In a recent intervention, Romanian philosopher and essayist Andrei Pleşu (Reference Pleşu2018) writes on the topic of destiny,

Archaeology is a discipline in the privileged position of engaging with things when they have seen their end lives and conclusions, at the end of the ‘spectacle’. The downside is that sometimes too much time has passed, and traces have got lost, while at other times we forget that any story had a development, alongside moments when things could have turned out quite differently. Thus, upon trying to interpret change in the past, we sometimes end up with what Arponen and colleagues tackle in their article, namely deterministic explanations. Their article raises some points directly related to the implications of a particular kind of data set – palaeo-environmental studies – for framing historical explanations. This range of studies has become more important in recent years, as part of a wider resurgence of scientific technologies applied to interpreting the past. This trend has been accompanied by important implications, revealing the problem of integrating data sets of different kinds, from natural sciences to social sciences towards explaining historical processes. As the authors highlight, most often the explanation proceeds by identifying patterns in different data sets, climate record and archaeology, which are then correlated, and if they match they are interpreted in a causal key. But is life that simple?

It is great to see an article critically discussing this important topic. Even though it is hard to discuss environmental projects in the abstract, as I am sure that they vary widely in design, in what follows my commentary will briefly elaborate on what I take as the two key points of the argument: (1) the peril of too-simple explanations in archaeology, and (2) the scale on which we want our questions to operate. I take the authors’ ideas more as a starting point for reflection, because at times it is not clear what position they support in relation to the themes discussed, e.g.: ‘Roberts et al. would presumably also accept biologism as a basic tenet of scientific thinking … yet the precise way the thresholds impinge upon human life is still up for debate’ (p. 5) (but we never hear if the authors agree with them or not). Similarly, ‘We bring up this branch of anthropology here not because we ourselves are champions of that perspective, but in order to provide a clearly contrasting picture of how to conceive the human–environment relationship’ (p. 7) leaves the reader wondering about what the authors actually support and suggest.

First, the main argument of the paper revolves around the issue of whether palaeo-environmental research is inherently deterministic or not: ‘We write as an interdisciplinary team of researchers working in a joint research project, and we are interested in gaining clarity about what it might mean, if anything at all, to be a determinist’ (p. 3). As a case study, the authors look at how palaeo-environmental research is designed, being ‘typically oriented to detecting certain parallels between societal and environmental changes’ (p. 3). While I agree with the authors in their diagnosis, I think the path they take to peel off the layers of the problem traps them in a yes–no game; that is, are environmental studies deterministic or not? How do they get here? From the start they frame the discussion within the two-cultures divide, and then they go through the processualism–postprocessualism debate, situating it within a distinction between conclusions of different kinds, i.e. comparative versus exploratory. Coupled with the topic of thresholds – ‘the gist of the determinism debate … exists precisely where, along the continuum from external influence to internal societal dynamics, to locate the moving forces of prehistoric transformations’ (p. 4) – this leads to an inevitable conclusion that it is up to the empirical data to evaluate whether in every particular instance the palaeo-environmental causal explanation is enough, or if it is deterministic: ‘At its heart, the question of determinism thus remains an empirical question, not an a priori charge levelled against a particular study’ (p. 5). However, I feel that this is not the most helpful path to take when framing or evaluating our investigations, because it says nothing about how an ideal project should work. At the same time, ‘empirical’ is a concept which is not really attainable: e.g. we know when the Soviet Union fell; many have witnessed it; we have economic, geographic, political data; and yet there is still disagreement among specialists about what best explains this historical moment.

So the issue does not lie with the quality of evidence alone, although this matters, of course, and neither with whether climate or environmental data are intrinsically deterministic, but instead the problem is when we deal with too-simplistic hypotheses. In other words, the problem is when disciplines meet, and a data set is brought in to shed light on another data set; when palaeo-environmental data (or any other data set) provides a reductionist account (see also Hulme Reference Hulme2011 for an interesting overview of climate reductionism). Philosopher of science Adrian Currie (Reference Currie2018) writes about what he calls ‘one-shot hypotheses’: ‘Such explanations identify a particular trigger for an event and then attempt to explain as much of the phenomena at hand as possible in light of that trigger’. Therefore, the problem is when complex biological, environmental and cultural interactions are reduced to one-shot hypotheses. Underlying this phenomenon which reduces complexities to simple A → B causal explanations are several elements, but one of the important ones is a power imbalance in the academic arena, a crisis of authority of the humanities. To this Mike Hulme (Reference Hulme2011, 245) refers as ‘a hegemony exercised by the predictive natural sciences over contingent, imaginative, and humanistic accounts of social life and visions of the future’.

Hence, maybe a more productive way forward is not to focus on which is the better explanation – environmental or social – and instead to design more complex interpretive models, which allow for multiple factors to be integrated. At the same time, especially for the prehistoric past, we need more data points and precise dating techniques that would allow for refined connections.

The second point raised by Arponen and colleagues regards the value of anthropological insights in framing human–environment relations, and here as well their position is also sometimes vague (‘We bring up this branch of anthropology here not because we ourselves are champions of that perspective’ (p. 7)). Of course, this is a necessary component in understanding any past social phenomenon. In this way we start fleshing out explanations which are actually more plausible, as they include human agency and power relations: who created the environmental crisis? Were humans responsible to any degree? Are certain groups more likely to be affected? And so on. From the examples given in the paper, from Mayan droughts or UN Climate Change Resilience reports, to research into preindustrial activities and their climate impact (see an interesting piece in The conversation by Karl-Heinz Wyrwoll (Reference Wyrwoll2012) on ‘How Aboriginal burning changed Australia’s climate’), what appears as a simple ‘natural’ phenomenon is quickly shown to be the result of complex nature–culture interactions.

This is not only a matter of ‘common sense’, but also a matter of explicitly recognizing that taking into account human agency is necessary if we are to reach any kind of historical understanding. As Artur Ribeiro (Reference Ribeiro2018, 116) writes on the margins of the example of Dust Bowl farmers’ migration to California in 1942, simple law-like explanations like Carl Hempel’s ‘people will tend to migrate to regions which offer better living conditions’ is not just too general, but also doesn’t really explain who migrates, and why to California. That is because human responses are always also cultural responses and understanding requires taking into account human intentionality. Consequently, the question which springs immediately to mind is: shouldn’t we then change the viewpoint, and the scale, of our questions from the large-scale processes to human scales? The point of view which takes into account intentionality is one operating at a human scale, of bottom-up, refined case studies, whereas climate data tend to operate on large spatial–temporal scales. This is not a matter of opposing big-data narratives to small-scale case studies, but of grounding general environmental phenomena in the local human responses as a way past reductionist frameworks.

Thus, in response to the paper by Arponen et al., interdisciplinary archaeology projects require us to follow a two-step process: (1) more refined methodologies, with more data points, and complex interpretive frameworks instead of ‘one-shot’ explanations, and (2) a shift in points of view, by bringing environmental data into anthropological frameworks. In this way we might avoid a fatalist perspective (i.e. droughts came, civilizations collapsed, the end), and instead we recover something of that ‘epic’ structure of past destinies that Andrei Pleşu was referring to.