TO THE EUPHRATES

AND BEYOND
Archaeological studies in honour
ofMaurits N.van Loon
Edited by
O.M.C.HAEX, H.H.CURVERS & P.M.M.G.AKKERMANS

OFFPRINT

A.A.BALKEMA/ ROTTERDAM /BROOKFIELD/ 1989

Halaf mortuary practices: A survey

PETER M.M.G. AKKERMANS

INTRODUCTION

In recent years, much attention has been given to the sociological impact of burials found
in excavations. Following the pioneering work of Saxe (1970) and Binford (1971),
virtually all studies of ranking and stratification of ancient societies rely to some degree
or other upon analysis of mortuary remains. Simply speaking, it has been suggested that
patterns in death bear a direct relationship to patterns of life in society. Any social
distinctions in life are assumed to be directly reflected in the burial record, thus enabling
the student to reconstruct a society's social organisation by analysing the mortuary
evidence. This approach has been severely criticised by, among others, Hodder (1980)
and Pader (1980; 1982) as being much too simplistic and as overlooking many of the
symbolic actions in human mortuary behaviour. Burial practices do not a priori reflect
social realities but, in contrast, may give an inverted or disguised picture of the world of
the living, due to prevailing societal attitudes towards death.
This paper intends to give an outline of Halaf mortuary practices. On the basis of the
burial evidence from several sites, I attempt to draw some general conclusions about
Halaf social structures and practices. The period involved ranges approximately from
5500 to 4500 B.C. At this time, Halaf society was widely dispersed along the northern
fringes of the Fertile Crescent, including modern northern Iraq, southeastern Turkey and
northern Syria. On the basis of the pottery, LeBlanc and Watson (1970) considered Halaf
being the first widespread cultural horizon in the Near East. If this view is correct, a
highly uniform burial record can be expected, more or less interchangeable from site to
site.

THE B U R I A L RECORD

Iraq
The first Halaf burials were found in the early 1930s at Tell Arpachiyah (Mallowan and
Rose 1935:42-43). Mallowan uncovered nine Halaf graves, six of which appeared in the
outlying areas whereas three others were found on the mound itself. The burial descrip-
tion is highly incomplete and allows some broad inferences only. All burials consisted of
simple and isolated pit inhumations. The dead seem to have been laid in a contracted

75
76 PMM.GAkkermans

position either on their left or on their right side. Orientation of the body was either
NW-SE or, in one instance, E-W. Two graves contained remains of infants, one of which
(G-57) had probably been laid in a mat-covered pit. One grave (G-56) was that of a child
whereas the others probably contained adults. In six graves burial gifts were found which
consisted of ceramic vessels, placed at the head and occasionally supplemented by bone,
obsidian, flint or other stone implements. In one case (G-55), a number of red-painted
shell beads were found at the neck, probably forming part of a necklace. Three graves,
two of which belonged to infants and one to an adult, yielded no grave goods at all.
In 1976, Ismail Hijara undertook soundings at Arpachiyah, thereby uncovering two
fractional adult burials (G1-G2) and one complete interment (G3) in the area south of
Mallowan'sTTS tholos (Hijara 1978,1980; Hijara et al. 1980). Burial Gl consisted of a
pit into which a large globular and tall-collared jar containing a skull had been placed.
Next to the jar two painted pottery bowls and one of stone were found. Burial G2
represented a collective grave of four skulls, three of which had been interred in pottery
bowls whereas the fourth had been placed in a squat jar. In addition, six ceramic vessels
and one stone bowl were found in the burial pit. Burial G3 consisted of a complete
skeleton, found in a contracted position, lying on its right side. Two pottery bowls and
two stone bowls had been placed near the feet. Mallowan (1935:34) suggested that the
tholoi found at Arpachiyah represented shrines built within a sacred precinct. Hijara
(1978) takes up Mallowan's point, thereby interpreting the skull burials as ritual
interments of people of high social rank within a religious centre.

At Tcpe Gawra, located about 15 km northeast of Arpachiyah, some Halaf graves have
been found in the soundings of Area A, at the foot of the mound (Tobler 1950). No
building remains have been uncovered here. Area A was probably used as a dump of
domestic debris by the inhabitants of Tepe Gawra. In this dump, three graves had been
dug, one of which (G36-158) belonged to a child whereas the two others (G36-159 and
G36-160) contained adults. The dead were laid in a flexed position either on their left side
(child) or on their right side (adults). The child inhumation was oriented N-S and
accompanied by a painted bowl. One of the adults (G36-159) was oriented SE-NW and
wrapped in a reed mat. Grave goods consisted of two painted jars, one placed in front of
the body whereas the other had been placed at the feet (cf. Tobler 1950, PI. LXIV.b). No
description is available for the second adult burial (G36-160).
In addition to these individual graves, some mass burials, containing 24 individuals in
total, have been found in a pit in the eastern corner of the sounding. This pit was five
metres in depth and maximally 3.10 metres in width. It had probably served as a well or
cistern for some time, but rapidly been fi lied with soil washed down from the slopes of the
mound. At a time when this pit was no longer used as a water-supply, it served as a burial
pit. Four different levels of interment are distinguished, termed Burials A to D (A being
the topmost; Tobler 1950:49).
Burial A consisted of two adults. Tobler (1950:49) suggested that whereas one of these
individuals had been formally interred in a contracted position on its left side, the other
one may simply have been thrown into the pit. However, judging from a photograph of
this burial (Tobler 1950, PI. C.a), both individuals were found in contracted position at
almost the same elevation. Moreover, a bowl was found directly above the so-called cast
individual, suggesting some kind of burial ritual. It seems more likely that both indi-
viduals were interred at the same time in a similar way. The appearance of impressions of
 Halaf mortuary practices 77

wooden poles between the elbows and knees and underneath one individual is interest-
ing, perhaps indicating some kind of burial construction or coffin.
Burial B was found about one metre below the topmost inhumations and consisted of
twelve adults, probably all thrown into the pit. Bones were scattered over the area without
any orientation or formal arrangement. A squat, lug-handled jar, two stone pendants and a
basalt pestle were associated with this mass interment. One well-preserved skull from
Burial B has been analysed and belonged to a female person of about 25 years old (Tob 1er
1950:217).
Burial C appeared only about 20 cm below burial B and consisted of nine adults. Like
in Burial B, these individuals had probably all been thrown into the pit, since no
orientation or arrangement of the bones was perceptible. Moreover, some individuals
were lying on top of others. No objects were directly associated with this burial but a
painted bowl and a painted jar were found at a slightly deeper level (10-15 cm underneath
the skeletons). Tobler (1950:49) points out that these vessels must have been cast into the
pit as well, since sherds were found scattered all over the area.
Burial D was found about 40 cm below Burial C and consisted of a single young adult.
The contorted position of this individual suggested that no formal interment had taken
place. No burial gifts were present.
The majority of the mortuary evidence for the Halaf period stems from the Russian
excavations at the sites of Yarim Tepe I, II and III, situated in the Sinjar plain west of
Mosul. In total, about forty graves have been found at this complex.
Yarim Tepe I belongs entirely to the Hassuna period. No later phases of occupation have
been attested. On the basis of some Halaf burials (at least six), sunk into the lower
Hassunan levels, Merpert and Munchaev (1969:131) suggest that the site served as a
burial ground for the nearby Halaf settlement of Yarim Tepe II.
The dead were buried in a flexed position on their left or on their right side. Orientation
of the bodies was E-W, N-S or NW-SE. Grave goods consisted of pottery and stone
vessels, beads of bone or shell, and stone axes and maces. Some of the graves showed a
vertical shaft giving access to a lateral burial chamber, containing one individual.
Another grave consisted of a long but narrow pit, measuring about 200 cm in length, 45
cm in width and 120 cm in depth. Only few human bones were found in this pit but it
remains unknown whether these represent a deliberate fractional burial or are, in contrast,
due to unfavourable conditions of preservation. A large skull of a bovoid was found
above the pit whereas in the pit about 200 astragalus bones were found as well as
fragments of three clay vessels and three stone objects, one of which has been identified
as a mace. It has been suggested that this grave contained the remains of a hunter
(Merpert and Munchaev 1971:17).
Most burials have been found at the Halaf settlement of Yarim Tepe II. Here at least thirty
graves were uncovered (the exact number is not clear from the various reports), most of
which belonged to infants and children up to eight years old. So far, only five or six adult
burials have been uncovered. Three types of mortuary treatment can be distinguished at
Yarim Tepe II, viz. simple pit inhumation, cremation and skull interment (fractional
burial).
Simple primary inhumations in pits are most common at Yarim Tepe II, comprising
about two-thirds of the burial sample. These graves all contained infants or children, laid
78 P.M.M.GAkkermans

in a contracted position either on their left or on their right side. Orientation of the body
was E-W, N-S or NW-SE. One burial (N59) contained two infants whereas another (N61 )
consisted of two adults and a child (Merpertand Munchaev 1987:25). Infant burials seem
to contain no burial gifts but in most other graves one to three pottery vessels were found,
occasionally supplemented by stone vessels or bone and stone implements. Thus a burial
of a seven-year-old child (N32), uncovered in level 4, yielded a miniature stone bowl, an
unfinished alabaster vessel, a painted pottery cup, a flint implement and two bone awls
(Munchaev and Merpert 1971:31 ). Another burial (N58), belonging to the earliest stages
of Halaf occupation at Yarim Tepe II (levels 9-8), contained a highly stylised figurine of
clay and a small cup filled with hundreds of beads of shell, carnelian and an unidentified
kind of stone (Merpert and Munchaev 1987:25-26). One pit burial of level 4 (N36) seems
to resemble the shaft graves found at Yarim Tepe I. This burial, described as 'a small
catacomb under the floor' (Munchaev and Merpert 1971:31 ), contained a three-year-old
child, lying in flexed position on its right side, facing west. Grave goods consisted of six
painted vessels, a necklace made of alabaster beads, and a pendant showing animal
heads. Two other graves (N42 and N48) need a short comment. Burial N42 consisted of
an irregular, rounded pit about 140/160 cm in diameter. The skeleton was lying on its
back with the arms spread and the legs bent and crossed. No grave goods were found.
Considering the position of the skeleton, it has been suggested that the corpse had been
thrown into the pit without a formal treatment (Merpert et al. 1977:91). The other burial
(N48) seems to represent a secondary interment. In a small and shallow pit, measuring
about 110 cm in length, 60 cm in width and only 20 cm in depth, the disordered remains
of a seven- or eight-year-old child were found. The skull had been laid on top of other
bones. No funerary gifts were present (Merpert et al. 1978:40).
Seven cremation burials have been found at Yarim Tepe II, all in the lower levels 7-8.
Cremation seems to have taken place either in an open fire-place or in a specially built
oven, sunk into lower levels. The latter has been attested in one instance and consisted of
a rectangular oven with rounded edges, measuring about 110 cm in length and 35/50 cm
in width. The walls of the oven narrowed towards the top (Merpert et al. 1976:51-52).
The size of this oven probably reflects the height of the person burnt (apparently a child).
During the cremation, three stone vessels and two pottery vessels seem to have been
deliberately broken and thrown into the oven. Some other objects were supposed to have
been thrown into the fire, too, but it is unclear from the publication of the facts of this
burial (Merpert et al. 1976:52) whether all finds listed were actually exposed to the fire or
were placed in the oven afterwards. The vast number of beads of various materials (shell,
gypsum, obsidian, clay and an unidentified kind of stone) and some pendants found in the
oven suggest that the deceased was wearing a kind of necklace at the time of cremation.
After the cremation had been completed, the remaining bone fragments were collected
and put into a small jar, together with twenty beads. The jar was put in the northwestern
corner of the oven whereas in the opposite corner some miniature vessels were placed.
Next, the oven was covered with soil.
The other six cremations (N43, N50-N54) uncovered at Yarim Tepe II do not give
evidence of specially built structures for cremation, although it should be noted that in
some cases the cremation and the subsequent interment seem to have been performed at
two separate places, leaving the archaeologist with tangible evidence on inhumation only
and without direct information about the circumstances of the burning of the corpse.
Burial N43 represents the cremated remains of a child of about ten years old. The burning
 Halaf mortuary practices 79

seems to have taken place in an open fire-place, marked by an accumulation of ashes and
charcoal. During the process of burning, two alabaster vessels and three pottery vessels
seem to have been deliberately broken and thrown into the fire. After completion of the
cremation, the remaining bones were gathered and stored into a lugged jar, buried close to
the fire-place (Merpert et al. 1977:91-92).
Burial N50 consisted of a small pit, being 70 cm long and 30 cm wide, in which the
cremated remains of an adult had been laid, mingled with ashes and charcoal. Since the
base and sides of the pit did not show traces of fire, it is assumed that the actual burning of
the corpse took place somewhere else (Merpert et al. 1978:40). Fragments of two painted
vessels were found in the northwestern comer of the pit and are thought to have been
broken and thrown into the pit during interment (Merpert et al. 1978:40).
Burial N51 represents the remains of another adult cremation. As with Burial N50, the
burial ground of N51 did not show any traces of burning. Moreover, no ashes or charcoal
were found, except for one piece of burnt wood, and it is again assumed that the actual
cremation was performed elsewhere (Merpert et al. 1978:40). The bones were laid in one
corner of the pit without any arrangement except for the leg bones which seem to have
been properly positioned. Two intentionally broken vessels were found, showing traces
of secondary firing (probably resulting from the process of cremation).
Burial N52 was found near N51 and consisted of a small pit measuring 35x25 cm and
containing the burnt remains of an infant. No traces of ashes were found nor any burial
gifts. Cremation must have been performed elsewhere.
Burial N53 was situated only one metre west of N52 and consisted of an oval pit,
measuring 116 cm in length and 53 cm in width. The sides and bottom of the burial pit
showed slight traces of burning in some places. The pit contained the cremated remains
of a child.
Finally, burial N54 consisted of a pit having a length of 125 cm and a width of about 50
cm. From the centre and the eastern half of the pit the burnt bones of an adult are reported,
lying on a ca. 15 cm thick layer of ashes. Mainly in the northern corner of the grave,
fragments of one painted and three undecorated vessels have been found, showing traces
of secondary firing. Moreover, two clay spindle-whorls and a red-coloured stone were
found. Although the excavators of burial N54 are inclined to see this burial as the interred
remains of a corpse cremated elsewhere (Merpert et al. 1978:41), the present evidence
strongly suggests that burial N54 represents the place both of cremation and of subse-
quent interment of the remains. If this view is correct, the size of the burial pit will reflect
the height of the person cremated, thus pointing towards a child or adolescent instead of
an adult.
In the same area as the cremation burials, three fractional burials (N49, N55, N56)
have been found, consisting of skull interments. Burials N49 and N55 each contained one
skull of an adult, laid on its left side and facing either East (N49) or West (N55). Burial
N56 yielded three skulls, buried together in a rounded and shallow grave having a
diameter of about 60/65 cm and a depth of about 14/16 cm. All skulls had been laid on
their left side, facing northeast. None of the three skull interments contained funerary
gifts. Similar burials have been found at Tell Arpachiyah although at the latter site the
skulls had been placed in bowls or jars. Moreover, at Arpachiyah the skulls were
accompanied by a number of stone and pottery vessels (see Hijaral 978).

The mound of Yarim Tepe III, situated at a close distance from Yarim Tepe I and II, has
80 P.M.M.GAkkermans

yielded only three Halaf burials so far (Merpert and Munchaev 1984). All graves were
found in domestic deposits and represented simple pit inhumations of children. Burial
N29 contained the remains of a two-year-old child, laid in contracted position on its right
side and facing south. Orientation of the body was E-W. No grave goods were present.
Burial N31 was found in the close vicinity of N29 and also contained a child of about two
years old. The child was lying in contracted position on its left side, facing north.
Orientation of the body was E-W. Again, no funerary gifts were found. Burial N30
belonged to a twelve-year-old child, buried in a contracted position with the head
westwards. Orientation of the body was N-S. One undecorated and three painted vessels
had been placed at the feet. Burial N30 belongs either to the final stages of the Halaf
period or to the beginning of the subsequent Ubaid period at Yarim Tepe III (Merpert and
Munchaev 1984:57).

East of the Yarim Tepe complex, some Halaf burials have been reported from Tell Azzo I.
Several decapitated human skeletons were found in a tholos but the relation between
building and interments is vague (Killick and Roaf 1983:206). The headless inhumations
from Azzo immediately recall the skull interments from Yarim Tepe II and Arpachiyah.
Apparently, the practice of dividing the head from the body is widely spread in the Mosul
region.

In the Tigris valley north of Mosul, some Halaf graves have been found at Tell Kulan and
Kharabeh Shattani. At the former, an adult inhumation was found but no further details
are available yet (Killick and Black 1985:235). At Kharabeh Shattani, south of Tell
Kutan, two pit inhumations of adults were found. One interment was accompanied by a
bone point and a stone bowl. Moreover, the individual had been wearing a girdle around
the hips. On the girdle there were about twenty beads of bone, coloured stone and
dentalium shell. A trapezoid bead of black stone was attached to the front of the girdle
(Watkins 1987:227).

Syria and the Levant

Inland Syria and the coastal regions along the Mediterranean are poorly understood in
terms of Halaf mortuary practices. Halaf burials have been reported from Chagar Bazar,
Shams ed-Din and Ras Shamra. It should, however, be kept in mind that Ras Shamra is
situated on the extreme western periphery of the Halaf culture. So far, only some of the
ceramics found on the site showed Halaf characteristics, leaving the question un-
answered whether Halaf influences also entered the more ritual-bound and tradition-
based spheres of society such as expressed in the treatment of the dead.

At Chagar Bazar, levels 12 to 6 have yielded eight Halaf burials (Mallowan 1936:18).
One more burial has been reported from level 12 but apparently no body was found
(Mallowan 1936:18). Because further information is lacking, nothing can be said on this
feature. The other graves mainly consisted of child inhumations. Adult remains have
been unearthed in three cases. All burials represented simple pit interments. The dead
were lying in a contracted position on their sides. In most cases, the body was oriented
E-W but in two instances the orientation was N-S. All burials except one were accom-
panied by grave goods, viz. a painted or unpainted pottery vessel (some of them were
miniatures).
 Halaf mortuary practices 81

At Shams ed-Din a badly preserved child inhumation was found. This burial consisted of
a small and rounded pit, about 80/90 cm in diameter, and contained the skull and some
bones of an incomplete child skeleton. No grave goods were present (al-Radi and Seeden
1980:106). It is not known whether the incomplete nature of the skeleton is due to
preservation or to secondary interment.

At Ras Shamra, levels IV C to IV A are ascribed to the Halaf period. These levels yielded
an unspecified number of simple pit inhumations, mainly of infants and children. Apart
from a skull in level IV B, adults only appear in the lower level IV C. The dead had all
been laid in a contracted position on their sides. No grave goods were present except for
the ornaments worn by the deceased at the time of interment (de Contenson 1973:93-94)

Southeastern Turkey
In Turkey graves of the Halaf period have been found at Mersin, Gerikihaciyan and, more
recently, at Cavi Tarlasi.

At Mersin, levels XIX-XVII are ascribed to the Halaf period (Garstang 1953:101 ff). The
pottery of Mersin is of a local, Cilician origin but shows evidence of strong Halaf
influences (Garstang 1953:101 ff; see also Davidson 1977). In analogy with Ras Shamra,
one may wonder whether the graves unearthed at Mersin also show Halaf traits or, in
contrast, fit entirely within local traditions. Evidently, this problem arises whenever one
reaches the periphery of Halaf society. At present, the highly restricted knowledge on
Halaf society in general, and on the marginal regions of this culture in particular, prevents
us from drawing any definite conclusions.
Six burials are ascribed to Mersin XIX-XVII. Two burials belong to level XVII, one to
level XVIII and three to level XIX.
The two graves sunk in from level XVn both represent simple pit inhumations. One
burial belonged to an infant, lying in a flexed position on its right side and facing north.
Orientation of the body was NW-SE. The other skeleton was lying in a contracted
position on its left side. Only the leg bones had been preserved. Both graves contained no
funerary gifts. The burial ascribed to level XVIII does not seem to represent an
intentional interment but an accidental feature. The individual was lying on its left side
but not in contracted position. Moreover, stones and mud bricks rested against the
shoulder blade and separated the skull from the body. The burials ascribed to level XIX
consist of two individual inhumations and one mass interment. One of the single
inhumations had an individual lying in a contracted position on its left side. Orientation
of the body was E-W. No grave goods were present. The other single burial seems to
represent a cremation of a child. The bones showed traces of blackening whereas the area
around and underneath the skeleton was red-coloured due to fire. Again, no grave goods
were found. The last burial ascribed to level XIX represents a mass cremation of an
unknown number of adults. With the bones some sherds and two crushed, imcomplete
pottery vessels were found. The ceramics were blackened and apparently accompanied
the deceased at the time of cremation. The cremation burials of Mersin immediately
recall to mind the occurrence of almost similar graves at Yarim Tepe II in northern Iraq. In
this respect, it is interesting to note that cremation burials were unknown at Mersin before
the introduction of Halaf at the site.
82 PMM.GAkkcrmans

Gerikihaciyan is a Halaf settlement situated in the hilly country near Diyarbakir. The site
is located on the periphery of the known distribution of Halaf society, like Mersin, but in
contrast to the latter, Gerikihaciyan is not so much Halaf-influenced as of true Halaf
derivation (Watson and LeBlanc, forthcoming). Excavations at Gerikihaciyan in 1968
and 1970 yielded four burials, all simple pit inhumations, three of which were complete
and one highly fragmentary. The graves could not be associated with a specific feature or
level of occupation.
Burial 1 showed an adult, probably male and aged between 25 and 40 years. The
individual was lying on its left side in a contracted position. Orientation of the body was
E-W. No grave goods were found. Burial 2 represents a female child, 6-7 years old and
lying in a contracted position on its left side. The body was oriented SE-NW. No grave
goods were present. Burial 3 belonged to a three-year-old child, lying in extended
position. The body was covered by a fragment of a large jar. Orientation of the body was
SE-NW. No grave goods were present. The last burial consisted only of five rib fragments
and a left humérus. The latter pointed towards a heigth of 160 cm for the person interred,
thus suggesting that this individual was an adolescent or an adult. Either this burial was
disturbed by later activities of occupation, or it represents a secondary interment. Again,
no grave goods were found.

Recent excavations at Cavi Tarlasi in the Euphrates valley have yielded 18 Halaf burials
(von Wickede and Misir 1985). The majority of these inhumations (16) belonged to
children, mainly buried in a contracted position on their right side. None of these child
interments contained funerary gifts except for two which each yielded a pottery vessel.
Two other graves (nos. 3 and 9) represented double interments, each consisting of an
adult male and a child. This is interesting since concomitant inhumations of adults and
children, without further determination of sex, are commonly viewed as 'mother-and-
child' burials, both of whom are assumed to have died in childbirth. One of the double
burials (no. 9) contained a pottery plate, a stone axe and some flint and obsidian
implements, all placed near the shoulder of the adult. In view of the limited number of
burials uncovered within the inhabited area of Cavi Tarlasi, it is assumed that there must
have been a cemetery somewhere in the vicinity of the settlement (von Wickede and
Misir 1985:105).

HALAF MORTUARY PRACTICES: SOME R E M A R K S

From the foregoing, it will be clear that any analysis of Halaf mortuary practices is
severely hampered by the fragmentary nature of our present data. The existing burial
record is deficient in various respects. Firstly, reports on Halaf burials are often inade-
quate, either because they generalise the burial evidence or because they omit many
details of utmost importance for further analysis (like sex and age). Secondly, vast regions
are virtually unknown in terms of Halaf mortuary practices (e.g., Syria). At present, most
information stems from sites situated in the Mosul region of northern Iraq. Thirdly, Halaf
mortuary behaviour may have changed through time and space. The Halaf period is of
long duration and the investigated sites cover various stages in various regions of Halaf
development. The overall picture may thus be distorted by temporal and spatial effects
(O'Shea 1984:14). Fourthly, as a result of the current emphasis upon settlement archaeo-
 Halqf mortuary practices 83

logy, all burial evidence comes from contexts of occupation. The rare occurrence of adult
graves within settlements leads to the conclusion that the present burial sample is highly
biased in favour of infants and children. Halaf cemeteries have not yet been found, except
perhaps at Yarim Tepe I. On the latter site only six Halaf graves have been found so far
and it may very well be the case that these burials belong to a now eroded Halaf
occupation level instead of a true cemetery (Breniquet 1987:239). Nevertheless, the
striking scarcity of adult interments at the investigated mounds strongly points towards
the existence of formal burial grounds outside these settlements. These graveyards will
have served to inhume most of the dead of a community, whereas interment within a
settlement is largely limited to children and perhaps individuals of a specific status. The
present, biased burial record shows only a part of Halaf mortuary variety. Hijara
(1980b:209-216) and Breniquet (1987:238) have, among others, suggested that normat-
ive Halaf treatment of the dead involves individual inhumation within settlements, a view
which needs some modification in the light of the present evidence. Moreover, the
common notion that the deceased are interred underneath the floors of inhabited
buildings (Breniquet 1987:238) is questionable. The stratigraphie relationship between
inhumation and associated building is often vague. In most cases, it cannot be ascertained
whether the burial was sunk into a floor of a contemporary building or covered by a floor,
thus making any association hazardous. It seems more likely that most burials took place
in deserted areas of a settlement, perhaps characterised by collapsed buildings still
standing to some height.
The present evidence shows that Halaf mortuary practices widely varied, from single
and double pit inhumations, mass interments, single and multiple skull sepultures to
individual or mass cremations. Moreover, within each kind of burial group considerable
variation is noted, e.g. in grave construction, orientation of the body, and number and
kind of funerary gifts. Apparently, the mortuary ritual is complex and defined by belief
systems and cultural circumstances, which may vary from case to case. Differences in
mortuary treatment may be related to status, age, sex, ethnic entity as well as to the causes
of death. Ucko (1969:271 ) already pointed out that those who died from illness, drowning
or lightning may have been treated each in a different way from those who died in battle
or through murder. As noted before, the existing burial record is biased, emphasizing the
interment of children. Others (e.g. Ucko 1969:270 ff; Saxe 1971:43; Poppa 1978:47)
have suggested that child interment generally proceeds along different lines than adult
burial. A high infant mortality rate, in combination with some kind of social ranking
according to age, may have prescribed specific rites of child burial. Actually, the Halaf
burial evidence seems to conform to this view. Infants and children of young age are
almost invariably inhumed, whereas the more exotic kinds of mortuary treatment
(cremation, skull interment) seem to be largely confined to adolescents and adults.
Within settlements, single and primary inhumations of children are most common.
The dead are formally inhumed, whereby the position of the body and the placement of
pottery vessels and other goods is taken into account. These single, formal interments
assert the individual and suggest that the individual's identity was seen as continuing into
death (Braithwaite 1984:104). Even the few double interments found at Tepe Gawra,
Yarim Tepe II and Cavi Tarlasi stress the individual by separating each person both
spatially and physically. Child inhumations were probably performed within family or
close kin circles. In view of social relationships and social implications, child death
hardly affected the community as a whole but merely those who were immediately
84 P.M.M.GAkkermans

related to the deceased. The dispersed interment of children within settlements, perhaps
in courtyards or other family properties instead of in a centralised community burial
ground, points towards a family-oriented mortuary procedure, too. The wide variation in
positioning and orientation of the body and in number and kinds of grave goods may
suggest that interment of children did not entirely proceed according to communal and
tradition-bound rules but was open to idiosyncratic wishes or needs of the relatives.
Remains of girdles or necklaces made of beads in burials from Yarim Tepe II and
Kharabeh Shattani suggest that at least some of the dead were clothed at the time of
interment. From Arpachiyah and Tepe Gawra we have some evidence that matting was
occasionally used to wrap the dead. The pottery or stone vessels accompanying the
deceased may have contained food, either in reality or in a symbolic manner, for the
hereafter. It may, however, also be the case that these vessels should be considered as the
deceased's personal belongings, used during his/her lifetime (cf. Ucko 1969:265). In this
respect, it is interesting to note that, whereas infant burials in most cases contain no
objects at all, children of older age are often accompanied by miniature vessels. The
so-called 'hunter burial' from Yarim Tepe I may also indicate that grave objects served to
express part of a person's social entity. In recent years, it has been repeatedly tried to infer
social ranking from grave wealth in child burials (cf. Brown 1981:30; Flannery
1972:403). Wealth is viewed as an indicator of social differentiation and positioning.
Since children, by their low age, are not capable of achieving wealth and associated status
by personal efforts, it is assumed that rank is ascribed and inherited. If we accept this
view and apply it to the existing Halaf inhumation record, no ranking other than the one
probably based on age is observed. Child inhumations show considerable variation in
number and kind of grave goods but no exceptional treatment is noted. The few adult
inhumations do not suggest any social ranking either.
Mass interments are known from Tepe Gawra and, on the periphery of Halaf society,
from Mersin. On the latter site, a mass cremation was uncovered whereas at Gawra two
mass inhumations were found in the so-called well of Area A. These mass burials deviate
from the other kinds of Halaf graves in various respects. Firstly, the characteristic
individual treatment of the dead has been abandoned. Secondly, at Tepe Gawra the
appropriate ritual discourse seems to have been largely omitted. Instead of having been
carefully positioned, the dead seem to have been simply thrown into the well. However,
the presence of a few grave goods, apparently also thrown into the pit, suggests that some
rites were performed but perhaps in a simplified manner. Thirdly, mass burials seem to
have been confined to adults only. As to the nature of the mass burials of Gawra and
Mersin, we can only speculate. Each burial contained a number of individuals who must
have died more or less at the same time. Tobler ( 1950:50) suggests that the dead of Tepe
Gawra were victims of famine or plague whereas Garstang (1953:111) considers the
cremated adults from Mersin to be victims of war, burnt by their conquerors. Garstang's
argumentation (1953:111-112), however, seems to be mere fiction. In case of Tepe
Gawra, one may wonder why the regular procedure of interment was not performed. In
analogy with the mass graves of the victims of pestilence in medieval Europe, it can be
postulated that hasty interment without an appropriate treatment can be neccessitated by
hygienic reasons and circumstances beyond one's control. Climatic conditions in the
Near East lead to a rapid decomposition of the corpse, thus requiring interment within a
short period after death had occurred (cf. Poppa 1978:25). The dead of Burials B and C at
Tepe Gawra must have been buried within a short time after each other.The lower mass
 Halaf mortuary practices 85

burial C, consisting of nine adults, was covered by a layer of soil only 20 cm thick, after
which the remains of another twelve adults (Burial B) were deposited.
So far, cremations have only been reported from Yarim Tepe II and Mersin. Crema-
tions seem to represent a late Halaf feature. At Yarim Tepe II cremations appear in the
lower levels 7 and 8, dated by some radiocarbon samples in the mid-fifth millenium
B.C.(Merpert et al. 1977:95; these dates, however, seem to be too late). According to the
pottery, Mersin belongs to the latest stages of the Halaf culture and probably dates from
around the middle of the fifth millennium (cf. Davidson 1977; Watkins and Campbell
1987). Most cremation graves represent secondary burials, i.e. the burning of the corpse
and the subsequent interment were performed at two separate places. Cremation seems to
be largely confined to adolescents and adults. The dead seem to have been extended upon
a crematory pyre within a shallow depression or, as illustrated by a burial from Yarim
Tepe II, in a specially-built oven-construction. The layout of some of the cremation pits,
reflecting the size of the deceased, suggests that the dead were cremated within a short
period after death had occurred. The occasional appearance of large numbers of beads
among the burnt remains points towards necklaces or girdles and suggests that at least
some of the dead were clothed at the time of cremation. During the burning of the corpse,
some pottery and stone vessels seem to have been deliberately broken and thrown into the
fire. Occasionally, some other objects were thrown into the fire, as well. After the burning
of the corpse, the remaining bones were gathered in most cases and either laid in a small,
shallow pit dug elsewhere or stored in an urn buried at the place of cremation. In some
other instances, the remainder of a cremation was simply covered by soil on the spot,
without further treatment. From one burial of Yarim Tepe II, we have some indications
that during interment of the cremation remains some miniature vessels were placed into
the grave as well, thus suggesting that, after the rites of cremation, some rituals associated
with inhumation burials were also performed. The deliberate smashing of pottery or
stone vessels does not seem to be restricted to the act of cremation only. In the case of a
burial from Yarim Tepe II, it has been suggested that two painted vessels were broken and
thrown into the pit during interment (Merpert et al. 1977:92). This practice reminds us of
Tepe Gawra where, slightly underneath the mass burial C, two painted vessels were
found, broken and scattered all over the pit (Tobler 1950:49). The ritual breakage or
'killing' of artefacts in association with funerals is known from many ancient and modem
societies (see e.g. Hodder 1980:164; Parker Pearson 1984:72; Tilley 1984:137 ff) and
may serve many purposes. Thus, among the Nuba of Sudan the breaking of a dead
person's personal belongings is associated with the removal of the impurity and ill effects
of death (Hodder 1980:166), whereas in other instances it has been suggested that, by
destroying the wholeness of form, the natural order is broken and, ultimately, the existing
traditional and 'natural' inequalities in social relationships are denied (Tilley 1984:137
ff). In relation to the latter view, Braithwaite's suggestion is worth mentioning that among
the people of Late Bronze Age Wessex the act of cremation itself may have been viewed
as the destruction of an individual's entity and as a ritual breakage of the continuity
between life and death (Braithwaite 1984:105). In case of the Halaf cremations, one can
only speculate about the nature of these graves. Cremations seem to have been performed
along lines of ritual behaviour different from simple, primary inhumations. Infants and
young children are excluded from cremation. Perhaps cremations are confined to persons
of specific rank or status although the present evidence gives few indications in this
direction. At Yarim Tepe II some cremation graves contained a considerable quantity of
86 RM.M.GAkkermans

grave goods but other burials yielded no objects at all. Moreover, the act of cremation
itself does not necessitate the use of resources restricted in distribution to some groups of
a community only. Thus it seems that reasons of a more spiritual nature underlie these
cremation burials. Although it can be argued that the inhumation of infants and children
of young age will have taken place largely within family or close kin circles, the
cremation of older members of a community may have been of a public nature. The latter
is illustrated by the mass cremation of adults at Mersin. The ritual burning of these
individuals, whatever caused their death, must have gone beyond the level of one family
or descent group, thus affecting a large part of the Mersin community in one way or
another.
Skull interments are known from Arpachiyah and Yarim Tepe II whereas some
decapitated skeletons have been found at Tell Azzo I, all situated in the Mosul region. The
skull sepultures probably represent secondary disposals; most likely the head was divided
from the body after complete or partial decay of the corpse. Thus, two stages of interment
are indicated: firstly, the corpse was temporarily disposed of to allow removal of the soft
tissue, and secondly, the remainder was finally interred. At the latter stage, it was decided
either to inhume the skeletal remains as a whole, or to divide the skull from the body and
bury each separately. A complete secondary interment was found at Yarim Tepe II, and
consisted of the disarticulated remains of a seven- or eight-year-old child. The skull had
been laid on top of other bones (Merpert et al. 1978:40). Sepultures of skulls only are
restricted to adults. At Yarim Tepe II, three such interments were found, each character-
ised by shallow pits on the floor of which one or more skulls had been laid. Two of these
burial pits contained one skull each whereas another yielded three skulls. No funerary
gifts were present in any of these graves. At Arpachiyah, the skulls had been treated in a
different way. Not only had the skulls been laid in pottery vessels, but they were
accompanied by a number of pottery and stone vessels as well. Two cranium graves
appeared at Arpachiyah, containing one and four skulls respectively. Hijara (1978:128)
suggests that the skull burials of Arpachiyah are associated with people holding an
exceptional social position. Concerning the burials of Yarim Tepe II, a similar suggestion
has been made by Merpert et al. (1978:40). These interpretations seem merely to have
been based upon the rare and seemingly exceptional nature of the skull sepultures.
However, not only these graves but adult burials as a whole are rare within settlements
and may thus all represent exceptional features, deviating from normative burial treat-
ment. On the other hand, in absence of comparative material (i.e. graveyards yielding a
random sample of both population and mortuary practices) any such interpretations seem
to be founded on quicksand. At present, no decisive conclusions can be drawn on
possible regularities or deviations in Halaf adult burial treatment.

REFERENCES

AL-RADI S. AND SHKDI-N H.

1980 'The AUB Rescue Excavations at Shams ed-Din Tannira', Berytus 28: 88-126.
BlNFORDL.R.
1971 'Mortuary Practices: Their Study and Potential', in: J.A. Brown (Ed.), Approaches to the
Social Dimensions of Mortuary Practices (Memoirs of the Society for American Archaeo-
logy 25). Pp. 6-29.
 Halaf mortuary practices 87

BRAITIIWAITI- M.
1984 'Ritual and Prestige in the Prehistory of Wessex c. 2200-1400 BC: A New Dimension to the
Archaeological Evidence', in: Daniel Miller and Christopher Tilley (Eds.), Ideology, Power
and Prehistory, Cambridge:Cambridge University Press. Pp. 93-110.
BRENIQUETC.
1987 'Nouvelle hypothèse sur la disparition de la culture de Halaf', in: Jean-Louis Huot (Ed.),
Préhistoire de la Mésopotamie, Paris: Éditions du CNRS. Pp. 231-241.
BROWN J.A.
1981 'The Search for Rank', in: Roben Chapman et al. (Eds.), The Archaeology of Death,
Cambridge: Cambridge University Press. Pp. 25-37.
DAVIDSON T.E.
1977 Régional Variation within the Halaf Ceramic Tradition, Edinburgh: University of Edin-
burgh.
Dr; CONTHNSONH.
1973 'La culture Halafienne du niveau IV de Ras Shamra d'après les campagnes de 1968 à 1972
dans le sondage préhistorique', Annales Archéologiques Arabes Syriennes 23: 85-103.
GARSTANOJ.
1953 Prehistoric Mersin, Oxford: Clarendon Press.
HUARA I.
1978 'Three New Graves at Arpachiyah', World Archaeology 10:125-128.
1980 The Halaf Period in Northern Mesopotamia, London:University of London.
HUARA L E T A L .
1980 'Arpachiyah 1976',/ra? 42:131-153.
HODDERl.
1980 'Social Structure and Cemeteries: A Critical Appraisal', in: Philip Rahtz et al. (Eds.),
Anglo-Saxon Cemeteries 1979, Oxford: BAR British Series 82. Pp. 161-169.
KILI.ICK R. AND BLACK J. (EDS.)
1985 'Excavations in Iraq, 1983-84', Iraq 47:215-239.
KlLLICK R. AND ROAF M. (EDS.)
1983 'Excavations in Iraq, 1981-82', Iraq 45: 199-224.
LEBLANC S.A. AND WATSON RJ.
1973 'A Comparative Statistical Analysis of Painted Pottery from Seven Halafian Sites',
Paleorient 1:117-133.
MAU.OWAN M.E.L.
1936 'The Excavations at Tell Chagar Bazar and an Archaeological Survey of the Habur Region
1934-5', Iraq 3:1-87
MALI.OWAN M.E.L. AND ROSE 1C.
1935 'Excavations at Tell Arpachiyah, 1933', Iraq 2: 1-178.
Ml-RPIiRT N. AND M U N C I I A E V R.
1969'The Investigation of the Soviet Archaeological Expedition in Iraq in the Spring 1969',
Sumer25: 125-131.
1971 'Excavations at Yarim Tepe 1970, Second Preliminary Report', Sumer 27:9-22.
1984 'Soviet Expedition's Research at Yarim Tepe III Settlement in Northwestern Iraq, 1978-
1979', Sumer 43: 54-68.
1987 'The Earliest Levels at Yarim Tepe I and Yarim Tepe II in Northern Iraq', Iraq 49:1-36.
MERPKRTN.KTAL.
1976 'The Investigations of Soviet Expedition in Iraq, 1973', Sunder 32:25-61.
1977 'The Investigations of Soviet Expedition in Iraq, 1974', Sumer 33:65-104.
1978 'Soviet Investigations in the Sinjar Plain, 1975', Sumer 34:27-70.
O'SIIEA J.M.
1984 Mortuary Variability. An Archaeological Investigation, Orlando, Fl.: Academic Press.
PADER E.-J.
1980 'Material Symbolism and Social Relations in Mortuary Studies', in: Philip Rahtz et al.
(Eds.), Anglo-Saxon Cemeteries 1979, Oxford: BAR British Series 82. Pp.143-159.
88 P.M.M.GAkkermam

1982 Symbolism, Social Relations and the Interpretation of Mortuary Remains, Oxford: BAR
International Series 130.
PARKER PEARSON M.
1984 'Economic and Ideological Change: Cyclical Growth in the Pre-State Societies of Jutland',
in: Daniel Miller and Christopher Tilley (Eds.), Ideology, Power and Prehistory, Cambridge:
Cambridge University Press. Pp. 69-92.
SAXE A.A.
1970 Social Dimensions of Mortuary Practices, Michigan:University of Michigan (Ph.D. thesis).
1971 'Social Dimensions of Mortuary Practices in a Mesolithic Population from Wadi Haifa,
Sudan', in: James A. Brown (Ed.), Approaches to the Social Dimensions of Mortuary
Practices, Memoirs of the Society for American Archaeology 25. Pp. 39-57.
TILLEY C.
1984 'Ideology and the Legitimation of Power in the Middle Neolithic of Southern Sweden', in:
Daniel Miller and Christopher Tilley (Eds.), Ideology, Power and Prehistory, Cambridge:
Cambridge University Press. Pp.111-146.
TOBLER A.J.
1950 Excavations at Tepe Gawra, Vol. 2., Philadelphia.
UCKOPJ.
1969 'Ethnography and Archaeological Interpretation of Funerary Remains', World Archaeology
1:262-280.
VON WlCKEDE A. AND MlSIR A.
1985 'Cavi Tarlasi 1984 kazi kampanyasi', VII. Kazi Sonuclari Toplantisi: 103-109.
WATKINST.
1987 'Kharabeh Shattani: an Halaf Culture Exposure in Northern Iraq', in: Jean-Louis Huot (Ed.),
Préhistoire de la Mésopotamie, Paris: Éditions du CNRS. Pp. 221-230.
WATKINS T. AND CAMPBELL S.
1987 'The Chronology of the Halaf Culture', in: O. Aurenche et al. (Eds.), Chronologies in thé
Near East, Oxford: BAR International Séries 379. Pp. 427-464.
WATSON RJ. AND LEBLANC S.A.
in press Gerikihaciyan: a Halaßan Site in Southeastern Turkey, UCLA Institute of Archaeology.