Stephen Turner writes in the 2023 book, The Oxford Handbook of Expertise and Democratic Politics:
Science is sometimes thought to be a self – correcting system: replication and the fact that other scientists must rely on the previous and related research results to perform their own experiments is thought to provide error detection. Sometimes, this works. But as the statistician John Iaonnidis (2005) has shown with respect to medical research, the idea that the system is self – correcting may be an illusion: “Simulations show that for most study designs and settings, it is more likely for a research claim to be false than true. Moreover, for many current scientific fields, claimed research findings may often be simply accurate measures of the prevailing prevailing bias” (40). Researchers searching for a particular drug effect, for example, may find it and report it, but they will not report failures to find it, or they will abandon research strategies that fail to find it. And this bias is a result of facts about the social organization of research — namely, the institutional reasons that cause people to look for confirming results, or, as Iaonnidis explains, where there is “financial and other interest and prejudice; and when more teams are involved in a scientific field in chase of statistical significance” (2005, 40).
We can see, sometimes in advance, that particular institutional arrangements, such as highly competitive grant systems, which rely on peer review, are likely to produce a great deal of conformism and far less high – risk innovative thinking. This was a fear of the physicists who produced the A – bomb. They used the risk – reducing device of setting up rival teams, with rival approaches, notably on the 600,000 – person Manhattan Project and throughout the postwar period. Lockheed pioneered the use of “skunk works,” innovation – oriented units created outside the normal organizational structure, to generate alternative technologies, which, at IBM, produced the personal computer (PC). And there are ongoing efforts in science to create institutional structures to correct for issues that become problematized. In recent years, for example, there have been organizations that publicize misconduct, such as Retraction Watch, and a large structure of research misconduct mechanisms was created over the last generation. Most recently, there have been such innovations as the online postpublication commentary forum on PubMed ( Marcus 2013 ) and funding for replication studies ( Iorn 2013 ).* Professions typically operate in markets, which they seek to control.
* An expert who speaks against the consensus risks losing the status of expert. And this is grounded in a basic feature of expertise itself: the legitimacy of expertise is closely associated with the legitimation provided by other experts who validate the expertise, the credentials if not the specific views, of the expert in question. So “intellectual capture” in the sense of the mutual dependence of experts on one another for legitimacy is a feature, not a bug, of expertise, and an organization that promotes opinions that fall outside of the range of what other experts treat as genuine expertise risks reputational loss or the loss of expert legitimacy.
* Consensus, even the limited kind of agreement necessary to produce a policy decision through the aggregation of expert knowledge, requires procedures.
From the chapter, “The Third Wave and Populism: Scientific Expertise as a Check and Balance,” by Harry Collins, Robert Evans, Darrin Durant, and Martin Weinel:
* Although it is important to challenge expertise to ensure accountability and legitimacy, in the last decades expertise has been steadily undermined in Western democracies to the point that, under some interpretations, everyone is counted as an expert.
* The actions of former US president Donald Trump are an iconic example of the confluence. His actions while in office were, in effect, “breaching experiments,” forcing us to think much harder about what democracy means and revealing things that we did not realize we already knew. 7 For example, we can now see that, before Trump, there was an unwritten constitution underlying the written Constitution of the United States. The unwritten constitution included the expectation that presidents will disclose their tax returns, divest themselves of their private business interests, and not appoint unqualified members of their families as senior advisers. It also assumed that they will refrain from attacking scientific expertise by denying its efficacy and shamelessly proclaiming the existence of an alternative set of truths, authorized by the government and its supporters, which better align with their preferred policies. 8 The Trump presidency and its aftermath have shown us, anew, how democracy works, or used to work, and where scientific expertise fits within.
* Under populism, in contrast to pluralist democracy, “the people” that the government claims to represent are no longer all citizens but only the subset that expressed a particular view — usually the majority view (though this is often substantially less than 50% of the electorate). Crucially, once expressed, this view is treated as a fixed, uniform, and collective view that encapsulates the legitimate aspirations and concerns of the entire society and can be understood and represented by a single leader or party, possibly in perpetuity. Minorities, or others who oppose this vision, are treated as deviants, and their refusal to accept the legitimacy of the populist claim is denounced as a betrayal of what is now defined as the organic view of the people. Under populism, the pluralist democratic principles of freedom and equality that uphold respect for minorities are set aside, and the diversity that pluralist democratic societies permit and even celebrate is seen as a sign of failure or danger.
* …we stress the importance of the right kind of representative institutions, including expert institutions, as opposed to giving ever wider responsibility to citizens. Broadly, we favor Walter Lippman’s views over John Dewey’s and elected representatives over continual referendums.
