Some Thoughts on Iconographic Relations between the Arabian Gulf and Syria-Mesopotamia during the Middle Bronze Age by Luca Pyronel (2008)

Society for Arabian Studies Monographs No. 7 Series editors D. Kennet & St J. Simpson

Intercultural Relations between South and Southwest Asia Studies in commemoration of E.C.L. During Caspers (1934-1996) Edited by

Eric Olijdam Richard H. Spoor

BAR International Series 1826 2008

This title published by Archaeopress Publishers of British Archaeological Reports Gordon House 276 Banbury Road Oxford OX2 7ED England [email protected] www.archaeopress.com

BAR S1826 Society for Arabian Studies Monographs No. 7

Intercultural Relations between South and Southwest Asia: Studies in commemoration of E.C.L. During Caspers (1934-1996)

© the individual authors 2008

ISBN 978 1 4073 0312 3 Cover illustration by © J.M. Kenoyer (Kenoyer and Meadow, Fig. 5.3)

Printed in England by Alden HenDi, Oxfordshire All BAR titles are available from: Hadrian Books Ltd 122 Banbury Road Oxford OX2 7BP England [email protected]

The current BAR catalogue with details of all titles in print, prices and means of payment is available free from Hadrian Books or may be downloaded from www.archaeopress.com

INTERCULTURAL RELATIONS BETWEEN SOUTH AND SOUTHWEST ASIA. STUDIES IN COMMEMORATION OF E.C.L. DURING CASPERS (1934-1996) E. Olijdam & R.H. Spoor (eds) BAR International Series 1826 (2008): 236-252

Some Thoughts on Iconographic Relations between the Arabian Gulf and Syria-Mesopotamia during the Middle Bronze Age Luca Peyronel

bifacial seals), which were directly influenced by Mesopotamian glyptic (Peyronel 1997; Denton 1997).4 The most recent effort in Arabian Gulf archaeology is devoted to problems of relative chronology, pottery types and re-assessment of stratigraphy, trying to substantiate the links between the Mesopotamian alluvium, the Oman Peninsula, the Iranian Plateau and the Indus Valley through a more refined periodisation.5 In this respect, how should we consider the seals’ iconography and stylistic analysis? First of all, when we look at the figurative patterns of Dilmun glyptic we have to face the problem of correctly reading the motifs engraved on the circular stone disc. Lacking a horizontal alignment of the scene —as with the cylinder seals— it is necessary to define a specific methodology so that confusion in the overall interpretation can be avoided. A useful approach might be based on three differentiated and hierarchical analytical steps:

The indigenous character of Dilmun glyptic has been underlined ever since the first seals were found at Failaka and Bahrain at the end of the 1950s (Glob 1954: Fig. 5; Bibby 1957: Fig. 13a-c; 1958). The Danish archaeological investigations in the Arabian Gulf allowed the identification of the provenance of some enigmatic seals from Woolley’s excavations at Ur (Gadd 1932; cf. also Mitchell 1986; Peyronel 2000: 190-199), relating them to “a new cultural group” (Glob 1959: 238). The stamp seal impressions on a cuneiform tablet from the Yale Collection dating from the reign of Gungunum of Larsa (Hallo & Buchanan 1965; Buchanan 1967) provided the first absolute date for the glyptic material discovered in the Gulf area. At the same time, it also established an unquestionable link between the land of Dilmun and the cities of the Mesopotamian alluvium, matching the textual evidence for the development of maritime commerce during the Ur III and Isin-Larsa periods (Oppenheim 1954; Leemans 1960; 1968). The complex pattern of interrelations that can be inferred from the seals’ iconography, which was only briefly pointed out by Buchanan (Hallo & Buchanan 1965: 207208), has been discussed by Porada (1971) in a seminal article presented in the occasion of the Third International Conference on Asian Archaeology, and, several years later, by Kjærum (1986), during the Bahrain Historical Conference, held in Manama in 1983.

A) decomposition of the figurative pattern into separate elements; B) identification of the relationships between single elements; C) definition of a scene and eventually separation of filling motifs. More than one hundred single motifs can be identified on Dilmun seals from Failaka and Bahrain. They are related to three main categories with six ‘possible’ schemes of relationship:

The origin of the Arabian Gulf glyptic at the end of the 3rd millennium B.C. is now quite well understood: stamp seals from stratified layers at Qala’at al-Bahrain Excavation 520 (or North Wall Sounding) show an evolution of distinctive typologies, which culminates at the beginning of the 2nd millennium B.C. with the ‘classic’ Early Dilmun seal (Kjærum 1994).1 The presence of an archaic type (Arabian Gulf type) and the identification of a ‘transitional’ group (proto-Dilmun type) from Qala’at level IIa (ca. 2100-1950 B.C.) allows us to reconstruct the first steps of the Dilmunite glyptic tradition.2 It is more difficult to establish the ‘lower’ chronological limit of Dilmun seals, which appeared during Qala’at period IIb-c (ca. 1950-1800 B.C.) and were in use at Failaka also during the Old Babylonian (Failaka 3A) and, perhaps, the Early Kassite periods (Failaka 3B-4A).3 Finally, during the later phases (late Early Dilmun and Middle Dilmun periods) new styles and types were introduced (i.e. Kjærum’s styles II-III, and

categories: human and divine figures; animals; objects and symbols; relationships: figure/figure; figure/animal; figure/object; animal/animal; animal/object; object/object. Thus, the concept of ‘relationship’ concerns either a direct human/divine/animal ‘action’ towards an object or towards another figure (i.e. a man holding a drinking tube, a bull-man holding a standard, a man killing an antelope, a lion attacking a gazelle), or ‘physical’ contact between elements (an object placed on a podium, an animal ‘above’ an altar, a figure standing on an animal, etc.). Steps B) to C) are critical, and in several cases it is quite hard to distinguish if an element could be considered a mere filling motif or if it has some direct (or indirect) 236

relation with another element. Additionally, several seals show differently orientated figures, which testifies to an absence of a main axis of reading. In fact, four basic patterns of representation can be distinguished:

tition of a unique scheme of representation; they could not use it to create the variety of images that was necessary to distinguish each seal from another. The limited surface of the stamp (with a diameter usually of less than 3 cm) calls for a high number of figurative motifs to create ‘distinct’ scenes, necessary to distinguish groups/ categories of people or individuals. The major changes that can be observed between Arabian Gulf stamps and proper Dilmun seals are surely the development of the human/divine sphere, which becomes the protagonist of many scenes; and the more explicit articulation of the relationships between figures on the Dilmun stamp seals. Arabian Gulf iconography is centred around animals and symbols, arranged to cover the stamp surface; specimens engraved with human figures are very rare (Kjærum 1994: 341-344). This situation is linked to the strong influence of Harappan glyptic during the end of the 3rd millennium B.C. in Bahrain and in the Arabian Gulf, witnessed by circular stone seals with Harappan inscriptions, cubical weights, painted pottery, etc. The distribution pattern of square and circular seals with Harappan inscriptions in the Gulf, or the Iranian plateau, and in the Mesopotamian alluvium seems to reveal different ‘stages’ of cultural and commercial interrelations between Mesopotamia and the East, during the period immediately before the beginning of the 2nd millennium B.C. (Parpola 1994; Collon 1996; Peyronel 2000). When Dilmunites became the direct referents of the Mesopotamian ‘seafaring merchants’, the maritime contacts changed and the flow of goods was controlled by settlements on Bahrain and Tarut, and by their outposts on Failaka at the head of the Gulf.8

1) single orientation for primary axis; 2) multiple orientation for main axis and secondary axis; 3) multiple orientation for equal axes; 4) circular or radial composition. The transmission of iconographies originating in another cultural sphere can be observed at all the levels mentioned above, with single motifs, peculiar relationships between figures, and ‘scenes’ adopted and elaborated again in the Dilmun glyptic repertoire. Trying to reconstruct ways and routes of these contacts is certainly only one aspect of the study of Dilmunite iconography and iconology. Understanding the meanings of the figures within the Dilmunite cultural sphere is more important, but also more difficult. It is not a problem of differentiation between ‘foreign’ and ‘local’: the focus is on building up a framework of the Dilmunite artistic expression, in which to place the significance of the iconographic transmission. The ‘high’ or ‘classic’ glyptic of the Early Dilmun period shows the predominance of Mesopotamian iconography: the kaunakes skirts, the horned headdress of the gods, the supernatural beings such as the bull-men, the erotic and symposia scenes, standards and astral symbols, peculiar animals as the monkey and the lion, and the harp with bull-shaped sound-box, represent only the most common features of an artistic world surely influenced by the Mesopotamian one. However, during the late Ur III and Isin-Larsa periods (ca. 2000-1800 B.C.) the situation is totally different: the ‘presentation scene’ and the ‘introduction scene’ which dominate Mesopotamian iconography during these periods, are not attested on the Dilmun seals.6 In more general terms this means that if we did not know the Early Dynastic and Akkadian glyptics we would consider Dilmunite and Mesopotamian sphragistic productions as two separate and almost totally different entities on both iconographic and stylistic grounds. A more detailed analysis reveals that relationships with the Early Old Babylonian seals (some standards and symbols, the square hatched-podium) do exist, but these are rather marginal.7 In this respect we should consider Mesopotamian iconography in a wider perspective, both from a synchronic and a diachronic point of view. On the one hand, we have to take other broad artistic categories into consideration, e.g. the clay figurines and plaques; and, on the other hand, we must take into account that the Mesopotamian repertoire which passed into the Arabian Gulf area during the last centuries of the 3rd millennium B.C. subsequently underwent a local transformation. The ‘evolution’ towards the presentation/introduction scene, which reflected Mesopotamian practice, did not find the right background in Arabian Gulf culture. The Dilmunites were not interested in the monotonous repe-

There are two main hypotheses that explain the origin of a large part of the Dilmunite iconographic repertoire: A. Mesopotamian provenance — The frequent commercial contact between the Arabian Gulf and the Mesopotamian alluvium attested during the second half of the 3rd millennium B.C. caused the transmission of certain glyptic representations during a timeframe spanning the Early Dynastic II-III and the late Akkadian period.9 B. Syrian and Cappadocian provenance — The classic Near Eastern iconographies came from the Syrian regions and were elaborated and assimilated in Dilmun, reflecting a strong ethnic presence of Amorites in the Gulf.10 What is clear is certainly the progressive retreat of Harappan sphragistics,11 which relates to the new role undertaken by the Dilmun civilisation in the Arabian Gulf and, at the same time, to the crisis and changes in the Indus Valley during the first centuries of the 2nd millennium B.C. This is not the place for an exhaustive iconographic analysis of the Dilmun repertoire, but it can be quite instructive to discuss some peculiar motifs to test the strength of the abovementioned hypotheses.

237

A

B

C

D

E

F

Fig. 1: Dilmun seals with gods. A) al-Sindi 1994: no. 17; B) al-Sindi 1994: no. 18; C) al-Sindi 1994: no. 19; D) Kjærum 1983: no. 185; E) Kjærum 1983: no. 186; F) al-Sindi 1994: no. 23.

The god and the bull-man

albeit sometimes a double crown is attested (Kjærum 1983: no. 185; al-Sindi 1994: no. 19) [Fig. 1c-d]. Gods are attested on seven seals from Failaka (Kjærum 1983: nos. 81, 185-186, 193, 212, 274; Pic 1990: no. 19) and on nine seals from Bahrain (al-Sindi 1994: nos. 17-20, 2324, 56-58), indicating a balanced situation between the two seal corpora. Finally, among the handful of Dilmunite seals found outside the Gulf, one from Ur depicts a scene with a seated garbed god with a bull and two bull-men (Gadd 1932: no. 14; Mitchell 1986: no. 10; Peyronel 2000:

The most evident ‘cultural’ traits linking Dilmun glyptic with Syria and Mesopotamia are the supernatural beings and divine figures: the bull-man and the god with horned headdress appear in several stamps from Failaka and Bahrain interacting with humans, animals and ‘cultic’ objects such as standards and hatched podia. We can distinguish the god only by the horned tiara, usually of ‘simple’ type with one pair of curved horns, 238

A

B

C

D

E

F

Fig. 2: Dilmun seals with gods. A) al-Sindi 1994: no. 24; B) al-Sindi 1994: no. 57; C) Kjærum 1983: no. 212; D) Kjærum 1983: no. 81; E) Kjærum 1983: no. 274; F) Kjærum 1983: no. 193.

no. 4.14). The divine figure is always depicted en profile with a horned crown and a sort of pigtail hanging over his shoulders, dressed in a kaunakes skirt, seated on a ‘throne’ (a square plain or hatched seat), involved in some ritual ‘acts’ or associated with animals. The god is the main person-age of the representation and only in two seals does he appear twice (Kjærum 1983: no. 81; al-Sindi 1994: no. 17). Naked male figures, bulls and gazelles, symbols and standards, and drinking tubes with jars are motifs

that recur with deities. The god interacts with humans, animals and objects: Relation with humans: direct/indirect: god with a naked or garbed attendant behind or before [Figs. 1c-f, 2e-f]12 direct/indirect: god with bull-men as attendants (Peyronel 2000: no. 4.14) [Fig. 3a] direct: god holding/touching a kneeling figure (Pic 1990: no. 19) 239

and Mature Old Syrian glyptic usually show a couple of personages holding a cup on both sides of a loaded table, or figures approaching a seated person with a bowl in his hand, relating the banquet to the kingship and possibly to the funerary ritual of royal ancestors (cf. Matthiae, Pinnock & Scandone Matthiae 1995: nos. 254, 290-291, 470). However, several cylinder seals are engraved with persons drinking through tubes, witnessing the survival of the iconographic theme during the 2nd millennium B.C. (de Graeve 1982: 20-23).20 The presence of drinking scenes on Arabian Gulf type seals firmly points to a transmission from Mesopotamia during the late Akkadian period and suggests an original elaboration during the first centuries of the 2nd millennium B.C. Moreover, the success of the motif in the mature Dilmun glyptic is testified by the theme adopted as a whole (pairs of drinking personages) and, additionally by the motif matched with other iconographic relationships. The god drinking through a tube seems to be a strong clue for direct Mesopotamian influence, albeit drinking gods are quite uncommon in Mesopotamia proper during the Akkadian period (de Graeve 1982: 19-20). What is certainly interesting to note is the similar process of adaptation which happened in Syria and in the Arabian Gulf during the Middle Bronze Age, revealing a common ‘cultural’ behaviour.

Relation with animals: direct: god grasping/touching a gazelle [Fig. 2c-d]13 indirect: god seated on a stool/throne ‘above’ a bull [Fig. 1d-e]14 Relation with objects: direct: god drinking through a tube leading to a jar; god holding a cup [Fig. 1a-f]15 direct: god holding/touching a crescent-standard (Peyronel 2000: no. 4.14) [Fig. 3a] indirect: god seated on a stool/throne16 When we look at the composition as a whole, two different ‘scenes’ can be recognised in which gods are possibly involved: A) god drinking with a tube — symposia scene; B) god stressing his powerful supremacy over the animals — master of animals scene. Generally speaking, the gods —who are attested on less than 5% of Dilmun seals— can be related to specific iconographic themes, which do not exclusively belong to the divine sphere but are also common to what appear to be human figures. In the first case, seals with a ‘drinking scene’ —the so-called symposia scene— constitute a rather substantial group on Failaka and Bahrain.17 The introduction of this motif probably happened during the end of the 3rd millennium B.C., as revealed by some Arabian Gulf and proto-Dilmun seals (Kjærum 1994: 343; al-Sindi 1994: nos. 10-11, 13-15, 26-27 from Saar, Hamad Town and Charnel House). The specimens show pairs of human figures drinking through short straws from a common jar; in the field above the vessel a symbol (crescent, star, schematic footprint, small animal) is usually placed, in the same way as later Dilmun seals. The first appearance of figures drinking through a straw from a vessel placed on the ground or on a table can be recognised in Mesopotamia during the Early Dynastic II, as testified by a gypsum plaque from the Inanna Temple VII at Nippur. The general theme of the banquet was quite popular during the Early Dynastic III and Akkadian period, when we find male and female figures related to drinking tubes or holding cups, in a scene ‘summarising’ a banquet probably connected to different ritual or ceremonial occasions (Selz 1983; Pinnock 1994). At the end of the 3rd millennium B.C. such a representation became uncommon and, at the beginning of the 2nd millennium B.C., it was probably connected only to representations of sexual intercourse. Terracotta figurines and moulded clay plaques depict a man and a woman performing a sexual act whereby the female figure bends down from the waist sometimes drinking from a jar through a tube and the naked standing man is shown penetrating her from behind (cf. Pinnock 1996).18 It is interesting to underline that these examples of ‘erotic art’ find exact parallels in a very limited amount of stamp seals from the Gulf, without doubt directly influenced by the Old Babylonian clay plaques.19 Contrary to Southern Mesopotamia, the banquet motif was preserved in the Syrian, Syro-Cappadocian and Anatolian glyptics of the Middle Bronze Age. Old Syrian relief

The iconography of the god as ruler of the animal world is also very common in the Dilmun repertoire. It is certainly linked to the peculiar role of the bull/gazelle/ antelope in the ritual performances and also by the strong links between other animals and ‘cultic’ symbols or presumed altars/podia attested on several Dilmun seals but also in Arabian Gulf specimens. Therefore, relationships between different personages and animals are a constant trait of Dilmun sphragistic, raising the problem of a correct interpretation of these figures: are they humans or gods? This is particularly difficult to establish for the master of animals scenes, where we can observe very similar compositions adopted for horned deities and nonhorned figures. We consider this theme to be a local elaboration reflecting Dilmunite religious belief, almost totally unrelated to the contemporary Syro-Mesopotamian spheres. The roots can be identified in the Mesopotamian and Iranian glyptic of the 4th and 3rd millennia B.C. (Amiet 1986), but the evolution should be better reconstructed within the context of Arabian Gulf cultures. Human figures (naked or with flounced skirts) only frequently appear in the proto-Dilmun and Dilmun seals introducing a new kind of representation, where relations and actions became the focus of what can be now called a scene. The animalistic repertoire (elaborated during the previous period) now interacts with the human sphere giving birth to several figurative schemes. One of the most popular is the ‘master of animals’, the conceptual sublimation of human predominance over nature, often translated on a ritual level. If humans at the high level of the social hierarchy with the ruler in primis answer for the cosmic order on earth, as in the Mesopotamian world, the two spheres overlap in the master of animals’ scene, obscuring any 240

A

B

C

D

F

E

Fig. 3: Dilmun seals with bull-men. A) Peyronel 2000: no. 4.14; B) Kjærum 1983: no. 208; C) Kjærum 1983: no. 121; D) Kjærum 1983: no. 122; E) al-Sindi 1994: no. 254; F) Kjærum 1983: no. 274.

help to identify specific gods.21 In this respect it seems more probable that Dilmun people prefer to represent ‘ritual’ or ‘cultic’ acts through these religious elements rather than referring to specific deities. We know from Mesopotamian cuneiform sources the connection between the land of Dilmun and the god Inzak with his spouse Meskilak/Ninsikila (Alster 1983; Nashef 1984). Inscriptions from Failaka and Bahrain on stamp seals, cylinder seals and steatite objects (Nashef 1984;

precise distinction. It is equally possible that the horned god only represents a deity with a Mesopotamian connection or the main god of the local pantheon. Moreover, we can think of symbolic representation of deities through standards, astral symbols, altars, etc., which are widespread elements in Dilmunite seals but appear to lack a direct relationship with certain recurrent figures. A marked difference between Dilmun and MesopotamiaSyria is in fact the absence of divine attributes which can 241

Glassner 1984) report the names of Enki, Inzak and PA.NI.PA and attest the presence of sacred buildings in Dilmun.22 In Nashef’s opinion, Inzak and Ninsikila might be considered the Dilmunite identification of the Mesopotamian pair Enki and Dagmalnuna, according to a syncretism process (Nashef 1984: 6-7). Moreover, Inzak of Dilmun (and/or Agarum) should have been worshipped as the god of the date-palm23 and, consequently, the schematic plant as well as the more realistic date-palm attested on stamp seals should be regarded as Inzak’s symbols. However when we turn to the glyptic evidence, out of more than a hundred examples that depict schematic vegetable elements only two seals show a direct relation between gods and plants [Fig. 2e-f]:24 the first shows a deity and a garbed figure each holding a branch in their raised hands (Kjærum 1983: no. 193), the second has a seated deity stretching out his arms towards a naked attendant holding a branch (Kjærum 1983: no. 274).25 Finally, it is interesting to point out that the only seal which attests a relation between bull-men and gods was found at Ur: the scene represents a garbed deity wearing a horned tiara, who is seated on a hatched stool apparently placed on the back of a bull, between two bull-men. Each of these holds a crescent standard, while one is also grasped by the god. Relationships between bull-man/ god, bull-man/crescent-standard, and god/crescent-standard never occur on Failaka and Bahrain, pointing to a stronger link for this seal with Mesopotamia: we are tempted to identify the owner as an alik Tilmun, who marked his seafaring commercial activity with an object that recalls specific Mesopotamian iconography.

Relationship with objects: direct: bull-man holding/touching a ‘ritual’ object or a branch [Figs. 3a, 4b, f, 5a-f]29 indirect: bull-man standing on a hatched podium (Kjærum 1983: no. 93) [Fig. 4a] When we look at the composition as a whole we can distinguish three main groups of ‘scenes’: a) a pair of bull-men or a single bull-man with a human figure placed on both sides of ‘ritual’ elements (standards, schematic gates, offering tables, podia) or animals — simple worshipping scene30 b) bull-men related to ‘ritual’ elements or animals together with other figures involved in different relationships — complex worshipping scene31 c) bull-man as centre of the representation —worshipped (?) bull-man’s scene32 It seems that bull-men participate quite exclusively in ‘ritual’ scenes, not only because of their own ‘religious’ connotations but also because they always occur together with peculiar cultic equipment. The relationship between bull-men and superimposed bulls or bull and gazelle (Kjærum 1983: nos. 247-249) [Fig. 4c-d] again suggests the complex pattern of ideological meanings which hides behind the animal repertoire in Dilmun stamp seals.33 Two crossed bull-men with raised hands stand across a net podium on a unique seal from Failaka (Kjærum 1983: no. 261) [Fig. 4e], according to a figurative scheme especially known for antelopes and gazelles (Kjærum 1983: nos. 256-260; al-Sindi 1994: nos. 195-197) and possibly related to the Syro-Anatolian milieu where crossed figures (also bull-men) are quite common (cf. Özgüç 1991: Fig. 31; Teissier 1994: nos. 157-165, 312, 341, 387-389).

Bull-men occur in 31 Dilmun seals, of which 26 come from Failaka, only 4 from Bahrain and one from Ur as mentioned above.26 Single figures are attested on ten seals, couples on 21 seals. Generally speaking, depictions of the bull-man occur from the Early Dynastic II onwards, with a human torso, bull’s legs and tail, human face with bull’s ears and horns (in profile or full-face), sometimes ithyphallic (cf. Black & Green 1992: 48-49). However, the Dilmunite bull-man is always represented in profile, apparently beardless with horns that resemble those of the god’s crown. Kjærum has pointed out the existence of two different styles of execution, reflecting a chronological evolution: the stamp seals found in the lower levels at Failaka (style IA) bear taller figures with a peculiar ‘sacklike’ body, while seals related to style IB show more linear figures with slender bodies (Kjærum 1980: 46).

Another meaningful seal is engraved with a schematic shrine or door with symbols inside (hatched podium, sunring, hatched lentoid, net podium), flanked by a bull-man and a garbed man grasping the door-frame (Kjærum 1983: no. 51) [Fig. 4f]. Rectangular structures appear on 9 seals (Kjærum 1983: nos. 51-54, 126; al-Sindi 1994: nos. 202-203, 205, 263): they have symbols or human figures within and they can be considered schematic gates or chapels/ shrines, without doubt linked with peculiar ritual functions, as revealed also by astral symbols, mythological figures (serpent monsters or bull-men) and worshippers on their sides.34 In Mesopotamia we find a winged ‘shrine’ placed on the back of a bull during the Old Akkadian period (Boehmer 1965: 105-109, nos. 589-619) and a representation of a ‘gate’ with heroes, bull-men or deities as guardians, through which the sun-god Shamash emerges at his rising from the mountains (van Buren 1947; Boehmer 1965: nos. 392-428). An original elaboration of the first theme can be recognised in the Old Syrian culture, where it occurs, sometimes winged and/or placed on bulls, together with the nude goddess flanked by bullmen (cf. Amiet 1960: 224-227; Matthiae 1987: 476-480).35 Therefore we can argue for a transmission of the sacred

The bull-man interacts with humans, animals and objects: Relationship with humans: direct: bull-man grasping/touching a naked male figure (Beyer 1986: no. 170) direct: bull-man grasped/touched by a naked male figure (Kjærum 1983: no. 208) [Fig. 3b] Relationship with animals: direct: bull-man grasping an animal [Figs. 3b-f, 4a, c-d]27 indirect: bull-man standing ‘above’ an animal [Figs. 3d, 5c-d]28 242

A

B

C

D

E

F

Fig. 4: Dilmun seals with bull-men. A) Kjærum 1983: no. 93; B) Kjærum 1983: no. 70; C) Kjærum 1983: no. 248; D) Kjærum 1983: no. 249; E) Kjærum 1983: no. 261; F) Kjærum 1983: no. 51.

‘gate’ or ‘chapel’ from Mesopotamia from the Akkadian period onwards, although we cannot exclude a link with the Old Syrian milieu.

transmission of the motif from the Syro-Mesopotamian regions, albeit we cannot also exclude a socio-economic reason which may link a specific group of Dilmunites residing at Failaka with the bull-man’s iconography. Since the foundation levels at Failaka roughly correspond to Qala’at IIb-c, we can conclude that the first appearance of the motif in Dilmun dates at the latest from the middle Isin-Larsa period, at the end of the 20th century B.C. In this case the Failaka settlement should be contemporary to late Qala’at IIb (cf. Højlund 1987: 157-158, fig. 707).

The balanced presence of divine figures on seals from Failaka and Bahrain cannot be matched with that of the bull-man. Only four seals from Bahrain attest the knowledge of this iconography in the Dilmunite heartland, whereas only a few examples were found at Failaka F3 and F6. This difference might be ascribed to a later 243

A

B

C

D

F

E

Fig. 5: Dilmun seals with bull-men. A) al-Sindi 1994: no. 115; B) al-Sindi 1994: no. 116; C) al-Sindi 1994: no. 117; D) Kjærum 1983: no. 115; E) Kjærum 1983: no. 116; F) Kjærum 1983: no. 141.

However it is equally reasonable to argue that the iconography was first introduced at Failaka and never became a figurative theme of the Bahrain seals, except for a few specimens (apparently three of style IA and one of style IB). It is certainly adopted at Failaka on ‘mature’ Dilmun seals during periods 2 and 3, until the beginning of Kassite period, and occurs also on style II seals.36 Bull-men are attested in Mesopotamian glyptic from the Early Dynastic II onwards (Karg 1984: 42-44). The iconographic elaboration probably happened at the end

of the 4th or at the beginning of the 3rd millennium B.C. in the Iranian milieu, where stamp and cylinder seals show hybrid creatures with mixed human and animal features since the prehistoric periods (cf. Amiet 1972: nos. 219-220, 1013-1017; 1986: 16-30). Bull-men were represented during the Early Dynastic period only in contest scenes together with rampant animals, the naked hero and the human-headed bull. A lenghty discussion on these figures has involved Near Eastern scholars, some proposing to identify Enkidu and 244

Gilgamesh with the bull-man and the hero with long hair with curls (cf. Afanesyeva 1971), others preferring to recognise in these figures different aspects of the god Dumuzi (Moortgat 1949). More recently a simplistic correlation between Early Dynastic supernatural beings and those known from mythological tales was submitted to a strong criticism (Lambert 1987), despite the unequivocal connection with the religious sphere.37 It is now widely accepted that the ‘nude hero’ must be considered a protective and beneficent deity, in later periods associated with Enki (Akkadian period) or Marduk (from the 2nd millennium B.C.), known by the name Lahmu (Black & Green 1992: 115). During the Akkadian period bull-men occur again in contest scenes, albeit some rare cylinders show the figure in relation to the sun-god Utu/Shamash (Collon 1987: no. 765; Amiet 1980: 39; Matthiae 1987: 481-482). A new elaboration of iconographic features and scenes in which bullmen were involved can be placed at the end of the 3rd millennium B.C., during the Ur III and Isin-Larsa periods. A ‘true’ horned tiara became the bull-man’s headdress and the figure is now shown also grasping standards, gate-posts, spears and carrying animal offerings.38 In the Syro-Anatolian regions, bull-men are frequently depicted involved in ‘ritual’ scenes, often as standardbearers, suggesting a more striking connection with the Arabian Gulf evidence (Özgüç 1965: 70-71; 1991: 308-309). The corpus of seal impressions from Kültepe karum II (ca. 1920-1850 B.C.) verifies the occurrence in the Anatolian, Syro-Cappadocian, Old Syrian, and Old Assyrian styles (Özgüç 1965; 1989; Leinwand 1992; Teissier 1994). The figure appears in the Old Babylonian group only in contest scenes and in the Old Assyrian group as a terminal element in introduction or supplication scenes, sometimes grasping a spear, or with both hands joined at the waist.39 A unique occurrence is a seal with a row of bearded bull-men in full-face holding a kind of crescent standard, apparently linked to the Assyrian style (Teissier 1994: no. 148). In the Anatolian group we quite frequently see bull-men (single or pairs) as standard-bearers, holding a flowing vase, struggling with animals, or carrying a bull.40 It is striking that the bull-man was adopted during the Old Anatolian period also in the local stamp seals, so far known especially from Kültepe, Karahöyük, Acemhöyük, Boghazköy and Alishar (Alp 1968; Özgüç 1980; von der Osten 1937: 210-229; Boehmer & Güterbock 1987: 1956).41 The quite heterogeneous Syro-Cappadocian (Syrian Colony) group often shows bull-men in different attitudes, pointing to the importance of the motif in the Old Syrian milieu (Pinnock 2000: 1402). They usually hold standards or spears, although they can also be placed on each sides of the nude goddess who is unveiling herself.42 Finally, this mythological being occurs in the mature Old Syrian glyptic from North Syrian workshops, mainly connected to the royal style of the Yamkhad-kingdom dating to the 18th century B.C.43 Moreover, the bull-man appears also in Old Syrian sculpture and relief, as testified by Ishtar’s stele, and is carved on basins and ritual bases from Tell Mardikh-Ebla dating from the Middle Bronze IIIA (ca. 2000-1700 B.C). The bas-reliefs show bull-men in full-face as apotropaic figures and, in the case of the stele

from Sanctuary G3, as guardians of Ishtar’s winged shrine placed on a bull (see Matthiae 1989: Pls. 121, 134-135; Matthiae, Pinnock & Scandone Matthiae 1995: nos. 291, 236, 238). Summing up, the bull-man’s role in Dilmun seals seems to be mainly ‘ritual’ or ‘cultic’. He was related to sacred paraphernalia such as standards, podia and altars, or to animals with a ‘symbolic’ meaning such as the snake. Moreover, he has a weaker ‘direct’ relationship with human figures. The main subjects of seals with bull-men are the worshipped ‘symbols’ or animals. A similar iconographic adaptation is never attested in Mesopotamia and infrequently in Syro-Anatolian regions; in the latter area we find bull-men with standards, albeit often as ‘secondary’ elements in ceremonial scenes with enthroned deities or kings. Generally speaking, it seems that there is a probable link with the Syro-Cappadocian evidence, which points out to a common ‘cultural’ sensibility towards specific artistic expression, rejecting the contest scene’s use in favour of ‘new’ religious functions. According to the distribution of seals with bull-men, we have seen that this theme was adopted, probably later in Dilmun, with respect to the iconography of gods and drinking figures. The offering table with bull hooves We would like to conclude this brief contribution with a discussion on what is certainly the most striking similarity between Dilmunite and Syro-Cappadocian glyptic. In a small group of stamp seals from Failaka a peculiar offering table with curved legs and bull’s hooves is visible (Kjærum 1983: nos. 163-168; Pic 1990: nos. 12-13?). All the specimens come from Failaka, where the motif is also attested in one seal of style II (Kjærum 1983: no. 43), probably dating from the Old Babylonian period. The table is formed by a slightly concave top, a central straight column, in one case moulded (Pic 1990: no. 12), and two legs with terminal bull hooves. Different kinds of symbols or objects are apparently placed on the table, and seated/standing garbed/naked figures or bull-men appear at its sides. Relationship with humans: direct: table grasped/touched by bull-men or humans (Kjærum 1983: no. 166; Pic 1990: no. 12) [Fig. 6c] Relationship with animals: direct: table grasped/touched by monkeys (Kjærum 1983: no. 168) [Fig. 6e] Relationship with objects: direct/indirect: symbols/objects apparently placed on the table [Fig. 6a-e]44 direct/indirect: table apparently placed on a ‘podium’ (Kjærum 1983: nos. 163, 166) [Fig. 6a, c] The overall representation often includes other symbols and animals (monkey, snake, plant, star and crescent, net podium), suggesting a ‘ritual’ character of the main scene, centred around the offering table. The figurative scheme with two bull-men or seated men flanking the table is 245

certainly linked with ceremonial activity in which the furniture fulfils a specific, pivotal role. We can distinguish between a ceremonial banquet with loaded tables, when the human figures hold cups, and a worship scene, when the figures ‘touch’ the table or raise arms toward it. Moreover, a seal from F3 (Kjærum 1983: no. 169) shows a naked standing man holding a table with the same shape of the offering tables but with human legs, repeating a figurative composition identical to the erotic scenes: it is a meaningful example of ritual elements combined together revealing a complex local ideology, often difficult to understand on the basis of the iconographic evidence. As correctly pointed out by Kjærum (1986: 272) this table resembles furniture which is very common in SyroAnatolian regions during the Middle Bronze Age, and known especially from cylinder seals but also from ivory objects, bas-reliefs and sculptures.45 The bottom of a basalt offering table was recently found at Tell Mardikh-Ebla, reused as a well-curb near the Temple P2 in the Sacred Ishtar Area (Matthiae 1994: 173-177, Figs. 3-4). It has established without doubt the existence of cultic furniture in temples of the Old Syrian period and the tripod shape of the equipment. The same furniture occurs on many cylinder seals in Syrian, SyroCappadocian and Anatolian styles,46 whereas it is virtually unknown from contemporary Mesopotamian glyptic.47 It must, therefore, belong to the Syrian milieu where it was used in ceremonial activities involving banquets and cultic libations. Several variants can be recognised from the glyptic evidence (Matthiae 1994: 175-176): the basic shape has a plain top, a central column diverging into two legs with bull’s hooves, and lateral supports connecting the legs with the top. The Eblaite basalt table demonstrates that the furniture may also have a solid base, which is lacking in the more schematic seal’s representation. The existence of similar tables in Dilmunite stamp seals testifies to a direct link to the Syrian region, albeit the iconography is different in the concave top and the lack of lateral supports. They are more similar to the motif known from ‘classic’ Old Syrian cylinder seals (ca. 18501720 B.C.),48 although it does not necessarily mean a ‘later’ transmission. It seems more probable that in the Gulf area an autonomous elaboration occurred during the 19th-18th centuries B.C., which could have been parallel to that of the Syrian milieu where this kind of offering table disappears at the end of the 18th century B.C. Dilmunite seal designs with offering tables might testify not only to the iconographic knowledge but also to a circulation of that type of ceremonial furnishing between Western Syria and Arabian Gulf, i.e. as real imports. From an artistic point of view the land of Dilmun again shows the trend to assimilate themes and figurative motifs pertaining to the ‘Amorite’ Western and Northern Syrian milieu dating from the very beginning of the 2nd millennium till the end of the 17th century B.C.

A

B

C

D

E Fig. 6: Dilmun seals with offering tables. A) Kjærum 1983: no. 163; B)

Concluding Remarks

Kjærum 1983: no. 165; C) Kjærum 1983: no. 166; D) Kjærum 1983: no. 167; E) Kjærum 1983: no. 168.

It seems that iconographic relationships with Syria and 246

Mesopotamia might be investigated by taking into account chronological differences and/or regional distinctions between material from Failaka and Bahrain. Some motifs strongly suggest an ‘earlier’ transmission and original elaboration in the Gulf area, others point to ‘later’ developments and, possibly, peculiar preferences of Dilmunites residing at Failaka. The inner articulation of the land of Dilmun certainly deserves more rigorous investigation, trying to reach a better understanding of socio-economic dynamics among the Failaka settlement, the coastal area from Kuwait to Tarut, and the Bahraini sites. The complex iconography of Dilmun seals might be analysed by looking for cultural links with the SyroMesopotamian world, but also for local elaboration related to highly specific ritual and ceremonial meanings. We have reviewed some of the most striking evidence of cultural Near Eastern connections, without exhausting the subject, which appears diversified and ‘stratified’ and begs an analytical and systematic study. We have omitted other iconographic elements, such as the whorl and radial composition of animal’s necks (cf. Porada 1971: 335, Boehmer 1986), the standards and podia, abstractions and pars pro toto (Kjærum 1986: 273), pointed out by several scholars, and instead focused on gods and bull-men to show that Dilmun imagery deals with these ‘classic’ Mesopotamian figures according to an original ideology, probably influenced by Western and North Syrian prototypes, according to a common spiritual sensibility.

3.

4.

Notes 1. The corpus of ‘Dilmun’ seals from Failaka and Bahrain comprises of several hundred specimens from temples, domestic structures, graves and public buildings (Kjærum 1983; Beyer 1986; 1989; Pic 1990; al-Sindi 1994). The typologic and stylistic criteria used for seal classification were singled out by Kjærum (1980), albeit not discussed in detail in the Failaka catalogue of glyptic materials from the Danish excavations (Kjærum 1983). 2. The presence of two different series was first recognized by Porada, who spoke of an ‘Earlier group’ for circular monofacial seals with a high button boss divided by one (exceptionally two or three) grooves, and a ‘Later group’ for circular monofacial seals with a broad domed boss divided as a rule by three thin parallel lines and four dot-in-circles (Porada 1971: 331-332, Figs. A-B); therefore the shape of the back is the feature adopted to classify the stamp seals, with more or less varied terminology: Potts (1990: 161, n. 57) introduced the terms ‘Persian Gulf’ for the earlier series and ‘Dilmun’ for the later series, finally changed to ‘Arabian Gulf type’ and ‘Dilmun type’ by Kjærum (1994: 319). A transitional group (protoDilmun type) can be also identified on iconographic/ stylistic grounds, with motifs characteristic of the Mature Dilmun glyptic and the style of the engraving still related to the Arabian Gulf series; some scholars have pointed out the unusual back of these seals, with shape and decoration not yet standardized (i.e. Beyer 1989: 137; Potts 1990: 168). However, Kjærum

5.

6. 7.

247

has recently shown that no less than 57% of the ‘proto-Dilmun’ stamps had the reverse ornament identical to the later group, the only definite criterion being the stylistic one (1994: 345). The archaeological context of most of the seals from the Danish excavations at Failaka F3 and F6 is quite difficult to establish on the base of the absolute level and square/trench references. In some cases it is possible to use the information given in the pottery publication (Højlund 1987: 130-137) but the results seem to be uncertain and debatable; however, in general, the distribution of Dilmun stamp seals in relation to the absolute level shows their continuity till Period 4A (i.e. the specimens from F3 Trench D above level 7.00: Kjærum 1983: nos. 18, 94, 175, 189, 230, 257). For an updated evaluation of the Failaka sequence see now Højlund & Andersen 1994: 173, 197; cf. also Højlund 1989: Fig. 1. The date of these productions is still controversial: Denton has suggested, starting from the evidence from cemeteries at al-Maqsha, al-Hajjar and Hamad Town, a possible initial dating for the ‘Style III’ seals to the middle/late Old Babylonian period, trying also to reduce the presumed ‘gap’ between Qala’at period IIc and IIIa (Denton 1997; 1999); Kjærum seems to connect this style with Failaka 3A-B periods (Kjærum 1994: 348, Fig. 1758); the writer prefers to date this group to 1650/1600-1400/1350 BC (Failaka 3B-4A), exploring a direct Kassite influence in ‘cultural’ and artistic meanings of glyptic representation; during the late Old Babylonian period this was instead elaborated into the ‘style II’ type, related to monofacial and bifacial specimens with a characteristic drilled style (Peyronel 1997). As regards Mesopotamian periodisation, the work of Gasche —proposing a revision of the traditional ‘middle chronology’ with dates ca. one hundred years lower— has recently raised new questions on the effectiveness of Near Eastern ‘conventional’ chronological framework (Gasche et al. 1998). On the introduction/presentation scene see Franke 1977; Winter 1986; Haussperger 1991. A sort of ‘introduction’ scene could perhaps be identified on one or two seals from Failaka (Kjærum 1983: nos. 96, 132), completely emptied of Mesopotamian canonical figurative parameters; several specimens show crescent or sun-crescent standards with bullmen or human figures apparently ‘worshipping’ these symbols (holding or stretching the arms towards the standard). Similar standards are known in the Mesopotamian glyptic of the Isin-Larsa and Old Babylonian periods, albeit never attested in the same kind of relations with human or divine personages (cf. Collon 1986: nos. 110, 143, 279-280, 284, 314-315, 556, 623). On the contrary more direct parallels can be drawn for style II and especially for style III seals, which show i.e. worshippers, suppliant goddess, bald priests, cuneiform signs or ‘pseudo-writing’ symbols, standards, with an almost complete dominance of the ‘human/divine’ world (in the style III) in a changed way of artistic/cultural communication (Peyronel

1997). 8. However, it must be remembered that the zenith of Indus Civilisation should be dated to 2400-2000 B.C., whereas the phenomenon of a sharp localisation during the first centuries of the 2nd millennium B.C. drastically reduced the trade relations towards the West and across the Indian Ocean (cf. Kenoyer 1991; Shaffer 1992; Jansen 1993). 9. The wide web of intercultural contacts during the second half of the 3rd millennium B.C. is well attested, for example, by the distribution of chlorite carved vessels (Lamberg-Karlovsky 1988) and by imports or objects with Harappan influence in Mesopotamia (i.e. square or circular stamp seals, etched carnelian beads, weights, clay figurines, dice, kidney-shaped inlays) (Possehl 1996). If it is very likely that some people from Meluḫḫa had settled in the alluvium (Brunswig, Parpola & Parpola 1977), but it is much more difficult to establish the presence of Mesopotamians in the Indus Valley on the basis of presumed Near Eastern ‘cultural’ traits in a handful of objects from Harappan cities (see, however, During Caspers 1971; 1972; 1979; 1982). 10. Some scholars have suggested that the Amorite ethnic element was a key factor in the rise of the Dilmun State at the beginning of the 2nd millennium B.C. (Højlund 1989; 1993). Direct contact with the Syrian kingdoms are still elusive: the only evidence so far comes from Mari, where some cuneiform texts attest to ‘diplomatic’ contact between Mari, Shubat-Enlil and Dilmun during the reign of Shamshi-Adad (Groneberg 1992; Eidem & Højlund 1993: 442-444; van Koppen 1997). 11. The Harappan roots of the Dilmunite glyptic can still be recognised in the multiple parallel lines of the animal’s necks, and in the fact that the bull is usually placed in the lower part of the seal, and in the motif of the water-carrier which probably derived from an Indus sign. 12. Male naked figure: Kjærum 1983: nos. 185-186, 274; alSindi 1994: nos. 19, 23; Male garbed figure: al-Sindi 1994: nos. 18, 20. 13. Kjærum 1983: nos. 81, 212; al-Sindi 1994: nos. 24, 5758. On Kjærum 1983: no. 81, two peculiar gods are holding the forelegs of a double gazelle with jointed backs. 14. Kjærum 1983: nos. 185-186; al-Sindi 1994: nos. 17-18, 23, 57-58. 15. Kjærum 1983: nos. 185-186; al-Sindi 1994: nos. 17-20. On al-Sindi 1994: no. 23, the god is depicted holding a cup in his raised hand. 16. All the specimens except for Kjærum 1983: no. 212 [Fig. 2c], where the god is standing and holds two opposed gazelles from the neck. 17. Drinking figures are attested on more than 40 stamp seals: they are naked or garbed, sitting or standing, and drink from separate or common jars with long straight or sometimes curved straws. We can single out a true symposium ‘scene’ or drinking iconography as part of a more articulated figurative pattern. See Kjærum 1983: nos. 95, 100, 158, 170-179, 181, 182,

18. 19.

20.

21.

22. 23.

24.

248

184-186, 188-189, 199, 207; Beyer 1986: nos. 167-168 (Failaka); al-Sindi 1994: nos. 9, 12, 18-22, 222; Crawford & Woodburn 1994: Fig. 19 (Bahrain); Peyronel 2000: nos. 4.13, 4.15 (Ur). Barrelet 1968: Pls. 50 no. 527, 63 no. 675, 71 no. 744; Opificius 1961: 166-168, nos. 604-614. Kjærum 1983: nos. 269-270; al-Sindi 1994: nos. 221226; Crawford & Matthews 1995: 57 no. 20. Seals depicting a sexual act are quite rare in Dilmun (two examples from Failaka and seven from Bahrain); we can also include in this group two interesting specimens (from Ur and from the Erlenmeyer collection) with the female figure represented in the same posture as the erotic scenes (Peyronel 2000: no. 4.13; Erlenmeyer & Erlenmeyer 1966: Figs. 17-18). We can distinguish two basic figurative patterns: the first with the women bent down with the body parallel to the ground line and with the phallic man behind her, representing coitus a tergo (Kjærum 1983: no. 269; cf. also Peyronel 2000: no. 4.13); the second —more common— with the usually nude (in one case garbed) phallic man standing in front of the woman, who has widely spread legs and hands on feet. The last motif appears less realistic but it gives a better lay-out on the stamp seals, where the women’s bodies form a triangular space delimited by arms and legs opposed to the axiality of the male figures. In Kjærum’s opinion (1994: 331) it represents a coitus with the woman in a superior position, but this hypothesis seems to be contradict by the fact that the man is standing on a podium in the Qala’at example (as Kjærum himself pointed out correctly) and by men holding gatesymbols, which again suggest a free-standing posture. We would like to emphasise that the findspot of 3 out of the 9 specimens can be related to sacred buildings (Barbar, Saar Temple, Failaka F3), like most of the clay moulded plaques from the Old Babylonian period in Mesopotamia, suggesting a possible link with ritual involving sacred prostitution (see Pinnock 1996: 2528-2529). Buchanan 1966: nos. 841-843; 1981: no. 1175; Teissier 1984: nos. 352-359; 1994: nos. 322-327, 517-520. It must be underlined that on these seals only one personage is drinking and that the straw is always of the curved-type. A differentiated pantheon can be singled out in artistic production from the very end of the Early Dynastic III period and during the Akkadian period (cf. Amiet 1980; Boehmer 1986; Lambert 1997; for the previous periods see Amiet 1986). For deities in the Old Babylonian glyptic see now Braun-Holzinger 1996. Several temples are identified on Bahrain (Barbar, Diraz, Saar) and Failaka (F3 and F6) (cf. Højlund 1992; Peyronel 2003). Nashef 1984: 7-8. On the controversial location of Agarum and the relation between Dilmun and Agarum see Glassner 1984: 47-48; Nashef 1984: 1, 11, nn. 4-4a. Another seal (Kjærum 1983: no. 54, cf. infra) shows a naked figure and a branch within a rectangular

25.

26.

27.

28. 29.

30. 31. 32. 33.

34.

‘shrine’, surmounted by a sun-crescent and flanked by horned dragons and garbed seated figures. This peculiar ‘religious’ scene could suggest a divine character of the personage inside the ‘temple-niche’ and, therefore, a link between a Dilmunite god and the schematic plant. It is possible that one of Inzak’s symbols could have been a date-palm branch, but there is no sure proof of this sense. The only piece of evidence is the socalled Durand stone from Bahrain engraved with the inscription ‘palace of Rimum, servant of Inzak of Agarum’ close to a incised schematic branch (Butz 1983). Nashef’s claim of a seal with a legend bearing a dedication to Enki together with a date-palm branch (Kjærum 1983: no. 350) was rejected by J-J. Glassner, who read the cuneiform signs a-na dPA-NI-PA (Nashef 1984: no. 16, contra Glassner 1984: no. 5). Failaka: Kjærum 1983: nos. 51, 70, 93, 115-119, 121122, 133, 141-142, 166-167, 208, 247-249, 261, 273, 276, 282; Beyer 1986: nos. 170, 173; Pic 1990: no. 14; Bahrain: al-Sindi 1994: nos. 115-117, 254; Ur: Peyronel 2000: no. 4.14. Gazelle: Kjærum 1983: nos. 121, 276; bull: Kjærum 1983: no. 249; al-Sindi 1994: no. 254; bucranium: Kjærum 1983: no. 93; snake: Kjærum 1983: no. 122; Beyer 1986: no. 173; composite animal: Kjærum 1983: no. 273. Antelope: Kjærum 1983: no. 261; al-Sindi 1994: no. 116; bull: Kjærum 1983: nos. 115, 122. Hatched podium: Kjærum 1983: nos. 70, 248; gate with symbols: Kjærum 1983: no. 51; standards: Kjærum 1983: nos. 115-118, 141; Beyer 1986: no. 170; Pic 1990: no. 14; al-Sindi 1994: nos. 115-117; offering table: Kjærum 1983: no. 166; branch: Kjærum 1983: nos. 142, 167. Kjærum 1983: nos. 51, 115-119, 121-122, 141-142, 166, 167, 273, 276; Beyer 1986: no. 170; Pic 1990: no. 14; alSindi 1994: nos. 115-117. Kjærum 1983: nos. 70, 133, 247-249, Beyer 1986: no. 173; Peyronel 2000: no. 4.14; al-Sindi 1994: no. 254. Kjærum 1983: nos. 93, 208, 261. Several elements underline this ritual function: animals placed on altars/podia or at each sides of standards, animals touched/grasped by gods, animals surmounted by astral symbols. Moreover, the bucranium, which occur on more than 30 seals (Kjærum 1983: nos. 13, 68-69, 72, 75, 86, 93, 117, 150, 154, 174, 176, 184, 206, 258, 296), has certainly a symbolic value linked with local cult (Kjærum 1986: 273). The motif has no parallel in Mesopotamian glyptic and resembles only in a general way the bull-heads from Cappadocian and Old Syrian glyptic (cf. Otto 2000: 264). Kjærum (1994: 333-334) considers the shield-like object (visible on more than 50 stamp seals from Failaka and Bahrain) as a symbol of the schematic sacred gate: this assumption is based on two seals where it is represented on both sides of a standing garbed man or placed standing on an element of the same shape (Kjærum 1983: nos. 66-67). Other scholars maintain the shield-hypothesis (Hallo & Buchanan 1965: 205, n. 16; Porada 1971: 337; Beyer 1989: 147, 150), especially in the case of objects which are taller and plain (called by Kjærum ‘notched podia’; cf. Kjæ-

35.

36.

37. 38.

39. 40. 41. 42. 43. 44.

45.

46.

47.

249

rum 1983: nos. 75-76; al-Sindi 1994: nos. 209-210). In our opinion the earlier glyptic material from Bahrain seems to confirm the latter claim, although it is quite reasonable to consider the object as ritual/ceremonial equipment released from any war-like connotation. Another god which appears inside a sacred structure is Enki/Ea, identified by streams of water around his shrine or enthroned and flanked by nude heroes holding gate-posts (Boehmer 1965: nos. 488-489, 500501, 518-520, 522-523; Barrelet 1970). Since Enki’s cult might be attested to in Dilmun, we cannot exclude a connection between the deity of the Abzu and the presumed shrines of stamp seals. Kjærum 1983: nos. 43, 44, 120, 143-144, 304, 307; Pic 1990: no. 21. Finally, an Old Babylonian cylinder seal with a contest scene formed by three pair of figures, among which is a bull-man fighting against a rampant lion, represents ‘direct’ evidence for the transmission of the iconography to the land of Dilmun (Kjærum 1983: no. 374). See the still enlightening discussion on the problem in Frankfort 1939: 62-67. See i.e. Porada 1947: nos. 370, 384, 421; Collon 1986: 41, nos. 21, 89, 144, 152, 155, 379, 460. For similar subjects on terracotta plaques cf. Opificius 1961: nos. 402420. Cf. Teissier 1994: nos. 136-139, 145-147, 156, 190 (Old Assyrian), 586, 615, 632, 636-639, 641, 643 (Old Babylonian). Cf. Özgüç 1965: Pls. I.1, II.7, XV.46, XIX.56, XX.61, XXI.63, XXVI.78; Teissier 1994: nos. 330-331, 335, 342, 349. Cf. in particular von der Osten 1937: Fig. 250.d745 (Alishar); Özgüç 1980: 75, Fig. III.41 (Acemhöyük). Cf. Teissier 1994: nos. 467, 478-481, 496-500, 502-503, 508, 525, 529b, 531, 541-542, 546-547, 550, 551-554, 564-566. See i.e. Collon 1981: nos. 12-13, 21; Buchanan 1981: nos. 1202, 1239; Teissier 1984: no. 543. Simple dots: Kjærum 1983: no. 163; flat breads and fish: Kjærum no. 165; bird: Kjærum no. 166; astral symbols: Kjærum nos. 167, 169; low hatched podium?: Pic 1990: no. 12. The evidence from carved reliefs includes two stelae from Hama and Furayjah (Pinnock 1992: 110-112), two ritual basins from temples B1 and D at Ebla (Matthiae, Pinnock & Scandone Matthiae 1995: nos. 290291), and the so-called ivory talisman from the “Tomb of the Lord of the Goats” at Ebla (Matthiae, Pinnock, Scandone Matthiae 1995: no. 470). Cf. Teissier 1984: nos. 350-353, 359, 361 (Archaic Old Syrian), 400, 416, 417, 418, 427 (Syro-Cappadocian), 459, 462, 464 (Mature Old Syrian); 1994: nos. 96, 192, 209, 219, 241, 269 (Old Assyrian group), 290, 291, 305, 320-322, 328, 331, 336, 347-348 (Anatolian), 466, 491, 515, 518, 541, 551 (Syro-Cappadocian), 586, 593-594, 626 (Babylonian Provincial). Kjærum (1986: 272, n. 27) has pointed out a unique tablet in the Yale Babylonian Collection with a seal impression belonging to Akalla, ensi of Umma under

ology and Early History of Bahrain. (BBVO 2). Berlin. Collon, D. 1981. The Aleppo Workshop. Ugarit-Forschungen 13: 3343. ———. 1982. Catalogue of the Western Asiatic Seals in the British Museum. Cylinder Seals II: Akkad, Post-Akkadian and Ur III Periods. London. ———. 1986. Catalogue of the Western Asiatic Seals in the British Museum. Cylinder Seals III: Isin-Larsa and Old Babylonian Periods. London. ———. 1987. First Impressions. Cylinder Seals in the Ancient Near East. London. Crawford, H. & R. Matthews. 1995. Seals and Sealings: Fragments of Art and Administration. Pp. 47-58 in H. Crawford, R. Killick & J. Moon (eds) The Dilmun Temple at Saar. Bahrain and its Archaeological Inheritance. London. Crawford, H. & M. Woodburn. 1994. London-Bahrain Archaeological Expedition: 1991-2 Excavations at Saar. Arabian Archaeology and Epigraphy 5: 89-105. Denton, B.E. 1994. Pottery, Cylinder Seals, and Stone Vessels from the Cemeteries of al-Hajjar, al-Maqsha and Hamad Town. Arabian Archaeology and Epigraphy 5: 121-151. ———. 1997. ‘Style III’ Seals from Bahrain. Arabian Archaeology and Epigraphy 8: 174-189. ———. 1999. More Pottery, Seals and a ‘face-pendant’ from Cemeteries on Bahrain. Arabian Archaeology and Epigraphy 10: 134-160. During Caspers, E.C.L. 1971. New Archaeological Evidence for Maritime Trade in the Persian Gulf during the Late Protoliterate Period. East and West 21: 21-44. ———. 1972. Harappan Trade in the Arabian Gulf in the Third Mill. B.C. Mesopotamia VII: 167-191. ———. 1979. Sumer, Coastal Arabia and the Indus Valley in the Protoliterate and Early Dynastic Eras. Journal of Economic and Social History of the Orient XXII: 121-135. ———. 1982. Sumerian Trade and Businessmen Residing in the Indus Valley Cities: A Critical Assessment of the Archaeological Evidence. Annali d’Istituto Universitario Orientale di Napoli 42: 337-380. Erlenmeyer, H. & M.-L. Erlenmeyer. 1966. Über Beziehungen des Alten Orients zu den Frühindischen Stadtkulturen. Archiv für Orientforschung 21: 21-31. Eidem, J. & F. Højlund. 1993. Trade or Diplomacy? Assyria and Dilmun in the Eighteenth Century B.C. World Archaeology 24: 441-448. Franke, J.A. 1977. Presentation Seals of the Ur III/Isin-Larsa Period. Pp. 15-23 in R.D. Biggs & McG. Gibson (eds) Seals and Sealing in the Ancient Near East. (BiMes 7). Malibu. Gadd, C.J. 1932. Seals of Ancient Indian Style Found at Ur. Proceedings of the British Academy XVIII: 191-210. Gasche, H., J.A. Armstrong, S. Cole & V. Gurzadyan. 1998. Dating the Fall of Babylon. A Reappraisal of Second-Millennium Chronology. (MHEM 4). Ghent & Chicago. Glassner, J-J. 1984. Inscriptions cunéiformes de Failaka. Pp. 31-50 in J.-F. Salles (ed) Failaka. Fouilles Françaises 1983. (TMO 9). Lyon. ———. 1996. Dilmun, Magan and Meluhha: Some Observations on Language, Toponymy, Anthroponymy and Theonymy. Pp. 235-248 in J. Reade (ed) The Indian Ocean in Antiquity. London. Glob, P.V. 1954. Bahrain, Island of the Hundred Thousand Burial Mounds. Kuml 1954: 92-105. ———. 1959. Archaeological Investigations in Four Arab States. Kuml 1959: 238-239. de Graeve, M.-C. 1982. A Drinking Scene on a Late Akkadian Seal. Pp. 17-24 in J. Quaegebeur (ed) Studia Paulo Naster Oblata II: Orientalia Antiqua. (OLA 13). Leuven. Groneberg, B. 1992. Mari et le Golfe Arabico-Persique. Pp. 69-80 in J.-M. Durand (ed) Florilegium Marianum. Recueil d’études en l'honneur de Michel Fleury. (Mémoires de N.A.B.U. 1). Paris. Hallo, W.W. & B. Buchanan. 1965. A “Persian Gulf” Seal on an Old

the reign of Ibbi-Sin of Ur (Buchanan 1981: no. 652). We can add one Elamite sealing from the same collection (Buchanan 1981: no. 1093A) and three postAkkadian/Ur III seals from the British Museum (Collon 1982: nos. 299, 415, 428). 48. Cf. Otto 2000: 265 and for some examples Porada 1948: nos. 913, 915, 944; Buchanan 1966: no. 856; Teissier 1984: nos. 459, 464. References Afanasyeva, V.K. 1971. Gilgameš and Enkidu in Glyptic Art and in the Epic. Klio 53: 60-75. Alp, S. 1968. Zylinder und Stempelsiegel aus Karahöyük bei Konya. Ankara. Alster, B. 1983. Dilmun, Bahrain and the Alleged Paradise in Sumerian Myth and Literature. Pp. 39-74 in D.T. Potts (ed) Dilmun. New Studies in the Archaeology and Early History of Bahrain. (BBVO 2). Berlin. Amiet, P. 1960. Notes sur le répertoire iconographique de Mari à l’époque du palais. Syria 37: 215-232. ———. 1972. Glyptique Susienne. Des origines à l’époque des Perses Achéménides. (MDAI 43). Paris. ———. 1980. The Mythological Repertoire in Cylinder Seals of the Agade Period. Pp. 35-60 in E. Porada (ed) Ancient Art in Seals. Princeton. ———. 1986. Le problème de l’iconographie divine en Mésopotamie dans la glyptique antérieure à l’époque d’Agadé. Contributi e Materiali di Archeologia Orientale 1: 1-66. Barrelet, M.T. 1954. Figurines et reliefs en terre cuite de la Mésopotamie antique. Paris. ———. 1970. Etude de glyptique akkadienne. L’imagination figurative et le cycle d’Ea. Orientalia 39: 213-251. Beyer, D. 1986. Les sceaux. Pp. 89-103 in Y. Calvet & J.-F. Salles (eds) Failaka. Fouilles Françaises 1984-1985. (TMO 12). Lyon. ———. 1989. The Bahrein Seals (Early Dilmun Period to Tylos Period). Pp. 135-166 in P. Lombard & M. Kervran (eds) National Museum Archaeological Collections. Vol. 1: A Selection of Pre-Islamic Antiquities from the Excavations 1954-1975. Manama. Bibby, T.G. 1957. The Hundred-Metre Section. Kuml 1957: 128-163. ———. 1958. The “Ancient Indian Style” Seals from Bahrain. Antiquity 32: 243-244. Boehmer, R.M. 1965. Die Entwicklung der Glyptik während der Akkad-Zeit. (UAVA 4). Berlin. ———. 1986. Einflüsse der Golfglyptik auf die Anatolische Stempelglyptik zur Zeit der Assyrischen Handelniederlassungen. Bagdhader Mitteilungen 17: 292-298. Boehmer, R.M. & H. Güterbock. 1987. Die Glyptik von Böğazköy. Vol. 2: Die Glyptik aus dem Stadtgebiet von Boğazköy 1931-1939, 19521978. Berlin. Braun-Holzinger, E.A. 1996. Altbabylonischen Götter und ihre Symbole ‒ Benennung mit Hilfe der Siegellegenden. Baghdader Mitteilungen 27: 235-362. Brunswig, R.H., S. Parpola & A. Parpola. 1977. The Meluḫḫa Village: Evidence of Acculturation of Harappan Traders in Late Third Millennium Mesopotamia? Journal of Economic and Social History of the Orient XX: 129-165. Buchanan, B. 1966. Catalogue of the Ancient Near Eastern Seals in the Ashmolean Museum. Vol. I: Cylinder Seals. Oxford. ———. 1967. A Dated Seal Impression Connecting Babylonia and Ancient India. Archaeology 20: 104-107. ———. 1981. Early Near Eastern Seals in the Yale Babylonian Collection. New Haven. van Buren, E.D. 1947. The Guardians of the Gate in the Akkadian Period. Orientalia 16: 312-332. Butz, K. 1983. Zwei kleine Inschriften zur Geschichte Dilmuns. Pp. 117-126 in D.T. Potts (ed) Dilmun. New Studies in the Archae-

250

Babylonian Mercantile Agreement. Pp. 199-209 in H.G. Güterbock & T. Jacobsen (eds) Studies in Honor of Benno Landsberger on his Seventy-Fifth Birthday April 21, 1965. (AS 16). Chicago. Haussperger, M. 1991. Die Einführungsszene. Entwicklung eines mesopotamischen Motivs von der altakkadischen bis zum Ende der altbabylonischen Zeit. (Münschener Vorderasiatische Studien XI). Munich & Vienna. Højlund, F. 1987. Failaka/Dilmun: the Second Millenium Settlements. Vol. 2: The Bronze Age Pottery. (JASP XXVII:2). Munskgaard. ———. 1989a. The Formation of the Dilmun State and the Amorite Tribes. Proceedings of the Seminar for Arabian Studies 19: 4564. ———. 1989b. Some New Evidence of Harappan Influence in the Arabian Gulf. Pp. 49-53 in K. Frifelt & P. Sørensen (eds) South Asian Archaeology 1985. (SIAS OP 4). London. ———. 1992. Holy Architecture in Bronze Age Bahrain. Dilmun 15: 73-79. ———. 1993. The Ethnic Composition of the Population of Dilmun. Proceedings of the Seminar for Arabian Studies 23: 45-53. Højlund, F. & H.H. Andersen (eds). 1994. Qala’at al-Bahrain. Vol. 1: The Northern City Wall and the Islamic Fortress. (JASP XXX:1). Munskgaard. Jansen, M. 1993. Pre-, Proto-, Early-, Mature-, Urban-, Late-, Post-Harappan: Linearity, Multi-Linearity: “Babylonian Language Confusion” or Ideological Dispute? Pp. 135-148 in A.J. Gail & G.J.R. Mevissen (eds) South Asian Archaeology 1991. Stuttgart. Karg, N. 1984. Untersuchungen zur älteren frühdynastichen Glyptik Babyloniens. Mainz. Kenoyer, J.M. 1991. The Indus Valley Tradition of Pakistan and Western India. Journal of World Prehistory 5: 331-385. Kjærum, P. 1980. Seals of “Dilmun-Type” from Failaka, Kuwait. Proceedings of the Seminar for Arabian Studies 10: 45-54. ———. 1983. Failaka/Dilmun. The Second Millennium Settlements. Vol. 1: The Stamp and Cylinder Seals. (JASP XXVII:1). Munskgaard. ———. 1986. The Dilmun Seals as Evidence of Long-Distance Relations in the Early Second Millennium B.C. Pp. 269-277 in Shaikha H.A. Al-Khalifa & M. Rice (eds) Bahrain through the Ages: The Archaeology. London. ———. 1994. Stamp-seals. Pp. 319-350 in F. Højlund & H.H. Andersen (eds) Qala’at al-Bahrain. Vol. 1: The Northern City Wall and the Islamic Fortress. (JASP XXX:1). Munskgaard. van Koppen, F. 1997. L’expédition à Tilmun et la révolte des bédouins. M.A.R.I. 8: 417-429. Lamberg-Karlovsky, C.C. 1988. The ‘Intercultural Style’ Carved Vessels. Iranica Antiqua 23: 45-73. Lambert, W.G. 1987. Gilgamesh in Literature and Art: The Second and First Millennia. Pp. 37-52 in A.E. Farkas (ed) Monsters and Demons in the Ancient and Medieval Worlds. Papers Presented in Honor of Edith Porada. Mainz am Rhein. ———. 1997. Sumerian Gods: Combining the Evidence of Texts and Art. Pp. 1-10 in I.L. Finkel & M.J. Geller (eds) Sumerian Gods and their Representations. (Cuneiform Monographs, 7). Groningen. Leemans, W.F. 1960. Foreign Trade in Old-Babylonian Period. Leiden. ———. 1968. Old Babylonian Letters and Economic History. Journal for the Economic and Social History of the Orient XI: 171-226. Leinwand, N. 1992. Regional Characteristics in the Styles and Iconography of the Seal Impressions of Liv. II at Kultepe. Journal of the Ancient Near Eastern Society XXI: 141-172. Matthiae, P. 1987. Una stele paleosiriana arcaica da Ebla e la cultura figurativa della Siria attorno al 1800 a.C. Sienze dell’Antichità 1: 447-495. ———. 1989. Ebla. Un impero ritrovato. Dai primi scavi alle ultime scoperte. Turin. ———. 1994. Old Syrian Basalt Furniture from Ebla Palaces and Temples. Pp. 167-177 in P. Calmeyer, K. Hecker, L. Jakob-Rost & C.B.F. Walker (eds) Beiträge zur Altorientalischen Archäologie und Altertumskunde. Festschrift für Barthel Hrouda zum 65. Geburt-

stag. Wiesbaden. Matthiae, P., F. Pinnock & G. Scandone Matthiae (eds) 1995. Ebla. Alle origini della civiltà urbana. Trenta anni di scavi in Siria dell’Università di Roma ‘La Sapienza’. Milan. Moortgat, A. 1940. Tammuz: der Unsterblichkeitsglaube in der altorientalischen Bildkunst. Berlin. Nashef, K. 1984. The Deities of Dilmun. Akkadica 38: 1-33. Opificius, R. 1961. Das altbabylonische Terrakottarelief. (UAVA 2). Berlin. Oppenheim, A.L. 1954. The Seafaring Merchants of Ur. Journal of the American Oriental Society 74: 6-17. von der Osten, H.H. 1937. The Alishar Hüyük. Season of 1930-32. Part III. (OIP 28) Chicago. Otto, A. 2000. Die Entstehung und Entwicklung der Klassisch-Syrische Glyptik. (UAVA 8). Berlin & New York. Özgüç, N. 1965. The Anatolian Group of Cylinder Impressions from Kültepe. (Türk Tarıh Kurumu, Series V/22). Ankara. ———. 1968. Seal and Seal Impressions of Level Ib from Karum Kanish (Türk Tarıh Kurumu, Series V/25). Ankara. ———. 1980. Seal Impressions from the Palaces at Acemhöyük. Pp. 61-100 in E. Porada (ed) Ancient Art and Seals. Princeton. ———. 1991. The Composite Creatures in Anatolian Art During the Period of Assyrian Trading Colonies. Pp. 293-317 in M. Mori, H. Ogawa & M. Yoshikawa (eds) Near Eastern Studies Dedicated to H.I.H. Prince Takahito Mikasa on the Occasion of His Seventy-Fifth Birthday. Wiesbaden. Parpola, A. 1994. Harappan Inscriptions. Pp. 304-315 in F. Højlund & H.H. Andersen (eds) Qala’at al-Bahrain. Vol. 1: The Northern City Wall and the Islamic Fortress. (JASP XXX:1). Munskgaard. Peyronel, L. 1997. La glittica a stampo dilmunita in età paleobabilonese classica e cassita. Contributi e Materiali di Archeologia Orientale 7: 407-435. ———. 2000. Sigilli harappani e dilmuniti dalla Mesopotamia e dalla Susiana. Note sul commercio nel Golfo Arabo-Persico tra III e II Mill. a.C. Vicino Oriente 12: 175-240. ———. 2003. L’architettura sacra del Paese di Dilmun. I luoghi di culto nel Golfo Arabo-Persico dell’Età del Bronzo. Contributi e Materiali di Archeologia Orientale 9: 269-338. Pic, M. 1990. Quelques éléments de glyptique. Pp. 125-140 in Y. Calvet & J. Gachet (eds) Failaka. Fouilles Françaises 1986-1988. (TMO 18). Lyon. Pinnock, F. 1992. Una riconsiderazione della stele di Hama 6B599. Contributi e Materiali di Archeologia Orientale 4: 101-121. ———. 1994. Consideration on the “Banquet Theme” in the Figurative Art of Mesopotamia and Syria. Pp. 15-26 in L. Milano (ed) Drinking in Ancient Societies. History and Culture of Drinks in the Ancient Near East. Padua. ———. 1996. Erotic Art in the Near East. Pp. 2521-2531 in J.M. Sasson (ed) Civilizations of the Ancient Near East (4 volumes). New York. ———. 2000. Some Thoughts about the Transmission of Iconographies between North Syria and Cappadocia, End of the Third-Beginning of the Second Millennium B.C. Pp. 1397-1416 in P. Matthiae, A. Enea, L. Peyronel & F. Pinnock (eds) Proceedings of the First International Congress on the Archaeology of the Ancient Near East. Rome. Porada, E. 1948. Corpus of Ancient Near Eastern Seals in North American Collection. Vol. 1: The Collection of the Pierpont Morgan Library. (Bollinger Series 14). Washington. ———. 1971. Remarks on Seals Found in the Gulf States. Artibus Asiae 33: 331-338. Possehl, G.L. 1996. Meluhha. Pp. 133-208 in J. Reade (ed) The Indian Ocean in Antiquity. London. Potts, D.T. 1990. The Arabian Gulf in Antiquity. Vol. I: From Prehistory to the Fall of the Achaemenid Empire. Oxford. Selz, G. 1983. Die Bankettszene. Entwicklung eines “überzeitlichen” Bildmotifs in Mesopotamien von der frühdynastischen bis zur Akkad-Zeit. (Freiburger altorientalische Studien, Bd. 11). Wies-

251

baden. Shaffer, J.G. 1992. The Indus Valley, Baluchistan and Helmand Traditions: Neolithic Through Bronze Age. Pp. 441-464 in R. Ehrich (ed) Chronologies in Old World Archaeology. Third edition. Chicago. al-Sindi, K.M. 1994. The Dilmun Seals in the Bahrain National Museum [in Arabic]. Manama. Teissier, B. 1984. Ancient Near Eastern Cylinder Seals from the Marcopoli Collection. Los Angeles. ———. 1994. Sealing and Seals on Texts from Kültepe karum Level 2. (PIHANS 70). Leiden & Istanbul. Winter, I.J. 1986. The King and the Cup: Iconography of the Royal Presentation Scene on Ur III Seals. Pp. 253-268 in M. Kelly-Buccellati, P. Matthiae & M. Van Loon (eds) Insight through Images. Studies in Honor of Edith Porada. (BiMes 21). Malibu.

L. PEYRONEL, Italian Archaeological Mission to Ebla, Via Palestro 63, I-00185 Rome, Italy. [email protected]

252