Material Culture & Craft Production in South India

Review of Craft Specialization, Technology and Social Change: A study of material culture in Iron Age and Early Historic South India (c. 1200 BCE – 400 CE), by Gwendolyn Ida Ortner Kelly

In her dissertation Gwendolyn Kelly analyzes a large body of data on craft activity (especially bead and ceramic production) in order to explore the relationship between craft production and society and economy in Iron Age (1000 B.C.E. – 400 B.C.E.) and Early Historic (400 B.C.E.- 400 C.E.) South India. This dissertation is one of very few research projects that methodically analyze data on craft production in early South Asia, and as such Kelly’s discussion and arguments are of great relevance to scholars interested in issues of craft production in general and South Asian studies in particular.

In Chapter 1, Kelly provides a brief overview of the archaeological history and geography of southern India and outlines her central research questions and the data used. The latter includes excavated materials from the sites of Kodumanal (Tamil Nadu), an Iron Age and Early Historic craft production site with both habitation and megalithic burial components; Pattanam (Kerala), a large Early Historic port and Kadebakele (Karnataka), a habitation site occupied from the mid third millennium B.C.E. to the 16th century C.E. In addition, Kelly analyzes published data from the large Early Historic port of Arikamedu (Tamil Nadu). This chapter contains a detailed discussion of the archaeology and context of each of these sites, providing an excellent and necessary background to the discussion of the sampling strategy used to select the materials to be analyzed from each site.

Following Kenoyer’s (1989, 1991, 1995 etc.) model of craft specialization, Kelly emphasizes two key components in her discussion: control over production and social stratification. As she writes:

Hence, the wide distribution of crafted goods such as beads and bangles, their standardization of forms and materials over a large area, and the evidence of control over production are all indicators of craft specialization – the process, as it is associated with complex stratified social organization, and the production of crafted goods that are used in the symbolic expression of wealth that reinforces social hierarchies (p. 22).

Archaeological variables used to analyze these dimensions of craft specialization include the relative ranking of both raw materials and finished products as measured by the availability of raw materials and the technological complexity involved in production; the spatial distribution of these differently ranked items within burial and settlement contexts as indicative of social differentiation and the spatial distribution of different stages of bead production as reflective of the organization of production.

Chapter 2 contains an overview of previous research, with an emphasis on issues of trade, social and political organization and local and regional economies. Kelly moves chronologically from the Neolithic (circa 3000 – 1200 B.C.E.) to the Iron Age (circa 1200 B.C.E. – 400 B.C.E.) and finally to the Early Historic (circa 400 B.C.E. – 400 C.E.), highlighting theoretical and methodological issues relevant to her own work. The chapter thus contains a critical ‘master narrative’ of major historical developments in South India and also makes explicit Kelly’s own contribution to these discussions, that lies in filling a considerable gap left by the paucity of systematic studies of craft production and organization in Iron Age and Early Historic South India.

Chapter 3 contains a discussion of beads (and some other stone ornaments) from Kodumanal, Pattanam and Arikamedu, framed within an analysis of the chaînes opératoires or operational sequence used in their production. The author recorded multiple physical attributes for the beads, and identified techniques of manufacture through microscopic study. The raw data used in this chapter is provided in monumental appendices I and II at the end of the dissertation. Kelly identifies the predominant chaînes opératoires at each site and argues that, in most cases, bead production was locally organized (although at Pattanam a large proportion of the assemblage was of non-local microcrystalline materials) and a hybrid tool-kit of technologies was used.

Kelly takes issue with Peter Francis’ (1991, 2002, 2004) argument that the differences in technologies (pecking vs. grinding), raw materials (macro-crystalline and local vs. microcrystalline and non-local) and chaînes opératoires (drilling before and after polishing) of bead production at Arikamedu could be correlated to the presence of two distinct communities of bead-makers: one indigenous to South India and the other migrants from Gujarat. Using her data on the beads from Kodumanal and Pattanam, Kelly shows that this pattern cannot be applied to South India as a whole. She highlights the considerable variability in raw materials used and in the technological choices for bead-making in (local) macro-crystalline materials (pecking and grinding; drilling before and after polishing; types of drills used and drilling technique etc.), all of them making it difficult of identifying distinct communities of bead makers. Kelly does, however, identify a strong correlation between non-local microcrystalline materials (primarily carnelian and agate) and specific technological choices (the use of grinding, drilling from both sides). In the absence of archaeological evidence for non-local material culture, she interprets this as possible evidence for the presence of small itinerant groups of bead makers who worked with these materials—an explanation that would also fit the patterns identified at Arikamedu.

The stone bead and other ornament (terracotta and bone beads and bangles) data from Kadebakele is discussed in Chapter 4. Kelly identifies patterns in the temporal and spatial distribution of these materials at the site; and – since the assemblage is typical of those at other contemporaneous sites – she also theorizes about the organization of trade and craft production in Iron Age South India. Due to the observed diversity of stone raw materials used as well as the low counts of beads produced from each material, Kelly uses Renfrew’s concept of a fall-off curve in the quantity of raw materials the greater the distance from the source (Renfrew 1975, 1977) to argue that trade in the Iron Age was down-the-line and opportunistic, without systematic access to non-local materials. The exception seems to have been trade in carnelian and steatite (with raw material sources likely outside of South India). The former was the most abundant bead material used for bead making at Kadebakele and beads of this material were deposited in a variety of contexts, with a roughly equal distribution between ritual/burial and habitation loci. Bead production at the site was confined to locally available raw materials (terracotta, bone, shell) and was a part-time pursuit conducted in domestic spaces, which is representative of Iron Age modes of production, as Kelly argues. She attributes the observed standardization of carnelian beads to the possibility that the common motifs were ideologically and socially significant across large parts of the subcontinent. Beads in South India likely were markers of status and identity. At Kadebakele they were furthermore found in ritual contexts, showing re-use over a considerable period of time.

In Chapter 5, Kelly examines the collection of finished beads and ornaments from Kodumanal to identify spatial and temporal patterns in social differentiation. Kelly ranks the materials used on a 10-point scale that measures four major attributes across a continuum (low, medium, high): abundance (standardizing the frequency by volume of soil excavated in each unit), distance from source (including considerations of methods of transport etc.), technological complexity (stages in the production process etc.), and quality (quality of raw material, skill needed, expediency in production). The ranking of raw materials thus obtained is used to analyze the assemblages (standardized by volume excavated) from each excavation unit, showing that while there is evidence for increasing wealth over time, there is limited evidence for wealth/social differentiation within the settlement. However, the assemblage in the burials (roughly contemporaneous to the habitation) is very different, containing little of the diversity of materials found in the habitation but primarily materials ranked above a 7 according to Kelly’s system, in particular agate and carnelian beads (n=3687 vs. 63 in the habitation areas). This indicates that the people buried were significantly wealthier and of higher status than those in the habitation and Kelly acknowledged the possibility that excavations might have missed elite residential areas. She also suggests that groups with the most social, political and economic power might have lived in urban centers larger than Kodumanal.

Chapter 6 moves away from merely identifying the presence or absence of specialized production and instead examines the organization of specialized craft production at Kodumanal by asking questions about the nature of production and the ways in which individuals and households invested labor in craft production. Analyzing excavated materials and notes from the 1985-1993 excavations at the site, Kelly identifies the spatial and temporal distribution of stages of manufacture across the site. In the case of quartz and other stone beads there was no identifiable trend towards more production but there were variable levels of intensity and volumes of production over time; in fact, there was some evidence of decline in production towards the end of occupation at the site. While some areas of the site which were loci of large-scale production, Kelly argues that the variability in amount and intensity of the various stages of production indicate the presence of part-time producers, although some households may have devoted more time to craft production than other. The segregation of the stages of production in different areas of the site could imply the presence of cooperative household or supra-household networks that divided labor. Due to the absence of large quantities of finished ornaments, it seems that Kodumanal, in fact, was an important center of production of bead and other stone ornaments for trade. In the case of shell bangle manufacture, however, from the small scale of production Kelly argues for the presence of itinerant bangle production specialists.

Chapter 7 contains an excellent typology of historic ceramics from the excavations at Kodumanal, which potentially has wider application for other archaeologists working in South India — a region for which there is no widely accepted ceramic chronology. As Kelly rightly notes, ware categories in South Asia are far too widespread temporally and spatially to be the sole criteria for classification, and her typology is based primarily on vessel form (bowl/ dish, lids, cooking pots/ storage jars), with sub-classifications based on rim form, and to a certain extent on wares. Separate classification systems based on decorative motifs and base forms are also provided. The metric and non-metric data includes attributes recording a variety of vessel dimensions, external decorative elements and methods of production (type of firing, use of wheel or various types of hand building etc.), and is presented in Appendix III. As an aside, Kelly’s appendices containing her raw data (on 3236 coded sherds for this chapter), while more than doubling the length of her dissertation, will be very helpful for comparative purposes and future ceramic analysis and research in South Asian archaeology. Similarly, Kelly’s discussion of the attributes used in analysis, the method used in sorting and classifying the ceramics and the detailed description of each form category is a great contribution to the field of ceramic analysis in general and South Asian ceramics in particular. Finally, Kelly also presents a preliminary seriation of vessel forms and wares (Appendix IV).

The final chapter places the sites discussed in the dissertation within a larger geographical and historical framework and outlines the broader implications of Kelly’s research for understanding the development of craft production and social organization in Iron Age and Early Historic South India. In particular, Kelly argues against the idea of a unilineal progress towards increasing complexity and suggests that the Iron Age pattern of varied levels of craft specialization for different crafts (seen at Kadebakele) and for ad-hoc down-the-line trade, changed by the Early Historic to one of complex, overlapping networks of movement and exchange. However, as at Kodumanal, there was no continuous trend towards increasing specialization but a variable pattern suggesting flexibility and choice on the part of the producers.

Kelly’s dissertation is a welcome addition to the study of craft specialization in early South Asia and is one of the few systematic, problem-oriented studies of craft production and technologies in the Iron Age and Early Historic periods. While some of Kelly’s arguments are necessarily preliminary, this dissertation is an important step towards beginning to formulate models of the organization of craft production in the region. The ceramic typology presented in the dissertation in particular is of exceeding relevance for archaeologists working in early South India and South Asia. Kelly has generously illustrated the assemblages studied with numerous figures and images. In addition, the inclusion of her raw dataset too makes her assemblage available for future comparative studies that will further refine our understanding of early craft production.

Uthara Suvrathan
Cornell Institute for Archaeology and Material Studies
Cornell University

Francis, Peter.
1991    ‘Beadmaking at Arikamedu and Beyond’. World Archaeology, 23(1): 28-43.
2002    Asia’s Maritime Bead Trade: 300 B.C. to the Present. University of Hawaii Press,
2004     ‘Beads and small finds from the 1989-92 Excavations’. In The Ancient Port of
Arikamedu: New Excavations and Researches 1989-1992, edited by V. Begley, N. Karashima,
K.V. Raman, S.E. Sidebotham, & E.L. Will, pp. 447 – 604. École Française d’Extrême Orient,

Kenoyer, J. M.
1989     ‘Socio-Economic Structures of the Indus Civilization as reflected in Specialized Crafts
and the Question of Ritual Segregation’. In Old Problems and New Perspective in the
Archaeology of South Asia, edited by J.M. Kenoyer, pp. 183-192. Wisconsin Archaeological
Reports, Madison.
1991   ‘ Harappan craft specialization and the question of urban segregation and stratification’.
Eastern Anthropologist, 44(3-4).
1995     ‘Interaction Systems, specialized crafts and culture change: The Indus Valley
Tradition and the Indo-Gangetic Tradition in South Asia’. In Indian Philology and South Asian
Studies, edited by A.A.M.W. Wezler, Walter de Gruyter, Berlin.

Renfrew, A. C.
1977    ‘Alternative Models for Exchange and Spatial Distribution’. In Exchange Systems in
Prehistory, edited by T.E. Earle & J.E. Ericson, pp. 71-90. Academic Press, New York.
1975     ‘Trade as Action at a Distance: Questions of Integration and Communication’. In
Ancient Civilization and Trade, edited by J. Sabloff & C.C. Lamberg-Karlovsky, pp. 3-59.
University of New Mexico Press and the School of American Research, Albuquerque.

Primary Sources
Published material from the sites of Kodumanal (Tamil Nadu), Pattanam (Kerala), Kadebakele (Karnataka), and Arikamedu (Tamil Nadu)
Original excavation notes from the 1985-1993 excavations at Kodumanal
Microscopic analysis on beads and other stone ornaments from all four sides conducted by the author

Dissertation Information
University of Wisconsin-Madison. 2013. 1560 pp. Primary Advisor: J. Mark Kenoyer, Professor, Anthropology.

Image: “Bleached Carnelian Beads from Kadebakele” by Gwendolyn Kelly.

Leave a Reply