Neolithic Statues from "Ain Ghazal: Construction and Form CAROL A. GRISSOM Abstract This paper focuses on fabrication of statues in the Reassembly of five large lime plaster statues from second of the two caches excavated at'Ain Ghazal. In the seventh millennium B.C. following their excava- particular, it will endeavor to show that the material tion in 1985 at 'Ain Ghazal, Jordan, provided an op- limitations of plaster used on a large scale played an portunity to examine evidence of their construction. important role in determining statue forms. More- For the most part, the statues had lain horizontally during fabrication, and they were made in stages by over, it will be argued that, in part because of formal applying plaster to reed bundles bound with cordage. limitations, the statues were likely adorned with Full-scale replication of a bust and a figure provided other materials, for example, clothing and wigs or additional information about construction. Plaster headgear, during display. probably had to be applied while armatures were hor- Three types of evidence are integrated to support izontal because of the statues' large size; broad, flat, and simple forms were made as a consequence. these In or-hypotheses: construction details visible in the der to fashion the complex standing figures, legs original were plaster, information acquired during repli- modeled separately and joined to the rest of the cation statue of two statues, and the forms and fabrication with plaster. Statues may have been accessorized of similarwith PPNB statues. Construction details were wigs, clothing, or other materials to simulate observed human during reassembly of statues in cache 2 at appearance.* the Smithsonian Institution and are considered pri- INTRODUCTION mary evidence, especially the excellent impressions left by entirely disintegrated reed-and-cordage arma- The discovery during the 1980s turesof twoon interior caches surfaces of of statue fragments. Such extraordinary PPNB (Pre-Pottery evidence Neolithic was thoroughly B) plas- documented during treat- ter statues at the site of 'Ain Ghazal ment in because Jordan it would has become inaccessible for study aroused considerable interest among after thearchaeologists statues were reassembled and prepared for as well as the wider public. Numbering museum more display.2 than Full-scale 30 replicas of a large bust in total, the caches' busts and standing and standing figures figure were are made in order to better in- among the world's oldest known large-scale terpret primary statues, evidence. The process provided an solidly dated to the seventh millennium unexpected B.C. dividend, (uncal- showing that statue forms had ibrated); they are exceptionally well to bepreserved. relatively flat, From shallow, and simple to be success- a visual standpoint the faces are fullycompelling, made. Closely with related statues in the first cache black bitumen decoration drawing excavated attention at 'Ain Ghazal to thein 1983 were examined eyes. Furthermore, because PPNB societies while they were in pre-the process of treatment at the literate, the meaning of the statues University remains of London, enig- giving access to construction matic, inviting speculation. Archaeology evidence. PPNB has plaster pro- statue fragments in the Ash- vided little additional information about the use of molean Museum in Oxford and the Musee du Louvre PPNB plaster statues because most have been exca- in Paris were examined in less detail, in part because vated from pits where they were apparently buried restoration has obscured much evidence of fabrica- when no longer wanted for their original purpose.' tion. References to other PPNB material are based on Moreover, the practice of making large-scale plaster published photographs and written accounts. statues ceased after the PPNB, so that there are no OVERVIEW OF PPNB PLASTER STATUES later exemplars to be used as an aid in interpretation. Thus, the principal source of information about The first modern discovery of PPNB plaster stat- meaning is the corpus of PPNB plaster statues itself. ues was made by Garstang at Jericho in 1935. He * I am especially indebted to Kathy Tubb and well Sueas Patricia Gil- Griffin, Ann Gunter, Zeidan Kafafi, Anne bert for access to the cache of statues excavated at 'Ain Liegey, Robert Mark, Gary Rollefson, and Denise Ghazal in 1983, and to Roger Moorey, Elisabeth Fontan, Schmandt-Besserat for their contributions to this paper. and Annie Caubet for access to Jericho material. I would I Garfinkel 1994. also like to thank my valued colleague Rae Beaubien, as 2 Grissom 1997. 25 American Journal of Archaeology 104 (2000) 25-45 This content downloaded from 160.111.254.17 on Thu, 20 Sep 2018 13:41:40 UTC All use subject to https://about.jstor.org/terms 26 CAROL A. GRISSOM [AJA 104 found two caches Givenof therelatively fragmentary nature of earlier realistically finds, dis- rendered statues, each cache covery said of the first to cache contain of plaster statues ata 'Ain"man" that was about life-size, a "woman" (with breasts) half that Ghazal in 1983 was a remarkable event.11 The num- size, and a doll-sized "child."3 The only known frag- ber of statues was large (26), consisting of busts and ments of these caches, however, are those of a "man" standing figures with arms, and most statues were in the collection of the Rockefeller Museum inJerusa- substantially complete although badly broken and lem, consisting of a pair of legs and a well-preserved distorted since burial. Radiocarbon dating of two head decorated with shell eyes and radially painted samples of charcoal closely associated with the stat- lines,4 and miscellaneous fragments in the Musee du ues produced almost the same dates, 6750 + 80 B.C. Louvre in Paris, including a shoulder, a smaller pair (uncalibrated) and 6710 + 80 B.C. (uncalibrated).'2 of legs (one with a six-toed foot), and a separate frag- Furthermore, the technique of blocklifting the entire mentary foot.5 Kenyon, excavating at Jericho in the cache for laboratory excavation and conservation 1950s, found a cache of at least three highly stylized treatment at the Institute of Archaeology in London busts without sculpted features.6 The most complete resulted in preservation of essentially all material.1 bust, painted yellow overall with eyes and nose pos- Discovery of the second cache of statues at 'Ain sibly outlined by dark paint, is in the Jordanian Ar- Ghazal occurred the following year at the end of the chaeological Museum in Amman. From the same excavation season, so that the contents of the pit find are a similar but more fragmentary yellow- were not blocklifted until 1985.14 They were subse- painted bust (1958.771) and a large red-painted sec- quently transported to a facility of the Smithsonian tion (1958.772), apparently the base of a bust; both Institution near Washington, D.C., where they were are in the collection of the Ashmolean Museum in excavated and conserved.'5 Based on radiocarbon Oxford.7 Kenyon believed the statues she found dating to of charcoal found below them, statues in the be later than those found by Garstang, but she notedsecond cache were buried after 6570 + 110 B.C. (un- that the stratigraphic position of the Garstang mate-calibrated),16 thus indicating that they are later than rial was "not very certain."8 Plaster fragments recently those in the first cache. Five statues were reassembled: found in the Nahal Hemar Cave have well-preserved two standing figures and three two-headed busts impressions of reeds and cordage.9 Although (fig. too 1). Unattached fragments include two additional limited to reconstruct, they are believed to be heads,parts one of which is fragmentary. Because of their of at least four statues because analyses show different four sizes and associated torso fragments, these distinct material compositions.10 two heads had probably been incorporated in sepa- -Garstang 1935, 166-7; Garstang and Garstang 1940, no. 1964.698). The latter have also been analyzed and 57-8. their results tabulated with those of bust fragments (Goren 4 Both the head and legs were found in deposit 195 and Segal 1995; Kingery, Vandiver, and Prickett 1988). The (Garstang 1935, pls. 25, 51, 52). Based on photographs interior surfaces of these fragments, however, are unusu- shown in Tubb and Grissom (1995, pls. 14, 15), the legs ally smooth, and I have found no convincing evidence that measure about 50 cm, essentially the same length as those the fragments formed parts of statues. of two-legged figures in cache 2 excavated at 'Ain Ghazal. 8 Kenyon 1960, 92. Based on a photograph shown in Garstang (1935, pl. 53), 9 Bar-Yosef and Alon 1988, 20-1, pl. 8. and Garstang and Garstang (1940, pl. 10), the head 10 Goren, Segal, and Bar-Yosef 1993. (IDAM acc. no. 35.3289; fig. 12 here) is a little smaller 1 Rollefson 1984. than that of statue #2 but larger than other heads found in 12 OxA-1473 and OxA-1472 respectively (Hedges et al. cache 2. Thus, the statue must have been somewhat less 1989, 228). Calibrated dates, in this case 7723 1 122 B.C. than life-size, like figures in cache 2. Radially painted lines and 7654 1 121 B.C., were done by probability distribu- on the head are often interpreted as depicting a beard. tion (Method B), % area enclosed (Gary Rollefson, per- 5 Measuring less than 40 cm in height, the legs in the sonal communication, 23June 1999). Louvre are smaller than those in Jerusalem and may be 13 Tubb 1985, 1987; Tubb and Grissom 1995. In a series those identified by Garstang as belonging to a woman. In of steps, surrounding soil was removed, and the deposit which of the two deposits these legs and the shoulder were was encased in polyurethane foam within a wooden crate, found is not known, but Garstang (1935, 167) notes that a built while it remained in situ.