Graham Philip
16
Warrior Burials in the Ancient Near-Eastern Bronze Age: the Evidence from Mesopotamia, Western Iran and Syria-Palestine
Graham Philip
ABSTRACT—Two clear instances of warrior burials are well-known to scholars and can be readily identified in the archaeological record. The first group occurs in Early Dynastic III Mesopotamia (the mid-third millennium BC), and is exemplified by the Royal Cemetery at Ur. The second, later group, occurs in the Middle Bronze Age of the Levant (roughly the first half of the second millennium BC). The present paper is intended to demonstrate that these two instances are simply the best known parts of a wider pattern. In addition to the well-known Syro-Palestinian burials, new evidence indicates that similar practices can be identified in Mesopotamia and western Iran in the early second millennium. There is now sufficient data to demonstrate continuity of tradition between these early second-millennium Mesopotamian burials, and those of the ED III graves at Ur. An examination of the third-millennium evidence from the Levant follows, in order to ascertain whether a local background can be traced for the tradition in that area also, or whether the evidence suggests its introduction from Mesopota-
Introduction
Definition of warrior burials
Warrior burials as defined here are burials interred with artefacts whose design indicates "weapon' as their primary function. In one sense I am attempting to Jimit "weapons' to items which would have been categorized as such by those producing and using them. The term "weapon* used in this sense does not embrace all tools which could have been used in fighting, chisels or lumps of rock for example. However; it includes weapons decorated with precious materials. These may not have been used in combat, but were weapons in the sense that they looked like the real thing. I suspect that appearance was an important part of the symbolism of ancient implements.
In practice the main grey area is cutting implements, which comprise a heterogeneous range of multipurpose knives or blades. While such were usable as weapons, they may have had quite different primary functions. Burials equipped with items from this rather heterogeneous range should be distinguished from genuine warrior burials. Within the latter there is an
element of overall unity, manifested through repeated combinations of particular artefacts. No single, consistent set of attributes is present in all cases, as there are both regional and temporal variations in the goods interred, not to mention short-term exigencies. However, there are enough superficial resemblances between the various warrior burials to suggest a unity of underlying structure.
Weapons in burial contexts: recent research
A glance at the published illustrations of metal-work1 reveals that there exists a complex of typologically similar artefacts, including forms of dagger, spearhead and axe, throughout Western Asia during the later third and much of the second millennia BC. While material from any one area shows greater internal homogeneity than that from distinct regions, the degree of similarity among the material as a whole is striking. Furthermore, the forms in use are such as would be inherently
1    E.g., Maxwell-Hyslop 1946; 1949; Deshayes 1960; de Margret 1976.
Chagar Bazar
Teil Shimshara
i Megidúo i
- J Betti-shan
SŘasji-AIn   "Tell el-Far*h (N) |Dhahrat|W •    .h-e„ ^-sultan Iel-Humrg'7- El/Gib Tell el-Ajjulf
Fig. 16.1: Some of the sites discussed.
unlikely to occur simultaneously, in widely separated areas, by chance alone. There is, I believe, an underlying theme traceable with a degree of localised variation, from the Nile valley at least as far as western Iran.
Despite the number of typological studies, little attempt has been made to consider the reasons for the patterns observed. Researchers have noted typological connections between material from different regions, but have not produced an explanatory framework, beyond presumed ethnic connections or .'diffusion'. Stewart, for example, writing in the 1940s, noted the occurrence of jvarrior-equipment in MB II graves at a number of Palestinian sites, and made a connection between this phenomenon and the Hyksos.2
The first detailed discussion was that of Oren in his publication of one such burial from Tomb 92 at Beth Shan. Despite the limitations of the evidence then available, Oren clearly perceived warrior burials as a phenomenon, and described their extent in Syria-Palestine during the earlier part of the Middle Bronze Age.3 His work was followed by Dever, who considered the continuation of this pattern into the later MB periods in the coastal Levant.4
A short article by Watkins drew attention to the existence of a similar phenomenon in Mesopotamia during the Early Dynastic III period, i.e. rnanycenturies before the burials discussed by Oren. Working from
Woolley's unpublished notes on the cemetery at Ur, he observed that burials containing weapons overlapped with, but were not identical to, the very rich burials. At Ur we see clearly the employment of standard weapon types in regular combinations, i.e. the patterning which is the keynote of the whole idea of warrior burials.5
Pollock, who has also worked on the original records, argued" that"for high status burials a clear.gender distinction can be observed at Ur, with two discrete sets of funerary artefacts.6 Weapons, along with whetstones and head-bands form part of the 'male' set. Knives and daggers were usually located around the waist, axes in the hand.7 Pollock bases her assessment on the few skeletons in which sex was identified, and which showed associated sets of grave-goods, supported by evidence from Cemetery A at Kish, where a certain amount of anthropological work was carried out.8 Pollock's suggestion is supported by textual evidence from Ebla and skeletal data from Tell el-Dab'a.
Moorey suggested that the preponderance of daggers and axes over spears and arrowheads in ED III graves at both Ur and Kish suggested that these represented the personal weapons of an elite, not the equipment of ordinary soldiers.9 In fact, grave-goods found in the richer graves at these sites resemble the items
2 Stewart 1974, 52.
3 Oren 1971.
4 Dever 1975.
Watkins 1983. Pollock 1991a, 373. Ibid., 376. Ibid., 375. Moorey 1982, 32.
140
141
Archaeology of Death in the ancient near East
described in the Grave Inventory of Billala, an important temple official at Kish. This document lists a silver spear, a knife and a copper axe, as well as a 'chariot' and harness donkey (translated by Zarins as a pair of animals for hauling a vehicle) among the gifts taken to the grave . by this particular individual.10
Noting that weapons, jewellery and vessels are present in considerable quantities, while there is little or no grave equipment relating to craft activities, agricultural labour, or textile manufacturing, Pollock suggests that the grave-goods from Ur relate to the individual's ritual or political position." Watkins observed that the many burials with jweajjons^were only "moderately rich".'2 Pollock attributes this to. the marking of male status at a lower absolute level than was the case with female burials, for whom mortuary evidence for a similar 'semi-elite' is lacking."
In summary, there is now good evidence to suggest that in ED III Mesopotamia we can detect a connection between the interment of weapons in funerary contexts and matters of status and entitlement. Can this be connected to the Syro-Palestinian phenomenon identified by Oren which dates to the beginning of the second millennium BC? Watkins has pointed out a general similarity in the equipment of west Asiatic warrior burials, which suggests that there may well be a connection of some kind.14 However, the plausibility of any reconstruction which seeks to unify the two groups of warrior burials is dependent on whether or not a clear connection between them can be demonstrated. These interments in their various regional and chronological manifestations encompass a vast area, and span a period of well over a millennium. The next task is to review the archaeological evidence.
Identification of warrior burials
The coherence of warrior burials as a group was initially obscured by the fact that many occur in tombs containing multiple successive interments, within which it is difficult to isolate distinct, associated groups of material. Examples include Middle Bronze-Age groups from Ras Shamra15 and Jericho Tombs 9, D9 and D22.'6 The key contexts for identifying warrior burials as such are single interments. A number of particularly clear examples has been excavated at the Second Intermediate Period site of Tell el-Dab'a in the eastern Nile Delta, e.g.,A/II-l/12Gr5(Fig. 16.3).17
Having used single burials to identify core sets of associated items, it is possible to discern their occur-
10 Foxvog 1980; cf. Zarins 1986.
11 Pollock 1991b, 180.
12 Watkins 1983,102.
13 Pollock 1991a, 376.
14 Watkins 1983, 102.
15 Schaeffer 1936, Fig. 17; 1938, Fig. 32.
16 Garstang 1932, 46, PI. XXXVII; Kenyon 1965, 259, 284, Fig. 111.
17 Bietak 1968, Fig. 3.
rence among the larger bodies of material found in multiple-use tombs. While the coastal Levant has been well studied, the significance of this phenomenon elsewhere in western Asia was concealed until recently by the lack of second millennium material with good archaeological context. In the last twenty years or so, excavations in an area extending from north Mesopotamia and western Iran to the Nile Delta, have provided sufficient new evidence to permit a re-assessment of the topic.
Warrior Burials by Region
It is best to begin with the southern Levant, where there is a useful body of well reported material. Here, early MB HA burials occur with a distinctive range of items: fenestrated axes, short socketed spearheads, and daggers (several possible types are acceptable). As much of the older evidence from Palestine has already been reviewed by Oren and by Dever, the reader is referred to these publications for fuller references.18 The present discussion is not intended as an exhaustive catalogue of possible warrior burials, and will concentrate on material published since 1971.
The Levant in the second millennium BC
Palestine: MB IIA
In Palestine, several good examples of MB HA warrior burials are known in the north Jordan valley. At Beth Shan Tomb 92, a re-used multi-chambered EB IV tomb disturbed in Roman times," produced a fenestrated axehead, grooved dagger, spearhead and a small projectile head.20 All are of early MB IIA types. In a cemetery at Tel Rehov south of Beth Shan,21 Tomb 2 revealed a single flexed burial inside a chamber which had been blocked with a large stone. The contents included a storage jar probably of early MB IIA type.22 The weapons comprised a classic set of warrior items: a dagger with central groove and crescent-shaped handle, placed by the pelvis, a fenestrated axe by the head, and two spearheads lying together below the feet of the burial (Fig. 16.2).
Both the Beth Shan and Tel Rehov graves show a grouping of fenestrated axe with a grooved dagger and small spearhead(s). Additional material of this date was recently reported from an early MB IIA cemetery at Gesher, some 13 km south of Lake Tiberias. Here 14 graves were investigated, mostly single, primary burials,
18 Oren 1971; Devcr 1975.
19 Oren 1973, 12. 2(1   Ibid., 61-67.
Yogev 1985.
21
22 Ibid., 99.
Graham Philip
lying in flexed position with head to east.25 Offerings include a few pottery vessels,24 fenestrated axes and small spearheads. Two tombs revealed a fenestrated axe and a spear, and one a single axe. The presence of a notched, narrow-bladed axe in another tomb indicates a late MB IIA date for this particular burial.25
Three substantial daggers with crescent-shaped pommels and two ribs flanking a central groove were found in the large Tomb HOOD at Megiddo.26 Like the fenestrated axes with which they are often found, these occur most often in northern Palestine. While no fenestrated axes were recorded from this tomb, it had been re-used extensively in the LBA, and these daggers are clearly of MB HA types. Such a large multi-chambered tomb would seem suitable as a setting for warrior burials. Despite certain typological changes including the introduction of notched narrow-bladed axes and daggers with ribbed blades,2' the essential pattern of axe, dagger and small spearheads continues into the later MB IIA.
Syria: MB IIA
Groups of similar composition are known from Tomb 1 at Sin el-Fil,28 the cemeteries at Kfar Jarrah and Lcbe'a in southern Lebanon29 and the 'tomheaux de partkuliers' at Byblos.50 In none of these however, can clear associations be made between individual artefacts and particular burials. More such material comes from the Depots des Offrandes at Byblos." Other coastal graves producing similar artefacts include built stone MB IIA tombs from Amrit, south of Tartous.52
Further north, similar material occurs at Ras Shamra. However, the contexts from which these weapons come are rather uncertain. Schaeffer ascribed the appearance of new metal types to the arrival at the site, at the beginning of Ugarit Moyen I, of the 'Porteurs du tores'.33 Much of the relevant material, fenestrated axes, daggers with triangular blades, rivetted butts and crescent-shaped handles, and the socketed spearheads, is identical to that found in warrior burials throughout Palestine and western Syria during the early second millennium BC. As at many other sites, this phase at Ras Shamra is marked by the presence of single burials with little associated pottery but a noticeable quantity of weapons and other metalwork.54 I would contend that
10
23 Garfinkel & Bonfil 1990.
24 Ibid., 106.
25 Cf. Philip 1989, 39.
26 Guy 1938, PI. 147:2-3; 5-7.
27 Philip 1989, 169.
28 Chehab 1939,807.
29 Guiges 1937; 1938.
30 Montn 1928,247.
31 Dunand 1939, 1958; cf. Philip 1988 for a possible interpretation of these deposits.
32 Dunand, Saliby & Kirichian 1955. The writer was able to see this largely unpublished material in Damascus Museum in 1984.
33 Schaeffer 1949,52-55.
34 Schaeffer 1948, 23-24, Fig. 56, PI. XIII.
Fig. 16.2: Set of warrior equipment from Middle-Bronze-llA Tomb 2 at Tel Rehov. After Yogev 1985, Fig. 4.
the material ascribed by Schaeffer to the 'Porteurs du tores', is simply the local manifestation at Ras Shamra of the phenomenon of warrior burials widespread elsewhere in the Levant at the beginning of the Middle Bronze Age.
Inland Syria
From inland Syria groups are known from Hama Tombs GI and GIV,55 in central Syria. Tomb GIV in particular produced one fenestrated axe, a dagger with central groove and two spearheads. Another MB HA tomb (or tombs), is known from Tell et-Tin in the Lake of Horns. The material includes fenestrated axes, spearheads and at least one dagger with ribbed blade.56 In both cases, the association of grooved daggers with fenestrated axes and spearheads is clear, although individual groups of material cannot now be isolated. Nearer Damascus, four fenestrated axes and several daggers and spearheads come from MB IIA graves in a cemetery at Yabrud, a multiple interment.57 Taken together, the evidence suggests that for the MB IIA period, it is possible to identify distinctive items of
35 Fugmann 1958, Fig. X.
36 Gautier 1895. The material from Tell et-Tin is now in Istanbul Archaeological Museum, where the writer was able to inspect the metalwork in 1985.
37 Assafl967.
14:
143
Archaeology of Death in the Ancient Near East
Graham Philip
warrior equipment, in some cases even complete sets, occurring in grave contexts throughout the Levant.
Palestine and Syria: MB IIB/C
Palestinian warrior burials of MB IIB/C date generally occur without spearheads, and with narrow-bladed axes, and daggers of styles different from their MB IIA precursors.38 Despite the typological changes the underlying structure, especially the dagger-axe pairing, remains the same, but for the disappearance of the small spearheads. It is worth noting that at Middle Bronze Age Jericho only 10 tombs out of 51 actually contained weapons, less than 20 percent. This includes four axes and around sixteen daggers, distributed between a .total number of interments which Kenyon, working from the number of crania, put at 681.39 It is clear then that weapons were very much a minority item. This is made even more so by that fact that three of the axes, and three daggers, were found, along with the sole instance of a metal belt from the site, in Grave ]3, which contained but a single burial."0 The evidence from Jericho supports the suggestion that the presence of weapons in graves is connected with the marking of status.
Few MB IIB/C burials are known from inland Syria. Those from Ras Shamra on the coast41 show a number of local peculiarities. All are from tombs used for multiple successive interments, preventing the isolation of individual grave groups. There is clear evidence for the use of socketed axes, but in a style closer to those of Cyprus than to contemporary Palestinian forms.42 Many of the daggers resemble Middle and Late Cypriot knives, rather than the dagger styles in vogue in the contemporary southern Levant.43 A final difference is that contrary to the pattern in the southern Levant, spearheads do occur in graves at Ras Shamra, but these are larger than the throwing spears found in MB IIA contexts.44 The presence of Cypriot pottery in those tombs45 reinforces the close stylistic ties between coastal Syria and Cyprus during the later Middle Bronze Age.
Despite the underlying structural unity, the typology of north and south Levantine warrior burials differs more in the later Middle Bronze Age than in the early part of the period. This suggests the gradual dissolution of a once unified tradition. If we assume the phenomenon to have a common origin, as the initial widespread homogeneity of MB IIA practices suggests, then it seems reasonable to expect that its introduction
should have taken place at the beginning of the Middle Bronze Age, or a little earlier.
The Nile delta
In recent years warrior burials similar to those already described have been encountered as far south as the eastern Nile Delta, where they are associated with a spread of Asiatic settlements. An early example of a warrior burial was found in a recently excavated single grave from Tell el-Dab'a (F/I-o/19 Gr 8), which produced a set of weapons very similar to that from Tomb 2 at Tel Rehov; namely, fenestrated axe, grooved dagger, and spearheads, but with the addition of a metal belt.47 This particular grave is assigned to an early phase (d/2 = H) of the Asiatic settlement at Tell el-Dab'a, and is one of a group of heavily plundered, mudbrick, chambered graves built in an Egyptian manner. The fact that both the tomb architecture and the pottery from this phase at the site are predominantly Egyptian in style48 makes the characteristically Levantine weapons from this, and the other graves, all the more striking.
Work at Tell el-Dab'a and Tell el-Maskhuta shows that warrior burials were in regular use in the eastern Delta well into MB IIB/C.49 Besides the typological similarities between the Tell el-Dab'a metalwork and that of the Levant,50 other, more structural relationships exist. For example, the grave A/II1/12 Gr 5 (Fig. 16.3) assigned to stratum F - that is, the transition from MB IIA to MB IIB/C51 - contains a single interment, with dagger at the waist and axe at the head. A study of the best preserved burials at Tell el-Dab'a reveals that this was the normal layout,52 one which resembles that of the weapons found in the Royal Cemetery at Ur. In fact, the same layout is apparent in many well preserved warrior burials found throughout Western Asia.
Also noteworthy is the occurrence of donkey burials in the vicinity of these tombs (Fig. 16.3), and the presence of sheep/goat bones within the graves, indicating meat-offerings. While relatively infrequent in the southern Levant, burials with equids are more common in Mesopotamia.54 The very fact that these practices are of West Asiatic origin contrasts with a willingness to adopt local Egyptian pottery and grave-construction techniques, and indicates that certain aspects of burial customs were of particular significance to the partici- ■ pants.
38 Philip 1989, 169-170.
39 Kenyon 1965, 569.
40 Kenyon 1960, 313.
41 Summarised in Courtois 1979,1204-1208.
42 Philip 1991a, 82, Fig. 10.
43 Ibid., 72, Fig. 8.
44 Philip 1989,170.
45 Courtois 1979, 1204-1208.
V i
46 Bietak 1979.
47 Bietak 1992, Abb. 6.
48 Bietak 1992.
49 Bietak 1979; 1991; Holladay 1982, 44-47.
50 Philip, in press (a).
51 Bietak 1991, Fig. 3, table of overall site phasing.
52 Philip, in press (a).
53 Pollock 1991a, 376.
54 Bietak 1992, 54, nn. 23-25; further discussion below.
144
Fig. 16.3: Tell el-Dab'a, Middle Bronze Age, Grave A/II 1/12 No. 5; note multiple donkey burials and position of weapons inside the coffin. After Bietak 1968, Fig. 3.
In j;ontrast_ to _the Syrian .situation, the Delta weapon-types are virtually indistinguishable from those found in Palestine.55 However, the pattern of alloys used at. Jericho and Tell el-Dab'a shows quite clearly that the metalwork from the two sites was the product of two distinct industries.56 The conclusion is that smiths in both areas were producing metalwork to essentially the same designs, which implies the existence of a stylistic norm, widely accepted throughout the southern Levant.
Additional paraphernalia
Alongside the weapons themselves, the most striking additional element to warrior burials is metal belts. These consisted of a thin metal sheet sewn onto a leather backing. The holes for stitching are clearly visible around the margins of the metal. Examples from good contexts include those from Jericho Tomb J3,57 Tell el-Far'ah (n),5s and four specimens from Tell el-Dab'a.59 A fragment of a possible example was found in Tomb LVI at Ras Shamra.60 Metal fragments likely to come from similar belts were found in a MB IIA grave at Tell et-Tin near Horns in Syria.61 Similar fragments of sheet metal with concentric circle decoration from the Depots at Byblos62 should probably be interpreted as belt components. All are of MB IIA or early MB IIB/C date. All Syro-Palestinian contexts from which metal belts have been reported have also produced weapons, including the Depots des Offrandes at Byblos. The association is
55 Philip 1989, 169-170; in press (b).
56 Philip, in press (b).
57 Kenyon 1960, 313 Fig. 117:3.
58 Vaux & Steve 1947, 432, PI. XX:1.
59 Philip, in press (a).
60 Schaeffer 1938, Fig. 4W.
61 Gautier 1895,459.
62 Dunand 1954, Nos. 8354-8358,189-90, PI. LVII.,
clear. As neither weapons nor belts appear in the majority of graves, this association is unlikely to be a chance one.
Additional examples are known from Cyprus.63 Although their contexts are generally poor, one was found in association with a narrow-bladed axe of the local Cypriot variety.64 Although none of the Cypriot examples are from tombs with undisturbed deposits, the occurrence of narrow-bladed axes65 and metal belts, both in styles resembling those common on the mainland is worthy of note. Both of these items are foreign to local Cypriot metalwork styles, and most likely represent the extension of the mainland idea of'warrior' burials to the island in the later Middle Bronze Age.65
The presence of these belts along with warrior equipment, in Middle-Bronze-Age graves throughout the east Mediterranean littoral and Cyprus, strengthens the argument for a structural unity beneath the superficial typological differences. The warrior burial clearly represents a concept understood and deemed appropriate throughout a wide area, and emphasises the long-range contacts between elites in otherwise quite different regions.
The end of warrior burials in the Levant
The uniqeness of the phenomenon of the MBA warrior burial is emphasised by its complete absence in the succeeding Late Bronze Age. Although burials with weapons continue to occur in the LBA, these are fewer in number and feature new and different types, mainly
63 Cf. Philip 1991a, 84-85.
64 Overbeck & Swiny 1972, 8, Figs 5-8.
65 Reviewed by Buchholz 1979; Philip 1991a.
66 For further discussion, cf. Philip 1991a.
145
Archaeology of Death in the Ancient Near East
Graham Philip
long daggers with cast-hilts. The weapons found in LBA graves bear little resemblence to the distinct sets of the earlier period. Axes decline sharply as a component of the grave record. Although occasional examples still occur in the LBA,68 these are related less to MBA forms than to the shaft-hole axes of north Mesopotamian styles.69 As with the daggers, the axe-styles of the LBA are of a more 'international' character, with regional stylistic variants being less pronounced than in the MBA.
The Middle Euphrates valley in the second millennium BC
Oren suggested that a similar concept of warrior burials applied in inland Syria.70 His argument was based on the evidence from the Middle-Bronze-Age cemetery at Baghouz in the Euphrates valley near Mari. Here, a number of single burials in stone-lined cist graves were excavated, some marked by tumuli.'1 Organic grave-goods were exceptionally well preserved in these graves, enabling the excavator to ascertain the spatial arrangement of the grave equipment relative to the deceased (Fig. 16.4:1).
De Mesnil du Buisson excavated around 185 Middle-Bronze-Age graves, and another 47 of this date showing evidence of later re-use. His tabulated data'2 reveals that of these, approximately 25 contained weapons, a little over 10% of the total. The commonest equipment was one or two ceramic vessels; most graves contained little else. The most common weapon in these tombs was a spearhead; followed by fenestrated axes, a strong typological connection with western sites. Wherever positions could be established, an axe was placed by the head (Fig. 16.4:1), as at more westerly sites, and in ED III Mesopotamia, although in several cases the axe could be seen to have rested on the wood and fabric 'bed' on which the deceased had been laid.
Most burials containing axes also contained a spearhead, and wooden grave-furniture, quite rare elsewhere in the cemetery. Furthermore, the bulk of these wealthier graves were concentrateTTin ?me~area of the cemetery - on a small rise termed Mamelon II by the excavator.7' This pattern provides additional weight to the weapons-status^connection. The axe was usually accompanied by a single spearhead, rather than a pair as is more common in the southern Levant (although in one case, Grave Z95, two spearheads were found). Daggers were markedly less frequent than in the coastal Levant. Another local difference is that at Baghouz daggers were generally found in association with the
67 Maxwell-Hyslop 1946, Tvpes 31-32.
68 Cf. Deshayes 1960, 186-191.
69 Philip 1989, 180-181. : 70 Oren 1971.
'71 Mesnil du Buisson 1948, 31-32.
72 Ibid., 63-93.
73 Ibid., 30.
Fig. 16.4
1. Baghouz burial Z121; note fenestrated axe by headof deceased and remains of -wooden grave-furniture. After du
Mesnil du Buisson 1948, PI. XLVIIl.l.
2. Tell Halatua Grave 3, crouched interment with dagger positioned at the back of the deceased; this burial is not
untypical of those from the second- millennium BC cemeteries excavated in the Hamrin basin. After excavator's unpublished plan.
remains of meat-offerings, not by the abdomen of the deceased as was more common at western sites. This suggests that they were playing the role taken by the distinctive curved knives found in many Syro-Palestinian graves.74 The types of both daggers and spearheads found at Baghouz differ from those of the Levant.73 Despite these differences, I agree with Oren that Baghouz represents yet another regional variation on the Middle Bronze Age warrior burial theme.76
Mesopotamia
Valuable though Baghouz is, it still leaves us some distance both chronologically and spatially from the Royal Cemetery. However, recent data from northern Iraq permit us to extend the pattern of second-millennium BC warrior burials to Mesopotamia, where the existence of axe-types related to those of the Levant has long been clear.77 The problem here has always been that apart from occasional isolated finds - from Kish,78 Assur75 and Shemshara80 - most Mesopotamian exam-
'74 Philip 1989, 141-142.
75 Ibid., 168.
76 Oren 1971, 126.
77 E.g., Maxwell-Hyslop 1949, 115, Type 23.
78 Langdon 1924, P1.XX.5.
79 Andrae 1922, PI. 60, below.
1+6
pies were museum purchases without site-provenance, let alone archaeological context.
The new material comes from early second-millennium BC cemeteries excavated at several sites in the Diyala valley in north-eastern Iraq. These were investigated during rescue work carried out in advance of the construction of a dam in the Hamrin basin during the late 1970s. Excavations revealed a number of occupation deposits and cemeteries dating to the Isin-Larsa and Old Babylonian periods (c. 2000-1600 BC). Close dating of these tomb-groups is difficult, owing to our more limited knowledge of the ceramics of these periods, and the paucity of published data. However, these burials can be placed securely in the first few centuries of the second millennium BC, broadly contemporary with those from Baghouz and the Levantine sites discussed above.
The material
Thje_Hjynrin.mexalw.oxk is clearly related to Syro-Palestinian types, but it forms a distinct regional group.. The forms of weapon (Figs. 5-6) are rather different from their western counterparts.
Japanese and Italian teams were prominent in the excavation of graves of these periods in the Hamrin. At Tell Songor B, 16 graves were assigned to the Isin-Larsa period.81 but none contained weapons. However, several _ofthe 21 graves at Tell Songor A were more reminiscent of traditions familiar in the Levant. These rectangular pit-graves contained single crouched interments, several of which were associaTed" with small socketed spearheads, sometimes in pairs.82 These were usually positioned by the arms or behind the back of the deceased, as at Baghouz. Note should be taken of the absence of axes and daggers from these graves. Several (including Gr 228 and 71) produced animal bones, no doubt the remains of meat-offerings, another trait observed in the Leant.
Many early second millennium graves were found in the Italian excavations at.Tell Yelkhi, which revealed in level V a substantial administrative building of Isin-Larsa date.83 Most of the richer graves are assigned to this phase and many were found in the domestic areas and in quarters adjacent to the administrative building.84 These produced various weapons including both fenestrated and socketed axes,85, daggers and spearheads, as well as meat-offerings, mainly the long bones of sheep and goats. Sormrof the graves with weapons may date to late in the third millenhiurrfBCJ86 suggesting a local background for the warrior burials of the
80
..'81 82
Laessoe 1963, PI. 12B. Yokoyama & Matsumoto 1989.
Kamada & Ohtsu 1988, 137-148; see Gr 228, Fig. 3, Bronze 3; Gr 71, Fig. 5:2, Bronzes 1-2; Gr 277, Fig. 8:2, Bronzes 22-23. Bergamini 1985, 49-50. Bergamini 19S4. 235-236. Quaramelli 1985. 316, Nos. 122-123. Fiorina 1985, 62-63.
10
Fig. I6.S: Axe-dagger set from Tell el-Seib Grave 20.
Isin-Larsa period. A rich female burial was found to have been dug from the floor of the main room of the level V administrative structure.87 The lack of weapons from this richly furnished grave further emphasises the male associations of such weaponry.
Graves of the Isin-Larsa period were also found at the small rural sites of Tell Hassan and Tell Abu Husseini, also excavated by Italian teams.88 Here, weapons were less well represented in grave contexts, as in the Tell Songor cemeteries described above. The contrast between the occurence °f Jmmej^s_wapons at the important Isin-Larsa period.administrativě centre of Tell Yelkhi, with their relative paucity at other contemporary sites in the Hamrin, emphasises ...the connection between arms and status. The concentration of rich graves in the vicinity of the level V administrative building might reasonably be taken as evidence for some connection between high status groups and this structure.
While these sites demonstrate the existence and socio-political setting of warrior burials in Isin-Larsa-period Mesopotamia, too little of the actual data have been published to permit detailed typological comparisons with the Levantine corpus. However, similar
87 Bergamini 1984,237.
88 Fiorina 1984, Fig. 27; 1985,63.
147
Archaeology of Death in the Ancient Near East
Graham Philip
Fig. 16.6: Socketed axe from Chagar Bazar, Level I; this axe is clearly a variation on that from Tell el-Seib depicted in Fig. 16.5. After MaUowan 1937, PI. XVBb. Actual length 27.0 cm
material from other excavations in the area will allow parallels to be drawn.89
Excavations at Tell el-Seib produced a total of 34 early second-millennium BC graves. The burials are generally in crouched position in rectangular pits. As at most sites, metalwork is relatively rare, and many graves contain only a limited range of ceramics. Grave 20, with two burials, produced an axe-dagger set (Fig. 16.5), a silver band, and a toggle-pin, as well as metal vessels.90 and the remains of a sheep at the north-west side of grave, presumably a meat-offering.91 The axe and dagger were both located in the vicinity of the abdomen of one of the burials, the former lying as if the handle had been pointing upwards.
Although the axe and dagger types are different from the Levantine examples, this is to be expected. It is the idea of the pairing which is^mportaht. The axe is an example of Deshayes' Type ©a/92 The only example of known provenance in his catalogue is the decorated axe from Chagar Bazar (Fig. 16.6).'3 There are now a number of (unpublished) axes of this type known from sites in the Hamrin, suggesting that it should be seen as a local north Mesopotamian style.
Grave 9 at Tell el-Seib had an equid (donkey ?) burial nearby.*"This practice is reminiscent of the donkey burials found at Tell el-Dab'a in the Nile Delta. These too are often located just outside graves.95
Tell Halawa, located near Tell el-Seib, revealed some 28 Old-Babylonian-period graves96 The pottery included many thin-walled cylindrical cups, both straight- and concave-sided, some with disc-base. Jars with flaring necks and flaring-sided bowls occur as well. It is interesting that cups and goblets are here such a
89 This material was excavated by Iraqi archaeologists working in the Hamrin area. Brief preliminary reports have been published on work at these and other Hamrin rescue sites (summaries in Postgate & Watson 1979; Roaf & Postgate 1981)
90 Hannoun 1984, Arabic section, Fig. 21.
91 Ibid., Arabic section, Fig. 19.
92 Deshayes 1960, 179-180.
93 Mallowan 1937,99, Pis. XVA-B.
94 Hannoun 1984.
95 Cf. above,.
96 Abbu 1984.
97 Ibid., Fig. 28.
98 Abbu. unpublished. •^99   Rumeidiyeh 1984.
10
Fig. 16.7: Axe-dagger set from Tell Halawa Grave 5.
feature of the grave repertoire, while the juglets so common in coastal Syria-Palestine are seemingly absent. The metalwork has not been published by grave-group, but a selection is illustrated." \
From a draft report in Arabic the following data can be established.98 Only a minority of graves produced weapons, and of these most contained only a single item, generally a knife or spearhead. The graves were mosdy single crouched interments. Grave 3 (Fig. 16.4:2) had a knife/dagger positioned at the back of the deceased. Grave 5 revealed an axe-dagger set (Fig. 16.7), with the axe placed by head. Grave 7, a poorly preserved adult burial, produced a set of axe, a fragmentary knife or dagger, and two spearheads. Two toggle-pins lay behind the head, while a necklace, two copper bracelets or anklets and a dish lying by the legs completed the set of metal artefacts.
Another large cemetery was excavated at Tell Slei-mah, where 85 graves are assigned to levels I and II which span the Isin-Larsa and Old Babylonian periods.99 As at other Hamrin sites, the grave-goods included metal vessels (beakers and bowls) as well as jewellery and weapons, the latter occurring in a relatively small proportion of the total number of graves. In short, this site conforms to the pattern established by
148
the other early second millennium BC sites in the Hamrin area.
The presence at Chagar Bazar of an axe with typological parallels in the Hamrin (see above) brings us to consider the north of Syria. This axe came from Chagar Bazar Grave 131, which also contained what Mallowan terms a spearhead.100 However, the photograph suggests that it might actually be a dagger with broken tang,101 which would indicate the presence of a dagger-axe set in this tomb. The grave also produced -i". two toggle pins, a copper wine-strainer, and one ceramic j vessel, a small/medium-sized jar. The limited ceramic
\j element in the tomb echoes thepajLternTqtmd la.Syio-
| Palestinian burials of the early second millennium BC.
' At Chagar Bazar only a few of the graves contained ~ weapons. Grave 143, a robbed vaulted grave produced a
' dagger and a spear. Grave 154, with three burials, also robbed, produced three spears, pottery, beads, and animal bones, probably a meat-offering.102 Another axe, this one with a ribbed socket, comes from Grave 200, and is also dated to the early second millennium BC.103 This grave also produced a small spearhead.
Summary of north Mesopotamia
Several differences between the contents of the graves in the Hamrin and contemporary cemeteries in the Levant are worthy of comment. Single burials are the norm in north Mesopotamia. The frequent re-use of tombs seen in the Levant is less common" here. The actual weapon types employed in the two areas differ somewhat and my impression is that there may be rather less emphasis on the dagger as a grave-item in Mesopotamia than in the Levant. The_frequency-of metal vessels in Mesopotamian graves presumably indicates their greater availability there, than in Palestine, as these are rare in Levantine contexts. Toggle-pins, so common in the Middle-Bronze-Age Levant,104 also occur in the Hamrin.
It has been suggested that the red-slipped and burnished, carinated bowls and jugs characteristic of MB IIA graves in coastal Syria and Palestine were inspired by the shapes and finish of metal vessels.105 This idea finds some support in that red-slipped and burnished pottery is absent in Mesopotamia, where copper/bronze vessels, often with carinated profiles, occur in the richer graves, in the Hamrin for example. Tubb has observed that this type of pottery is not found irTThTand SyriaT106'"Unfortunately, the Middle-Bfohze"-Age graves of this area are little known, but if the Mari texts reflect accurately the prevailing conditions,107 then
100 Mallowan 1937, 120.
101 Mallown 1937, PI. XIVE.
102 For details of the graves, cf. ibid., 118-124.
103 Mallowan 1947, 85-86, PI. XLI:1. «104 Henschel-Simon 1938.
105 Amiran 1969,90.
106 Tubb 1983, 55.
107 Cf., the summary by Dalley 1984.
metal vessels were likely to have been quite widely available, perhaps rendering superfluous the production of red-slipped and burnished ceramics in this area, as well as in Mesopotamia. In other words, the familiar red-slipped vessels may have been widely used only where metal vessels were in restricted supply, namely the less economically developed coastal and southern areas of the Levant.
Warrior burials in Mesopotamia in the third millennium BC
The early second-millennium BC warrior burials in the Hamrin just described have good local predecessors. Tell Madhhur grave 5G revealed a crouched, young adult male burial interred in a large rectangular pit with two equid skeletons.108 A dagger lay across the waist of the deceased and a socketed adze some 0.25 m from the skull.109 Again we see the classic pattern of axe by the head and dagger close to the body. Other burial goods included metal beads, pins and silver jewellery as well as copper-base vessels and the remains of food offerings. This grave is dated to the late ED III or early Akkadian period, and is one of four similar burials in varying states of preservation. The earliest of these may be of ED I date.110 Tell Sabra too has produced an ED III burial containing both a dagger and a socketed adze.111 Such adzes occur quite often in Mesopotamian graves and are used in the same way as axes. At Tell Razuk a tomb dated to the early Akkadian period housed a burial in one chamber, while the other contained copper tools and weapons along with two equid skeletons lying side-by-side.
The evidence from these three sites shows that there is a strong, local thread of continuity in the Hamrin which will allow us to trace structured warrior burials at least as far back as the later third millennium BC. In fact, as hinted at by the evidence from Tell MadKhur, burials with weapons are known in the Hamrin as early as the ED I period. At ED I Kheit Qasirn, for example, copper objects are restricted to the larger tombs, presumably those of the more important personages, who may have had some connection to the elaborate ED I building at nearby Tell Abu Qassem.112 However, in these early examples the weapon types are less clearly defined and the structuring is less tight, suggesting that the essential form of the. warrior burial crystalised somewhere between ED I and ED III when it appears in a developed form in the Royal Cemetery at Ur, that is at some point in the first half of the third millennium BC.
108 Roaf 1984,116.
109 Ibid., Fig. 24.4-24.7..
110 Ibid., 116.
111 Tunca 1987, 32, PI. 33:1-4.
112 Forest 1984, 112-113.
149
archaeology of death in the ancient near east
The end of warrior burials in Mesopotamia
As in the Levant, the tradition does not continue beyond the middle of the second millennium" BC. Although little material from second-millennium cemeteries has been published, neither Ur'13 nor Tell Zubeidi in the Hamrin"4 provides any evidence for burials with warrior items of the kind described above. Occasional burials with weapons occur, but these are generally equipped with long daggers with cast-hilts. Examples are known from Nippur,115 Ur,116 'Aqar Quf"7 and Tell Zubeidi."8 These weapons stand at the head of a tradition which continues, with some variation, into the Iron Age.
The evidence from western Iran
The sketch drawn of developments in Mesopotamia receives support from the evidence from western Iran. The publication of the metalwork from Susa has revealed the sheer quantity of weapons uncovered in the excavations at the site, much of it dating on typological grounds to the third and earlier second millennia BC.119 Many of these weapons can only be assigned to general chronological periods,120 as exact contexts are frequently missing. However, the bulk of these are almost certainly from the largely unpublished graves excavated at Susa over many decades.121
To make sense of this material, we must begin with those grave-groups which are adequately documented. Carter notes a group consisting of axe, metal belt and a decorated mace-head from Grave 555. This she assigns to her period IVa, the mid-third millennium BCj? The material from this group is clearly in the warrior burial tradition, albeit a distinctly Iranian variant. Grave 507 with rich goods including a spearhead and the remains of a meat-offering she assigns to period V (late third and beginning of the second millennia BC).123 These~twtT graves reveal traits associated with warrior burials elsewhere - weapons, the metal belt and meat-offerings - suggesting that the general phenomenon extends to western Iran.
However, these graves are but two examples. In order to demonstrate clearly the eastward extension of the phenomenon, a larger body of data is required. This exists in outline published form in the numerous graves
113 Baker, in this volume.
114 Boehmer 4 Dammer 1985, 35-44.
115 McCown & Haines 1967, 22, Pi. 30:4-5.
116 Wooliev&Mallowan 1976,119, n.l:I84.
117 Curtis 1983.
•118 Boehmer & Dammer 1985, 63-64, PI. 149:646-647.
119 Over 70 socketed axes and more than 50 daggers are published byTaiion (1987).
120 Ibid., Catalogue.
121 A coherent publication of the burials from Susa has not yet appeared.
122 Carter 1980, 75-77, Fig. 22. The periodisalion used here is that of Carter 1978, 198, Table 1, as used by Tallon 1987, 28-32.
123 Carter 1980, 107-113, Fig. 44.
excavated during the pre-war seasons at the site, in particular among the 357 graves from the 'Chantier de Donjon' published briefly by de Mecquenem.124 While this rather cryptic report may not be entirely reliable,125 it should provide an indication of the nature of the mortuary evidence from Susa.
Of the 357 graves listed, 24 are recorded as containing weapons, about 7%.126 Most of these occur singly, and include axes, daggers and spears, although a rich group from Grave 89b produced a dagger, an axe, spears and other metal goods.127 Two graves (16 and 125) produced a pairing of axe and macehead. Rarely found outside Iran, this seems to be a distinctive local set.128 In one instance, Grave 294, an axe comes from an infant burial,129 which might argue for a definite association between status and weapons, the interment of the latter being a matter of entitlement rather than related to the deceased's role in the living world. Alternatively, it could be an aberration. Another is from what is alleged to be the burial of an adult female.130 However, there is little indication of the basis on which the sex of the deceased was determined, other than that the grave contained numerous items of jewellery, a notoriously unreliable indicator. Without full documentation, it is hard to date individual graves exactly, although it is clear that collectively they span a considerable period. A great number are contemporary with Carter's period TV, others including the distinctive coffin burials date to the early second millennium BC.131
Tallon dates axe-type B3 from Susa to the Isin-Larsa-Old Babylonian periods.132 This type of socketed axe features a blade with a very wide cutting edge with concave upper and lower margins which narrow abruptly rendering the blade quite slender at the point where it joins the socket. More than twenty examples are reported from the site.133 This form is rare at more westerly sites, and represents a local style.
In summary, Susa provides good evidence for warrior burials, clearly related to the standard international pattern but showing certain local peculiarities, just as observed in other regions. Chronologically weapon finds are concentrated in periods IV and V, that, is from ED III through to Isin-Larsa/Old Babylonian in Mesopo-tamian terms. Again this fits with the picture from Mesopotamia. The key feature of Susa, however, is that as far as it can be established from the available data, burials with weapons span the whole of this period
124 Mecquenem 1943, 76-106.
125 Cf. the remarks of Tallon 1987, 58-59.
126 For illustrations of typical examples see Mecquenem 1943, Fig. 73.
127 Ibid., 89.
128 For discussion of the distinctive maceheads from Susa, cf. Tallon 1987, 128-134, Table 6. These artefacts are not treated in detail in the present paper.
129 Mecquenem 1943, 102.
130 Tallon 1987,88.
131 Carter & Stolper 1984, 147.
132 Tallon 1987, 83.
133 Ibid., Catalogue Nos. 46-67.
150
Graham Philip
without interruption. Susa provides strong supporting evidence to that cited from the Hamrin for the continuity of a tradition of warrior burials over a period of a millennium or so. In other words, the burials from the Rtjya^Ctmctery'and those from second millennium BC graves in the Hamrin are parts of a unified phenomenon, which on present evidence has its_origins__in_ Mesopotamia in the early third millennium BX-
A similar phenomenon, the .placing of quantities of metalwork in graves, can also be seen in Luristan during the laier third and early second millennia BC. While most of this material has only been published in preliminary form, summary articles134 and the illustrations in the interim reports are sufficient to show that burialswith weapons, generally of Mesqpotamian. inspiration, were widespread in the Pusht-i-Kuh area of, western Iran.'Examples include material from Bar.i Surmah,13VDar tanha136 and Kaileh Nisar.137
Typblogically this material has wide parallels, and compares well with that from Mesopotamia and Susa.138 However, it should be clearly distinguished from the later 'Luristan' metalwork which dates to the end of the second and the beginning of the first millennia BC. This second group of metalwork is different in many ways from the earlier material and represents a truly local style.139 Some scholars would go so far as to see a clear break between the earlier material and later 'Luristan' metalwork140 The implication is that despite the existence of a late tradition of metalworking in Luristan, the Iranian component of the wider West Asiatic phenomenon, represented by warrior burials at Susa and in the Pusht-i-Kuh, ceases at around the same time as it does in other regions - that is a little before the middle of the second millennium BC.
The Levant in the third millennium BC
The discussion above should have clarified two points. Firstly that the warrior burials of the Levant are simply a local version of a phenomenon widespread throughout Western Asia. Secondly that this tradition has a long and continuous history in Mesopotamia and neighbouring regions, and that its cessation in Mesopotamia and western Iran, is broadly contemporary with its disappearance from the archaeological record of Syria-Palestine around the end of the Middle Bronze Age.
The next step is to assess whether the Levantine warrior burials have local third-millennium BC precursors, or represent a tradition introduced from outside
the region. To do this we must examine the evidence for warrior burials in third millennium BC Syria-Palestine.
Third millennium BC Syria
The funerary archaeology of third millennium Syria is poorly known. Individual burials with weapons occur, but sets of the kind, easily identifiable in the Middle Bronze Age, are unknown.141
The remains of three equids were found in association with burial H-70 at Halawa in north Syria.142 In Mesopotamia, interments with equids are often_under-stood as jhose.oT high3tpusIndividuals.143 This early-mird-rnillennium BC grave also included a male burial associated with two daggers placed by the chest. The association of_the equids and_the_weapons js_ clear, and although the ingredients differ from the classic Mesopo-tamian types, the underlying pattern may well be similar.
The only large quantity of Early Bronze Age weaponry from a single context comes from the Hypogeum, a large stone-built grave at Til Barsip on the.Euphrates. This structure is later than the burial from Halawa, and should be placed somewhere in the second half of the third millennium BC. Only two adult burials were recorded,1** although the site was excavated many years ago and important details may have gone unnoticed. Among other things, the chamber produced a large amount of pottery, metal bowls, a wide array of axes, spears and daggers and two rein-rings, suggestive of a vehicle of some kind.143 Although no equid bones were mentioned in the report, it does state that large quantities of sheep and goat bones were noted, suggesting the presence of meat-offerings.146 All these points indicate an episode of conspicuous consumption and display. In many ways this tomb echoes elements of the rich Early Dynastic graves of Mesopotamia.
AtMari, some distance down the Euphrates, several big, stone-built tombs, excavated in the 1930s have recendy been published in full.147 Although largely robbed, these are reminiscent_of the tomb at Til Barsip, and the surviving mafenal includes jewellery and personal items, weapons including a crescentic axe, metal vessels and various ceramic examples.1*8 These tombs date to the late ED I, i.e. the earlier third millennium BC,14' arguing for a long history of such rich graves in the Euphrates valley, of which the Mari graves and the Hypogeum are simply the best known instances.
More evidence for the importance of weapons in third millennium BC Syria comes from recent work on
134 Vanden Berghe 1973.
135 Vanden Berghe 1968, 55-58.
136 Vanden Berghe 1970a, 15-16, Fig. 12.
137 Vanden Berghe 1970b, 69-70, Fig. 72.
138 Moorey 1971, 286; Waele 1982, 36-38.
139 Cf. Moorey 1971, 288-289.
140 Waele 1982, 39.
\ ,'141 Philip 1989, 164-165. \f 142 Orthmann 1981, 54-55, PI. 39. ,143 Cf. Zarins 1986.
144 Thureau-Dangin & Dunand 1936,97, Fig. 28. '145 Cf. ibid., 106-108, Pis. 28-31.
146 Ibid., 97.
147 Jean-Marie 1990.
148 Ibid., 314-316, Pis. IX, XV, XVI.
149 Lebeau 1990, 351.
151
I
Archaeology of Death in the Ancient Near East
texts from Tell Mardikh. The frequent mention of weapons, including daggers, spears and axes, often decorated with precious metals is a striking feature of these texts. The items concerned seem to feature as part of a gift-obligation system, linking both internal and external elites to the central palace economy.150 While Archi has observed the importance of weight in these artefacts, suggesting that it was the amount of precious metal which was of primary concern,151 the very choice of weapons as one of the appropriate forms of transferring precious materials (as opposed to using ingots, for example) is itself of interest, and argues for an important symbolic role for these artefacts. Archi also notes the existence of a differentiation by sex in the kinds of artefacts given to different individuals, with weapons being a favoured 'gift' for males.152
A detailed study of the textual references to weapons at Ebla observes that it is decorated, status items, not military equipment which dominate.153 These were available in large numbers. In one case there is mention of 540 spears being provided with silver decoration,154 while other texts clearly relate to weapons fashioned for different gods.155
Several forms of dagger feature, frequently with decorated sheaths and handles. Waetzoldt points out that the familiar gir-Manu often appears in specific weights and may represent a particular style of dagger.156 They too are often decorated with precious metal, thus taking the form of prestige items. Axes were treated in the same way.157 Like elaborate daggers these were not made for the use of ordinary soldiers. The textual evidence from Ebla indicates that weaponry was an accepted way of making gifts to men, and of displaying rank, status and connections. Although few of these precious metal objects have survived, I would suggest that the existence of this symbolic role indicates that weapons played an important part in the ideology of Syria in the third millennium BC, in particular in definitions of high status and 'maleness'.
Given this situation, the appearance of weapons in graves is predictable. While the evidence remains sparse, I suspect that in time Syrian equivalents of the rich cemeteries of ED III Mesopotamia, with a number of warrior burials, will be excavated, and will confirm the existence in both Mesopotamia and parts of Syria, of a common notion regarding the expression of status messages, from an early phase of the third millennium BC.
,130 Archi 1985, 31.
151 Ibid., 30.
152 Ibid., 28-29.
153 Waetzoldt 1990.
154 Ibid., 3.
155 Ibid., 3.
156 Ibid., 17.
157 Ibid, 24.
First appearance of warrior burials in Palestine The picture is less clear in the case of Palestine There are few weapons from EB I-III grave contexts 15* Despite the existence of walled towns, and good evidence for some sort of social hierarchy during the EB II-III,159 unequivocal evidence for the expression of status through a warrior idiom is lacking prior to EB IV. The sole published example of an identifiable, rich EB H-UI burial, the EB II grave from Kinneret near Khirbet Kerak is completely lacking in weapons.160 A further difference is provided by the lack of evidence for meatofferings reported from the numerous EBA burials from sites such as Jericho161 and Bab edh-Dhra'.162
The major increase in the deposition of weapons in graves actually comes in EB IV, a phenomenon once linked to incoming 'Amorites'.163 However, more recent studies have stressed the degree of ceramic continuity from EB HI to EB IV, and interpret the archaeological evidence in terms of changes in settlement patterns and socio-economic organisation rather the arrival of new peoples.164
Although there is some regional variation among the weapon types employed in EB IV burials, a standard underlying structure can be identified. They usually feature a narrow-bladed dagger, sometimes accompanied by a hooked-tang spearhead.165 In many cases these are the sole (surviving) items accompanying the deceased. Sets there are, but the types are very different from those in use in Mesopotamia and Syria. Weapons in Syrian styles are few, and concentrated in north Palestine.166 EB IV metalwork is generally the product of a local industry, making objects in traditional styles.
The contrast between EB IV individual burials and the multiple successive tombs favoured during the EB H-ni is striking.167 In the latter the isolation of particular individuals is virtually impossible, as is any association between any one burial and particular grave goods. The notion of separating and identifying individuals on burial is more clearly expressed in the EB IV period. More importantly this pattern continues into the succeeding Middle Bronze Age. Although multiple successive burial was the norm at some MBA sites, such as Jericho161 and el-Gib,169 this practice may owe to the local availability of large EB IV tombs. Many MBA burials at sites such as Megiddo170, Ras el-'Ain,171 Dhahrat el-Humraiya172 and Tell el-'Ajjul,173 as well as
158 Philip 1989, 164.
159 Palumbo 1990; Esse 1991.
160 Amiran&Haas 1973.
161 Kenyon 1960; 1965.
162 Schaub & Rast 1989.
163 Kenyon 1966, 12-14.
164 Dever 1980; Richard 1980; Palumbo 1990.
165 Philip 1989, 165-166.
166 Ibid, 167.
167 Palumbo 1990, 125.
168 Kenyon 1960; 1965.
169 Pritchard 1963.
170 Loud 1948.
171 Oryl937.
172 Oryl948.
Graham Philip
the Svarrior' burials at Tel Rehov and Beth Shan described above, are single interments. Single burials., with weapons provide a strong threadJinking_Pjdesttn:. ian Middle Bronze Age burial practices to. those _o.f. the EB IV period. ,
"Many of the typological differences between the grave-goods of the EB IV and MB IIA periods result from the different technological possibilities open to the craftsmen of the two periods, in particular the greater use of two-piece moulds in the later period, and the different bases on which they may have worked.174 These should not distract us from an important element of continuity between the two periods, namely single burials with sets of weapons.
Besides a preference for single burials and a marked increase in the use of weapons as grave-goods, the EB IV period also sees the appearance in the southern Levant of the practice of placing., m?at-offerings" in graves!175. While this practice continues into local MBA graves, we have already noted that it was a i component of Mesopotamian, and probably Syrian, , warrior burials of the third millennium BC. Perhaps then, the warrior burials of EB IV Palestine and Jordan are simply a local attempt to emulate practices already well-established to the north, but using locally available metal artefacts. The presence together of weapons and meat-offerings, combined with the clear association of particular items with individual burials, suggests that a common structure may underlie the EB rv burial practices of both the northern and southern Levant, one which contrasts with Palestinian EB II-III practices.
The adoption in the late third millennium BC in the southern Levant of a system of marking status through warrior goods, which had been acknowledged in Syria by the mid-third millennium BC at the latest, provides a possible explanation for the appearance of burials equipped with costly metal goods in a period -' 5 the EB IV - which is generally considered to be impoverished, and which lacks major fortified centres. We might consider the possibility that the well-documented settlement shifts,176 have no relation to the changes in ^burial practices, other than that they were broadly 'contemporary.
Why warrior equpiment at all?
Possible reasons for the attribution of significance to warrior equipment have been discussed elsewhere. Briefly, they were associated with power and conquest, with the ability to take tribute and lives, i.e. the acquisition of wealth and reputation. The image of the king as conqueror, the warrior-leader as hero, occurs repeatedly
173 Petrie 1931-34.
174 Philip 1991b, 101.
175 Cf, e.g, the graves from Jericho published by Kenyon 1960; 1965.
176 Palumbo 1990.
177 Philip 1989, 156.
in ancient Near Eastern literature, and probably reflects the values of that society. The fact that the concept of weapon-symbolism cuts across a range of ancient ceramic, political and linguistic boundaries, as demonstrated above, argues that it expressed widely held, culturally formed, notions concerning appropriate male high-status behaviour. These views were current throughout a large part of the ancient Near East during much of the third and the early second millennia BC. It is this, the very unity of the phenomenon, which underlies the widespread typological similarities observable in weapon types throughout Western Asia at this time.
Decline of the Warrior Burial
One of the most striking aspects of warrior burials is that they cease at approximately the same time throughout most of Western Asia. In the southern Levant and Nile Delta, the last classic warrior burials date to the later phases of the Middle Bronze Age (we cannot be more precise than this); a similar date would seem reasonable for burials in Syria.1'8 In Mesopotamia, there is no evidence that they continue beyond the end of the Old Babylonian period.179 For Iran the existence of a possible break between the early second-millennium BC metalwork and the later 'Luristan' bronzework has already been discussed.
The uniqueness of the warrior burial phenomenon is emphasised by its complete absence in the later second millennium BC. The reasoiTfof this doubtless lies in the network of complex changes taking place towards the middle of that millennium. However, three main possibilities can be identified.
1. The widespread adoption of the combination of the chariot, the composite bow and scale-armour around the beginning of the LBA180 had a major impact on the conduct of warfare. It may also have had a concomitant effect on the mode of representing high status, with an associated decline in the deposition of sets of weapons designed for hand-to-hand combat. The growth of a group of highly trained, and high-status chariot troops, the mari-arma, may have been an important factor in the decline of the status value of the more traditional weapon-sets. However, Heltzer has pointed out that at LBA Ugarit chariot troops were dependent on royal stores and workshops, and that many man-anna were themselves royal dependents.111 Changes in the range of grave artefacts may well reflect the greater degree of control by palace organisations of
178 Cf. Philip 1989,217.
179 Cf. Baker in ihis volume.
180 Moorey 1986.
181 Heltzer 1982,114.
152
153
Archaeology of Death in the Ancient Near East
the new-style elite military equipment such as chariotry and horses. Such costly material may not have been available for use in the graves of individuals.
In the Levant, at least, the LBA has a distincly cosmopolitan nature.'12 There was a significant increase in long-distance trade in the Late Bronze Age, with more, and more varied, international trade and diplomatic connections, best seen through imported pottery.183 While there is too little pertinent archaeological data to allow an assessment of the situation in contemporary Mesopotamia, the general ambience of international relations during this time (see below) suggests an increase in long-range political relations. In these circumstances, a greater range of alternative, perhaps imported, prestige-type material would have become available, perhaps contributing to a diminution of the importance of the more traditional weapons in this role. The decline in warrior equipment may also reflect the changing political landscape during and after the sixteenth century BC. The Late Bronze Age is generally understood as seeing the replacement of the small Middle Bronze Age 'Amorite' kingdoms by larger political units, organised along different lines, the Hittite and Egyptian empires for
example, and the kingdom of Mitanni.184 Foreign powers, governing from relatively distant centres such as Hattusas and Thebes played a greater role in the political life of the region than had hitherto been the case. The appearance of new political units, and the different systems of power relations which would have resulted from this, might well have altered the mode of interchange between elites, causing shifts in the nature of their material expression. Such developments may have contributed to a decline in the importance, and hence the production, distribution and deposition, of 'personal' warrior equipment, artefacts which had only been meaningful in terms of previous sociopolitical structures.
Acknowledgements
i wish to express my gratitude to the staff of the Directorate-General of Antiquities and Heritage and other individuals in Iraq, who kindly provided access to material in the Iraq Museum, Baghdad, and at Mosul University, and to various notes and records. Helpful comments on the text were provided by the various participants at the Manchester conference, and by Stephen Bourke, Elizabeth Carter and Stjohn Simpson in particular. i also owe a debt to Trevor Watkins, some of whose insights I have hijacked as my own. For the illustrations I am indebted to my wife, Carrie Philip.
17
Aegean Influence in Late Bronze Age Funerary Practices in the Southern Levant
Garth Gilmour
ABSTRACT—The two dominant burial practices in Canaan during the Late Bronze Age were single burials in pits or cists in the coastal plain, and multiple burials in caves and rock-cut tombs in the interior. Several other burial customs are present, four of which have been ascribed to Aegean influence: chamber tombs, stone-built corbelled tombs, lamakes and cremation.
Chamber tombs are best represented in the VOO'and '500'cemeteries at Tell el-Far'ah (South), though they occur also at Tell el-Ajjul and Beth Shean. Several studies of these tombs have attributed them to either Mycenaean or Cypriote influence. It is here concluded that the 900 cemetery tombs represent a local development of the bilobate tomb tradition of MB UC, which itself was a concept imported from Cyprus; and the 500 cemetery tombs were cut by the Philistines who drew on their own Aegean as well as Cypriote and the local Canaanite traditions.
Stone built corbelled tombs in LB Canaan have been found at Megiddo, Dan and Aphek. It is here concluded that the origin of this tradition lies to the north in Syria, and not in the Aegean.
Late Bronze Age lamakes have been found only twice in Canaan, one each in Gezer and at Acco (the 'Persian Garden'}. They are typical of burials in Crete in the Middle and Late Minoan periods, and are also found in mainland Greece during LH III. While there seems little doubt about the Aegean origin of the tradition, factors such as the dates and associated finds in each case require caution before ascribing these burials to Aegean groups resident in Canaan.
Cremation appears in three different forms, at Jericho, Tell Beit Mirsim, and the Amman Airport building. Some scholars have associated cremations in Syria with the coming of the Sea Peoples, while others have attributed them to the Hittites. However, cremation may be performed for any number of reasons, and these three isolated and different examples suggest that they may be local aberrations; any outside influence is to be ascribed to Anatolia rather than the Aegean.
182 Leonard 1989.
183 Mazar 1990,261-263; Gonen 1992b, 247-249.
184 Astour 1981; Gonen 1992b, 211-214.
Introduction
Burial practices in Late Bronze Age Canaan followed a consistent pattern of single burials in cists or pits in the coastal plain and multiple burials in caves and rock-cut tombs in the Shephelah and mountainous interior.' Nevertheless there were numerous individual exceptions to this pattern, and four of these have been variously ascribed to Aegean influence. These are cut chamber tombs, stone-built corbel vaulted tombs, larnakes and cremation. In each case a detailed study of the practice reveals that the simple invoking of influence from Aegean groups is at best inadequate, and
more often unable to explain the practice, and other factors need to be considered.
Chamber Tombs
Introduction
At the site of Tell el-Far'ah (South), excavations in the 1920s and '30s exposed a settlement and several surrounding cemeteries. The '900' cemetery, which ranges in date from Late Bronze II into the Iron Age,
1    Gonen 1992a.
154
155