Indian Prehistory I: The Lower Palaeolithic from the Soan Pebble Tools and the Madrasian Acheulian of the Kortallayar Valley to Attirampakkam, Didwana, the Belan Valley, Bhimbetka, Chirki-Nevasa, Hunsgi, the Krishna Valley and Hathnora; the Middle and Upper Palaeolithic; and the Mesolithic of Bagor, Tilwara, Langhnaj, Adamgarh, Chopani Mando, Sarai Nahar Rai and Birbhanpur

Indian prehistory began on 30 May 1863, when Robert Bruce Foote of the Geological Survey of India picked up a quartzite handaxe in a laterite pit at Pallavaram near Madras; later that year, with William King, he found the rich Acheulian site of Attirampakkam in the Kortallayar valley. For the first half of the twentieth century Indian Lower Palaeolithic cultures were described as two traditions. The Soanian, named from the Soan river in the Potwar plateau near Rawalpindi (now in Pakistan), was defined by Helmut de Terra and T.T. Paterson of the Yale–Cambridge expedition of 1935 (published in 1939, Studies on the Ice Age in India and Associated Human Cultures): a pebble-tool industry of choppers (flaked on one face) and chopping tools (flaked on both) on quartzite river pebbles, with flakes, divided into an Early Soan and a Late Soan in which flakes and prepared cores became commoner, and placed by the two authors in a sequence of Himalayan glaciations and Soan terraces. Hallam Movius (1948) made the Soan part of his "chopper-chopping tool complex" of South and East Asia, set against the handaxe cultures of the West across his Movius line. The Acheulian of the peninsula — handaxes, cleavers, choppers and flake tools on quartzite, and later on basalt, dolerite and limestone — was called the Madras handaxe industry or "Madrasian", after the region of Foote's finds, and its sequence was worked out in the Kortallayar valley near Chennai by Paterson and V.D. Krishnaswami. That two-tradition picture has since dissolved. Soanian choppers occur in the Siwalik foothills of Punjab and Himachal Pradesh (the Beas–Banganga and Sirsa valleys) often in contexts no older than the Acheulian, and many are now read as a regional facies or as tools of a later period, while Acheulian sites cover almost the whole of India outside the Ganga plain, the far north-east and the Western Ghats' wettest slopes. The Acheulian is the Lower Palaeolithic of India. It is also one of the oldest outside Africa: cosmogenic-nuclide burial dating of Attirampakkam's lower levels (Shanti Pappu and colleagues, 2011) gave about 1.5 million years, and Isampur in the Hunsgi–Baichbal valleys about 1.2 million.

Read together, the sites show what Acheulian life in India was like. Toolmakers chose their stone deliberately — quartzite in the south and in the Belan, basalt and dolerite on the Deccan Traps, limestone at Hunsgi — and worked it at quarry-workshops such as Isampur and Chirki, where roughouts, flakes and hammerstones lie as they were dropped. Large-flake cleavers, struck from giant cores and trimmed, are a distinctive Indian and African trait and set the Indian Acheulian beside Africa's rather than Europe's. Sites cluster by rivers, lakes and springs on the edge of hills, as at Hunsgi and Didwana, where the tools lie in fine lake sediment, not rolled gravel, and so preserve activity areas. Bhimbetka shows that rock shelters were occupied from the Acheulian onward and gives a continuous sequence into historic times. And Hathnora supplies what the rest lack: the Narmada calvarium, found by Arun Sonakia in the Narmada's boulder conglomerate with handaxes, cleavers and the bones of Stegodon, Bos namadicus and Hexaprotodon, now read as an archaic Homo sapiens of the heidelbergensis grade — the maker, or a relative of the maker, of India's later Acheulian. The Acheulian ended in India, as elsewhere, by about 200,000–150,000 years ago in most places, giving way to the flake industries of the Middle Palaeolithic, though at Attirampakkam the transition came earlier.

The Middle Palaeolithic of India, about 385,000 to 40,000–30,000 years ago, is a flake-tool culture: side-scrapers, points, borers and small handaxes on flakes struck from discoid and Levallois-type cores, made in fine-grained jasper, chert, chalcedony and agate rather than the Acheulian's quartzite. H.D. Sankalia defined it at Nevasa on the Pravara in the 1950s as the "Nevasian" (his "Series II", above the Acheulian "Series I"), and it is known across the peninsula and central India. In the Belan valley it lies in Cemented Gravel II; at Bhimbetka in the shelters above the Acheulian floors; in the Narmada valley in the gravels above the boulder conglomerate; at Didwana in the 16R dune. Two recent findings have reshaped it. At Attirampakkam, Levallois technology appears about 385,000 years ago, far earlier than the arrival of anatomically modern humans in India, which suggests a local development by archaic hominins. At Jwalapuram in the Jurreru valley of Andhra Pradesh, Middle Palaeolithic tools lie both below and above the ash of the Toba super-eruption of about 74,000 years ago, and the argument over who made them — archaic hominins or early modern humans who had arrived before Toba — is still open. The Upper Palaeolithic, about 40,000 to 12,000 years ago, is marked by blades, burins, backed blades, scrapers and points, by bone tools and by ostrich-eggshell beads and engravings; its best-known sites are the syllabus's. Renigunta in Chittoor district, Andhra Pradesh, where M.L.K. Murty in the 1960s described a blade-and-burin industry, was the first clear Indian Upper Palaeolithic; the Billa Surgam caves in the Kurnool district, explored by Robert Bruce Foote and his son Henry in 1884, yielded late Pleistocene fauna with bone tools; Patne in the Jalgaon district of Maharashtra (S.A. Sali) gave a sequence from the Middle Palaeolithic to the Mesolithic with ostrich-eggshell beads and an engraved eggshell fragment of about 25,000 years; Bhimbetka's shelters carry Upper Palaeolithic floors; the Son and Belan valleys (Gravel III in the Belan) hold Upper Palaeolithic sites, among them Baghor I in the Son valley, where J.D. Clark, G.R. Sharma and J.M. Kenoyer found a late Upper Palaeolithic platform with a natural triangular stone bearing concentric laminations, read as a shrine because the Kol and Baiga of the area still worship such stones as the goddess; Visadi in Gujarat and the dune sites around Pushkar in Rajasthan extend it to the west; the syllabus also lists a site in the Gunjan valley. Microlithic assemblages in India go back to about 45,000–35,000 years at some sites, which blurs the line between the Upper Palaeolithic and the Mesolithic.

The Mesolithic of India, from about 10,000 BCE and in many regions to the arrival of farming several millennia later — and among some forest peoples far later still — is the culture of the post-Pleistocene world. The end of the last glacial brought a warmer climate and, in the early Holocene, a stronger south-west monsoon: the Rajasthan lakes (Didwana, Sambhar, Lunkaransar) filled, dunes were stabilised by vegetation, the Ganga plain's oxbow lakes teemed with fish and turtles, and grassland and forest spread; later, in the mid-Holocene, aridity returned in the north-west. People responded with a broad-spectrum economy — hunting deer, antelope, wild cattle and pig, fishing, fowling, collecting tubers, fruit, honey and wild grain — and with a new technology. Microlithic tools, typically under three centimetres, were made on bladelets struck from small fluted and conical cores of chert, chalcedony, agate, jasper and quartz, first as non-geometric forms (backed bladelets, points, lunates) and later as geometric ones (triangles and trapezes), with the microburin technique; they were set in grooves of bone or wood with resin to make composite tools — arrows barbed and tipped with several microliths, sickles, knives and harpoons. The bow and arrow, shown in hunting scenes in the rock paintings of Bhimbetka and Adamgarh, transformed hunting. Mesolithic society was organised in bands, some seasonally mobile between the plains and the hills, others semi-sedentary where food was rich, as by the lakes of the Ganga plain, where the sites have hut floors, hearths, querns and cemeteries; the dead were buried extended under the floors or near the huts, sometimes in pairs, with ornaments of bone and shell and grave goods, and the burials yield the tall, robust Mesolithic people studied at Sarai Nahar Rai, Mahadaha and Damdama. Towards its end, the Mesolithic shows the first signs of change: domestic sheep and goat at Bagor and Adamgarh, handmade pottery at Chopani Mando and Langhnaj, and exchange with farming neighbours — the long coexistence of foragers and farmers that still marks India.

The sites fall into regional groups whose differences are ecological. In Rajasthan and Gujarat, Bagor, Tilwara and Langhnaj sit on dunes and river banks of a semi-arid land that the early Holocene made wetter, and they show the transition to herding: Bagor's lowest phase, with its thousands of microliths and its bones of sheep and goat alongside wild deer and pig, is the classic case of a microlith-using people who kept domestic animals before they made pottery, and its later phases show copper, pottery and eventually iron arriving among people who kept their microlithic kit. In central India, Adamgarh and Bhimbetka are rock-shelter sites in the Vindhyas and Satpuras, with painted walls and continuous occupation. In the middle Ganga plain, Sarai Nahar Rai, Mahadaha and Damdama are open-air lake-side settlements whose cemeteries, hearths, bone ornaments and rich animal remains point to a semi-sedentary life, with radiocarbon dates spread across the early Holocene. In the Vindhyan valleys, Chopani Mando's three phases lead from late Pleistocene microliths to beehive huts, wild rice and cord-impressed pottery, which G.R. Sharma saw as the prelude to the Neolithic of Koldihwa and Mahagara in the same valley (next chapter). In the east, Birbhanpur shows a simpler quartz industry on the laterite uplands. The syllabus names Bagor twice; it is one site, and the repetition should be read as a slip.