<< Back to Project Gallery

Antiquity Vol 82 Issue 317 September 2008

The political economy of iron in Late Prehistoric South India

Matthew D. Gallon


During the last centuries of the second millennium BC, smiths in South India began experimenting with iron to produce the region’s first locally made metal objects. The subsequent expansion of the production and consumption of iron objects coincided with the emergence of regional styles of ceramics, megaliths and more hierarchical forms of socio-political organisation throughout much of South India during the Late Prehistoric (c. 1200-300 BC).

In 2003 and 2005 the Early Historic Landscapes of the Tungabhadra Corridor project (EHLTC), co-directed by the Karnataka Department of Archaeology and Museums, Kathleen Morrison and Carla Sinopoli, excavated Late Prehistoric habitation and megalithic contexts at the settlement of Kadebakele, located in the Bellary District of northern Karnataka. The project recovered 37 metal objects; 32 of which are made of iron. In order to better understand the political economy of iron at both the site and regional level, I compare these iron objects and their depositional contexts at Kadebakele with those from Brahmagiri (Wheeler 1947) and Maski (Thapar 1957), contemporaneous sites located within 120km of Kadebakele (Figure 1). The distribution of production debris provides evidence that the control of iron production provided a new means for constructing socio-economic inequalities during the South Indian Late Prehistoric.

Figure 1
Figure 1. Extent of the South Indian Late Prehistoric and locations of the study sites. Click to enlarge.
Figure 2
Figure 2. Late Prehistoric sites in the Coimbatore region (adapted from Rajan 1994: Figure 2) (Tungabhadra Region indicated by grey box in inset). Click to enlarge.

Late Prehistoric iron production

Most South Indian Late Prehistoric habitation sites exhibit low densities of slag on their surface, which is indicative of the smithing of bloom into finished iron objects (Brubaker 2001). Evidence of smelting, or the reduction of iron ores to form iron bloom, is less common. Surveys in the Coimbatore region in western Tamil Nadu identified a few Late Prehistoric sites with typical smelting debris, such as a high density of iron slag, fragments of tuyeres, baked clay and roasted ore (Rajan 1994: 93; Figure 2). The distribution of iron production debris in this region indicates that Late Prehistoric smelting occurred at only a few sites, while smithing was much more widespread. A smaller but more intensively surveyed area around Kadebakele (Sinopoli & Morrison 2007) suggests a similar distribution of iron production activities in the Tungabhadra Region.

Smelting communities traded iron bloom to other communities to be made into finished objects by resident or itinerant smiths. The ability to control access to iron bloom provided emerging elites with a source of influence within their own communities and with neighbouring groups. The presence of several ‘featureless iron bars’ (22-60cm in length) in megalithic burials at Brahmagiri (Wheeler 1947; Figure 3) suggests an important association between burial in this higher-status context and the possession of bulk unfinished iron.

Figure 3
Figure 3. Unfinished bulk iron from megalithic burials at Brahmagiri (from Wheeler 1947: Figures 36 and 37). Click to enlarge.
Figure 4
Figure 4. Late Prehistoric iron objects from Kadebakele. Click to enlarge.
Figure 5
Figure 5. Iron point found in situ with a bovid maxilla in a small megalith at Kadebakele. Click to enlarge.

Late Prehistoric iron consumption

A total of 134 metal objects, 89 of which are made of iron, were recovered from late prehistoric habitation and megalithic (mortuary and non-mortuary) contexts at Kadebakele (Figure 4), Brahmagiri (Wheeler 1947) and Maski (Thapar 1957). If objects such as beads and bangles are made of metal, they are most often made of copper, bronze or gold, which suggests distinctions between the social value of iron and other metals.

Metal objects that are not ornaments are almost exclusively made of iron. These iron objects fall into three categories based on use: weapons (e.g. swords, daggers, projectile points), tools (e.g. chisels, sickles, small knives) and construction materials (e.g. nails, spikes, hooks). Many of the weapons and some of the tools show a high degree of similarity in form across the three sites.

A contingency table (Table 1) comparing the depositional contexts, in either megalithic or habitation areas, of objects from the three use-categories shows that the distribution significantly deviates from a random distribution. Z-scores for the individual use-categories indicate that construction materials are positively associated with habitation areas, weapons are negatively associated with habitation areas, and tools appear in both contexts at expected frequencies.

Table 1
Table 1. Categories of iron objects by depositional context. Click to enlarge.

Of the objects from the three sites examined in this study, weapons are the most numerous and the most commonly found in megalithic contexts (Figure 5). Since burial in megaliths was somewhat restricted to higher status individuals (Brubaker 2001), weapons probably served as symbols of social inequalities and as mechanisms for physically maintaining such differences. The high visibility of these objects and their similar forms across the study sites may indicate that they also carried messages regarding inter-group affiliations.


The production and consumption of iron objects provided new and important opportunities for the Late Prehistoric inhabitants of Kadebakele, Brahmagiri and Maski to construct and maintain social inequalities and affiliations. Control over the production and trade of iron bloom enabled emerging elites to gain influence and to increase social inequalities both within and between communities. Within this social context the consumption of iron objects, especially weapons, played an important role in creating and symbolising social differences and affiliations.


This paper would not have been possible without the support and guidance of the Karnataka Department of Archaeology and Museums, Peter Johansen, Kathleen Morrison and Carla Sinopoli.


  • BRUBAKER, R.P. 2001. Aspects of mortuary variability in the South Indian Iron Age. Bulletin of the Deccan College Post-Graduate and Research Institute 60-61: 253-302.
  • RAJAN, K. 1994. Archaeology of Tamil Nadu. Delhi: Book India Publishing Co.
  • SINOPOLI, C. & K.D. MORRISON. 2007. Vijayanagara Metropolitan Survey. Volume 1 (Memoirs of the Museum of Anthropology, University of Michigan 41). Ann Arbor: Museum of Anthropology, University of Michigan.
  • THAPAR, B.K. 1957. Maski 1954: a Chalcolithic site of the Southern Deccan. Ancient India: Bulletin of the Archaeological Survey of India 13: 4-142.
  • WHEELER, R.E.M. 1947. Brahmagiri and Chandravalli 1947: Megalithic and other cultures in the Chitaldrug District, Mysore State. Ancient India: Bulletin of the Archaeological Survey of India 4: 180-310.


  • Matthew D. Gallon Museum of Anthropology, University of Michigan, Ann Arbor, Michigan 48109, USA (Email: mgallon@umich.edu)

Published as part of a group presentation

Back to Top