Eastern India

The nature of early Neolithic societies in eastern India has been less well studied than other parts of the subcontinent. However, there is a growing corpus of information from various streams of evidence available in the archaeological literature.

Only a few archaeological sites in this area have been radiocarbon dated. What dates we have suggest that the Neolithic began around 2500 bce, but these dates largely come from later deposits of Neolithic-Chalcolithic sites, leaving the earliest levels undated.^[537] The beginning of the Neolithic in this area is often placed at 3500 bce^[538] on the basis of typological similarities with other, better-dated parts of South Asia. The results of several regional surveys and excavations indicate that there are two main settlement types present in the eastern Neolithic. The first of these, the Upland tradition, can be found in the foothills and uplands. The sites are very shallow with few archaeobotanical and zooarchaeological remains. They are often identified through lithic and pottery scatters and appear to represent seasonal camps. The second tradition, of the coastal plains of Odisha, is characterized by mounded settlement sites up to 8 m high, from which both archaeobotanical and zooarchaeological material has been recovered.

Our best evidence for early farmers in east India comes from the coastal plain settlements, which have been referred to as the Eastern Wetlands tradition: Khameswaripali, Golbai Sasan, and Gopalpur.^[539] The large settle­ment mounds of Golbai Sasan and Gopalpur have been systematically excavated and submitted to full archaeobotanical analysis. The mounds are around 150-200 m in diameter and up to 8 m high. Scant evidence for structures in the Neolithic, in the form of floor layers and occasional post holes, suggests roughly oval hut dwellings. Red-ware and grey-ware ceramics were recovered from both sites, as were bone tools and several celts. Detailed macrobotanical and phytolith analyses have provided much of the informa­tion about the type of agriculture practised in the east Indian Neolithic. In this region agricultural production was dominated by rice, with approxi­mately 50 per cent of the crop remains recovered from each site identified as rice grains.^[540] The sites are situated on alluvial soils close to seasonally flooded rivers, and rice grains, rice-type phytoliths, and spikelet bases have all been recovered from the very earliest levels. Other crops include pulses (horse­gram and pigeon pea), plus a few small millets (browntop millet and Setaria sp.). This diet was largely based on rice and beans complemented by livestock meat. Cattle rearing and possible pig rearing occur by the Chalcolithic, but probably also took place in the Neolithic. Pig rearing is of particular interest as modern genetic data indicate that Indian pigs and wild boar are highly distinct from those elsewhere in Asia,^[541] and that pig bones are generally rare on sites elsewhere in Neolithic-Chalcolithic South Asia. Faunal remains also demonstrate hunting, with wild species such as chital deer and nilgai appearing in the archaeological record.^[542] Chalcolithic levels provide evidence for fishing in the form of bone harpoons and marine fish remains; however, similar remains have not been found from Neolithic levels.

Of key interest to archaeologists studying early crop cultivation is the discovery of pigeon pea (Cajanus cajan) at both Gopalpur and Golbai Sasan. While both the frequency and ubiquity of finds were low, it appears likely that this was a cultivar domesticated within the region.^[543] Wild progenitors of pigeon pea are present in the Eastern Ghats today and it is likely that their range was slightly broader in the past, though still centred on the Eastern Ghats. The eastern Neolithic therefore appears to have an independent, local domestication of a pulse crop. The east Indian Neolithic has been considered as a possible area for the management or cultivation of native wild rices, but recent work has challenged this theory. The earliest rice in this region is found alongside domestic-type spikelet bases (around 70 per cent of the identifiable spikelet base assemblage, which is expected in domesticated populations), and morphometric data suggest that most rice grains are consistent with domesticated populations.^[544]

In addition, there is some evidence for fruit and nuts at Gopalpur and Golbai Sasan. Citrus-type peel has been found in deposits at Gopalpur and unidentified nutshell fragments have been recovered from both sites. Whether the nuts were collected from wild trees or represent some form of arboriculture is not clear, but it is likely that tree species were managed to some extent at this time. Indeed, citrus trees are native to a belt that runs across the Himalayas. Therefore the presence of citrus-type peel in this area may indicate the movement by people of trees across the subcontinent and the early stages of arboriculture. The movement of economically valuable tree species, such as citrus, could be used to track shifting cultivation by providing a terminus ante quem.^[545] In this case it is rather late, c. 1500 bce, but the evidence provides a glimpse of a usually unobserved activity.

The small, ephemeral sites of the foothills, such as Bajpur, Banabasa, and Kuchai, have little or no archaeological remains outside of lithic scatters.^[546] Several have Mesolithic and Neolithic layers (often identified through the presence of pottery sherds), indicating continuity of at least some groups during the transition to agriculture in this region. The almost total lack of archaeobotanical evidence has been taken to indicate the seasonal nature of these settlements. It seems that they are temporary camps, possibly occupied by hunter-gatherers or shifting cultivators, or perhaps specialists involved in lithic production.^[547] The environment in which they are located could have been used for agriculture. However, there is what appears to be a cultural demarcation in the landscape of Neolithic east India that separated settled agriculturalists from more mobile groups.

It is possible that people using these upland sites consumed tubers and gourds (various Cucurbitaceae), which are more ephemeral in the archaeolo­gical record. Both can be pit-roasted, removing the need for pottery, of which there is very little at these sites. Unfortunately, both are hard to identify archaeobotanically and are less likely to be preserved than cereal grains. More work needs to be done before we can truly begin to develop hypotheses about the subsistence regimes at these sites; however, both taro and yams are found in wild forms in Odisha and it is possible that these are wild progenitor species.

One of the richer streams of evidence on the Neolithic societies of east India comes from stone tool analysis. In particular, several production sites have been located in the Sundargarh district of Odisha.^[548] Here Neolithic sites are concentrated in areas of dolerite rock with which most tools were made.

While evidence is relatively thin on the ground, it is clear that east India had a unique and complex Neolithic society. It is certainly one that deserves further investigation. The interplay of farmers and mobile groups is particu­larly interesting, as is the domestication of the native wild pigeon pea.

Although there is little archaeological evidence for the Neolithic cultures of northeast India and Bangladesh, the region deserves brief discussion. It is worth flagging that recent genetic work on rice identifies the aus rices, which include several landraces of Bangladesh and northeast India, as being as genetically distinct from other Oryza sativa types as indica is from japonica.^[549] Although conventionally classified morphologically as indica, genetic data suggest these represent a third group, plausibly from a separate domestica­tion somewhere in this region.^[550] It has often been proposed that this area saw an influx of tools, pottery traditions, and agricultural plants and animals from Southeast Asia.^[551] Unfortunately, to date we do not have the data to securely back up this hypothesis.