D D KOSAMBI: THE MAN AND HIS WORKEconomic & Political Weekly EPW july 26, 200871Kosambi’s ArchaeologyShereen RatnagarD D Kosambi offered remarkable insights into the history of ancient India. Does his archaeology measure up to his history? The approach in this paper is to view the internal logic of his hypotheses in archaeology, and to ask if Kosambi did justice to the data available in his time. Did he present a sound data analysis that could be emulated, enlarged, or reworked? The answer has to be “no”. Kosambi’s site locations were not precise; he was not interested in the typology of stone tools; and his correlations of tool occurrences with sacred sites, of the tribe with an absence of plough agriculture, and of iron technology with agricultural surpluses, were flawed. Perhaps Kosambi’s archaeology does not measure up to his history because for him archaeology was only an extension of history. But in order to work with the entities of archaeology, typology and classification are indispensable: as indispensable as is the knowledge of an ancient language for the historian. Failure to engage in the grammar of these entities and an ignorance of site formation processes give rise to faulty generalisation.For his exploration of the development of Indian civilisation D D Kosambi utilised not only written sources but also archaeology and ethnography. History has been made, he said [Kosambi 1957: xii], not by the glittering autocrat but by ordinary people, often backward and ignorant. Once our pre- occupation with ruling dynasties is discarded, it is essential to consider the material conditions in which production was organ-ised in society, to ask how land was held, how metals were pro-cured and so on. And it is archaeology that gives us information about the humdrum, including the tools of production utilised by ordinary people since the dawn of history. Because societies advanced in each age as far as their technology allowed (“man makes himself”, he said (ibid: 6), echoing Gordon Childe) the role of archaeology became important to Kosambi’s enquiries. Except for a perceptive piece on Harappan silver [Kosambi 1941], Kosambi’s published thoughts on Indian prehistory and protohistory date to the 1950s and 1960s (more precisely, to 1966, the year of his death). By then, several Indus civilisation sites had been excavated by the Archaeological Survey of India (ASI), and the archaeology departments of both the M S University, Baroda and the Deccan College, Pune not to speak of departments fur-ther afield in Banaras, Chandigarh, Nagpur, etc, that were active in the field with other kinds of sites. B Subbarao published his pathbreaking The Personality of India in 1958, in which he tried to find regional patterns in the development of prehistoric cultures across the country. In Kosambi’s own city, what was then Poona, the department of archaeology at Deccan College was flowering, with specialists such as a geomorphologist, epigraphist, pre-historians, art historian, Sanskritist and archaeological chemist on the faculty.Deccan College faculty and students had found the Stone Age sites in Gujarat, along the river valleys of the northwest peninsula, Madhya Pradesh and Andhra Pradesh. The department had excavated, by the mid-1960s, sites from Rajasthan to northern Karnataka: Langhnaj, Ahar, Navdatoli and Maheswar, Nasik, Jorwe, Brahmapuri, Sangankallu and Tekkalakota. These spanned the Neolithic, Chalcolithic and Iron Ages. A volume entitledIndian Prehistory 1964followed an all-India conference organised by the Deccan College: it was an evaluation and inter-pretation of material gathered in the last 15 years.1 Thus there were varied resources available to any scholar residing in Poona who was interested in archaeology.It would not be fair to evaluate Kosambi’s archaeology at a dis-tance of four decades by citing evidence unearthed since his death. The approach here is instead to view the internal logic of his hypotheses and to ask if he did justice to the data available in his time. Was his a body of work that could be carried forward, Shereen Ratnagar (rshereen@vsnl.com) is an archaeologist who is currently working on tribal communities, past and present.

D D KOSAMBI: THE MAN AND HIS WORKjuly 26, 2008 EPW Economic & Political Weekly72did he present a sound data analysis that could be emulated, enlarged, or reworked? Did his articulations provoke problem-oriented research in the decades to follow, or were his ideas repeated by succeeding generations just because they came from a Marxist?Before we proceed, it may be well to note that Kosambi’s ar-chaeological discoveries (and his interpretations of them) were not published in peer-reviewed archaeological journals such as Ancient India (Delhi) orAntiquity(Cambridge). Three notes (not full-fledged articles) appeared in 1962 in the anthropological journal, Man (London). 1 The Indus CivilisationKosambi made several casual but provocative suggestions con-cerning the Indus (or Harappan) civilisation. His inference that it was not forceful domination by elites, but an ideological glue that held this urban society together was not particularly original, as it came from the Orientalist thinking of Marshall, Piggott, and others who were steeped in the paradigm of India as a land of tradition and religion (in contradistinction to the dynamism of the west). Moreover, it was based on the oft-repeated observation that the excavated weaponry of the civilisation was scant and technologically poorly developed [Kosambi 1957: 62-65, 1970: 64], a generalisation that was not quite satisfactory: there were excavated projectiles of metal and baked clay for attacking the enemy at a distance, and there were also a few stout metal axes and stone mace-heads for face-to-face fighting. Besides, there were what Mortimer Wheeler termed the “defences” of the citadel at Harappa.2 One may contrast this weaponry with contemporary well-developed metal axes and spearheads from western Asia, but the point is that on the Indus plains it was not Mesopotamians or Egyptians, but south Asian people whom the Harappan armies would have fought. And in this region we have no cultures with a well-developed metal technology or armoury. More important, while Kosambi realised that no urban society can be non-stratified or classless, he failed to appreciate that social stratification itself cannot rest on a purely theocratic base (i e, rule by an organised priesthood [Kosambi 1957: 68-69]) – ultimately it must be backed by the threat of the use of force, however much the dominance of the rulers is disguised in the trappings of ritual paraphernalia or divine sanction. Kosambi’s belief in religious unification is to me curiously un-Marxist. Other insights on the Indus civilisation came from his reading of the archaeological evidence in tandem with that of contem-porary Mesopotamia and are more valuable. Already in 1955 [Kosambi 1955a: 62], he had seen that the urban fraction in the Indus area was much lower than in Mesopotamia (i e, the cities and urban population of Mesopotamia were, proportionately, very much larger): no one else had cared to engage in such ana-lytic comparison till then. Kosambi suggested – quite rightly – that this contrast was due to differences in agricultural output. For him, much of the reason lay in the absence of canal irrigation and the lack of a heavy iron plough [Kosambi 1970: 58-62] in the Indus region. The iron plough repeatedly reappears like a leitmotif, in his writing; we shall deal with it later, but suffice it to state here that differences in productivity could have been due to the necessity of lift irrigation in the ‘rabi’ season in the Indus zone, which meant that agriculture here was much more labour- intensive and required considerable cattle power. This would have severely limited the area a peasant could put under crops in any one rabi season, whereas in Mesopotamia, with its huge cities, irrigation was a matter of gravity flow, canals having to be opened and closed at the right time [Ratnagar 1986]. Thus Kosambi did indeed articulate an idea that proved fruitful. A point of considerable interest is Kosambi’s reading (1941) of cuneiform marks on a rectangular sheet of silver (hitherto uunpublished) excavated in a jar below a Mohenjo-daro house floor. This small piece occurred together with about a dozen others in the jar, most of them bearing “incisions”, but it was the only one carrying what looked like Mesopotamian cuneiform writing. Kosambi is put out by the fact that the cuneiform inscrip-tion is not mentioned in the final reports on Mohenjo-daro. This may have been due to doubts about the reading. The sign as drawn by him is not a recognisable one in the syllabary of the period, and inIntroductionKosambi (1957: 78, note 10) admits that the experts he consulted were not convinced that this was cuneiform writing. This evidence if substantiated supports the view that silver was coming to India from Mesopotamia, but it may not support Kosambi’s suggestion that there was a connec-tion between cuneiform and writing in Sanskrit.2 Cult Spots and Prehistoric TracksWhat about Kosambi the field archaeologist? In his books and in a few articles he made cultural connections between surface assem-blages of microlithic tools, non-brahmanical cult spots, and pre-historic movements or migrations [Kosambi 1962b: 110-51]. When he located a folk shrine where blood sacrifice was occasionally offered and found a microlithic-tool assemblage in the vicinity, he inferred that the cult-spot had hoary origins. (In Myth and Reality (1962b: 82-109) the logic is that the earliest food-gathering cultures had goddess worship, whereas male gods came with pastoralist intruders.) A short distance away, Kosambi would find another scatter of Late Stone Age tools or else another open-air shrine with an aniconic stone. In Introduction [Kosambi 1957: 263-67], he states that the best concentrations of tools were in the vicinity of the cult spots, which, for him, marked the sites of pre-historic camps located at the junctions of tracks used by animal herders and for trade. Kosambi (1962b: 82-85)was struck by the importance of crossroads as places of blood sacrifice through the centuries in this country. He also tried to relate prehistoric tracks with early first-millennium trade routes that affected the location of Buddhist monasteries in western Deccan (ibid: 95-96).Kosambi was able to conceive of landscapes as dynamic rather than backdrops to human activity. He could transport himself to a prehistoric landscape and people it with a facility that is remarkable and largely accounts for the fascination of his writ-ing. And yet, on deeper investigation, we realise that there are as many holes in this cultural construction as there are in the “pierced microliths” he reported having found [Kosambi 1962a].3 (According to Sankalia (1979: 117) the latter were bits of stone with holes caused by natural processes.)

D D KOSAMBI: THE MAN AND HIS WORKEconomic & Political Weekly EPW july 26, 200873First, the tools. Kosambi (1957: 51, note 5) says his collections were verified by F R and B Allchin, although it is not specified which particular collections were shown to them. Actually, in a volume dedicated to his memory, the Allchins (1974: 45) remark that Kosambi’s archaeology was less successful than his other works. It is not clear whether Kosambi’s drawings are of waste flakes or of actual tools. None of his finds feature inIndian Archaeology – A Review, the annual published by the Archaeologi-cal Survey of India that reports all discoveries in brief, giving the locations and naming the finders. And then, the simple proximity of tools and cult spots cannot be adequate reason to infer a his-toric or cultural association, especially when it is admitted that these two kinds of phenomena do not occur exclusively in associ-ation with one another [Kosambi 1962b: 94]. Links with Stone AgeNeither is there a justification to interpret folk shrines as primor-dial in existence since prehistoric times. To be fair, Kosambi does warn the reader that the continuity of cult spots does not mean that those who worship there are descendants of stone-age peo-ple, still worshipping the same goddess. But in spite of upheavals, migrations, epidemics and population displacements, old cult sites would not be ignored by later settlements, he thought [Kosambi 1962b: 110].4 Thus after the decline of the Buddhist monastic esta-blishment, mother-goddess worship was re-established in the old places. In my experience among tribal people of eastern Gujarat, however, shrines set up along the roads and/or in sacred groves usually endure for about 75 to 150 years; they then fall into disuse and the old lamps, daubed stones and the terracotta offerings are pushed aside in a heap and are forgotten. No person who worships at these sacred places has claimed in my presence that they are centuries-old, leave alone primordial. Their short duration is con-firmed by the quantity of offerings that accumulates at these spots.Not only did Kosambi not care to report his sites in the official annual or journal meant for the purpose, he avoided making a typology of the stone tools,5 confining himself to remarks that certain sites could not be tool-making loci because of the pro-portion of cores in the assemblages (ibid: 127). He used a number of descriptions such as “excellent”, “in profusion”, a marked “density of tools”, or a “striking deposit of microliths”, terms that lack precision. As for the ethnographic analogy with the use of flaked stone by the Dhangar herdsmen to castrate their sheep, why did he think to mention it repeatedly in different pub-lications, when he was honest enough to admit that the Dhangar method of hammering out a tool had nothing to do with pre-historic flaking, that the tools are totally dissimilar and that the Dhangar were unable to identify microlithic flakes of chalcedony as tools [Kosambi 1962a: 435-7; 1967: 35-36 and photograph on p 34; 1970: 42]? Rural people pick up stones for many uses as they go about their everyday lives, but this does not indicate a link with the Stone Age. Amateurish ArchaeologyNow to the prehistoric tracks that are said to have developed into trade routes of the historic period [Kosambi 1962b: 95 ff]. The Late Stone Age endured for many centuries, so that we cannot be certain that all the assemblages of microlithic tools in a certain area were contemporary. Students may also beware the notion that intensive fieldwork and on-the-ground observation will reveal the exact placing of millennia-old roads or tracks. It is one thing to do basic map work and realise that, at the regional level, Ujjain rather than Chhota Udepur, or Nasik rather than Dahanu, lie on major routes. But on the ground, routes are hard to pin-point. For instance, river fords and crossings are important points on routes, but a ferry crossing means moving downstream with the flow of the river, so that it is not always easy to pick up the continuing route on the opposite bank. We may suppose that the Mula-Mutha could have been forded in the dry weather, for centuries, at Yeravada (in Pune). But do clusters of tools in Bund Garden and at the Yeravada traffic lights (on opposite banks) prove this? What if the tools on the left bank lie not at the traffic lights, but on Parnakuti hill behind them? Perhaps a family had stopped on one river bank during a hunt, or had spied fish in the river, discarding its tools on the spot after having caught its prey. A flaked-tool assemblage on Parnakuti hill may indicate that after crossing the river, some hunters once went up to scout out the land on the left bank; but it could also be the result of hunters repeatedly using the hill to watch for game, hunters who did not often cross the Mula-Mutha. Besides, not even with careful tool counts and comparative quantification can we distinguish repeated visits from a one-off event. Lastly, surely the traffic of the Buddhist-period traders, carry-ing bulky or heavy or precious things over long distances from source to urban consumption point or port, and interacting en route with the monastic establishment, was qualitatively differ-ent from the movements of prehistoric hunter-gatherers. Hunter-gatherers characteristically move their camps over four-five kilometre distances, perhaps wandering off the route a short dis-tance to chase game; they use tracks through forest to stalk game, or to set up traps. Several camps may converge on scarce hot- season water bodies. Hunter-gatherers are known to occupy areas for short periods. Only if one believes that all routes are pre- existing givens carved into the physiography of the land, can one claim that hunters’ tracks developed into trade routes of the his-toric period. While the major highways of the country do follow the topography, they fall in place as routes according to and in tandem with the growth of settlements, forest clearance and the laying of fields, herdsmens’ movements, and the movements of goods and people. It may be more realistic, then, to look on land-scapes as palimpsests. So Kosambi’s work has all the freshness and colour of amateur archaeology, but not the rigour required of archaeological reasoning. In the Poona region Kosambi claimed to have found megaliths: these were not the Karnataka or Vidarbha-type stone-slab sepul-chres or cists or stone cairns surrounded by boulder circles. Instead, they were piles of boulders that Kosambi believed were manmade even though there were no burials beneath them. He was tempted to make this inference because he detected grooves in the forms of circles and other designs that he thought could not have been natural marks [Kosambi 1962c]. When Sankalia (1979: 117), however, went to these sites in the company of Kosambi, he did not accept that they were megaliths. Even if they

D D KOSAMBI: THE MAN AND HIS WORKjuly 26, 2008 EPW Economic & Political Weekly74do turn out to be manmade memorials, it is difficult to accept Kosambi’s (1970: 38-39) reasoning that they are the work of cattle herders who made “thicker” flake tools than did sheep and goat herders, the reasoning being that cattle hides are thicker than those of the smaller animals. The leap from the perceived formof microlithic flakes (that were set in pieces of wood or bone for composite tools) to differences in subsistence patterns is not acceptable.3 TechnologyAs indicated at the outset, the materialist approach entails delin-eating a sequence of social formations (the relations and the means of production) and their respective forms of property, labour, and surplus mobilisation. The tools of the past are relevant in that they indicate to us the conditions in which ordinary people engaged in production. Apropos of this paradigm is Kosambi’s oft-repeated, but never substantiated or thoroughly argued, view that when the heavy, iron plough was taken into tribal areas hitherto cultivated in the shifting, slash-and-burn, pattern, agricultural production came into its own and large surpluses were produced. Only (iron) plough agriculture was “real agriculture” [Kosambi 1962b: 147], “regular agriculture” [Kosambi 1970: 44]. The other was “primitive” (ibid) and could not produce a surplus.6 This is an echo of colonial thinking on the subject. Ravi Rajan (1998: 218), for instance, quotes a forester’s contrast (1920) between shifting cultivation and “real agriculture”. The British considered shifting cultivation wasteful, a cause of erosion, primitive, and backward [Prasad 1998]. For Kosambi (1970: 34), there could be no “proper society” until there was a system of surplus produc-tion – which takes political incorrectness to extremes, even for the context of the 1960s.In the greater Indus valley, in Harappan times, Kosambi felt, surplus production had been possible to a limited extent. Even though the plough used then could not have been of iron (which came into use centuries later), the soils here required only a light ploughing [Kosambi 1957:57, 1970: 58-59]. Thus the discovery in the 1960s of the pre-Harappan ploughed field at Kalibangan does not invalidate his argument. Kosambi (1957: 23) echoed Gordon Childe’s observation that iron, which is plentiful on the surface of the earth, is cheap. The “real” plough (ibid: 70) was iron-shod and this alone made possi-ble the systematic settlement of the thickly-forested Ganga plains. As for Maharashtra, it was emphatically stated [Kosambi 1962b: 111] that the village economy spread into this region only in the 6th century BC. Even though the caveat was lodged [Kosambi 1957: 52; 1970: 44] that shifting cultivators of recent times are also capable of wielding the plough,7 the theory is flawed. This merits discussion here, not the least because the equation of the coming of iron with adequate ploughing of the soil and the rais-ing of good harvests that could support ruling elites has been repeated so often that it appears to have become irrefutable. One wonders how an academic working on archaeology in the 1960s could have failed to note that a number of archaeological sites of the pre-iron, pre-“real-plough” centuries, had been exca-vated in peninsular India by the M S University and the Deccan College. There were Neolithic-Chalcolithic sites in Maharashtra, Malwa, Bellary district and the Raichur doab. Some of these have yielded the seeds of multiple crop plants and the bones of dome-sticated animals. Given the general superimposition, in vertical sequence, of house floors, the evidence of storage facilities, a depth of some nine feet of habitation debris at Jorwe, a one-period Chalcolithic site and intramural burials, these could only have been the villages of sedentary mixed-farming communities. Already in 1955 the excavators of Nasik mound had suggested that its pre-iron occupation lasted several centuries, between the later second millennium and the mid-first millenniumBC. Already in 1958 Subbarao had taken many of these sites into account in working out his interpretation of the pattern of cultural development across the various regions of the country. (In fact Subbarao (1958: 150) gave importance to the few copper axes found at some of the Chalcolithic sites and suggested, for his part, that it was these copper tools that enabled groups to clear the peninsular forest and grow crops in the river valleys.) 8Use of the PloughNext, agricultural production cannot be reduced to a matter of tools. Ploughing means the dragging of a heavy wooden contrap-tion across a field, to create a furrow in which crumbled soil will give a good micro-environment for seeds to sprout with the first rains and for crop plants to grow. Animal traction introduces energy and manure on the field, but animals require space to walk and turn around: not every jungle clearing will be suitable. Sometimes stony ground, sometimes a slope without terracing, or a field with standing trees, will rule out the use of the plough. The red soils on hill slopes tend to be leached soils, with heavy annual monsoon rain carrying water-soluble minerals deep into the profile and out of the reach of the roots of the crop plants. It may be preferable to leave a substantial tree cover in place so that their roots conserve the soil and to do “garden cultivation” around these old trees with the hoe and digging stick, for a short period, after which the patch will be left fallow. Even on fertile river plains, deep ploughing is not the secret of productivity: in a monsoon region in which there are eight months of dry weather, the less the soil is exposed to the sun and wind, the better. Thus it is known that some tribal people have used iron-blade hoes for tilling the soil. In his monograph on the Munda country Sarat Chandra Roy (1871-1942)9 referred to tribal agriculture in which the plough played a central role. Conversely, when peasant groups became refugees in the forested highlands of western India, finding the current prices of ragi high, they too would take to slash-and-burn cultivation of this crop [Pouchepadass 1995]. Pouchepadass (ibid: 151) says that a slash-and-burn patch may give a high output even when no ploughing is done on the adjacent plot below. True, Pouchepadass’ work is much later than Kosambi’s, but even in co-lonial times Voelcker (1893) had spelt out the flaws in the colonial policy that considered heavy ploughing the only respectable form of tillage: he cited the monsoon climate and its long months of dry weather and the dangers of precious moisture evaporating from the ground; he did not believe that Indian agriculture could be improved with the introduction of heavy iron ploughs; most significant, he appreciated the sound ecological sense behind the

D D KOSAMBI: THE MAN AND HIS WORKEconomic & Political Weekly EPW july 26, 200875system of ‘rab’, a local form of shifting cultivation on the Western Ghats. And further, if Kosambi had been able to read V Elwin’s Agaria, a classic monograph on a marginalised tribe of iron smelters,published as early as 1942, his theory about the role of iron in agriculture and peasant and tribal life would have acquired more content and balance. Thinking in terms of a dichotomy, tribe versus organised society [sic], Kosambi (1967: 31) wrote that as the use of the plough advanced, there was a mutual acculturation, a two-sided “adjustment” providing “both the fabric and the pattern of India’s past”. He said [Kosambi 1970: 172] the co-option of tribal chiefs by stratified society was not sufficient for the transformation of the tribe: it was the brahmins who carried plough agriculture into tribal villages, and with it, new crops, knowledge of distant markets, and trade. (Perhaps he is right in saying that it was the brahmins who carried a calendar into these areas, but students should not confuse the calendar with a knowledge of the seasons, as is sometimes done.) In contrast, his contemporary Subbarao had seen the fuller picture of tribal life, including hunting- gathering, root-crop agriculture, and a range of specialised crafts, including metal work and weaving. Subbarao noted that tribal habitats were for centuries the sources of good wood and metal required in peasant villages and towns. And he underlined the fact that no tribe today uses stone tools [Subbarao 1958: 141ff]. An even richer and more subtle tapestry was woven by Nirmal Kumar Bose – who, incidentally, was also academically active in the 1960s and also combined, in his scholarship, archaeology, Indology, and ethnography.10 Bose (1994) saw many processes of tribal absorption at work in Indian history, among them the advent of craft specialisation amongst tribal groups and the affiliation of some of them, as specialists, with the regional economy; weekly markets and annual fairs that are meeting grounds for people and hubs for the dissemination of knowledge; the institution of the temple; and various local festivals and pilgrimage centres [Later, K S Singh (1985) was to elaborate on these processes].114 Tribal PeopleIn the first great transformation in history, said Kosambi (as did many others, notably Gordon Childe, before him), appeared agri-culture and animal herding. Kosambi (1955b: 308) said the agrarian village economy came to replace the “tribal” [sic] way of life. He tends to mistake Stone-Age hunter-gatherers for tribes, when in fact hunting-gathering-fishing constitutes the pre-tribe stage of cultural development. The tribe as a social formation came into existence only with the coming of agriculture.12Tribal people and their subsistence and forms of ritual, we have seen, were relevant to Kosambi in that they represent “sur-vivals” of “primitive” ways that help us to interpret the bare b(st)ones of the archaeological record. Long survivals were feasi-ble in our country because biodiversity made hunting-gathering-fishing relatively easy [Kosambi 1970: 34]. Indian history is a his-tory of long continuities, he said (which was not a new idea) and it is of primary importance to investigate the tribal presence within the interstices of mainstream society, or within the cities, because the absorption of tribes into the mainstream as castes is the signature of Indian civilisation [Kosambi 1957: 7, 27-30; 1970: 13]. Kosambi (1967: 30) went so far as to state that among tribes the way of life has remained “largely unchanged” since prehis-toric times. This amounts to denying marginalised groups a his-tory and it was an assumption handed down in the Orientalist tradition. Elsewhere, however [Kosambi 1957: 24], he acknow-ledged that customs may persist down the centuries more in form than in content. Yet to me it is significant that it was his archaeo-logist contemporary, B Subbarao (1958: 144) of M S University, who noted that in 1953, within an Oraon area in Chhotanagpur, there were old, abandoned (undated) village sites that yielded large numbers of stone tools – which Oraons in the 1950s did not use – and a form of pottery that was in sharp contrast to that in use in 1953.13 Kosambi was also mistaken in the notion that tribalism equates with matriarchy. He suggests that those who came last to plough agriculture are groups among whom there remain vestiges of matriarchy. This argument loses clarity when he asserts that pastoralist Aryans introduced male gods, because the Vedic Aryans were tribal people. In any case, in standard anthropological theory today there is no such thing as matriarchy – matrilineal succession is another matter. Matriarchy and matriliny are analytically separate phenomena.PrejudicesObservations of tribal people appear to have been made by Kosambi within the urban metropolis of Poona rather than in tribal villages. He was thus observing small numbers of people who had been dispossessed – deracinated people, in fact. Not only this, he swept all tribal groups – Kadar, Bhil, Santal, Munda – into the “primitive” [Kosambi 1957: 26-27] category, “fossilised” by their inability to take to regular food production, an “obsolete form of society”. In contrast, Bose (1940: 44-45), working among tribal people in their habitat, recognised change (as did Subbarao). He showed some prehistoric hand-axes to forest hunter-gatherers who ate roots dug up from deep down in the soil. They said these hand-axes would not enable them to dig out the roots. Even the iron tips of their digging sticks became blunted in the course of digging and the points of these stone axes would break if used for digging. And then there was Roy who corrected his own references to tribal people as “primitive”, “superstitious”, or at a “low level” of culture, and in the 1930s wrote appreciatively about tribal village councils and oral poetry (even though he continued to speak of “uplift”) [Dasgupta 2007: 158-67]. Kosambi made no such self-correction. The settlement of the Vaddar people (today labelled a denotified tribe) was fast turning into a slum, which in his view was “a problem for the Poona municipality” [Kosambi 1957: 32, emphasis supplied]; given small patches of land, the Pardhi began to grow vegetables [which contradicts his statement about their refusal to progress to agriculture] and would, hopefully, begin to purchase soap with their earnings (ibid: 33) – their body odour is mentioned (ibid: 34) in a passage probably more malodorous than anything he could have smelled. In his posthumously published book is the statement [Kosambi 1970: 13-14] that production for self consumption is “precisely the backward, inefficient and local”

D D KOSAMBI: THE MAN AND HIS WORKEconomic & Political Weekly EPW july 26, 200877Notes 1 This volume maintains its historiographic impor-tance until today. Although Kosambi had pub-lished short notes on Late Stone Age material (that he had himself found) in the journal Man and had written a discursive text on the cultural significance of microlithic sites in hisMyth and Reality(1962b), he did not participate in the sym-posium. Was he cold-shouldered by the archaeo-logical establishment as an outsider and a Marx-ist? No: in H D Sankalia’s department all were welcome. Sankalia and his students, in fact, went over to see some sites that Kosambi had discov-ered (it appears that Sankalia rubbished most of these “discoveries” [see Sankalia 1979: 117]). This meeting occurred shortly before Kosambi’s death, says Sankalia (ibid), which would make it in 1965 or early 1966. Judging from a review by Kosambi in 1964 of a book by Sankalia, it was Kosambi who had for the latter a disdain that he did not care to disguise. 2 Today we also know [Ratnagar 1991: 78-88] that across a few sites the strata that represent the transition from the earlier developmental phase to the mature Indus are characterised by ash lay-ers and cinders, together with the debris of walls and artefacts – obviously the remains of fire and destruction, warfare in other words. 3 In this brief note, Kosambi was suggesting that the distribution of microlithic assemblages and cult sites may delineate the prehistoric begin-nings of the Pandharpur ‘jatra’. 4 In one of his posthumous articles [Kosambi 1967: 48], the emphases are somewhat different. The continuity of worship at mother-goddess cult sites is dismissed as implausible because hunter-gatherers and pastoralists were not sedentary peoples. 5 The drawings of microliths published inMyth and Reality (pp 133, 134, 151) are amateur work. Stone- tool drawings are tantamount to an analysis of their form and function, an indication of how they were made. In Kosambi’s drawings the concavity/convexity of flake scars is not indicated; no bulb of percussion occurs on any diagram; nor are small retouch flakes shown along any working edge. An archaeological drawing should depict two if not three sides of a tool, so that the typo-logy becomes evident in the drawing itself, but Kosambi’s diagrams except in three cases remain views of one face. Were these, then, actual tools or were they just waste flakes? 6 Today the thinking is that “surplus” production is not the production of excess food over basic wants; that it is a factor of distribution and not of production. In other words, the populace surren-ders surplus (in grain, labour, or rent/tax) not because it has plenty, but because it is ruled by a dominant elite that has the monopoly of the use of force. 7 For a description of this see Hardiman 1995.8 Further down in the text Subbarao says the economic development of Magadha was largely due to the development of iron technology (1958: 155 ff). 9 As far as I am aware, Kosambi made only a pass-ing reference to Roy’s Mundas and Their Country, first published in 1912. This culture was indisput-ably tribal, with exogamous lineages, fission and fusion among them over time, the burial of the dead, bride-price, ancestor worship and offerings to tutelary deities in sacred groves, witchcraft, and, most important, communal land tenure. And here, the hill slopes were terraced, embanked, and irrigated and tilled with the plough. 10 Bose’s essay on the Hindu ways of absorbing tribal culture was published in 1941. He died in 1972.11 Singh (1985) contested the idea of a one-way flow of technology from brahmin to tribe, pointing out that there was also the role of artisan tribes, trad-ers, and pedlars. There were also intermarriages, and sometimes when tribal groups were power-ful, emulation was in the opposite direction. For instance, in the northern hills, brahmin migrants turned non-vegetarian and married hill women. 12Many hunter-gatherer groups are classed as scheduled tribes according to the provisions of the Constitution. This does not mean that they are analytically the same thing as tribes – the Constitution does not set out to define the tribe or the caste, its purpose being to select groups for positive discrimination. Students with doubts on this issue are referred to Steward 1955; Kent 1996 (especially 9-17); and Ingold et al 1988, among others, on hunter-gatherer societies; and to Sahlins 1968 and 1972; Mahapatra 1993; Godelier 1977 (especially 70-96); Kuper 1999: 159-200; and Claessen 2000 (especially 123-132). 13 Nowhere does Kosambi refer to the fieldwork or interpretations of Subbarao.14 In a book on Mortimer Wheeler’s archaeology and its legacy in India [Ray 2008], is a chapter enti-tled ‘The Indus Civilisation and the Archaeology of Fortifications’. This chapter fails to refer to Wheeler’s landmark contribution: his deep trench, about 30 m long, up to 6 m wide, and 14 m deep, across the Harappa citadel wall and “Sec-tion HP XXX” that gave the relative date of a forti-fication; Wheeler was able to distinguish in the cut a mud rampart, a brick wall, a revetment, and a platform abutting on the wall within that cut and the serial order in which they were built; moreover, his stratification was clear enough to show up the horizontal strata against which this perimeter was built, thus giving the whole a rela-tive date. That this was an outstanding achieve-ment of stratified digging is not even acknowl-edged, leave alone discussed, in this book.ReferencesAllchin, F R and B Allchin (1974): ‘The Relationship of Neolithic and Later Settled Communities with Late Stone Age Hunters and Gatherers’ in R S Sharma and V Jha (eds),Indian Society: Historical Probings in Memory of D D Kosambi, People’s Publishing House, Delhi, pp 45-66.Bose, N K (2006) [1940]: ‘Prehistoric Researches in Mayurbhanj’, Science and Culture 1940, rpt Selected Writings, Centre for Archaeological Studies and Training, Kolkata. – (1994): The Structure of Hindu Society,A Beteille (ed), revised edn, Orient Longman, Hyderabad (first published 1975).Claessen, H J M (2000):Structural Change: Evolution and Evolutionism in Cultural Anthropology, Universiteit Leiden, Leiden.Dasgupta, S (2007): ‘Recasting the Oraons and the Tribe’ in P Uberoi, N Sundar and S Deshpande (eds),Anthropology in the East, Permanent Black, Ranikhet, pp 132-71.Elwin, V (1942): The Agaria, Oxford University Press, London.Godelier, M (1977): Perspectives in Marxist Anthropology, Cambridge University Press, Cambridge.Hardiman, D (1995): ‘Small Dam Systems of the Sahy-adri’ in D Arnold and R Guha (eds),Nature, Culture, Imperialism, Oxford University Press, pp 185-209.Ingold, T et al (eds) (1988): Hunters and Gatherers: History, Evolution, Social Change, Berg, Oxford (two volumes).Kent, S (ed) (1996):Cultural Diversity among Twentieth-Century Foragers, Cambridge University Press, Cambridge.Kosambi, D D (1941): ‘On the Origin and Development of Silver Coinage in India’, Current Science,Vol X, pp 395-400, reprinted pp 85-94 in D D Kosambi, Indian Numismatics, (ed) B D Chattopadhyay, Orient Longman, Hyderabad. – (1955a): ‘Stages of Indian History’, Journal of the Indo-Soviet Cultural Society, Vol 1, No 1 and re-printed pp 57-72 in B D Chattopadhyay (ed),Com-bined Methods in Indology, Oxford University Press, Delhi.– (1955b): ‘The Basis of Ancient Indian History’, Journal of the American Oriental Society, Vol 75, and reprinted pp 308-26 in B D Chattopadhyay (ed), Combined Methods in Indology, Oxford University Press, Delhi. – (1957): An Introduction to the Study of Indian History, Popular Prakashan, Bombay.– (1962a): ‘Pierced Microliths from the Deccan’, Man, January 1962 and reprinted pp 433-37 in B D Chattopadhyay (ed),Combined Methods in Indology, Oxford University Press, Delhi. – (1962b):Myth and Reality, Popular Prakashan, Bombay. – (1962c): ‘Megaliths in the Poona District’,Man, May 1962, note 108, and reprinted pp 438-43 in B D Chattopadhyay (ed),Combined Methods in Indology, Oxford University Press, Delhi.– (1967): ‘Living Prehistory in India’, Scientific American, February 1967 and reprinted pp 30-48 in B D Chattopadhyay (ed),Combined Methods in Indology, Oxford University Press, Delhi. – (1970): The Culture and Civilisation of Ancient India in Historical Outline, Vikas, Delhi.Kuper, A (1999):Culture: The Anthropologists’ Account, Harvard University Press, Cambridge Mass.Mahapatra, L K (1993): ‘Customary Rights in Land and Forest’ in M Miri (ed), Continuity and Change in Tribal Society, IIAS, Shimla, pp 84-96.Pathy, Suguna (1987): ‘Class Formation in an Indian Tribe’,Social Science Probings, Vol 4, No 4, pp 396-435.Pouchepadass, J (1995): ‘British Attitudes towards Shifting Cultivation’ in D Arnold and R Guha (eds), Nature, Culture, Imperialism, Oxford University Press, pp 123-51.Prasad, A (1998): ‘The Baiga’, Studies in History, Vol XIV, No 2, pp 325-48.Rajan, Ravi (1998): ‘Foresters and the Politics of Colonial Agroecology: Shifting Cultivation and Soil Erosion’,Studies in History, Vol XIV, No 2, pp 217-36.Ratnagar, S (1986): ‘An Aspect of Harappan Agricul-tural Production’,Studies in History, Vol 2, No 2, pp 137-153. – (1991): Enquiries into the Political Organisation of Harappan Society, Ravish, Pune. Ray, H P (2008): Colonial Archaeology in South Asia: The Legacy of Sir Mortimer Wheeler, Oxford University Press, Delhi.Roy, S C (1970): The Mundas and Their Country, Asia Publishing House, Bombay.Sahlins, M D (1968): Tribesmen, Englewood Cliffs, Prentice-Hall. – (1972): Stone Age Economics, Aldine, Chicago.Sankalia, H D (1979): Indian Archaeology Today, Ajanta Publications, Delhi.Singh, Kumar Suresh (1985):Tribal Society in India, Manohar, Delhi.Steward, J (1955): Theory of Cultural Change: The Methodology of Multilinear Evolution, University of California, Urbana.Subbarao, B (1958): The Personality of India, M S University, Baroda.Voelcker, J A (1893): Report on the Improvement of Indian Agriculture, Eyre & Spottiswoode, London.Given the current disposition in which academic status often rests on ideological positions, one does not expect much self-correction. Yet in the long term it is not enough to be proclaimed as a leftist/secularist/Marxist – the gloss of identification with a charmed circle will inevitably wear off over time, and then the scrutiny of a person’s scholarship will be of a different order al-together. An object lesson, perhaps, for young and not-so-young scholars with professional ambitions.