“I became a social scientist,” explains Iver Brynild Neumann, “because I was flabbergasted by the fact that most humans, in most places, most of the time, manage to live in relative peace with one another, and wanted to find out how this could be.” And so the Norwegian-born academic set off to study political science and social anthropology, ultimately with a clear emphasis on international relations (he earned a double doctorate in social anthropology and IR and he helped start the first international relations master’s program in Norway, for example).
Now on the faculty of the London School of Economics, his inquiry of late has been almost geographical –where does IR fit into the social sciences?
The question isn’t so much whether international relations is a social science – Neumann says he takes that for granted – but rather how it fits into the social science landscape, and furthermore, whether that landscape is part of a larger academic continent or just an island.
Neumann addressed these questions in the inaugural lecture for the Montague Burton chair in International Relations at LSE (he assumed the chair in 2012 and the speech was delivered in January 2013). His speech, “International Relations as a Social Science,” and a roundtable of learned responses to it, appear in the current issue of the journal Millennium, Journal of International Relations. (If you prefer to watch his speech, click HERE.)His concerns, and even his forum, are familiar (so familiar that one reviewer, the LSE’s Chris Brown, complains, “It seems to me that IR spends far too much time on this kind of navel gazing; we are much too concerned with ‘all that meta stuff’ as Fred Halliday used to call it”). Neumann cites a 2001 article in Millennium, “Why International Relations Has Failed as an Intellectual Project and What to Do about It,” by Barry Buzan and Richard Little. Those two, he summarizes, noted an ‘assumption that IR is a “backward” social science’ in and out of the discipline, and themselves charged that IR ‘remains curiously insulated from the other social sciences.”
Insular, eh? So back to that island vs. inland line. Neumann for his part suggests that a discipline should be both discrete and connected; perhaps a peninsula?
The social sciences emerged out of history, jurisprudence and sundry now half-forgotten disciplines such as social arithmetics. Sheer volume made specialization a necessity. Still, the ongoing splitting up of academic inquiry has consequences for each and every discipline, and I do not think we discuss those consequences enough. By the very logic of science, every one of its disciplines may have a certain autonomy, but in order to maintain its membership, it still has to answer to certain common standards.
Those standards – to extend the metaphor, where the peninsula joins the mainland—Neumann places at three junctures: history, methods, and autonomy (with a side trip to interdisciplinarity).
Looking at history, Neumann sees “ample room for improvement” and tangible benefit from studying “the full range of [IR] cases.” In Neumann’s telling, social science and history were somewhat twinned at IR’s emergence, and that family resemblance somewhat embarrasses IR: “History was what the social sciences tried to limn itself off from.” But by defining itself in part by what it wasn’t, the two disciplines were even more linked. Worse, when history has been invoked in IR, ham-handed uses created problems.
Now Neumann wants to reunite the family, saying that the possible misuse of anecdote is less problematic than the amnesia that replaced it. Such amnesia Neumann called a “road to perdition.”
The reason why we would be lost without history is very simple. Any science worthy of the name must aim to cover the full universe of cases that are pertinent to it. The subject matter of IR is political and social life that plays out in a setting where there is a plurality of polities. More specifically, the subject matter of IR is sovereign and suzerain relations between polities as they existed in the past, as they exist in the present and as they have been and are imagined to exist. It follows that IR has three, and only three, sets of cases to pursue, namely the logics and effects of extant relations between polities, historical relations between polities and imagined relations between polities. I note, ruefully, that we are not doing too well even on extant relations.
A second path to the mainland runs through methods. “Any science worthy of the name has a running debate about methods,” Neumann insists. “As Weber put it, ‘The dilettante differs from the expert … only in that he lacks a firm and reliable work procedure.’”
If history gives us something to think about, there is also the question of what we need to think with. To take a leaf from the debate about research programmes, a scientific discipline worthy of the name must have at its disposal tools for data collection, ways of ordering the data, ways of discussing the relationship between the data and the wider world, and debates about what that world consists of.
As someone who himself focused on theory and qualitative efforts at the start of his career, Neumann isn’t defecting in the age-old fight between qualitative and quantitative styles. Instead, he’s just looking for a little rigor.
In IR, the discipline’s quantitative practitioners are passionate about the problems surrounding data programming, in such a degree that one sometimes wonders if counting is not being substituted for thinking, but at least the quants do engage in a debate about methods. Those of us who mostly do qualitative stuff, however, must be severely faulted for having largely neglected methods.
Thirdly, Neumann expresses a concern that could be crudely shorthanded as ‘beware of becoming too relevant,’ which could be seen as a long-standing trait of IR if you accept the notion that “the adviser to the King is a predecessor of the IR scholar.” While relevance might seem an unalloyed positive, Neumann argues it can crowd out important (and durable) scholarship:
We tend to concentrate on phenomena that are politically important and agents that are politically powerful, to the detriment of studying stuff that would have added to our general knowledge of the subject matter that we call our own.
Being policy-relevant and being the handmaiden of specific forces, be that states, NGOs or some imagined collective, amounts to the same thing, namely to produce instrumental knowledge. With the government playing an ever greater role in the allocation of research money, and with scholars within universities being under ever more pressure to apply for external funding, we must expect there to be ever more instrumentally produced knowledge. The risk is that we end up drowning in papers, reports and articles that are of no interest to anybody a year after they were published.
While Neumann suggests a stronger autonomy from the applied, he also calls for closer ties to other academic areas. He seeks a greater connection with other disciplines, specifically two ostensible “competitors” of psychology and biology, which “share our key interest in group conflict and alterity.” Neumann suggests the collaboration won’t necessarily be friendly – “I think a major debate between biology and IR is brewing” — but it is necessary. “As social scientists,” he says, “we are not only interested in specific phenomena, but also in the human condition, and so we cannot simply ignore the physical corpse.”
This isn’t a universal impulse. Reviewer Jonathan Mercer of the University of Washington argues, in a sense, for keeping a wary eye on the social science purity of IR:
A risk is that political scientists will fall prey to physics envy (or neuroscience envy) and become high-tech phrenologists, using brain activity evident from brain scans to identify where one can find trust or fear in the brain. It is almost irresistible to refer to [insert brain jargon here] to bolster one’s ‘science cred’ and locate apparently subjective emotion as a brute fact of the brain and body
Where Mercer suggests taking a care, reviewer Lauren Wilcox at the University of Cambridge found greater collaboration with behavioral and biological science potentially “disheartening,” based in part on the fear of mayhem:
To put it bluntly, all sciences are social practices with histories, and Neumann’s article elaborates on some of the deeply problematic types of sociobiology and evolutionary psychology that have gained popularity in IR in recent years. From the vantage point of many associated with feminism, anti-militarism, postcolonialism, critical race studies, disability studies, queer theory, and other fields of study grounded in struggles on behalf of marginalized peoples, there is a deep scepticism of the uses of scientific discourses to justify claims about the social and political world, with reasons based precisely in considering the history (and the present) of discourses and their use in legitimizing various forms of oppression, violence and inequalities around the world.
In his own lecture, Neumann offers a pre-emptive response that doesn’t defend ham-handed uses but does imply that the absence of data can cause as much mayhem as the misuse of it:
I also think that it is wrong that we, as a discipline, tend to rate most highly the books that deliver abstract theoretical arguments and nothing else. Social scientists should lean on philosophy, but they are not primarily philosophers. Our niche is between history, which tends to fetishise empirics, and philosophy, which tends to fetishise metaphysics. For me, it follows logically that the social scientific ideal is the sustained, theory-led empirical argument.