CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART LATE NEOLITHIC POTTERY AND AMBIGUOUS SYMBOLS IN THE SOUTHERN LEVANT K. GIBBS Abstract: Compared to the Pre-Pottery Neolithic B (PPNB), evidence for potentially symbolic imagery is rare in the Late Neolithic of the Southern Levant, consisting primarily of figurines and decorated pottery. While the symbolic roles of figurines have occasionally been discussed, studies of decorated pottery have usually focused on developing chronologies or defining the boundaries of archaeological cultures. In an attempt to better understand pottery symbolism this paper suggests that over the course of the Late Neolithic pottery decoration became more varied but also became simpler. These developments may reflect an increase of symbolic ambiguity. The decline in figurine use over the course of the Late Neolithic similarly suggests a focus on ambiguous symbols during the later part of the period. A tentative suggestion is that this shift may be related to a change in community organization. Some early Late Neolithic sites are relatively large and each one may have comprised a single, spatially contiguous community. In the later part of the period, at least in some areas, community groups may have been dispersed across several small sites. A symbolic system emphasizing ambiguity may have facilitated the occasional interactions that were a part of such dispersed communities. Résumé : Dans la région du Levant Sud, les témoins potentiels d’une imagerie symbolique sont peu nombreux au Néolithique récent, contrairement au Néolithique précéramique B, et se composent principalement de figurines et de poterie décorée. Si le rôle symbolique des figurines a parfois été abordé par les chercheurs, les études portant sur la poterie décorée ont surtout servi à la mise en place de chronologies et à la définition de cultures archéologiques. Afin de mieux comprendre le rôle symbolique joué par la poterie, le présent article suggère que le décor des céramiques s’est non seulement diversifié, mais également simplifié tout au long du Néolithique récent. Ces développements pourraient être le reflet d’une ambiguïté symbolique croissante. Une diminution de l’utilisation des figurines au cours du Néolithique récent pourrait aussi refléter une nouvelle préférence pour les symboles ambigus à la fin de cette période. Il est possible qu’une telle transformation soit reliée à un changement dans l’organisation sociale des communautés qui est perceptible à la même époque. Au début du Néolithique récent quelques sites très étendus suggèrent l’occupation d’un espace contigu par des communautés entières, tandis que la fin du Néolithique récent voit une dispersion des communautés entre plusieurs petits sites, au moins dans certaines régions. Un système symbolique mettant l’accent sur l’ambiguïté a pu faciliter les interactions sociales épisodiques que nécessitait une telle dispersion des communautés. Keywords: Levant; Late Neolithic; Pottery; Figurines; Symbolism; Communities. Mots-clés : Levant ; Néolithique récent ; Poterie ; Figurines ; Symbolisme ; Communautés. INTRODUCTION Evidence for potentially symbolic imagery is relatively rare in the Late Neolithic (or Pottery Neolithic) of the Southern Levant (ca 6500-5200 cal. BC) (fig. 1). While the preceding Pre-Pottery Neolithic B (PPNB) period has produced examples Paléorient, vol. 39.2, p. 69-84 © CNRS ÉDITIONS 2013 of human statuary, figurines, plastered human skulls and stone masks (Kuijt and Goring-Morris 2002), in the Late Neolithic potentially symbolic imagery is limited largely to figurines and decorated pottery (Verhoeven 2002). Although several scholars have commented on the possible symbolic meanings of figurines and their roles in Late Neolithic society (e.g., Garfinkel et al. Manuscrit reçu le 2 mai 2012, accepté le 12 juin 2013 CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART 70 K. GIBBS have facilitated a shift to a more dispersed system of community organization in the later part of the Late Neolithic. MATERIALS AND METHODS Fig. 1 – Map of Southern Levant showing Late Neolithic sites mentioned in the text. 2010; Gopher and Orrelle 1996; Miller 2002; Stekelis 1972), studies of decorated pottery have more often focused on developing chronologies and defining the boundaries of cultural groups. Considerations of the social or symbolic roles of pottery and the potential meanings pottery decoration communicated within Late Neolithic communities are rare. In fact, identifying the specific symbolic meanings of particular designs or decorative motifs on Late Neolithic pottery may be extremely difficult and, indeed, it is probably wrong to assume that any symbol in the Late Neolithic had a single fixed meaning. However, by examining the development of decorated pottery within a wider social context it may be possible to gain some insight into its symbolic roles and how these changed over time, even if it is not possible to comment on specific meanings (Campbell 2010). Without trying to interpret the specific meanings of decorative motifs, this paper suggests that pottery symbolism became more ambiguous over the course of the southern Levantine Late Neolithic. It considers the possibility that this change gave Late Neolithic people the potential to interpret the symbols of community identity in different ways. This may This paper discusses pottery from Late Neolithic sites in the Southern Levant, relying on published data as well as evidence from sites in Wadi Ziqlab, Northern Jordan. It is restricted to a discussion of decorated pots because these may have played particular ritual or symbolic roles (Verhoeven 2002). Furthermore, decorated pots are often better represented than undecorated vessels in published reports resulting in a greater availability of data. It is worth noting, however, that it is somewhat artificial to separate decorated and undecorated pottery in discussions of pottery’s social and utilitarian roles and future research would benefit from more integrative approaches to Late Neolithic pottery. I consider the degree of variability of decoration types within two phases of the Late Neolithic as well as the complexity of decorative motifs. While statistical methods can be used to assess the degree of variability in pottery design (e.g., Graves 1994) the general lack of suitable quantifiable data from Late Neolithic sites renders these kinds of analyses problematic. As a result, I rely on more general observations of the types of decorations present in each phase. While this is far from ideal, it may provide a useful step towards understanding the significance of variation in Late Neolithic pottery decoration. Similarly, the complexity of decorative motifs is assessed by general observations as noted by others (e.g., Kerner 2001). Ideally, a comparison of two phases of the Late Neolithic will be based on stratified sites where both phases are present. This would help to eliminate some of the issues that can arise when comparisons are made between sites, such as researchers having different ways of classifying and quantifying pottery, and differential levels of preservation between sites. However, few sites in the Southern Levant span multiples phases of the Late Neolithic period. Munhata is an exception and, importantly, pottery from the site has been published (Garfinkel 1992 and 1999), so I rely heavily on that site in this paper. It should be mentioned, however, that in some pit contexts at Munhata pottery was assigned a phase (Yarmukian or Wadi Rabah) depending on its typological characteristics rather than its stratigraphic context. Also, data is not available for pits that were perceived to have a ‘mixed’ pottery assemblage (Garfinkel 1992: 19). This may have some implications for understanding change in pottery decoration as potentially transitional contexts may be overlooked. The site of Paléorient, vol. 39.2, p. 69-84 © CNRS ÉDITIONS 2013 CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART Late Neolithic Pottery and Ambiguous Symbols in the Southern Levant Sha’ar Hagolan is also considered as it is broadly contemporary with the earliest phase of occupation at Munhata. It was analysed using a comparable approach; and published data is available (Garfinkel 1999; Garfinkel and Miller 2002). I also refer to evidence from two sites in Wadi Ziqlab, al-Basatîn and Tabaqat al-Bûma. These sites both date to the later part of the Late Neolithic (e.g., Banning et al. 2011; Gibbs et al. 2010). LATE NEOLITHIC DECORATED POTTERY In addition to its symbolic imagery, the southern Levantine Pre-Pottery Neolithic is well known for its contributions to the origins of sedentism and agriculture, and for the later emergence of ‘mega-sites’ with populations estimated in the thousands (Kuijt and Goring-Morris 2002; Simmons 2007; Twiss 2007). While the Pre-Pottery Neolithic has generally been viewed as a period of increasing material and cultural complexity, the Late Neolithic has often been interpreted as a reversal to a less complex way of life characterized by smaller agricultural settlements and, perhaps, groups of nomadic pastoralists. Kenyon (1970: 60), for example, described the Late Neolithic occupants of Jericho as foreigners who brought about ‘a great retrogression’. Although some have pointed out that this picture of the Late Neolithic is coloured largely by biases in the history of Neolithic research (Banning et al. 1994; Garfinkel and Miller 2002; Simmons 2007: 199), study of the period remains underdeveloped compared to the seemingly more socially, symbolically, and technologically complex PPNB. The introduction of pottery is the most obvious technological advancement of the Late Neolithic and pottery has figured prominently in attempts to define Late Neolithic cultural groups and to date archaeological sites (e.g., Garfinkel 1999). A general cultural framework consisting of three cultural groups—the Yarmukian, Jericho IX, and Wadi Rabah—was established by the 1950s and has changed little since (BenDor 1936; Kaplan 1958; Stekelis 1972). In this framework the Yarmukian is the earliest pottery using group, dating roughly to the second half of the 7th millennium cal. BC. Dating the Jericho IX is more problematic due to a scarcity of radiocarbon dates from relevant sites. It may be somewhat later than the Yarmukian but the degree of temporal overlap is unclear (Banning 2007). The Wadi Rabah group is the latest, dating to the 6th millennium cal. BC (Banning 2007). Some sites have produced pottery assemblages that have resisted assignment to neat cultural groupings and there is sometimes disagreement about how particular assemblages relate Paléorient, vol. 39.2, p. 69-84 © CNRS ÉDITIONS 2013 71 to one another. In part, this is due to the low intensity of Late Neolithic research, which has often resulted in small, poorly published pottery assemblages that lack supporting radiocarbon dating evidence. However, another part of the problem may be that clearly bounded cultural groups did not, in fact, exist in the Late Neolithic. If so, attempts to impose rigid organizational frameworks will ultimately prove unsuccessful no matter how many sites are excavated and published. In this paper, I retain the terms Yarmukian, Jericho IX, and Wadi Rabah because their familiarity renders them convenient for representing temporal divisions but it is important to acknowledge that the spatial and temporal boundaries between any Late Neolithic groupings were likely rather permeable. The diachronic changes that I highlight in this paper should be viewed more as gradual shifts than sudden cultural replacements. The later part of the Late Neolithic, including sites attributed to the Wadi Rabah cultural group, has stimulated considerable debate and disagreement (Banning 2002). For example, Gopher and Gophna (1993) attempt to differentiate ‘normative’ Wadi Rabah sites from ‘variant’ ones. Garfinkel (1999), however, attributes many of these same variant sites to a later ‘Middle Chalcolithic’ period, which Braun (2008) has in turn rejected. As a more specific example, many archaeologists acknowledge that the site of Abu Hamid in the Jordan Valley has evidence for a Wadi Rabah (or ‘variant’) occupation but disagree on whether it is the basal-level assemblage or the middle-level assemblage (Banning 2007; Lovell et al. 2007). In general, Wadi Rabah pottery assemblages contain a range of vessel shapes, including deep and shallow bowls with rounded, straight, flaring, inverted or carinated profiles; platters and large bowls; pedestalled bowls (or chalices); pithoi; and a range of jars, including necked and holemouth jars. A jar with a bow-rim is considered characteristic of Wadi Rabah assemblages. Pots have a wide range of decorative treatments including slipped, burnished, incised, impressed, combed, painted and applied motifs (fig. 2) (Garfinkel 1999; Gopher and Gophna 1993). In contrast to the later part of the Late Neolithic, there tends to be more agreement about what pottery assemblages should be classified as “Yarmukian” or “Jericho IX”, although the nature of the relationship between these two earlier groups is itself somewhat unclear. Yarmukian and Jericho IX assemblages consist of a range of vessel shapes including a variety of bowls and cups, chalices, platters, basins, pithoi, necked jars and holemouth jars (Garfinkel 1999; Gopher and Gophna 1993). While the majority of Yarmukian sherds are either undecorated or have evidence of a red slip or paint entirely covering one surface, some pots have evidence of a characteristic surface CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART 72 K. GIBBS Fig. 3 – Examples of typical Yarmukian (A-D) and Jericho IX (E-F) pottery decoration. Incised herringbone and red paint or slip (A-B), other incised decoration (C-D), and paint with burnish (E-F). Further examples can be observed in Garfinkel (1999). Fig. 2 – Examples of Wadi Rabah decoration. A-C) combed; D-E) incised; F-G) stylus-impressed; H) comb-impressed; I) painted. Further examples can be observed in Garfinkel (1999). treatment consisting of a combination of straight or zigzag incised frames, typically filled in with incisions in a herringbone pattern (fig. 3). These bands are often adjacent to fields or bands of red paint or slip. The primary distinguishing feature of Jericho IX pottery is the dominance of red painted and burnished bands, zigzags and triangles on a cream or pale pink slip. It should be noted that incised decoration does occur in limited amounts in Jericho IX pottery assemblages just as small numbers of painted motifs occur in Yarmukian assemblages. Although the proportion of incised versus painted decoration is different between Yarmukian and Jericho IX assemblages, many of the motifs are similar, especially the prevalence of ‘V’ shapes, triangles, bands and zigzags, perhaps indicating a degree of cultural overlap or permeable cultural boundaries. Accepting the difficulties with attributing specific Late Neolithic pottery assemblages to rigid cultural groupings, especially during the later part of the period, in this paper I want to draw attention to two general developments in pottery decoration that seem to have occurred between the earlier part of the Late Neolithic and the later part, apart from the differences in the actual motifs and techniques used. First, the range of decorative techniques and motifs appears to increase over time. Given the nature of available datasets this is somewhat difficult to quantify and pottery decoration can be described and grouped in different ways. Nevertheless, decoration in the earlier part of the Late Neolithic seems to be less variable. For example, at Munhata (layer 2B) Yarmukian decoration is limited primarily to slip, paint and incisions (table 1). Although other motifs do occur, incised straight and zigzag bands filled in with herringbone pattern are particularly obvious and dominant in the site’s Yarmukian assemblage. Some sherds have the incised frames characteristic of this type of decoration but these are left blank or filled in with short incisions rather than herringbone pattern. When taken together these account for 42.4% of decorated sherds from layer 2B at Munhata and 30% of decorated sherds from Sha’ar Hagolan. Painted decoration comprises 17.9% of the decorated sherds from layer 2B at Munhata and 9.9% of the decorated sherds from Sha’ar Hagolan. Many of the primary motifs are similar to those found on incised pots, consisting of straight lines, zigzag pattern, and triangles, although other motifs such as painted dots do occur. It is worth pointing out that the actual proportion of vessels with incised herringbone decoration could be higher than the numbers reported by Garfinkel (1992; Eirikh-Rose and Garfinkel 2002) if many of the red-slipped or painted sherds initially derived from the un-incised parts of vessels with herringbone pattern. The motifs used to decorate Wadi Rabah pottery tend to be more varied and new decorative techniques appear during the later part of the Late Neolithic (Garfinkel 1999: 142-147). Gopher and Gophna (1993: 332) recognize this “broad range” of surface treatments as characteristic of Wadi Rabah pottery. At Munhata, burnished, combed, impressed, roulette, and Paléorient, vol. 39.2, p. 69-84 © CNRS ÉDITIONS 2013 CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART Late Neolithic Pottery and Ambiguous Symbols in the Southern Levant Table 1 – Percentages of pottery decoration types at four sites (including data from Garfinkel 1992; Eirikh-Rose and Garfinkel 2002). Note that percentages for indented entries are also counted in the main heading for each type of decoration. An ‘x’ indicates that the decoration type is present at a site but percentages are not available. Also note that for Munhata 2B and Sha’ar Hagolan the entries for “slip” do not include slipped and burnished sherds, while at Munhata 2A, Tabaqat al-Bûma and al-Basatîn the entries for “slip” include slipped sherds with or without a burnish. For Sha’ar Hagolan the entry for burnish includes slipped sherds with or without a burnish while at Munhata 2A, Tabaqat al-Bûma and al-Basatîn the entries for burnish indicate sherds with a burnish and no evidence for a slip. In this table decoration consisting of herringbone incisions within an incised frame is combined with incised frames containing other kinds of incisions or no incisions (under “frames”). This is due to the general similarity of technique and primary motifs. Decoration Munhata 2B Sha’ar Hagolan Munhata 2A Tabaqat al-Bûma alBasatîn Slip 38,6 57,9 86,4 85,7 51,0 1,7 x 4,7 0,8 0,4 0,8 x x Burnish Paint 17,9 9,9 Wide painted lines 13,0 8,0 Thin painted lines 4,9 1,9 Net pattern x x Incisions 43,5 30,0 3,2 2,4 Frames (includes herringbone) 42,4 28,5 x x Other 1,1 1,5 Combing Impressions <0,4 Stylus, various shapes ‘Fingernail’ x Comb 4,1 x x x 8,3 4,4 38,1 1,0 1,2 4,8 x x x x x x x Roulette Appliqué x x <0,4 0,6 0,7 1,1 Table 2 – Percentages of decorated sherds within pottery assemblages at four sites. The entry for al-Basatîn is derived from the Late Neolithic layers of one area (P33) within the site. Sites Yarmukian Wadi Rabah Munhata 12,8 6,2 Sha’ar Hagolan 15,4 Tabaqat al-Bûma (levels 3-5) 5,2 al-Basatîn (area P33) 5,3 appliqué decorations appear for the first time in Wadi Rabah contexts (layer 2A), while slip and thin painted lines continue. These techniques were used to apply a range of motifs. For example, combing can be applied as straight, intersecting, or curvilinear bands, or can cover the entire surface of a sherd. Paléorient, vol. 39.2, p. 69-84 © CNRS ÉDITIONS 2013 73 Impressions can be made with a comb, or with a stylus having a triangular, round, rectangular, or ‘fingernail’ shape. Incised herringbone pattern persists into Munhata layer 2A but according to Garfinkel (1992: 91) it is “not so closely controlled” as Yarmukian herringbone decoration and typically is not contained within bands of incised frames. Sites in Wadi Ziqlab dating to the later part of the Late Neolithic also show evidence for a range of decorative motifs. The increase in decorative techniques during the later part of the Late Neolithic cannot simply be attributed to a greater prevalence of decoration in later assemblages. In fact, at Munhata the proportion of decorated sherds in the pottery assemblage actually dropped from 12.8% in the Yarmukian levels (layer 2B) to 6.2% in the Wadi Rabah levels (layer 2A) (Garfinkel 1992). Available data from Sha’ar Hagolan and Wadi Ziqlab are consistent with these numbers (table 2). At the Yarmukian site of Sha’ar Hagolan 15.4% of sherds were decorated (Garfinkel 1999: 59). At the site of Tabaqat al-Bûma, which dates to the later part of the Late Neolithic, approximately 5.2% of sherds show evidence of decoration, although it is worth noting that heavy carbonate deposits sometimes obscure the surface of sherds at this site (Banning et al. 2011). At the contemporary site of al-Basatîn it is more difficult to quantify the overall proportion of decorated sherds in the Late Neolithic assemblage and work is on-going. In some contexts there is the possibility of intrusive material from later Early Bronze (EB) deposits, particularly undecorated EB body sherds that can sometimes be difficult to differentiate from undecorated Late Neolithic sherds. However, in one secure Late Neolithic context the proportion of decorated sherds is 5.3%, which is similar to Tabaqat al-Bûma and to Wadi Rabah pottery from Munhata. The second general development that seems to have occurred over the course of the Late Neolithic is a shift to simpler decorative motifs during the Wadi Rabah phase. Kerner (2001) demonstrates that typical Yarmukian decoration would require up to six or more production steps to execute: incising parallel horizontal lines on the body of the vessel, incising parallel zig-zag lines, filling in each of the resultant bands with herringbone pattern, painting the adjacent areas while leaving the herringbone-incised zones unpainted, and adding additional decoration to other parts of the vessel. In contrast, Wadi Rabah vessels would only require two to four production steps, such as the application of slip and burnish over the entire surface of a vessel or decorating with fields of small impressions (Kerner 2001). How might we account for these shifts? One could argue that a greater range of decorative motifs in the later part of CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART 74 the Late Neolithic could result from communities becoming increasingly isolated and cultural ties breaking down; however, there is little evidence for this. Indeed, Goren’s (1992) petrographic analysis of Late Neolithic ceramics from Munhata shows an increase of non-local pottery during the site’s Wadi Rabah phase. Evidence for longer-distance contacts also occur at some sites attributed to the Wadi Rabah cultural group, such as the chlorite bowls found at Hagoshrim that were likely manufactured in the Northern Levant or Anatolia (Rosenberg et al. 2010). Another possibility is that changes in decoration relate to increasing craft specialization. The possibility of kiln-made Wadi Rabah pottery may support this (e.g., Goren 1992: 341) although more evidence is needed to demonstrate this conclusively, such as the excavation of kiln structures. At any rate, at some sites the fine black and red-burnished wares that are seen as evidence for kiln-made pottery are relatively rare; at Tabaqat al-Bûma, for example, these make up only about 1% of the ceramic assemblage. It seems possible, then, that much pottery production was not specialized and remained at the level of the household even if kilns were sometimes used. Kerner (2001 and 2010) also considers increasing craft specialization in her explanation for the overall simplification of pottery decoration during the later part of the Late Neolithic. She suggests that as a result of increasing specialization pottery became more standardized over time due to an increase in production efficiency, with the decoration of Wadi Rabah pottery requiring fewer production steps to complete when compared to earlier Yarmukian pottery. However, although Wadi Rabah decoration was simpler in that it involved fewer discrete steps to produce, it is not clear that its execution would always require an overall lower investment of time, especially the extensive burnishing and fields of small impressions that occur on some Wadi Rabah pots. Moreover, the suggestion of increased standardization does not easily explain the wider range of decorative motifs that seems to occur on Wadi Rabah pottery. It may be useful, then, to consider decorative changes in terms of pottery’s potential symbolic function. In the Southern Levant there have been few attempts to include pottery in interpretations of Late Neolithic symbolic imagery, with more emphasis being placed on figurines (Simmons 2007: 217-219; see below). Goren and Gopher (1995) do emphasize the symbolic potential of early pottery in their discussion of changing pottery fabrics over the course of the Late Neolithic. They point out that early pottery was often made using calcareous pastes. Although these pastes were not particularly efficient for the production of cooking pots, Gopher and Goren suggest K. GIBBS that they were intentionally selected because they resulted in light-coloured vessels that were conducive to being decorated with symbolically important designs in slip or paint. They note a decrease in calcareous pastes over the course of the Late Neolithic and presumably a decrease in pottery’s symbolic potential. However, even during the later part of the Late Neolithic, light-coloured calcareous pastes dominate, comprising 64% of Wadi Rabah pottery (Goren and Gopher 1995: 24). Gopher and Goren (1995) only hint at the specific messages symbolized by Late Neolithic pottery decoration. Orrelle and Gopher (2000) make a more explicit attempt to interpret the meanings of this symbolism, which they use to explain changes in decoration over time. They suggest that pottery imagery reflects Late Neolithic gender roles, which complements their earlier interpretation of Late Neolithic figurines (Gopher and Orrelle 1996). Their interpretation of Yarmukian pottery emphasizes incised and painted zigzag and ‘V’ motifs, bands filled in with herringbone or ‘nested V’ incisions, and the red colour of the paint. Looking to Palaeolithic imagery as an analogy, they argue that the repeated ‘V’ or triangle motifs on Yarmukian pottery represent female genitalia while red paint represents menstrual blood. Significantly, a single Yarmukian vessel could have multiple ‘V’ motifs, thus representing multiple females. Their discussion of pottery decoration from Wadi Rabah sites calls attention to both the lack of repeated ‘V’ motifs and the high proportion of decoration on the rims of Wadi Rabah vessels, typically as a band of red slip or paint around the orifice (Orrelle and Gopher 2000). Like the red paint and ‘V’ motifs on Yarmukian pottery they suggest that the decorated rim symbolizes a menstruating vulva; they state that: “It is possible, for example, that the decorative transition from the Yarmukian triangle/vulvae arranged in a unified pattern to the single isolated, Wadi Raba vulva symbols on the woman pot, represent a deliberately altered use of the vulva symbol from one of solidarity to one of isolation. By extension this may imply that the view of menstruation changed from one of empowering to that of polluting” (p. 302). Orrelle and Gopher’s attempt to explicitly discuss the symbolic value of pottery decoration and to integrate it into a wider symbolic context is laudable. However, the specific meanings attributed to particular decorative motifs (e.g., ‘V’ = vulva, red = blood) are debatable and the system of communication suggested by their study is perhaps overly rigid. Moreover, while their discussion could contribute to understanding why particular pottery motifs become simpler over time (i.e., moving away from complex motifs that represent multiple women) it does Paléorient, vol. 39.2, p. 69-84 © CNRS ÉDITIONS 2013 CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART Late Neolithic Pottery and Ambiguous Symbols in the Southern Levant not directly consider the greater diversity of decorative motifs found on pottery dating to the later part of the Late Neolithic. AMBIGUOUS SYMBOLS Gopher and Goren (1995) and Orrelle and Gopher (2000) demonstrate the potential for addressing symbolism in discussions of southern Levantine Late Neolithic pottery, including why pottery decoration changed over the course of the Late Neolithic. However, they do not entirely account for the two general shifts suggested above—an increase in variability of decorative motifs during the later part of the period, and a shift to simpler designs. Here I offer a different suggestion that emphasizes symbolic potential in terms of these general trends rather than looking for more specific or rigid meanings for particular decorative motifs. I suggest that these developments reflect an increase in the ambiguity of pottery symbolism. In other words, the symbolic messages—whatever they may have been—conveyed by the less varied and more complex decoration on Yarmukian and Jericho IX pots were relatively fixed, while Wadi Rabah symbolism had a greater potential range of meanings. This increase in symbolic ambiguity is reflected in the greater variability observed within and among Wadi Rabah pottery assemblages and in the simpler motifs used to decorate Wadi Rabah pots. In fact, all material symbols contain an element of ambiguity. As Cohen (1985: 21) notes, “Symbols are effective because they are imprecise”. Symbols may be meant to communicate some meaning, but this meaning emerges through engagements with people rather than being inherent in the symbol itself (Thomas 1996: 97-98; Robb 1998: 337-339). A person’s experiences affect how meanings are interpreted, making symbols potentially polyvalent or multi-referential. However, a more rigid symbolic framework that includes a narrower range of regularly repeated imagery may serve to focus one’s interpretation to a more fixed intended meaning, leaving less room for interpretation. This perspective recalls McLuhan’s (1964) distinction between “hot” and “cool” media. McLuhan is perhaps better known for coining the phrase “the medium is the message” which he used to explore the impacts of new technologies or media: “The ‘message’ of any medium or technology is the change of scale or pace or pattern that it introduces into human affairs” (McLuhan 1964: 24). The significance of a technology, then, is not simply how it is used, but rather, how its introduction or development impacts and shapes people’s lives and their environments. In the case of prehistoric pottery, the spe- Paléorient, vol. 39.2, p. 69-84 © CNRS ÉDITIONS 2013 75 cific foods and other products that were cooked, processed or stored in pots were of less importance than the ways pottery shaped scheduling patterns, and altered economic and social strategies. In viewing technologies in this way McLuhan’s work may be compatible with recent studies of materiality and other archaeological approaches that emphasize the relationship between people and things (e.g., Hodder 2012). McLuhan recognized that material technologies (or media) play a crucial role in forming social environments but he suggested that the level of human participation in the dynamic between people and things was variable, with some media encouraging little participation and others requiring more. To him, a “hot” medium is one that provides abundant information and therefore requires little participation on the part of the human “audience”. In contrast, “cool” media contain little information and therefore encourage more participation. This allows more flexibility in the ways cool media are approached and interpreted; in other words, cool media are more ambiguous. The greater range of human engagement permitted by cool media is less likely to exclude particular perspectives or interpretations. In McLuhan’s terms, Yarmukian and Jericho IX pottery decoration could be described as relatively “hot” while Wadi Rabah pottery is comparatively “cool”. As a more ambiguous symbol, Wadi Rabah pottery decoration would be potentially more inclusive of a wider range of interpretations. It is worth stressing here that in this context the medium that is changing is pottery decoration, rather than pottery itself. At present it is difficult to say how the impact of pottery technology more generally developed over the course of the Late Neolithic. The introduction of new methods of production (e.g., kilns, moulds) has been proposed (e.g., Goren 1992: 341; Yannai 1997) but more research is needed to assess the evidence for these changes and their potential impacts on Late Neolithic society. For this reason this discussion is restricted to pottery decoration as a distinct medium with its own developmental trajectory over the course of the Late Neolithic. Ambiguous symbols have been discussed by archaeologists and ethnoarchaeologists working in other contexts. For example, in a discussion of the British Neolithic, Thomas (2005) notes the prevalence of ambiguous symbols, including decorated Grooved Ware pottery: “These symbols had associations that were fully understood, but their meanings were ambiguous and multiple, and that ambiguity or polyvalence was the source of their efficacy as tools of ritual practice” (p. 173). He suggests the emphasis on ambiguous symbols may help to explain the scarcity of depictions of the human form CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART 76 in the British Neolithic, especially when compared to southeast Europe. Representations of a specific, unambiguous form would not have fit into the accepted system of symbolism. In a discussion of Halafian painted pottery from the Northern Levant and upper Mesopotamia, Campbell (2010) suggests that naturalistic designs were more explicit symbols than abstract designs. Explicit symbols were required when new meanings and narratives were introduced into the ceramic repertoire. It should be noted, however, that Campbell views the abstract designs on Halafian pottery as comprising a relatively rigid symbolic system with a high level of convention rather than an ambiguous symbolic system. Yarmukian and Jericho IX decoration, being both abstract and relatively rigid, may have comprised a similar system. It is possible that in the specific context of Halafian pottery there is no particularly ambiguous system of symbolism comparable to the one proposed for Wadi Rabah pottery in the Southern Levant. Spielmann et al. (2006) argue that potters in New Mexico in the 17th century AD began to decorate glaze ware bowls with ambiguous symbols and obscured images. They see this as a deliberate act of resistance; potters hid the meaning of their symbols from Spanish missionaries who were attempting to eradicate Pueblo religious practices. Cunningham (2009) discusses how decorated enamel pots in modern Mali are ambiguous symbols of the status of recently married women. As objects with functional ties to food preparation and serving, they suggest “acquiescence to patriarchal authority” (Cunningham 2009: 289), yet as decorative items displayed in the home as part of a woman’s trousseau they symbolize her economic independence and power. The narrower range of decorative motifs on Yarmukian and Jericho IX pottery may indicate that any symbolic information conveyed by pottery decoration may have been more focused and fixed in the earlier part of the Late Neolithic than messages communicated by later pottery, which, with less of an emphasis on specific motifs and design elements, seems to have been more ambiguous. The apparent simplicity of later decoration further suggests its potential ambiguity as a material symbol. Danesi (2009) argues that simple symbols tend to have a greater potential range of meanings than more complex ones. He uses the simple symbol X as an example, showing how in the modern world it can stand for a wide range of things, including youth culture (GenX), a mistake, the Roman numeral ten, a kiss, Christ, comic book superheroes (X-men) and the mathematical function of multiplication, among other things. In the Late Neolithic of the Southern Levant, the simple decoration of Wadi Rabah pottery may have similarly permitted the conveyance of a range of meanings. K. GIBBS Fig. 4 – Yarmukian figures. A) after Garfinkel et al. (2002: fig. 13.11); B) after Stekelis (1972: pl. 46.1). It is worth briefly commenting on the ways figurines changed over the course of the Late Neolithic as they represent the period’s other obvious class of potentially symbolic imagery (Verhoeven 2002). Several types of figurines have been recovered from Yarmukian sites (Gopher and Orrelle 1996), with seated cowrie-eyed figurines and incised pebble figurines receiving the most attention (fig. 4). A number of explanations have been proposed to explain their occurrence on Yarmukian sites. Some emphasize the female characteristics of many of the seated cowrie-eyed figurines and suggest that they played a role in fertility magic or were representations of a female deity (Garfinkel 2004; Stekelis 1972), or perhaps represented a specific deified ancestor or ‘matron’ figure (Miller 2002: 232). Others have suggested that some of these figurines combine female and male characteristics and that they should be interpreted in the context of gender mutability (Gopher and Orrelle 1996; Peterson 2010: 258). A small number of seated cowrieeyed figurines are decorated with incised herringbone patterns that recall the major decorative motif of Yarmukian pottery (Garfinkel 2004), suggesting that there may be parallels in the symbolic roles of some pots and figurines. Incised pebbles have also been interpreted in different ways. Some incised pebbles have what appear to be facial features, especially eyes, and Garfinkel (2004) argues that these Paléorient, vol. 39.2, p. 69-84 © CNRS ÉDITIONS 2013 CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART Late Neolithic Pottery and Ambiguous Symbols in the Southern Levant may also represent a goddess figure. Other incised pebbles have more abstract designs, including parallel or cross-hatched lines. Gopher and Orrelle (1996) suggest that these represent different stages of female physical development. Eirikh-Rose (2004), however, suggests that these abstract pebbles are not representations of the human form at all but, rather, are representations or symbols of individual or family identities. In contrast to Yamukian sites, figurines or other representations of the human form from Jericho IX and Wadi Rabah sites are almost entirely absent. An incised pebble recovered at the Wadi Rabah site of Abu Zureiq (Garfinkel and Matskevich 2002: Figs. 18.1 and 21) and one from a reportedly Jericho IX context at Tel Te’o (Eisenberg et al. 2001: fig. 9.6: 7) both exhibit stylized human features. The most notable potential Wadi Rabah representation of the human form, however, is not a figurine. Rather, it is found as decoration on a unique Wadi Rabah pottery vessel from the site of ‘En Jarba (Kaplan 1969). This pot is decorated with an appliqué figure on each side. Garfinkel (2002) interprets these as male dancers, perhaps wearing animal masks, although they could represent animals or animal skins instead (Bar-Yosef 1992: 39) or perhaps animal-human hybrids. There have been few attempts to explain why figurines and representations of the human form decline in the later part of the Late Neolithic. Rollefson (2008a) suggests that this represents the culmination of a long-term trend and that figurines had, in fact, been declining in number since the Middle PPNB. However, a general shift to more ambiguous symbolism could also explain the decline. Realistic depictions that clearly refer to a specific form could serve to focus symbolic interpretations to rather specific meanings, which may not fit with a symbolic system that emphasizes ambiguity. As noted above, Thomas (2005) suggests that the scarcity of representations of the human form in the British Neolithic might be explained by an emphasis on ambiguous symbolism. The same could hold true for the later part of the Late Neolithic in the Southern Levant. CHANGING COMMUNITIES Decorated pottery and figurines can both be interpreted in terms of an increase in symbolic ambiguity during the later part of the Late Neolithic. But what might account for this shift? The current extent of data makes any answer to this question necessarily tentative. However, considering the potential roles of decorated pottery in Late Neolithic ritual and society Paléorient, vol. 39.2, p. 69-84 © CNRS ÉDITIONS 2013 77 may be informative. Certainly, pottery may have been valued as a useful container technology but it was also an object with potential symbolic or social significance. Nieuwenhuyse et al. (2010: 83) suggest that the material properties of durability and portability made early Near Eastern pots useful tools in the manipulation of social relations. Decorated pots, in particular, could have been ‘marked’ with individual biographies or roles in processes of display and exchange. Similarly, Campbell (2010) suggests that painted Halafian pots can be viewed as agents that conveyed and enforced social information. In a discussion of Neolithic ritual symbolism in the Levant, Syria and southeast Anatolia, Verhoeven (2002) proposes that decorated pottery was particularly important for regulating social relationships at the level of the community. It was used to transmit messages concerning group membership and therefore played an important role in community organization. More generally, ethnoarchaeological evidence from a range of context supports the idea that pottery can play a role in conveying social information and marking social boundaries. For example, the Kalinga of Northern Philippines have a social organizational unit above the level of the settlement that is referred to as a region. Graves (1994) demonstrates how Kalinga pottery is distinctive between regions and homogeneous within each region and argues that this is due, in part, to potters intentionally selecting contrastive designs. Hodder (1982) argues that pottery is used to mark group membership by people living around Lake Baringo in Kenya. This process is most obvious in areas where different groups are competing for scarce resources. In other cases pottery styles may correspond with social groups without actively being used to mark social boundaries. For example, in western Kenya the Luo people make pots that result in ‘micro-styles’ that correspond to particular potter communities. Differences between microstyles may be recognizable by both potters and non-potters but they do not seem to be actively used as identity markers, particularly by pottery consumers (Herbich 1987: 202). Given the possible role of pottery in negotiating social relationships it may be productive to consider changes in pottery decoration in the context of changing social organization, although we should not assume that early pottery in the Southern Levant necessarily had the same role as pottery in other regions of the world including other parts of the Near East. Building on Verhoeven’s (2002) suggestion that pottery played specific roles in rituals at the level of the community, an increase in symbolic ambiguity could have been related to change in the way communities were organized during the later part of the Late Neolithic, with a new organizational system emerging that benefited from a less rigid symbolic system. CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART 78 Unfortunately, there are few studies of Late Neolithic community organization in the Southern Levant, either at the level of the individual site or from a regional perspective (Banning et al. 1994). However, some recent fieldwork projects may hint at a change in settlement patterns that could reflect a shift in community organization between the Yarmukian and Wadi Rabah phases. The lack of data related to Jericho IX sites and settlement patterns means that little can be said about relevant community organization. Villages developed in the Southern Levant over the course of the Pre-Pottery Neolithic with ‘mega-sites’ appearing to the east of the Jordan Valley towards the end of the PPNB. These were large, aggregate sites over 10 ha in area with evidence for dense occupation, including two-storey houses (Kuijt and Goring-Morris 2002: 424). While determining population size for these sites is complicated (Kuijt 2008: 289-290), estimates range into the thousands (Kuijt 2000). The end of the PPNB has been described as a collapse that resulted in the abandonment of sites, a drastic decrease in population, and a shift to much smaller settlements. As recently as the 1970s scholars could speak of a settlement hiatus that left the Southern Levant unoccupied after the PPNB until new pottery-using settlers re-colonized it centuries later (de Vaux 1966; Kenyon 1970; Perrot 1968). However, there is evidence for a degree of settlement continuity, with some sites remaining occupied after the ‘collapse’ of the PPNB (e.g., Rollefson et al. 1992; Simmons et al. 2001) and new sites being founded soon after, and there is evidence that during the earlier part of the Late Neolithic some sites were quite large. The Yarmukian occupation at ‘Ain Ghazal, which remained occupied throughout the ‘collapse’, may have still covered 12 hectares (Rollefson 2008b) although population density seems to have declined resulting in a population of 400-500 people (Simmons 2007: 215). The renewed excavations at Sha’ar Hagolan have uncovered the remains of an extensive Yarmukian settlement, perhaps 20 hectares in area (Garfinkel and Miller 2002), with evidence for large courtyard structures, passageways and a well (Ben-Shlomo and Garfinkel 2009; Garfinkel et al. 2006). These features have led the excavators to interpret the site as a community with a three-tier system of social organization: public features like the streets and the well are thought to indicate organization at the level of the wider settlement; courtyard structures are seen as a means of organizing smaller groups of people, perhaps extended families; and groups of dwelling and storage rooms inside each courtyard structure are interpreted as the residential units of individual nuclear families (Ben-Shlomo and Garfinkel 2009). As Banning (2010) notes, however, it is not clear that all the K. GIBBS structures were contemporaneous and we cannot be certain of the development of each building complex. It is possible that architecture developed as a “gradual infilling of the space” (Banning 2010: 58) rather than as a result of top-down decision making at the scale of the community or extended family. Excavations at other Yarmukian sites have not been as extensive so it is more difficult to assess community size. Wadi Shu’eib, like ‘Ain Ghazal and Sha’ar Hagolan, was probably quite large (Rollefson 2008b) but other sites such as Jebel Abu Thawwab (Kafafi 2001) were considerably smaller. BenShlomo and Garfinkel (2009: 198) suggest that Sha’ar Hagolan was a regional centre in a hierarchical settlement system but, in fact, the relationship between large and small sites is not well understood and at present there seems to be little indication of social inequality or large sites having influence over smaller ones. What does seem clear, however, is that some Yarmukian communities were large and contiguous, continuing in some form a pattern that existed in the preceding PPNB. Evidence from Wadi Ziqlab in Northern Jordan suggests that a different system of social organization occurred here during the 6th millennium cal. BC (Banning 2001; Kadowaki et al. 2008). Recent excavations at the site of al-Basatîn recovered evidence for a small agro-pastoral hamlet that was engaged in a range of domestic activities (Banning et al. 2005; Kadowaki et al. 2008; Gibbs et al. 2010). Sickle elements and grinding stones point to the harvesting and processing of grain. Faunal remains are dominated by sheep and goat, with fewer cattle and pigs (Kadowaki et al. 2008). Small numbers of spindle whorls indicate limited textile production. Pottery from the site shows parallels with assemblages attributed to Wadi Rabah assemblages, including small amounts of red and black burnish, and impressions, combing and incisions on vessel surfaces (fig. 5) (see also Kadowaki et al. 2008: figs. 9 and 10). The distribution of artifacts and lack of substantial architecture at the site suggest a small population, perhaps consisting of one or two households. Al-Basatîn is one of several sites in Wadi Ziqlab that have been interpreted as small agro-pastoral hamlets. Earlier excavations at Tabaqat al-Bûma, about 6 km to the east of al-Basatîn, also produced sickle elements, grinding stones, spindle whorls, and pottery exhibiting similarities with Wadi Rabah assemblages (Banning et al. 1992 and 1996). An architectural analysis suggests that only one or two households was occupied at the site at any given time in the later part of the Late Neolithic (Kadowaki 2007). Interaction among the small sites in the wadi very likely occurred—they are too small to have been endogamous, occasional interaction could have facilitated access to certain kinds of materials such as Paléorient, vol. 39.2, p. 69-84 © CNRS ÉDITIONS 2013 CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART Late Neolithic Pottery and Ambiguous Symbols in the Southern Levant 79 Fig. 5 – Pottery from Tabaqat al-Bûma (A-G) and al-Basatîn (H-Q). Painted (A-C); combed (D,I-M); incised lines (E, O); appliqué (F-G, N); stylus-impressed (P) and comb-impressed (Q). basalt and shell (Kadowaki et al. 2008: 123), and interaction could have encouraged labour-sharing for certain tasks such as grain harvesting (Banning 2001). In some ways, the pottery from al-Basatîn and Tabaqat al-Bûma is similar, suggesting a degree of contact between the sites. For example, necked jars are not common at either site and the absence of bowrim jars at both sites is notable as these often occur on Wadi Rabah sites (Garfinkel 1999). But there are enough differences between al-Basatîn and Tabaqat al-Bûma, in terms of architectural style, material culture, and faunal remains, to suggest that interaction was not always regular or sustained. There is no clear evidence for a larger, regional centre that could have co-ordinated interaction among the sites and, as in the earlier part of the Neolithic, there is no indication of a hierarchical settlement pattern. Instead, the sites may have comprised a dispersed community that was arranged linearly along the course of the wadi (Banning 2001; Kadowaki et al. 2008), with different small sites or congeries of individuals engaging in occasional and intermittent interaction. In contrast to PPNB mega-sites and large Yarmukian sites like Sha’ar Hagolan and ‘Ain Ghazal, the Wadi Ziqlab community during the later part Paléorient, vol. 39.2, p. 69-84 © CNRS ÉDITIONS 2013 of the Late Neolithic appears not to have been spatially contiguous. The boundaries of the community and those of the small sites that comprised it were not congruent. Indeed, the boundaries of dispersed communities can potentially be flexible and rather ill-defined. The work of Goody (1956) in Ghana provides an ethnographic example of a linearly-organized, dispersed group with no central focus and lacking clear boundaries: “When we examine the system of group designations used in this region, we finds that it is based not upon a series of exclusive tribal names, but upon a ‘directional’ system in which a number of contiguous peoples refer to themselves obliquely by means of two names” (p. 17). In Goody’s research area, the term ‘Lo’, along with a related term ‘Dagaba’ (or ‘Dagaa’), was used by congeries of people to refer to their neighbours and, depending on context, sometimes themselves. Loose clusters of people referred to those living to the east as ‘Lo’ and those living to the west as ‘Dagaba’. But adjacent groups in either direction used the same system of naming, calling groups to the east Lo and groups to the west CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART 80 Dagaba. As a result, group boundaries continually shifted, as any particular congeries of people perceived themselves to be at the boundary between Lo and Dagaba. While Goody’s work may represent a somewhat extreme example, any shift to a more dispersed system of community organization is likely to have altered the nature of community relationships and boundaries. In the Late Neolithic of the Southern Levant, large sites like Sha’ar Hagolan may have comprised distinct communities; residents would have interacted on a regular basis and it would have been relatively clear who was a member of the community and who was not. As the boundaries of the community were more or less congruous with those of the site, co-residence would be a primary factor of membership or affiliation. The similar backgrounds of the residents, based on regular proximate interactions, would mean that material symbols were more likely to be interpreted in similar ways, which would be solidified by a relatively rigid symbolic system, such as that represented by Yarmukian pottery. Everyone engaging with Yarmukian pottery would have realized their affiliation in a particular community so opportunities for interpretation of material symbols would be less crucial. In contrast, in a dispersed community with potentially flexible boundaries affiliation would be less obvious. While interaction likely occurred among the various small groups of people that collectively comprised Wadi Ziqlab’s Late Neolithic dispersed community, this would have been less regular than in a large congruous community. This may have been particularly true for people residing at different ends of a dispersed community that was arranged linearly along a river valley. In this case it is likely that community affiliation was somewhat flexible. Living apart, individuals would have had more freedom to emphasize or play down relationships with others who lived in different locations, depending on specific social or economic contexts. If for whatever reason a group of people did not want to emphasize their affiliation with those living to the west, they could emphasize their alignment with those living to the east. As a result, community identity in this context may have been more situational than during the earlier part of the Late Neolithic. In such a dispersed community characterized by intermittent and flexible interaction between its members, rigid or unambiguous symbols of community affiliation may have been undesirable. By emphasizing fixed meanings these symbols would not have encapsulated the variability in backgrounds and experiences that characterized the members of the dispersed community. A rigid symbolic system may have served to isolate some members by strongly emphasizing certain specific mean- K. GIBBS ings that may be unfamiliar or unappealing. In contrast, a more ambiguous symbolic system would have provided people from different parts of the dispersed community greater freedom to interpret the symbols of community affiliation in their own way, thus mitigating against possible tensions in the ways symbols are interpreted. By promoting more flexible interpretations, the simple yet varied symbolism of Wadi Rabah vessels would be perceived as familiar to all community members even if the specific meanings assigned to the symbols were not identical. Ambiguous symbols could have facilitated the creation of new relationships within and, perhaps, beyond the boundaries of the community. Interaction among sites may have included social occasions where the serving of food was involved, such as marriages, initiation rites, and work-sharing parties (Banning 2001; Kadowaki et al. 2008). Decorated pots could have played a role during these occasions as serving vessels, making it a potential symbol of community membership. The participants in these events would be more able to interpret these ambiguous symbols of community membership in their own way and in the context of their own backgrounds, potentially easing interactions with others who were seen only irregularly or perhaps not obviously members of the same community. CONCLUSION This paper discusses two possible developments that occurred over the course of the southern Levantine Late Neolithic. First, there was an emphasis on more ambiguous symbolism in the later part of the period, as indicated by changes in pottery decoration as well as a decline in figurine use. Second, there was a shift in some cases from large contiguous communities to a more dispersed system of community organization. Furthermore, it suggests a connection between these two developments, with a system of more ambiguous symbolism facilitating interactions that occurred on an irregular basis amongst people who were less clearly part of the same community group. Given the current state of Late Neolithic evidence, these two developments and the connections between them must remain somewhat speculative. Evidence for pottery decoration becoming more ambiguous relies on limited data from a small number of sites. It would be preferable to have in-depth quantitative evidence from a large number of sites but this is currently not available. It is encouraging, however, that others have noted the broad range of decoration that occurs in Wadi Rabah assemblages symbols and the new decorative tech- Paléorient, vol. 39.2, p. 69-84 © CNRS ÉDITIONS 2013 CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART Late Neolithic Pottery and Ambiguous Symbols in the Southern Levant niques developed in the later part of the Late Neolithic (e.g., Garfinkel 1999: 147; Gopher and Gophna 1993), which I have interpreted as evidence for an increase in symbolic ambiguity. Furthermore, the documented decline of figurines as a nonambiguous symbol seems to corroborate this suggestion. Evidence for a shift from large, contiguous settlements to a more dispersed system of social organization is also somewhat limited. In fact, some sites dating to the earlier part of the Late Neolithic appear to have been quite small when compared to contemporary sites like Sha’ar Hagolan and ‘Ain Ghazal, so the picture is undoubtedly more complex than the one painted in this paper. The general lack of regional surveys designed to look for Late Neolithic settlement patterns makes it difficult to assess the significance of these sites or how widespread dispersed communities were during the later part of the Late Neolithic (Banning et al. 1994). It is possible that the differences in community organization outlined here were also influenced by geographical location rather than being the result of any straightforward diachronic shift. It is worth pointing out, however, that the area around Hazorea in the Jezreel Valley shows a fairly dense concentration of sites dating to the 6th millennium and perhaps a similar system of dispersed settlements existed there (Banning et al. 1994; Yannai 2006: 279-280). At any rate, it is safe to say that more work on Late Neolithic settlement patterns and social organization is needed. Given the current state of research, it may be tempting to refrain from engaging with more theoretical topics in Late Neolithic archaeology until further empirical date becomes available through fieldwork, analysis and publication. However, proposing new ideas and models, even somewhat speculative ones, may serve to stimulate interest in the period, which may in turn result in increasing research and improved datasets. Philip (2011: 193) acknowledges the importance of suitable data but also notes that Late Neolithic research is lacking a set of theories and concepts through which debate can be conducted. This has resulted in an ‘intellectual gap’ leaving the Late Neolithic one of the most poorly understood in the archaeology of the Southern Levant, despite a number of recent projects that have focused on the period. Many studies of Late Neolithic pottery have emphasized the creation of typologies for chronology-building and the definition of cultural groups. However, pottery studies can certainly be taken in other directions and can help to fill this gap by contributing to the development of new theoretical approaches. Paléorient, vol. 39.2, p. 69-84 © CNRS ÉDITIONS 2013 81 It is becoming increasingly clear that perspectives that view the Late Neolithic as a period of devolution or collapse are difficult to justify. In the earlier part of the period sites were sometimes large, continuing a pattern that existed in the PPNB, while a new system of dispersed communities may have emerged by the 6th millennium cal. BC, at least in some areas. In both cases, people would have developed complex relationships to maintain their communities, although the nature of these relationships may have changed over time. Symbolic imagery was likely one dimension to this process and symbolism would have been important during the Late Neolithic much as it was during the PPNB (e.g., Gopher and Orrelle 1996; Orrelle and Gopher 2000). Indeed, appreciating the symbolic dimension of pottery may help resolve other issues in southern Levantine Late Neolithic research, including basic typological classification. As mentioned above, there is considerable disagreement about how 6th millennium cal. BC sites should be grouped even at a relatively basic level, especially outside of the Wadi Rabah core area in the Jezreel Valley. The limited intensity of studies that have focused on Late Neolithic sites has certainly contributed to this confusion. But another contributing factor may be that, in some cases, Late Neolithic pots were intended to be ambiguous by the people who made and decorated them. ACKNOWLEDGEMENTS An earlier version of this paper was presented to the Department of Archaeology at the University of Manchester and the current version has benefited from discussions stemming from that presentation. I would like to thank Lisa Maher, Karina Croucher and three anonymous reviewers for commenting on a previous draft of this paper, and Karine Taché for assistance with the French abstract. I would also like to thank Ted Banning for giving me the opportunity to examine Late Neolithic material from Wadi Ziqlab. Excavations at al-Basatîn were directed by T. Banning, S. Kadowaki and the author. This paper was written with financial support from the Social Sciences and Humanities Research Council of Canada. Kevin GIBBS Department of Archaeology University of Aberdeen St. Mary’s Building Aberdeen AB24 3UF Scotland – UNITED K INGDOM kevin.gibbs@abdn.ac.uk CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART 82 K. GIBBS BIBLIOGRAPHY BANNING E.B. 2001 Settlement and Economy in Wadi Ziqlab During the Late Neolithic. Studies in the History and Archaeology of Jordan VII: Jordan by the Millennia: 149-156. Amman: Department of Antiquities. 2002 Consensus and Debate on the Late Neolithic and Chalcolithic of the Southern Levant (Reviews). Paléorient 28,2: 148-155. 2007 Wadi Rabah and Related Assemblages in the Southern Levant: Interpreting the Radiocarbon Evidence. Paléorient 33,1: 77-101. 2010 Houses, Households, and Changing Society in the Late Neolithic and Chalcolithic of the Southern Levant. Paléorient 36,1: 49-87. BANNING E.B., GIBBS K. and K ADOWAKI S. 2005 Excavations at Late Neolithic al-Basatîn, in Wadi Ziqlab, Northern Jordan. Annual of the Department of Antiquities of Jordan 49: 229-243. 2011 Changes in Material Culture at Late Neolithic Tabaqat Al-Bûma, in Wadi Ziqlab, Northern Jordan. In: LOVELL J.L. and ROWAN Y. (eds.), Culture, Chronology and the Chalcolithic. Theory and Transition: 36-60. Oxford and Oakville: Oxbow Books and the Council for British Research in the Levant. BANNING E.B., R AHIMI D. and SIGGERS J. 1994 The Late Neolithic of the Southern Levant: hiatus, settlement shift or observer Bias? The perspective from Wadi Ziqlab. Paléorient 20,2: 151-164. BANNING E.B., R AHIMI D., SIGGERS J. and TA’ANI H. 1996 The 1992 Season of Excavations in Wadi Ziqlab, Jordan. Annual of the Department of Antiquities of Jordan 40: 29-49. BANNING E.B., DODS R., F IELD J.J., KUIJT I., MCCORRISTON J., SIGGERS J., TA’ANI H. and TRIGGS J. 1992 Tabaqat Al-Buma: 1990 Excavations at a Kebaran and Late Neolithic Site in Wadi Ziqlab. Annual of the Department of Antiquities of Jordan 36: 43-69. BAR-YOSEF O. 1992 The Neolithic Period. In: BEN-TOR A. (ed.), The Archaeology of Ancient Israel: 10-39. New Haven: Yale University Press. BEN-DOR I. 1936 Pottery of the Middle and Late Neolithic Periods In: GARSTANG J., BEN-DOR I. and FITZGERALD G.M. Jericho: City and Necropolis (Report for Sixth and Concluding Seasons). Annals of Archaeology and Anthropology 23: 77-90. BEN-SHLOMO D. and GARFINKEL Y. 2009 Sha’ar Hagolan and New Insights on Near Eastern Proto-Historic Urban Concepts. Oxford Journal of Archaeology 28,2: 189-209. BRAUN E. 2008 Deconstructing Y. Garfinkel’s ‘Beth Shean Ware’ and ‘Middle Chalcolithic’: A Realistic Look at Two Archaeological Chimeras. Journal of the Serbian Archaeological Society 24: 81-108. CAMPBELL S. 2010 Understanding Symbols: Putting Meaning into the Painted Pottery of Prehistoric Northern Mesopotamia. In: BOLGER D. and MAGUIRE L.C. (eds.), The Development of Pre-State Communities in the Ancient near East: Studies in Honour of Edgar Peltenburg: 147-155. Oxford: Oxbow Books. COHEN A.P. 1985 The Symbolic Construction of Community. London: Tavistock. CUNNINGHAM J.J. 2009 Pots and Political Economy: Enamel-Wealth, Gender, and Patriarchy in Mali. Journal of the Royal Anthropological Institute 15,2: 276-294. DANESI M. 2009 X-Rated: The Power of Mythic Symbolism in Popular Culture. New York: Palgrave Macmillan. EIRIKH-ROSE A. 2004 Geometric Patterns on Pebbles: Early Identity Symbols? In: PELTENBURG E. and WASSE A. (eds.), Neolithic Revolution: New Perspectives on Southwest Asia in Light of Recent Discoveries on Cyprus: 145-162. Oxford: Oxbow Books. EIRIKH-ROSE A. and GARFINKEL Y. 2002 The Pottery. In: GARFINKEL Y. and MILLER M.A. (eds.), Sha’ar Hagolan, Volume 1: Neolithic Art in Context: 86-138. Oxford: Oxbow Books. EISENBERG E., GOPHER A. and GREENBERG R. 2001 Tel Te’o: A Neolithic, Chalcolithic, and Early Bronze Age Site in the Hula Valley. Jerusalem: Israel Antiquities Authority (IAA Reports 13). GARFINKEL Y. 1992 The Pottery Assemblages of the Sha’ar Hagolan and Rabah Stages of Munhata (Israel). Paris: Association Paléorient (Les Cahiers du CRFJ 6). 1999 Neolithic and Chalcolithic Pottery of the Southern Levant. Jerusalem: The Hebrew University of Jerusalem (Qedem 39). 2004 The Goddess of Sha’ar Hagolan: Excavations at a Neolithic Site in Israel. Jerusalem: Israel Exploration Society. GARFINKEL Y. and MATSKEVICH Z. 2002 Abu Zureiq, a Wadi Rabah Site in the Jezreel Valley: Final Report of the 1962 Excavations. Israel Exploration Journal 52,2: 129-166. GARFINKEL Y. and MILLER M.A. 2002 Sha’ar Hagolan Volume 1: Neolithic Art in Context. Oxford: Oxbow Books. GARFINKEL Y., BEN-SHLOMO D. and KORN N. 2010 Sha’ar Hagolan 3: Symbolic Dimensions of the Yarmukian Culture: Canonization in Neolithic Art. Jerusalem: Israel Exploration Society and the Institute of Archaeology, Hebrew University. GARFINKEL Y., KORN N. and MILLER M.A. 2002 Art from Sha’ar Hagolan: Visions of a Neolithic Village in the Levant. In: GARFINKEL Y. and MILLER M.A. (eds.): 188-208. GARFINKEL Y., VERED A. and BAR-YOSEF O. 2006 The Domestication of Water: The Neolithic Well at Sha’ar Hagolan, Jordan Valley, Israel. Antiquity 80,309: 686-696. Paléorient, vol. 39.2, p. 69-84 © CNRS ÉDITIONS 2013 CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART Late Neolithic Pottery and Ambiguous Symbols in the Southern Levant 83 GIBBS K., K ADOWAKI S. and BANNING E.B. 2010 Excavations at al-Basatîn, a Late Neolithic and Early Bronze I Site in Wadi Ziqlab, Northern Jordan. Annual of the Department of Antiquities of Jordan 54: 461-476. GOODY J. 1956 The Social Organisation of the Lowiili. London: Her Majesty’s Stationary Office. GOPHER A. and GOPHNA R. 1993 Cultures of the Eighth and Seventh Millennia BP in the Southern Levant: A Review for the 1990s. Journal of World Prehistory 7,3: 297-353. GOPHER A. and GOREN Y. 1995 The Beginning of Pottery. In: LEVY T.E. (ed.), The Archaeology of Society in the Holy Land: 224-225. New York: Facts on File. GOPHER A. and ORRELLE E. 1996 An Alternative Interpretation for the Material Imagery of the Yarmukian, a Neolithic Culture of the Sixth Millennium BC in the Southern Levant. Cambridge Archaeological Journal 6,2: 255-279. (Bibliotheca Neolithica Asiae Meridionalis et Occidentalis, and Yarmouk University, Monograph of the Institute of Archaeology and Anthropology 3). KAPLAN J. 1958 Excavations at Wadi Rabah. Israel Exploration Journal 8: 149160. 1969 ‘Ein El Jarba: Chalcolithic Remains in the Plain of Esdraelon. Bulletin of the American Schools of Oriental Research 194: 2-39. K ENYON K.M. 1970 Archaeology in the Holy Land (3rd ed.). New York: Praeger. K ERNER S. 2001 Pottery Decoration as a Medium to Examine Specialised Production in the Sixth to Fourth Millennia BC. Studies in the History and Archaeology of Jordan VII: Jordan by the Millennia: 157-162. Amman: Department of Antiquities. 2010 Craft Specialisation and its Relation with Social Organisation in the Late 6th to Early 4th Millennium BCE of the Southern Levant. Paléorient 36,1: 179-198. KUIJT I. 2000 GOREN Y. 1992 Petrographic Study of the Pottery Assemblage from Munhata. In: GARFINKEL Y. (ed.), The Pottery Assemblages of the Sha’ar Hagolan and Rabah Stages of Munhata (Israel): 329-348. Paris: Association Paléorient (Les Cahiers du CRFJ 6). 2008 GOREN Y. and GOPHER A. 1995 The Beginnings of Pottery Production in the Southern Levant: A Model. In: VINCENZINI P. (ed.), The Ceramics Cultural Heritage: 21-28. Faenza: Techna (Monographs in Materials and Society 2). GRAVES M.W. 1994 Kalinga Social and Material Culture Boundaries: A Case of Spatial Convergence. In: LONGACRE W.A. and SKIBO J.M. (eds.), Kalinga Ethnoarchaeology: Expanding Method and Theory: 13-49. Washington: Smithsonian Institution Press. HERBICH I. 1987 Learning Patterns, Potter Interaction and Ceramic Style among the Luo of Kenya. The African Archaeological Review 5: 193-204. HODDER I. 1982 2012 Symbols in Action: Ethnoarchaeological Studies of Material Culture. Cambridge: Cambridge University Press. Entangled: An Archaeology of the Relationships Between Humans and Things. Chichester: Wiley-Blackwell. K ADOWAKI S. 2007 Changing Community Life at a Late Neolithic Farmstead: Built Environments and the Use of Space at Tabaqat al-Bûma in Wadi Ziqlab, Northern Jordan. Unpublished PhD Dissertation. University of Toronto. K ADOWAKI S., GIBBS K., A LLENTUCK A. and BANNING E.B. 2008 Late Neolithic Settlement in Wadi Ziqlab, Jordan: al-Basatîn. Paléorient 34,1: 105-129. KAFAFI Z.A. 2001 Jebel Abu Thawwab (er-Rumman), Central Jordan. The Late Neolithic and Early Bronze Age I Occupations. Berlin: ex oriente Paléorient, vol. 39.2, p. 69-84 © CNRS ÉDITIONS 2013 People and Space in Early Agricultural Villages: Exploring Daily Lives, Community Size, and Architecture in the Late PrePottery Neolithic. Journal of Anthropological Archaeology 19: 75-102. Demography and Storage Systems During the Southern Levantine Neolithic Demographic Transition. In: BOCQUETAPPEL J.-P. and BAR-YOSEF O. (eds.), The Neolithic Demographic Transition and Its Consequences: 287-313. Dordrecht: Springer. KUIJT I. and GORING-MORRIS N. 2002 Foraging, Farming, and Social Complexity in the Pre-Pottery Neolithic of the Southern Levant: A Review and Synthesis. Journal of World Prehistory 16,4: 361-440. LOVELL J., DOLLFUS G. and K AFAFI Z. 2007 The Ceramics of the Late Neolithic and Chalcolithic: Abu Hamid and the Burnished Tradition. Paléorient 33,1: 51-76. MCLUHAN M. 1964 Understanding Media: The Extensions of Man. New York: Signet. MILLER M.A. 2002 The Function of the Anthropomorphic Figurines: A Preliminary Analysis. In: GARFINKEL Y. and MILLER M.A. (eds.), Sha’ar Hagolan. Volume 1: Neolithic Art in Context: 221-233. Oxford: Oxbow Books. NIEUWENHUYSE O.P., AKKERMANS P.M.M.G. and VAN DER P LICHT J. 2010 Not So Coarse, nor Always Plain–the Earliest Pottery of Syria. Antiquity 84: 71-85. ORRELLE E. and GOPHER A. 2000 The Pottery Neolithic Period: Questions About Pottery Decoration, Symbolism, and Meaning. In: KUIJT I. (ed.), Life in Neolithic Farming Communities: Social Organization, Identity, and Differentiation: 295-308. New York: Kluwer Academic/Plenum Publishers. PERROT J. 1968 La Préhistoire palestinienne. In : Dictionnaire de la Bible. Supplément 8, 43 : 286-446. Paris : Letouzey et Ané. CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART 84 K. GIBBS P ETERSON J. 2010 Domesticating Gender: Neolithic Patterns from the Southern Levant. Journal of Anthropological Archaeology 29: 249-264. SPIELMANN K.A., MOBLEY-TANAKA J.L. and POTTER J.M. 2006 Style and Resistance in the Seventeenth Century Salinas Province. American Antiquity 71,4: 621-647. PHILIP G. 2011 STEKELIS M. 1972 The Yarmukian Culture of the Neolithic Period. Jerusalem: The Magnes Press. ROBB J.E. 1998 The Later Prehistory of the Southern Levant: Issues of Practice and Context. In: LOVELL J.L. and ROWAN Y. (eds.), Culture, Chronology and the Chalcolithic. Theory and Transition: 192209. Oxford: Oxbow Books and the Council for British Research in the Levant (Levant Supplementary Series 9). The Archaeology of Symbols. Annual Review of Anthropology 27: 329-346. ROLLEFSON G.O. 2008a Charming Lives: Human and Animal Figurines in the Late Epipaleolithic and Early Neolithic Periods in the Greater Levant and Eastern Anatolia. In: BOCQUET J.-P. and BARYOSEF O. (eds.), The Neolithic Deomographic Transition and its Consequences: 387-416. Dordrecht: Springer. 2008b The Neolithic Period. In: ADAMS R.B. (ed.), Jordan: An Archaeological Reader: 71-108. London: Equinox. ROLLEFSON G.O., SIMMONS A LAN H. and K AFAFI Z. 1992 Neolithic Cultures at ‘Ain Ghazal, Jordan. Journal of Field Archaeology 19,4: 443-470. ROSENBERG D., GETZOV N. and ASSAF A. 2010 New Light on Long-Distance Ties in the Late Neolithic/Early Chalcolithic Near East. Current Anthropology 51,2: 281-293. SIMMONS A.H. 2007 The Neolithic Revolution in the Near East: Transforming the Human Landscape. Tucson: University of Arizona Press. THOMAS J.S. 1996 Time, Culture and Identity: An Interpretive Archaeology. London: Routledge. 2005 Ambiguous Symbols: why there were no figurines in Neolithic Britain. Documenta Praehistorica 32: 167-175. T WISS K.C. 2007 The Neolithic of the Anthropology 16: 24-35. Southern Levant. Evolutionary DE VAUX R. 1966 Palestine During the Neolithic and Chalcolithic Periods. Cambridge Ancient History, Vol. 1: 498-520. Cambridge: University of Cambridge Press. VERHOEVEN M. 2002 Transformations of Society: The Changing Role of Ritual and Symbolism in the PPNB and the PN in the Levant, Syria and South-East Anatolia. Paléorient 28,1: 5-13. YANNAI E. 1997 The Possible Origin of the Tournette? A Group of Ceramic Bowls Made in Stone Moulds from ‘En Asawir. Tel Aviv 24: 253-257. 2006 ‘En Esur (‘Ein Asawir) I: Excavations at a Protohistoric Site in the Coastal Plain of Israel. Jerusalem: Israel Antiquities Authority (IAA Reports 31). SIMMONS A.H., ROLLEFSON G.O., K AFAFI Z., MANDEL R.D., ALNAHAR M., COOPER J., KÖHLER-ROLLEFSON I. and DURAND K.R. 2001 Wadi Shu’eib, A Large Neolithic Community in Central Jordan: Final Report of Test Excavations. Bulletin of the American Schools of Oriental Research 321,1: 1-39. Paléorient, vol. 39.2, p. 69-84 © CNRS ÉDITIONS 2013