The limits of uniformitarianism
Whilst the uniformitarian principle is of importance in the analysis of the mechanistic processes of formation that contributed to the form of their archaeological residues, we might wonder if it is of similar importance for gaining an understanding of the historical conditions of the various kinds of humanity that have also engaged with that material. Because the uniformitarian principle delivers the belief that formation processes occur quite generally and can be read back from the predictability of their residues, this also implies that processes that once operated only locally and in historically specific ways cannot be read off from the record. This has resulted in the assumption that the archaeological record is limited simply because key aspects of earlier, historical conditions are represented in ways that are unpredictable. For example, Margaret Smith once asserted that ‘[i]t would obviously be impossible to understand the relics of the Trobrianders from the evidence of the material remains alone: they don’t even act rationally, on standards we can comprehend, when engaged in growing their subsistence crop’ (Smith 1955, 5; cf. Chapman & Wylie 2016, 16).
Upon this basis it appears that the very things in which we might be most interested, namely the specific ways that others once lived and which diverged from our own ways of living, lie beyond the possibility of archaeological understanding (cf. Hawkes 1954), such that our understanding of what it might be to act ‘rationally’ is quite different from that which was employed by those whom we study.
In negotiating the uncertainties that link specific historical processes to the contemporary data, Adrian Currie has advised that the broad scope of these uncertainties cannot be resolved by any single historical method. In his view the historian/ archaeologist should therefore abandon any commitment that they might hold to the ideal of a single method of analysis. I will assume that Currie means us to question the assumption that all archaeological data must be treated as if it recorded the processes of its formation. Instead Currie advises taking a more pragmatic approach, one that develops methods more concerned with what it is that we want to know and that are sensitive to the historical context being analysed (Currie 2018). Thus, for data to become the evidence by which we might gain an understanding of more than the mechanical processes of its formation, we will need to find a way to use that evidence as the evidential traces that are specific to the historical processes that interest us (cf. Wylie 1989).
Lewis Binford recognised that the application of uniformitarian assumptions should enable him to resolve the problem of determining the formation processes that had resulted in the bone debris recovered from a number of French Palaeolithic cave deposits (Binford 1983, 95-108). The predictability offered by the uniformitarian link between the formation process and its resulting faunal deposit meant that the proposed cause-effect relationships could be tested by contemporary observations on environmental conditions that were similar to those that had occurred during the European Palaeolithic. This became the essential step that Binford took towards establishing archaeology as an experimental science, whereby the validity of observations on static residues could be tested by their replication in conditions that were available to contemporary observations.
There is no doubt that such work has had considerable impact on archaeological interpretation, with the greatest impact of this so-called ‘middle range’ analysis being on our understanding of early hominin behaviour, and on hunter-gatherer archaeology more generally (Binford 1981a; cf. Binford 2001). In the classic example, analysis has allowed patterns of debris resulting from animal behaviour to be distinguished from the patterns that might have resulted from human behaviour.
The South African anthropologist Raymond Dart asserted that he could link a number of animal bone deposits to the activities of very early hunters. By this means Dart claimed to identify the presence of early humans in the archaeological record, not by the presence of fossilised skeletal remains of hominin, but by the apparent presence of residues resulting from behaviour that he regarded as being typically human (cf. Binford 1981a, 13). Dart asserted that the relevant deposits were characterised by the discarded debris of early hominin (australopithecine) who were meat eaters and therefore hunters, who had butchered their kill and transported selected joints to prepare and consume at a home base, and who had then worked some of the residual animal bone as a source for new tools and weapons (Dart 1949). The humanity of the australopithecine was thus attested for by the claimed behaviour of maintaining a home base, sharing food, and by tool manufacture, with all these factors coming at a key point in the evolutionary line that led ultimately to modern humans. This appeared to present hunting and meat sharing, along with tool manufacturing, as if they were key behavioural attributes leading to the evolution of modern humans (cf. Ardrey 1976). Obviously, the link that Dart proposed between australopithecine behaviour and animal bone deposits cannot be replicated today, but it is possible to link the formation of contemporary bone deposits with the behaviour of contemporary hunting and scavenging animals. It was upon this basis that C.K. Brain demonstrated that the structure of the fossil bone deposits that Dart had identified as resulting from australopithecine behaviour actually matched the structure of bone deposits found today in the dens of hyena. By comparing contemporary scavenger deposits, which included gnawed and split bones, with the fossil record, and by also considering the hunting behaviour of large cats, in particular the bone debris resulting from leopard kills, Brain successfully challenged Dart’s behavioural assumptions and relocated the early hominin populations from amongst the hunters to amongst the hunted (Brain 1981).
Binford stressed the need for archaeologists to recognise that complex, tapho- nomic mechanisms, extending from human, animal, and plant activity to processes of chemical and mechanical erosion, will have contributed significantly to the patterned palimpsest of residues that are the object of archaeological classification. Binford’s criticism was that archaeologists had been far too ready, as in the case of Dart, to assert the significance of deposits. This had resulted in archaeological categories of material, such as the assemblages of bones and artefacts, being claimed to derive from so-called ‘living floor’ deposits that supposedly attested to human behaviour, when in fact these assertions, and the categories that they sustained, were based upon prejudice rather than testable observations. So-called ‘living floor’ assemblages, for example, have been treated as if they represented the results of short periods of human behaviour, a fossilisation of momentary acts akin to a kind of‘Pompeii premise’, rather than, as Binford argued, a palimpsest of complex and ongoing material transformations that had extended across lengthy periods of time (Binford 1981a, 1981b, & 1982; cf. Schifier 1976 & 1985).
Uniformitarian processes are obviously at work in the world. Hyenas, wherever they might be encountered, will reduce the scavenged carcases of a bovine to a similar pattern of residues. The physical reduction of a flint core by means of soft hammer percussion will generate a similar range of flake debitage whenever and wherever it is executed. We might refer to all such processes as being mechanical. The American philosopher Charles Sanders Peirce offered a typology' of signs in which one thing is taken to indicate, or to represent, another when observed by a particular interpretant, and he termed the mechanical relationship between the thing acting as a sign, and what it represented, as an indexical relationship. The classic example of an indexical relationship is that of the weather vane’s response to the wind’s direction, where the sign of the weather vane’s direction is mechanically determined by the direction of the wind. The uniformitarian principle therefore works when the cause has a mechanical consequence, and the indexical reading of archaeological deposits remained Binford’s focus for building a methodologically sound process of archaeological interpretation (Binford 1987).
It would be easy enough to present the arguments of Binford as if they provided the basis for an archaeological methodology that calibrated archaeological data indexically and thus ensured that archaeological data were secure in being taken to represent a particular process. However, the uniformitarian principle implies that past dynamics are only securely recognised because they also operate today. If we accept that archaeological data are the product of historical dynamics, and that archaeology should offer not only an account of the formation of those data but also an understanding of how the historical processes had arisen to bring that material into being, then the idea of understanding the cause and effect relationship proposed by Binford presents us with two problems.
First, we obviously have no way of seeing those processes that did not result in a material record, nor those processes where a non-uniformitarian principle links them to a material residue. This is hardly a trivial problem, and, as I have already indicated, it has been taken to limit the relevance of any archaeological analysis of historical processes. If, for example, processes that are of central concern to disciplines neighbouring upon archaeology, such as the ways that kinship systems have tended to operate in anthropological studies, cannot be securely recognised archaeologically, despite the fact that various kinship practices must, presumably, have left material residues, then archaeological analysis is likely to be treated as of marginal relevance to anthropology.
The second problem is that archaeology is surely more than a matter of establishing that currently familiar mechanisms once operated in the past. Frere, for example, believed that what mattered for him was the dramatic implication for human antiquity of his observation that humanity had produced artefacts buried beneath a sequence of geological materials, indicating that such a humanity had existed a very long while ago. Thus, whilst the mechanisms that have transformed the material conditions that comprise an archaeological record are recognisable, archaeological work, if it is to have any value, should surely tell us something that we might not previously have known and allow us to understand the nature of historical conditions that are otherwise unfamiliar to us. If we were to characterise an enquiry into the archaeological record as an attempt to understand why certain recognisable processes happened in the ways that they did, then those processes must include the activities of humans whose activities have long been regarded as historically and culturally specific. Archaeology' thus encounters the diversity, complexity, and unpredictable messiness of human history', and if methodological security simply depends upon uniformities linking past processes with their surviving residues, then it remains unclear how we might understand the residues generated by historically specific actions, and by human motivations and intentions that might seem distinctly ‘odd’ to us today.