Late Pleistocene to Early Holocene Lithic Industries in the Southern Fringes of the Himalaya

Quaternary International xxx (2010) 1e11

Contents lists available at ScienceDirect

Quaternary International

journal homepage: www.elsevier.com/locate/quaint

Late Pleistocene to Early Holocene Lithic industries in the southern fringes of the Himalaya

Claire Gaillard a,*, Mukesh Singh b, Anne Dambricourt Malassé a a UMR 7194 CNRS, Département de préhistoire du Muséum national d’Histoire naturelle, Paris, France b Society for Archaeological and Anthropological Research, Chandigarh, India article info abstract

Article history: The Siwalik Range especially in its western sector is known for numerous surface sites yielding large Available online xxx quantities of cobble tools. These tools have for long been considered to belong to the Lower Palaeolithic but they are sometimes associated with other apparently more evolved tool types having a transversal trimmed cutting edge. These types look like adzes/axes and are made from large ﬂakes, split cobbles or ﬂat cobbles. Moreover, such assemblages composed of choppers along with these adze/axe-like tools occur on geological surfaces formed in the late Pleistocene; they are probably later than these deposits in age. These assemblages compare well with the industries occurring further east in Nepal or northern South-East Asia related to the Hoabinhian industries. It is suggested that they are linked through subsistence and technical behaviours to the particular environment of the sub-Himalayan belt and its eastern extension characterised by densely forested hilly landscapes, generously irrigated by perennial streams. Besides, this region was not severely disturbed during the last glacial maximum and could have been a refuge for many animal and vegetal species. It has certainly favoured “latitudinal” circulation (precisely circulation along the geomorphological features) of human and animal populations despite the global climatic changes. Ó 2010 Elsevier Ltd and INQUA. All rights reserved.

1. Introduction However, lithic artefacts occur in plenty, especially in the western Siwaliks. In the sub-Himalaya region, all along the highest mountain The earliest of these artefacts are represented by a few flakes in range of the world (Fig. 1), late Pleistocene and early Holocene the Soan valley (Indus basin) dated to ca. 2 Ma (Dennell et al., 1988). human settlements are poorly known. The Neolithic begins around At the time of the first surveys in the Potwar, in the 1930s, two types 5 ka BP in the Swat Valley, with hand-made pottery associated with of lithic assemblages had been identified. The Acheulian, often rolled a lithic assemblage rich in cobble tools (Stakul, 1981), and in the and in a lower stratigraphical position, was considered as older, while Kashmir Valley it begins around 4.5 ka BP with a possible aceramic the Soanian (further sub-divided into two stages) was considered as phase at Burzahom (Agrawal, 1982; I.A.R., 1961e1962; I.A.R., later (Teilhard de Chardin, 1936, 1937). In synthesising the data 1964e1965). At the opposite end of the sub-Himalayan belt (in regarding the Lower Palaeolithic of South and East Asia, Movius a wide sense) in Northern Thailand, the Neolithic emerges in the (1944, 1948) considered the early Soanian and the Acheulian of the mid-Holocene with a local facies called Bacsonian, in continuation western Siwaliks as belonging to the same time period; in his of the Hoabinhian (see Forestier et al., 2006). Except in a few sites in conclusions he proposed a sketch of the Old World distinguishing Nepal (Corvinus, 2007), there is no trace in this 2500 km long strip two Lower Palaeolithic cultural areas: “handaxe culture” and of hills, covered with forests and generously supplied in water by “chopper-chopping tool culture”. Subsequently this concept was many streams, of the microlithic technical stage that characterises oversimplified and became “the Movius line”. Although the Acheu- the late Pleistocene and early Holocene of peninsular India. lian is acknowledged even by Movius himself in both cultural areas, this famous line is still a matter of debate (Keates, 2002; Corvinus, 2004; Petraglia and Shipton, 2008), recently leading to the “Movius * Corresponding author. Department of Prehistory, Institut de Paléontologie line sensu lato” (Norton et al., 2006; Norton and Bae, 2008)that Humaine, 1 rue René Panhard, 75013 Paris, France. Tel.: þ33 (0)1 55 43 27 37; fax: actually matches well with the original proposition of Movius. þ33 (0)1 43 31 22 79. E-mail addresses: [email protected] (C. Gaillard), [email protected] However, geological and chronological contexts suggest that the (M. Singh), [email protected] (A.D. Malassé). early Soanian, at least in a number of cases, is later than the

Fig. 1. Map of south and south-east Asia showing location of the main sites or regions mentioned in the text.

Acheulian. Moreover, in central Nepal assemblages mostly the richest one and was first found at the base of the Potwar loess, comprising cobble tools are dated between 24 and 20 ka (Kadereit assigned to the Upper Pleistocene, and therefore was regarded as et al., 2007), and in eastern Nepal a rich site yielding cobble tools the youngest. It was considered as a local Middle Palaeolithic, along with adzes/axes akin to Hoabinhian types dates back to Early especially rich in cobble tools. The latter was usually occurring in Holocene (Corvinus, 1987, 1989). This latter site known as Patu is the form of rolled but very typical specimens in the basal gravel of a key for understanding the cobble tool industries of the western the Potwar loess and was considered as belonging to the underlying Siwaliks (Soanian), and probably also those of the Irrawaddy Boulder Conglomerate, the last formation of the Upper Siwaliks terraces (Anyathian) in Myanmar (Movius, 1944, 1948). Looking in (Teilhard de Chardin, 1936, 1937). a broader perspective, it appears that comparable industries are Since then many assemblages of different facies have been common further east in late Pleistocene and early Holocene collected in the Western Siwaliks (Rendell et al., 1989; Dennell, contexts, for instance in Thailand and Vietnam. They precede the 2004, 2007, 2009; Chauhan, 2007, 2008), especially between Hoabinhian tradition and progressively merge into it. At places they Indus and Sutlej (Mohapatra, 1966, Mohapatra and Singh, 1979a,b). already include some of the characteristic Hoabinhian tool types, The Acheulian, characterised by handaxes and cleavers, fits well to which are much more significant than the time-transgressive and the Acheulian’s variability known in Africa and Eurasia (Gaillard universal cobble tools. et al., 2008; Singh, in press). The Soanian has been sub-divided Recent studies of the early Soanian artefacts in the western into different evolutionary stages according to the stratigraphical Siwaliks lead to the identification of these tool types, suggesting origin and state of preservation (Terra and Paterson, 1939; Paterson that the “early” Soanian in many cases if not all probably belongs to and Drummond, 1962). However, the stratigraphical context was this later phase of the Pleistocene. The abundance of cobbles as raw severely questioned by later field studies (Rendell et al., 1989) and material along the Himalayan rivers might have prompted prehis- only two Soanian stages are now referred to (Early and Late Soa- toric populations to use them with minimal modification, just as nian), apart from a very rare Final Soanian (Mohapatra, 1974). The cobble tools and flakes. As well, the particular landscape in the hilly late Soanian mostly including flakes and cores (sometimes well environment of the Siwaliks and its frequent modifications due to organised), along with cobble tools, is technologically related to the continuous tectonic activity might have been one of the reasons for Middle Palaeolithic (Teilhard de Chardin, 1937; Movius, 1948; the originality of this tradition when compared to the sharply Paterson and Drummond, 1962; Sankalia, 1974; Karir, 1985; different contemporaneous industries in peninsular India. Lycett, 2007). It is dated to between >57 ka and 20 ka in the Sirsa valley (Fig. 2; Suresh et al., 2002). 2. Western Siwaliks The Early Soanian is represented by numerous assemblages mainly composed of cobble tools (choppers and chopping tools), From the 1930s, the Western Siwaliks became well known for flakes and cores being occasional. This industry, like the Late Soa- their rich assemblages of prehistoric industries, especially in the nian, is made from the well silicified quartzite cobbles, usually of Potwar plateau and Soan Valley (Teilhard de Chardin, 1936; Terra dark colour, transported by the Himalayan rivers from the inner and Paterson, 1939). Two technical groups were identified: mountain ranges. It displays simple technical features and there- “Sohan/Soan industry” (Soanian) and Acheulian. The former was fore has for long been assigned to the Lower Palaeolithic like the

Fig. 2. Map of the sub-Himalayan sector between Beas and Ghaggar Rivers, focussing on the Janauri Anticline of the Siwalik Frontal Range and showing the sites where tools with transversal trimmed cutting edge occur (small circles: 1 tool; medium: 2e5 tools, large: 6e10 tools; very large circles: 11 tools or more).

Acheulian. It was then suggested that people bearing different Mishra, 2007; Gaillard and Dambricourt Malassé, 2008). On the “cultures” or technical traditions were alternatively “invading” the one hand this latter tradition is dated to between 0.7 and 0.4 Ma in region (Terra and Paterson, 1939; Movius, 1944, 1948; Sankalia, the Jhelum valley, at Dina and Jalapur (Pakistan; Rendell and 1974). However, the early Soanian appears to be later than the Dennell, 1985) and it is probably older than 0.6 Ma at Atbarapur Acheulian (Gaillard and Mishra, 2001;Chauhan, 2003, 2008; (Punjab, India; Gaillard et al., 2008, 2010a,b), on the other hand the

Please cite this article in press as: Gaillard, C., et al., Late Pleistocene to Early Holocene Lithic industries in the southern fringes of the Himalaya, Quaternary International (2010), doi:10.1016/j.quaint.2010.06.023 4 C. Gaillard et al. / Quaternary International xxx (2010) 1e11 abundant assemblages of cobble tools representing the early Soa- and the “axes/adzes” found by Corvinus (1987, 1989, 2007)in nian mainly occur on the 2nd and 3rd terraces of the Himalayan Eastern Nepal (see below). rivers, which formed after the post-Siwalik tectonic phase in the It is interesting to note that some of these tools show a polished second half of the Middle Pleistocene. Besides, field observations facet on one or two lateral side(s), perpendicular to the preferential show that the cobble tools are usually much better preserved than cutting edge (Fig. 6(1)). However, no gloss is observed except on the Acheulian cleavers or handaxes, and at places they occur in one of the rectangular tools, whose surface is almost entirely glossy. dense clusters in a very fresh condition suggesting they have been Similarly the flakes do not show any gloss on the edge, unlike those made on the spot and have not been disturbed afterwards. from the Nangal-Barmla sites on the left bank of the Sutlej, on the Recent reassessment of the early Soanian assemblages collected lower terrace dated to 6.2 0.8 BP (Soni et al., 2008; G.C. pers. between the Beas and Sutlej Rivers, in association with new field observation 2009) or from Amapur in Nepal (Corvinus, 2007). work by the authors (Singh et al., 1999, 2008), resulted in the Close scrutiny of the cobble tools also shows severe abrasion of identification of several specimens looking more standardised than some of the edges, comparable to small polish facets (Fig. 6(2)), the choppers; with their rectangular, square or sometimes trian- while some others have been heavily chipped as if they had been gular shape, they may recall the adze/axe-like tools of the early used for battering or pounding. Many choppers present this type of Hoabinhian of South-East Asia (Colani, 1936; Forestier, 2000; wear on one part of their edge only, while the other part looks Forestier et al., 2008; Zeitoun et al., 2008), or of the Sonvian (pre- unworn for a large flake scar having re-sharpened it. The corre- Hoabinhian) of North Vietnam (Hà Van Tan, 1997, 2001), which are sponding flakes do occur in the field at places: they are charac- attributed to the Late Pleistocene and Early Holocene. In the terised by frequent step-retouch on the striking platform’s dorsal Western Siwaliks these adze/axe-like tools, or at least tools with ridge. The cobble tools therefore appear as evolving tools, recur- a transversal trimmed cutting edge (Figs. 3e5), usually occur along rently re-sharpened by a large removal that affects one part of their with typical Soanian choppers and chopping tools, in different cutting edge only. In general the early Soanian collections actually localities of the Siwalik Frontal Range between Sutlej and Beas include all the stages of modification, from the cobbles only Rivers, a sector forming the Janauri Anticline. Their maximum showing a few removals to the intensively reduced cobbles. concentration is observed in the area which is facing the Sutlej’ deflection zone downstream of Nangal, i.e. when the Sutlej 3. Nepal encounters the anticline and turns from NE-SW to NW-SE, (Fig. 2). In this area the final uplift occurred quite late in the Pleistocene and During her 20 years work in Nepal, Gudrun Corvinus laid the even possibly in the Holocene, as the Janauri anticline results from foundations of the prehistory in the central sub-Himalayas, in close the junction of two segments that have been growing towards each relation with the Quaternary stratigraphy. Not only did she discover other (Delcaillau et al., 2006). As long as they were not connected, the first Acheulian elements (Corvinus, 1990, 1995) in this part of the Sutlej continued flowing southwest, at 90 to the tectonic the central Siwaliks (dun and frontal range), and excavate a rich structure (Malik and Mohanty, 2007). The top of the ridge is Middle Palaeolithic site belonging to MIS 5 or early MIS 4 (Corvinus, broader and flatter in this junction area, forming a sort of plateau, 1994, 2007), but also she found many cobble tool assemblages and wherein at places the landscape is littered with quartzite cobbles showed that they came from Late Pleistocene deposits. Moreover, at deposited by the ancient Sutlej. These were selected by the Patu site in Eastern Nepal she discovered a unique “macrolithic prehistoric craftsmen for making their tools, mostly choppers of Mesolithic” assemblage and compared it to the Hoabinhian of early Soanian type, but sometimes also thinner tools with one or South-East Asia, thus suggesting a technical link between the two transversal cutting edge(s). These assemblages are not dated, Siwaliks and the South-East Asia (Corvinus, 1987, 1989, 2007). but they are definitely later than the deposition of the cobbly Her contribution regarding the cobble tool industries is espe- gravels by the Sutlej, during a high energy (tectonic rather than cially interesting when related to the comparable assemblages climatic) phase. In the Sirsa-Soan dun (longitudinal valley along the found in the Western Siwaliks. In Nepal this techno-cultural phase north-eastern flank of the Janauri and Chandigarh Anticlines; Fig. 2) is abundantly represented in the dun as well as in the top of the the last aggradation phase ended around 20 ka (Kumar et al., 2007). river terraces and hill wash deposits (Fig. 1); given such a promi- It would have been the same in the Sutlej Valley, to where these nence, she named it the “Brakuti industry”. It is characterised by two rivers merge upstream the Siwalik crossing, and this date may various types of cobbles tools but the main component are flakes, correspond to the period when the river was blocked by the apparently resulting from the shaping of the cobble tools; a few emerging Janauri Anticline. It is difficult to know whether the adze/axe-like tools are also noted within these assemblages. This populations who were using the cobbles as raw material had type of industry is usually found in the upper part of the yellow settled in this sector at a time when the river was still flowing silts in the upper Babai formation, dating to 25e15 ka BP (MIS 2). “straight” through the Siwalik structure, or whether they came In the lower part of the same silts, dating to 43e34 ka BP, an later, probably for encampments of short duration only, in the assemblage characterised by end-choppers occurs at Sanpmarg on absence of a close source of water. a buried paleosol. Corvinus never compared these industries to the Within the assemblages dominated by choppers, these tools early Soanian assemblages from the Western Siwaliks, probably with transversal trimmed cutting edge look significantly more for the latter being known as Lower Palaeolithic largely dominated refined. Most are made on flakes (nearly 40%) and otherwise on flat by cobble tools and practically devoid of flakes. Actually sites cobbles or on split cobbles. They are trimmed on 2e4 of their edges yielding large quantities of flakes are extremely rare in the Janauri by unifacial or bifacial flaking, the transversal edge being generally Anticline, but they do exist as shown by recent field work bifacially trimmed. The shaping never hides the original shape of (Dambricourt Malassé, 2008; Singh et al., 2008). Salient features in the blank support, whose selection or initial production (by flaking both the Brakuti industry and the so called “early” Soanian from or splitting) therefore strongly determines the final tool the Janauri Anticline are the “rejuvenation flakes” as Corvinus also morphology. The contour is mostly trapezoidal (35%; Fig. 3(1) and interpreted them. These flakes bear a characteristic “step-flaking (3)). Square, rectangular (Figs. 3(2) and 4(2)) or pentagonal (Fig. 4 or step-retouch at their platform edges on the dorsal face” (1)) tools are also common. The only 3 triangular tools (Fig. 5(1)), (Corvinus, 2007: 93). along with some of the trapezoidal ones, clearly resemble the In Eastern Nepal, the site of Patu (Fig. 1) covers an area of “primitive axe/adze” identified by Colani (1929) in North Vietnam 2 3 km on the higher terrace along the Rato Khola. It has yielded

Fig. 3. Tools with transversal trimmed cutting edge (adze/axe-like tools) from the Janauri anticline, Western Siwalik Frontal Range (drawings).

Fig. 4. Tools with transversal trimmed cutting edge (adze/axe-like tools; shorter shapes) from the Janauri anticline, Western Siwalik Frontal Range. a rich (2590 artefacts) “macrolithic Mesolithic” industry mainly sites the lithic assemblages are mainly composed of cobble tools, composed of flakes (80%) along with unifacial and bifacial “adzes”, core tools and flakes. Some triangular “primitive axes” (Colani, choppers and core-scrapers. It also includes 6 “sumatraliths”, which 1927, 1929) and rectangular tools are also reported from North are typical tools of the Hoabinhian. These artefacts are imbedded in Vietnam (especially from the Xom Trai cave, near HoaBinh; Hà Van a red soil developed in the upper part of the alluvial silt capping the Tan, 1997) and from Northern Thailand (Zeitoun et al., 2008). The terrace. Originally they might have been abandoned on the surface various studies on these lithic assemblages (reviewed by Marwick, of the silt and later buried in it by bioturbation. A charcoal found in 2007) usually conclude that they were mainly used for working the same horizon as the artefacts provided a date of 7 ka BP and wood and various vegetal matters. therefore a minimum age for the industry. Nearly half of the 83 In Northern Vietnam, the Xom Trai cave providing a typical “adzes” show a very distinct gloss on their transversal edge. This Hoabinhian industry is dated by 14C to between 18 and 17 ka BP (Hà assemblage made from quartzite cobbles and boulders by splitting Van Tan, 1997). In Huang Cho rockshelter, the first occupation and flaking is interpreted as mainly oriented towards wood or occurred around 30 ka BP (Yi et al., 2008), but the first stages may bamboo working (Corvinus, 1987, 1989, 2007). include only cobble tools and flakes without the typical large unifacial tools known as “sumatraliths”. In several Vietnamese sites the 4. South-east Asia industries preceding the Hoabinhian are identified as Sonvian (Hà Van Tan, 1997). In Thailand, especially in the rich province of Kan- The Hoabinhian was identified by Madeleine Colani in the late chanaburi, the Hoabinhian covers the time range of about 32e7ka 1920s on the basis of lithic assemblages from the region of HoaBinh, BP (Santoni et al., 1990; Shoocongdej, 2000). In the north, the lower east of Hanoi in North Vietnam (Fig. 1; Colani, 1927, 1929). It has levels in Tham Lod rockshelter are dated TL to 35 ka (Shoocongdej, been recognised in most parts of Mainland South-east Asia to South 2006) but it is not clear whether Hoabinhian tools are from these China, Myanmar, and Malaysia (Zuraina Majid, 1990) and in lower levels. Similarly in South Thailand, the Hoabinhian starts Sumatra (Forestier et al., 2005), where the “sumatraliths” had first around 35 ka BP at Lang Rongrien (Anderson, 1990), while at Moh been described. These large oval tools, unifacially shaped are the Khiew, the lower level yielding a local facies of the Hoabinhian most typical stone artefact of this “industry” rather than “techno- (Moser, 2001) is dated to 25 ka BP (Pookajorn, 2001). However, complex” or “material culture” (Bowdler, 2008) but in most of the unlike in Vietnam, it is not sure in South Thailand and Malaysia that

Fig. 5. Tools with transversal trimmed cutting edge (adze/axe-like tools; triangular and elongated shapes) from the Janauri anticline, Western Siwalik Frontal Range. there is continuity between the earlier technical stages and the forested environments and the stone artefacts are associated with typical Hoabinhian (Bellwood, 1997). large quantities of fresh water or sea shells (often boiled or burned Hoabinhian settlements are found both in caves or shelters and at Moh Khiew; Pookajorn, 2001) and bones from small to medium in the open. The formers provide good information regarding the sized animals (Shoocongdej, 2000; Auetrakulvit, 2004). Some sites, environment and subsistence behaviour of people who made and as in Vietnam, also yield abundant nuts, provided by an environ- used such tool kits. These people were foragers and moving ment that was forested even during the colder but not dry Last according to the seasons; they probably had different strategies Glacial Maximum (Viet, 2008). Open-air sites are especially char- during the wet and dry seasons. Their sites are usually located in acterised by huge accumulations of shells. A large number of such

Fig. 6. Polished marks on (1) the cortical side of a tool with transversal trimmed cutting edge (adze/axe-like tool) and (2) both ends of a chopper edge.

shell middens occur along the north-east coast of Sumatra and in then the entire last glacial cycle as well as the subsequent warming the Malay Peninsula. They attest a human occupation from 12 to and cooling phases of the first part of the Holocene. During this 3ka(McKinnon, 1990; Forestier, 2007). More than the cave sites, period the stone tool kit did not change much, apart from an they indicate an intensive exploitation of the marine or river food increasing number of tools with a transversal trimmed cutting resources, but game from the forest is also found. edge. Wood and especially bamboo were probably used as main raw In the Nepali sub-Himalayas, these industries date back to 25 ka materials, as they are nowadays, for all the activities of the day to BP. They are even earlier (34e43 ka BP) at Sanpmarg, where the day life (Pope, 1989; Forestier, 2003). This might have given its artefacts are lying on a red soil indicating milder conditions specificity to the “Hoabinhian”, which has to be better defined, not (Corvinus, 2007). They probably last till the beginning of the only on technological terms but also in relation to the chrono- Holocene, as observed at Patu site, where they are also associated to climatic variations, subsistence strategies and territories (Zeitoun a red soil. In the Janauri Anticline of Western Siwaliks, between et al., 2008). Beas and Sutlej, they may date back to 20 ka or older, and in Northern South-east Asia, the Hoabinhian (and Sonvian in Viet- 5. Discussion nam) cover the same time period. In the North-West, comparable assemblages have been observed in the aceramic Neolithic of Sri- All the industries presented here, mainly composed of unre- nagar valley (Pant et al., 1982) or in the upper Yarkhun Valley of touched flakes, cobble tools and a few tools with transversal Northern Pakistan (Gaillard et al., 2002; Dambricourt Malassé and trimmed cutting edge (adze/axe-like tools), occur in sites located in Gaillard, this issue). These industries did not change much till the the sub-Himalayan belt (between the Main Boundary Thrust and mid-Holocene, despite the global climatic alternation of cold and the Himalayan Frontal Fault) and its eastern continuation in warm phases. northern South-east Asia. They belong to a geographic area occu- However, the sub-Himalayan belt provides specific conditions pied, nowadays, by the monsoon forest (clearly visible on satellite during the climate changes, which may be favourable to continuous views such as Fig. 1), of which bamboo is a typical component. This (or nearly continuous) human occupation. The strong gradient of forest grades into the tropical rain forest towards the south-east the mountain slopes offers refuges for the vegetation, which and to the mountain forest on the Himalayan slopes; it is bordered remains quite stable despite the climatic changes, since slight by the Alpine and savannah biomes and may be replaced by them upwards or downwards shifts allows the biomes finding again their according to the climate fluctuations. specific climatic conditions. This is observed for instance on the It is difficult to reconstruct the exact landscape and environment slopes of Tibetan Plateau in Eastern China (Qiu et al., 2009). associated to these particular lithic industries, all the more as they Moreover the last glacial maximum (LGM) was not as severe as in encompass several climatic phases including the end of the MIS 3, most parts of the world. In the Himalayas, the maximal extension of

Please cite this article in press as: Gaillard, C., et al., Late Pleistocene to Early Holocene Lithic industries in the southern fringes of the Himalaya, Quaternary International (2010), doi:10.1016/j.quaint.2010.06.023 C. Gaillard et al. / Quaternary International xxx (2010) 1e11 9 the glaciers occurred during MIS 3, earlier than the global blade and bladelet technology is widely applied in the same time maximum (Owen et al., 2002). In the Kumaun Himalaya and period, confirm that this vast plain was a barrier, or at least a border Ladakh, for instance, the lake deposits have recorded detailed separating two types of behaviours. As the Movius line is slowly climatic alternations. Many lakes in this sector have formed as vanishing from the concepts regarding the Asian Lower Palae- a consequence of a regional tectonic event that happened around olithic, this new line, partly superimposed to the former one, 40 ka. At that time the climate was rather dry and cold; then it appears as clearly related to the late Pleistocene and early Holocene became more humid between 35 and 30 ka (Kotlia et al., 1997, environment. 2000). This latter phase may be in relation with the paleosol at Sanpmarg in Nepal with cobble tool industry lying on it. After a short alternation (cold and dry, then warm and humid), the major References glacial phase is recorded between 29 and 21.7 ka cal. BP in Kumaun (Kotlia et al., 2000) or between 28 and 21 ka BP in Ladakh Agrawal, D.P., 1982. The archaeology of India. Scandinavian Institute of Asian Studies Monograph Series 46, 294. (Bhattacharya, 1989), while only a short cold event is observed Anderson, D., 1990. Lang Rongrien Rockshelter: A Pleistocene, Early Holocene around 18 ka BP (Sharma, 1995), at the time of the global LGM. In Archaeological Site from Krabi, Southwestern Thailand. The University North Thailand the late Pleistocene vegetation is forested with pine Museum, University of Pennsylvania, Philadelphia. Andrefsky Jr., W., 2007. Cobble tool or cobble core: exploring alternative typologies. and oak as characteristic taxa (White et al., 2004). In North Vietnam In: McPherron, S.P. (Ed.), Tools Versus Cores: Alternative Approaches to Stone charred plant remains and especially walnut trees indicate that Tool Analysis. Cambridge Scholars Publishing, Newcastle, p. 280. during the LGM the conditions were rather cool but not dry (Viet, Auetrakulvit, P., 2004. Faunes du Pléistocène final à l’Holocène de Thaïlande: 14 approche archéozoologique. Université de Provence, Faculté des lettres et 2008). Therefore, the global LGM (18 2 C ka BP, corresponding sciences humaines, unpublished Ph.D. Thesis, 433 p. to 21 2 cal ka BP) was not the harshest period of the late Pleis- Bellwood, P., 1997. Prehistory of Indo-Malaysian Archipelago. University of Hawaii tocene in the Himalayas: due to dryness and snowfall decrease, the Press, 384 p. Bhattacharya, A., 1989. Vegetation and climate during the last 30,000 years in ice cover could not spread. The glaciers did not extend beyond Ladakh. Palaeogeography, Palaeoclimatology Palaeoecology 73, 25e38. 10 km of the present day ice margin (Owen et al., 2002) and among Bowdler S., 2008. Hoabinhian and non-Hoabinhian. In: Pautreau, J.P., Coupey A.S., all the tropical and sub-tropical glaciers of the world, the Hima- Zeitoun V., Rambault E. (Eds.), From Homo erectus to the living traditions. layan glaciers show the smallest equilibrium-line altitude drop Papers from the 11th International Conference of the EurASEAA, Siam Ratana Ltd. Part, Chiang Mai, pp. 59e66. (100 m) from the present day altitude (Mark et al., 2005). Moreover, Chauhan, P., 2003. An Overview of the Siwalik Acheulian and Reconsidering Its it appears that the climate variations are less abrupt in the Asian Chronological Relationship with the Soanian e A Theoretical Perspective. monsoon zone (Shakun et al., 2007). In the Andaman Sea the Assemblage 7. Available from: http://www.shef.ac.uk/assem. Chauhan, P.R., 2007. Soanian cores and core-tools from Toka, northern India: temperature during the LGM is estimated to be about 3 C less than towards a new typo-technological organization. Journal of Anthropological during the late Holocene (Rashid et al., 2007). Archaeology 26 (3), 412e441. The relative stability of the forested environment and the Chauhan, P.R., 2008. Soanian lithic occurrences and raw material exploitation in the Siwalik Frontal Zone, northern India: a geoarchaeological perspective. Journal abundance of perennial rivers and other water sources such as of Human Evolution 54, 591e614. lakes, possibly linked to the constant tectonic activity, have Colani, M., 1927. L’âge de la Pierre dans la province de Hoa-Binh (Tonkin). Mémoires certainly favoured a particular way of life that remained more or du service géologique de l’Indochine 14 (1), 95. Colani, M., 1929. Quelques paléolithes hoabiniens typiques de l’abri sous roche de less the same throughout the late Pleistocene and early Holocene, Lang-Kay. Bulletin de la Société Préhistorique Française 6, 354e384. until the Neolithic. The use of bamboo, wood and various types of Colani, M., 1936. Recherches sur le préhistorique indochinois. Extraits du Bulletin de vegetal raw materials may have induced the particular aspect of the l’École Français d’Extrême Orient 30 (3e4), 126. Corvinus, G., 1987. Patu, a new stone age site of a jungle habitat in Nepal. Quartär lithic tool kit; some other tasks may also be involved. For example 37/38, 135e187. the cobble tool and flake industry from Votaw site on the Snake Corvinus, G., 1989. The Patu industry in its environment in the Siwaliks in eastern River (Washington State, USA) is supposed to be associated with Nepal. Quartär 39/40, 95e123. fi fishing activities (Andrefsky, 2007). The abundance of shellfish in Corvinus, G., 1990. A note on the rst discovery of handaxes in Nepal. Man and Environment 15 (2), 9e11. the Hoabinhian sites supports the idea of a subsistence strategy Corvinus, G., 1994. Prehistoric occupation sites in the DangeDeokhuri Valleys of closely linked to the water foods. Why not apply this pattern to the western Nepal. Man and Environment 19, 73e89. Western Siwaliks, as the Soanian sites are mostly settled along the Corvinus, G., 1995. The Satpati handaxe site and the Chabeni uniface site in Southern Nepal. (A handaxe site in tectonically folded alluvial sandstones at the rivers? To date, the cobble tool assemblages have not been shown frontal Himalayan thrust zone). Quartär 45/46, 15e36. to be associated with shells; the absence of faunal remains in these Corvinus, G., 2004. Homo erectus in East and Southeast Asia, and the questions of open-air sites prevents any approach of the subsistence pattern the age of the species and its association with stone artifacts, with special e fi fi attention to handaxe-like tools. Quaternary International 117, 141 151. (large or small game, shell sh, sh, etc.) and strategies (hunting, Corvinus, G., 2007. Prehistoric cultures in Nepal. Harrassowitz Verlag Wiesbaden 1 trapping, vegetal food collection, etc.). There is no possibility either and 2, pp. 383, 263. to know how long the settlements lasted at each place, whether Dambricourt Malassé, A., 2008. Le peuplement humain en Eurasie: l’Asie centrale montagneuse et les piémonts sous-himalayens du Plio-Pléistocène à l’Holo- they were seasonally recurrent, and how wide were the territories cène, origines, évolution humaine et migrations. L’Anthropologie 112 (3), of these foraging populations. The continuity of the forested envi- 370e403. ronment all along the sub-Himalayan belt and its south-eastern Dambricourt Malassé, A., Gaillard, C., this issue. Relations between climatic changes and prehistoric human migrations during Holocene between Gissar Range, extension was certainly favourable to population movements, Pamir, Hindu-Kush and Kashmir: the archaeological and ecological data. especially as the upland regions (above 400 m) have been shown to Quaternary International. be particularly rich in resources during the entire Palaeolithic, Delcaillau, B., Carozza, J.M., Laville, E., 2006. Recent fold growth and drainage providing opportunities to people to migrate eastwards (Schepartz development: the Janauri and Chandigarh anticlines in the Siwalik foothills, northwest India. Geomorphology 76, 241e256. et al., 2000) and probably westwards too. This “latitudinal” circu- Dennell, R.W., 2004. Early Hominin landscapes in the Pabbi Hills, Northern Pakistan. lation, or rather circulation along the geological structures in British Archaeological Reports International Series 1265, 454. following similar geomorphological features, might have been Dennell, R.W., 2007. Resource-rich, stone poor: early hominin land use in large river systems of Northern India and Pakistan. In: Petraglia, M., Allchin, B. (Eds.), The forced by the marked contrast between the hilly forested envi- Evolution and History of Human Populations in South Asia. Springer, ronment, rich in quartzitic or other siliceous raw material (as far as Netherlands, pp. 41e68. stone is concerned), and the indo-gangetic plain lacking such Dennell, R., 2009. The Palaeolithic Settlement of Asia. Cambridge World Archae- ology, Cambridge University Press. material (Dennell, 2007) and therefore implying other behaviours. Dennell, R.W., Rendell, H., Hailwood, E., 1988. Early tool-making in Asia: two- The sharply different lithic industries in peninsular India, where million-year-old artefacts in Pakistan. Antiquity 62, 98e106.

Forestier, H., 2000. Chaînes opératoires lithiques en Asie du Sud-Est au Pléistocène Mishra, S., 2007. The Indian Lower Palaeolithic. Bulletin of the Deccan College Post- supérieur final et au début de l’Holocène. L’Anthropologie 104, 531e548. graduate and Research Institute 66e67, 47e94. Forestier, H., 2003. Des outils nés de la forêt. L’importance du végétal en Asie du Mohapatra, G.C., 1966. Preliminary report of the exploration an excavation of the Sud-Est dans l’imagination et l’invention technique aux périodes pré- stone age sites in eastern Punjab. Bulletin of the Deccan College Research historiques. In: Froment, A., Guffroy, J. (Eds.), Actes du colloque: Peuplements Institute 25, 221e237. anciens et actuels des forêts tropicales, Ermes-IRD, Coll. Colloques et sémin- Mohapatra, G.C., 1974. Lithic industries of Himachal Pradesh. In: Ghosh, A.K. (Ed.), aires, pp. 315e337. Perspective in Palaeoanthropology. Firma K.L, Mukhopadyay, Calcutta, pp. Forestier, H., 2007. Les éclats du passé préhistorique de Sumatra: une longue his- 199e212. toire des techniques. Archipel 74, 15e44. Mohapatra, G.C., Singh, M., 1979a. Stratified occurrence of lithic artifacts in the Forestier, H., Simanjuntak, H.T., Guillaud, D., Driwantoro, D., WiradnyanaI, K., Siwalik frontal range of western sub-Himalaya. Panjab University Research Siregar, D., Due Awe, R., Budiman, 2005. Le site de Tögi Ndrawa, île de Nias, Bulletin (Arts) 10 (1e2), 65e77. Sumatra nord: les premières traces d’une occupation préhistorique Hoa- Mohapatra, G.C., Singh, M., 1979b. Prehistoric investigations in a sub-Himalayan binhienne en grotte en Indonésie. Compte Rendu Palevol (Académie des Valley, Himachal Pradesh, India. Current Anthropology 20, 600e602. Sciences, Paris) 4, 727e733. Moser, J., 2001. Hoabinhian, geographie und chronologie eines steinzeitlichen Forestier, H., Zeitoun, V., Maitay, C., Winayalai, C., 2008. The Hoabinhian site of Huai technocomplexes in Südostasien. AVA-Forshungen 6, 256. Hin (Northern Thailand). In: Pautreau, J.P., Coupey A.S., Zeitoun V., Rambault E. Movius, H.L., 1944. Early man and Pleistocene stratigraphy in Southern and Eastern (Eds.), From Homo Erectus to the Living Traditions. Papers from the 11th Asia. Papers of the Peabody Museum 19 (3), 125. International Conference of the European Association of Southeast Asian Movius, H.L., 1948. The Lower Palaeolithic cultures of Southern and Eastern Asia. Archaeologist, Bougon 2006, Siam Ratana Ltd. Part, Chiang Mai, pp. 67e72. Transactions of the American Philosophical Society 38 (4), 329e420. Forestier, H., Zeitoun, V., Tiamtinki, C., Boonyai, S., Winayalai, C., Auetrakulvit, P., Norton, C.J., Bae, K., Harris, J.W.K., Lee, H., 2006. Middle Pleistocene handaxes from 2006. Lames du Neolithique dans les carrières de Nan (Nord de la Thaïlande). the Korean Peninsula. Journal of Human Evolution 51, 527e536. Aséanie 18, 47e81. Norton, C.J., Bae, K., 2008. The Movius Line sensu lato (Norton et al., 2006) further Gaillard, C., Dambricourt Malassé, A., 2008. Les principales étapes de l’occupation assessed and defined. Journal of Human Evolution 55, 1148e1150. humaine en bordure de l’Himalaya occidentale, vol. 112. L’Anthropologie, Paris, Owen, L.A., Finkel, R.C., Caffee, M.W., 2002. A note on the extent of glaciation pp. 404e422. throughout the Himalaya during the global Last Glacial Maximum. Quaternary Gaillard, C., Dambricourt-Malassé, A., Magraner, J., Maîtrerobert, A., Taj, Ali, Science Reviews 21 (1e3), 147e157. Voisin, J.-L., Abdul, Nasir, 2002. Discovery of recent lithic industries with archaic Pant, R.K., Gaillard, C., Nautiyal, V., Gaur, G.S., Shali, S.L., 1982. Some new lithic and features in the Hindu Kusch range (Chitral district, North Pakistan). Bulletin of ceramic industries from Kashmir. Man and Environment 6, 37e40. the Indo-Pacific Prehistory Association 22, 25e33. Paterson, T.T., Drummond, H.J.H., 1962. Soan the Palaeolithic of Pakistan, vol. 2. Gaillard, C., Mishra, S., 2001. The Lower Palaeolithic in South Asia. In: Sémah, F., Memoir of the Department of Archaeology, Pakistan, Karachi. Falguères, Ch., Grimaud-Hervé, D., Sémah, A.-M. (Eds.), Origine des peuple- Petraglia, M.D., Shipton, C., 2008. Large cutting tool variation west and east of the ments et chronologie des cultures paléolithiques dans le Sud-Est Asiatique. Movius Line. Journal of Human Evolution 55, 962e966. Semenanjung-Artcom, Paris, pp. 73e91. Pookajorn, S., 2001. New perspectives for Palaeolithic research in Thailand. In: Gaillard, C., Mishra, S., Singh, M., Deo, S., Abbas, R., 2010a. Lower and early Middle Sémah, F., Falguères, C., Grimaud-Hervé, D., Sémah, A.-M. (Eds.), Origine des Pleistocene Acheulian in the Indian sub-continent. Quaternary International. peuplements et chronologie des cultures paléolithiques dans le Sud-Est Asiat- doi:10.1016/j.quaint.2009.08.021. ique. Semenanjung-Artcom, Paris, pp. 167e187. Gaillard, C., Mishra, S., Singh, M., Deo, S., Abbas, R., 2010b. Reply to Chauhan 2010: Pope, G.G., 1989. Bambou and human evolution. Natural History 10, 49e57. comments on lower and early Middle Pleistocene Acheulian in the Indian sub- Qiu, Y.X., Guan, B.C., Fu, C.X., Comes, H.P., 2009. Did glacials and/or interglacials continent. Quaternary International. doi:10.1016/j.quaint.2010.01.022. promote allopatric incipient speciation in East Asian temperate plants? Phy- Gaillard, C., Singh, M., Rishi, K.K., 2008. Technological analysis of the Acheulian logeographic and coalescent analyses on refugial isolation and divergence in assemblage from Atbarapur in the Siwalik Range (Hoshiarpur District, Punjab). Dysosma versipellis. Molecular Phylogenetics and Evolution 51, 281e293. Man and Environment 33 (2), 1e14. Rashid, H., Flower, B.P., Poore, R.Z., Quinn, T.M., 2007. A w25 ka Indian Ocean Hà Van Tan, 1997. The Hoabinhian and before. Bulletin of the Indo-Pacific Prehistory monsoon variability record from the Andaman Sea. Quaternary Science Review Association 16, 35e41. 26, 2586e2597. Hà Van Tan, 2001. Perspectives de recherche sur le Paléolithique au Viêt Nam. In: Rendell, H., Dennell, R.W., 1985. Dated Lower Palaeolithic artefacts from northern Sémah, F., Falguères, C., Grimaud-Hervé, D., Sémah, A.-M. (Eds.), Origine des Pakistan. Current Anthropology 26 (3), 393. peuplements et chronologie des cultures paléolithiques dans le Sud-Est Asiat- Rendell, H.M., Dennell, R.W., Halim, M.A., 1989. Pleistocene and Palaeolithic ique. Semenanjung-Artcom, Paris, pp. 139e148. investigations in the Soan Valley, Northern Pakistan. In: British Archaeological I.A.R., 1961e1962. Indian Archaeology e A Review. Archaeological Survey of India, Reports International Series, vol. 544. Oxford, 346 p. New Delhi. Sankalia, H.D., 1974. Prehistory and Protohistory of India and Pakistan. Deccan I.A.R., 1964e1965. Indian Archaeology e A Review. Archaeological Survey of India, College Postgraduate and Research Institute, Poona. 529. New Delhi. Santoni, M., Pautreau, J.P., Prischanschit, S., 1990. Excavations at Obluang, Province Kadereit, A., Wagner, G.A., Corvinus, G., 2007. OSL/IRSL fine-grain dating. A prelimi- of Chiang Mai, Thailand. In: Glover, I., Glover, E. (Eds.), Southeast Asian nary chronology of Quaternary slopewash and alluvial deposits from the Dun Archaeology 1986. Proceedings of the first conference of the Association of Valleys of Dang and Deokhuri in western Nepal. Appendix II. In: Corvinus, G. (Ed.), Southeast Asian Archaeologists. British Archaeological Research International Prehistoric Cultures in Nepal. Harrassowitz Verlag, Wiesbaden, pp. 372e379. series, 561, Oxford, Archaeopress, pp. 37e54. Karir, B.S., 1985. Geomorphology and Stone Age Culture of North-Western India. Schepartz, L.A., Miller-Antonio, S., Bakken, D.A., 2000. Upland resources and the Sundeep Prakashan, Delhi, 156 p. early Palaeolithic occupation of Southern China, Vietnam, Laos, Thailand and Keates, S.G., 2002. The movius line: fact or fiction? Bulletin of the Indo-Pacific Burma. World Archaeology 32 (1), 1e13. Prehistory Association 22, 17e24. Shakun, J.D., Burns, S.J., Fleitmann, D., Kramers, J., Matter, A., Al-Subary, A., 2007. A Kotlia, B.S., Bhalla, M.S., Sharma, C., Rajagopalan, G., Ramesh, R., Chauhan, M.S., high-resolution, absolute-dated deglacial speleothem record of Indian Ocean Mathur, P.D., Bhandari, S., Chacko, S.T., 1997. Palaeoclimatic conditions in the climate from Socotra Island, Yemen. Earth and Planetary Science Letters 259, upper Pleistocene and Holocene Bhimtal-Naukuchiatal lake basin in south- 442e456. central Kumaun, North India. Palaeogeography, Palaeoclimatology, Palae- Sharma, K.K., 1995. Quaternary stratigraphy and prehistory of the Upper Indus oecology 130, 307e322. Valley, Ladakh. In: Wadia, S., Korisettar, R., Kale, V.S. (Eds.), Quaternary Envi- Kotlia, B.S., Sharma, C., Bhalla, M.S., Rajagopalan, G., Subrahmanyam, K., ronments and Geoarchaeology of India. Geological Society of India, Bangalore, Bhattacharyya, A., Valdiya, K.S., 2000. Palaeoclimatic conditions in the late pp. 98e108. Pleistocene Wadda Lake, eastern Kumaun Himalaya (India). Palaeogeography, Shoocongdej, R., 2000. Forager mobility organization in seasonal tropical environ- Palaeoclimatology, Palaeoecology 162, 105e118. ments of western Thailand. World Archaeology 32, 1. Archaeology in Southeast Kumar, R., Suresh, N., Sangode, S.J., 2007. Differential features of alluvial fans in the Asia, 14e40. Pinjaur-Soan Dun, NW Himalaya, India: controlling factors. Himalayan Geology Shoocongdej, R., 2006. Late Pleistocene activities at the Tham Lod rockshelter in 28 (1), 37e46. highland Pang Mapha, Mae Hong Son province, northwestern Thailand. In: Lycett, S.J., 2007. Is the Soanian techno-complex a Mode 1 or Mode 3 phenomenon? Bacus, E., Glover, I., Pigott, V. (Eds.), Uncovering Southeast Asia’s Past. Singapore A morphometric assessment. Journal of Archaeological Science 34, 1434e1440. University Press, pp. 22e37. Malik, J.N., Mohanty, C., 2007. Active tectonic influence on the evolution of drainage Singh, M., in press. Acheulian in the Siwaliks of North-Western India. Proceedings of and landscape: geomorphic signatures from frontal and hinterland areas along the international conference on Les cultures à bifaces du Pléistocène inférieur et the Northwestern Himalaya, India. Journal of Asian Earth Sciences 29, 604e618. moyen dans le monde. Émergence du sens de l’harmonie, Tautavel 25e30 Mark, B.G., Harrison, S.P., Spessa, A., New, M., Evans, D.J.A., Helmens, K.F., 2005. juillet 2007. Tropical snowline changes at the last glacial maximum: a global assessment. Singh, M., Chopra, S.N., Karir, B.S., 2008. Prehistoric Investigations in the Siwalik Quaternary International 138e139, 168e201. Frontal Ranges of the Punjab (Part II: Between Jaijon-Doaba and Roopnagar). Marwick, B., 2007. Approaches to flaked stone artefact archaeology in Thailand: Society for Archaeological and Anthropological Research, and Department of a historical review. Silpakorn University International Journal 7, 49e88. Cultural Affairs, Archaeology and Museum, Government of Punjab, Chandigarh, McKinnon, E.E., 1990. The Hoabinhian in the Wampu/Lau Biang valley of North- 218 p. (unpublished report). eastern Sumatra: an update. Bulletin of the Indo-Pacific Prehistory Association Singh, M., Karir, B.S., Chopra, S.N., 1999. Prehistoric and Quaternary Studies of the 10, 132e142. Roop Nagar District of Punjab, Research Project Sponsored by the Government

of Punjab. Department of Cultural Affairs, Archaeology & Museums, Chandigarh, Terra, H. de, Paterson, T.T., 1939. Studies on the Ice Age in India and Associated 265 p. (unpublished report). Human Cultures, vol. 493. Carnegie Institution of Washington Publications, Soni, A.S., Soni, V.S., Dhillon, D.S., 2008. Large assemblages of flakes and cores found Washington, 350 p. on dated young terraces of River Satluj and its tributaries. Current Science 94 Viet, N., 2008. Hoabinhian macrobotanical remains from archaeological sites in (5), 577e580. Vietnam: indicators of climate changes from the Late Pleistocene to the Stakul, G., 1981. Cultural change in the Swat Valley and beyond, c. 3000e1400 BC. Early Holocene. Bulletin of the Indo-Pacific Prehistory Association 28, In: Allchin, B. (Ed.), South Asian Archaelogy 1981, Proceedings of the Sixth 80e83. International Conference of the Association of South Asian Archaeologists in White, J.C., Penny, D., Kealhofer, L., Maloney, B., 2004. Vegetation changes from the Western Europe, Cambridge University 5e10 July 1981. Cambridge University late Pleistocene through the Holocene from three areas of archaeological Press, pp. 205e212. significance in Thailand. Quaternary International 113, 111e132. Suresh, N., Bagati, T.N., Thakur, V.C., Kumar, R., Sangode, S.J., 2002. Optically stim- Yi, S., Lee, J.J., Kim, S., Yoo, Y., Kim, D., 2008. New data on the Hoabinhian: inves- ulated luminescence dating of alluvial fan deposits of Pinjaur Dun, NW Sub tigations at Hang Cho cave, Northern Vietnam. Bulletin of the Indo-Pacific Himalaya. Current Science 82, 1267e1274. Prehistory Association 28, 73e79. Teilhard de Chardin, P., 1936. Joint geological and prehistoric studies in the late Zeitoun, V., Forestier, H., Nakbunlung, S., 2008. Préhistoire au sud du Triangle d’Or. Cenozoic in India. Science 83 (2149), 233e236. IRD Éditions, 252 p. Teilhard de Chardin, P., 1937. Notes sur la Paléontologie humaine en Asie Mér- Zuraina Majid, 1990. The Tampanian Problem Resolved. Modern Quaternary idionale. L’Anthropologie (Paris) 47, 23e33. Research in South-East Asia 11, 71e96.