G. A. Clark*
Scientists have been trying to arrive at a consensus about modern human origins (MHO) for more than a century. How is it then that key questions such as whether modern humans evolved only in Africa and migrated from there or evolved in other regions across the world from local archaic ancestors remain unanswered? Many would say that we simply do not have enough data to answer the question of our origins and that with the eventual accumulation of more data, many MHO issues will be resolved. Insufficient data is only part of the answer, however. MHO researchers come from various research traditions. In each of these traditions, different assumptions about the remote human past determine what is considered relevant data, which questions are asked of the data, and how the data are interpreted. More data do not remove the paradigmatic bias implicit within each research tradition, and in consequence people from different fields fail to communicate effectively.
The geneticist Henry Harpending once expressed this problem succinctly when he described MHO research as "a highly visible, yet intangible field" (1). The disciplines that contribute to the field (archaeology, human paleontology, and molecular biology) tend to be discovery-driven and focused on methodology. Following a strictly empirical approach ("the facts speak for themselves"), they often have little concern for the logic of inference underlying knowledge claims.
Although these observations apply to all aspects of MHO research, MHO archaeology in Europe is a particularly good example of such epistemological naïvete. Like the larger debate of which it is a part, it can be summarized in terms of two competing models: the continuity model, which contends that modern humans in Europe and elsewhere evolved from their local archaic predecessors, and the replacement model, in which modern humans evolved only in Africa, migrated out of Africa, and replaced other hominids that were the products of earlier, similar radiations (2, 3).
Each of these models is based on a set of assumptions that favors some groups of variables at the expense of others, and both define and weight variables thought to be held in common differently (4). By making the tenets of the replacement and continuity paradigms explicit, it should in principle be possible to develop tests for their validity--patterns in the archaeological and paleontological records that should hold if in fact the paradigm is an accurate descriptor of reality (5). However, such a critically self-conscious approach is often lacking.
According to conventional archaeological systematics, the transition between neandertals and modern humans in Europe coincides with the Middle-Upper Paleolithic transition, 35,000 to 45,000 years ago (see the figure). Although consensus has remained elusive, a dominant "replacement" view of the European archaeological transition is embodied in textbook generalizations (see the table). The adequacy of the systematics underlying these generalizations needs to be evaluated.
A matter of timing. Comparison of the standard and the demographic
compression models for the appearance of symbolic behavior in Europe between 50,000 and
10,000 years before present (yr B.P.). The standard model argues for an
"explosion" of evidence for symbolism coincident with the Middle-Upper
Paleolithic transition, 40,000 to 35,000 yr B.P. The demographic compression model sees
change as much more gradual, with the sharp increase in evidence for symbolism occurring
only after 20,000 yr B.P., caused by demographic changes and mainly confined to southern
France and northern coastal Spain (12).
| A shift in stone tool technology from predominantly "flake" technologies to "blade" technologies, achieved by means of more economic techniques of core preparation. |
| A simultaneous increase in the variety and complexity of stone tools involving more standardization of shape and a higher degree of "imposed form" in the various stages of production. |
| The appearance of relatively complex and extensively shaped bone, antler, and ivory artifacts. |
| An increase in the rate of technological change accompanied by increased regional diversification of tool forms. |
| The appearance of beads, pendants, and other personal ornaments made from teeth, shell, bone, stone, and ivory blanks. |
| The appearance of sophisticated and highly complex forms of representational or "naturalistic" art. |
| Associated changes in the socioeconomic organization of human groups, marked by (i) a more specialized pattern of animal exploitation, based on systematic hunting, (ii) a sharp increase in the overall density of human population, (iii) an increase in the maximum size of local residential groups, and (iv) the appearance of more highly "structured" sites, including more evidence for hearths, pits, huts, tents, and other habitations. |
The European approach to paleolithic archaeology originated in France almost a century ago. It is based heavily on a typological systematics that emphasizes retouched stone tools. A cultural transition is usually demarcated by changes in the retouched tool components of archaeological assemblages, that is, the totality of archaeological remains contained in a site. The tacit assumption is made that these tools represent the remains of stylistic microtraditions akin to those produced by more recent ethnic or linguistic units like tribes, peoples, or nations, transmitted from one generation to the next through the medium of culture. Modes of tool fabrication are equated with social learning, and it is assumed that distributions of particular types of stone tools in time and space, to a degree, also define the boundaries of identity-conscious social units analogous to the peoples and cultures of history. This reasoning is then extended to modes in the frequency of entire assemblages of artifacts. However, both the spatial extent and the temporal persistence of these hypothetical entities would have been enormous, orders of magnitude beyond those of any known or imaginable social unit at the time in question. Changes in the character of retouched stone tools over the Middle-Upper Paleolithic transition have been interpreted in three ways. Some workers see the transition as an in situ phenomenon with clear evidence of continuity between late Middle Paleolithic (LMP) and early Upper Paleolithic (EUP) assemblages. Others argue that certain EUP industries such as the Châtelperronian and the Szeletian are "adaptive responses" by neandertals to the arrival of modern humans making Aurignacian industries. Although it is by no means clear what an adaptive response is, this scenario implies that neandertals modified their existing Mousterian technologies because of contact with modern humans and produced assemblages with mixed Middle and Upper Paleolithic characteristics. A third hypothesis is that no such intermediate industries existed. Contemporaneous LMP and EUP assemblages in the same site or region are taken to imply that the EUP must have been intrusive; that is, modern humans migrated into Europe and replaced existing hominid groups. This scenario implies that the creators of LMP and EUP industries can be known with certainty (they cannot) and that archaic and modern groups coexisted for extended periods of time but did not interact with one another to any substantial extent (6). These different perspectives are inextricably entangled with the classifications for Middle and Upper Paleolithic retouched stone tools. Empirical support for replacement of neandertals by modern humans rests on the notion that Middle Paleolithic stone tools do not exhibit "imposed form" and "morphological standardization" whereas Upper Paleolithic stone tools do (7). However, quite distinct and incompatible typological systems are used to characterize these assemblages. Discontinuity is thus "built into" the interpretation right from the start. Harold Dibble showed successfully for Middle Paleolithic stone artifacts that the size and shape of the starting flake (called blank) determine the form of the resulting tool and that tools previously classified as distinct types were simply part of a few, generalized lithic reduction sequence (8). The same arguments can and have been made about Upper and even Epipaleolithic artifacts (9). If we do not abandon typological systematics altogether, we should at least uncouple them from the historicist biases that are invoked to explain them. The archaeological aspects of the MHO debate turn on the expected behavior and adaptations of two different hominids occupying Europe over tens of thousands of years. Differences in adaptation would clearly be expected under all replacement scenarios published so far. But the common archaeological monitors of adaptation do not show such differences. Patterns in stone tool technology, typology, raw material variability, lithic reduction strategies, blank frequencies, bone and antler technologies, paleolithic art (see the figure), and subsistence strategies change at different rates in different places over the entire transition interval, in a manner that cannot be reconciled with biological replacement (10). In this context, some recent approaches are noteworthy. Since the late 1980s, some researchers have used a community ecology approach to understanding ancient forager adaptations. Grounded in evolutionary ecology, these workers reject ethnographic models, study paleolithic foragers as members of guilds of social carnivores (like wolves and hyenas today), and use ecological niche theory to analyze and interpret prehistoric hunting, scavenging, and foraging behaviors, which are then contrasted and compared with those of other large predators living in the region at the time (11). Paleolithic adaptations almost certainly constituted a range of options very broadly distributed in space and time, held in common by all circum-Mediterranean hominids and invoked differentially according to context. The challenge is to determine the factors that may have constrained choice among these options. Such factors probably included access to raw materials, distribution of food resources, forager mobility strategies, anticipated tasks, group size and composition, changes in occupation of a site during an annual round, and, more generally, duration of site occupation. On the surface, the voluminous literature on the MHO debate paints a picture of informed and sophisticated interdisciplinary research in which data are absorbed and digested, arguments assimilated, and methodologies understood, compared, and evaluated. I suggest, however, that this is a gross oversimplification of a much more complex reality. We are, in effect, consumers of one another's research conclusions, but we select among alternative sets of research conclusions in accordance with our biases and preconceptions. These biases and preconceptions must be subjected to critical scrutiny. As long as there is no explicit concern with the logic of inference--how we know what we think we know about the past--there can be no consensus. References
The author is in the Department of Anthropology, Arizona State University, Tempe, AZ
85287-2402, USA. E-mail: gaclark@asu.edu Volume 283, Number 5410 Issue of 26 Mar 1999, pp. 2029 - 2032
|