I think Nature needs to work on how it thinks about interdisciplinarity. A recent article (Rafael Núñez , Michael Allen, Richard Gao, Carson Miller Rigoli, Josephine Relaford-Doyle and Arturs Semenuks, “What Happened to Cognitive Science?”, Nature Human Behavior, Published 10 June 2019, https://doi-org.ezproxy.net.ucf.edu/10.1038/s41562-019-0626-2) answers the title question as follows: “cognitive science appears to have failed to generate “a successful research program”.
The evidence for this is a study of the bibliometric and socio-institutional authors’ affiliations from over 1000 articles in the journal Cognitive Science. The authors found that psychology was overwhelmingly dominant in these affiliations and citations. They note that in the central disciplinary contributors of cognitive science (as defined by a 1978 Sloan Foundation report on the area), which included philosophy, linguistics, anthropology, neurosciences, computer science, and psychology, there remains a wide variation (and even incommensurability) on how “cognition” is understood. They also note that after 2000, less than 10% of affiliations come from departments or programs in cognitive science.
So, where to begin with all this? It seems like there are significant problems with their study at two levels – methodological and conceptual. As for methodology, we can quickly point out that one journal was seen as representative of the whole field. It is worth noting that, for instance, had a journal like Phenomenology and the Cognitive Sciences been included, the presence of philosophers would have jumped. Work in neurosciences is scattered across several other journals. And then there’s Trends in Cognitive Sciences, Cognition, and a bunch of others. And then, there is the work on specific topics, which might fit in more specialist journals. So, for instance, a study in cognitive impairment or disability might show up in a journal devoted to that.
In other words, it seems clear that the sample set is already very small and highly skewed towards a particular conclusion. And this also assumes that the only relevant research data comes in the form of the journal article. While that is surely the major outlet, there are also books which contribute significantly. The authors list some significant books, but it is noteworthy that the books they see as relevant are all at least 10 years old, and some much older than that. Articles are clearly the fastest way to get material out which is experimentally based, but the cognitive sciences are a meeting-space between science and the humanities. So, the sample set takes out of the picture any real contribution of the humanities, which might come in a more extended fashion.
As for conceptual issues, that gets us into thinking about how interdisciplinarity works. The authors seem to think that a successful interdisciplinary program is one which has contributions distributed fairly evenly across the member areas, and/or has developed programs or departments in which research can be calibrated and controlled administratively (in other words, the research program can be shaped through hirings, curricula, etc.).
Interdisciplinarity isn’t always like that. It is always a space of negotiation, since concepts are defined within disciplines using their own resources and when disciplines start talking to each other the differences become obvious. Furthermore, the kinds of questions that are asked will differ. And, the kinds of resources available to address questions will differ. Take, for example, a concept as slippery as “meaning”. This will at the very least have a vastly different sense between the sciences and the humanities on questions of cognition, and that difference might never go away. At least some philosophers might want to say that cognition happens in an already meaningful world, not just in a world of symbolic manipulation, and that meaning must take into account our bodiliness in the world and a host of other things. That’s going to mean that models will not be simple and explanations will not be easy. Does that mean that cognitive science isn’t a “successful research program”? Does the very presence of humanities, even in the small amounts that the authors identify, serve to undermine the success of the area? Or does it mean that, unlike some other areas of research, this one will always be irreducibly complex, and will never look like other recent areas such as genome research?
What would a “successful research program” look like, for the authors? I suspect it would be based on a single or small cluster of models. That is not the case in this area. They acknowledge that – they mention connectionism, enactivism and other non-computational models, but the fundamental implication of that is that it is more likely that those working on a particular model will talk to each other, and find the venues to do that. In other words, it’s likely that in an area like the cognitive sciences, the emergence of research will look exactly like it does – centered in different places, still figuring out questions that exist at the edges of both the constituent disciplines and the nodes of existing research.
I don’t think a successful research area should be defined by either the levels of involvement of its constituent areas, nor by whether it has anything like a unified model for the phenomena it studies. The success is in the ways that new questions are generated. The life-span of an academic area has different modalities at different times. This area is still in its infancy. The questions raised at the end of the article are actually good ones – what exactly is the subject matter of the field? If there’s to be a degree in the area, what should that look like? What should employers expect graduates to know? To what extent is cog. sci. necessary for forming a new generation of scientists of the mind? These questions could have been arrived at without the questionable understanding of interdisciplinarity and without the limited sample set.