PERSONAL ADORNMENT IN THE EPI-PALEOLITHIC
OF THE LEVANT
CHERRA WYLLIE,1 FRANK HOLE2
Abstract
Material remains from burials at el-Wad, Mallaha and Hayonim provide evidence for the important role of personal adornment in the Early Natufian. Reconstruction drawings based on beaded headdresses, jewelry and clothing reveal a surprising level of sophistication. While the social implications of these costume elements can only be tenuously inferred, the skillful manipulation of natural products suggests a cultural consciousness of shared aesthetic and personal augmentation, as well as presaging forms and materials of subsequent decorative arts in the Near East.
Introduction
The Early Natufian of the Levant (12,500-11,000 cal BC) is notable for the number and geographic range of sites, sedentism, architecture and burial goods, all of which stand in contrast to older cultures of the region (Belfer-Cohen 1991). This paper considers one aspect of the Early Natufian, personal adornment, an attribute serving to define social identities.
The 1930s excavation of el-Wad in Palestine, recovered five skulls with dentalia shell and bead coverings, as well as bead necklaces and beaded arm and leg bands (Garrod 1940). The skulls are known from the field photographs, but there is no reconstruction showing the appearance of distinctively different headdress styles, nor any study of how the different types of beads may have been worn. Through the use of artistic renderings we offer an interpretation of headdresses, necklaces and limb bands. These images, more than published statistics, give insight into how some Natufians displayed their individuality. Their rare occurrence reinforces suggestions of social signaling.
1 University of Hartford, Humanities Department, Hillyer College, West Hartford, CT 06117, U. s. A. wyllie@hartford.edu.
2 Yale University, Department of Anthropology, Box 8277, New Haven, CT 06520-8277, U. s. A. frank.hole@yale.edu.
708
Cherra Wyllie, Frank Hole
Archaeological Data
Over 450 burials have been recovered from Natufian sites: most of these come from el-Wad, Mallaha and Hayonim. El-Wad, in Mount Carmel, the first such site described in detail, has just under 100 burials, depending on whose counts you use (Belfer-Cohen 1995). Only 45 were described in Garrod's excavation reports (Garrod and Bate 1937; Garrod 1940) and, of these, only the best preserved received attention. Hayonim has some 48 burials and Ain Mallaha 105. Byrd and Monahan (1995: 257) reckon that from these three sites there are 186 burials 'that are reported in sufficient detail to be analyzed'. Of these, 87 pertain to the Early Natufian and are relevant to the present discussion.
We based our reconstructions of el-Wad on Garrod's basic descriptions and photographs (Garrod and Bate 1937; Garrod 1940), augmented by additional information in Wright (Wright 1978), Byrd and Monahan (1995: especially Table 4) and Belfer-Cohen (1995). Data for Hayonim are in Belfer-Cohen (1988) and Byrd and Monahan (1995), and those for Mallaha are published in Perrot and Ladiray (1988). For considerations of space we present only abbreviated details and some photographs from the original publications showing the beadwork and its placement on the skeletons. Fuller descriptions and counts of ornaments are in the publications cited above.
El-Wad3
Garrod described group graves with six clusters of skeletons containing individuals whose skulls and other body parts were decorated. Each skeleton was given a number, e.g. H 41. The number of skeletons in each cluster ranged from 2-10 and five clusters featured an individual with significant decoration. Garrod described the skeletons or groups of skeletons, the bead patterns found on the skulls, as well as beads usually found in the neck area but also possibly on the arm and leg (Garrod 1940). Three of the skulls (H 23, H 25, H 57a) have particularly well preserved sets of dentalium shell beads; one (H 41) has an unpatterned scatter of shells across its surface; and a child's skull (H 28) was covered with 32 gazelle phalange beads rather than with shells. Garrod speculated that the shells on all the skulls had originally been attached to a cap, no longer preserved
3 We are grateful to Dr. Iman Saca who brought to our attention the collections of beads, worked and carved bones and other objects from the Epi-Paleolithic sites of el-Wad and Kebara that are in the Anthropology collections of the Yale University Peabody Museum. These artifacts were acquired in the 1930s by George Grant McCurdy, then at Yale and Director of the American School of Prehistoric Research, which participated in the excavations of the caves.
Personal Adornment in the Epi-Paleolithic of the Levant
709
The Reconstruction Drawings
The archaeological data are bare statistics lacking much interpretation other than that given by the numbers themselves. We asked, 'What did the adornment look like on living people?' This required considering how the decorations were constructed and worn. Immediately alternatives came to mind, leading us to investigate comparative material, including contemporary beadwork. We made trial reconstructions. Eventually we settled on the illustrations included here, recognizing, of course, that others may find different alternatives. The drawings attempt to faithfully represent the verbal descriptions, although some of the figures are composites based on more than one skeleton.
Archaeological drawings contribute to the tasks of archaeological description and interpretation by offering or accentuating specific visual information. They supplement photography by delineating details lost through taphonomic processes of decay.
Archaeological illustration is an exercise in which the artist provides a visual rather than textual interpretation of the archaeological data. In our reconstructions we considered the best ways to combine discrete categories of information based on existing photographs and published accounts. For example, creating individual portraits, we not only considered the images and descriptions of skeletons and beads, but needed to question what we knew about phenotype and ethnicity. We discussed practical hairstyles and clothing based on what we knew of Natufian lifeways and the late Pleistocene climate. We went through several stages of drawings and revisions, each one generating further discussion and investigation. As Atkins and Atkins (1989: 131) point out one of the values of the process of making a reconstruction is that 'it requires the archaeologist to face up to and even challenge the evidence, since the meaning and function of the surviving evidence has to be examined and interpreted with precision to see whether a reconstruction is plausible.' We agree with their choice of the term 'process,' rather than looking at an interpretive drawing as a finished product.
In the following paragraphs we discuss how we made decisions reflected in the drawings. In order to show that men, women and children were adorned we show a funerary scene with individuals wearing the different styles of adornment found at el-Wad and Mallaha (Fig. 1). Here, we base our reconstruction (including setting, burial pit and position of the deceased) on the el-Wad cave environment and burial practices. We follow this group picture with an illustrated discussion showing individual features of adornment found on particular skeletons.
H. 23 (Fig. 2)
Male adult. 'A number of dentalium shells were sticking to the occiput, and when the skull was removed, remains of an elaborate head-dress were found on the forehead' (pi. VII, fig. 2). 'This consisted of two fan-shaped motifs made of dentalium shells,
710
Cherra Wyllie, Frank Hole
spreading from the temples to the middle of the forehead' (Garrod and Bate 1937: 18). 'Each fan was made up of three groups of dentalium shells, carefully matched so as to be approximately equal in length, the number of rows increasing from back to front. The fans must have been fastened to a cap of some kind, made perhaps of skin or netted hair-string' (Garrod 1940: 124). 'On one femur was a band or garter of eight rows of shells....the shells and pendants were firmly cemented to the bones by a calcareous concretion' (Garrod and Bate 1937:18).
We can imagine the headdress forming a peak or small flattened plane just above and between the eyes. We do know that the headdress fit the head snugly since, as Garrod notes 'a number of dentalium shells were sticking to the occiput, and when the skull was removed remains of an elaborate headdress were found on the forehead' (Garrod and Bate 1937: 18) The wearer's head, then, likely supported the headdress, which slid further down the face after internment. The photograph (Garrod and Bate 1937: pi. VII 2) indicates that the front of the headdress would have been quite high, obscuring the wearer's hair at the front. Garrod imagined that a skin cap or netted string supported what she refers to as 'fan-shaped motifs' (Garrod 1940: 124). According to this reconstruction we imagine a cap with dentalium bead ornamentation, rather than a dentalium bead headdress. Yet another possibility is that skin or corded panels joined the two side panels to hold the headband together. A cord may also have supported the elliptical bone pendant found on the right side of the skull, which could have been suspended from the headdress or served as a hair ornament of some kind—either woven or braided into the hair.
H 23 also wore a shell bead leg band, and a stunning necklace composed of bilobate pendants hanging like stamen or conjoined droplets (Fig. 2) (Garrod 1940: pi. 11; Garrod and Bate 1937: pi. XIV 2). The so-called twin type ornaments dangle from a collar ofthreaded dentalium shells, executed with utmost artistic skill. Our reproduction gives some indication of the stately effect produced by H23's ensemble.
H25 (Fig. 3)
The male referred to as El Wad H 25 wears a seven-row beaded headdress, adhering to the skull (Garrod and Bate 1937: pi. VI, 1, 2). Garrod (1940: 124) wonders whether the beads may have been sewn on 'to the edge of some kind of cap' (discussed above) although we prefer the simpler explanation that it consisted of circular rows of dentalium shells forming a beaded headband—an interpretation explored in our reconstruction drawing. This form would have conformed to the head, requiring no other material to secure it, and appears to comprise the basic design elaborated upon in other, more ornate, headdresses at el-Wad. Additionally H 25 had 'a strip on the shaft of the right humerus made up of eighteen dentalium shells spreading fan-wise, which may be a remnant of a cloak or some other garment' (Garrod 1940: 126).
Personal Adornment in the Epi-Paleolithic of the Levant
711
H 41 (Fig. 4)
The weave of dentalium shells covered most, if not all, of the individual's hair. Bits of dentalium shells and tibio-tarsus bird bone pendants were found sticking to the skull. Shells were also found adhering to the crown of the head. Based upon this information we drew a headdress similar to that for H 57, but one that expands from a skullcap of shells into rows of beads, like later Egyptian faience bead headdresses. Overall this individual possessed the greatest number of intricate costume elements, including an armband, a leg band, and what was possibly a shell cloak (Garrod and Bate 1937: 126). In each item of adornment el-Wad artisans appeared to be experimenting with the potential of not merely stringing tubular shell beads, but forming and connecting what Garrod calls fan-fold motifs, assembling these forms into headdresses, detailed costumes and accessories.
H 57 (Fig. 5)
Figure 5 portrays an adult wearing another of these elaborate shell and bone bead headdresses, like the one Garrod (1940: 125) describes for H 41 (above). Garrod's photograph reveals how the remaining dentalium shells are still affixed to the skull, and suggested the arrangement used in our reconstruction drawing (Garrod and Bate 1937: pi. VII, 1). The 'three rows of shells' encircling the head, as the excavation photo confirms, are horizontal rows, and form a narrow headband easily supported by the natural contours of the wearer's head (Garrod 1940: 125). This headband creates the armature supporting the weight of attached ornamentation, including the layers of hollow, reed-like fringe, like that shown pressed against the skull. A similar group of dentalium beads had become detached from the headdress. Garrod describes these as two rows of 'fringe' (dentalium beads strung vertically rather than horizontally) lying on the ground alongside the skeleton. By Garrod's reckoning the intact headdress— 'four shells deep...must nearly have reached the shoulders' (Garrod 1940: 125)
On the bases of the photograph and data it is doubtful that the headdress would have hung as far down as Garrod suggests. Our drawing provides an extra row of vertical beads ('fringe') to illustrate this point. The design of the headdress would have been ideal in a ritual or performance contexts, especially dance where the horizontal headband would have stayed firmly in place while the fringe spanned out in various directions as the performer moved and spun. In our reconstruction the six tibia-tarsus bird bone pendants dangle from the front edges of the headdress, framing the face on either side. Several of these 'were also found adhering to the skull,' although it is not certain precisely how they conformed to the design of the headdress (Garrod and Bate 1937: 19). Ours is but one possibility.
The necklace resembles the one accompanying H 25, apart from the fact that a few of the twin pendants on the H 57 piece were carved from a single piece of bone. The original H 57 necklace was made up of '37 twin bone pendants and 6 phalange beads' (Byrd and Monahan 1995).
712
Cherra Wyllie, Frank Hole
H 28 Young Child (Fig. 6)
In Figs 2 and 6 we explore children's decorative attire using descriptions from el Wad H 28 and Ain Mallaha H 43. Based on Garrod and Bate's description (1937: 18) for the young child from el Wad, our reconstructed boy wears a netted cap. Gazelle phalanges dangle from a band forming the rim of the headpiece. We also dress him in a dentalium bead waistband referred to in Byrd and Monahan's description of a similar belt interred with an infant (H 43) from Ain Mallaha. Several ways of wearing this beaded sash occurred to us; two based on Biernert's photograph (1995: fig. 3). In the reconstructed group (Fig. 2) a child wears such a 'belt' as a sling circling his small body diagonally from his shoulder to his hip. The beaded belt in Biernert's photograph appears to be angled over the infant's profiled skeleton, so that it is not clear whether it is a waistband or a beaded sling.
Figurines from PPNB Nevali Qori (Morsch 2002) wear a variety of folded sashes, possibly the very type buried with the Ain Mallaha infant. The figurines inspired the reconstruction of the belt on Fig. 6. Because of the fragility of the shell beads the sash would have been cinched and overlapped loosely, or possibly affixed to a backing of some kind.
Social Significance of Adornment
The only certainties are that neither age nor sex correlate with adornment, and that most people were not buried with beads. This does not preclude that all people wore beads in life, a possibility that is difficult to verify despite the fact that loose beads were commonly encountered during excavations. Perhaps there were other equivalent forms of decoration that are archaeologically invisible, such as body paint, tattoos, feathers and the like.
It is hard to argue that beads were a significant form of wealth when they could be picked up on the beach and strung by whoever had the patience to do it. That they could make a strong social statement, however, is certain, as the drawings show. Given their essential fragility, it is likely they were worn only on special occasions, such as during dances, initiations, rituals of passage, as in burial, and so on. If there were sumptuary rules among the Early Natufians, perhaps only selected individuals were entitled to wear beads, but the range of age and sex would tend to refute that. Similarly, if the group burials were those of families, one might expect that status was shared among the individuals rather than confined to one person.4 At the least family status would seem more likely to have been conferred on the elders. While there was similar use of adornment among the Early Natufians, there were also subtle differences between individuals and among the sites, suggesting that while the practice was general, it was
4    This is not to imply that all members of any particular family would be adorned. Indeed, not only is adornment rare, but it seems that only one person in each burial group wore beads.
Personal Adornment in the Epi-Paleolithic of the Levant
713
carried out locally and individually.
The presence of dentalium shells is common in the Early Natufian, and all three sites discussed here have individuals with necklaces, armbands and bracelets. El-Wad is unique in having headdress beads still attached to the skulls so as to allow reconstruction. The use of large numbers of beads as possible chest garments or pectorals is unique to Hayonim where headdresses are not attested. While dentalium shells are used in all sites, the various kinds of bone beads display site-to-site differences.5 Bar-Yosef and Belfer-Cohen (1999: 403) suggest that such differences among sites relate to different territorial groups yet, as noted above, differences exist within sites as well. It is interesting that both men and women, all seemingly relatively young, and including children, are decorated. The facts do not provide us with clear answers as to why certain individuals wore beads and most people did not. Nor do they inform on the reasons that beaded ornaments fell from favor in the Late Natufian.
Reconstruction drawings cannot answer these questions, but they are an attempt to give a living context to what has been a largely statistical discussion. The drawings bring the subject to life, in some cases almost literally putting flesh on the bones' (Atkins and Atkins 1989: 131); conversely, 'they have a tendency to fossilize a certain aspect of archaeology' (ibid. 132). In our stages of drawing and revisions, we discovered that the data lent themselves to multiple interpretations and possibilities, whereby the rendering itself became a vehicle for the testing of ideas. We believe that archaeological drawings, like any textual reconstruction or preliminary analysis, should be viewed as an active form of conceptual discourse.
Bibliography
Atkins, L., Atkins, R.
1989     Archaeological Illustration, Cambridge. Bar-Yosef, O., Belfer-Cohen, A.
1999      Encoding information: unique Natufian objects from Hayonim Cave,
Western Galilee, Israel: inAntiquity 73, pp. 402-411. Belfer-Cohen, A.
1988      The Natufian graveyard in Hayonim Cave: in Paleorient 14 (2), pp. 297-308.
1991      The Natufian in the Levant: in AnnualReview of Anthropology 20, pp. 167-186.
5   The manufacturing and wear patterns on beads from el-Wad are the subject of a separate study by Geoffrey Cunnar.
714
Cherra Wyllie, Frank Hole
Belfer-Cohen, A.
1995 Rethinking social stratification in the Natufian culture: the evidence from burials: in S. Campbell, A. Green (eds), The Archaeology of Death in the Near East, Vol. 1, Oxford, 9-16.
Biernert, H.
1995 The human image in the Natufian and Aceramic Neolithic period of the Middle East: in W. Waldren et al. (eds), Rituals, Rites, and Religion in Prehistory, Vol. 1, Oxford.
Byrd, B., Monahan, C.
1995      Death, mortuary ritual, and Natufian social structure: in Journal of
Anthropological Archaeology 14 (3), pp. 251-287. Garrod, D., Bate, D.
1937      The Stone Age of Mount Carmel. Excavations at the Wady el-Mughara, Vol.
1, Oxford. Garrod, D.
1940      Notes on some decorated skeletons from the Mesolithic of Palestine: in The
Annual of the British School at Athens 37 (1936-37), pp. 123-127. Morsch, M.
2002      Magic figurines? Some remarks about the clay objects of Nevali Cori: in H.
Gebel et al. (eds), Magic Practices and Ritual in the Near Eastern Neolithic,
Volume 8, Berlin, 145-162. Perrot, J., Ladiray, D.
1988      Les Sepultures: in J. Perrot (ed.), Les homes de Mallaha, Eynan, Israel,
Paris. Wright, G.
1978 Social differentiation in the Early Natufian: in C. Redman et al. (eds), Beyond Subsistence and Dating, New York.
Personal Adornment in the Epi-Paleolithic of the Levant
715
Fig. 1: Reconstructed group of individuals wearing personal adornment based on el-Wad and Mallaha prototypes.
Fig. 2: Reconstruction of el-Wad H 23.
Based on photograph and text in Garrod and Bate 1937: 18, PI. VIE.
716
Cherra Wyllie, Frank Hole
Fig. 3: Reconstruction of el-Wad H 25.
Based on photograph and text in Garrod and Bate 1937: PL VI2 and Garrod 1940: 124.
Personal Adornment in the Epi-Paleolithic of the Levant
717
Fig. 6: Reconstruction of MallahaH43.
Based on Perrot and Ladiray 1988: Fig. 22, and el-Wad H 28 (Garrod 1940:126).