In the 1990s Harry Collins was considered one of the principal actors in the so-called Science Wars, which pitted postmodern thinkers, including some sociologists and philosophers, against scientific realists. The postmodernists argued that scientific theories were social constructs; scientific realists, like physicist Alan Sokal of the infamous 1996 Sokal hoax, argued that science was objective, based solely on reason.
Collins is a sociologist—a distinguished research professor and director of the Centre for the Study for Knowledge, Expertise and Science at Cardiff University in the UK. He is also a fellow of the British Academy and recipient of the 1997 John Desmond Bernal Prize from the Society for Social Studies of Science. Like the postmodernists, Collins has sought to redefine and reinterpret scientific expertise and scientific consensus in most of his 18 books.
;However, regarding Collins’s latest treatise, Are We All Scientific Experts Now? (Polity, 2014), Physics Today reviewer Thomas Vogt writes that Collins denies “in no uncertain terms, the specialist status to anyone who is not a member of the scientific community and is not able to make ‘judgements from the platform of the norms and aspirations that drive the community.’”
In fact, says Collins in the interview below, his research group was never engaged in a campaign against science. In the new book, and in his responses below, Collins focuses on some benefits of his group’s “wave-two” science studies (which includes the notion of “interpretive flexibility”) and cites examples of where some “social analysts” and journalists take it too far. He also describes what it means to have “interactional expertise,” an idea he explores through discussion of his more than three decades of engagement with the gravitational-wave-detection research community. Collins is developing a new technique he calls the “Imitation Game” for exploring expertise and comparing the extent to which minority groups are integrated into societies.

PT: Some may see this book, Are We All Scientific Experts Now?, as a remarkable turn of events, given previous positions you’ve held on the nature of the scientific enterprise. What led you to write this book, and how and why have your positions changed?
COLLINS: In 1980 my philosophical position shifted from the joyful iconoclasm of the 1960s to seeing relativism as a matter of methods. But my generation of sociologists of science, who in the early 1970s invented the sociology of scientific knowledge, were never engaged in a campaign against science. What we were against was misrepresentations of its degree of perfection. Overinterpretation of my 1993 coauthored book, The Golem: What You Should Know About Science (Cambridge University Press), sucked me into the so-called Science Wars, but that book demands, “Let us admire [scientists] as craftpersons: the foremost experts in the ways of the natural world.” (page 140)
My group was trying to show that science was a craft embedded in society, not a logical machine. What people said about us is another matter and did not always reflect great credit on those saying it. It is true that our early work became swamped in the much larger movement known as postmodernism, which did have an antiscience flavor reflecting the conflict between “the two cultures.” [With my and Robert Evans’s] 2002 paper, called “The Third Wave of Science Studies: Studies of Expertise and Experience,” I changed the emphasis of my work. This was in response to social analysts interpreting the new and richer analysis of the nature of science as meaning that any contribution to a scientific debate was equally valid.
The term “lay expertise” came into vogue and led analysts to justify vaccine scares and the like because ordinary people had a kind of folk wisdom that was on a par with scientific understanding. This seemed crazy to me: Scientists had something special to say even if it was not perfect and even if they often disagreed with each other. A society in which all expertise was leveled out would be abhorrent. But as [Evans and I] said in Rethinking Expertise [University of Chicago Press, 2007], “the speed of politics is faster than the speed of scientific consensus formation.” So my group shifted attention to expertise, which can be recognized long before the truth is established. We argued that scientists’ expertise justifies their special contribution to policy. The point about this change in the direction of gaze is that it is entirely compatible with the craft practice model—it is not a volte-face or anything like that.
PT: What do you regard as the greatest accomplishments, and greatest mistakes, of what you call wave-two science studies, which challenged the scientific mythos?
COLLINS: I think wave two accomplished many things. It led to a better understanding of tacit knowledge and its implications for the use of experimental replication as a test. See, for example, my books Changing Order: Replication and Induction in Scientific Practice (Sage, 1985) and Tacit and Explicit Knowledge (University of Chicago Press, 2010). A scientist cannot know if a replication falsifies an experimental finding unless they also know that the work has been done adequately, and since experimental skill is tacit-knowledge-laden, the argument can go on forever. I call this “the experimenter’s regress.” Wave two revealed that in science, as in love, “distance lends enchantment”: The further away from the real scientific work you are, the more decisively perfect—or wrong—it looks. That’s why outsiders can be more certain than insiders, and that explains a lot about how public arguments work.
The same applies to distance in time. When historians travel backward they find that those classic experiments were rarely as simple, sharp, and clear as they look now. Wave two showed that because there is a lot of interpretative flexibility when scientific data are turned into findings, outside interests can have an effect on results. It showed that statistical analysis is misleadingly and damagingly exact-looking. It showed that the same data can be made more or less acceptable to the scientific community depending on how ambitiously they are presented. It showed that technologies are only good or bad in the context of what consumers want from them—a dangerous, “high-wheeler” bicycle is better than a “safety” bicycle if you want to show off. It showed that what people do with failed experiments is quite different in physics, where you struggle to find out why, and biology, where you throw them away and start again. It showed that the public are not going to agree about things like global warming just because you teach them more science—after all, scientists don’t always agree, and they know far more than the public.
In sum, wave two got rid of the scientist in the white coat with authority to pronounce definitively on anything scientific, and it got rid of the idea that all the forensic science you need you can get from the prosecution. Journalists stopped reporting revelations and started reporting debates. The trouble is that journalists now think every scientific story has to be balanced and that consensus in the science, ironically, creates a dangerous imbalance. And some analysts still think that if scientists disagree, then democracy demands that everyone has equal right to join the argument.
PT: Can you briefly explain your “taxonomy of expertise” and what impact you think that framework could have on science policy and education?
COLLINS: The key is to treat the crucial part of expertise as a body of tacit knowledge. That means, to be an expert you have to be part of the expert community or you won’t acquire the tacit knowledge, including the knowledge of which published work to take seriously and which to ignore. So that reduces the rights of those who are not members of the community to contribute to the debate unless they have something special to offer, like some narrow body of experience. Policymakers can draw on this to justify giving extra weight to the views of the expert community. And educators ought to give up the idea that distance learning will solve all their problems, because membership of groups is the key to understanding.
Psychologists and philosophers need to grasp the idea that some expertises are ubiquitous because we learn them as we grow up. Natural language speaking is an example; it is an extraordinarily difficult feat even though we can all do it. If one sees this, one sees why computers are so bad at handling speech and doing many other things: They are not proper members of societies. Meta-expertise is judging the value of experts and their expertise; that, too, is a social skill and an unreliable one unless one has local knowledge and can be a whistle-blower or equivalent.
Interactional expertise, which is expertise that is learned from immersion in the spoken discourse of a domain, is turning out to be an important new concept, explaining how narrow practical expertises can be combined, how good management works, and how societies hold together. Combining this new kind of understanding of expertise with “distance lends enchantment” shows just how dangerous and misleading “www dot et cetera” can be.
PT: Many scientists are optimistic that we’ll soon see gravitational waves either directly or indirectly, as imprints on the cosmic microwave background. Are these exciting times for you personally, having spent a significant portion of your career interacting with the gravitational-wave community?
COLLINS: If I live long enough, I’ll be excited for my physics friends and colleagues and excited at the prospect of finishing a project that I started in 1972. I’m nearly the last man standing from that era. But I’m a social scientist, not an astrophysicist, and the least interesting thing for me, as opposed to my colleagues, will be if a clear and sharp signal shows up in the more sensitive detectors. Interesting social science comes out of tension, not triumph.
My guess is I have no need to worry. The blind injections I have written about in a couple of books have told me a lot of what there is to learn about the process of discovery, but something unexpected nearly always turns up. I should add that, social scientist or not, it was a huge privilege and huge thrill to be around the initial LIGO as it came on air and started to work in the face of what seemed impossible odds; that sense of the near miraculous may return with the new devices.
PT: What books are you currently reading?
COLLINS: Novels! Robert Harris’s novel of the Dreyfus affair (An Officer and a Spy, Knopf, 2014) is extraordinary.