Printer Friendly

Ethnic identity, biological history and dental morphology: evaluating the indigenous status of Maharashtra's Mahars.


The idea of indigenous people in South Asia is more complex than elsewhere, in part because it involves longstanding and intimate contact between `tribal' and non-tribal peoples (Beteille 1998; Gardner 1985; Lukacs in press). Additional complications arise from the hierarchal and endogamous structure of Hindu social and ritual organization, including the plight of people who occupy the lowest stratum of the hierarchy -- `untouchables' (Charsley 1996; Deliege 1992; 1993). Because the system of socioreligious stratification known as caste does not encourage social mobility, new ethnic identity is often sought by groups whose position in the hierarchy is low (Dumont 1980; Klass 1980; Kolinda 1978). Biological anthropologists are interested in the caste system for the opportunities it offers to understand the interaction of cultural behaviour with the biological patterning of human genetic and phenotypic diversity (Majumder 1998; Majumder et al. 1990; Malhotra 1974). Although most westerners perceive caste as an immutable category, in which membership is ascribed, and hierarchal rank is forever fixed, many accounts of castes changing their occupational and ritual status have been documented (Silverberg 1968). Some castes seek to elevate their ritual or economic position by claiming higher status and adopting an appropriate new caste name, while others lay claim to indigenous origins seeking to benefit from rights and privileges that accompany autochthonous status. Such claims often involve adopting new or different patterns of behaviour commonly associated with the new social, religious, indigenous or occupational position claimed. This process is sufficiently common in India to be labelled `Sanskritization' when a Hindu caste emulates higher castes (Srinivas 1968), `Hinduization' when tribal or non-caste groups emulate Hindu castes, or more generally, `elite-emulation' (Lynch 1969).

The Mahars, a low-status, ex-untouchable caste that comprises approximately 9% of the population of Maharashtra, have sought to improve their status in two ways:

1 through mass conversion to a new religion -- Buddhism, and

2 by claiming to be the autochthonous inhabitants of the Deccan Plateau of west-central India, through their folklore, which identifies them as `creators of culture' and `sons of the soil' (Zelliot 1992a).

The modern state name -- Maharashtra -- may derive from the caste name `Mahar', as in `Mahar rashtra' -- meaning people of the great country (Enthoven 1922: 401; Robertson 1938:44). By contrast, other likely derivations include `maha' `ari' -- great enemy, and `mrit' `har' -- he who removes dead animals (Zelliot 1992b). Remarks note the wide distribution of Mahars throughout the state, such as, `Wherever you find Mahars, there is Maharashtra' and `Wherever there is a village there is a maharwada' (a residential area on the village periphery for exclusive occupation by Mahars -- due to their untouchable status) (Karve 1959). According to Zelliot (1992a: 70), `In Maharashtra, the idea that the Mahar was the original settler is one of the most persistent images among Mahars, Dalits, and Buddhists'. Nevertheless, caste mobility is feasible and claiming autochthonous status is one strategy for achieving social mobility. An anthropological evaluation of the accuracy and authenticity of such declarations constitutes an intriguing bio-cultural research problem. If the Mahars are autochthonous inhabitants of Maharashtra, biological evidence of genetic continuity between prehistoric populations of the Deccan Plateau and the modern Mahars should confirm it. Hence, the questions this paper seeks to answer are two-fold. Firstly, are the Mahars recognizably distinct in their biological attributes from other ethnic groups of Maharashtra? Secondly, do Mahars represent the autochthonous inhabitants of Maharashtra, descended from ancient occupants of the Deccan Plateau? These questions will be investigated through an examination of polymorphic variants of the tooth crown among three living ethnic groups of Maharashtra and one prehistoric human skeletal series from the early farming site of Inamgaon.

Context of research: deciphering population history

The following questions and their answerswill contribute to the resolution of three broad yet interrelated issues of population history in the subcontinent of India:

1 What complex of dental traits characterize populations of west and central India, and can the pattern of dental morphology distinguishing these populations be regarded as intermediate between the dental complexes characteristic of North and South Indian populations?

2 Is there any differentiation in dental morphology between Hindu castes and non-caste or `tribal' populations?

3 Were prehistoric inhabitants of the Deccan Plateau ancestral to modern tribal populations?

This final question re-examines conclusions derived from the analysis of craniometric variation in Chalcolithic human skeletal series of the Deccan Plateau. Regarding the first issue, little is known about the dental morphological pattern of either tribal or Hindu caste populations of western and central India (Lukacs 1977; Zubov & Gashimova 1982; Walimbe & Aphale 1987). Yet this data has tremendous potential for answering questions regarding the peopling of the Indian subcontinent, and for understanding the biological nature of the interaction between Indo-European speaking populations of the north Indian culture area and those populations speaking Dravidian languages in south India. In addition to the linguistic differences which separate them, north and south India also display different preferential mating systems which result in dissimilar levels of inbreeding as well as distinctive matrimonial distance values (Malhotra 1984; Malhotra et al. 1978; Sundar Rao 1984). In fact, the key geographical position of Maharashtra, at the transition zone between these two distinctive linguistic and cultural areas enhances the anthropological significance of this study (Fairservis & Southworth 1989).

Regarding the second issue, recent study of the apportionment of tooth size among Hindu castes and non-caste tribal groups of Andhra Pradesh found that regardless of linguistic group affiliation, or geographic distance, Hindu castes exhibit greater phenetic similarity to one another than to tribal groups (Hemphill 1991; Hemphill et al. 1992). Using five samples (high caste (Pakanati Reddy) and low caste (Gampadhompti Madiga) Hindus, Maharashtrians (a mix of Hindu caste groups), Bengalis (a mix of Hindus caste groups), and a tribal group (Chenchu)), in conjunction with multivariate statistical methods (cluster analysis and principal components), these studies revealed that the two south Indian castes were most similar to one another, and that the Chenchus were clearly differentiated from all Hindu samples. If the results of tooth size apportionment analysis of the Andhra Pradesh samples are applied to the groups included in the present analysis, Marathas and Mahars will predictably be most similar in dental morphology, while the Gonds should be divergent from these Hindu castes. Genetic studies of Hindu castes and tribes of Maharashtra reveal a clear-cut division between them (Mukherjee et al. 1979a; 1979b), a finding that will be examined here from the perspective of dental morphology.

The third question leads to an alternative hypothesis of population relationships derived from the study of craniometric variation among prehistoric inhabitants of Maharashtra. This alternative hypothesis holds that the descendents of Chalcolithic populations of the Deccan Plateau are to be found among the tribal populations of the subcontinent, not among the Hindu castes of urban centres and villages (such as Mahars). While human remains of infants and children have been recovered in some abundance, well-preserved adult human crania, suitable for craniometric study, are few. For example, only one partial adult cranium was recovered from Daimabad, where the total number of individuals recovered was 35 (Walimbe 1986). Other Chalcolithic sites are similar, with Chandoli yielding one adult and 24 immature skeletons (Malhotra 1965), Kaothe produced two adults and three immature skeletons (Walimbe 1990) and at Nevasa, while 130 individuals were recovered, all but four are immature or fragmentary. Since one of the adults was a late Indo-Roman burial, only three Chalcolithic specimens could be included in the craniometric analysis of the Nevasa series (Kennedy & Malhotra 1966). From their osteological analysis of the Nevasa adult skeletons, Kennedy & Malhotra (1966: 120) conclude:

Comparisons of the anthropometric data of the Chalcolithic and Indo-Roman Nevasians with the living inhabitants of Maharashtra, Uttar Pradesh and Gujarat indicate that phenotypic similarities can be traced not among the present urban and village populations, but among the tribal peoples, particularly among the Bhils and Gonds. [italics added]

This conclusion from osteological research stands in direct contrast to the cultural history evidence cited above that consistently identifies the Mahars, a low-status, ex-untouchable, village caste group as the indigenous, aboriginal population of Maharashtra. The goal of this research is to use biological evidence from polymorphic variation in dental morphology to test which of these competing hypotheses is most probable.

Materials and methods

The materials for this investigation consist of 569 plaster dental casts of three ethnic groups of Maharashtra State, India, and one prehistoric human skeletal series. Groups selected for study include the Marathas, the locally dominant, high-caste Hindu group; the Mahars, an ex-untouchable, low-status caste; and a tribal group, the Madia Gonds of eastern Maharashtra, who are not part of the Hindu caste system (TABLE 1). Karve's (1961: 67, diagram 1) schematic representation of important ethnic groups in the State of Maharashtra, includes Marathas (A) and Mahars (B, [B.sup.1]) along with other tribes and Hindu castes (FIGURE 1). A diachronic perspective is provided by a prehistoric sample from the Chalcolithic site of Inamgaon (1700-700 BC), located near Pune (Dhavalikar et al. 1988). The geographic location of each study group is depicted in FIGURE 2. Observation and scoring of dental crown features of the 569 central India samples was conducted in the summer of 1990 by Lukacs. The analysis of dental morphology for the prehistoric skeletons from Inamgaon was initiated by Lukacs at Deccan College, Pune, in 1979 with the collaborative assistance of Dr G.L. Badam. The permanent dental morphology data used in this study are taken from Lukacs (1987), which illustrates the expression of selected dental traits used in this analysis (see also Turner et al. 1991). Prior to multivariate analysis, trait frequencies, based on presence/absence dichotomization, were arc-sine transformed according to the method of Green & Suchey (1976) in order to stabilize the variance. Three comparative methods were used;

1 arc-sine transformed trait frequencies were subjected to cluster analysis (euclidean distance, Ward's minimum variance method),

2 mean measure of divergence values (MMD) were calculated, together with their variances and standard deviations (Sjovold 1973), then standardized MMD values were multidimensionally scaled, and

3 principal components analysis was performed on the arc-sine transformed trait frequencies and the derived components plotted in multidimensional space.

TABLE 1. Dental survey of India study groups.

 geographic provenience
status northwest central southeast

high caste Rajput (RAJ) Maratha (MRT) Pakanati
 Reddi (PNT)

low caste Garasia (GRS) Mahar (MHR) Gampadhompti
 Madiga (GPD)

tribal Bhil (BHL) Madia Gond (MDA) Chenchu (CHU)

prehistoric -- Inamgaon --

 this study

The statistical evaluation of population relationships was conducted by Hemphill in two stages. In the first stage three modern Maharashtrian samples (Maratha, Mahar, Madia Gond) were compared with the prehistoric skeletal sample from Inamgaon, and two `out-group' samples -- Native Americans (Papago Indians) and American Whites. In the second stage of analysis, nine prehistoric South Asian skeletal samples were added to the comparison. TABLE 2 provides a list of all study groups used in these comparisons. A more detailed description of analytical procedures, graphic depiction of group relationships and interpretation of statistical results is available elsewhere (Lukacs et al. 1998).
TABLE 2. Antiquity, size, and source of comparative dental data.

sample abb. date

Mahars MHR living
Madia Gonds MDA living
Marathas MRT living
Inamgaon INM 1600-700 BC
Papago PAP living
American whites AMW living
Neolithic Mehrgarh NeoMRG 6000 BC
Chalcolithic Mehrgarh ChlMRG 4500 BC
Harappa HAR 2300-1700 BC
Sarai Khola SKH 200-100 BC
Timargarha TMG 1400-850 BC
Djarkutan DJR 2100-1950 BC
Sapalli Tepe SAP 2300-2150 BC
Kuzali KUZ 1950-1800 BC
Molali MOL 1800-1650 BC

sample [N.sub.max](1) source

Mahars 195 this study
Madia Gonds 169 this study
Marathas 198 this study
Inamgaon 41 Lukacs (1987)
Papago 178 Scott (1973)
American whites 77 Scott (1973)
Neolithic Mehrgarh 49 Lukacs (1988)
Chalcolithic Mehrgarh 25 Lukacs & Hemphill (1991)
Harappa 33 Lukacs (1992)
Sarai Khola 15 Lukacs (1983)
Timargarha 25 Lukacs (1983)
Djarkutan 39 Hemphill et al. (1998)
Sapalli Tepe 43 Hemphill et al. (1998)
Kuzali 24 Hemphill et al. (1998)
Molali 41 Hemphill et al. (1998)

(1) [N.sub.max] represents the greatest number of individuals scored for a non-metric trait.


Trait frequencies for morphological variants of the dental crown are presented in TABLE 3 for the three living Maharashtrian groups, the prehistoric Inamgaon skeletal series, and the two `out-groups' (Native Amerindian Papago, Americans of European descent). Results of the first comparison using six groups:

1 consistently identified the Papago and American Whites as distantly related to the four Maharashtrian groups, but

2 produced highly variable and inconsistent results regarding the degree of biological affinity among the living and prehistoric Maharashtrian groups.
TABLE 3. Frequencies of dental morphology traits -- comparison one.


dental trait tooth abb. P N freq.

shovelling UI1 ShovI1 22 24 0.917
shovelling UI2 ShovI2 13 19 0.684
med. lingual ridge UI1 MlrI1 14 25 0.560
med. lingual ridge UI2 MlrI2 1 20 0.050
hypocone size UM1 HypoM1 27 41 0.659
hypocone size UM2 HypoM2 0 20 0.000
Carabelli's trait UM1 CaraM1 13 40 0.325
metaconule UM1 MtclM1 6 41 0.146
metaconule UM2 MtclM2 3 20 0.150
entoconulid LM1 C6M1 4 37 0.108
entoconulid LM2 C6M2 0 24 0.000
metaconulid LM1 C7M1 2 36 0.056
metaconulid LM2 C7M2 1 25 0.040
cusp number LM1 CspnM1 32 39 0.821
cusp number LM2 CspnM2 4 24 0.167
Y-groove pattern LM1 YgrvM1 32 35 0.914
Y-groove pattern LM2 YgrvM2 7 24 0.292


dental trait tooth abb. P N freq.

shovelling UI1 ShovI1 77 186 0.414
shovelling UI2 ShovI2 22 181 0.122
med. lingual ridge UI1 MlrI1 106 177 0.599
med. lingual ridge UI2 MlrI2 41 174 0.236
hypocone size UM1 HypoM1 163 195 0.836
hypocone size UM2 HypoM2 10 164 0.061
Carabelli's trait UM1 CaraM1 140 187 0.749
metaconule UM1 MtclM1 43 191 0.225
metaconule UM2 MtclM2 33 153 0.216
entoconulid LM1 C6M1 13 191 0.068
entoconulid LM2 C6M2 3 174 0.017
metaconulid LM1 C7M1 25 191 0.131
metaconulid LM2 C7M2 3 177 0.017
cusp number LM1 CspnM1 170 192 0.885
cusp number LM2 CspnM2 30 178 0.169
Y-groove pattern LM1 YgrvM1 115 127 0.906
Y-groove pattern LM2 YgrvM2 30 161 0.186

 Madia Gonds

dental trait tooth abb. P N freq.

shovelling UI1 ShovI1 80 163 0.491
shovelling UI2 ShovI2 23 161 0.143
med. lingual ridge UI1 MlrI1 60 153 0.392
med. lingual ridge UI2 MlrI2 21 149 0.141
hypocone size UM1 HypoM1 155 169 0.917
hypocone size UM2 HypoM2 10 153 0.065
Carabelli's trait UM1 CaraM1 86 165 0.521
metaconule UM1 MtclM1 36 156 0.231
metaconule UM2 MtclM2 34 138 0.246
entoconulid LM1 C6M1 12 158 0.076
entoconulid LM2 C6M2 5 152 0.033
metaconulid LM1 C7M1 27 165 0.164
metaconulid LM2 C7M2 7 158 0.044
cusp number LM1 CspnM1 149 161 0.925
cusp number LM2 CspnM2 32 158 0.203
Y-groove pattern LM1 YgrvM1 112 115 0.974
Y-groove pattern LM2 YgrvM2 31 133 0.233


dental trait tooth abb. P N freq.

shovelling UI1 ShovI1 81 198 0.409
shovelling UI2 ShovI2 24 194 0.124
med. lingual ridge UI1 MlrI1 95 194 0.490
med. lingual ridge UI2 MlrI2 42 190 0.221
hypocone size UM1 HypoM1 170 197 0.863
hypocone size UM2 HypoM2 4 179 0.022
Carabelli's trait UM1 CaraM1 122 198 0.616
metaconule UM1 MtclM1 56 193 0.290
metaconule UM2 MtclM2 32 169 0.189
entoconulid LM1 C6M1 17 194 0.088
entoconulid LM2 C6M2 5 191 0.026
metaconulid LM1 C7M1 15 198 0.076
metaconulid LM2 C7M2 1 197 0.005
cusp number LM1 CspnM1 166 195 0.851
cusp number LM2 CspnM2 28 192 0.146
Y-groove pattern LM1 YgrvM1 117 128 0.914
Y-groove pattern LM2 YgrvM2 51 181 0.282


dental trait tooth abb. P N freq.

shovelling UI1 ShovI1 168 171 0.982
shovelling UI2 ShovI2 150 168 0.893
med. lingual ridge UI1 MlrI1 72 133 0.541
med. lingual ridge UI2 MlrI2 68 144 0.472
hypocone size UM1 HypoM1 170 178 0.955
hypocone size UM2 HypoM2 28 147 0.190
Carabelli's trait UM1 CaraM1 115 160 0.719
metaconule UM1 MtclM1 28 126 0.222
metaconule UM2 MtclM2 3 79 0.038
entoconulid LM1 C6M1 71 135 0.526
entoconulid LM2 C6M2 17 129 0.132
metaconulid LM1 C7M1 53 158 0.335
metaconulid LM2 C7M2 22 128 0.172
cusp number LM1 CspnM1 175 175 1.000
cusp number LM2 CspnM2 98 140 0.700
Y-groove pattern LM1 YgrvM1 83 90 0.922
Y-groove pattern LM2 YgrvM2 16 95 0.168

 American Whites

dental trait tooth abb. P N freq.

shovelling UI1 ShovI1 12 46 0.261
shovelling UI2 ShovI2 5 52 0.096
med. lingual ridge UI1 MlrI1 40 60 0.667
med. lingual ridge UI2 MlrI2 43 65 0.662
hypocone size UM1 HypoM1 77 77 1.000
hypocone size UM2 HypoM2 35 35 0.467
Carabelli's trait UM1 CaraM1 55 74 0.743
metaconule UM1 MtclM1 3 44 0.068
metaconule UM2 MtclM2 3 46 0.065
entoconulid LM1 C6M1 3 47 0.064
entoconulid LM2 C6M2 0 70 0.000
metaconulid LM1 C7M1 18 67 0.269
metaconulid LM2 C7M2 9 61 0.148
cusp number LM1 CspnM1 57 60 0.950
cusp number LM2 CspnM2 4 67 0.060
Y-groove pattern LM1 YgrvM1 7 9 0.778
Y-groove pattern LM2 YgrvM2 5 13 0.385

Altogether four different patterns of relationship were derived using three multivariate statistical techniques:

a Madia Gonds closest to prehistoric Inamgaon (2-dimensional plot of multidimensionally scaled MMDs);

b Inamgaon equally close to tribal Gonds and dominant Marathas (3-dimensional plot of multidimensionally scaled MMDs);

c Maratha closest to Inamgaon with Madia Gonds most divergent (2-dimensional ordination of principal component scores);

d no living Maharashtrian group displays close affinity to the prehistoric people of Inamgaon (3-dimensional plot of principal components and cluster analysis).

The heterogeneity of these results may be attributable to inclusion of study groups with vastly different biological identities in a single comparison. While in taxonomic analysis out-groups are often used to balance the analysis and validate results, the use of two out-groups with significantly different biological identities may have blurred finer-level distinctions among the living and prehistoric groups of Maharashtra. For this reason a second stage of comparison was conducted in which dental morphological data for nine prehistoric south and central Asian samples were added to the analysis.

Results of the second stage comparisons rely heavily on recent research by Hemphill and colleagues on the dental morphology of central Asian populations of Bactria and Margiana (Hemphill et al. 1998). Dental trait frequencies for groups included in the second comparison are provided in TABLE 4. These results are more consistent with regard to interrelationships of south Asian prehistoric peoples and also present a clearer picture regarding the biological affinities of living and prehistoric Maharashtrian samples. Cluster analysis of arcsine transformed trait frequencies yields a close association among the three living Maharashtrian groups (FIGURE 3), with the two caste groups closely linked, and the tribal Gonds somewhat more distant. Two- and three-dimensional plots of multidimensionally scaled MMD values provide consistent insight into relations between Inamgaon and modern Maharashtrian groups. The Marathas are positioned close to the prehistoric people of Inamgaon in both plots, while the Mahars are situated more distantly. In the 2-dimensional plot (FIGURE 4), the association between Marathas and prehistoric Inamgaon is quite close, while in the third dimension of the 3-dimensional plot Inamgaon is intermediate and equidistant from both Marathas and Madia Gonds.

TABLE 4. Frequencies of dental morphology traits -- comparison two.

 Sarai Khola

dental trait tooth abb. P N freq.

shovelling UI1 ShovI1 3 9 0.333
shovelling UI2 ShovI2 2 9 0.222
med. lingual ridge UI1 MlrI1 2 9 0.222
med. lingual ridge UI2 MlrI2 0 9 0.000
hypocone size UM1 HypoM1 11 14 0.786
hypocone size UM2 HypoM2 2 13 0.154
Carabelli's trait UM1 CaraM1 2 9 0.222
metaconule UM1 MtclM1 3 9 0.333
metaconule UM2 MtclM2 2 14 0.143
entoconulid LM1 C6M1 1 14 0.071
entoconulid LM2 C6M2 0 15 0.000
metaconulid LM1 C7M1 1 15 0.067
metaconulid LM2 C7M2 0 15 0.000
cusp number LM1 CspnM1 9 15 0.600
cusp number LM2 CspnM2 1 15 0.067
y-groove pattern LM1 YgrvM1 5 7 0.714
Y-groove pattern LM2 YgrvM2 5 14 0.357


dental trait tooth abb. P N Freq.

shovelling UI1 ShovI1 7 7 0.143
shovelling UI2 ShovI2 2 7 0.286
med. lingual ridge UI1 MlrI1 3 8 0.375
med. lingual ridge UI2 MlrI2 0 7 0.000
hypocone size UM1 HypoM1 17 22 0.773
hypocone size UM2 HypoM2 0 13 0.000
Carabelli's trait UM1 CaraM1 9 18 0.500
metaconule UM1 MtclM1 4 19 0.211
metaconule UM2 MtclM2 0 13 0.000
entoconulid LM1 C6M1 0 22 0.000
entoconulid LM2 C6M2 1 18 0.056
metaconulid LM1 C7M1 2 24 0.083
metaconulid LM2 C7M2 2 20 0.100
cusp number LM1 CspnM1 19 25 0.760
cusp number LM2 CspnM2 3 17 0.176
y-groove pattern LM1 YgrvM1 12 17 0.706
Y-groove pattern LM2 YgrvM2 3 18 0.167


dental trait tooth abb. P N Freq.

shovelling UI1 ShovI1 25 28 0.893
shovelling UI2 ShovI2 31 37 0.838
med. lingual ridge UI1 MlrI1 15 26 0.577
med. lingual ridge UI2 MlrI2 2 29 0.069
hypocone size UM1 HypoM1 35 42 0.833
hypocone size UM2 HypoM2 2 41 0.049
Carabelli's trait UM1 CaraM1 7 27 0.259
metaconule UM1 MtclM1 7 28 0.250
metaconule UM2 MtclM2 10 25 0.400
entoconulid LM1 C6M1 3 37 0.081
entoconulid LM2 C6M2 0 44 0.000
metaconulid LM1 C7M1 4 40 0.100
metaconulid LM2 C7M2 0 43 0.000
cusp number LM1 CspnM1 39 43 0.907
cusp number LM2 CspnM2 3 49 0.061
y-groove pattern LM1 YgrvM1 23 25 0.920
Y-groove pattern LM2 YgrvM2 12 37 0.324


dental trait tooth abb. P N Freq.

shovelling UI1 ShovI1 21 25 0.840
shovelling UI2 ShovI2 21 24 0.875
med. lingual ridge UI1 MlrI1 14 25 0.560
med. lingual ridge UI2 MlrI2 7 24 0.292
hypocone size UM1 HypoM1 22 22 1.000
hypocone size UM2 HypoM2 10 18 0.556
Carabelli's trait UM1 CaraM1 11 18 0.611
metaconule UM1 MtclM1 5 19 0.263
metaconule UM2 MtclM2 6 18 0.333
entoconulid LM1 C6M1 5 23 0.217
entoconulid LM2 C6M2 1 23 0.043
metaconulid LM1 C7M1 3 25 0.120
metaconulid LM2 C7M2 0 24 0.000
cusp number LM1 CspnM1 20 23 0.870
cusp number LM2 CspnM2 2 24 0.083
y-groove pattern LM1 YgrvM1 15 21 0.714
Y-groove pattern LM2 YgrvM2 6 22 0.273


dental trait tooth abb. P N Freq.

shovelling UI1 ShovI1 8 15 0.533
shovelling UI2 ShovI2 10 16 0.625
med. lingual ridge UI1 MlrI1 8 12 0.667
med. lingual ridge UI2 MlrI2 6 13 0.462
hypocone size UM1 HypoM1 16 16 1.000
hypocone size UM2 HypoM2 2 18 0.111
Carabelli's trait UM1 CaraM1 4 9 0.444
metaconule UM1 MtclM1 6 13 0.462
metaconule UM2 MtclM2 4 16 0.250
entoconulid LM1 C6M1 1 20 0.050
entoconulid LM2 C6M2 0 28 0.000
metaconulid LM1 C7M1 1 22 0.045
metaconulid LM2 C7M2 0 28 0.000
cusp number LM1 CspnM1 17 20 0.850
cusp number LM2 CspnM2 0 33 0.000
y-groove pattern LM1 YgrvM1 15 17 0.882
Y-groove pattern LM2 YgrvM2 3 31 0.097

 Sapalli Tepe

dental trait tooth abb. P N freq.

shovelling UI1 ShovI1 2 19 0.105
shovelling UI2 ShovI2 10 17 0.588
med. lingual ridge UI1 MlrI1 4 17 0.235
med. lingual ridge UI2 MlrI2 5 17 0.294
hypocone size UM1 HypoM1 36 36 1.000
hypocone size UM2 HypoM2 23 32 0.719
Carabelli's trait UM1 CaraM1 8 25 0.320
metaconule UM1 MtclM1 3 37 0.081
metaconule UM2 MtclM2 2 34 0.059
entoconulid LM1 C6M1 3 25 0.120
entoconulid LM2 C6M2 0 40 0.000
metaconulid LM1 C7M1 1 38 0.026
metaconulid LM2 C7M2 0 43 0.000
cusp number LM1 CspnM1 22 28 0.786
cusp number LM2 CspnM2 2 41 0.049
y-groove pattern LM1 YgrvM1 19 24 0.792
Y-groove pattern LM2 YgrvM2 7 38 0.184


dental trait tooth abb. P N freq.

shovelling UI1 ShovI1 3 16 0.188
shovelling UI2 ShovI2 8 22 0.364
med. lingual ridge UI1 MlrI1 3 17 0.176
med. lingual ridge UI2 MlrI2 4 22 0.182
hypocone size UM1 HypoM1 30 30 1.000
hypocone size UM2 HypoM2 21 32 0.656
Carabelli's trait UM1 CaraM1 3 21 0.143
metaconule UM1 MtclM1 1 29 0.034
metaconule UM2 MtclM2 0 32 0.000
entoconulid LM1 C6M1 1 20 0.050
entoconulid LM2 C6M2 0 36 0.000
metaconulid LM1 C7M1 1 32 0.031
metaconulid LM2 C7M2 1 39 0.026
cusp number LM1 CspnM1 20 21 0.952
cusp number LM2 CspnM2 2 36 0.056
y-groove pattern LM1 YgrvM1 20 22 0.909
Y-groove pattern LM2 YgrvM2 11 35 0.314


dental trait tooth abb. P N freq.

shovelling UI1 ShovI1 1 13 0.077
shovelling UI2 ShovI2 6 14 0.429
med. lingual ridge UI1 MlrI1 2 13 0.154
med. lingual ridge UI2 MlrI2 6 14 0.429
hypocone size UM1 HypoM1 23 23 1.000
hypocone size UM2 HypoM2 11 22 0.500
Carabelli's trait UM1 CaraM1 2 20 0.100
metaconule UM1 MtclM1 2 21 0.095
metaconule UM2 MtclM2 1 24 0.042
entoconulid LM1 C6M1 0 14 0.000
entoconulid LM2 C6M2 0 15 0.000
metaconulid LM1 C7M1 0 18 0.000
metaconulid LM2 C7M2 0 18 0.000
cusp number LM1 CspnM1 10 15 0.667
cusp number LM2 CspnM2 1 14 0.071
y-groove pattern LM1 YgrvM1 10 14 0.714
Y-groove pattern LM2 YgrvM2 5 15 0.333


dental trait tooth abb. P N freq.

shovelling UI1 ShovI1 4 25 0.160
shovelling UI2 ShovI2 14 27 0.519
med. lingual ridge UI1 MlrI1 9 23 0.391
med. lingual ridge UI2 MlrI2 8 25 0.320
hypocone size UM1 HypoM1 41 41 1.000
hypocone size UM2 HypoM2 23 37 0.622
Carabelli's trait UM1 CaraM1 13 36 0.361
metaconule UM1 MtclM1 3 39 0.077
metaconule UM2 MtclM2 3 37 0.081
entoconulid LM1 C6M1 3 33 0.091
entoconulid LM2 C6M2 0 35 0.000
metaconulid LM1 C7M1 2 39 0.051
metaconulid LM2 C7M2 1 36 0.028
cusp number LM1 CspnM1 29 33 0.879
cusp number LM2 CspnM2 2 35 0.057
y-groove pattern LM1 YgrvM1 25 34 0.735
Y-groove pattern LM2 YgrvM2 5 33 0.152

The outcome of the principal components analysis of dental traits in this second stage of comparison provide confirmation of the MMD results described above. Both 2- and 3-dimensional plots of principle components reveal a close biological association between the Marathas and the people of Inamgaon, while the low- status Mahars are distantly positioned from the Inamgaon sample. In all these comparisons, expected associations consistently occur: the central Asian samples are closely linked and separate from south Asians and the Papago and American White out-groups are the most distantly positioned samples in the plots. The consistency of results based on the known relationships of certain groups to one another enhances our confidence in the reliability of interrelationships among the Maharashtrian samples.

Summary & conclusions

Frequencies of 17 dental morphology traits among three contemporary social groups of Maharashtra have been contrasted with other samples in two phases of comparison. In the first comparison, these groups were compared with a prehistoric sample from Maharashtra (Inamgaon), and with living Papago Indians and American whites. Contingency chi-square analysis indicated that 15 of these tooth-trait combinations differed significantly across all six samples. Analysis of inter-populational variation successfully identified Papago Indians and American whites as strongly separated from one another and from all Maharashtrian samples. However, patterns of affinity among Maharashtrian samples remained unclear. In the second comparison, dental morphology data from nine additional prehistoric samples from central and south Asia were considered. Contingency chi-square analysis indicated that 16 of 17 tooth-trait combinations differed significantly across all 15 samples. Analysis of inter-populational variation continued to provide a strong separation between Papago Indians and American whites from all central and south Asian samples. Variation among prehistoric samples provided additional evidence to support the pattern of relationships among prehistoric south Asians and central Asians identified in previous analyses. Genetic relations derived from our analysis of dental morphology confirm prior studies that suggest Mahars and Marathas are more closely related than either caste group is to the Madia Gond, who are more distantly linked to both the people of Inamgaon and to the Maharashtrian castes. Closest affinities between contemporary Maharashtrians and the prehistoric people of Inamgaon are found with the high-status Marathas. The low-status Mahars tend to be most the divergent from Inamgaon and from the perspective of dental morphology their role in the population history of Maharashtra and their claim of indigenous status remains unclear.

Acknowledgements. The co-operation of Mahar and Maratha villagers and schoolchildren in and around Pune, Maharashtra (India), and the Madia Gonds of Chandrapur District is greatly appreciated. Financial support from the American Philosophical Society, the Smithsonian Institution (Foreign Currency Program), and the National Geographic Society, for the fieldwork phase of research, and from the University of Oregon Summer Faculty Research Award Program for laboratory analysis is deeply appreciated. Many individuals assisted in logistical support and field assistance including Dr Pradeep Mehendiratta (American Institute of Indian Studies), Mr Madhav Bhandari (AIIS Pune) and Dr V.N. Misra (Director, Deccan College Post-Graduate and Research Institute, Pune). We thank the post-graduate students and faculty of Deccan College whose assistance in gathering bio-data, recording anthropometric measurements, and taking dental impressions was indispensible to the project's success (Dr P.K. Thomas, Mr Ravi Murthy, Mr Simadri Bihari Ota, Mr Joshi. The professional assistance of Dr U.S. Moorti, who took time from his busy dental clinic practice to assist us in the field, is sincerely appreciated. Officers of the US Consulate (Bombay), including Messrs Scott, McGee and Barucha, assisted in arranging air shipment of dental casts to Portland (OR). Many colleagues assisted us in research by providing reprints, citations, or by reading and providing comments on earlier versions of the manuscript. We thank K.C. Malhotra, W.D. Merchant, Traude Pillai-Vetschera, and Eleanor Zelliot for their co-operation.


BETEILLE, A 1998. The idea of indigenous people, Current Anthropology 39: 187-91.

CHARSLEY, S. 1996. `Untouchable': What is in a name?, Journal of the Royal Anthropological Institute n.s. 2(1): 1-23.

DELIEGE, R. 1993. The myths of origin of the Indian untouchables, Man 28: 533-49.

1992. Replication and consensus: untouchability, caste, and ideology in India, Man 27: 155-73.

DHAVALIKAR, M., H. SANKALIA & Z. ANSARI. 1988. Excavations at Inamgaon 1. Pune: Deccan College Post-Graduate and Research Institute.

DUMONT, L. 1980. Homo hierarchicus: the caste system and its implications. Chicago (IL): Univeristy of Chicago Press.

ENTHOVEN, R.E. 1922. The tribes and castes of Bombay 2. Bombay: Government of Bombay.

FAIRSERVIS, W.A. & F.C. SOUTHWORTH. 1989. Linguistic archaeology and the Indus Valley Culture, in J.M. Kenoyer (ed.), Old problems and new perspectives in the archaeology of South Asia: 133-41. Madison (WI): Wisconsin University Press. Wisconsin Anthropological Papers 2.

GARDNER, P.M. 1985. Bicultural oscillation as long-term adaptation to cultural frontiers: cases and questions, Human Ecology 13: 411-32.

GREEN, R.F. & J.M. SUCHEY. 1976. The use of inverse sine transformations in the analysis of non-metric cranial data, American Journal Physical Anthropology 45: 61-8.

HEMPHILL, B.E. 1991. Tooth size apportionment among contemporary Indians: an analysis of caste, language, and geography. Unpublished Ph.D dissertation, University of Oregon, Eugene (OR).

HEMPHILL, B.E., A.F. CHRISTENSEN & S.I. MUSTAFAKULOV. 1998. Biological affinities of Bronze Age Bactrians II: Dental morphology, in Lukacs (ed.): 51-77.

HEMPHILL, B.E., J.R. LUKACS & V. RAMI REDDY. 1992. Tooth size apportionment in modern India: factors of caste, language and geography, in J.R. Lukacs (ed.), Culture, ecology and dental anthropology: 231-53. Delhi: Kamla-Raj Enterprises.

KARVE, I. 1959. What is caste? IV: Caste society and Vedantic thought, The Economic Weekly 11: 49-63.

1961. Hindu society -- an interpretation. Poona: Deccan College.

KENNEDY, K. & K.C. MALHOTRA. 1966. Human skeletal remains from Chalcolithic and Indo-Roman levels from Nevasa: An anthropometric and comparative analysis. Poona: Deccan College.

KLASS, M. 1980. Caste: the emergence of the south Asian social system. Philadelphia (PA): Institute for the Study of Human Issues.

KOLINDA, P. 1978. Caste in contemporary India: Beyond organic solidarity. Menlo Park (CA): Benjamin-Cummings.

LUKACS, J.R. 1977. Anthropological aspects of dental variation in north India: A morphometric analysis. Unpublished Ph.D dissertation, Cornell University, Ithaca (NY).

(Ed.). 1984. The people of India: The biological anthropology of India, Pakistan and Nepal. New York (NY): Plenum. 1987. Biological relationships derived from morphology of permanent teeth: recent evidence from prehistoric India, Anthropologischer Anzeiger 45: 97-116.

(Ed.). 1998. Human dental development, morphology and pathology: A tribute to Albert A. Dahlberg. Eugene (OR): University of Oregon. Anthropology Papers 54.

In press. Hunter-gatherers, pastoralists, and agriculturalists in prehistoric India: A biocultural perspective on trade and subsistence, in K. Morrison (ed.): Long-term histories of tropical foragers and traders in south and southeast Asia. Cambridge: Cambridge University Press.

LUKACS, J.R., R.K. BOGORAD, S.R. WALIMBE & D.C. DUNBAR. 1986. Paleopathology at Inamgaon: a post-Harappan agrarian village in western India, Proceedings of the American Philosophical Society 130: 289-311.

LUKACS, J.R., B.E. HEMPHILL & S.R. WALIMBE. 1998. Are Mahars autochthonous inhabitants of Maharashtra?: A study in dental morphology and populations history of south Asia, in Lukacs (ed.): 119-53.

LYNCH, O. 1969. The politics of untouchability. New York (NY): Columbia University Press.

MACY, J. & E. ZELLIOT. 1980. Tradition and innovation in the contemporary Buddhist movement in India, in A.K. Narian (ed.): Studies in the history of Buddhism: 133-53. Delhi: BR Publishers.

MAHAR, J.M. 1972. The untouchables in contemporary India. Tucson (AZ): University of Arizona Press.

MAJUMDER, P.P., B. UMA SHANKAR, A. BASU, K.C. MALHOTRA, R. GUPTA, B. MUKOPADHYAY, M. VIJAYAKUMAR & S.K. ROY. 1990. Anthropometric variation in India: a statistical approach, Current Anthropology 31: 94-103.

MAJUMDER, P.P. 1998. People of India: biological diversity and affinities, Evolutionary Anthropology 6:100-110.

MALHOTRA, K.C. 1965. Human skeletal remains from Chandoli, in S.B. Deo & Z.D Ansari (ed.), Chalcolithic Chandoli: 143-84. Poona: Deccan College. Building and Silver Jubilee Series 55.

1974. Some models for the study of human population genetics in India: a review, in Indian Society Of Human Genetics (ed.), Human population genetics in India: 157-72. New Delhi: Orient Longman.

1978. Morphological composition of the people of India, Journal of Human Evolution 7: 45-53.

1984. Population structure among the Dhangar caste-cluster of Maharashtra, India, in Lukacs (ed.): 295-324.

MUKHERJEE, B.N., K.C. MALHOTRA, S.K. DAS, P.P. MAJUMDER, M. ROY, S.L. KATE & G.S. SAINANI. 1979a. Genetic polymorphism analysis among nine endogamous population groups of Maharashtra, India, Journal of Human Evolution 8: 555-66.

MUKHERJEE, B.N., P.P. MAJUMDER, K.C. MALHOTRA, S.K. DAS, S.L. KATE & R. CHAKHAKRABORTY. 1979b. Genetic distance analysis among nine endogamous population groups of Maharashtra, India, Journal of Human Evolution 8: 567-70.

NICHOL, C.R. & C.G. TURNER, II. 1986. Intra- and inter-observer concordance in observing dental morphology, American Journal of Physical Anthropology 69: 299-315.

ROBERTSON, A. 1938. The Mahar folk. Calcutta: YMCA & Oxford Publishing House.

SILVERBERG, J. 1968. Social mobility in the caste system in India: an interdisciplinary symposium. The Hague: Mouton.

SJOVOLD, T. 1973. The occurrence of minor non-metrical variants in the skeleton and their quantitative treatment for population comparisons, Homo 24: 204-33.

SRINIVAS, M.N. 1968. Mobility of the caste system, in M. Singer & B.S Cohen (ed.): Structure and change in Indian society: 189-200. Chicago (IL): Aldine.

SUNDAR RAO, P.S. 1984. Inbreeding in India: concepts and consequences, in Lukacs (ed.): 239-68.

TURNER, C.G., II, C.R. NICHOL & G.R. SCOTT. 1991. Scoring procedures for key morphological traits of the permanent dentition, in M.A Kelley & C.S. Larsen (ed.), Advances in dental anthropology: 13-31. New York (NY): Wiley-Liss.

WALIMBE, S.R. 1986. Paleodemography of protohistoric Daimabad, in S.A Sali (ed.), Daimabad 1976-79. New Delhi: Archaeological Survey of India: 641-740.

1990. Human skeletal remains, in M.K. Dhavalikar, V. Shinde & S. Apte (ed.), Excavations at Kaothe: 111-231. Pune: Deccan College Post-Graduate and Research Institute.

WALIMBE, S.R & N.S. APHALE. 1987. Dental anthropology of the female population of Marathas and Mahars of Maharashtra: an odontometric analysis, Deccan College Post-Graduate and Research Institute Bulletin 46: 151-68.

ZELLIOT, E. 1978a. Religion and legitimation in the Mahar movement, in B.L. Smith (ed.), Religion and legitimation in south Asia: 88-105. Leiden: E.J. Brill.

1978b. Dalit -- new cultural context of an old Marathi word, in C. Maloney (ed.), Language and civilization change in south Asia: 77-97. Leiden: E.J. Brill.

1986. The Buddhist literature of modern Maharashtra, in H. Van Skyhawk (ed.), Minorities on themselves: 134-49. Heidelberg: Heidelberg University, South Asian Institute.

1992a. The folklore of pride: three components of contemporary Dalit belief, in G.D. Sontheimer (ed.), Folk culture, folk religion, and oral traditions as a component in Maharashtrian culture: 62-85. Heidelberg: University of Heidelberg Press.

1992b. Mahar, in P. Hockings (ed.), Encyclopedia of world cultures 3: South Asia: 163-65. Boston: G.K. Hall.

ZUBOV, A.A. 1980. Anthropological characteristics of the Union Territory of Delhi on the basis of odontological features, in M.G. Abdushelishvil & K.C Malhotra (ed.), New anthropological data for northern India: Results of the joint Indo-Soviet expedition 1971: 218-40. Moscow: Academy of Sciences of the USSR.

ZUBOV, A.A. & U.F. GASHIMOVA. 1982. Odontological characterization of the population of western India, in M.G. Abdushelishvil & K.C Malhotra (ed.), New materials for the anthropology of western India: 157-74. Moscow: Nauka. (In Russian.)

BRIAN E. HEMPHILL, JOHN R. LUKACS & SUBHASH R. WALIMBE, Hemphill, Department of Sociology & Anthropology, California State University, Bakersfield CA 93311, USA. Lukacs, Department of Anthropology, College of Arts & Sciences, 1218 University of Oregon, Eugene OR 97403-1218, USA. Walimbe, Department of Archaeology, Deccan College, Pune 411 006, India.
COPYRIGHT 2000 Antiquity Publications, Ltd.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2000 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Article Type:Statistical Data Included
Geographic Code:9INDI
Date:Sep 1, 2000
Previous Article:Colonial Indology and identity.
Next Article:Double identity in Orissa's Golden Triangle.

Terms of use | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters