We would like to begin by thanking the journal and the commentators for their time and attention.

For us, the comments to our paper illustrate a certain diversity pertaining to how the scientific field positions itself regarding environmental determinism and connected issues. A discussion of this diversity will lead us to revisit some of the key themes of our paper in the context of the comments.

For example, in the responses of Kristiansen and Riede, we see a bold embrace of something one might term archaeological science (Killick Reference Killick2015). From that point of view, the idea of ‘social and historical construction of vulnerability and resilience’ is something that – while inspiring – has ‘[n]o ready operationalizations’ so far available in archaeology (Riede, p. 18). In a comment like this we see that, for Kristiansen and Riede, archaeology is fundamentally a data-driven science. Interpretive efforts stand or fall with how data and interpretations are coupled; that is, with how interpretive ideas are operationalized. The longer-standing critique that archaeological theory is dead unless theories make an explicit effort at operationalizing their key concepts (see e.g. Bintliff Reference Bintliff, Bintliff and Pearce2011; Kristiansen Reference Kristiansen2017; Johnson Reference Johnson2006) comes to mind. In any case, for Riede and Kristiansen, the dimension in which archaeological explanation and understanding can fundamentally improve appears to be the development of always sharper and sharper methods of data extraction and analysis. For them, all systems are already go and essentially just need more time to operate, iterate and refine themselves. Famously, in recent years, Kristiansen has been using the term ‘third scientific revolution’ to designate the new archaeological science (Kristiansen Reference Kristiansen2014).

It is worth observing here that in our original paper, we in fact sought to adopt a mediator position seeking to understand the contours of the debate about determinism, rather than go on branding this or that piece of research deterministic or not – given the heterogeneous make- up of the present author group, that was only natural. That may explain some of the charges of vagueness raised against us by a number of commentators. In either case, for the archaeological science our original paper wanted to offer the idea of implicating governance structures as a potential dimension in which the social production of vulnerability and resilience can be seen to operate. This would be a dimension that also archaeological science knows how to approach insofar as quantifiable (if indirect) data pertaining to social stratification – as evidenced in changing burial practices, architecture and the rest of the standard archaeological fare – are available.

A second effort contained in our original paper by way of bringing a social construction of vulnerability and a resilience perspective to archaeological science – which, however, does not seem to have attracted attention from the commentators – concerned the difference in the temporal scope of study taken regarding hazards and disasters. As argued in our original paper, where in archaeology our research designs tend to look for convergence of environmental/climate proxies and archaeological proxies with a decided temporal focus placed on the post-change moments in time, the literature on the social construction of vulnerability would place the temporal focus upon the longue durée before the disaster.

In any case, in Lafrenz Samuels’s and Ion’s comments, we see a more reserved attitude to archaeological science. Lafrenz Samuels, for example, picks up on the aforementioned idea of forms of governance as a possible pathway towards studying the social construction of vulnerability and resilience. She argues that such an approach merely threatens to revert us to bygone ‘comparative, typological, generalizable, even “testable” approaches’, while more modern, ‘more productive, or at least anthropological’, approaches are also available and, according to her, preferable (p. 15). In its place, a more anthropological approach sets off from the concepts of ‘situated knowledges’, a concept that Lafrenz Samuels briefly illustrates with reference to prominent work such as that by Donna Haraway as well as Clifford Geertz. From the point of view of archaeological science, following Riede and Kristiansen, a central challenge would certainly seem to be the operationalization of this type of concept in archaeology: how are such knowledge claims to be evaluated by reference to data? The present point is not to seek to answer that much-discussed question (Wylie Reference Wylie2002, Chapter 3), but to point out that in discussions like these, something like the processual/postprocessual contrast seems to be alive and well. We will return to this theme shortly below.

That said, we want to explicitly side with Lafrenz Samuels’s call for sensitivity to accountability in the way a science forms its object of study – that, after all, was effectively the closing statement of the original paper as well. That is to say, there are (at least) two ways of framing the character of the human relationship with changing climate and environment. We may frame it apolitically as a technological challenge of humans developing and using technology and other innovations to cope with and maybe even take advantage of the changes; or we frame it politically by implicating different sociopolitical forms of organization of human social and cultural life as producing vulnerability and resilience respectively. These competing ways of framing the situation are very much with us in today’s environmental discourse and we need to hold ourselves accountable for whichever framing we use, for example, in archaeology. Why? Because the way we frame the issue may in its own perhaps small way set an example of how others can begin to think about framing it.

We believe that most readers would side with political framing and accountability, arguing, however, that this ought not to mean a rejection of the point of view that the environment and climate set the threshold or envelope within which humans produce their material basis of existence, an envelope that human technological and innovative resourcefulness may also alter. This idea of a synthesizing, multidisciplinary approach again takes us to the argument made by some commentators that our original paper propounded the tired, old processual/postprocessual division in archaeology.

Thus Ion’s comment casts deterministic explanations as one-shot, reductionist explanations to which multifactor views with ‘more data points, and complex interpretive frameworks’, are preferable (p. 12). On the face of it, Riede and Kristiansen would readily agree in that they too champion the (in their view already ongoing) interdisciplinary archaeological science enterprise.

There is more than meets the eye here, however. Recently elsewhere, Ion argued that in archaeology we are still some way from ‘what a truly integrative narrative would look like’ (Ion Reference Ion2017, 179). A key issue identified by Ion is epistemological: a given phenomenon is defined in a particular way that implicitly structures research. Her example is how aDNA technology has seemingly allowed archaeology to capture the so-called Neolithic Revolution in a new, more perceptive way:

Ion points out that what is going on in a research design like this is the epistemological, semantic, equation of the distribution of aDNA with the concept of an expanding Neolithic Revolution. That is to say, where we have evidence of aDNA as it were travelling from place to place, there we find the Neolithic Revolution spreading out.

Now, arguably, Ion’s argument is formally closely related to ours in the original paper. We argued that there is a popular form of research design that searches for parallel developments in environmental and climate proxies as compared to archaeological proxies. Where there are parallels, we have a reason to believe that the changing environment and climate are somehow influencing human development. As for Ion, there is an implicit epistemological definition at issue here in the manner in which a phenomenon of a certain order such as sociocultural change is conceptualized in relation to a phenomenon of another order, namely changing environment and climate or changing distribution of aDNA.

Again, similar to Ion, the denial of such an epistemological equation does not amount to saying that changes in the environment and climate – or in aDNA – are unrelated to sociocultural changes. As Ion puts it, ‘I also think that scientific data brings an important contribution, given that humans are the result of contingent histories of genetic, biologic, environmental, and cultural interactions’ (Ion Reference Ion2017, 180).

In the original paper, we spoke of something called biologism as possibly accounting for why the changing environmental and climate envelope, in tandem with sociocultural changes, is taken to indicate the presence of a relationship of influence here. Biologism can be described as the view that, as biological beings, humans are subject to and dependent on the affordances of their environment in terms of the material basis of human existence. In such reasoning, there indeed is a prima facie connection between environment/climate and socioculture. We suspect something similar is going on in the equation of aDNA distribution with the advancing Neolithic Revolution: humans reproduce biologically, handing down both their biological and their sociocultural heritage to the next generation. Therefore, prima facie, a particular distribution of some markers of biological heritage is indeed an indication of the distribution of sociocultural heritage.

The problem Ion has with the equation of aDNA with a sociocultural phenomenon is that the equation limits our understanding of the sociocultural dimension – it limits ‘the understanding of human beings to genetic entities’, she argues (Ion Reference Ion2017, 187). However, alongside key tenets such as biologism, scientific thought builds upon the idea that surface complexity can be comprehended by ‘a representation of the system’s critical components’, as Coombes and Barber (Reference Coombes and Barber2005, 305) put it in the passage quoted in our original paper. To put it provocatively, where Ion sees limited understanding and Lafrenz Samuels sees ignorance of ‘situated knowledges’, others might see a superior reduction to the essentials. This difference might describe the hard core of the debate about determinism and archaeological science.

To conclude, on the question whether our original paper perpetuates the ‘two-cultures’ distinction, from engagements in conferences and the like over the past several years, we can tell that there is a lot of goodwill around for a synthesizing position in archaeology. At the same time, exchanges such as our original paper has given rise to here might be taken to demonstrate that some fundamental differences in research orientation remain.