The Antiquity of Man

   Home   |   About   |   What is Evolution   |   World Archaeology & Palaeoanthropology   |   North African Archaeology   |   Pseudoscience   |   Recommended Readings   |   Reviews   |   Links   |   Search   |   Contact

Mike Brass. 1999 essay. How the concept of landscape archaeology has been applied to the Early Stone Age

The term landscape is frequently used by archaeologists to categorise an activity, whether mental or physical, that is engaged by hominids with their surrounding enviornment, i.e. subsistence or ritual (in the form of, for example, interconnected ritual sites) interactivity usages. Therefore landscape "refers to the integration of natural and human phenomena on a portion of the earth's surface… Landscape, as systems of settings, are taken to be intimately related to human life, and are primarily for living and working in rather than for just looking at. They are always "symbolic": i.e., they always have meaning - which is another way of saying they are cultural" (Rapopoort 1992).

These definitions are in themselves very broad and leave open a large scope for varying degrees of application to and interpretation of remaining physical evidence. To a large degree this is because human action on landscapes occur over time and therefore the end result examined by archaeologists (and also by other disciplines such as art historians and cultural geographers) is one of complex historical make-up.

Different groups of people, comprising of varying "cultures", have their own beliefs as to the "ideal" life-style. In essence these are changes and variations in how phyical landscape with their natural and cultural attributes are evaluated (Rapopoort 1992). For example, during the Middle Ages in Europe streets became the center of social activities and where thereby embued with positive attributes, with the city's quarters opening onto them (Rapopoort 1992). This is in stark contrast to many city-states of classical Greece.

Rapopoort (1992) put it perfectly when he stated: "… A cultural landscape is a system of settings within which particular systems of activities taken (including their latent aspects) place in space and time, incorporating particular proximities, linkages and separations, and boundaries among settings (and the ways in which these are expressed physically). All of these, in turn, reflect and influence communication, and also have meaning. It follows that a dwelling, for example, is not only part of a larger system, but is itself a particular subsystem of settings. Both that subsystem and the way in which it relates to other settings in the larger system in which it is set, are highly variable cross-culturally. In other words, the question "who does what, where, when, including or excluding whom (and why?)" receives very different answers. Put differently, the settings comprising the systems are linked in space and time through the systems of activities. Moreover, the extent of the various systems needs to be discovered…"

An example of just how variable the interpretations of archaeological landscapes can be is best demonstrated by the study of rock art. With regard to prehistoric Atlantic Europe, Bradley hypothesises that rock art within striking range of habitual settlements revealstart contrasts with the distribution of both carvings and paintings across a broader landscape context (Bradley 1997). The latter distribution of sites portrayed instead "pathways" in the landscape,and assocations with water sources; while the extent and character of important social networks can be traced in both cases through the different styles of art. By contrast Whitley interprets the landscape symbolism of North American rock art as portraying masculine and feminine activities in a landscape that was previously an intregal part of a symbolic sysetm embued with nuanced meaning (Whitley 1998). Manhire, Parkington and van Rijssen worked on the distribution of rock art in the south-western Cape in terms of settlement change, in which the paintings might reveal patterns of concentration and dispersal of the groups the painters were an integal part of (Manhire, Parkington & van Rijssen 1983).

By the very nature of the forces of preservation the further back in the past the archaeologist deals with, the scantier the fossil, artefactual and habitational evidence. The challenge facing the archaeologist is how to go about reconstructing the hominids' relationship with the surrounding ecological landscape. Back at roughly 1.5 million years ago the scanty remains, together with the fact that archaeologists and paleoanthropologists are not dealing with Homo sapien sapiens but with our Homo predecessors, renders any attempt to determine the cultural landscape of those remote epochs very tenious. It is a debatable point where early Homo possessed a capacity for symbolism or not. These factors, taken together, immediately impose strict limits on the investigator as to the various research and interpretative paradigms he/she can apply to a given problem, thus leaving the door open for novel and innovative methods. Rather than searching for symbolic explanations of the placement of archaeological features in the landscape, archaeologists and paleoanthropologists instead concentrate on functional, socio-economic probabilities through a combination of recorded studies of our nearest surviving relatives the chimpanzee, through typological analyses of the stone aretfacts, through microscopic examinations of the damage exhibited on animal bone remains, and through the range, distribution and clustering density patterns of both the stone tools and animal bones.

The examination of early hominid land-use patterns through landscape archaeology is accomplished through varying research programs on the regional distribution of sites, how these sites are inter-related as well as their ecological contexts. The four models examined here begin from essentially differing premises, yet all have the same structuring theme of their "emphasis [being on] the paleogeographic and ecological contexts of synchronic hominid behaviour" (Blumenschine & Masao 1991).

Schick (1987) endeavours to sample specific horizontal archaeological layers at "frozen" moments in time, these frozen moments consisting for 100 000 years or longer. Isaac (1978), and Rose and Marshall (1996) produce general models to account for artefact distributions through space and time in the early hominid tool-production period. Peters and Blumenschine (1998) reconstruct paleoenvironments and use it as a basis for predicting the distribution and function of early hominid "sites".

These exercises have been collectively termed "landscape archaeology." The aim, with regard to hominid paleoenvironments, is to map the density distribution of artefacts and their compositional degree of spatial variability, explaining these sites' socio-economic function within an interpretative framework of the surrounding ecological context.

The Home Base hypothesis

Three of the best currently known occurrences of artefact assemblages in distinct horizon layers are the FxJj 20 site complex of Koobi Fora, and the FLK Zinj and the FC West Floor of Olduvai Gorge. They were originally termed "living floors" by the Leakeys, which carried the implications of artifactual debris deposited on a ground surface within either one or more temporally fixed occupations areas (Isaac 1978). A modern analogue would be a hunter-gatherer camp where a particular social group of people would carry out their diverse day-to-day domestic activities of survival, for a limited period.

This model failed to explain adequately the site formation processes of the "living floor" occurred. This was a challenge taken up subsequently by Isaac in the development of his "home base" hypothesis (Isaac 1978). The foundation of the home base hypothesis rests on the assumption that there exists, at any given time in hominid history, human social groups foraging and hunting independently over the surrounding terrain who come together at a particular focal point in the landscape for the purpose of various activities. It is this focal point which Isaac classifies as a "home base" (Isaac 1978).

Isaac (1978) drew on two geographical areas to elaborate on, and provide examples for, his hypothesis. These are Koobi Fora and Olduvai Gorge. A locality at Koobi Fora, termed the hippopotamus/artifact site (HAS) dated to 1.6 mya, has yielded the faunal remains of hippopotami in association with stone tools. Isaac turned to a second site, the Kay Behrensmeyer site (KBS), at Koobi Fora to answer the questions posed about the hominid behavioural patterns that resulted in the HAS assemblage. KBS was situated in the same volcanic layer as HAS, and thereby provided good stratigraphic analogies between assemblages of comparatively the same age. Over a distance of 19 meters in diameter several hundred stone tools were excavated, with abundant animal bone remains in association. The nearest suitable material, both in the form of rocky outcrops and naturally occurring stones, for the manufacture of stone tools occurs three kilometers away (Isaac 1978). Thus the hominids carried the stone tools to the particular sites with the aid of a carrying device. In addition, Isaac believed it was unlikely that all the faunal remains were the results of killings that took place with a short time interval at these sites (Isaac 1978). This led him to tentatively conclude that both the meaty bones and the stone tools were transported to the chosen localities.

Isaac (1978) proposed that these behaviours could also account for the dense concentrations of broken-up bones and stone artifacts found by Mary Leakey at the Zinjanthropus site of Olduvai Bed I (1.7 mya). Other stone concentrations at Olduvai Gorge, Koobi Fora and sites like Omo have very few faunal remains. These could either have been stone tool manufacturing sites with other economic and social activities occurring which are not archaeologically visible. Isaac thus hypothesised that these hominids undertook foraging rounds, during which period they would split up and later regroup at a predetermined locality on the landscape (Isaac 1978). He took this hypothesis one step further by analogies with modern hunter-gatherer societies in proposing that during the foraging rounds the social group fissured further into males and females, males hunting and females collecting plant foods. These foods would then be redistributed at the home base among the social group for consumption (Schick 1987).

Isaac's home base model, therefore, draws analogies between the Plio-Pleistocene sites and present-day hunter-gatherer camps. The archaeological material is a combination of primary refuse and materials that had been deposited there for intended future purposes. Criticism has been levelled at this hypothesis as the result of its ahistorical assumption that early hominid behavioural activities were largely unchanged in terms of land-use patterns and social organisation from those of modern hunter-gatherer communities (Schick 1987). Also the question of how and why these faunal remains and manuports were created in such dense concentrations is essentially unanswered.

The Habitual Transport hypothesis

Schick (1987), who approaches the question from a novel angle, took up this challenge. Schick defines a site "as a place where stone import rates exceeded exports" (Schick 1987). The definition is used as a basis for building a model explaining the formation of archaeological deposits in concentrated space in any given stratigraphic horizon, the occurrence of vertically diffuse artifactual deposits and to hypothesise about the hominid behavioural patterns which resulted in contrasting spatially large and small concentrations of artefacts in the landscape.

The site of FxJj 50 (Koobi Fora), dated to 1.6 mya, contains 1405 flaked stone artefacts, 76 unshaped cobbles and cobble fragments, in association with 2100 faunal remains (Schick 1987). The intensive studies carried out on this site have revealed that the manufacturing activities occurred removed from the site locality. This brings into question what the possible hominid transport patterns of these stone tools to, and most likely also from, the locality.

General experimental replications, and the resultant predictive models, are consistent with Toth's analysis of Koobi Fora stone tool technologies from a number of suitably selected sites. This analysis shows that the cores present are in the late stage of reduction, and thus were imported from other localities (Schick 1987). On this basis, Schick believes it is possible to extend her hypothesis to cover all the Koobi Fora sites.

The question arises about how and why these artefacts were deposited at these sites, thus effectively taking them out of the broader equation of the stone transport system. Although some portions can be regarded as "de facto" waste, this would only be applicable to debris with a maximum dimension of less than 2cm that resulted from the on-site flaking processes (Schick 1987). This leaves the problem of the cores and core tools unaccounted for. Explanations such as deliberate disposal and abandonment of tools no longer required are not satisfactory. They cannot explain the volume of incomplete and complete cores, and large, sharp flakes so visibly in abundance at a great number of sites.

Schick hypothesises that, instead, "these concentrations of stone artifacts developed as a by-product of a habitual transport and discard behaviors centered on specific locales, rather than as a deliberate stockpile" (Schick 1987). Thus at individual points within the hominid movement and transport system and worked stone tools dropped out of this system (becoming archaeological sites with archaeological recognised materials), and these were places where the import of stone exceeded its export. There remains the question of why this import-export equation occurred.

Plausible reasons why this could have transpired include:

(1) the necessity to transport other materials, either gathered or processed (such as plant and animal foods, bedding and perhaps wood), as they moved on to their next locality in their hunting-foraging rounds;
(2) the sudden abandonment, of smaller sites, due to predator threat and;
(3) the next anticipated hunting-foraging locality having known stone resources, which the hominids would have been able to easily exploit (Schick 1987).

The localities that were rich in exploitable faunal resources were likely to have been the sites continuously visited periodically for planned hunting-foraging activities. The high rates of stone tool transport demanded by Schick's model is proposed to have occurred due to the uncertainty over suitable raw material availability during the course of the hunting-foraging rounds (Schick 1987). When moving onto a locality with known or anticipated high sources of raw materials it is hypothesised (see (3) above) that the artefacts would be left at the previous locality, being now surplus requirements. Provided that these sites were visited regularly enough on the hunting-foraging rounds, a substantial, accentuated development of stone deposition density would occur through a circular feedback mechanism (Schick 1987).

Medium density sites would have arisen through more sporadic occupation by the hunter-foragers due to their less attractive surrounding environments. Hominids moving to other hunting-foraging localities within this environment would have had an incentive to carry away with them some of their stone tools, thus increasing the exportation rate and maintaining the artefact density at a moderate level (Schick 1987).

By contrast, low-density depositions are those most frequently recorded in surface surveys. These localities may represent places of infrequent habitation or simply areas where large stone resources were both imported and exported by their hominids inhabitants due to the general scarceness of new stone resources in that particular environment.

Some localities contain vertically diffuse deposits. This may have been caused by frequently visited sites that had rapid rates of sedimentary deposition in comparison with the rate of cultural deposition (Schick 1987). Factors like vertically mixed deposits also have to be taken into consideration. Rock source sites provide an interesting way to test this hypothesis, although most of them are to be found in erosional or non-depositional environments, as well as within stream gravels, which severely hinder preservation. However, the MNK Chert Factory lends supports to Schick's model of importation of materials. It contains chert imported from other sources (Schick 1987).

Schick's hypothesis supports several predictions concerning the formation of stone tool assemblages in differing circumstances: intersite variability in assemblage size, raw material and technological diversity, and the frequency of visits by the early hominids to the localities.

Prime foraging areas are expected, in this model, to have attracted the making of the larger archaeological sites through frequent visitations by the early hominids and, by extension, are more likely to possess variably complex sets of technological systems (Schick 1987). These artefact systems would be comprised of imported artefacts and manuports, on-site manufactured artefacts, and artefacts which were exported from the locality. The frequency of the visits in favourable areas increases the possibility of their being deposited stone tools and unfinished stone tools made from raw materials found at distant sources. At Olduvai Gorge, for instance, some of the raw stone tool materials come from sources as far as 10 km removed.

Smaller-scale assemblages would, according to the predictions made by Schick's hypothesis, develop at sites less frequently visited, and where much debitage might occur with little conjoinable material. At these sites this would be a result of a higher export than import ratio, examples of them being FxJj 1, 10, 11 and 17 at Koobi Fora (Schick 1987).

The diverse stone resources available in the vicinity of Olduvai Gorge resulted in the stone transport patterns evident at those localities exhibiting much complexity, compared with their Koobi Fora counterparts. The hominids of Olduvai Gorge lived in rich lake margin environments with diverse stone resources. At the Koobi Fora basin the stone tool materials came from cobbles found in and by the streams that drained from east and northeast volcanoes (Schick 1987). These streams decreased in gradient and, with it, the size range of available cobbles to such an extent that the early hominids utilising the localities of FxJj 1, 3 and 10 (which are part of the Lower Member within the channels of the delta distributary adjacent to the lake) would have had to travel an estimated 4 km for appropriate raw materials to manufacture their stone tools.

The Resource-Defence model

Rose and Marshall (1996) present a revised version of Isaac's home base hypothesis, which is termed "the resource-defence model". They take a novel approach in examining what the effects of early hominid meat eating have on their relations with the Plio-Pleistocene carnivore compatriots, from a broad ecological perspective.

One of the central arguments against the home base hypothesis has been the stressing of the potential danger posed by the carnivores attracted to the carcasses. Rose and Marshall believe the underlying assumption of this argument is the early hominids possessed a behaviourally limited capacity for hunting and that scavenging was the easy way out (Rose & Marshall 1996). The inference is these hominids were incapable of defending either their kill or the carcasses to which they first had access. In this way the scavenging and home base models have been seen as mutually exclusive behavioural strategies. However, it is insufficient to argue on the basis of carnivore presence alone against the home base hypothesis, according to the view of Rose and Marshall (1996) in which they envisage carnivore predation and the competition resulting instead in increased co-operative social behaviour.

Due to our incomplete knowledge of habit use by both Plio-Pleistocene carnivores and the hominids, it is difficult to assess their degree of interaction. There was likely to have been a similar carnivore: prey ration to that existing today and most of the carnivores would have occurred their own environmental niche (Rose & Marshall 1996). Taking this into consideration, it is unlikely that the hominids would have been exposed to higher risks of predation than some extant primates.

Studies of primate behaviour in the face of predation danger are taken as the basis for the model. Living group primates have a wide range of alarm calls and co-operative defensive systems (Rose & Marshall 1996). "Data from studies of extant primates suggest that the most common response to predation risk is increased group cohesion and Cupertino in antipredator defence. We suggest that early hominids would have responded similarly to the risk of carnivore predation by intensifying co-operative behaviours."

Rose and Marshall avoid drawing a straight parallel between primate and early hominid behaviour, pointing out that meat comprises only a small portion of primate sustenance. As meat consumption increased with early Homo, so did the risk of increased carnivore competition and an increased chance of becoming the prey themselves. Under this threat, Rose and Marshall hypothesise (1996), early Homo developed to a greater extent the primate trend towards co-operative defence to actively defend both themselves and their resources.

The view of the hominids participating in passive scavenging holds that there would have been little meat and bones to take back to a particular locality for processing. However if these hominids were instead engaging in active, also termed "confrontational", scavenging, they possessed the ability to appropriate carcasses from carnivores. This implies significant amounts of meat would have been obtained for carcass transport, processing and group sharing at a chosen site (Rose & Marshall 1996). This competitive ability would have been sufficient not only in obtaining the carcass or carcasses, but further defending both their food resources and themselves at the focal localities in the landscape.

"Co-operative defence of high-quality resource patches has also been proposed as a primary selective pressure in primate social evolution. By living in-groups and co-operatively defending resources against others, individuals are expected to gain more in terms of improved nutrition and enhanced reproductive success than they lose through within-group resource competition. Wrangham (1980) argues that because the primary reproductive strategy of females is food acquisition (while that of males is mate acquisition), females are most likely to co-operate in resource defence, resulting in the predominance of a female-bonded social structure among primate species." (Rose & Marshall 1996) This is a major revision to Isaac's model of seeing the origins of modern hunter-gatherer social divisions in the Plio-Pleistocene.

Fruit trees are spatially fixed resource patches for primates. Rose and Marshall (1996) hypothesises that this concept can be extended to early hominids by regarding hunted and scavenged carcasses as movable high quality resource patches requiring defence from both carnivores and other hominid groups. This way of life would also have had to take into account stone tool raw material availability, water, trees for shade and potential sleeping places. Rose and Marshall propose that "an efficient strategy would have been to bring carcasses or carcass parts, which are perishable and movable resources, to a strategic area of fixed resources where tools were kept and carcass processing could be carried out co-operatively" (Rose & Marshall 1996).

Land Use Predictions

The basic principle on which Peters and Blumenschine's predictive model, for the lowermost Bed II Olduvai Basin, is built is that "inter-facet variability in the scales and characteristics of larger mammal scavenging opportunities for hominids will drive most aspects of the composition, density, and clustering of the recoverable bone and stone artefact assemblages" (Peters & Blumenschine 1998).

With the possibility of artefact assemblages created by wood cutting and nut pounding being negligible (although visible at Large Spring on the Eastern Lacustrine Plain), most of the stone tools are predicted to be remnants of early hominid scavenging of larger mammal carcasses, and early hominid hunting of larger mammals. The composition of these assemblages are said to be the result of facet-specific constraints: competition with carnivores as well as other hominids for larger mammal carcasses; and the degree of carnivore predation risk involved in scavenging (Peters & Blumenschine 1998). These constraints are, in turn, ultimately determined by the geographically ecological variability of cover abundance.

Factors of refuge and transport of carcass parts comes into consideration. The prediction made is that transport of stone and the carcass bones will be minimised by endeavouring to seek the nearest natural or artificial source of stone raw materials or stone tools, and the transport of some of the tools and all of the carcass parts to the nearest facet which would offer relative safety. This is something which Peters and Blumenschine (1998) have termed the "least-effort principle".

This minimising of distance ensures that lost and discarded stones and stone tool would be reused, but only to the degree that its locality is predictable. The clusters of stone tools that can be reused for butchery and predator defense are not found generally in unwooded localities, but are common in closed areas with a fresh supply of water nearby (Peters & Blumenschine 1998). It is unclear, according to Peters and Blumenschine's model (1998), whether stone carrying devices were used. If such devices were utilised, however, this would allow "reduce the unintentional storage of usable stone at places [the early] hominids visited repeatedly, while allowing them greater mobility untethered to secondary sources of raw material" (Peters & Blumenschine 1998).

There is a dearth of archaeologically detectable criteria for determining the presence or lack of food. Stable carbon isotopic studies are, in addition, unlikely to ever be of the calibre of determining good vs. bad refuge tree covering. Thus, according to this predictive model, it is unrealistic to expect "to determine whether observed archaeological clustering results from carcass processing by acquirers in adequately safe settings or through provisioning of dependants at maximally safe arboreal central places" (Peters & Blumenschine 1998).

Comparing the competing hypotheses

For the early hominids transport of artifacts and raw materials for artifact manufacture were an integral part of a well-developed behavioural pattern, according to Schick (1987). In summary, "this appears to have been habitual behaviour, materials intended for future use having been carried around in the course of hominid foraging and Oldowan sites of variable artifact densities having formed as a by-product of repeated stone transport behaviours. Variability among Oldowan sites in their overall size or artifact density, artifact types, and raw materials utilized, may depend largely upon which stone resources were encountered during hominid foraging rounds, variable distances of stone sources to individual sites, and the frequency of site utilization as determined by food resource distribution" (Schick 1987). Thus the stone and faunal assemblages are not seen in social terms of homes bases, but as the result of technological and ecological variables.

The main difference between Isaac's and Rose and Marshall's versions of the home base hypothesis, and the cause for Rose and Marshall renaming it their version the resource-defence model, lies in the greater body of knowledge currently available on meat consumption and processing patterns of modern carnivores in East Africa from primate perspectives (Bietti 1996). Rose and Marshall have put forward hypothetical answers to the questions posed over Isaac's home base model of why and how it happened; and, in essentially incorporating and expanding on Schick, provide the social perspective glaring lacking in Schick.

Rose and Marshall's "resource-defense hypothesis does incorporate many elements of traditional home bases, including regular use of multi-purpose sites, a mixed strategy of gathering, hunting, and scavenging, food sharing, and some division of labour, while rejecting the strict sexual division of labour, monogamy, and provisioning. It stresses ecological factors, primarily predation risk and resource competition, and, most important, hominid behavioural responses in evolutionary perspective." (Schick 1996)

Peters and Blumenschine's (1998) predictive model is designed to provide an ecological interpretative framework for future landscape archaeological work at Olduvai Gorge. It examines the differential transport of carcass parts and stone from open, dangerous areas to relatively safe localities. Up to this point their model is in general agreement with the three models outlined above. However, their model differs in suggesting that "the repeated and sequential use of a single archaeological-site-scale "central place" is plausible only if refuge trees were clustered in small groves at a small number of locations within the hominids' day range. More extensive strands of trees distributed throughout a facet and among a number of adjacent facets would dilute the spatial and temporal concentration of hominid activities and the traces thus produced. This circumstances would result in spatial distributions of archaeological traces more in accord with what we might expect from a "route foraging" or an "individualistic-feed-as-you-go" model of hominid land use" (Peters & Blumenschine 1998).


The study of the organisation of, and interaction with, the land in spatial and temporal terms by hominids is landscape archaeology at its most basic definition. It becomes more difficult as the archaeologist attempts to define his views of what our predecessors viewed as liminal space and identity in their contrasting surroundings.

It is thus of crucial importance for the investigator to define what he/she means by landscape archaeology in the context used. It can range from the spatial layout of Dutch East India Company towns and settlements, to the distribution of rock art and to tracing the transhumance movements of pastoralists. This very loose definition of landscape archaeology, therefore, permits the contrasting interpretations of the inter-relations between the artifactual remains and the land on which they were imposed. It is impossible to determine what kind of hierarchies existed with early Homo habitation and foraging groups, and how these different levels of society used and viewed their place in the landscape by comparison both with their group peers and other groups inhabiting the same landscape, i.e. the relations of production and the dynamics of interaction; and also what obvious or subtle differences, if any, existed between different Homo groups at opposite geograpgical spectrums of the Rift Valley and their interaction with varying environmental conditions and pressures.

Isaac's model was very influential in its time, being a novel attempt to explain the site formation processes of contextually related artifact and bone assemblages. However, Isaac's primary weakness was a crucial argument employed within his hypothesis: uncritically using modern hunter-gatherer ethnography as analogy for hominid behaviour 2 million years ago. Allied with this oversight hindrance is that while Isaac's Home Base hypothesis strives to explain the occurrences of the assemblages, it fails to address the question of the interconnectedness of these assemblages in the landscape; i.e. whether their distribution is random or whether there is a detectable patterned use of the landscape by the early hominids in a given archaeological stratigraphic horizon.

Schick contrasts different landscape features and the flow of stone raw materials and artifacts through them, but fails to demonstrate whether there is a relationship between the different assemblages in a particular environment or not. She also does not take into account why an assemblage was sited where it was within the different landscape; i.e. within a wooded environment would the site be in the shade and protection of the trees or would it be where the cover was less dense and more likely closer to good killing and scavenging sites.

Rose and Marshall essentially produce a revamped version of Isaac's Home Base hypothesis, but with enough essential difference for them to reconsider naming their hypothesis the Resource-Defense model. This model essentially irons out the problems highlighted above with Isaac's model, but in doing so the authors produced another critical problem of their own: it is not time-restricted. What is presented is a general model based mainly on analogies with primate behaviour and sociality. They fail to test their hypothesis on any early hominid time-period environment and the assemblages accrued. Admittedly this could be due to the lack of good excavated horizons yielded the paleo-environment detail necessary, but it does not explain why at least no qualified attempt was made mention of.

Peters and Blumenschine's predictive model attempts to reconstruct the differing paleo-environments and which animals would have inhabited them. Based on these results, they go on to make predictions about which animal skeletal parts will be most commonly found at hominid sites and which animals they would represent, in conjunction with predictions made about the variability and placement of these sites in the landscape. These predictions are based on their hypothesised model of hominid behaviour, that the early hominids were primarily non-confrontational scavengers.

The differing hypotheses of Peters & Blumenschine & Marshall are essentially irreconcilable because they start from different premises: non-confrontational vs. confrontational scavenging and the resulting patterns of hominid behaviour that arise from these competing behavioural and land-use strategies. Schick's model can potentially combine well with Rose and Marshall's by not only subjecting Rose and Marshall's model to a specific time-period test, but by explaining the movement of raw materials utilised by the hominids between the various resource-defense bases. Schick can not be reconciled with Peters and Blumenschine for the simple reason that Peters and Blumenschine view the assemblage occurrences as randomly chosen, used once and then discarded; in other words, Peters and Blumenschine's model predicts that the early hominid had no concept of central territoriality.

Projecting the concept of landscape archaeology back into the distant past is therefore problematic in some respects: a whole model can be based on a premise as equally valid as a competing premise. This task is made even more frustrating by the last of written records to tell us in greater detail about the movement of these early hominids across the landscape: could all the early hominid Homo groups have been behaving in the same manner 1.8 million years ago? Did they practice seasonal exploitation and, if so, to what extent and how exactly did that affect their movements across the landscape? What kind of hierarchical system did these early hominids have? What kind of territoriality did they practice? These are the typical archaeological questions posed for more recent eras of hominid history. In many cases, such as analysing pastorialists and pastroalism in the 18th century along and within the Northern Cape Frontier Zone in South Africa (Penn 1986), it has been proven possible to fruitfully combine historical records with the available archaeological data to produce a more well-balanced coherent hypothesis. The distant time period of early Homo dealt with here does not permit that luxury.

Further complicating the numerous problems confronted by the archaeologists and paleoanthropologists alike is that "the reconstruction of prehistoric landscape activity patterns is complicated for any time period and region by uneven preservation of remains, incomplete exposure of ancient landscapes, and difficulties of establishing contemporaneity among sites. Some of these problems are magnified in the earliest archaeological record. For example, unlike the large horizontal exposure of many recent archaeological landscapes which lie close to the modern land surface, those of the Plio-Pleistocene are severely limited in area because they are defined by a narrow and meandering strip transect along the erosional front of outcrops" (Blumenschine & Masao 1991).

Archaeological and paleontological studies into the nature of early Homo's interaction with his surrounding enviroment (which perhaps should be more rightly called "paleo-landscape archaeology" rather than be classified under the all-emcomposing term "landscape archaeology"). These studies strip away the symbolic essence of more historically recent examinations and re-apply the analytical paradigm techniques in different contexts using more appropriate investigative methods best suited to the vast deep time period in which they are operating. These differences have been outlined and explained, as well as their problems. All these are tantalising questions, beckoning the archaeologist to investigate further and deeper with ever increasing refined scientific methods of analyses. A goal which scientists will get increasingly closer to, but one which can never be fully reached and understood.

back to top of page