Morphological affinities and demography of metal-age Southwest Asian populations.код для вставкиСкачать
AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 59:47-60 (1982) Morphological Affinities and Demography of Metal-Age Southwest Asian Populations TED A. RATHBUN Department of Anthropology, University of South Carolina, Columbia, South Carolina 29208 KEY WORDS Skeletal remains, Southwest Asia, Metal ages, Morphology, Demography ABSTRACT Multivariate and cluster analyses of human skeletal remains of seven regional groups and nine individual sites and periods reveal different patterns of relationships for the sexes. This complex pattern of morphological a n ities probably depends on differential demographic features. Traditional morphological studies of the pop- than proper excavation and reportage of huulations of Southwest Asia from the metal ages man remains have hampered attempts to inhave attempted to trace migrations and recon- terpret the population dynamics in the area. struct racial histories of the area (Buxton and Preservation has been good at some sites, but Rice, 1931; Cappieri, 1969, 1970; Field, 1932, local conditions and vagaries of excavation in 1947; Finkel, 1974; Krogman, 1940; Rathbun, early periods of less rigorous archaeological 1972; Swindler, 1956). Much of the archaeo- control have resulted in skeletal collectionsthat logical literature focuses on trade, contact, and often are fragmentary, lack individual temmigration in discussions of the formation dy- poral designations, and in some instances have namics of state level organizations (Speiser, not been analyzed by professional physical an1930; Gelb, 1961; Halder, 1965, Moorey, 1970; thropologists. Even though such a state of afGibson, 1972; Adams, 1974; Cowgill, 1975; fairs is regrettable, lack of precision and ideal Wright and Johnson, 1975; Kohl, 1978). Al- data sets can be offset by analyses which prothough migration and population movements duce suggestions for subsequent investigation documented by protohistorical sources have as better samples become available. Even limbeen invoked to explain sociocultural dynam- ited results and interpretations, with defects, ics, univariate analyses of skeletal material are better than nothing at all, especially if they have revealed a basic stability of the mor- are regarded as suggestive and not necessarily phology of the populations in the area over definitive with provisions for revision as cirtime (Cappieri, 1969,1970; Finkel, 1974;Rath- cumstances warrant. bun, 1972, 1975). Metrics The present attempt at synthesis of regional The primary data in this study are drawn grouping data uses multivariate analysis to document the generalized morphological dis- from my own work with the skeletal collections tances among the populations in that area. The from Kish, Iraq; Hasanlu, Iran; and Dinkha cluster analyses of seven regional groupings Tepe, Iran (Rathbun, 1975, 1972, n.d.1. The and nine individual sites and periods indicate compiled data for the regional groups are from distinct patterns of relationships for males and Cappieri (1969).Data from individual sites used females. The analysis of individual popula- for comparisons include Nippur, A1 'Ubaid and tions from the Mesopotamian valley and the Ur in Iraq, and Tepe Hissar, Iran (Swindler, Iranian plateau reveals a more complex pat- 1956; Cappieri, 1970; Krogman, 1940). All of the populations analyzed data from the tern of morphological affinities which probably depends on the demographic features of the metal ages. The areal groups, A1 'Ubaid, and Ur are from the Copper and Bronze Ages, and groups. the remaining populations span the Iron Age. MATERIALS AND METHODS The approximate dates range from 3,500 B.C. Although archaeological excavation has been h i v e d September 8,1981;accepted March 16,1982 intense in Southwest Asia for some time, less 0002-9483/82/5901-0047$04.00 0 1982 ALAN R. LISS. INC. 48 T.A. RATHBUN to 200 B.C. Such large time spans are less than ideal for precise interpretations of population dynamics, but univariate analyses of individual sites and their components as well as superpopulations and regional groupings revealed a basic stability in the areas and major cultural divisions (Finkel, 1974; Cappieri, 1969, 1970). Small sample sizes from shorter time components at each site or in each region during the Copper through Iron Ages preclude accurate statistical manipulation. Lack of time precision may be partially compensated by larger pattern documentation which will be subject to revision as new data allow. Sites, regional groupings, sample sizes, sources, and temporal data are summarized in Table 1. The metric data on individuals from Kish, Hasanlu, and Dinkha Tepe were collected following the techniques described by Bass (19641, Vallois (1965), and Howells (1973). The reliance on published data carries with it the possibility of interobserver error which can cloud definitive statements, but the measurements included here are standard ones not requiring sophisticated instruments or extensive training. The landmarks are distinctive and readily located. Published methodological accounts for the analysis of groups not personally examined seem to be comparable, although slight interobserver error might contribute to minor mean differences. In a few cases small sample sizes necessitated the insertion of the average mean value and standard deviation from the Kish pooled data for a missing component in individual periods at Kish. The pooled data values from the site, rather than other statistical estimates, were used since the pooled data would span the range of variation for each period. The distance measures were calculated from the same variables for all groups. Standard descriptive statistics were calculated by the “SPSS condescriptive” procedures for Kish, Hasanlu, and Dinkha Tepe. Summary statistics for other groups came from published sources. The variables, values, and measures of generalized distances are presented in tabular form. Numerous statistical tools have been developed to study the affinities of human populations (Weiner and Huizinga, 1972; Constandse-Westermann, 1972). As Conuccini (1975) points out, each technique has its limitations and should be used for data that meet the basic statistical assumptions. After searching the literature for comparative materials and assessing the descriptive statistics from the Kish cultural periods, the Penrose (1954) generalized distance ( C 2 / ~ seemed ) most appropriate since it utilized means and standard deviations rather than demanding raw data unavailable for the published comparative materials. The correspondenceof the Penrose statistic with the more sensitive Mahalanobis one has been shown to be high (Hiernaux, 1964). TABLE 1 . Populations and sources for distance and cluster analysis 1. Troad 35 adult crania (Cappieri, 1969) 3200-1500 B.C. Ha-Nai-Tepe(16). Hissarlik (ll),Kum-Tepe (4), Muskebi (31, Baba Koy (1). 2. Anatolian: 93 adult crania (Cappieri, 1969) 3750-1300 B.C. Alishar Huyuk (271, OsmankayasiBoghaz Koy (16), Alaca Huyuk (13), Tilki-Tepe (12), Sheysk Huyuk-Tell Esh-Sheikh (5), Kusura (5), Ahlati-Li Be1 (5), Yumuk-Tepe (3), Polati Huyuk (2), Altin-Tepe (21, Buyuk Gulucek (l), Acem Huyuk (l),Evdi Tepesi (1). 3. Syro-Palestinian:78 adult crania (Cappieri, 1969) 3000-1400 B.C. Ras Shamra-Minet El Beida (33), Megiddo (32), Byblos (lo),Kafer Edj-Djarra (3). 4. Turkmenian: 77 adult crania (Cappieri, 1969) 4000-1200 B c Kara Tepe-Geoksur (371, Koktcha (28), Takhibray (3), Serakhs (2), Namazga-Tepe (2), Anau (2), Tup-Khona (l),Chust (l),Vuadil (1). 5. Indian: 103 adult crania (Cappieri, 1969) 3000-1500 B.C. Harappa (64), Mohenjo-Daro (12), Langhnaj (E), Lothal (8), Nevasa (4), Gedrosia (2), ChanhuDaro (l),Chandoli (l),Nal (l),Sialkot (11, Quetta (1). 6. Nippur: 35 adult crania (Swindler, 1956) 626-141 B.C. 7. A1 ’Ubaid 11 adult crania (Cappieri, 1970) ca. 3000 B.C. 8. Ur Pooled 11 adult crania (Cappieri, 1970) 3500-1700 B.C. 9. Tepe Hissar: 155 adult crania (Krogman, 1940) 3500-2000 B.C. 10. Hasanlu IV: 52 adult crania (Rathbun, 1972) 1100-800 B.C. 11. Dinkha Tepe: 30 adult crania (Rathbun, n.d.1 1900-800 B.C. 12. Kish neo-Babylonian:8 adult crania (Rathbun, 1975) 1150539 B.C. 13. Kish Early Dynasty 111: 22 adult crania (Rathbun, 1975) 2600-2350 B.C. 14. Kish Unknown: 61 adult crania (Rathbun, 1975) 2900-141 B.C. 15. Mesopotamian: 56 adult crania (Cappieri, 1969) 4800-1500 B.C. Kish (22), Ur 1141,A1 ’Ubaid (12), Tepe Gawra (4), Assur (3), Jemdet Nasr (1). 16. Iranian: 190 adult crania (Cappieri, 1969) 4100-1300 B.C. Tepe Hissar (156), Tepe Sialk (14), Shah-Tepe (ll),Hasanlu (51, Luristan (Tepe Bad Hora, Tepe Giyan, Tepe Jamshidi) (3). Tchoga Zambil (1). METAL-AGE SW ASIA The program was written and checked using the data and results of a similar study of North American Indians (Bass et al., 1971). The generalized distance values were used as input for the “Osiris hierarchical” cluster procedure. These values expressed the relative similarities of a whole set of variables as viewed in a matrix, and do not account for significance levels. The cluster procedure is relative and searches out the most similar populations to form an initial cluster. Each subsequent population added to the original cluster must show mutual affinities to all the previously included samples. Thus when a new population is added to the matrix, the whole pattern of relationships must be reevaluated. The results of the cluster analysis for area groups and individual sites and periods are discussed subsequently. 49 age. This limitation is expecially true for the Kish material because of the fragmentation and the necessity for the “indeterminate” adult category. Sex diagnosis depended on traditional Torphological criteria (Krogman, 1962; Hrdlicka, 1952; Anderson, 1962; Phenice, 1967; Bass, 1971). Age determinations were based on criteria from various functional areas (McKern and Stewart, 1957; Krogman, 1962; Montagu, 1960;Schour and Massler, 1944;Johnston, 1962; Bass, 1971). Because of the fragmentary condition of much of the material, less reliable gauges of age such as dental attrition and degenerative change (Hrdlizka, 1952; Brothwell, 1963) were sometimes used. Because of the limitations of the data, specific life tables were not produced, but the age at death by category, average age at death, and the population distribution for the Kish material and selected Southwest Asian sites were examined. Since the age at death for most individuals was indicated by a range, the average age at death was derived by multiplying the frequency by the midpoint of each age category and dividing by the total number of individuals of the same known sex. The percentages reflect the number of males, females, or subadults as the denominator in each category and exclude the sex and age indeterminates. The open-ended category 35+ was given an artificial end point of 45 years to calculate the average age a t death even though a few individuals undoubtedly lived beyond this age. This, as well as differences in aging criteria, partially accounts for the slightly younger average age at death than that found by other authors for related populations. It should be noted that many analyses combine the sex samples, whereas I have distinguished between them. Demography The demographic aspects of prehistoric populations provide a base for dealing with the dynamics of the groups. Reconstructing demographic features from skeletal material has certain limitations, but as long as these basic limitations are recognized, an attempt at reconstruction is justified. Howells (1960) has pointed out the problems of estimating total population size and the related features of circumscription of the group and the contemporaneity of the specimens, which is a problem at many complex Southwest Asian sites excavated without rigorous controls. Vallois (1960) mentioned the important factors of accurate sampling, considerations of the same degree of preservation within the sample, the conditions of excavation, and estimates of age at death. Hassan (1979) and Ubelaker (1978) have also discussed the potential and limitations of demographic analysis of skeletal populations. The limitations of the demographic data from Kish and selected Iranian sites should be menRESULTS tioned at the outset: 1) The total population Metrics size is almost impossible to determine, so the The basic cranial morphology of the Mesodata present represent only the individuals actually recovered, and 2) the subadult segments potamian and Iranian populations during the of the populations are underrepresented ap- metal ages remained relatively stable. This parently owing to poor preservation and the general Mediterranean complex included a long, vagaries of excavation. Differences in burial narrow, and medium height vault with a relpatterns do not seem to pose a problem, and I atively narrow face. Few traits usually assohave limited the analysis to individuals com- ciated with African or Central Asian populaplete enough to provide an 80-9070degree of tions such as alveolar prognathism, nasal form, flaring zygomatics, or shoveled incisors were acccuracy for sex and age determination. Although the techniques for estimating age notable. The major trend through time a t Kish, a t death from skeletal material have been re- Hasanlu, and Dinkha Tepe has been a defined, we are still often limited to statements creased robusticity, although both robust and in terms of categories rather than a specific more gracile individuals were present in most 50 T.A. RATHBUN cultural periods. Indications of robusticity reduction were notable in both the cranial and postcranial morphology of both sexes, but males tended to be significantly larger than females, especially for the mandible and face. At Kish, Hasanlu, Dinkha Tepe, and Tepe Hissar, both sexes in the earlier populations seem to have been more robust than the later populations a t the same sites. However, the proportions and shapes of the vaults and faces remained stable. A slightly divergent pattern existed at Kish, with the later neo-Babylonian males being more robust than others. The metric data for individual sites and the areal groupings are presented in Tables 2 and 3. When the areal groups are compared with the pooled Kish data and the cluster analysis (Fig. 1) is examined, interesting patterns of affinities appear. The male samples form two distinct clusters divided by the Syrian desert. The Kish males are definitely part of the eastern group. The relative distance of the Anatolian males from the more southern and eastern groups was not expected owing to the archaeological and historical evidence of con- tact between the areas. The total pattern of clustering does not indicate marked migration patterns, and a sudden influx of alien people is not evident. The clustering of the females in the areal comparison (Fig. 2) is more difficult to interpret. They do not form two distinct clusters. The Syro-Palestinian females are highly divergent from all the other samples. Two low level clusters are of interest. The close affinity of the Anatolian and Iranian females is different from the males. The cluster of the Kish females with this group instead of with the Mesopotamian grouping is surprising. Except for the highly distinct Syro-Palestinian females, the areal cluster analysis supports the hypothesis of basically similar and relatively homogeneous populations throughout Southwest Asia during the metal ages. Although comparing regional superpopulations with the pooled Kish sample proved suggestive of the population dynamics in the area, generalized distances for specific sites and the cultural groups at Kish were computed for finer comparisons (Table 4). It was also hoped TABLE 2. Data for Penrose C 1 / ~males ; Anatolian Troad N Max. Lt. Max. Br. Ba-Br. Ht. Up. Face Bizygo. Nasal Ht. Nasal Br. Max. Lt. Max. Br. Ba-Br. Ht. Up. Face Bizygo. Naeal Ht. Nasal Br. Mean S.D. N 19 186.60 7.90 18 141.80 6.40 14 138.60 5.30 10 68.60 4.30 6 131.20 3.20 10 51.10 3.20 10 23.90 0.80 Nipper pooled 52 52 26 26 27 25 26 N S.D. N 193.70 5.38 136.90 5.53 136.74 5.96 68.91 4.69 133.32 5.72 53.63 4.04 26.71 2.13 Dinkha Tepe pooled 7 8 5 6 7 7 7 23 22 19 23 22 22 24 Mean N Mean 14 12 4 10 8 10 13 188.29 137.42 137.25 72.50 134.50 52.70 23.69 S.D. N S.D. N 184.50 7.80 140.60 5.30 133.40 5.70 67.90 4.20 129.20 4.50 50.60 3.60 25.30 1.90 A1 'Ubaid Mean 29 30 19 17 18 17 17 Mean S.D. 192.90 5.00 140.60 3.60 136.80 4.20 70.50 5.20 127.70 4.60 53.50 2.90 25.80 1.60 Kish neoBabylonian Mean S.D. N 6 6 3 3 5 4 5 Mean S.D. 183.60 7.00 139.70 5.20 131.80 7.30 67.30 4.40 130.80 6.80 50.60 3.20 24.50 2.30 Ur Pooled N N 30 30 15 30 30 27 27 Mean S.D. N 191.70 137.20 144.30 76.70 132.40 53.00 25.80 7.00 3.90 2.50 1.20 3.20 3.40 2.10 112 Kish E.D.III Mean Mean S.D. N 189.50 11.20 135.70 6.40 138.30 6.90 70.40 5.40 129.70 6.40 50.70 2.70 25.60 1.80 Tepe Hissar Mean 188.206.20 111 134.10 99 134.70 108 70.00 96 127.10 103 50.70 107 25.40 7.33 5.90 5.56 3.34 7.98 3.30 2.87 'Values from all Kish periods pooled 4 4 2 1 1 1 1 30 Mean 184.50 31 134.90 13 134.77 29 69.10 14 125.00 31 51.26 31 24.00 Mean S.D. 187.20 7.50 136.00 5.10 134.40 5.20 69.70 4.70 130.10 6.90 51.00 3.80 25.90 2.10 Hasanlu N Mean S.D. 4.90 5.10 4.90 5.40 1.80 3.20 34 28 11 28 18 30 29 188.75 132.29 135.89 69.58 121.21 51.76 23.07 7.25 7.90 6.05 4.50 5.70 3.11 2.80 Unknown-Kish S.D. N 193.25 10.15 14 190.14 7.81 135.25 7.41 14 136.50 5.56 121.00 9.90 1 133.19' 7.39' 60.00 5.63* 1 78.00 5.63l 133.00 7 . W 2 126.00 7 . W 49.00 3.04' 1 56.00 3.04l 28.00 1.81' 4 24.00 2.00 47 94 36 39 22 43 42 S.D. S.D. ~~ Max. Lt. Max. Br. Ba-Br. Ht. Up. Face Bizygo. Nasal Ht. Nasal Br. Indian Turkmenian Syro-Palestinian 9.20 5.48 5.59 5.07 7.09 2.77 1.77 51 METAL-AGE SW ASIA that a time dimension would appear in the clusters to indicate direction of influence. The new matrix excluded Cappieri'sMesopotamian and Iranian regional groupings since data for individual sites were available. The cluster analysis for males including individual sites (Fig. 3) is generally similar to that of the areal one. The neeBabylonian males from Kish appear very different from all the other groups. Since there seems to be a trend T A B U 3. Data for Penrose Czl,. females Troad N Max. Lt. Max. Br. Ba-Br. Ht. Up. Face Bego. Nasal Ht. Nasal Br. Max.Lt. Max. Br. Ba-Br. Ht. Up. Face Bego. Nasal Ht. Nasal Br. 14 15 6 5 3 4 4 Mean Anatolian S.D. N Mean S.D. N 11 12 8 11 9 13 14 181.64 133.25 134.50 63.73 124.33 49.85 25.21 8.73 4.53 3.25 4.73 5.15 3.26 2.36 Dinkha Tepe p l e d Max. Lt. Max. Br. Ba-Br. Ht. Up. Face Bizygo. Nasal Ht. Nasal Br. S.D. N Mean 181.40 6.90 32 178.10 6.80 136.80 7.20 32 136.00 6.40 132.80 3.60 5 130.00 5.50 62.80 5.60 9 65.40 3.40 123.30 10.00 15 118.90 3.90 48.30 3.90 9 48.10 2.90 23.30 1.30 11 23.70 1.90 Nipper p l e d A1 'Ubaid N - 181.50 3.30 140.00 1.60 132.50 6.50 64.30 1.50 122.70 4.50 48.00 2.90 23.40 0.90 Kish neoBabylonian 5 5 1 1 1 5 5 Mean 187.20 133.80 136.00 65.00 120.00 46.00 24.00 Kish N Mean S.D. N Mean S.D. N 176.69 137.58 126.63 67.50 124.75 50.90 23.77 7.92 6.89 2.62 4.77 7.87 3.57 1.92 178.00 136.00 132.00 67.67 124.00 51.00 23.00 5.88 9.85 5.9P 4.16 5.88z 6.00 1.73 8 8 1 1 2 1 1 Mean Turkmenian S.D. N 38 38 22 37 36 31 31 Mean Indian S.D. N 183.10 7.30134.40 5.90 136.40 7.10 68.50 3.30 127.00 7.30 49.80 2.50 24.30 2.10 Tepe Hissar 40 41 34 34 21 45 42 Mean S.D. 179.70 5.70 131.80 9.40 128.60 6.30 63.60 4.30 122.50 5.30 46.60 4.10 24.30 2.10 Haeanlu S.D. N Mean S.D. N Mean S.D. 7.90 5.40 5.18l 3.97l 5.941 2.47 1.00 179.20 132.00 129.90 66.30 120.70 48.20 23.80 5.50 4.70 4.60 4.30 4.50 2.80 1.70 181.42 134.18 129.16 67.75 118.92 50.56 24.06 7.40 6.10 4.39 3.12 4.06 2.01 1.55 E.D.III 16 13 8 10 8 10 13 4 3 1 3 1 3 3 Mean 34 179.00 5.10 34 135.30 5.60 20 129.30 6.90 13 68.50 4.90 18 25.40 5.80 13 50.40 4.20 13 24.00 1.70 Ur pooled S.D. N Mean 4 4 4 3 2 4 4 Syro-Palestinian 43 43 38 42 32 41 39 18 15 4 14 10 14 13 Unknown-Kish S.D. N 181.25 5.87 132.00 4.24 131.49* 5.95z 65.6a2 3.732 115.00 7.07 48~34~3.222 29.00 2 . W 30 27 10 16 15 16 18 Mean S.D. 178.47 133.44 131.30 65.31 119.13 48.44 24.06 8.17 5.32 6.27 3.66 5.85 2.56 1.92 'Average value from all populationsin the series ~alues from all Kish periods pooled. 2 TABLE 4. Distance values for Penrose's CH 1 2 3 4 5 6 7 8 9 1 0.000 1.684 1.919 2.805 3.170 4.750 3.121 5.645 4.201 2 1.906 0.000 0.355 2.792 1.421 3.906 2.620 8.105 2.673 3 3.916 2.342 0.000 3.570 1.771 4.191 3.723 9.235 3.335 4 2.994 3.718 2.299 0.000 0.753 2.071 2.318 2.608 0.597 5 2.107 1.920 3.136 3.533 0.000 1.670 2.274 4.825 0.502 6 1.984 3.160 3.894 2.202 1.913 0.000 2.134 5.492 2.409 7 0.502 1.461 3.821 3.618 2.880 3.416 0.000 5.062 2.221 8 3.362 4.396 8.383 4.204 3.881 4.161 3.770 0.000 4.347 9 2.501 0.790 1.474 2.210 1.126 2.355 2.831 4.010 O.OO0 10 4.329 1.389 1.844 3.881 3.531 4.066 4.133 6.171 1.078 11 4.928 2.896 0.703 5.097 4.588 5.884 4.394 11.834 3.006 12 3.033 2.193 0.888 2.201 4.512 4.107 3.039 7.995 1.981 13 16.331 11.369 15.486 15.679 10.721 9.827 16.641 13.365 11.059 14 2.059 0.362 2.407 2.802 1.496 1.921 2.276 3.672 0.379 Note Males above the diagonal, females below. See Table 1 for population names. 10 11 12 13 14 6.913 2.474 22.596 11.716 3.108 7.152 4.069 14.768 10.923 2.397 8.031 3.425 13.896 11.210 2.902 3.670 2.597 18.957 9.079 1.583 4.814 2.710 14.207 8.399 1.479 6.585 3.512 13.541 9.635 4.031 3.810 4.017 18.744 6.162 2.270 6.142 3.245 33.031 7.844 5.010 2.570 3.453 16.598 7.164 0.861 0.000 6.077 27.011 5.626 1.566 2.765 0.000 23.485 6.750 3.181 2.235 1.561 0.000 32.960 20.976 10.772 17.590 18.607 0.000 6.067 1.250 3.686 2.330 9.496 O.Oo0 52 T.A. RATHBUN Fig. 1. Cluster analysis of male regional groups. for the other groups to cluster along areal lines, it is unfortunate that no data for sites around the Persian Gulf or the Arabian Peninsula were available. The Caucasus groups should also be investigated. The Eastern Mediterranean males seem to form a tight cluster in biological, geographical, and temporal dimensions. Another tight cluster is formed by the Turkmenian, Indian, and Tepe Hissar populations. Although the Nippur and A1 ‘Ubaid males show biological affinities with the eastern groups, geographical and temporal factors are more difficult to invoke. The analysis of the cluster patterns of females including individual sites (Fig. 4) differs markedly from the males. Although the SyroPalestinian females were the most distinct when the regional groups were compared, the Early Dynasty I11 females from Kish emerge as the most distinct group in the inclusive matrix. None of the other Mesopotamian female samples cluster closely with Kish, and the females from the later periods a t Kish show affinities with widely dispersed populations. The clustering pattern of the females does not seem to be so geographically harmonious as that of the males. Instead, we see a close affinity among the Dinkha Tepe, Syro-Palestinian, and Kish neo-Babylonian females. The cluster formed by other Mesopotamian Sam- METAL-AGE SW ASIA 53 Fig. 2. Cluster analysis of female regional groups. ples and widely dispersed groups is also perplexing. Although there is a tendency for the female groups to cluster at slightly lower levels than the males, the high divergence of the Early Dynasty I11 females and the neo-Babylonian males from Kish remains problematic. The small sample sizes and some inferred variation data preclude definitive evaluation for the individual components at Kish. In summary, the analyses of both regional groups and individual populations suggest a basically similar and stable morphology for Southwest Asian populations during the metal ages. When relative variation is considered, male-female differences appear. The males tend to reflect localized patterns, whereas the female morphological patterns seem similar over widely dispersed areas. The metric results are generally supported by analysis of nonmetric traits for some of the sites (Rathbun, 1979). Demography Discussions of the problems and advantages of reconstructing demographic properties of past populations from the skeletal remains have recently become more numerous. Especially applicable to our understanding of demographic problems in Southwest Asian prehistory are those works that discuss the cultural-biological interface in population dynamics (Angel, 54 T.A. RATHBUN 33 03 I 11 72 Fig. 3. Cluster analysis of males at individual sites and regions. 1954,1969,1971, 1972;Cook, 1972;Hassan, 1973,1979Peterson, 1975;Weiss, 1973;Swedlund, 1975;Ward and Weiss, 1976).McNeill (1976,1979)has also pointed out possible interactions of disease and migration. Most of these studies indicate the continuing problem of sufficient sample sizes and hard data in generating meaningful discussions. Unfortunately, published demographic data from Mesopotamian and Iranian sites are scant. This analysis of the demographic situation at Kish and surrounding areas should be viewed as preliminary. Tables 5 and 6 present the summary materials on distribution of the populations and comparative materials for average age at death. At most of the Mesopotamian and Iranian sites the males typically died around 30, and the females died at approximately 28 (Swindler, 1956;Krogman, 1940;Vallois, 1960). These means are similar to those from Dynastic and pre-Dynastic populations in Egypt and for Chalcolithic and Copper Age populations of Anatolia (Masali and Chiarelli, 1973;Senyurek, 1957).A later age at death is reported for Bronze and Early Iron Age Greek populations METAL-AGE SW ASIA 55 Fig. 4. Cluster analysis of females at individual sites and regions. and the Lerna site (Angel, 1954,1971).At the Eneolithic site, Vovnigi, in Hungary, males typically died a t 47 and females a t 35. Nemeskeri (1973)suggested that the Hungarian pattern is very similar to that revealed by the data from Afalou and a late Paleolithic Nubian series. It should be noted that the methods of age determination used by Acsadi and Nemeskeri (1970)use different criteria from those used in this study, and in some instances bones, especially for more advanced years. The artificial end point of 45 that was established for the older age category may magnify the dif- ference when considering average age at death. The relative frequency in the three categories of adulthood for the sexes would not be affected. Probability of dying and life expectancy have been shown to be useful in paleopathology and paleodemographic study even when segments of the population have been undernumerated (Buikstra and Cook, 1980:453-454). The Southwest Asian data support the argument by Weiss (1973)for a short life expectancy when the total population is considered from birth. His adjusted ranges from 20 to 29 years for protoagriculturalists, living pri- 56 T.A. RATHBUN TABLE 5. Avemge age at death Males Females Subadults N 18-24 25-35 35+ N 18-24 25-35 35+ N Kish pooled E.D. I11 Neo-Babylonian Unknown period 104 21 10 70 28% 24% 60% 24% 43% 48% 20% 46% 29% 29% 20% 30% 74 9 7 54 42% 33% 57% 43% 35% 23% 22% 29% 20% 101 6 Hasanlu pooled Period IV Dinkha Tepe 77 57 17 32% 33% 29% 32% 42 35% 33% 26 33% 53% 15 18% After reaching age 40% 46% 20% 18 31% 35% Culture Average age at death 44% 14% 37% 47% 29% 19% 33% Males Females Subadults 30.4 30.7 26.6 30.8 30.6 30.3 32.6 30.2 30.0 37.0 25.0 30.0 28.5 29.2 27.7 28.2 29.2 27.8 31.5 28.9 27.0 37.0 25.0 30.0 7.6 8.3 Kish pooled E.D. III Neo-Babylonian Unknown period Hasanlu pooled Period IV Dinkha Tepe, Iran Nippur, Iraq Tepe Hissar, Iran Tepe Sialk, Iran Dynastic Egypt Pre-Dynastic Egypt B 2- 3-6 7-12 13-18 25% 22% 30% 24% 33% 0% 33% 33% Too Few For Analysis-----------------82 24% 22% 29% 24% *3-12 35 37% 40% 23% 29 45% 31% 24% Too Few For Analysis------------------ From birthSexes pooled 21.6 26.6 24.5 20.9 24.7 23.2 - 7.5 7.1 6.7 ~ TABLE 6. Total population age distribution by percent Females Males 18-24 Kish pooled E.D. I1 E.D. 111 Neo-Babylonian Unknown Hasanlu pooled Hasanlu IV 25-35 5.29 8.21 3.85 3.85 7.58 15.15 17.65 5.88 4.31 8.12 &+ 5.47 - 9.09 5.88 5.33 16.23 17.53 16.23 16.96 16.96 16.96 ~ Adult 18-24 25-35 35+ 5.66 13.87 15.38 24.24 4.55 20.59 11.76 9.90 5.84 11.04 10.71 mitives, classic, and medieval populations accommodate the Kish and Hasanlu populations nicely. These means are consistently lower than those reported by Acsadi and Nemeskeri (1970). When age at death after reaching 18 is considered, the Southwest Asian groups consistently died at a younger age and fall within the lower range of variation. It should be noted that none of the Southwest Asian groups, except Dinkha Tepe and Tepe Sialk, are within the range of life expectancy of urban agriculturalists (mean 40.3,range 31.9-49.6)characterized by Weiss. Most of them are in the lower ranges for modern hunting-gathering groups and protoagriculturalists. Problems of comparable aging criteria may account for some of the discrepancy. 4.74 3.85 6.06 2.94 5.08 8.44 8.04 3.10 7.69 3.03 5.88 2.79 7.79 4.46 Adult 8.58 11.54 10.61 8.82 7.87 Indet. Adult 26.64 26.92 10.61 14.71 29.95 - Subadults B-2 4.56 - 3.63 2.94 5.08 3-6 7-12 4.01 11.54 5.47 11.54 3.03 - 2.94 4.57 3-12 8.44 9.09 11.61 8.04 - 13-18 4.38 3.85 3.03 - - 6.09 5.08 13-17 5.19 6.25 It is unlikely that the skeletal samples included here meet all of the assumptions of model life tables: stable population, representativeness of the sample, and freedom from systematic aging errors (Weiss, 1973:14).After estimating mortality rates from the skeletal samplesfrom Kish and related groups, the model table MT = 15.0-45.0 seemed best to accommodate the groups and to produce the hypothetical demographicsituation: crude birth rate, 0.655;completed family size, 4.77;mean family size, 4.44;generation length, 25.87;adults past 15 reaching 50,8.8; population under 15,55.3%; fertile adults, 42.1%; adults past the age of reproduction, 2.6%;dependency ratio, 1.37;average age of population, 16.4;average age of adults, 28.2 (Weiss, 1973:118). METAL-AGE SW ASIA 57 These characteristics support the argument sive rate. This situation is similar to that among of H. Weiss (1975) for a per annum growth rate the Egyptian early agriculturalists (Masali and of less than 0.16% at Kish as estimated by Chiarelli, 1973; Nemeskeri, 1973). Reference settlement size and pattern for the Early Dy- should also be made to estimates of population growth rates in the area by Carneiro and Hilse nasty period. Carneiro and Hilse (1966:177-180) estimated a Neolithic growth rate in the area (1966) and Hassan (1973). The Southwest Asian groups exhibit a comof 0.08-0.12%, and Hassan (1973540) suggests a fluctuating growth of 0.4-1.3% with periods mon pattern of more males living longer than females. The higher female death frequencies of stagnation and depopulation. It should be in the younger ranges can possibly be attribremembered that theoretical models are useful uted to pregnancy-related problems. The higher tools, but should not be considered ends in frequency of cribra orbitalia with infants and themselves. They do, however, provide a base young females may also suggest a nutritional for general comparisons, and when applied to similar populations may indicate avenues for component through anemic conditions during lactation and breast feeding. In Mesopotamia empirical testing. From the examination of the model life table both subadults and adult females exhibited a and the actual population distribution, we see higher infection occurrence than among parallel Iranian groups, but stress, as indicated that the populations in Southwest Asia were quite young during the metal ages. In most by linear enamel hypoplasia, was more frequent in Iran for both males and subadults premodern populations the proportion of the (Rathbun, 1980). total population in the subadult range is usuIn summary, the metal-age populations in ally between 50%and 60%, but with the Kish sample only 18.42%of the skeletons in the col- Southwest Asia were young and apparently relatively stable. Mortality was high for all lection were subadults. Sampling techniques remain a problem for paleodemographic stud- ages, and females had a shorter life expectancy than males. The metal age agricultural sociies. Vallois’ (1960:203-204) statement remains applicable: . . . archaeologists must neces- eties do not seem to reflect a dramatic improvement in health as indicated by average age a t sarily try to preserve systematically all the human remains they discover, without discrim- death. From a cultural evolutionary perspective a strong case can be made in support of inating against young or very fragmentary the argument that “the agricultural revolution bones.” cannot be viewed as being a great demographic A probably high subadult mortality rate during the metal ages is supported by ethno- change with regard to the individual. Although the denser, more predictable food supgraphic data on agricultural Iranian villages. At Hasanlu 35.86% of the 145 children born to ply allowed a great many more individuals to 35 married men from the present village were live, it did not dramatically change their life dead (Rathbun and Bass, 1977). The average expectancies. Longevity did not jump signifinumber of births for each marriage was 4.14. cantly until industrialization” (Weiss, 1973:30). Similar circumstances were evident with Kur- Agriculture appears not to have changed the dish villagers at Ziwiye, Iran, where 39.35% basic biological patterns very much in Southof the 61 children born to 15 married men had west Asia. Mortality is still the dominant popdied. These high infant mortality rates (393.4/ ulation growth inhibitor, but urban agricul1,000 births at Ziwiye and 358.6/1,000 at Has- turalists may have had fertility and mortality anlu) are considerably higher than the 216.51 approximately equal in influence. 1,000 live births reported for the Iranian central plateau (Mashayekhi et al., 1953). In a DISCUSSION AND CONCLUSIONS Khuzistan, Iran, village more than half the Morphological and demographic changes of deaths in a 10-year period were of children less Southwest Asia populations during the metal than 6 years old (Salmanzadeh, 1969).The high death rate among infants and children in an- ages appear to have been minimal. The major cient Southwest Asia is illustrated in the per- discernible trends from populations a t urban centage districutions of skeletal populations, centers on the Iranian plateau and in the Meand is comparable to those reported for other sopotamian valley include a reduction of roarchaeological populations (Howells, 1960; busticity of the entire skeleton and reduction of the crown area of the posterior maxillary Vallois, 1960). The high rates of infant mortality and young dentition (Rathbun, 1975). These trends may adult death, especially among females, indi- be responses to both cultural and genetic faccate that although the total populations were tors, but environmental change was not of a young, they were not expanding at a n explo- marked nature. Genetic drift does not seem to “ 58 T.A. RATHBUN be a significant feature (Finkel, 1974).Selec- their daughters who may then be married to tive pressures do not seem marked for any par- sedentarists and contribute to that gene p l . ticular zone in the ecological mosaic except for Impoverished herdsmen who cannot maintain the trend for mountain populations to be more a large enough number of animals and male robust. Widespread and sudden alien migra- kin to function as a viable unit also settle. tion does not appear to be detectable. Instead, Daughters are oRen married to sedentary men, the morphological pattern that emerges is of a but sons may remain pastoral as hired shepbasic stability and similarity of most popula- herds. Most of the gene flow seems to be from tions in the area. These findings based on mul- the pastoral groups to the sedentary ones via tivariate techniques agree with univariate the females. Only occasionally will sedentarstudies. The question then becomes one of ac- ists take up the nomadic pastoral existence. This combination of social, economic, and decounting for the basic stability and relative relationships of the groups in their biological mographic factors would create conditions for maintaining the morphological stability in and cultural milieu. The cluster patterns and levels of affinity of Southwest Asia, and partially account for the the sexes suggest differential mobility, with differences in cluster levels between the sexes. the males tending to be locally stable and the Such an account of the population dynamics females more mobile. Possible cultural expla- could explain the existence of both gracile and nations of the patterns include the possibility robust components of the populations a t Tepe of patrilineality and patrilocality and the dif- Hissar, Hasanlu, Kish, and other early Meferential inclusion of females from adjacent sopotamian sites. The population dynamics of the Iranian plapopulations as the result of warfare, trade, political alliance, or demographic adjustment. The teau and Mesopotamian valley have been commutual influence of the major geographical and plex and variegated. The general situation, as cultural areas within Southwest Asia during well as the specific situation a t Kish, seems the metal ages and urbanization process is being compatable with Gibson’s (1972:2)suggestion increasingly emphasized (Adams, 1974;Cowgill, that the historical dynamic has been “a grad1975;Wright, 1971;Wright, and Johnson, 1975; ual, uneven, infiltration of the new element Kohl, 1978). These complex and interdepen- into the region. . . not as an invasion, but as dent processes are probably reflected in the a gradual consolidation, domination and abmating patterns and resultant morphological sorption of older settled populations by a group mosaic in the area. that may once have been predominately noThe urbanized areas in the mountain and madic.” river valleys might be viewed as recipient reAlthough there is an increasing awareness gions, whereas the surrounding mountain and of the potential contributions of osteological pastoral groups served as donors (Binford,1968). studies in interdisciplinary research in SouthThe cluster analysis of morphological dis- west Asia, problems of sample sizes, presertances suggests that the females were more vation, and systematic skeletal excavation remobile than the males. The higher female mor- main. The developing specialties of tality rates and earlier age a t death in the paleopathology, paleodemography, and nonurban centers would create marriage difficul- metric multivariate analysis may provide adties for widowers with dependent children and ditional information on the complex culturalfor young males. Recruitment of females from biological interface in the area. Mellaart’s surrounding groups already in political or eco- (1974:274)comment seems appropriate: “It is nomic contact seems a viable solution. Eth- only in the last few decades that scholars have nographic examples such as Barth‘s (1961)de- learned to pay proper attention to plant and piction of the Basseri of Southern Iran indicate animal remains, but how many have realized that pastoral groups often have a high fertility the potential of anthropological studies, dewith a surplus of marriageable females in the mography and the pathology of the human tent unit. With the limited carrying capacity skeletal remains?” and critical ratios of animals and male herdsACKNOWLEDGMENTS men for continued economic success, the males Since various phases of this research were of the group tend to remain pastoralists except in two segments of the population. Analysis of carried out over several years, a number of sedentarization indicates that successful her- individuals have contributed to its completion. ders invest their capital in arable land near Funds for travel, analysis, and preparation were villages and towns and often settle there with provided by Henry Field of Field Research METAL-AGE SW ASIA Projects; the University of Kansas Greater University Fund; and the University of South Carolina Committee for Productive Scholarship. A number of students contributed to various phases of the project, and I thank them collectively. Glen Cole of the Field Museum of Natural History was supportive, and Robert H. Dyson of the University of Pennsylvania made materials available. Dorothy Tart,Cheryl Fowler, and Patricia Rhudy typed several versions of the manuscript. Final thanks go to Babette Rathbun for editorial suggestions, analytical comments, and sympathetic understanding. LITERATURE CITED Acsadi, G, and Nemeskeri, J (1970)History of Human Life Span and Mortality. Budapest Akademiai Kiado. Adams, RM (1974)Anthropological perspectives on ancient trade. Curr. Anthropol. 15239-259. Anderson, J E (1962)The Human Skeleton: A Manual for Archaeologists. Ottawa: National Museum of Canada, Department of Northern Affairs and National Resources. Angel, J L (1954)Human biology, and health, and history in Greece from first settlement until now. In Yearbook of the American Philosophical Society, Philadelphia: American Philosphical Society, pp. 168-174. Angel, J L (1969)The bases of paleodemography. Am. J . Phys. Anthropol. 30:427437. Angel, JL (1971)The People of Lerna, Analysis of a Prehistoric Aegean Population. Washington: American School of Classical Studies a t Athens, and Smithsonian Institution Press. Angel. J L (1972)Ecology and population in the Eastern Mediterranean. World Archaeol. 4r88-104. Barth, F (1961)Nomads of South Persia. Boston: Little, Brown. Bass, WM (1964)The variation in physical types of prehistoric plains Indians. Plains Anthropol. Vol. 9, No. 24, Memoir 1, pp. 65-145. Bass, WM (1971)Human Osteology: A Laboratory and Field Manual of the Human Skeleton. Columbia: Missouri Archaeological Society. Bass, WM,Evans, D, and Jantz, R (1971)The Leavenworth Site Cemetery; Archaeology and Physical Anthropology, No. 2. Lawrence, Kansas: University of Kansas Publications in Anthropology. Binford, L (1968)Post pleistocene adaptations. In L Binford and S Binford (eds.)New Perspectives in Archaeology, Chicago: Aldine Publishing, pp. 313-341. Brothwell, DR (1963)Digging Up Bones. London: British MWU. Buikstra, JE, and Cook, DC (1980)Palaeopathology: An American account. Annu. Rev. Anthropol. 9:433-470. Buxton, LH, and Rice, DT (1931)Report on the remains found at Kish. J . R.Anthropol. Inst. Great Britain Ireland LXZ:57-119. Cappieri, M (1969)The Mediterranean Race in Asia before the Iron Age. Occasional Paper No. 8. Miami: Field Research Projects. Cappieri, M (1970)The Mesopotamians of the Chalcolithic and Bronze Ages. Occasional Paper No. 12.Miami: Field Research Proiects. Carneiro, RL, A d Hilse, DF (1966)On determining the probable rate of population growth during the Neolithic. 59 Am. Anthropol. 68:177-180. Constandse-Westermann, TS (1972)Coefficients of Biological Distance. New York: Humanities Press. Cook, SF (1972)Prehistoric Demography. Module 16, J Casagrande, W Goodenough, and E Hammel (eds.)Reading, Massachusetts: Addison-Wesley Modular Publications, pp. 1-42. Cormccini, RS (1975)Multivariate analysis in biological anthropology: Some considerations.J. Hum. Evol. 4:l-19. Cowgill, GL (1975)On muses and consequences of ancient and modern population changes. Am. Anthropol. 77:505-526. Field H (1932)Human remains from Jemdet Nasr, Mesopotamia. J. R. Asiatic SOC.,October, Part 4,pp. 967-970. Field, H (1947)Human remains from Kish, Iraq. American Documentation Institute Microfilm, No. 2345,pp. 1-118. Washington: Library of Congress. Finkel, DJ (1974)The Dynamics of Middle Eastern Skeletal Populations. Ph.D. Dissertation, University of Oregon. Gelb, IJ (1961)The early history of the west Semitic peoples. J. Cuneform Stud. XV:2747. Gibson, M (1972)The City and Area of Kish. Miami: Field Research Projects,Study No. 69. Halder, A (1965)Wohr Kamen Die Sumerer? Bibliotheca Orientalis WI:131-140. Hassan, FA (1973)On mechanisms of population growth during the Neolithic. Curr. Anthropol. I4:535-543. Hassan, FA (1979)Demography and archaeology. Annu. Rev. Anthropol. 8:137-160. Hiernaux, J (1964)The measuring of morphological differences between populations for a set of variables. Yearbook Anthropol. 12:127-135. Howells, WW (1960)Estimating population numbers through archaeological and skeletal remains. In R Heizer and S Cook (eds.) The Application of Quantitative Methods in Archaeology. New York: Viking Fund Publications in Anthropology, No. 28,pp. 158-186. Howells, WW (1973)Cranial Variation in Man: A Study by Multivariate Analysis of Patterns of Difference among Recent Human Populations. Cambridge: Papers of the Peabody Museum of Archaeology and Ethnology, Harvard University, Vol. 67. Hrdlitka, A (1952)Practical Anthropometry. Philadelphia: Wistar Institute. Johnston, FE (1962)Growth of the long bones of infants and young children at Indian Knoll. Am. J. Phys. Anthropol. 21:249-254. Kohl, PL (1978)The balance of trade in Southwestern Asia in the mid-third millennium B.C. Curr. Anthropol. lX3):464-492. Krogman, W M (1940)Racial Types from Tepe Hissar, Iran, from the Late Fifth to the Early Second Millennium B.C. A Chapter in the Protohistory of Asia Minor and the Middle East. Verhand. der Konik, Nederl. Akad. van Westensch., Natuurk. Tweede Sectie, Dee1 39,No. 2,Amsterdam. Krogman, WM (1962)The Human Skeleton in Forensic Medicine. Springfield, Illinois: Charles C. Thomas. Masali, M, and Chiarelli, B (1973)Demographic data on the remains of ancient Egyptians. In D Brothwell and B Chiarelli (eds.) Population Biology of the Ancient Egyptians. London: Academic Press, pp. 161-169. Mashayekhi, MB, Mead, PA, and Hayes, GS (1953)Some demographic aspecta of a rural area in Iran. Milbank Mem. Fund Q. XXXI:l49-165. McKern, TW, and Stewart, TD (1957)Skeletal Age Changes in Young American Males: Analyzed from the Standpoint of Age Identification. Natlick, Massachusetts: Technical 60 T.A. RATHBUN &port EP-45, Quarter-master Research and Development Command. McNeill, WH (1976) Plagues and Peoples. New York: Doubleday. McNeill, WH (1979) Historical patterns of migration. Curr. Antbropol. 20(1):95-102. Mellaart, J (1974) The Neolithic of the Near East. New York Scribner’s. Montagu, MFA (1960) An Introduction to Physical Anthropology. Springfield, Illinois: Charles C. Thomas. Moorey, PRS (1970) Cemetery “ A at Kish Grave groups and chronology. Iraq 11:86-128. Nemeskeri, J (1973)Some comparisonsof Egyptian andearly Eurasion demographic data. In D Brothwell and B Chiarelli (eds.) Population Biology of the Ancient Egyptians. London: Academic Press, pp. 171-186. Penrose, LS (1954) Distance, size, and shape. Ann. Eugenics m337-343. Peterson, W (1975) A demographer’s view of prehistoric demography. Curr. Anthropol. 16:227-247. Phenice, TW (1967) A newly developed visual method of sexing the 0 s Pubis. Am. J. Phys. Anthropol. 30:297302. Rathbun, TA (1972) A Study of the Physical Characteristics of the Ancient Inhabitants of Hasanlu, Iran. Coeonut Grove, Miami: Field Research Projects, Study No. 68. Rathbun, TA (1975) A Study of the Physical Characteristics of the Ancient Inhabitants of Kish, Iraq. Coconut Grove, Miami: Field Research Projects, Study No. 87. Rathbun, TA (1979) Metric and discrete trait variation among Southwest Asian populations. Am. J. Phys. Anthropol. 50(.3):473. Rathburn, TA (1980)Patterns of pathology among metal age Iranian and Mesopotamian populations. Am. J. Phys. Anthropol. 52269. Rathbun, TA (n.d.1 Skeletal material from Dinkha Tepe, Iran. Manuscript. Rathbun, TA, and Bass, WM (1977) Physical and demographic features of two groups in northwestern Iran. Anthropologie XV/l:55-61. Salmanzadeh, C (1969) Social Survey of Naddafiaeh village. Ahwaz, Iran: Jundi Shapur University, College of Agriculture. Schour, I, and Massler, M (1944) Development of the Human Dentition. Chicago: American Dental Association. Senyurek, MS (1957) The duration of life of the Chacolithic and Copper Age populations of Anatolia. Anadolu Rev. Annu. Archeol. (Universite dAnkara) 295-1 10. Speiser, E (1930) Mesopotamian Origins: The Basic Population of the Near East. Philadelphia: University of Pennsylvania Press. Swedlund, AC (ed.)(1975) Population Studies in Archaeology and Biological Anthropology: A Symposium. Memoirs of the Society for American Archaeology, No. 30. Swindler, DR (1956) A Study of the Cranial and Skeletal Material Excavated at Nippur. Philadelphia: University Museum, University of Pennsylvania. Ubelaker, DH (1978) Human Skeletal Remains. Chicago: Aldine. Vallois. HV (1960)Vital statistics in prehistoric population as determined from archaeological data. In R Heizer and S Cook (eds.) The Application of Quantitative Methods in Archaeology. New York: Viking Fund Publications in Anthropology, No. 28, pp. 186-222. Vallois, HV (1965) Anthropmetric techniques. Curr. Anthropol. 6:127-143. Ward, RH, and Weiss, KM (1976) The Demographic Evolution of Human Populations. New York: Academic Press. Weiner, JS, and Huizinga, J (1972) The Assessment of Pop ulation Amnities in Man. Oxford Clarendon Press. Weiss, H (1975) Kish, Akkad and Agade. J. Am. Oriental SOC. 95(3):434-453. Weiss, KM (1973) Demographic Models for Anthropology. Memoirs of the Society for American Archaeology, No. 27, American Antiquity, Vol. 38, Number 2, Part 2, April. Wright, GA (1971) Origins of food production in southwestern Asia: A survey of ideas. Curr. Anthropol. 12:447475. Wright, HT, and Johnson, GA (1975)Population, exchange, and early state formation in southwestern Iran. Am. Anthropol. 77267-290.