The Origins of the Urartians in the Light of the Van/Karagündüz Excavations

Author(s): Veli Sevin

Source: Anatolian Studies, Vol. 49, Anatolian Iron Ages 4. Proceedings of the Fourth Anatolian
Iron Ages Colloquium Held at Mersin, 19-23 May 1997 (1999), pp. 159-164
Published by: British Institute at Ankara
Stable URL: http://www.jstor.org/stable/3643071 .
Accessed: 17/05/2014 16:10

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp

.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.

British Institute at Ankara is collaborating with JSTOR to digitize, preserve and extend access to Anatolian
Studies.

http://www.jstor.org

This content downloaded from 81.165.57.70 on Sat, 17 May 2014 16:10:17 PM

All use subject to JSTOR Terms and Conditions
 The origins of the Urartiansin the light of the
excavations
Van/Karagiindtiz

Veli Sevin
CanakkaleOnsekiz MartUniversity

The UrartianKingdom,as is well known, played a major According to cuneiform records and especially to
power role on the stage of history in easternAnatolia in philological studies, the Urartianswere based in the area
the second half of the ninth century BC and remained well to the south of Lake Van, at presentthe borderarea
powerfuluntil the second half of the seventhcenturyBC. of Turkey,IranandIraq,aroundthe holy town of Musasir
With their highly advanced architecturaltraditions and (Salvini 1982: 31; Haas 1986: 23, 26). In fact infor-
organised state structure,the Urartianstake their place mation about the location and cultureof the Urartiansis
among the most exciting civilisations of the first half of inconsiderable before they appeared in history around
the first millenniumBC in the Near East. 830 BC as a strong state with their capital city at Tushpa
Extensive detailed researchand publicationhas been (Van Kale). There is no agreement,for instance, on the
carried out on Urartiancivilisation for over a hundred location of cities such as Arzashkun and Sugunia
years, but the origin and dynamicsof the developmentof mentionedby the Assyrianking ShalmaneserIII (Salvini
this civilisation are still obscure. The Assyrian annals, 1982: 5; Burney 1966: 60). On the presentevidence, like
which startfrom the 13th centuryBC, are at presentthe a shooting star from the darkness of the past, the
only source for understandingthe early periods. These Urartiansappearto us all of a sudden as a strong state
records were intended as propagandaand their accuracy and an importantcivilisation.
is in many instances thus questionable. Is the actual situationlike this? Did this civilisation
The Assyrianroyal annalsmentionpeoples named as suddenly appeararoundLake Van? Did they emigrate
Nairi and Ur(u)atriwho lived in the northof theirregion, from anotherregion? Alternatively,did they undergo a
and they also mention the existence of a large numberof development stage in the same area, which may have
kings and kingdoms. The Assyrian king ShalmaneserI taken hundredsof years? Archaeologicalinformationis
(1274-1245 BC), for example, recorded that Uruatri the most importantevidence with which to attempt to
consisted of eight countries (tribes) (Grayson 1976: no solve these problems,althoughthis informationis sparse.
527). Tukulti-NinurtaI (1244-1208 BC) and Tiglath- Additionally, long-term systematic excavations at
pileser I (1115-1077 BC) mentioned that Nairi had Urartian fortresses have not provided evidence with
between 23 and 60 kings (Grayson 1976: nos 715, 721, which to solve the problem of Urartianorigins. The
760, 773, 803). It was later understood,however, that archaeologicalevidence indicatesthat a pastorallifestyle
these rulers who were called kings were in fact only took over from a settled lifestyle at the end of the third
chiefs of tribes which did not develop into centralised millennium in eastern Anatolia. According to this, the
statesin easternAnatolia. Accordingto this inforination, transition back to settled life only occurred after the
it was believed that Nairi extended from the Tur-Abdin second half of the ninth centuryBC, an intervalof about
mountains in the south to the mountainous area 1000 years. Certainly there is adequate evidence in
southwestof Lake Van in the north. Uruatri,however, is easternAnatoliato demonstratea majorhiatusduringthe
believed to have lain more in the area to the north and period from the late third millennium to the mid ninth
east of Lake Van. The recordsalso mentionthe existence century BC. Some settlements were destroyed by fire
of many cities and fortressesbelonging to these domains, duringthe Early Bronze Age, and the majoritywere not
some of which were surroundedby strong defensive subsequentlysettled until the Iron Age, if at all. Places
walls. In addition, Urartianfortresses are depicted in such as Dilkaya, Van Kalesi Hoytik, Karagtindtizand
reliefs on the Balawat Gates from the period of even Erci?/?elebibag, all in the region of Van, and
ShalmaneserIII (Gunther1982: 104, pl Ia, d). In spite of several hoytiks now flooded by the waters of the Keban
these records,thereis still minimalinformationaboutthe reservoir, such as Elazig/Degirmentepe, Han Ibrahim
natureof pre-Urartiansettlementin easternAnatolia. ?ah, and Nor?unTepe, and sites such as Malatya/Deg-

159

This content downloaded from 81.165.57.70 on Sat, 17 May 2014 16:10:17 PM

All use subject to JSTOR Terms and Conditions
 Anatolian Studies 1999

irmenTepe, Imamogluand Ko6kerbabain the land of the

Karakayareservoir are all witnesses to this proposal.
This gap can also be detected in Transcaucasia and
IranianAzerbaijan,with a few exceptions (for example
at Ktiltepe II in Nakhicevan, and Haftavan Tepe in
IranianAzerbaijan,see Bumey, Lang 1971: 86).
So, who were these Urartians with their strong
characterof settled life? We will to try to address this
issue in the following discussion.
There is new information from the cemetery of
Karagtinduzabout the UrartianKingdom around Lake
Van. Karagtinduzis a village located 35km northeastof
the centre of Van at an elevation of 1890m, on the north-
easternborderof the mineral waters of Lake Erqek. An Fig 2. Karagiindiizcemetery,dromos tombKl
importantIronAge cemetery was uncoveredin 1991 and
rescue excavations startedin 1992 (Karagtinduzrescue As the roofs have collapsed, the original height is not
excavations are supported by the Governate of Van clearbut the existing heightsvary between 1.50 to 2.50m.
Province and IstanbulUniversity ResearchFund,project The roofs are similarto the Emis (Unseli) graves, which
number613/210494). The cemetery is 1.5km east of the were covered with the technique called false arch or
modem village, on an alluvial plain (fig 1). corbelling(Sevin 1987: 36). The walls areeven, only one
sample (KI) having a large niche near the middle of the
southwall (fig 2). This kind of largeniche is encountered
later as an early characteristicof classic Urartiangrave
architecture(Sevin 1986: fig 3; 1987: fig 11).

Fig 1. Karagiindiizcemetery

During the excavations from 1993 to 1996, nine

tombs (Kl-10) were excavated in Karagiinduzcemetery.
Six of these were chamber tombs with dromoi (fig 2).
Burialchamberswere formed as pits dug into the ground
and surroundedwith rough stones and mortar. All of
them are orientated northeast to southwest. All the
chambers, with one exception (K4), have a little front
entrance(dromos)at the narrowsouthwestside, which is
at a higher level than the chamber(fig 3). Thereis a low
door, reaching into the chamber. After the burial
placement the entrance was blocked with a vertically
positioned heavy stone slab, and the dromos was filled
with stones and earthuntil a new burialwas to be made.
Some of the burial chambers with stair access are
rectangularin shape. The ground plans are approxi-
mately 2.10m by 4.20m, 1.75/1.85m by 3.30m,
1.30/1.50m by 3.20m and 1.25/1.64/1.45m by 4.30m. Fig 3. Karagiindiizcemetery,tombK5

160

This content downloaded from 81.165.57.70 on Sat, 17 May 2014 16:10:17 PM

All use subject to JSTOR Terms and Conditions
 Sevin

solution to the lack of space. One of the graves, K6-7

(fig 4), for example, has a secondary chamber at the
southeastcomer, reachedby a low passageway. An oval
hollow was dug into the base of this chamber and ten
skeletons and their grave goods placed therein (Sevin,
Kavakli 1994a: fig 2).
In anothergrave, K10 (fig 5), there were ten skulls
with grave goods in a simple earthhollow dug througha
niche-like little window in the lower rearwall nearto the
middle and close to the surface. This kind of enterprise
can be considered as a prototype of the multi-roomed
Urartianexamples. In the same way as it is encountered
in Urartianrock tombs, and as it is easy to distinguish
Fig 4. Karagindiiz cemetery,tombK6-7 these side rooms with their unfinished workmanship
from the main chamber,it may be thoughtthat these side
One of the tombs in the cemetery, K2, has very rooms were used as stores for skeletons and grave goods
distinctive features (Sevin, Kavakli 1994: 335; 1996a: (Sevin, Kavakli 1996a: 19; 1996b: 5).
19, 1996b:4, fig 3). This tomb is orientatednortheastto
southwest like the others, and the entrance on the low
side faces west. Perpendicularlyplaced monoliths were
used, ratherthan stones placed on top of each other, to
build the walls, and on top of these heavy flat slabs were
used to facilitate the setting of the flat roof stones. The
characteristicsof this tomb suggest a stone box in its
appearance,but the entrance at the southwest makes it
clear thatit is a very interestingandprimitivelow type of
grave chamber. The entrance,0.53m wide, was blocked
with a thin stone slab. The whole length of this grave
chamber is 2.50 to 2.70m and there is a corridor-like
passage 0.53 to 0.65m wide and 0.70m deep which leads
to the grave chamber. The narrow and shallow room
widens towards its eastern comer, reaching a width of
2.30m. There is a pit, dug into the hardclay soil, with a
diameterof 1.20m and a depth of 0.45m, in which were
found skulls, skeletons and grave goods, which help in
the interpretation of this structure. This situation
reminds us that bottle-shapedpits, of obscure purpose,
occur in some Urartiangraves (Burney 1966: 107, fig 22; Fig 5. Karagiindiizcemetery,tombKO1
Sevin 1994a: fig 5; Kleiss 1974: fig 18). This grave,
however, could not be understoodin detail because of A general featureof the architecturalstructureof the
damage at the northeastend. Karagiinduzgrave chambers is the fact that the grave
Grave K2 from Karagiinduzappearsmore archaicin chamberscontainedmany skeletons. Mostly depending
its architectureand finds. There is a similar one to this on the size of the grave, the numbervaries from 20 to 80.
but a little smaller in the Van/Dilkayacemetery (no 1) Burials were placed in the chamberin a flexed position,
(qilingiroglu 1985: 153, figs 4-5, plan 4; 1991: 30, fig on one side without any distinct orientation(figs 4, 6).
03.1). These examples of grave chamberswith opening When a new burial was made the old ones were pushed
door systems also have features characteristicof stone aside to open a space for the new one, thus creatinga pile
boxes, such as being very shallow and having big stone of older skeletons at the back throughtime. Only the last
plates on top of the graves. These examples can be inter- burialthus kept its original position. This aspect is also
preted as being transitionalfrom stone boxes to grave known as a feature of classic Urartianchamber tombs
chambers. (Ogtin 1978a: 661; 1978b: 62, fig 2). The same tradition
One of the most interesting characteristicsof the was appliedto Assyriangrave chambersfrom the Middle
Karagtinduzgraves is the enterpriseshown in finding a Assyrianperiod (Haller 1954: 102, fig 136a).

161

This content downloaded from 81.165.57.70 on Sat, 17 May 2014 16:10:17 PM

All use subject to JSTOR Terms and Conditions
 AnatolianStudies 1999
Fig 6. Karagiindiizcemetery,tombK5
In addition to this, a smaller number of the grave
chambers contain burnt bones, mostly belonging to
youngsters(Sevin, Kavakli 1996: 23; 1996b:6). As with
Urartianexamples, there is no trace of these cremated
bones and ashes having been kept in urns. It is another
characteristicof Urartiangraves that one may encounter
inhumationand cremationnext to each other in the same
grave (Ogun 1978a: 660; 1978b: 6).
Variousgrave goods were deposited with the burials,
which were either wrappedaroundwith a cloth or buried
in garments. Amongst these grave goods some are
universalfor all burials,such as a potterybowl with open
rim and a high or shallow-necked and narrow-rimmed
cup (fig 7). Generally some vertebraeof lamb or young
goat were placed in the first type of cup and a liquid was
contained in the second type. The cups were usually
placed near the head of the individual. In one of the
graves, K6-7, the cups contained some food as well as
pieces of mandible belonging to a sheep-goat type of
animal. In addition,therearehearthsnext to every grave.
It is probable that the sacrificed animal was cooked in
these hearths. The occurrence of so many pots in the
graves may indicate thatthe feast includedall the people
who attendedthe funeralceremony.
Iron ornamentsand ceremonialweapons are some of
the most interestingof the grave finds. They comprise
bracelets, anklets and rings made by hammering Fig 7. Cupsfor foodstuffs and liquids
162
This content downloaded from 81.165.57.70 on Sat, 17 May 2014 16:10:17 PM
All use subject to JSTOR Terms and Conditions
techniques,toggle-pins, daggers, spear points, knives in
differentsizes, a mace, a chain (?), and long iron pieces
which are interpretedas swords (Sevin, Kavakli 1996a:
pls 12-25). The use of iron for ceremonialpurposesand
ornamentsmight be an indicationthat these early graves
in the cemetery are contemporary with Hasanlu IV,
KordlarTepe IIA and the beginning of DinkhaTepe II.
Bronze works of art are very few comparedto iron.
The most significant are a small disk, a shuttle-shaped
bead wrappedwith wire, barrel-shapedbeads, one pairof
ear-ringswith frit beads and assortedtypes of rings and
pins. It is very common to encountersimilarpins in the
period of the Urartian Kingdom, often with a head
adornedwith a pairof cocks back to back, a poppy,a bud
or an eagle (Sevin, Kavakli 1996a: pls 26-7).
Many pots were encountered in the Karagtinduz
cemetery. These pots can be divided into two major
groups according to their technology, shape and
decoration. In the first, larger,group the colour of the
fabricis usually pink, less often brownor reddish. These
pots are usually wheel-made, but there are some hand-
made examples as well. The most common technical
characteristicof these pots is the differentiationin colour
on burnishedsurfaces,caused by varying temperaturein
the kiln. This characteristiccan be seen in Karagtinduz,
the contemporaryEmis cemetery and HasanluIV (Sevin
1996: 441, note 6; Dyson 1989: 108, fig 7). The
commonestforms are inverted-rimmedearthenwarepots
with horizontalfluted decorationunderthe rim, pots with
angularprofiles and verticallyperforatedhandles (fig 7),
and vessels with a flaring rim usually with a groove
inside, short body, S-shaped profile and flat base. In
additionto these, pots with a simple rim, carinationwith
or withouthandle and flat base, andjars with narrowand
high necks are significant. Small blobs often occur on
the shouldersof pots as decorativemotifs. Some of the
cups are decoratedby incision, and a few of them bear
motifs made by stamping.
 Sevin

Overall, the pots with diagonal incision or horizontal

)
grooves underthe rims, made while the paste was moist,
are similar to those of the Iron Age in their techniques \ /

and form. These are found scatteredfrom Elazig-Bingol 2

region in the west, to the western part of Lake Urmia in

the east (Sevin 1991; 1994b: 222, fig 21.5; 1996b;
Pecorella, Salvini 1984: figs 24, 26, 30, 33, 62). The
closest relatives, however, can be seen at Emis in the
region of Lake Van (Sevin 1996a), Ahlat (Ozfirat 1993:
362, fig 7, pl 3), Van Kalesi Hoyuk (Sevin 1994b: 222,
fig 21:5) and, admittedly a little later, at Dilkaya
(gilingiroglu 1992: 475, figs 12-7).
The second group of pots is less numerousthan the
first, and all of the examples are wheel-made. The most
interestingcharacteristicis that the surface of the pots is
coated with a thick, highly polished red-brownslip. The
commonest forms are carinatedcups with simple flaring
rims (fig 8:1-2), and horizontallyribbedpots with ring or
flat bases. Included in this group are trefoil jugs with
ribbed shoulders(fig 8:3-4). The examples with a short
thick neck, squat body and handles that are the exact
copy of metallic vessels, are the earliest representatives
of this group. An example with a globular body and
elegant neck is the closest to classic Urartianones in its
shape (fig 8:3). It is clear thatthis groupand the metallic
vessel imitation types compare well with the character-
istics of classic Urartianpottery in their techniques and
shapes. In contrastto the first group, which occurs over
a large area,the shiny red-slippedware, also known from
the cemetery of Emis (Onseli), belongs to the Early Iron
Age, that is to say before the UrartianKingdom, and is
found only in the Vanregion (Sevin 1996a).
The existence of a cultural unity predating the
UrartianKingdomis clear in the areato the northof Lake
Ercek and Lake Van, even on its east and west shores.
From the aspect of materialculture,this area comprised
a distinctive zone in its own terms. Contraryto this, it is
clear that ratherdifferentculturalprocesses were taking
place in the region of west and south Urmia at the same
period. For instance,in contrastto multipleburialsin the
Van region, there are simple inhumationsin northwest
Iran (Musceralla 1968: 189, 194; 1974; Burney 1972:
134, fig 8, pl Ila; Dyson 1989: fig 3), apartfrom burials
in Geoy Tepe K (Burton-Brown1951: 142; Dyson 1965:
196) and Dinkha Tepe IV (Rubinson 1991). The
common grey ware andespecially its repertoireof shapes
in the Early Iron Age of western Iran does not occur in
the region of Lake Van.
The Early Iron Age culture which we are startingto
recognise from grave architecture,burialpractices,metal
technology and pottery from Ernis and Karagtindiuz
shows that it is closely related to the Urartianculture.
Therefore it is understandablenot to acknowledge a
culturalgap duringthe transitionfrom the EarlyIronAge
0
4
,
1,0O cm.
3
Fig 8. Red polished ware
to the Middle Iron Age, which starts with the estab-
lishment of the Urartianstate. We can, moreover,assert
that the culture attested at the cemeteries of Emis and
Karagiinduzis the progenitor of Urartianculture, and
that the roots of classic Urartiancivilisation should be
searchedfor in the Van region.
It is reasonable to surmise that the people who are
buried in Karagiinduzcemetery may have lived in the
hoyiik located 1.5km away. However, the excavatedpart
of this mound gives very little evidence aboutsettled life
in the Early Iron Age. It is thereforefeasible to suggest
that the people buried in the middle of the valley of the
Memedik _ay were members of a tribe with a pastoral
lifestyle.
Just as there is some evidence of pastorallife in the
Lake Van region until the Middle Iron Age, the excava-
tions in Dilkaya, Van Kalesi Hoytik and Karagiinduz
Hoyuk revealed very little evidence of architectural
levels from the Early Iron Age, indicating that the
settlement mounds were not occupied immediately
before the Urartian Kingdom. These three mounds,
however, played a strong settlement role during the
period of the UrartianKingdom.
After all this, it is possible to say that some, at least,
of the eight nations or tribes mentioned in the Assyrian
royal annals followed a pastoral lifestyle. The
cemeteries of Karagiinduz,Emis and Dilkaya may very
well belong to such tribes.

163

This content downloaded from 81.165.57.70 on Sat, 17 May 2014 16:10:17 PM

All use subject to JSTOR Terms and Conditions
 AnatolianStudies 1999
Bibliography
Burney,C A 1966: 'A first season of excavations at the
Urartiancitadel of Kayalidere' Anatolian Studies
16: 55-111
1972: 'Excavationsat HaftavanTepe 1969' Iran 10:
127-42
Burney,C A, Lang, D M 1971: The Peoples of the Hills
- AncientArarat and Caucasus. London
Burton-Brown,T B 1951: Excavations in Azerbaijan,
1948. London
(ilingiroglu, A 1985: 'Van Dilkaya Hbyigti 1984
Kazilan' Kazi Sonu?larlToplantisi7: 151-62
- 1991: 'The EarlyIronAge at Dilkaya' in A (ilingiro-
glu and D H French (eds), Anatolian Iron Ages 2
(British Institute of Archaeology at Ankara
Monograph13). London:29-38
- 1992: 'Van-DilkayaHoyiiguiKazilarl Kapani' Kazi
Sonu?larlToplantisi14.1: 469-91
Dyson, R H Jr 1965: 'Problems of protohistoricIran as
seen from Hasanlu' Journal of Near Eastern
Studies 24: 193-217
1989: 'The Iron Age architecture at Hasanlu'
Expedition31.1-2: 107-27
Grayson, A K 1976: Assyrian Royal Inscriptions I-II.
Wiesbaden
Gunther, A 1982: 'Representations of Urartian and
westernIranianfortressarchitecturein the Assyrian
reliefs' Iran 20: 103-12
Haas, V 1986: 'Die altesten Nachrichtenzur Geschichte
des armenischenHochlands' in Das Reich Urartu.
Xenia. Konstanz:21-30
Haller, A 1954: Die Grdber und Griifte von Assur
(WVDOG 65). Berlin
Kleiss, W 1974: 'PlanaufnahmenurartaischerBurgen
und Neufunde urartaischerAnlagen in Iranisch-
Azerbaidjan im Jahre 1973' Archdologische
Mitteilungenaus Iran 7: 79-106
Muscarella, 0 W 1968: 'Excavations at Dinkha Tepe,
1966' Bulletin of the MetropolitanMuseum of Art
27: 187-96
1974: 'The Iron Age at Dinkha Tepe, Iran' Metro-
politan MuseumJournal 9: 35-90
Ogtin, B 1978a: 'UrartaischeBestattungsbrauche'in S
?ahin, E Schwerteim and J Wagner(eds), Studien
Religion und KulturKleinasiens. Festschriftfir F
KDo5rner.Leiden: 639-78
164
This content downloaded from 81.165.57.70 on Sat, 17 May 2014 16:10:17 PM
All use subject to JSTOR Terms and Conditions
1978b: 'Die urartaischenGraberin der gegend von
Adilcevaz und Patnos' in E Akurgal (ed),
Proceedings of the Tenth Congress of Classical
Archaeology.Ankara: 61-7
Ozfirat,A 1993: 'M. 0. II. Binyil Dogu AnadoluBoyali
Seramik Kulturleri Uzerine Ara?tirmalar'
ArastirmaSonu?larlToplantisi11: 359-77
Pecorella, P E, Salvini, M 1984: Tra Lo Zagros e
l'Urmia.Roma
Rubinson,K S 1991: 'A mid-secondmillenniumtomb at
Dinkha Tepe' American Journal of Archaeology
95: 373-94
Salvini, M 1982: 'Researchesin the region between the
Zagros Mountainsand Urmia Lake' Persica 10: 1-
35
Sevin, V 1986: 'UrartuMezarMimarisineYeni Katkilar'
AnadoluAra;tirmalarl10: 329-40
- 1987: 'Urartu Oda-Mezar Mimarisinin Kokeni
Uzerine Bazi Gozlemler' in A 1ilingiroglu (ed),
AnadoluDemir (aglari. Izmir:35-55
- 1991: 'The Early Iron Age in the Elazig region and
the problem of Mushkians'Anatolian Studies 41:
87-97
- 1994a: 'ThreeUrartianrock-cuttombs from Palu' Tel
Aviv 21: 58-67
- 1994b: 'The excavations at Van castle mound' in A
(ilingiroglu and D H French(eds), AnatolianIron
Ages 3 (BritishInstituteof Archaeology at Ankara
Monograph16) London:221-8
- 1996a: 'Van/Emis (Unseli) Nekropolii Erken Demir
_agC(anak?(omlekleri'AnadoluAra;tirmalari14:
439-67
- 1996b: 'La ceramiquede l'age du fer ancien dans le
region de Van (Turquiede 1'est)'OrientExpress 3:
89-90
Sevin, V, Kavakli, E 1994: 'Van-Karaguindiiz Erken
Demir _agi Nekropolii KurtarmaKazilarn1992-
1993' Kazi Sonu?larlToplantisi16.1: 331-50
- 1996a: Van-Karagiindiiz. An Early Iron Age
Cemetery.Istanbul
- 1996b: 'Van-Karagtindiiz Erken Demir (agi
Nekropolii'Belleten 227: 1-20