By Paul Newall (2005)
Philosophy has always been involved in archaeology. This is an argument we will develop by considering the so-called New Archaeology and the debates surrounding it that have taken place over the past thirty years or so.
Archaeology is generally taken to have become established in the nineteenth century with Jacques Boucher de Perthes' discovery of chipped stones in Somme river gravel quarries, alongside the bones of now-extinct animals. He interpreted the former as human artifacts (such as hand axes) and hence claimed that humans had existed for much longer than Biblical accounts apparently allowed (due to the inference that they appeared much older than anything mentioned in the Bible). Although there was some skepticism to begin with, the antiquity of humanity was accepted soon thereafter. Notice, of course, the inevitable speculative dimensions to this early conclusion: how do we know when we have a genuine artifact and when a mere product of nature? How do we date them?
One of the reasons for the relative ease with which archaeology took root was its association with two other intellectual currents. Firstly, James Hutton and later Charles Lyell had investigated rock formations and used stratification to support a principle of uniformitarianism, according to which conditions in the past were the same as (hence uniform with) our own, enabling us to infer things about the past from current geological arrangements. It was also argued that the processes that could account for stratification are still in operation today, such that the Earth had to be far more ancient than was otherwise believed.
Secondly, Charles Darwin's On the Origin of Species was published in 1859 and also implied a lengthy process by which modern humans evolved, giving archaeologists the chance to look for signs of this in the material remains. Evolution hinted, too, that perhaps cultures developed in a fashion similar to plants and animals, with this possibility influencing anthropology and (later) social theory. The confluence of these three strands helped archaeology move from speculation (as it was denigrated by some) to the firmer foundation of the assumption of antiquity and the principles of uniformity and evolution. Moreover, as the world became smaller with travel over long distances the new peoples discovered were beginning to be investigated by curious scholars, especially given the (widespread) assumption that so-called "primitive" cultures could provide us with insights into how our own culture had changed over time. The idea that progress from savagery to civilisation was possible and had occurred in the past again informed social theories and suggested that we could come to understand the way we are today by searching for our origins in the archaeological record. The discoveries made in Egypt by Napoleon's teams and the increasing excavations in Mesopotamia also provided archaeologists with plenty of new data on which to test their ideas, their efforts motivated in large part by Biblical accounts and the Ancient Greek poets.
If early archaeologists had been content to classify the remains they found and attempt to build a chronology from them, the influence of V. Gordon Childe meant that questions were being asked about why collections of artifacts were located in one place and not another, or what it meant to assume that such a collection implied a group had existed there previously. In short, archaeologists began to realise that a chronological ordering - even if "correct" (whatever that means and however we would determine it) - would tell us nothing about the past unless an interpretive step was added. In 1949 radiocarbon dating (C14) was invented by William Libby and suddenly archaeology seemingly had a scientific basis to back up the placing of finds in historical order, although it took some time for the consequences to become clear. Now archaeologists apparently had an independent means to determine the age of a site without needing to resort to written records (that may not exist) or comparisons with other cultures, removing speculation altogether and providing a firmer footing for chronologies. Other scientific techniques were employed, including chemical analyses, and soon the number of methods involved exploded. Unfortunately for this spirit of optimism, however, the surety of radiocarbon dating soon gave way to arguments over its application, validity and the approach as a whole. We will return to this controversy in due course.
As a result of these scientific aids, some archaeologists claimed that the discipline was no longer plagued by quesitons of dating and began to be dissatisfied with the conclusions that were drawn by their colleagues. After all, providing a timeframe for a site was one thing but it became more important to explain what had happened. Gordon Willey and Philip Phillips suggested a processual approach, wherein the archaeologist should look at the processes involved in the histories of cultures. This is where a major turning point in the philosophy of archaeology occurred: impressed with the superior epistemological model that science appeared to provide, some sought to set archaeology on a surer footing by reproducing the supposed "scientific method".
Following the lead of Lewis Binford, several archaeologists in the late 1960s began to argue for what came to be called the New Archaeology. Inspired by developments within the philosophy of science, they wanted to do more than just describe and believed that genuine explanations could be achieved by changing direction in archaeology. In the past, they claimed, archaeologists had made inductive inferences, collecting pieces of evidence and try to infer conclusions from them. There was (and is) a significant problem with this, however: given that the archaeological record is incomplete, how can inferences be accurate? One possible response is to wait until all the evidence is in, but this is impractical (or rather impossible, if we accept that the record must inevitably be incomplete); another is to give up making inferences at all, yet this leaves archaeology as solely as descriptive enterprise. In general, this was the well known problem of induction in action once again.
Nevertheless, according to the New Archaeologists the alternative was to adopt the methodology of science and formulate hypotheses, deriving their consequences deductively and - most importantly - testing them. This model, based largely on Hempelian deductivism, was a form of logical positivism (explained here). The New Archaeology also included functional approaches, wherein generalisations are made about changes in political, social and economic systems and how these cultural processes can aid in explanation - hence Processual Archaeology, another term often used in discussion.
Whatever the merits of the New Archaeology, one thing at least was clear: archaeology could never go back to what has since been called its "state of innocence". In order to practice archaeology it would be necessary to question presuppositions that had previously been implicit and reject them if required, as well as to consider the impact that the philosophy of science would have as debates therein changed the image of archaeology that had been crafted to date. This did not mean that philosophy came to archaeology from without, but rather that it had always been involved (although previously with little or no appreciation of the impact of assumptions). As Alison Wylie put it, "[w]hat you find, archaeologically, has everything to do with what you look for, with the questions you ask and the conceptual resources you bring to bear in attempting to answer them." This was a familiar point in science, especially in physics thanks to the philosophical investigations of Einstein and others following the advent of quantum theory, but now it would affect archaeology, too.
Although the New Archaeology introduced several currents, including the realisation that perhaps the best way to arrive at explanations would be to study societies of today, hence the advent of ethnoarchaeology with the adding of an archaeological aspect to ethnography. Even so, no sooner had archaeology seen an influx of positivistic thinking than criticisms in the philosophy of science began to chip away at the influence of the Vienna Circle thinkers and the earlier confidence began to look jaded. The most famous work of this period is probably Thomas S. Kuhn's The Structure of Scientific Revolutions, but others can justifiably said to have played an even greater role. Then, as now, there were some archaeologists who paid little attention to theoretical debates and believed (along with some physicists and biologists) that it is possible to do science without worrying about the tortured timewasting of philosophers. This seemed especially so for field archaeologists, apparently far removed from debates in academia. We will see, however, that this separation becomes increasingly difficult to sustain.
Criticisms were made of the New Archaeology from several directions. Firstly, the philosophers of science N.R. Hanson and Paul Feyerabend discussed the supposed objectivity of science, one of its greatest assumed virtues, via the concept of theory-ladenness. They stressed that there can be no neutral observation statements; in archaeological terms, this is to say that data recovered from sites do not form a class of "facts" independent of the observer (or archaeologist) but are already filtered by us. Indeed, this insight was later developed by Paul Churchland in neurological terms such that this inevitable layer (there might be several) of interpretation is what makes our perception cognitive in the first place. The earlier example of Jacques Boucher de Perthes' is a case in point: he did not unearth human artifacts but some stones he understood in this way. Through this reading of the ostensibly objective data, they became evidence for the claim that humans had existed for far longer than had previously been thought. There was nothing about the stones themselves that made this conclusion a certainty, however; on the assumption that a natural process could occasion the same results, say, he might have scarcely given them a second glance. On the contrary, though, he came upon the stones with many preconceptions, not all of them explicit, including that stones cannot be weathered in this fashion; and so on. Today we might consider this a commonplace but the point is that we do not just arrive at virgin archaeological data but unavoidably bring our other ideas with us, without which we could not make sense of anything in the first place. This seemed catastrophic for the goal of objectivity.
Secondly, and again via the philosophy of science, Ian Hodder and others emphasised that interpretations of archaeological date are underdetermined; that is, many readings are possible (some would say an infinity of them) for the same set of data and hence our choices between them must rely on extra-empirical factors. This is to say that appealing to the evidence alone is inadequate to account for our decisions (hence the death of the more naive forms of empiricism - see here for more detail) and so we inevitably go beyond it in arguing for our conclusions. Much like theory-ladenness, this may seem obvious with the benefit of hindsight but it is important to remember that at the time some philosophers of science and New Archaeologists with them were contending that the sciences could avoid the arbitrary and in particular solve the demarcation problem, or how to separate science and non-science. There were and is no shortage of people claiming that this task can be done, in spite of the difficulties with the criteria proposed. What theory-ladenness and underdetermination did not mean was that archaeology - along with science as a whole - would have to descend into relativism, that poorly understood bugbear and the stuff of rationalist nightmares, even if drawing a boundary between archaeology (and history, too) and mere stories became somewhat troublesome.
If objectivity was problematic, that did not mean that it could not serve as a rhetorical trope. More importantly, though, its apparent demise led other archaeologists to consider what else had been lying hidden in their discipline, unexamined. This helped bring about what came to be known as postprocessual archaeology, in spite of there being much debate over whether the critiques of the New Archaeology could properly be said to have superseded it or rather complemented it. Nevertheless, neo-Marxists and others emphasised that the responsibility of the archaeologist should not just be to describe or explain the past but to use whatever they gained from both to make positive contributions to the present, striving to make the contemporary world a better place. For them it was not enough to promote archaeology as an objective science going about its business while politicians worried about poverty and power structures. In a manner similar in some aspects to Feyerabend previously, archaeologists such as Christopher Tilley went further and suggested that science is itself part of a system of economic and social hegemony, set up as an ideal without sufficient inquiry into whether or not its benefits would outweigh the costs to the individual or whether scientific (and thence archaeological) work actually aided power structures throughout the world. Once again, this was a criticism of the supposed objectivity of science by pointing to its (potential or actual) consequences, especially for the poor and disadvantaged. At the very least, archaeology led to some difficult questions that continue to be asked today. What happens, for example, when some remains are found on land ostensibly "belonging" to native inhabitants of a country? Do archaeologists have the right to investigate? What if the tribal group, say, forbids unearthing anything? That instances of just this issue have occurred recently (such as Kennewick man) show that archaeology is not able to pursue a neutral path, avoiding political and cultural conflicts, particularly when its results may impact on the wider world and social debates.
At any event, one of the results of the debates in the philosophy of science and in postprocessual archaeology was widespread agreement that there is no such thing as the "scientific method". Just as physicists do not behave as biologists or geologists do, with significant differences within these disciplines, too (compare organismic and molecular biology, say, or condensed matter and particle physics), so do archaeologists employ a range of methods. This does not mean that "anything goes", possibly the least understood argument in the philosophy of science, but only that the myth of science as a unified epistemology has had to be discarded in favour of a far more nuanced appreciation of what scientists actually do - aided, in a nice irony, by sociological investigations of their actual behaviour. It has also not stopped the so-called scientific method being used rhetorically to deny credibility to some ideas, in legal as well as public discourse, but it is interesting to consider how these issues impact upon archaeology.
We have looked at some of the philosophical difficulties faced by the New Archaeology and now we will expand on them in turn, considering the problems confronting archaeology and how the New Archaeologists proposed to solve them. In so doing we will again come to appreciate that philosophy was not a distraction from the business of archaeology proper but an inevitable part of the discipline that could not be ignored.
As we have seen, developments in the philosophy of science had led some archaeologists by the 1960s and 70s to question what archaeology is and how its practice should be understood. Note, however, that the clean break with the past suggested by the very name for this so-called movement – the New Archaeology – has been subject to skepticism itself. In 1955 B.J. Meggars had written of "the coming of age of American Archaeology" and raised many of the same philosophical issues that were discussed at length in the following decades, while criticism of the supposed foundations of archaeology was already well advanced in the late 1930s and the 40s in the writings of Kluckhohn and Bennett, amongst others. (See Wylie's How New is the New Archaeology? for an extended analysis.) Indeed, it may be that the view of the New Archaeology as a revolutionary break with archaeology thus far was (and is) itself influenced by currents in the philosophy of science, particularly Thomas S. Kuhn's famous The Structure of Scientific Revolutions. The most forceful objection to Kuhn's ideas (due to Lakatos, Feyerabend and others) was that the periods of "normal science" – in which scientists working within a paradigm are resistant to anomalies until it is finally toppled after the fashion of a revolution – never really existed in the first place. In archaeological terms, this is to say that there was no period of philosophical naivety wherein archaeologists ploughed ahead with scant concern for wider intellectual debates, but rather a continual internal dialogue and questioning of assumptions.
In any case, the problems with what was called traditional archaeology were threefold: two philosophical (specifically epistemological) and one methodological. Firstly, the New Archaeologists complained that the traditional version relied on a form of empiricism that was hopelessly outdated, which involved only the observable (i.e. archaeological data) and systematising it. This had been shown to be untenable. Secondly, and resulting from this, the study of cultures had to be restricted to inferring motivating beliefs from material remains. After all, people of the past are unobservable, too, and hence if archaeology were limited to the narrow empiricism the New Archaeologists opposed then it would be impossible to set anthropology on an archaeological basis. Lastly, and perhaps most significantly, the belief that these were methodological restrictions meant that archaeologists could not go beyond the resulting form of archaeology, whether that was considered to be antiquarian cataloguing or unbridled speculation. Where a method is assumed to be fixed, of course, it is not long before its advocates claim that their investigations are therefore neutral and use their "method" as a rhetorical tool to deny recognition and support to alternative ideas not following it.
It is worth dwelling on these concerns because it was not that empiricism was the problem, or philosophy on general, but bad philosophy. No one lamented the intrusion of philosophy but rather that concepts that had been shown to be flawed by the philosophers were still alive in the social sciences. As Kluckhohn wrote in 1939, it was not a case of choosing
… between theory and no theory or a minimum of theory, but between adequate and inadequate theories, and, even more important, between theories, the postulates and propositions of which are conscious and hence lend themselves to systematic criticism, and theories the premises of which have not been examined even by their formulators.
This indispensability of theory combined with the necessity of continually challenging and reassessing the concepts we employ in the sciences was also recommended by Einstein. When archaeologists or scientists at large proceed as though philosophical issues are irrelevant to their work, then, they do so not because they have examined their method at length and found it to be neutral but through ignorance and without realising that the assumptions we bring to inquiry can influence what we find. This insight was formalised as the problem of theory-ladenness: when the clear distinction between facts and theories failed, the New Archaeologists insisted that their discipline could no longer be viewed as the collections of "facts" and that the presuppositions that implicitly guide research should be stated openly and questioned forcefully at every available opportunity. Moreover, archaeologists would have to recognise that they are involved in their investigations rather than passive collectors of these "facts". Indeed, from the 1930s an increasing number of critics were noting that the huge volume of data accumulated was not matched by a richness of interpretation.
A nice example of a debate entered into by the New Archaeologists (although it was being addressed well beforehand) was that surrounding classification. Are the categories used by archaeologists to sort material – systematising it – inherent in the remains or just instruments to help us make sense of it? This is a particular instance of the problem of universals, an issue in metaphysics that has been discussed for thousands of years. When archaeologists develop taxonomies by which data are sorted, are they arbitrary or are they to some extent forced by a "natural order", so to speak? For the New Archaeologists the claim that classifications could exist with or without archaeologists to employ them was just another instance of the idea that archaeology was a neutral science with the preconceptions of its practitioners not affecting the conclusions reached, as opposed to facts unavoidably being theory-laden.
For the New Archaeologists, then, traditional archaeology was crippled by philosophical problems. In order to avoid the complaint (often levelled at them) that the discipline was thus entirely subjective, they sought to replace na�ve empiricism with a far stronger theoretical basis. This could be achieved, they thought, by accepting that interpretations of the archaeological data are underdetermined but that nevertheless a form of empiricism can be used because the evidence can be appealed to when judging between different hypotheses. Data could then be admitted to be theory-laden but still used to test claims – or so they hoped. In addition to their criticisms of traditional archaeology there was also an effort to develop alternatives, including an argument that archaeology should move away from descriptive accounts and provide explanatory ones. It was here that the New Archaeologists appealed to the work of the logical positivists, especially C.G. Hempel, insisting that an archaeological explanation must be governed by laws. For Binford, this meant going beyond the particular to the general, aiming ultimately at understanding "the total range of physical and cultural similarities and differences characteristic of the entire span of man's existence".
As we have noted, however, this was occurring just as positivist ideas in the philosophy of science were beginning to crumble. When he appealed to a Hempelian form of explanation, in which an observed event is said to be explained if it fits an already established regularity such that we should have expected it to happen, Binford did not realise that he was invoking exactly the restrictions of na�ve empiricism by relying on observed regularities. In short, his version of positivism was as bad as the philosophical approach he had objected to. Moreover, Binford advocated Hempel's hypothetico-deductive model – that is, offering an explanatory hypothesis, deducing its consequences and testing for them – even though it was quickly shown to be flawed. The most common example used in this context is a proposed law such as "all swans are white"; if true, it would follow straightforwardly that all swans are white, but in order to confirm this we would have to check all swans – so it can never be confirmed deductively. In archaeological terms, this would be much like explaining the collapse of civilisations by a specific set of circumstances: in order to confirm that we have a law, we would have to check all civilisations anywhere and at any time. The hypothetico-deductive model can thus only apply in restricted (usually trivial) domains. Perhaps even more significantly, though, even if it were possible to arrive at laws explaining present conditions and behaviour, to infer anything about past cultures from the (relatively scant) material data available to us requires inductive steps – precisely what the New Archaeologists were supposed to be moving away from.
There is a rhetorical dimension to the New Archaeology that is worth considering, too. It has been suggested that moves toward positivism in archaeology was but an extension of a larger endeavour on the part of naturalists to make the social sciences harder, invoking models that might apply to physics, traditionally the "hardest" science of all (hence the conceit that its methods are the ideals to be aimed at everywhere else). This was a reaction to a widespread concern that Enlightenment ideals of reason and civilisation were under threat from relativists and subjectivists, and particularly philosophers of science who claimed that science was irrational and no better than astrology or voodoo. These charges were – and are – empty but it is rare to lose money betting on how zealously people will defend simplistic models of science if civilisation itself is alleged to be under attack and losing ground. (Indeed, it is not difficult to find instances of this behaviour today in several contexts.) Nevertheless, by placing the social sciences on a positivistic base it was hoped that they could be saved from degeneration and contrasted with so-called pseudoscience and groundless speculation. The great irony was – and again, still is – that those who so desperately wanted to improve on a naive and untenable empiricism in order to counter a crude version of archaeology (and science in general) and to oppose "cranks" resorted to a positivism that was just as unsophisticated and relied on just the same uncritical empirical foundations. The problem was that too many archaeologists were caught up in a (false) dilemma that defined the rhetorical backdrop to the debate: either archaeology had to follow the other sciences and be rigorously recast along positivist lines or the entire game was lost and would unravel into skepticism. This is a tactic that – once more – is still employed today for much the same reasons, but a bad idea does not become a good one just because the only alternative provided is supposed to be even worse.
Meanwhile, the insistence that archaeologists should seek explanations rather than descriptions led to a great deal of discussion concerning what form such explanations should take. While New Archaeologists were advocating a Hempelian theory of covering laws, critics like Merrilee Salmon came up with counterexamples in the form of generalisations that would cover a phenomenon but not explain it. For instance, men who take birth control pills do not get pregnant, but this is an explanation we would not consider. Likewise, prisoners who are deprived of writing materials do not reproduce verbatim the works of Shakespeare. Why not? There are (implicit) generalisations here that cover and yet we do not accept them as explanations. We thus see that covering is not enough; it is necessary, we might say, but not sufficient and on its own gives us no guidance as to whether or not the true explanation has been found. Some theorists responded by appealing to higher-level regularities, but this led to a regress problem because resorting to a presumed regularity of a different order to explain which of several possible regularities was the correct one led to the same question on the new level, and so on. Indeed, whenever we can explain a phenomenon because we have good (empirical) reasons to suppose the existence of the causes invoked by the explanation, we can always ask where these causes came from and how they operate – requiring explanation all over again.
Perhaps the most important aspect of traditional and New Archaeology and the criticisms of both is that everyone involved (along with others in the philosophy of science) was directly or otherwise addressing the questions of what science is and what it aims at. Some opponents of positivism advocated a realist perspective, which they thought could improve on the philosophical dead ends, and hence we come full circle to having to understand the role of philosophy and the part it invariably plays in the discussion and practice of archaeology.
- Binford, L.R., In Pursuit of the Past: Decoding the Archaeological Record (Berkeley: University of California Press, 1983).
- Feyerabend, P.K., Against Method (London: Verso, 1975).
- Hodder, I., Archaeological Theory Today (Cambridge: Polity Press, 2001).
- Kuhn, T.S., The Structure of Scientific Revolutions (Chicago: University of Chicago Press, 1962).
- Renfrew, C. and Bahn, P., Archaeology: Theories, Methods and Practices (London: Thames & Hudson, 2004).
- Wylie, A., Thinking From Things: essays in the philosophy of archaeology (Berkeley: University of California Press, 2002).