When first asked to comment upon the contribution of Arponen et al. on environmental determinism in archaeology – a red flag to many – I became excited that the topic is receiving attention again and, not least, that this attention is translating into printed debate. I commend the authors on their effort, also for bringing together multiple voices in their article. All too rarely do theoretical contributions translate into multiple authorship. I cannot in any way disagree with their key conclusions, namely that investigations of deep-time relations between humans and the environment are not just timely but important, and that archaeology should make full use of its rich array of data, cases and dissemination possibilities to investigate them and make them relevant in the present.
That said, it is worth, I believe, adding some commentary to the way in which Arponen et al. frame their discussion, to how they gloss over critical issues of analytical scale and purpose, and to the solution they offer. Arponen et al. acknowledge that the tired paradigmatic straw men of processual and postprocessual archaeology may no longer be of much use, but they nonetheless swiftly link processualism with environmentally deterministic positions. In referencing Arkush (Reference Arkush2011), they pitch studies of the postprocessual non-deterministic variety focusing on ‘contingency’ and ‘agency’ against those of the processual determinists focusing on ‘process’ and ‘conditions’ respectively. As much as this perceived contrast has been canonized in textbook versions of disciplinary history, it is fundamentally flawed and should be filed under ‘of historical interest only’. Archaeology is a historical science whose epistemological undergirding, data and explanatory goals mirror most closely disciplines such as evolutionary biology and historical linguistics rather than, as many processualists fatally assumed, physics or chemistry. Critically, in the historical sciences, attention to contingency, to process, to agency and to conditions is ever-present – after all, without the interplay of these key factors there would simply be no history, be it cultural or organismal. Moreover, meaningful historical causation – that is, patterns and processes of culture change over time – are only ever inferred from correlations and are only ever probabilistic. The epistemic and data-structural likeness amongst the historical sciences was already stressed by philosophers of science such as Whewell (Reference Whewell1847) and has recently been discussed again by Mesoudi, Whiten and Laland (Reference Mesoudi, Whiten and Laland2006) and Currie (Reference Currie2018b). Granted, the common institutional placement of archaeology within the humanities or social sciences has resulted in student recruitment, education and socialization trajectories as well as practices that perpetuate the opposition between the humanities and the natural sciences. It is these practices that continue to feed the spectre of C.P. Snow’s (1959) antagonistic and aggressive ‘two-cultures’ argument also referred to by Arponen et al. – but it is my belief that we really must no longer pander to this out-dated way of viewing the sciences if indeed our concern is to explain, in a replicable and transparent manner, real-world phenomena that have come to us from an actual past.
Along with Arponen and colleagues, I do not reject the notion that archaeology is conducted in the present and responds to personal, social and political concerns. To the contrary, I am very much concerned with the ethical dilemmas posed precisely by conducting research on the causal connections between past environmental change and human responses (Riede, Andersen and Price Reference Riede, Andersen and Price2016; Riede Reference Riede2018). Archaeology is an accommodating discipline and there are many ways in which to generate new knowledge of the past and just as many ways to make this knowledge do good work in the present. As Arponen et al. rightly note, being cognizant of our discipline’s role in the present does not automatically lead to a radical (and ultimately counterproductive) social-constructivist position (Shennan Reference Shennan and Bintliff2004). Furthermore, as our colleagues in history are increasingly pointing to the causal role of environmental change in cultural change over the last centuries, hard-line positions that decouple cultural, social and political histories from environmental histories ring increasingly hollow (Chakrabarty Reference Chakrabarty2009). The growing sub-field of environmental history is showing us, thanks to the commonly much more dense availability of sources, just how intricately human individual and societal trajectories, political authority, warfare and migration – all major analytical concerns also in archaeology – are intertwined with environmental changes and extreme events (e.g. Mauch and Pfister Reference Mauch and Pfister2009; Behringer Reference Behringer2010; Parker Reference Parker2014).
In suggesting a solution to the conundrum of determinism as perceived by Arponen and colleagues, they propose that archaeologists turn to the anthropology of disaster that argues for the social and historical construction of vulnerability and resilience. This literature is indeed inspiring – it has definitely inspired me (e.g. Riede Reference Riede2017; Reference Riede2014) – but it is also notoriously diverse, if not outright confusing (Otto et al. Reference Otto, Reckien, Reyer, Marcus, Le Masson, Jones, Norton and Serdeczny2017; Lorenz Reference Lorenz2013). Most available frameworks are difficult to align with historical analyses beyond description. No ready operationalizations are available, which is particularly problematic as Arponen and colleagues also claim that understanding human–environment interactions ultimately is an empirical matter. But better data will not bring salvation without robust theory and transparent methodologies. My counterproposal is therefore to look towards the powerful conceptual and methodological apparatus developed by quantitatively minded biologists who similarly struggle to understand the causal relations between the myriad environmental parameters and both past and current distributions of plants and animals, their complex interactions and their histories. The terminology of disaster-risk-reduction studies can be aligned with that of cultural evolution (cf. Brewer and Riede Reference Brewer and Riede2018), and human behavioural ecology offers a robust framework for how environmental cues shape key life-history decisions (Boone Reference Boone2002). With these ecological envelopes in place, methods providing reasoned and replicable insights into precisely which factors influenced the presence/absence of particular human groups as identified by artefactual proxies can be developed (Franklin et al. Reference Franklin, Potts, Fisher, Cowling and Marean2015; Robinson and Riede Reference Robinson and Riede2018); an integration with quantitative cultural-historical (i.e. phylogenetic) approaches is also possible. At all times, the intended scale and object of explanation must be kept firmly in mind, however. Environmental changes only ever underdetermined the historically contingent changes occurring within any given past community; many features we encounter in the archaeological record – for instance, the specifics of projectile-point shape beyond ballistic requirements – cannot be explained meaningfully with regard to climate and the environment but rather are the product of contingency and cultural transmission processes. Yet, when it comes to the core concerns and explanatory strengths of prehistoric archaeology – palaeodemography, migration, political change and warfare – ignoring the environment most likely will lead to flawed explanations. If, then, we bring the epistemology of the historical sciences and transparent and replicable methods to bear on the question of prehistoric human–environment relations, determinism actually becomes a red herring.
