Palaeolithic research at the Institute of Archaeology

conceptualization and planning, suggesting that a broader view of the evo­ lutionary foundations of modern human technological competence may be needed. This research highlights the fact that stone toolmaking is a concrete visual-motor skill rather than an abstract conceptual capac­ ity, and that such skill may be developed only through deliberate effortful practice. What might this imply about the social contexts that encouraged and supported such investments in skill acquisition in prehistory? Ethnographic and experimental research with modern traditional stone toolmakers can provide a valuable reference point to address this question. Although Oldowan 14 Figure 1 Positron emission tomography (PET) scans of brain activation during experimental Oldowan toolmaking. Successful flaking requires the precise delivery of powerful blows, as well as the identification of appropriate targets based on visual assessment of the core. These demands are reflected in the activation of motor planning and execution regions in the posterior frontal lobe, seen in the mid­ dle of the brain in the central (right lateral) and lower (dorsal) images, and of higher­ order visual cortex in the occipital and posterior parietal regions, as can be seen towards the back of the brain in the upper

Human technology and cognitive evolution (Dietrich Stout) Human beings are technological animals. Although many species make and use tools, Homo sapiens is distinguished by obligatory reliance on complex technolog ical modes of adaptation. How did we become this way, and what does it indicate about the nature of human intelligence? These are complex questions that require a similarly complex approach, crossing tra ditional disciplinary boundaries between archaeology, anthropology, neuroscience and psychology.
The earliest known technological arte facts in the world -simple stone flakes struck from river cobbles -come from the site ofGona in the Afar region of Ethiopia.5 When such early Oldowan stone tools were first described from Olduvai Gorge in Tanzania, they were thought to indicate the emergence of distinctly human cul tural and cognitive capacities involving "the imposition of arbitrary form on the environment". 6 Experimental replication subsequently revealed that these typolog ical forms are more likely to have been imposed by the interpretations of modern archaeologists than by the intentions of ancient toolmakers/ leading some to argue that the Oldowan provides evidence of nothing more than ape-level cognition.8 However, times are again changing as an increasing body of archaeological evidence reveals the technical skill and sophisti cation of the earliest toolmakers.9 At Gona, this includes both mastery of flaking tech niques and a pronounced degree of selec tivity for high-quality raw materials. 10 But what do these early indications of technological ability actually reveal about human cognitive and brain evolution? Modern imaging techniques of functional brains provide an exciting new oppor tunity to plumb the neural foundations of early technology. Positron emission tomography11 applied to human subjects immediately after they have engaged in experimental manufacture of Oldowan type stone tools have revealed the impor tance of complex perceptual-motor and spatial integration (Fig. 1), as opposed to abstract conceptualization and planning, suggesting that a broader view of the evo lutionary foundations of modern human technological competence may be needed. This research highlights the fact that stone toolmaking is a concrete visual-motor skill rather than an abstract conceptual capac ity, and that such skill may be developed only through deliberate effortful practice. What might this imply about the social contexts that encouraged and supported such investments in skill acquisition in prehistory?
Ethnographic and experimental research with modern traditional stone toolmakers can provide a valuable reference point to address this question. Although Oldowan knapping may have been mastered in a relatively short time, more refined tools from the later Acheulean (more than 500,000 years ago) and Middle Palaeo lithic (more than 250,000 years ago) pro vide evidence that hundreds of hours or more were probably invested in practising their manufacture. A particularly extreme modern example of heavy investment in

ARCHAEOLOGY INTERNATIONAL
the acquisition of toolmaking skills comes from the stone-adze makers of Langda village in the Highlands of New GuineaY These remarkable artisans spend as long as ten years mastering their craft, supported as much by aspects of identity, prestige and social motivation as by the explicit demonstration, instruction and facilita tion they receive as apprentices. fossiliferous chambers of a limestone karstic cave system. 16 Excavations in the 1960s exposed c. 13-14m of deposits from the Postglacial to the Lower Palaeolithic, with clear evidence of human occupation throughout much of that period, including part of a hominin mandible described as pre-Neanderthal. 17 Unfortunately, these excavations were not well recorded and early publications provide limited infor mation.
In 2002 systematic exploration of fossil and archaeological materials were under taken by a multidisciplinary international team, with a focus on the undisturbed deposits of Azokh 1 (beds v-n).16 A stra tigraphical study enabled geologists to check and refine conclusions reached in earlier excavations (Fig. 2). Survey and mapping of the entire cave system revealed Azokh 2 and Azokh 5, both chambers with a long sequence of undisturbed deposits that may correlate with those of Azokh 1 (Fig. 3).
In the present investigations finds are recorded three-dimensional! y, and all excavated sediments are passed through dry and wet sieving processes in order to retrieve micro-fauna!, lithic debitage and botanical remains. Large animal remains include red and fallow deer, wild goat, wild boar, rhino, wolf, hyaena and, in particular, cave bear, the latter probably having died during hibernation. Bones, including bear bones, with cutmarks made by stone tools indicate dismembering and butchering processes by humans, although it is impossible to determine if humans killed the animals or scavenged dead car casses.
Human occupants of the cave exploited D limestone bedrock rockfalls Figure 3 The topography of the Azokh cave system (modified from an original by P. Andrews & P. Ditchfield, with permission).
a range of stone types for tool manufacture, among them chert, limestone, quartzite, and obsidian. Most of the stone is of local origin; however, the nearest source of obsidian is about 60 km away which raises the question of how and why hominins obtained obsidian. Study of stone-tool technology indicates a change in produc tion systems between stratigraphical bed V (see Fig. 2 ) , where a simple technology is employed, and bed II, where the prepared core, or Levallois, technique indicates a more complex approach to manufacture. These technological differences could be the handiwork of different hominin spe cies. In Europe, Levallois production (as evidenced in bed II ) is often associated with Neanderthals; however, the pre Neanderthal mandible (indicating a differ ent hominin species) that was found in the 1960 excavations came from bed V, where stone-tool evidence indicates less techno logical complexity. A recent study of charcoal samples pro vides useful environmental information. Identified species include fruit-bearing trees, which could have supplied food as well as fi rewood. Analysis of sediments adhering to stone tools has revealed starch grains, although the particular species of plants from which these were derived have not yet been identified. Charcoal and starch analyses are at a preliminary stage, but future work, augmented by pollen and phytolith analyses, will increase the archaeobotanical database. Using a range of modern scientific techniques, the present excavations have begun to reveal tantaliz ing aspects of how Azokh's early occu pants interacted with their environment. The prospect for future research is excit ing.
The Palaeolithic occupation of inner Spain during the Upper Pleistocene (Ignacio de la Torre) Cooling temperatures affecting Europe at the end of marine isotopic stage 5 ( MIS5 )18 culminated in glacial conditions at the beginning of MIS4, around 74,000 BP. Ice sheets in northern Europe displaced popu lations of plants and animals (including humans) to refugia in warmer, more south erly latitudes, such as the Iberian Penin sula.
The portrait of the Iberian Peninsula as a refuge for human populations during the Upper Pleistocene is demonstrably valid for the Cantabrian and Mediterranean coasts, where relevant archaeological sites are plentiful. In such regions, intense archaeological research undertaken dur ing the nineteenth and twentieth centuries continues today. However, the density of Upper Pleistocene sites is significantly lower in the interior of the Peninsula and this has traditionally been attributed to the dry and cold climatic conditions that pre vailed during the marine isotope stages MIS4, MIS3 and MIS2.
Nonetheless, the fragmentary record of Palaeolithic colonization of continental Spain could equally reflect a lack of research rather than extreme climatic con ditions. Our research project is fo cused on an evaluation ofPalaeolithic settlement in the interior during MIS4, MIS3 and MIS2 (the so-called Pleniglacial period, which was the coldest known period of the Iberian Peninsula) . Current fieldwork is in two regions (Fig. 4): the Pre-Pyrenees of Cata lonia and the Spanish Southern Plateau at Cuenca (Castilla-La Mancha). The project at Lleida in the Pre-Pyrenees aims to reconstruct the behavioural adap tations of the Neanderthals and early mod ern humans who occupied the caves and rockshelters of this mountainous region. 19 Several field seasons at La Roca dels Bous (Fig. 5), have revealed many Middle Pal aeolithic levels. Studies of hearths, lithics and lithic refits, as well as analysis of fossil bones, the spatial configuration of the Cava Gran (Fig. 6) is a second rockshel ter currently under investigation as part of the Upper Pleistocene human settlement project of the Catalonian Pre-Pyrenees. Discovered in 2002, Cava Gran has a thick stratigraphical sequence the excavation of which has not yet reached bedrock. We have identified Mesolithic and late Upper Palaeolithic levels in one area of the site and Middle to Upper Palaeolithic transi tion levels in another. Therefore, this site constitutes a compelling case for evaluat ing the adaptations of both Neanderthals and early modern humans and, along with La Roca dels Bous, will contribute to understanding the Upper Pleistocene set tlement of the Pre-Pyrenees.
How did modern humans colonize areas farther south in the Iberian Peninsula? We are seeking to answer this question through a new project in the Southern Plateau at Cuenca. Excavations in 2005 at Buendia rockshelter (Figs 4, 7) and will continue over the next few years. So far, more than 20 archaeological levels have been found, each containing a high density of cultural remains. Magdalenian (a later Upper Pal aeolithic culture, c. 14,000 BP ) materials have been found in the upper levels, but there are also many earlier Upper Palaeo lithic levels below these. Although the information is still preliminary, Buendia rockshelter has become crucial for recon structing the way in which Late Palaeo lithic hunter-gatherers lived in the harsh conditions of the Spanish Plateau during the most recent glacial maximum. reconstruct cultural adaptations ofPalaeo lithic hunter-gatherers to the changing cli matic conditions of the Upper Pleistocene in extreme environments such as those of the mountainous Pre-Pyrenees and the Southern Plateau. At the same time we are seeking to understand those adaptive responses in evolutionary terms through the comparison of behavioural strategies developed by Neanderthals and modern humans during the most recent ice age.

The Late Palaeolithic and Neolithic in the Levant (Andrew Garrard)
Southwest Asia is regarded as one of the earliest centres of village-based farming in the world; as a consequence, extensive research has been undertaken on the adap tations of the Late Pleistocene hunter gatherers in the region, and the factors that may have led to the development of sed entism and plant and animal domestica tion. The most detailed field studies have been undertaken in the natural habitat areas of the main plant cultivars, and par ticularly in the moist steppe and open woodland environments of the Levantine Corridor -stretching fr om southern Israel and western Jordan to the Euphrates Valley of northern Syria and southeastern Turkey (Fig. 8). For many years, far less was known of the Late Palaeolithic and Early Neolithic communities that inhabited both the more arid and the mesic areas to either side of the Corridor-in particular, the dry steppe and sub-desert environments of the Syrian Jordanian plateau, and the mesic forested mountain environments of Lebanon, west ern Syria and southern Turkey. Much of the author's field research over the past 20 years has been attempting to redress this imbalance, by reconstructing the environ mental history of selected areas in these arid and mesic habitats and examining their role in regional developments through The goal of our research in Spain is to Figure 7 Th e Buendia rocksh elter in Castejon (Cuenca, cen tral Spain).

ARCHAEOLOGY INTERNATIONAL
the Late Pleistocene and Early Holocene.
In the 1980s, the author undertook an extensive field project in the steppe and oasis environments of the Azraq Basin in eastern Jordan (Fig. 8). Following an initial survey programme, small-scale excava tions were undertaken at ten Late Upper Palaeolithic and Epi-Palaeolithic sites dat ing to 28,000-12,000 BP and larger-scale excavations at five Neolithic sites dating to 12,000-8,400 BP. The majority of these sites contained dense midden deposits with very substantial lithic and animal bone assemblages as well as carbonized plant remains. Their analysis has provided valuable insights into many aspects of human settlement, technology and sub sistence at the southeastern margins of the Levantine Corridor. 21 In the mid-1990s, the author began similar research in the formerly forested environments ofthe Sak c;:agi:izi.i region of southern Turkey (Fig. 8), where a detailed survey project was under taken. 22 Since 2002, the author has been under taking field research with Dr Corine Yaz beck from St Joseph's University in Beirut, looking closely at Palaeolithic and Neo lithic adaptations to the thickly forested environments of the northern Lebanese Mountains, which rise to over 3,000m to the southeast of Tripoli. Following an initial survey, excavations were begun at two adjacent caves at Moghr el-Ahwal in the Qadisha Valley (Fig. 9). In 2004, exca vations were undertaken in the smaller cave, where well preserved Late Neolithic, Natufian and Geometric Kebaran material (17,500-7,500 BP) was found; and, in 2005, excavations were begun in the adjacent larger cave, where there is evidence for similar periods of occupation, plus earlier material possibly dating to the late Lower or Middle Palaeolithic. Preliminary analy sis of faunal remains from the Epipalaeo lithic levels has clearly demonstrated a high dependence on forest resources, but, by the Late Neolithic, domestic livestock were also being kept in this region. It is hoped that the detailed investigations on the artefacts, bio-remains and sediments from the survey and excavations will pro vide a strong foundation for reconstructing Late Pleistocene and Early Holocene envi ronments, settlement and subsistence in this otherwise poorly known region. 23