Printer Friendly

Active tectonics and land-use strategies: a Palaeolithic example from northwest Greece.

Tectonics and human behaviour

Humans live at the interface between the solid earth and the unstable atmosphere. The earth appears constant and safe, while the atmosphere is changeable and seems to have the greatest effect on environment. Droughts or floods, erosion or deposition of sediment and the consequent fertility or barrenness of soils are most readily explained by climate, or in recent millennia by human interference. Large-scale tectonic processes of mountain building or subsidence, driven by the motions of the continental plates, seem to operate too slowly, and to belong to a too far-distant geological past, to have any impact within the time span of human occupation. Even the most rapid changes -- earthquakes and volcanic eruptions -- appear disruptive and temporary in effect unless they seem to trigger some far-reaching social change.

Here we argue that this perspective is too simple. The inexorable changes in landscape geometry caused by deformation of the solid earth can determine the way in which climate affects the landscape, and the way in which the landscape in its turn affects, and is affected by, human activity. In many parts of the world, an appreciation of tectonics is needed to understand the distribution of agricultural soils or of animal herds. On the longer time-scales of prehistory, tectonically driven landscape change may become an active agent of selection, creating pressures or opportunities for changes in behaviour. It is easy to see that tectonic uplift can destroy an environment by triggering erosion, but active deformation is by no means always negative in effect. Tectonic subsidence can create well-watered sediment traps and the accumulation of fertile soils. Uplift and subsidence also create natural barriers to animal movement that can be exploited by humans. It is interesting to note that many early Palaeolithic sites are in regions of tectonic activity. The East African Rift Valley, North Africa, the Levant and Sub-Himalayan India are obvious examples. Undoubtedly the ease of site discovery can be assisted by tectonic changes. Tectonically active areas are those most likely to develop thick, rapidly formed sequences of terrestrial sediments favouring burial and fossilization of archaeological evidence, followed by rapid erosion and exposure to discovery. But the reverse is true and tectonically active regions can destroy or obscure human evidence within decades or centuries. In our discussion of Epirus, we discuss why ease of discovery is not the main reason why a correlation between early human settlement and tectonic activity is to be expected.

Active tectonics

Since the advent of Plate Tectonics, it has been evident that deformation of the Earth is a continuing process. Relative motions at major plate boundaries have rates that can approach 20 cm per year (Demets et al. 1990). In oceanic regions deformation can be localized to ridges, trenches or transform faults with widths of less than 100 km. Where plate boundaries cross continental crust, contraction, extension or sideways motion (strike-slip) can be spread over regions with widths exceeding 1000 km. The overall strike-slip motion of 5 cm per year between the Pacific and North American plates, commonly regarded as being accommodated by the San Andreas fault, actually extends from a short distance off the Californian coast to central Utah (Slemmons et al. 1991). At most, 50 % of the deformation (2.5 cm per year) occurs as strike-slip on the San Andreas, while the remainder is distributed on contractional, extensional and strike-slip structures which have lower rates. These lower rates can nonetheless result in substantial changes. The Coastal Ranges south of San Francisco, rising at 1 mm per year (Valansise 1992), are 1000 m high and less than 1 million years old, while the floor of Death Valley has dropped relative to its uplifting flanks by a similar amount in the same period of time (King & Ellis 1990).

By global standards, deformation of the western USA is not unusually rapid. Although overall deformation rates in the eastern Mediterranean are comparable, at |is greater than~2 cm per year (Tapponier 1977), locally they are much higher: in the Aegean region |is greater than~10 cm per year (King et al. in press), and from eastern Turkey to the Himalayas |is greater than~7 cm per year (Demets et al. 1990; Tapponier & Molnar 1977). Again the zone of active deformation can be very wide. In the western zone of the Alpine-Himalayan belt, deformation extends for about 1500 km from the Maghreb to France and Germany; in the eastern zone from the Indus-Ganges basin to Lake Baikhal in Siberia, about 3000 km.

Deformation is not uniformly distributed within these zones. Death Valley or Lake Baikhal owe their present form to local extension and subsidence, while the Coastal Ranges in California, or the foothill folds of the Himalayas, are localizations of contraction. The rates in these regions accommodate only a part of the relative plate motions. Other active features create local uplift and subsidence in these deforming zones and even in the western USA not all have been identified, particularly features with annual rates of motion of less than 1 mm per year.

Deformation also occurs at a wide range of scales. Over long periods of time vast areas rise to form mountains such as the Alps or the Himalayas, or subside to form basins or valleys such as the Aegean trough or the Red Sea rift. These features result from the cumulative effect of motion on many smaller features which individually may have a shorter existence but nonetheless exhibit substantial rates of motion when active. At scales of 5 to 50 km, motion can be very rapid, and uplift or subsidence of 100 m is possible within historic times. In regions of contraction, not all motion is progressive uplift; in regions of extension, not all motion is progressive subsidence. Areas of local uplift and local subsidence can lie close together, only kilometres apart, or even hundreds of metres or less.

Temporal variability occurs and can cause activity to be overlooked, particularly on smaller features. Even features such as Death Valley and the associated Panamint Valley show little activity at present. But they are not dead. Ample morphological evidence exists that several major earthquakes have occurred in the last 10,000-15,000 years (Zhang et al. 1990). Similarly, historical data indicate that features between Turkey and Iran have had periods of major earthquakes lasting for tens of years separated by centuries of inactivity (Ambraseys 1989). In the absence of direct observations from instruments or historical reports, the best evidence for tectonic activity is commonly evidence for local disruption of rivers or evidence for local regions of erosion and deposition of sediment.

Human modes of land-use can be affected by modest amounts of uplift or subsidence. Even very small rates of movement can be significant in modifying water-table levels in the short term (100s or 1000s of years), or in creating substantial morphological features in the longer term. Overall, the effects of tectonic deformation may be expected to be widespread and not just restricted to regions where tectonic activity is widely acknowledged to be violent. Furthermore, understanding how tectonics influences human behaviour is not solely of use to the study of the past. Data collected for this purpose are also useful in understanding tectonic processes, and for addressing contemporary problems such as the vulnerability of critical engineering structures like chemical or nuclear facilities to earthquake hazard (Site Characterization Plan 1988; Carrigan et al. 1991). Such information becomes particularly important in places like the British Isles, commonly but incorrectly thought to be inactive, where even a low probability of a major event poses a very serious hazard.

Sediment traps and climatic insensitivity

Local concentrations of sediment or water are almost self-evidently advantageous to a variety of plant and animal species and to a range of human subsistence economies under a diversity of climatic regimes. Local sediment traps are characteristically formed in river catchments on geological structures subject to normal faulting or compressional folding. Active structures produce a series of local folds or uplifted blocks separated by troughs, with river profiles which alternate between stretches of down-cutting of uplifted structures and aggradation in the subsiding region between them. Climatically induced cycles of incision and aggradation are superimposed on this pattern, but it is the tectonic processes that provide the underlying long-term determinants on the pattern of erosion and sedimentation in the landscape.

Topographic change and sediment traps are not exclusively associated with regions of active tectonics. Vertical motion due to isostatic depression and rebound in regions of continental glaciation may have analogous effects, as can sea level change in coastal regions. Whether they result from glacial or tectonic causes, nearly all these features are associated with rates of vertical movement of metres or tens of metres per 1000 years (Slemmons et al. 1991). In regions of little activity, where the basic geometry of the landscape can be considered essentially static from the point of view of human land-use, climatically induced and humanly induced changes in erosional and sedimentary regimes may take prominence. Indeed, such regions can be said to be particularly vulnerable to such changes. Arid and semi-arid regions are especially fragile environments, where relatively small climatic or anthropogenic effects may have catastrophic consequences for economic viability and human survival. Conversely, in arid regions which are tectonically active, sedimentary traps will acquire an enhanced attraction through a degree of insensitivity to changes of climate and land-use, maintaining essentially stable environmental conditions for plant and animal life. The precise advantages of these sediment traps will vary depending on their size and other local factors, but will tend to become greater in conditions where soil and water are limiting factors and in regions which are otherwise environmentally fragile. Similar considerations apply at a much larger geographical scale, with regions of uplift providing rainfall catchment and a sediment supply which is concentrated in adjacent lowlands surrounded by an otherwise barren environment.

Topographic barriers and herd control

The notion of topographic barriers as ecologically advantageous features is not intuitively obvious, but follows from a consideration of humans as predators of medium and large sized mammals. One does not have to subscribe to a 'Man the Hunter' view of human origins, with its over-emphasis on meat-eating and male activities, to recognize that dependence on animals as a source of subsistence has been a significant selective factor in the course of human evolution, and in the survival and expansion of human populations up to the present day. Yet animal prey pose a double problem for the human predator: of accessibility; and of competition with other carnivores. This can be illustrated by taking the two extremes of the spectrum of 'man-animal' relationships. At one extreme is scavenging, where the human population relies on other carnivores to do the difficult work of immobilization and takes what is left over. At the other extreme is full domestication, in which the animal population is under permanent human control and protection. In between are a range of ill-defined relationships commonly labelled as 'hunting', which it is one of the main tasks of Pleistocene archaeology to explore and define.

Given that the initial reaction of a prey population to a predator is to flee, the hypothetically featureless plain offers a very poor chance of success to a technologically simple human hunter, unaided by efficient means of killing at a distance such as guns and horses, or by constraints on animal movement such as fences or other artificial barriers. Fertile local environments such as sediment traps and lake basins may help to 'tether' mobile prey to some extent, but do not solve the problem of their ability to resist capture by flight or defence, nor the problem of competition with other carnivores better adapted in body form to the needs of chase and attack. A complex topography with hill and mountain barriers, on the other hand, facilitates the monitoring, prediction and control of animal movements, creates bottlenecks where prey can be located and trapped and provides places to live or camp which are within easy access of the prey population, but hidden away so as to minimize disturbance of the animals -- all features which an intelligent but otherwise unspecialized predator may be able to turn to advantage.

As in the discussion of sediment traps, tectonics plays a fundamental role in the creation of such landscape features, and these features may be advantageous both on a local scale -- a matter of kilometres -- where barriers are often juxtaposed with local sediment traps by the inherent nature of tectonic displacement and folding, and on a regional scale, where topographic barriers may circumscribe large blocks of favourable grazing terrain and confine the movements of animals within them (Sturdy 1975).

Palaeolithic Epirus

Investigation of the Palaeolithic archaeology of the Epirus region of northwest Greece over the past 30 years has been accompanied by an unusual emphasis on off-site archaeology and the reconstruction of local and regional palaeo-landscapes (Bailey 1992; Bailey & Gamble 1990; Bailey et al. 1983a; 1986a; 1990; Dakaris et al. 1964; Higgs 1978; Higgs & Vita-Finzi 1966; Higgs et al. 1967; Higgs & Webley 1971; King & Bailey 1985; Sturdy & Webley 1988; Vita-Finzi 1978). Northwest Greece is also one of the most seismically active areas of Eurasia and may be expected to highlight the underlying effects of tectonically active structures on the prehistoric archaeology and land-use of a region.

Regional topography and seismicity

The topography of Epirus is rugged. A series of sharp ridges with elevations of 1000 m or more separate plateau regions that are in places deeply dissected by rivers. To the northwest in Albania and the northeast in the Pindus mountains, the terrain has higher elevations and even under the present interglacial conditions is relatively inaccessible. To the southwest the region is bounded by the Ionian Sea and to the south by the Gulf of Arta and the more subdued terrain of Akarnania.

The seismicity map shows that Epirus is subject to earthquakes with a rate of activity comparable to regions such as Japan, New Zealand and parts of the Middle East, where uplift rates of between metres and tens of metres per millennium are well established. Both studies of earthquake mechanisms and the geology indicate a region subject to compression, and its broad features can be understood in such a context. However, more recent work indicates that strike-slip motion also plays a role and is important to a more complete understanding of recent evolution of the topography (King et al. in press). The deformation is concentrated on a series of structures oriented roughly 20 |degrees~ east. The structures shown must extend into Albania, and seismicity indicates that features parallel to those on land must be active offshore. Although the features south of the Gulf of Arta have a similar strike, net compression gives way to the net extension that is characteristic of the Aegean and Peloponnese. Compression and uplift does, however, restrict the mouth of the Gulf of Arta and extends down the coast of Levkas, where it becomes associated with the very active Hellenic arc subduction system.

Elements of the topography can be understood in terms of continuing tectonic evolution. The plateau-ridge system is associated with the telescoping of a series of fault-bounded limestone islands separated by sediment-filled marine basins. These basins are uplifted by progressive tectonic movements to form flysch and flysch-like rocks, which are the other dominant rock type alongside limestone. These are younger than the limestone, made from coarser and shallower sediments, and create softer, sandstone rocks, with varying proportions of clay, silt and siliceous material. They show spectacular folding and compression resulting from tectonic displacement. A schematic diagram of the mechanism is shown in FIGURE 6 and the main features of the geology in FIGURE 7. The limestone regions are separated from regions of flysch by narrow limestone ridges that are the topographic expression of the most active fault zones.

On a more local scale, features such as Lake Ioannina or the Gulf of Arta are associated with active relative depression and many more local regions of subsidence can be identified. Uplift has resulted in spectacular gorges in some places with river profiles that alternate between down-cutting and deposition over short distances.

The same processes continue offshore. Kerkyra (Corfu) is a limestone island, and rivers such as the Kalamas supply sediment that will become flysch when compacted and uplifted. The offshore tectonics is therefore a continuation of that seen on land. During the Last Glacial, when sea level receded, the land revealed would not have been a uniform plain, although the relief would have been more subdued than inland. Uplifted ridges would have been cut by valleys containing river terraces, with lakes and ponded sediments occurring between these features. For example, the Gulf of Arta deposition could have continued throughout the glaciation as a result of the uplift at its mouth.

Different areas of the landscape thus have very different susceptibilities to tectonically induced surface disturbance. The limestone plateau regions are relatively undisturbed today while the flysch basins are tectonically very active. The flysch, being soft, erodes almost as fast as it deforms and the result is a heavily gullied 'badlands' landscape. Although it is tempting to attribute gullying and badlands of this kind to intensive pastoral and agricultural activity in the late Holocene and historical periods, this is a secondary factor in the geological circumstances of Epirus. The underlying primary cause of instability is tectonic.

This is demonstrated by a variety of indicators. Erosion of flysch is chronic in Epirus and extends well back into the Pleistocene. In the case of the river terraces of the Voidomatis, some units are high in material derived from local flysch areas while others are dominated by glacial outwash materials (Bailey et al. 1990; Lewin et al. 1991). When glacial processes are not producing large amounts of removable material, it is the flysch erosion which supplies most of the sediment. River terraces dated to |is greater than~150,000 years, c. 20,000 and c. 1000 years BP are all dominated by flysch-derived materials, and while human activities can be implicated in the most recent case, they can hardly be invoked for the earlier episodes.

To the west of Klithi in the Doliana basin, the terraces of the river Kalamas are of Holocene date and are largely composed of flysch-derived materials. The base of one terrace is dated to 9160|+ or -~80 BP (OxA-3943), a higher unit to 6240|+ or -~70 BP (OxA-3942), while sediments near the top are dated by TL to 4200|+ or -~600 (BM lab ref. DOL6). This suggests a prolonged period during the early to mid Holocene, a period of climatically optimal conditions for vegetation growth and of less than intense agricultural activity, during which flysch erosion in the river catchment maintained an apparently steady supply of sediment.

Finally, over the past 30 years there has been a substantial reduction in grazing pressure throughout Epirus, with consequent expansion of woodland and scrub vegetation, sometimes in the form of extensive and almost impenetrable thickets. Yet slope failure and gullying in the flysch basins continue unabated.

Palaeo-environment and palaeo-economy

Today Epirus is one of the wettest regions of Greece, with annual precipitation ranging from about 900 mm on the coast to 1300 mm inland, a pronounced dry summer season, and extensive winter snow cover in the hinterland and at higher altitude. Pollen studies indicate that, during the Last Glacial, cold and semi-arid conditions prevailed with a vegetation of Artemisia steppe and occasional stands of evergreen oak and pine, creating a predominantly open, or at best a parkland, landscape (Bottema 1974; Willis 1989). The combination of steppe landscapes with high lake levels, as recorded for example by beach deposits in the Kastritsa rock-shelter on the shores of Lake Ioannina, is best explained by a climate of cold winters, some 5 |degrees~ -6 |degrees~ C colder than at present, high winter precipitation, and summer aridity with temperatures 2 |degrees~ -3 |degrees~ C lower and less precipitation than today (Prentice et al. 1992). The Pindus mountains created locally wetter conditions, providing refuges for deciduous tree species, although the latter probably persisted only as isolated stands in protected localities, for example on the south-facing cliffs of the Vikos Gorge (Bennett et al. 1991; C. Turner pers. comm.; Willis 1989). Glacial moraines on the heights of the Pindus have been linked to valley-fill sediments which can be dated to between about 28,000 and 16,000 years BP, and the presence of local glaciation seems to have created conditions severe enough to inhibit any human use of the upland interior at the Last Glacial Maximum, i.e. from c. 20,000 to 16,000 years BP |Bailey et al. 1990. Lewin et al. 1991; Woodward 1990). Low plant biomass, snow cover in winter and water availability in summer would clearly have been important ecological limiting factors, with a major impact on regional population movements and individual settlement locations.

The Palaeolithic research conducted by Higgs in the 1960s culminated in the well-known hypothesis of seasonal transhumance, in which the rock-shelters of Asprochaliko and Kastritsa were interpreted as the winter and summer camps, respectively, of people who followed herds of red deer in their seasonal migrations between coast and hinterland (Higgs et al. 1967). The palaeoenvironmental data that has become available since then has, if anything, reinforced the environmental basis for this hypothesis, emphasizing the low plant and animal biomass on a regional scale, and the seasonal contrast between summer and winter climates and between coast and hinterland.

Subsequent archaeological investigations, however, have refined this picture through new surveys, excavations, radiometric dates and faunal analyses, particularly at Klithi, demonstrating greater complexity and variability in the palaeoeconomy through time and space (Bailey 1992). Settlement on the coast is of long duration, extending back at least 100,000 years (Bailey et al. 1992; Huxtable et al. 1992). The hinterland was only visited sporadically or intermittently, if at all, until after the Last Glacial Maximum. The high density of artefacts and faunal remains in the upper levels of Kastritsa (Bailey et al. 1983b) and the thick, culturally sterile deposits at the base of the Klithi sequence (Bailey et al. 1986b; Bailey & Thomas 1987) suggest that intensive exploitation of the hinterland occurred only after about 16,000 years.

Faunal remains show a greater diversity than predicted by the original Higgs hypothesis. Ibex is at least as common in the Upper Palaeolithic levels at Asprochaliko as red deer (Bailey et al. 1983b); bovids and equids appear well represented at Grava (Sordinas 1969) and at Kastritsa; the large faunal collection from Klithi is dominated by remains of ibex and chamois (Bailey et al. 1986a), and suids are present in low frequency at several sites. In view of the environmental constraints discussed above, we think it likely that all these species would have made seasonal migratory movements of some degree, with deer and equids travelling over the longest distances, pig the shortest, and ibex and chamois shifting altitudinal range over relatively restricted distances. Hence we believe that seasonal mobility would have dominated land use strategies, although we do not now believe that this would have involved herd following in the strict sense, or have been shaped solely by the movements of red deer.

Carnivores are represented by bones of lion, lynx, wolf, fox and pine marten, which would have been potential competitors with human populations, although they appear to have made virtually no contribution to the accumulation of the archaeological bone assemblages, at least by the end of the Upper Palaeolithic period to judge from the Klithi assemblage, where evidence of carnivore bone chewing or modification is almost totally absent.

Although the possibility of substantial plant-food gathering is often raised, usually on the basis of inappropriate environmental and ethnographic analogies from other regions, we do not believe the local evidence justifies much attention to this. Pollen data for the Last Glacial indicate a regional environment with a generally low plant biomass and one that is poor in edible plant-foods. Even in refuge areas with more favourable microclimates and a higher plant biomass, as in the vicinity of Klithi, plant macrofossils and pollen data from local lake sediments indicate an environment which is unlikely to have supported any density of edible plant-food (K. Willis pers. comm.). At Klithi itself, full flotation and wet-sieving procedures were carried out during excavation to check for plant remains, but none were recovered in spite of favourable conditions for their preservation. Grinding equipment, found in late Palaeolithic sites in the Near East and North Africa, and putatively associated with processing of plant-foods, is absent from the Epirus sites. The backed bladelets which abound, and which Clarke (1976) in a famous piece of devil's advocacy suggested could have been used in slotted knives to cut and grate vegetable matter, have been shown at Klithi by microscopic study of edge damage to have been used as projectile tips or hide-working awls (E. Moss pers. comm.). We do not totally discount plant-foods. Chenopod seeds could have been collected in the Last Glacial environment, and perhaps the roots of plants growing on lake margins. To the extent that plant-foods were collected in the Palaeolithic period, we note that their exploitation would have benefited from the local tectonic features described below, no less than the exploitation of herd animals. But we doubt that plant foods could have contributed any more than they do to the modern Sarakatsani, who, notwithstanding the dominant pattern of animal transhumance in their lives, also scour the landscape for all available naturally occurring edible plant-foods as relishes and supplements to the staple diet (Campbell 1967). Even less of a case can be made for a significant contribution from sea-foods (Bailey 1982; Bailey et al. 1983a).

Regional barriers and animal distributions

Placing animals into the landscape is a matter of combining the known or inferred habitat preferences and behaviour of the various species with sub-divisions of the landscape based on their relative edaphic potential (i.e. their potential to provide food and essential nutrients for animals). For edaphic categories we rely on the underlying geology, soils where these are known to have existed during the Pleistocene, terrain, and general climatic and vegetational parameters supplied by pollen data. A fuller discussion of the method is given elsewhere (Sturdy & Webley 1988). Here we focus on the differences between limestone and flysch geologies and the environmental and topographic barriers that they create.

Soils on the limestone plateaux are thin and patchy, except where sediments are concentrated in small basins, but they are much the most favourable for vegetation that is attractive to animals. Soluble phosphate, the principal vegetational contribution to the calcium compounds from which growing animals make bones, is between 2 and 4 times higher than in the flysch soils, and only the limestone soils have adequate trace elements for animal growth such as copper and cobalt (Sturdy & Webley 1988). In contrast, the flysch produces only thin and immature ranker soils and their edaphic quality remains low. It is important to emphasize that this characteristic of the flysch only applies to environments which are subject to repeated disturbance by tectonic factors. In stable environments productive soils eventually develop on the flysch, and this potential can be realized on a small scale in Epirus by terracing and intensive horticultural practices in the vicinity of villages. On a large scale, however, the extensive flysch basins have a poor grazing potential for animals, and have been treated as marginal for this purpose even in recent and historical times, whereas the limestone continues to be extensively grazed. The only advantage that flysch offers to grazing animals is that its soft sediments allow deep rooting plants to reach moisture, and hence may offer some browse of last resort on shrub and scrub vegetation during dry seasons in areas distant from well-watered lake and sedimentary basins.

Landscapes in a comparable geological and tectonic environment in North Island New Zealand have been used extensively for sheep and cattle since the early 1950s, but only as a result of aerial dressing with the trace elements the soils lack (Gibbs 1964). The withdrawal of government subsidies for appropriate aircraft in the 1980s is causing much of this land to become unproductive. The natural thick vegetation on these New Zealand soils or the flysch soils of Epirus superficially seems to be an ideal habitat for herbivorous mammals, compared to the thin vegetation on the Epirus limestone. We are, however, observing cause and effect. The limestone areas become heavily grazed because they are attractive to animals, while the adjacent flysch is relatively untouched.

Precipitation and temperatures were never such as to create a desert-bounded environment of the type indicated in FIGURE 2. However, topography and geology create analogous barriers. The narrow limestone ridges have very steep profiles and scree slopes and cliffs which act as formidable barriers to human and animal movement. The inadequacy of disturbed flysch to develop soils that produce satisfactory feedstock for herbivorous mammals could have inhibited animal movements as effectively as desert barriers in arid regions.

The effect of these barriers in combination with areas favourable for grazing reveals a large horseshoe-shaped area of grazing for the larger animals, enclosed by barriers of mountain, sea and flysch. The eastern arm of the horseshoe is an area of gentle limestone topography, forming an ideal grazing territory for the larger herbivores, bounded to the southwest by a flysch barrier and to the north and east by mountains and flysch, reinforced by permanent ice-fields at high elevation. These topographic features have the effect of making the eastern area of accessible grazing an effective cul-de-sac some 100 km long by 10-20 km wide, with important consequences for the control and prediction of animal movements. This is all the more significant given that deer, cattle and horse could only have used this area during summer because of the severity of the Last Glacial winters.

The western arm would have formed a similar large-scale enclosure of limestone plateaux, with optimal conditions for use in winter, enhanced by extensive areas of swampy ground in large basins like the area of the Gulf of Arta, where scrub and shrub vegetation would have provided important winter browse for deer and cattle, and insect infestation in summer would have powerfully stimulated seasonal herd dispersal. Some of the water-retentive soils could also have been important areas of summer grazing, and it is possible that this western coastal region formed an annual system of animal movements and economy at least partially independent of the eastern arm of the horseshoe (Bailey et al. 1986a). The eastern arm, however, would not have been viable without the complementary winter grazing of areas near the coast.

The Palaeolithic rock-shelters clearly have a patterned relationship to these regional features, controlling major points of entry and exit to the limestone plateaux. Many more Palaeolithic sites are known from extensive surveys throughout the region, including numerous open-air sites, but these are without exception found within the favoured areas of limestone terrain or around their edges. The absence of finds from the flysch basins is notable, if difficult to evaluate, given that any Palaeolithic artefacts deposited there are likely to have been removed or obscured by the intensive erosion to which the flysch is susceptible.

Local site environments

This phenomenon of topographic closure is reproduced at a smaller geographical scale in relation to the immediate surroundings of individual sites. Asprochaliko is in a small gorge hidden away to the west of the main deer migration routes, but well placed to intercept deer diverting from the main route into a small enclosed limestone plateau west of the site. This would have provided attractive spring grazing, especially for older stags and hinds, which are the animals most likely to leave the main herd at this season. The well-watered sediment trap to the northeast would also have been attractive towards the end of the summer, when dry conditions would have reduced grazing opportunities elsewhere, and lower temperatures were beginning to drive the herds back towards their winter territories nearer the coast. The site is also well placed to intercept ibex in local movements from high ground in the northwest to lower and more sheltered terrain in the southeast.

Kastritsa is on the northwest slope of a limestone 'island' protruding through lake sediments, and partly surrounded by the waters of the lake. Shallow gravel fans and water retentive lake sediments fringing the open water would have attracted horse and cattle respectively, with deer concentrating on the gentle limestone terrain to the south and west, feeding closer to the lake at the end of the summer. As at Asprochaliko, the site is well placed to one side of the main migration routes, but also controls a local enclosure formed by the eastern arm of the lake and steeper ground further to the east and north, while being secluded from it. Deer are most numerous amongst the bone remains, while cattle and horse are important secondary species, and the presence of all major anatomical elements suggests that the animals were killed near-by, rather than at more distant butchering locations (Kotzambopoulou 1988). The site location offers a combination of attractions facilitating exploitation of all three species: local enclosure, a locally fertile and water retentive environment and strategic but secluded proximity to routes of animal movement.

Similarly, Klithi is in a sheltered gorge, close to, but out of the way of, the main routes of seasonal movement of chamois and ibex, but well placed to control their movements into and out of an enclosed area of high summer grazing. The faunal remains suggest a very effective exploitation, with the introduction of whole carcasses into the site (Bailey et al. 1986a). In addition the occurrence of isolated deer bones associated with artefacts in the neighbouring rock-shelter of Megalakkos and at the mouth of the nearby plain suggests that the area may also have been important in monitoring deer movements at the extreme of their summer range to the west. The alluvial sediments that filled the valley in front of the site, and extended out over the large Konitsa basin to the northwest, appear to have consisted of active outwash gravels during the period when the site was occupied, between 16,000 and 10,000 years, and presented a bare and stony landscape with perhaps at best a light spring flush of fine grasses. The finer sediments and soils which provide a fertile basis for present-day agriculture are of late Holocene data (Lewin et al. 1991; Sturdy & Webley 1988). This represents an important difference from the other two sites, but is consistent with the emphasis on caprids in the site fauna, since these are animals which are less critically dependent on water supplies than the larger herbivores, and which could have mitigated any effects of summer aridity by moving to high altitude pastures.

Long-term dynamics

The above reconstructions describe the average conditions for the Last Glacial, and as such give a largely static picture of the landscape and of human economic activities within it. However, in assessing the long-term success of human and other populations in the regional environment, it is not so much the average conditions that are relevant as the shorter-term fluctuations, and especially the extremes in the range of such variability (King & Lindh in press). It is these extremes which are critical in defining the selective pressures which shape longer term trends, giving a competitive edge to those species and populations that are better equipped to cope with periods of environmental deterioration, and hence are better placed to respond rapidly when environmental conditions improve.

Winter cold and snow cover, especially in bad winters, is an obvious limiting factor of the Last Glacial environment, which would have placed a high premium on conditions of local shelter or at low altitude near the coast. The rock-shelter of Asprochaliko has long been recognized, with its south-facing aspect, as a favourable location for winter use (Legge 1972). The south-facing Klithi in its protected gorge offers a similar attraction, important even in the summer months for habitation near high altitude terrain in close proximity to a permanent ice-field. Recent palaeo-environmental data underline this point while also demonstrating that summer aridity would have been another powerful limiting factor, placing an equally high premium on areas of water-retentive sediments. The importance of tectonics in sustaining local areas of fertility under these circumstances cannot be overemphasized. The fact that the three sites discussed offer advantages of shelter, access to locally fertile and well-watered sediments, or both, is thus of great significance. By controlling those local areas of the landscape best protected against extremes of cold and aridity, the human population would have been well placed to cope with temporal variability in environmental conditions, both the large-scale changes associated with the Last Glacial Maximum, when aridity and cold would have been at their maximum, and smaller-scale fluctuations from year to year.

The hinterland region, being available for humans on only a seasonal basis, was probably always a marginal area for human exploitation. It certainly appears to have been visited only sporadically until quite late in the regional sequence of Palaeolithic occupation, and was therefore an area in which other carnivores could have competed successfully for prey. With climatic deterioration at the Last Glacial Maximum, all animal activity in the hinterland would have been severely restricted, and the carnivores would have been forced into the coastal lowlands into closer competition with humans who were already well established there. When conditions began to improve again, the human population, by virtue of their ability to control key topographic features in the landscape, would have been able to respond quickly, and thus to secure and maintain an increasingly effective control of the hinterland region and its seasonal population of herbivorous mammals. To the extent that tectonic processes have not simply created those features of the landscape that human populations have turned to their advantage, but have actively sustained and even accentuated them during the course of human occupation, tectonics needs to be considered as a far more active agent in human-environment interactions than has previously been the case, with further consequences for an understanding of other dynamics -- climatic, vegetational, cultural and social.

The ecological and evolutionary 'drama' has traditionally been viewed in terms of ecological interactions between human populations and the biological organisms which provide food or act as predators, competitors and parasites. Quaternary climatic and vegetational change provide changes of scenery and sometime more active players. Changes in the solid earth, however, have been treated as external, adventitious and disruptive, causing a temporary halt in production or a change of venue, but not otherwise actively shaping the performance. This view is clearly over-simplified, and tectonics, like other environmental or biological processes, needs to be more actively incorporated as a variable shaping the long-term interactions which affect the course of human biological and socio-cultural change.


This case-study raises an issue of more general interest, and that is the nature of the different temporal scales at which landscape processes operate, the ways in which human groups perceive and interact with those processes, and the problem of how these different scales of phenomena are related. On very short time-scales of decades, essentially the time-scale of individual perception or living memory, the only visible impact of tectonic processes is the occasional earthquake, which must seem at best neutral, at worst disruptive. Obviously at this scale, the advantages of living in a tectonically active landscape are likely to be perceived in terms of superficial productivity of resources, rather than in terms of underlying tectonic processes, and the connection between the two may not be apparent.

On longer time-scales of thousands to tens of thousands of years tectonic processes in the Epirus context appear to have an essentially constructive impact on human land use, creating or sustaining local barriers and sediment traps and their concomitant advantages, resulting in human population stability in the face of adversely fluctuating climatic circumstances, or population growth at the expense of other ecological competitors. On longer time-scales again, hundreds of thousands to millions of years, tectonic processes may have a disruptive effect at the local scale, transforming local areas of subsidence into areas of uplift and erosion, as has happened with the Kokkinopilos red beds (Bailey et al. 1992), while on the longest time-scales of millions to tens of millions of years, whole regional landscapes are likely to be destroyed or transformed out of all recognition.

It is clearly important to be aware of the different effects that continuing tectonic evolution can have at different time-scales, the ways in which this can interrupt or modify other sorts of environmental processes and the ways in which these different scales of interaction can affect human behaviour at short and long time scales. For example, if our time perspective were confined solely to the Holocene, the principal cause of erosion on the flysch might be ascribed to agricultural practices, climatic changes or grazing pressures. All of these processes have been invoked as powerful agents of landscape change (and sometimes of landscape stability) at various times and places elsewhere in the Greek context (Davidson 1980; Pope & Van Andel 1984; Van Andel & Zangger 1990; Vita-Finzi 1969; Wagstaff 1981), and the controversy that surrounds the interpretation of, for example, Vita-Finzi's 'Younger Fill' is indicative of the difficulties of disentangling cause and effect. In the Epirus case, a longer time perspective reveals the underlying tectonic instability and the simple geological explanations for it, and many of the factors commonly invoked as causes of erosion may simply be triggers acting on the underlying instability. Without an appreciation of differential time-scales, we could easily draw the wrong conclusions, confusing proximate and ultimate causes, and playing havoc with decisions on future land-use policy.

At the longer end of the time-scale, tectonically induced landscape change may impose selection pressures resulting in evolutionary change. The location of archaeological sites in the complex landscape of Epirus has been discussed largely within the time scale of anatomically modern humans. But there is no reason why such an approach should not contribute a valuable understanding to much earlier periods and other parts of the world. Uplift can divert the course of rivers or cause them to create deeply incised channels. Fold and fault fronts produce local enclosure or natural fences. Such features lead to a complex topography, which, by channelling or concentrating nutrients and resources, offers much more potential than a featureless plain to a predator that must use its wits to be competitive. Given the complex topography of tectonically active areas, the human brain had additional material from which to fashion a unique technical competitiveness.

Acknowledgements. Fieldwork was supported by grants from the British Academy, the British School at Athens, the National Geographic Society and the Society of Antiquaries, and by permits issued through the good offices of the British School at Athens, the Ministry of Culture, Athens, the Institute of Geological and Mineralogical Research (IGME), Athens, the Ephoreia of Palaeoanthropology and Speleology, Athens and the Ephoreia of Prehistoric and Classical Archaeology, Ioannina. We are also grateful for financial support to the Archaeomedes Project on Desertification in Southern Europe, funded by Directorate XII of the European Economic Commission, and to the Oxford Radiocarbon Accelerator Unit and the British Museum Research Laboratory for radiometric dates.


AMBRASEYS, N.N. 1989. Temporary seismic quiescence: SE Turkey, Geophysical Journal 96:311-31.

BAILEY, G.N. 1982. Coasts, lakes, and littorals, in M.R. Jarman, H.N. Jarman & G.N. Bailey (ed.), Early European agriculture: 72-107. Cambridge: Cambridge University Press.

1992. The Palaeolithic of Klithi in its wider context, Annual of the British School of Archaeology at Athens 87: 1-28.

BAILEY, G.N., P.L. CARTER, C.S. GAMBLE & H.P. HIGGS. 1983a. Epirus revisited: seasonality and inter-site variation in the Upper Palaeolithic of north-west Greece, in G.N. Bailey (ed.), Hunter-gatherer economy in prehistory: a European perspective: 64-78. Cambridge: Cambridge University Press.

1983b. Asprochaliko and Kastritsa: further investigations of Palaeolithic settlement and economy in Epirus (north-west Greece), Proceedings of the Prehistoric Society 49: 15-42.

BAILEY, G.N., P.L. CARTER, C.S. GAMBLE, H.P. HIGGS & C. ROUBET. 1984. Palaeolithic investigations in Epirus: the results of the first season's excavations at Klithi, 1983, Annual of the British School of Archaeology at Athens 79: 7-22.

BAILEY, G.N. & C.S. GAMBLE. 1990. The Balkans at 18,000 BP: the view from Epirus, in C.S. Gamble & O. Soffer (ed.), The world at 18,000 BP: 148-67. London: Unwin Hyman.

BAILEY, G.N., C.S. GAMBLE, H.P. HIGGS, C. ROUBET, D.A. STURDY & D.P. WEBLEY. 1986a. Palaeolithic investigations at Klithi: preliminary results of the 1984-1985 field seasons, Annual of the British School of Archaeology at Athens 81: 7-35.

BAILEY, G.N., C.S. GAMBLE, H.P. HIGGS, C. ROUBET, D.P. WEBLEY, J.A.J. GOWLETT, D.A. STURDY & C. TURNER. 1986b. Dating results from Palaeolithic sites and palaeoenvironments in Epirus (north-west Greece), in J.A.J. Gowlett & R.E.M. Hedges (ed.), Archaeological Results from Accelerator Dating: 99-107. Oxford: Oxford University Committee for Archaeology.

BAILEY, G.N., J. LEWIN, M.G. MACKLIN & J.C. WOODWARD. 1990. The 'Older Fill' of Epirus, north-west Greece and its relationship to the Palaeolithic archaeology and glacial history of the region, Journal of Archaeological Science 17: 145-50.

BAILEY, G.N., V. PAPACONSTANTINOU & D.A. STURDY. 1992. Asprochaliko and Kokkinopilos: TL dating and reinterpretation of Middle Palaeolithic sites in Epirus, north-west Greece, Cambridge Archaeological Journal 2: 136-44.

BAILEY, G.N. & G. THOMAS. 1987. The use of percussion drilling to obtain core samples from rockshelter deposits, Antiquity 61: 430-39.

BENNETT, K.D., P.C. TZEDAKIS & K.J. WILLIS. 1991. Quaternary refugia of north European trees, Journal of Biogeography 18: 103-15.

BOTTEMA, S. 1974. Late Quaternary vegetation history of northwestern Greece. Groningen: Groningen University Press.

CAMPBELL, J.K. 1964. Honour, family and patronage. Oxford: Clarendon Press.

CARRIGAN, C.R., G.C.P. KING & G.E. BARR. 1991. Water-table excursions induced by seismic and volcanic events at Yucca Mountain, Geology 19: 1157-60.

CLARKE, D.L. 1976. Mesolithic Europe: the economic basis, in G. de G. Sieveking, I.H. Longworth & K.E. Wilson (ed.), Problems in economic and social archaeology; 449-81. London: Duckworth.

DAKARIS, S.I., E.S. HIGGS & R.W. HEY. 1964. The climate, environment and industries of Stone Age Greece, part I, Proceedings of the Prehistoric Society 30: 199-244.

DAVIDSON, D.A. 1980. Erosion in Greece during the first and second millennia BC, in R.A. Cullingford, D.A. Davidson & J. Lewin (ed.), Timescales in geomorphology: 143-59. Chichester: Wiley.

DEMETS, C., R.G. GORDON, D.F. ARGUS & R.S. STEIN. 1990. Current plate motions, International Geophysical Journal 101: 425-78.

ELLIS, M. & G.C.P. KING. 1991. Structural control of flank volcanism in continental rifts, Science 254: 839-42.

GIBBS, H.S. 1964. Soils of Northland, in National resources survey, Part III -- Northland region: 25-37. Wellington: Government Printer.

HIGGS, E.S. 1978. Environmental changes in northern Greece, in W.C. Brice (ed.), The environmental history of the Near and Middle East since the last ice age: 41-9. London: Academic Press.

HIGGS, E.S. & C. VITA-FINZI. 1966. The climate, environment and industries of Stone Age Greece, part II, Proceedings of the Prehistoric Society 32: 1-29.

HIGGS, E.S. & D.P. WEBLEY. 1971. Further information concerning the environment of Palaeolithic man in Epirus, Proceedings of the Prehistoric Society 37: 367-80.

HIGGS, E.S., C. VITA-FINZI, D.R. HARRIS & A.E. FAGG. 1967. The climate, environment and industries of Stone Age Greece, part III, Proceedings of the Prehistoric Society 33: 1-29.

HUXTABLE, J., J.A.J. GOWLETT, G.N. BAILEY, P.L. CARTER & V. PAPACONSTANTINOU. 1992. Thermoluminescence dates and a new analysis of the early Mousterian from Asprochaliko, Current Anthropology 33: 109-14.

KING, G.C.P. & G.N. BAILEY. 1985. The palaeoenvironment of some archaeological sites in Greece: the influence of accumulated uplift in a seismically active region, Proceedings of the Prehistoric Society 51: 273-82.

KING, G.C.P. & M. ELLIS. 1990. The origin of large local uplift in extensional regions, Nature 348: 20-27.

KING, G.C.P. & A.G. LINDH. In press. Predictability and the evolution of culture, in S. van der Leeuw (ed.), Proceedings of the Cambridge conference on dynamical descriptions and human systems. Edinburgh: Edinburgh University Press.

KING, G.C.P. & R.S. STEIN. 1983. Surface folding, river terrace deformation rate and earthquake repeat time in a reverse faulting environment: the Coalinga, California, earthquake of May 1983, in The 1983 Coalinga, California, earthquake: 61-9. Sacramento (CA): California Division of Mines and Geology. Special Publication 66.

KING, G.C.P. & C. VITA-FINZI. 1981. Active folding in the Algerian earthquake of 10 October, 1980, Nature 292: 22-6.

KING, G.C.P., D.A. STURDY & J. WHITNEY. In press. Landscape geometry and the active tectonics of northwest Greece, Bulletin of the Geological Society of America.

KOTZAMBOPOULOU, E. 1988. Faunal analysis of the Kastritsa Cave, Greece. Unpublished MPhil. thesis, Cambridge University.

LEGGE, A.J. 1972. Cave climates, in E.S. Higgs (ed.), Papers in economic prehistory: 97-103. Cambridge: Cambridge University Press.

LEWIN, J., M.G. MACKLIN & J.C. WOODWARD. 1991. Late Quaternary fluvial sedimentation in the Voidomatis basin, Epirus, northwest Greece, Quaternary Research 35: 103-15.

POPE, K.O., & T.H. VAN ANDEL. 1984. Late Quaternary alluviation and soil formation in the Southern Argolid: its history, causes and archaeological implications. Journal of Archaeological Science 11: 281-306.

PRENTICE, I.C., J. GUIOT & S.P. HARRISON. 1992. Mediterranean vegetation, lake levels and palaeoclimate at the last glacial maximum, Nature 360: 658-60.

Site characterization plan. 1988. Yucca Mountain Site, Nevada research and development area, Nevada. Oak Ridge (TN): US Department of Energy, Office of Scientific and Technical Information.

SLEMMONS, D.B., E.R. ENGDAHL, M.D. ZOBACK & D.D. BLACKWELL (ed.). 1991. The neotectonics of North America: Boulder Colorado, decade map 1. Boulder (CO): Geological Society of America.

SORDINAS, A. 1969. Investigations of the prehistory of Corfu during 1964-1966, Balkan Studies 10: 393-424.

STEIN, R.S. & G.C.P. KING. 1984. Seismic potential revealed by surface folding: the 1983 Coalinga, California, earthquake, Science 224: 869-72.

STURDY, D.A. 1975. Some reindeer economies in prehistoric Europe, in E.S. Higgs (ed.), Palaeoeconomy: 55-95. Cambridge: Cambridge University Press.

STURDY, D.A. & D.P. WEBLEY. 1988. Palaeolithic geography: or where are the deer? World Archaeology 19: 262-80.

TAPPONIER, P. 1977. Evolution tectonique du systeme alpin en Mediterranee: poinconnement et ecrasement rigide-plastique, Bulletin de la Societe Geologique de France 19: 437-60.

TAPPONIER, P. & P. MOLNAR. 1977. Active faulting and tectonics in China, Journal of Geophysical Research 82: 2905-30.

VALANSISE, G. 1992. Geological records of combined tectonic processes in the central Santa Cruz mountains. Loma Prieta: United States Geological Survey.

VAN ANDEL, T.H. & E. ZANGGER. 1990. Landscape stability and destabilization in the prehistory of Greece, in S. Bottema, G. Entjes-Nieborg & W. Van Zeist (ed.), Man's role in the shaping of the eastern Mediterranean landscape.: 139-57. Rotterdam: Balkema.

VITA-FINZI, C. 1969. The Mediterranean valleys: geological changes in historical times. Cambridge: Cambridge University Press.

1978. Archaeological sites in their setting. London: Thames & Hudson.

WAGSTAFF, J.M. 1981. Buried assumptions: some problems in the interpretation of the 'Younger Fill' raised by recent data from Greece, Journal of Archaeological Science 8: 247-64.

WILLIS, K.J. 1989. Late Quaternary vegetation history of Epirus northwest Greece. Unpublished Ph.D dissertation, Cambridge University.

WOODWARD, J. 1990. Late Quaternary sedimentary environments in the Voidomatis Basin, northwest Greece. Unpublished Ph.D dissertation, Cambridge University.

ZHANG, P., M. ELLIS, D.B. SLEMMONS & F. MAO. 1990. Right-lateral displacements and the Holocene slip rate associated with prehistoric earthquakes along the southern Panamint Valley fault zone: implications for southern Basin and Range tectonics and coastal California deformation, Journal of Geophysical Research 95: 4857-72.
COPYRIGHT 1993 Antiquity Publications, Ltd.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1993 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Bailey, Geoff; King, Geoff; Sturdy, Derek
Date:Jun 1, 1993
Previous Article:Rock art and changing perceptions of southern Africa's past: Ezeljagdspoort reviewed.
Next Article:The Neolithic timber hall at Balbridie, Grampian Region, Scotland: the building, the date, the plant macrofossils.

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters