbanistic layout, was better
preserved than at most other sites of the same time period.
The most recent excavations have brought to light
material from the fourth millennium, about which more
will be said below.
Three major third millennium public structures are
currently under excavation. (1) Of the Palace of Tupkish
1
Here, we will only refer to some of our publications where the documentation from Urkesh, pertaining to these problems,
is treated in more detail: refer to online versions
142 Giorgio Buccellati, Marilyn Kelly-Buccellati
Bull. Georg. Natl. Acad. Sci. vol. 175, no 2, 2007
(Fig. 3) we have completely exposed the service wing,
and have started work on the formal wing (on top of
which there is a thick deposit of the second millennium
settlement, which grew above the abandoned palace).
The most important finds have been sealings which had
been used to seal containers and doors. Many of them
were inscribed: they have given us the name of the city,
of the king who built it (Tupkish) and of one of his successors
(Ishar-kinum), of one and presumably two
queens, and of several courtiers. The style of the seals is
very distinctive, and different in several ways from the
standard glyptics known from the Sumero-Akkadian
south.
(2) A deep underground structure (Fig. 4), next to
the Palace, would not even have been visible except for
a small covered entrance, but it was monumental in character
and of great ideological significance. While no written
document has been found associated with it, the
nature of the finds makes it certain that this corresponds
to what is known in Hurrian as âbi – a sacred area where
the spirits of the Netherworld were summoned as part of
complex rituals that are well preserved in later Hurrian
texts preserved in the Hittite archives. It was presumably
used primarily by the king and the queen. While the structure,
in its simplest form, is a deep circular shaft going
back to the early third, and possibly the fourth millennium,
it was at about 2300 B.C. that it came to be covered
by a corbelled vault and expanded by means of an antechamber.
This was also the time when the Palace of
Tupkish was built, in a position immediately adjacent to
the underground structure.
(3) Another religious structure pre-dated the Palace
of Tupkish – a Temple sitting atop a Temple Terrace (Fig.
5) that towered high above the Palace. This was clearly
Fig. 2. Urkesh Monumental Urban Complex: reconstruction (P. Pesaresi)
Fig. 1. Urkesh Monumental Urban Complex: view of excavations
Urkesh and the Question of the Hurrian Homeland 143
Bull. Georg. Natl. Acad. Sci. vol. 175, no 2, 2007
the main temple of Urkesh, and we have good reasons to
believe that it was dedicated to the main God of the
Hurrian pantheon, Kumarbi. For reasons that are detailed
below (section ý4), it is very likely that an earlier version
of this Temple Terrace dates back to the fourth millennium.
The top of the Terrace (hence, the floor of the
Temple) is about 25 meters above the ancient plain level,
and about 12 meters above the level of the palace. It is
linked with the Palace through a wide Plaza, thereby creating
a vast urban complex which, from one end of the
Palace to the end of the Temple Terrace is more than 200
meters wide (Figs. 1-2).
2. Connections with the north
Against this background, we may reflect on questions
we have had in mind since the beginning of our
excavations concerning the existence of relations with
the north and the nature of such relations. To the north
in eastern Anatolia, Armenia and Georgia for most of the
third millennium the material culture is characterized by
burnished or polished black and red ceramics, hearths
and andirons decorated with anthropomorphic or geometric
decorations and a settlement pattern centered for
the most part around villages and small towns. This culture,
called the Early Transcaucasian culture, is well documented
from Georgia to eastern Anatolia and into northwestern
Syria with extensions north of the Caucasus and
to the southwest into Palestine. Many cultural patterns
Fig. 3. View of the Palace of Tupkish [with inset showing relationship to city (about 2300 B.C.)]
Fig. 4. View of the âbi, the necromantic underground structure
(about 2300 B.C., with earlier levels going back much
earlier
144 Giorgio Buccellati, Marilyn Kelly-Buccellati
Bull. Georg. Natl. Acad. Sci. vol. 175, no 2, 2007
link Urkesh with the north, and we will focus on these
first, and then (in section 3) on those traits which are
more distinctly indicative of ethnic affiliation.
One characteristic connecting Urkesh with Anatolia
is the presence in Urkesh seal iconography of elements
that appear later in the seal corpus of Kultepe, level 2.
Prominent among these is the depiction of a bull standing
on an altar (Fig. 6). Not only the iconography of our
seal impressions, but also the style, anticipate these later
impressions. Other elements found both in Urkesh and
later in Kultepe include a scene depicting the slaying of
a reversed bull using the same type of long triangular
knife. This is seen in a recently found cylinder seal in
Urkesh (Fig. 7) where a reversed bull has just been sacrificed
by two priests; the head of the bull rests at the
base of a palm column topped with a necked jar. In the
later Hurrian-Hittite texts the usual offering to the Weather
God was a sacrificed bull. In the seal impressions from
Kultepe, level 2, the bull cult is prominent as is the
Weather God. One seal impression from Kultepe 2 depicts
a Storm-god threatening a reversed bull; he steps
on the head of the bull while holding one of its rear legs
with his right hand. In his left hand he holds a triangular
shaped knife strikingly similar in shape to the one used
to slay the bull in the Urkesh seal.
Another aspect that is distinctive of Urkesh and
seems to link it with the north pertains to fashion, in
particular two forms of headdress. The first is the one
Fig. 5. A night view of the main staircase of the Temple Terrace (about 2400 B.C., but remained in use until
about 1300 B.C.g)
Fig. 6. Composite drawing from a number of seal impressions of a scene with a bull standing on a platform.
Urkesh and the Question of the Hurrian Homeland 145
Bull. Georg. Natl. Acad. Sci. vol. 175, no 2, 2007
shown on the two priests depicted on the Urkesh seal
with the bull sacrifice: they wear distinctive head coverings
that come under their chin, of a type found also on
a seal of Tupkish, where the figures are carrying offerings.
A second type of headdress is a basque-like hat
that is found in several seal impressions (Fig. 8).
More direct connections with the north can be found
through EarlyTranscaucasian sherds found in the Urkesh
excavations. The black and red-brown burnished pottery
so characteristic of the north has only been found in small
amounts in Urkesh. It has been discovered both in the
main temple BA connected with strata dating to the mid
third millennium and in the later third millennium palace of
Tupkish. From this distribution it appears that there was a
continuity of contact even if the evidence is scarce. That
they are few is clear but their importance must be connected
with the fact that they show some importation of
either these ceramics themselves or what they contained.
But the trade with the north must not have been in
foodstuffs or other small items best transported in ceFig.
7. Seal with ritual scene showing a bull that has just been slain and whose head lies at the
base of a palm column.
Fig. 8. Seal impression with two figures wearing a basque-like hat
146 Giorgio Buccellati, Marilyn Kelly-Buccellati
Bull. Georg. Natl. Acad. Sci. vol. 175, no 2, 2007
ramic vessels, but rather the city of Urkesh, being the
closest city to the main trade route to the north through
the Mardin Pass, must have benefited from the trade in
the raw materials the EarlyTranscaucasians controlled.
Of major importance must have been metals but could
also have included certain types of wood. From our environmental
studies we know that in the third millennium
the area around Urkesh not only produced grain but also
that the area was sheltered by a savannah type of light
tree coverage. Other imported raw materials must have
included certain types of stones for building. We know
from the animal bone inventory and from the seal iconography
that wild mountain animals were also present
in Urkesh; whether they were hunted by the elite of
Urkesh or imported cannot be determined.
More important for connections between Urkesh and
the north is the presence of andirons in our excavations.
Both permanently installed and movable andirons in the
Early Transcaucasian culture are horseshoe shaped and
decorated on the two end portions with incised and applied
decoration, usually anthropomorphic but also with
geometric elements. The Urkesh examples have both geometric
and anthropomorphic decorations and date mainly
to the Khabur period (ca 1800 B.C.) (Fig. 9). At Urkesh
andirons have been found permanently installed in contexts
of both domestic and funerary nature. In addition
we have found at Urkesh small portable andirons with
the same type of decoration. For an interpretation of these
objects we can turn to the letters from Zimri-Lim, king of
Mari, to the rulers he had installed in Urkesh, i.e., Terru
and Haziran. The content of the letters shows how disturbing
to the ruler the conditions in Urkesh were. They
indicate clearly that the local people in Urkesh did not
accept the overlordship of these foreign Amorite rulers
appointed by him. Given this chaotic political situation,
it may very well be that the local Hurrian urban population,
wanted to reaffirm their ethnic identity and their
connections to the rural population in the mountains to
the north and east through the use of specifically Hurrian
cultural artifacts. In this case these artifacts are also
closely linked to the Early Transcaucasian culture. Were
the people identified with the Early TransCaucasian culture
Hurrians? Our andirons used in the Hurrian city of
Urkesh are the clearest evidence thus far of the link between
the two.
3. The Hurrian identity of Urkesh
Two important questions underlie our understanding
of ethnicity, most especially when dealing with archaeological
evidence. The first is historical: is the notion
of an ethnos, as different from other types of social
groupings, a reality that plays a role in the dynamics of
historical development? The second question assumes a
positive answer to the first and is historiographical: what
criteria can we use to recognize an ethnic group?
Semiotics can help us in dealing with both questions.
Symbolic notation is of paramount importance in
providing coherence and integration to an ethnic group.
There is no single element that in and of itself is ethnic.
Rather, any trait (from language, material culture, ideology)
can become a powerful ethnic marker to the extent
that it proclaims distinctiveness. The task of historiography
(broadly understood, and therefore referring also to
the archaeological record) is to recognize these markers,
and to show whether consciousness of their existence
was an operative factor in holding the group together.
Semiotics helps us to recognize minimal pairs with contrastive
values, to see what may be charged with ethnical
valence because it is opposed to counter-ethnical
traits (traits that distinguish other ethnic groups).
Early in the history of the research on the Early
Transcaucasian culture, the ethnic identity of the population
had been tentatively connected with the Hurrians
on the basis of second millennium written documentation
which showed them to be present more or less in
the same geographical area. While there continues to be
no written evidence for the vast spread of this culture in
the early periods, and therefore no direct evidence of
their ethnic affiliations, we think that our excavated evidence
from ancient Urkesh provides sufficient evidence
to show that at least this city was indeed Hurrian, for
here we have a cluster of traits that can be linked specifically
with this ethnic group.
The linguistic evidence is the most apparent. From
Urkesh come the most ancient texts written in the language
that came later to be known as Hurrian, the Tishatal
inscription in the first place, but also, in a minor key,
the legends of the seals of Tupkish and his court. What
is important about the Tish-atal inscription in particular
Fig. 9 Large andiron typical of the Khabur period in Urkesh
(A11.34)
Urkesh and the Question of the Hurrian Homeland 147
Bull. Georg. Natl. Acad. Sci. vol. 175, no 2, 2007
are two facts. The first is that the text is entirely written
in Hurrian (in contrast, e.g., with the texts from Ebla which
are written in Eblaite only to a very limited extent). The
second is that it is a foundation inscription for a temple
built by a king, wherein the choice of Hurrian holds special
meaning as an affirmation of political pride and dis-
tinctiveness.
An important aspect of the linguistic evidence is the
use of the Hurrian term endan for the king. It occurs
only in Urkesh, and is the only known case of a non
Sumero/Akkadian title in third millennium Syria. Since it
is documented at the height of the political expansion of
the Akkadian empire, it reflects an emphatic display of
self-assertion that contrasts not only with the prevailing
fashion of the time, but also with the imperial policy of
the conquering southern power. While possibly at home
in other Hurrian cities of the third millennium, the title is
so far attested only for Urkesh.
Along with the use of the language come the
onomastics. The evidence of proper names can be used
improperly, for two major reasons: we must not automatically
equate the linguistic nature of the name with the
ethnic affiliation of the name-bearer, and the isolated evidence
of one or even a few names must not incontrovertibly
be taken to reflect demographic density for a
given ethnic group. A way out of the impasse is to consider
the distribution of the names: our assumption is
that they can be taken to reflect ethnicity if they affect
distinct classes. That is the situation at Urkesh where,
even though the available evidence is still limited, the
kings (with one exception) and the courtiers have Hurrian
names, including the nurse and the cook of the queen,
while the queen herself has an Akkadian name, presumably
reflecting a dynastic marriage.
The architectural and stratigraphic evidence pertaining
to the underground passage to the Netherworld,
known as âbi in Hurrian, points to cults and rituals that
are specifically linked to the Hurrian sphere. Even more
importantly, these cults are distinctly non-Mesopotamian,
both in tenor and for what they tell us of the relative
conceptions of the Netherworld. Whether or not there is
merit in the suggestion that the structural re-arrangement
at the time of the Palace of Tupkish reflected an adaptation
on the part of the Hurrian milieu to Akkadian sensitivities
(see above, ý1), the fact remains that there are
here two sharply contrasting religious perceptions, that
go well beyond the mere terminological difference in divine
names or cultic objects.
In Hurrian mythology Urkesh occupies a central role.
It is preserved in Hurrian texts from Hittite times that are
much later than the heyday of Urkesh, but two considerations
are in order. First, by the time they were written,
Urkesh had lost all its importance: still specifically Hurrian
in nature, it was reduced to the status of a small and
marginal shrine with a limited service support system. It
would therefore be curious to assume that it was “invented”
at that time as the seat of the father of the Hurrian
pantheon. Second, the myths retain a very distinctive
archaic tone, which may reasonably be assumed to reflect
a much earlier tradition, one that goes back to the
formative periods of the mythology itself.
It is interesting to note that the Mesopotamian ideological
landscapes in the third millennium block out altogether
the northern region. While ample mental space is
given to the East, the South and the West (see Fig. 10),
it is as though the North did not exist, even though it is
geographically closer than any of the other landscapes.
We interpret this to mean that the North had a strong
Fig. 10. Ideological landscapes of Syro-Mesopotamia in the third millennium
148 Giorgio Buccellati, Marilyn Kelly-Buccellati
Bull. Georg. Natl. Acad. Sci. vol. 175, no 2, 2007
cultural consistency of its own, which is precisely reflected
in the rich Hurrian mythological world. There was
no opportunity for the Sumerians to appropriate an empty
landscape, because it was not empty. And there was no
opportunity to adopt any of the Hurrian mythology because
it was too foreign and well developed. This, at
least, is our hypothesis. It seems further supported by
the most recent discoveries at Tell Mozan, which seem
to push back into the fourth millennium the full blossoming
of Urkesh as a major urban and religious center.
4. Evidence for a fourth millennium
Hurrian Urkesh?
The 2006 excavations have brought to light limited
but important evidence relating to the fourth millennium.
We will first describe briefly the typology of the finds,
then we will discuss their stratigraphic situation and finally
we will draw the conclusions regarding their possible
significance for the question of ethnic affiliation.
4.1. Ceramics and seal impressions
The fourth millennium ceramics from the temple terrace
and the exterior base of the revetment wall are typical
Late Chalcolithic 3 and 4 local production. The major
part are coarse wares with large pebble inclusions. Shapes
include open bowls, hammer head bowls (Fig.11) and
casseroles. Fine ware sherds are much fewer in number
but do include a characteristic small carinated bowl. Only
one small body of a beveled rim bowl has been found
together with this material.
Connections with the south in this time period are
evident from a number of cylinder seal impressions of an
Uruk 3 style. A series of seal impressions show intertwined
ribbons or snakes (Fig. 12a) One set of impressions
were made from a seal with a complex iconographic
motif of what appears to be two offering bearers carrying
standards on a vertical pole (Fig. 12b). The top of only
one of these standards is preserved in our rollings and it
is semi-circular with a crosshatched central motif. This
type of complex motif is not common even in the south
of Mesopotamia or Susa where the Uruk culture is best
attested.
4.2. Stratigraphy
Some Late Chalcolithic ceramics had already been
found in 2005 in two places: immediately below the escarpment
that leads from the Plaza to the revetment wall,
in unit J1, and below the glacis just inside the revetment
wall, in unit J3. In each case, the exposure was minimal
and so we assumed, conservatively, that these finds
might represent isolated pockets without any particular
contextual significance. In 2006 we found more Late
Chalcolithic material, in similar contexts: seal impressions
in a shallow pit just below the surface of the glacis in J3,
and ceramics in a sounding to the east of the stairway, in
J4 – this too, in a setting that seems to match the layers
below the glacis in J3 where the seals were found.
The wide spatial distribution of the finds and their
coherence suggest that they are not isolated pockets.
The consistency of the matrix is also an important factor:
wherever we probed below the surface of the third millennium
installations (escarpment and glacis), the same
type of material has appeared. All of this indicates, in our
view, that the third millennium Terrace was built on a preexisting
structure that was not very dissimilar in size.
This aspect is significant. The area covered is quite wide
(from J1 to J4 there are about 40 ms), and the elevation
above the plain level is quite high (at least 20 ms). It is
Fig. 11 Hammer Rim bowls from Late Chalcolithic 3/4 strata.
Urkesh and the Question of the Hurrian Homeland 149
Bull. Georg. Natl. Acad. Sci. vol. 175, no 2, 2007
plausible to infer that such a high and wide structure
was very similar, in its basic traits, to the high and wide
structure of the third millennium – i.e., a Temple Terrace.
At the base of the staircase, there is evidence of an
earlier staircase. The associated material is not sufficient
to provide a good date for it, but it seems plausible to
assume that it goes with the fourth millennium strata
found below the third millennium escarpment and glacis.
The nature of the structure that hides beneath the
third millennium monumental Temple Terrace can hardly
be different from that of later Temple Terrace. If so, already
by the middle of the fourth millennium we would
have a towering structure, that would speak to both the
fully developed urban character of Urkesh and the religious
prestige associated with it as the seat of the father
of the gods. Structural and stratigraphic similarities between
the Urkesh situation and the organization of the
Tell Chuera “Steinbauten,” conceived as a well planned
sacred space, suggest that a similar situation occurred at
both sites.
4.3. Significance for the ethnic
question
A Temple Terrace such as the one we see for the
third millennium and can infer for the fourth implies a
continuity of religious traditions. In other words, it is
reasonable to assume that the earlier monumental structure
served the same purpose as the later. And this means
that we can project, into a much earlier period than was
ever suspected, a well established urban Hurrian pres-
ence.
This argument is fragile on two counts. First, we
arrive at our conclusion on the basis of an inferential
chain that, while plausible, remains just that, inferential.
We feel that the nexuses are sufficiently clear to warrant
our conclusion, but it does remain hypothetical. The second
caveat is that the third millennium Terrace itself is
not demonstrably Hurrian. This, too, is based on a set of
inferences that we have developed elsewhere: we assume
that the Temple Terrace is nothing less than the very
seat of Kumarbi described in the later myths.
But for all the fragility of the full argument, the facts
that are certain are sufficient to give us pause when we
hear the communis opinion repeat that the Hurrians are
late comers in the Fertile Crescent. There is no question
that (1) from the mid third millennium on Urkesh is, demonstrably,
a Hurrian city; that (2) this city does not
begin at that point in time, but at least a thousand years
earlier; that (3) there is no evidence for a solution of
continuity from the very earliest beginning; and that (4)
the third millennium Temple Terrace was not erected from
the plain level, but rather rested on an earlier structure.
We must consider one additional piece of circumstantial
evidence. The âbi, as we have it, is a structure
that is uncontrovertibly Hurrian. It may be seen as the
monumental frame for a ritual that is known to us in detail
from later periods. The frame is so specific in its typological
features and in its import that as far back in
time as we can go it must be considered to remain specifically
Hurrian. Now, the current depth of the âbi, at 8
ms from the top, reaches to the same date as the extant
Temple Terrace, the mid third millennium. The circular
wall that defines the structure gives every indication of
going deeper, and it seems plausible that the emplacement
of the original âbi may in fact be at the plain level.
Since that level is some 6 ms below the current excavation
limit, we may assume a depositional history for the
âbi that also takes us back, as with the Temple Terrace,
to much earlier periods, very possibly the beginning of
the city in the fourth millennium.
5. Towards the identification of the
earliest Hurrian homeland
5.1. The hinterland
The significance of the finds from Urkesh reaches
beyond the realm of the city itself. The hinterland of
Urkesh extends, it seems to us, to the northern highFig.
12. Photos of Uruk period cylinder seal impressions excavated in Urkesh showing standard being carried
150 Giorgio Buccellati, Marilyn Kelly-Buccellati
Bull. Georg. Natl. Acad. Sci. vol. 175, no 2, 2007
lands of the central Tur-Abdin and possibly beyond. We
do not have direct evidence for this, but the cumulative
import of the indirect evidence is significant. The strong
and pervasive connections to the north, discussed
above, are not found to the same extent at other coeval
sites. Particularly important are those connections that
pertain to ideology, of which the andirons are the most
telling, because they affect popular consciousness.
A possible negative argument might be seen in the
limited amount of ceramics specifically linked with the
north, in particular the Early Trans-Caucasian ware. However,
this may simply mean that (a) no significant transshipment
of goods from the hinterland played any significant
role, and (b) no production of such ceramics took
place in the urban centers linked with the highlands (such
as Urkesh), because it was not deemed to have either
ideological or prestige value.
We may also consider the geographical location. The
cities in the piedmont area immediately adjacent to the
highlands, among which Urkesh played a major role, were
in an ideal position to act as gateways for the shipment
of metals, and perhaps timber, from the mountains. The
distinctiveness of the Urkesh material culture indicates
that this was no outpost of Mesopotamian civilization,
but rather an autonomous center, with its own strong
urban character. Thus, rather than a peripheral outgrowth
from the south, Urkesh can best be understood as the
urban efflorescence of the mountainous north. Consider
the contrasting situation with Tell Brak/Nagar. It certainly
looks to the south, not the north.
The argument we have developed in support of a
Hurrian ethnic identity of Urkesh also goes along with
this. The only possible linguistic links of Hurrian are with
the north, reaching possibly all the way to the Caucasus
and to Georgia in particular. We certainly do not have
Hurrian cities south of Urkesh (even the neighboring
Nagar does not meet the criteria proposed above). What
we have called “Hurrian urban ledge,” i.e. the arc represented
by Chuera, Urkesh, Nineveh and (presumably)
Kumme, remains in the third millennium just that, a narrow
ledge. And it is the ledge of the northern highlands
that provide subsistence and wealth.
We have often suggested that the success of Urkesh
in withstanding the Akkadian onslaught was due not to
any military strength that they possessed, but to the
unique socio-political formula with which the city could
maintain control of the mountains and exploit its resources.
Rather than through an articulate administrative
and bureaucratic network, Urkesh could efficiently mine
the resources of the mountains because its ethnic ties
with the population were acknowledged, by the mountain
villages, as binding and operative. The Hurrian myth
of Silver provides an eloquent literary projection of this
state of affairs. Silver is a boy living with his mother in
the mountains, and has rough encounters with the other
children (read: the villages squabble). He then sets out
to look for his father, Kumarbi, because, his mother tells
him, he administers justice for all the lands from his main
seat in Urkesh (read: the villages are aware of the unifying
and regulating role played by the lowland city). When
Silver reaches Urkesh, he does not find his father,
Kumarbi, because he is walking in the mountains (read:
the city maintains active control over the villages by
being recognized as the ancestral dimension of public
life). The wider significance of this situation, so interpreted,
is that it outlines a novel socio-political model for
ancient Syro-Mesopotamia, and for Syria in particular –
next to the Mesopotamian style city-states (like Ebla),
the macro-regional states (like Mari), the tribal configurations
(as with the Amorites). Such a wealth of diverse
political realizations is indicative of the historical vibrancy
of Syria at the beginning of history.
5.2. The homeland
The conclusion we propose builds on the premises
that we have been articulating in this article. (1) The archaeology
of Mozan shows that we have there a city
already by the middle of the fourth millennium, and we
have every reason to believe that this city is indeed
Urkesh from the beginning. (2) This city has clear and
distinctive links with the north, which is its real hinterland,
in terms of material culture, language and ideology.
(3) Urkesh can be shown to have a specific ethnic affiliation
with the Hurrians, and, we may also argue, the northern
hinterland, as well.
This hinterland would obviously be coterminous with
the earliest period of the city’s history, which is at least
the fourth millennium. But it seems inescapable to assume
that, in fact, the hinterland preceded the urban development
of Urkesh – to assume, in other words, that
we have Hurrians in the highlands to the north of Urkesh
by at least the fifth millennium. How far north this extended
we have no way of knowing. But we have at least
good reasons to see the southern part of the Tur-Abdin
as an area where the Hurrians were at home since late
prehistoric times. We cannot of course say that they are
autochthonous in the sense that they did not replace
any other population, in other words, that this is their
original homeland. But if not, it is at least the southern
bridgehead of this homeland, a bridgehead that was critical
in the formation of new urban structures in the low-
lands.
Urkesh and the Question of the Hurrian Homeland 151
Bull. Georg. Natl. Acad. Sci. vol. 175, no 2, 2007
arqeologia
urqeSi da xuritebis pirvelsacxovrisis sakiTxi
jorjo buCelati*
, merilin keli-buCelati**
* direqtori, mesopotamiis laboratoria, kotsenis arqeologiis instituti, kaliforniis universiteti,
los anjelesi, aSS;
sapatio profesori, axlo aRmosavleTis enebisa da kulturebis departamenti, kaliforniis universiteti,
los anjelesi, aSS;
sapatio profesori, istoriis departamenti, kaliforniis universiteti, los anjelesi, aSS
** kotsenis arqeologiis instituti, kaliforniis universiteti, los anjelesi, aSS
(warmodgenilia akademiis wevris T. gamyreliZis mier)
zogjer arqeologiuri gaTxrebi garkveul adgilze ufro mets dagvanaxebs, vidre es Sida,
adgilobrivi urTierTobebiT Cans. swored ase moxda es Crdilo-aRmosavleT siriaSi, tel mozanis
didi yorRnis gaTxrebze muSaobisas. 1984 wels, roca gaTxrebi daviwyeT, vvaraudobdiT, rom am
adgilebSi SeiZleboda Zveli urqeSis naSTebi aRmogveCina. es varaudi sabolood 1995 wels gamarTlda,
rodesac mTeli rigi warwerebis gaSifvram dagvanaxa, rom Cven gagviTxria am dromde ucnobi
“urqeSis mmarTvelis (endan)” sasaxle, romelic xuritebis saxels – tupqiSs – atarebda. samefo
saxlis zogierT sxva wevrsac xurituli saxelebi erqvaT.
onomastika iyo mxolod pirveli mraval damaxasiaTebel niSan-TvisebaTagan, romelmac damajereblad
daadastura am uZvelesi qalaqis eTnikuri identuroba. didma materialurma kulturam,
romelmac dRis sinaTle Cveni gaTxrebis Sedegad ixila, gamoavlina eTnikur identifikatorTa
konstelacia, rac sxva gaTxrebis dros ar momxdara. am mxriv ori ram aris gansakuTrebiT mniSvnelovani:
upirveles yovlisa is, rom urqeSis gareubnebi vrceldeboda sakmaod Sors Crdilo
zeganze; meorec is, rom siRrmeSi gadavinacvleT im dromde, roca urqeSma rogorc qalaqma Camoyalibeba
daiwyo. am sxvadasxva faqtoris arseboba uflebas gvaZlevs wamovayenoT sakiTxi, rogorc
SesaZlebeli hipoTeza, xuritTa yvelaze adreuli dasaxlebis Sesaxeb. winamdebare statias, romelSic
vajamebT Cven sxva SromebSi ganxilul sakiTxebs, warmovadgenT mimoxilvis saxiT, rasac
erTvis garkveuli raodenobis fotomasalac.
Received June, 2007