Preliminary Remarks
The palaeolithic sites of India have been divided into primary, semi-primary and secondary categories, depending on their relationship with the place where the tools were first manufactured: original position, slightly removed from the original position through colluvial forces, and indeterminably removed from the original positions through alluvial agencies. The tools reported from various cliff sections, which occupy a vast space of the Indian literature on palaeolithic archaeology, belong unhesitatingly to the last group. Manufacturing places, known as ‘factory sites’ in literature, occur widely on the surface, and if there is no indication of major post-depositional disturbance, they may be assigned at least a semi-primary status. Such factory sites provide insight into how the tools were manufactured and if preference was shown for a particular type of raw material. For instance, in the Rorhi hills of Sind, Bridget Allchin noticed that the middle and upper palaeolithic people showed a distinct preference for a dark grey-coloured chert with the mottled black and rust-coloured patina or encrustation at the top, whereas the much later people of the Harappan civilization using the same area for their stone tools preferred light buff chert with a cream-coloured patina. In the Lalmai hills of Bangladesh, it has been noted that the local tool manufacturers invariably selected only those pieces of fossil wood which possessed a flinty core below the fibrous wooden matter at the top. Further, when such sites of a limited area are studied in their totality, they may throw light on some hitherto unperceived dimensions of the contemporary life. In the Raisen district of Madhya Pradesh, Jerome Jacobson located 94 Acheulian surface sites within an intensively surveyed area of 175 sq km and lying almost exclusively in a plateau plain: some near present stream channels, some on open plains away from the present streams and some (about 10 per cent of the total number) clustered at the base of the local hills. With only one exception, the in situ (i.e. the original position) status of these occurrences have been confirmed by geologists ‘who could comprehend no natural agency capable of transporting the artefacts to their present positions.’
One locality—Minarwala Kund or a large forest pool in a stream bed—showed a large number of clusters ranging from c. 1500–4500 sq m, with six of them covering c. 2000–3000 sq m (or between half and three-quarters of an acre) each. On the gridded map of the entire locality ‘the patterns of what seemed to be at least seven individual concentrations of tools could be discerned, dense towards the centres, thinning toward the peripheries’. The behavioural implications are not clear, but a pattern emerges out of the surface occurrence of tools.
The occurrence of the first type of sites, i.e. primary sites, is still very rare in the subcontinent, no doubt because of the inadequacy of research. The cave sites are undoubtedly primary, but there are only three excavated cave sites: the upper palaeolithic Sanghao and Kumool and the Acheulian rock-shelter III.F-23 at Bhimbetka. Sites buried in the alluvium or other deposits, such as those found in west Rajasthan, at Hunsgi in Karnataka, Paisra in Bihar and site 55 of Riwat in Pakistani Punjab are primary sites too. At Hunsgi, Paisra, and site 55, as will be seen below, there is some structural evidence. Hunsgi and Paisra are lower palaeolithic whereas site 55 is upper palaeolithic. As far as the upper palaeolithic is concerned, there is more evidence from Baghor I in Sidhi, Madhya Pradesh, and the Kurnool caves of Andhra. No such case can be cited for the middle palaeolithic; even at the excavations at Samnapur, which have been argued to belong to a semi-primary context, no cultural material except tools and the associated debris has been found.
There is, of course, a vast amount of literature on the palaeolithic tool types in India, which comfortably fit in the various categories of palaeolithic tools found in Africa, Europe and elsewhere in Asia. The modern classificatory system depends on a close analysis of different morphological attributes of the tool-type concerned, which is often statistically represented. Another field of statistical enquiry in this field is the inter-site variability of different assemblages. Such studies are necessary both to quantify and represent the data, although their cultural implications may not always be clear. A line of study which has not yet begun in the Indian context—at least not in a major way—is to obtain close insights into the actual manufacturing process of stone tools by replicating them. The difference in the distribution between pebble-based ‘chopper–chopping’ tools and the handaxe–cleaver assemblages, which was once an important theme of Indian palaeolithic literature, is no longer valid. If the pebbles constitute the dominant raw material of a given area, its palaeolithic industry will contain many pebble-based types. It is now acknowledged to be as simple as this. Besides, it may also be a question of functionality; some functional needs may perhaps be better fulfilled by making tools out of pebbles. Another field of study which is still largely undeveloped in the Indian situation is the use–wear analysis of tools, which is a good technique to judge the probable function of a particular tool and based essentially on microscopic studies of the wear marks at the edge of the tools and their ethnographic interpretation. The broad evolutionary sequence of the Indian palaeolithic industry from the Acheulian to the upper palaeolithic is beyond doubt, and in the field of the Acheulian itself, Misra’s postulate of two developmental stages seems to be valid.
One, probably chronologically earlier, is characterized by such core tools as handaxes, choppers, polyhedrons, and spheroids, a low proportion of crudely made cleavers and of flake tools, the predominant use of the stone-hammer technique, and the absence of the Levallois technique. … The second, and probably younger one, is marked by the low proportion of bifaces, the high ratio of cleavers to handaxes, the very high proportion of flake tools like scrapers, the extensive use of the soft hammer technique and the knowledge of the Levallous and discoid core technique.33
Earlier, we noted that the assumption of an evolutionary sequence of handaxe in India from a pebble-base is still uncertain. Although the very idea seems to be ignored in the current literature, it is worthwhile to draw attention to the results of excavations at Durkadi near Maheshwar where fresh pebble artefacts (choppers and crude flakes) were observed in stratigraphic association with one irregularly shaped, pebble-butted handaxe and two protohandaxes in the top 15 cm of a boulder conglomerate deposit overlain by a thick layer of clay. The association between a large number of pebble tools and these three pieces of handaxe in the Narmada boulder conglomerate has led to the hypothesis that there could be an evolution of handaxe from the pebble-based industry here.
The use of bone tools, which is a common feature of European upper palaeolithic, is limited in the Indian upper palaeolithic context only to the Kurnool caves, 90.30 per cent of the tools excavated in the cave called Muchchatta Chintamanu Gavi being made on shafts of long animal bones and including finished forms of scrapers, perforators, chisels, scoops, shouldered points, barbs and spatulae. However, we do not know if the use of bone tools was limited only to this part of Andhra or was common in other areas too.
Beyond Tools
Those who are not interested in the minutiae of tools will wonder why the evidence of living beyond the manufacture of tools is so very limited in the Indian context, especially in view of the fact that such evidence is not at all rare in the suitably investigated palaeolithic localities elsewhere in the world. This may be due partly to the lack of discovery of many primary sites, and partly to the fact that the spots where tools were manufactured were not the spots where people put up their hearths and spent nights. Considering that we are dealing with peripatetic hunting-gathering groups living in temporary shelters made of reeds, branches of trees and possibly even leaves in a basically arid/semi-arid landscape where considerable erosion has taken place since then and where the monsoonal rainfall will obliterate traces of temporary abodes of the last summer, we really cannot expect to find much evidence of actual living beyond the lithics in Indian palaeolithic sites, unless they were actually using stone blocks as parts of dwellings or living in caves and rock-shelters.
Hunsgi in Gulbarga, Karnataka, is one of the few open-air Acheulian sites in India to have revealed some indications of structural evidence. The evidence comes from Locality-V. In an excavated area of about 63 sq m, the Acheulian occupation level
was compact and consisted of slab-like pieces … as also pebbles and cobbles … of limestone. … The limestone pieces as well as the small number of pieces of sandstone, quartzite, dolerite, and chert, also found in the level, were obviously introduced by man from outside to serve as raw material. The floor also contained numerous fragments of granite, which form part of the rock outcrop at the locality. The fresh physical condition of artefacts and the lack of any preferred orientation in the positions of their long axes as well as those of rock fragments clearly exclude the role of fluvial agency in the formation of this level. … An interesting aspect of this floor was that its limits were defined by granite blocks (up to two metres across) on the eastern, western and northern sides. While those on the eastern and western sides form part of the granite outcrop itself, the blocks on the northern side appear to have been arranged in a line by man because they occur as individual blocks and do not extend below the level of the occupation floor. This feature, taken together with the occurrence of granite pieces … of various sizes in the occupation level, suggests that the Acheulian inhabitants selected for occupation an open area with hard surface and already surrounded by large granite boulders; improvements were made by arranging a row of boulders on the northern side. Apart from serving to demarcate the camping ground and providing protection from wind, these blocks must have facilitated the erection of temporary shelters of leaves/grasses and branches. From the fact that very few artefacts were found outside the limits set by these boulders, it is clear that the occupation floor we are dealing with is a complete one.34
Paisra is in the area of Bhim Bandh, a hot spring located in the Kharagpur forests near Munger in Bihar. In one of the Acheulian working floors excavated here (Locality D), P.C. Pant and Vidula Jayaswal noted that
The entire excavated area was found extremely rich in finished and semi-finished tools, cores, flakes, chips and blocks of quartzite, some of which were presumably meant as raw material for manufacturing tools, and as anvils. The most significant evidence yielded by this area was a series of post-holes indicative of actual habitation of the locality by the Acheulians. As many as 8 post-holes were revealed. … The post-holes were of unequal size, the larger ones measuring about 20 cm in diameter and the smaller ones only about 12 cm. Apparently, they do not seem to exhibit any definite pattern of the form of structures. However, they seemed to occur usually in groups of two, separated from each other by a distance of about 80 to 120 cm. It is quite interesting that in the case of at least two such groups, one of the post-holes was considerably larger than the other. They had been dug into the very compact deposit of haematite pellets and were easily recognizable because of their filling of light yellow earth. They varied in depth from 20 to 30 cm. The entire evidence was distinctly suggestive of thatched superstructures of temporary nature, like which is constructed by tribal nomads even today.35
Fig. 3 Lower Acheulian knife and chopping tool. Locality VI, Trench I, Hunsgi (Paddayya, 1982)
Fig. 4 Lower Acheulian cleavers from Hunsgi, Trench I, Locality VI (Paddayya, 1982)
Pant and Jayaswal also found ‘significant evidence in the form of various alignments of rocks’ which they tend to interpret as evidence of Acheulian structural activities of the site. The results of excavations in the rock-shelter III.F-23 at Bhimbetka to the south of Bhopal in Madhya Pradesh are only briefly published.
Several floors made by arranging rocks of various sizes were encountered in the Upper Acheulian, Middle Palaeolithic and Upper Palaeolithic levels. … There was no evidence of the use of fire, and no organic remains have survived. … The occupation in this and other shelters is likely lo have been seasonal, perhaps during the monsoon and winter seasons. … During the summer months human groups are more likely to have camped in the open.36
The results of recent excavations in the Sanghao cave near Peshawar are not accessible, and although the Kurnool caves of Andhra have yielded bone tools, the actual living evidence does not extend beyond the traces of a hearth fire, the only evidence of its kind in Palaeolithic India, although the direct evidence of fire goes back early in the palaeolithic sequence in China and Europe.
A positively breathtaking piece of evidence emerges in the upper palaeolithic context at Baghor I. In the centre of a roughly circular rubble-built platform of about 85 cm in diameter the excavators of Allahabad and Berkeley universities located a triangular piece of natural stone (15 cm high, 6.5 cm wide and c., 6.5 cm thick). According to the excavators, ‘… there is absolutely no doubt that the rubble platform with its unique stone, and the chert artefacts throughout the rest of the site, are contemporaneous and were made by a group of final upper palaeolithic hunter-gatherers.’ A piece of natural stone found in the centre of the platform has generated great interest. Such stones are found on the top of the Kaimur escarpment nearby and show triangular or ellipsoidal laminations which are yellowish-brown to reddish-brown in colour. They are placed on rubble-made platforms and worshipped as female principle or Shakti, in the countryside, passing for one Mai, (mother) or another. The similarity between the upper palaeolithic specimen, dated 9000–8000 BC and the modern village shrines is undeniable.37
There also seems to be a regular arrangement of rubble at the excavated upper palaeolithic site of Site 55 (minimum date 45,000 BP) at Riwat in the Potwar plateau, and the possibility of its supporting a superstructure of brushwood cannot be denied. About seven post-holes were also located on either, side of this wall-footing, but their identification as post-holes is not beyond doubt. The existence of a post-socket has also been inferred, but this, too, is not beyond doubt.38
Fig. 5 Site map of Paisra (Pant and Jayasawal, 1991)
Fig. 6 Acheulian occupation in Paisra valley (Pant and Jayasawal, 1991)
Fig. 7 Paisra, Acheulian stone alignments. Locality G (Pant and Jayasawal, 1991)
Ethnographic Approach
Although the preceding evidence from the lower palaeolithic contexts of Hunsgi, Bhimbetka rock-shelter III.F-23 and Paisra, and the upper palaeolithic contexts of the Kurnool caves, Baghor I and Site 55 of Riwat introduces an impressive and rather forbidding array of prehistoric tools of the subcontinent, its impact must be considered still very restricted. One of the ways in which scholars have tried to get around the problem is by focusing on the plausible ethnographic dimensions of the lithic surface clusters of different areas. Around Hunsgi, K. Paddayya noted the difference between a cluster of sites around a water-source and occurrences dispersed away from it in different parts of the valley. People would have congregated around water-sources in summer and would have moved away and taken advantage of a wider availability of food-sources in nature during the wet season. An ethnobotanical investigation of the area revealed about 40 species of wild edible plants, comprising 26 species of fruits/berries/pods, 10 species of leafy vegetables, mushrooms and at least one species of edible seed. Small game, including birds, must have been abundant, and formed, along with the plant resources, an ecologically suitable niche for the prehistoric hunter-gatherers. D.R. Raju has investigated a large number of prehistoric sites in the Gunjana valley of Cuddapah district of Andhra, the upper palaeolithic sites being the most visible of them (extent up to 2 sq km and the density of tools 5 to 67 per sq m). In addition to the usual tool-types of the period, a trial excavation at the site of Vodikalu also revealed 84 pieces of flat grinding stone slabs, 19 rubbers and 135 split pebbles suggesting the possible processing of plant foods at this site. Raju has also carefully listed the naturally available food-sources of his area: 30 species of game animals, 45 species of freshwater fish, 68 species of birds, 13 species of yams, wild roots and tubers, 10 species of edible leaves and 25 species of fruits. Taking clues from the Yanadi, a local group of people known for their hunting-gathering activities, Raju writes:
Fig. 8 Baghor upper palaeolithic shrine. The numbers 1 to 10 denote the fragments of vulva-shaped rock (Kenoyer et al., 1983)
… it is reasonable to predict that the Upper Palaeolithic hunter-gatherers relied on aquatic food resources for their subsistence as much as on game and wild vegetal foods. All the upper palaeolithic sites are located close to perennial water resources. Further, consumption of roots and tubers must have been an important part of the upper palaeolithic diet as the Yanadis are known to have lived during seven years (1950–57) of continuous drought mainly on wild yams … supplemented to some extent by roots, honey, fish and game. Such habits and practices possibly have a high antiquity and indicate the traditional subsistence patterns among the late Pleistocene hunter-gatherers.
In Andhra again, M.L.K. Murty makes a detailed list of the plant food used by the hunting-gathering groups of the state and concludes that ‘it seems reasonable to assume that the range of exploited wild plant foods during prehistory was no smaller than that of the present.’39
In the lower palaeolithic context at Paisra, Pant and Jayasawal make a distinction between the camp-base and resource-base areas. The Paisra valley as a whole is about 18 sq km, whereas the actual concentration of artefacts is noticed only over 2 sq km around the Paisra village. The latter area has been called camp-base whereas the wider valley which contains sporadic artefact occurrences throughout formed the general resource base zone.
A good source of the possible range of animals sharing the landscape with prehistoric humans is, of course, the extensive series of Pleistocene fossil animals found in different areas. The ancestral forms of modern cattle (Bos), horse (Equus), and elephant (Elephas) abound, and along with them occur forms of buffalo, varieties of deer, nilgai (Boselaphus tragocamelus), wild cats including lions, tigers and panthers. Interestingly, in the upper Pleistocene fossils from Susunia in Bankura, West Bengal, there is lion and in the fossils from the Kurnool caves of Andhra there is rhino. Sheep/goat (Ovis/Capra), remains come from the upper Pleistocene contexts of the Kurnool caves and the upper Mahanadi valley. Ostrich roamed the grasslands of Maharashtra and Madhya Pradesh and, possibly, parts of Rajasthan. The richness of the Indian fauna was impressive even 50 years ago, and looked at from this perspective, one can only marvel at the possible range and density of animals sharing the prehistoric landscape with humans. It would be interesting to know more about the Pleistocene background of the present day living higher plants of the Indian region (exceeding 21,000 species, which is more than one-tenth of the total number of species in the world). The blue pine (Pinus wallichiana), is supposed to have arrived early in the Pleistocene, and out of the immigrant Cedrus libanica, from west Asia emerged during this period the most majestic of trees in the Himalayas, the Cedrus deodara.40