Book review: The Crisis of Expertise

Gil Eyal is one of the most prominent sociologists studying the role of experts, intellectuals and professionals in contemporary societies. In the last 20 years, he has written a series of empirical studies on the role of experts in the diffusion of neoliberalism in Eastern Europe, in the breakup of Czechoslovakia, in the formation and cultural justification of the Israeli State and in the rise in autism diagnoses. In these carefully documented works, he has argued in favour of the sociology of expertise as an original approach superior to the sociology of professions. This is because it focuses not only on jurisdictional struggles (i.e. who controls a task or problem?) but also, and crucially, on what makes it possible for a certain form of expertise, conceived as networks of objects, actors, techniques and institutional arrangements, to be gradually assembled. The latter is now commonly known as the ‘network’ theory of expertise and is often opposed to the ‘realist’ theory of Collins and Evans (2007) and to the ‘institutional’ theory of Jasanoff (2004). In his last book, Eyal takes a step back from these debates and addresses the very topical question of whether expertise is experiencing a crisis, why this is so, and what can be done about it. In recent years, claims that we are witnessing a crisis of expertise have circulated across different social environments, including statements from prominent politicians, debates in media outlets animated by pundits and journalists, and academic publications. Among a variety of explanations, authors have attributed the crisis of expertise to the dumbing down effect of new technologies and to the excessive democratisation and commercialisation of education (Nichols, 2017), to a general erosion of trust in established authorities, exacerbated by increasing inequality (Drezner, 2017), to the influence of false relativist doctrines such as postmodernism and deconstructionism (Pinker, 2018), to widespread and long-standing magical beliefs shared by large portions of the population of any advanced country, aggravated by increased deprivation and uncertainty (Oliver & Thomas, 2018), and to the agnotological strategies deployed by powerful groups to instil doubt on the public on complex issues (Oreskes & Conway, 2010). Although most scholars who speak of a crisis of expertise see it as pernicious, some authors have suggested that this decline of trust might in part reflect a healthy reaction against elite encroachment and technocratic exclusion of informed and concerned lay publics, following a script already rehearsed in many important works of what Collins and Evans (2017) call the ‘second wave’ of science studies. European Journal of Social Theory 2021, Vol. 24(2) 306–310 a The Author(s) 2020

Eyal's account is less categorical. In The Crisis of Expertise, he tries to understand what this growing concern about expertise means in the first place. Inevitably, this leads him to reconsider some of the most apocalyptic diagnoses advanced in recent years, especially those that, like Nichols' (2017), blame non-sociologically the wilful ignorance of the public. First, what we are witnessing is not simply a crisis of expertise, but rather what he calls a 'two-headed pushmi-pullyu of unprecedented reliance on science and expertise coupled with increased suspicion, scepticism, and dismissal of scientific findings, expert opinion, or even of whole branches of investigation ' (p. 4). Second, the whole concept of an 'assault on science' is misleading, because no single, unified, capital-S Science exists, and because not all science is under assault, but rather what may be called the 'regulatory' and 'policy' sciences, that is, those fields that occupy an intermediary position between academic research and the policymaking process. Third, and most importantly, the very concept of expertise has always presupposed a fundamental uncertainty about who deserves to be called an expert. Expertise, Eyal argues, is a 'historically specific way of talking' occasioned by a situation in which 'the number of contenders for expert status has increased, the bases for their claims have become more heterogeneous and uncertain, and the struggles between them have become more intense ' (pp. 19-20).
Since the relatively recent diffusion of the vocabulary of expertise, which dates back to the post-war period, debates about technical matters of public concern have been riddled with uncertainty about how to adjudicate competing claims to expertise. Expertise is not the product of a 'knowledge society' but rather the product of a deeply politicised, agonistic society. The 'problem of extension', that is, where to draw the boundary between the experts and the laypeople, and the 'problem of trust', that is, whether we should trust explicit abstract procedures or the trained judgments of experts, have been the two main dilemmas around which academic theories of expertise as well as real-world decision-making have been structured. Gradually, these problems have engendered a legitimation crisis. Through an insightful reading of Jürgen Habermas and Niklas Luhmann, Eyal argues that legitimacy cannot be conceived simply as deception orchestrated by powerful groups, nor as the achievement of a rational consensus. Justifying authority, that is making commands legitimate, takes time and involves a wide range of rhetorical tools and a motivated group of people who will uphold and enforce public choices.
In sum, the rhetorical repertoire of expertise was one of the most significant sources of legitimacy employed by the liberal state to justify its policies, and it is within this practical endeavour that something misfired. Eyal proposes a narrative in three stages. The first stage is the 'scientisation of politics', coinciding with the post-war period, at a time of massive state intervention to promote growth, increasing regulation of the economy, and massive involvement of experts at every level. Science is mobilised by the liberal state to justify political decisions, as illustrated by the Kefauver Harris Amendment, which strengthened the US Food and Drug Administration's (FDA) control over the requirements to market new drugs. The second stage is the 'politicisation of science': science becomes polluted because the prescriptions decided by unelected scientists, such as the 'acceptable levels of safety' decided by an expert body, were not grounded in objective truth and sometimes clearly benefitted certain groups over others (the FDA, for example, was regularly accused of having an 'industry bias'). Lay stakeholders mobilised to 'demand transparency and a seat at the table' (p. 102). Finally, in a third stage, the liberal state rescues science by minimising exposure and adopting a variety of strategies that mirror the dominant theories of expertise. This is arguably the most interesting part of the book. These strategies are outlined by crossing two axes representing the problem of trust and extension. The first strategy is 'exclusion', that is, reinforcing the boundary between experts and laypeople to 'generate trust in technocratic expert judgment' (p. 105), under the assumption that introducing economic and political interests directly into scientific deliberations is detrimental. This strategy is encapsulated in the Daubert standard and finds a theoretical defence in Collins and Evans' works. The second strategy is 'mechanical objectivity', which shares the technocratic thrust of the first but prioritises formal procedures over expert judgment. This is illustrated by the widespread reliance on randomised controlled trials (RCT) by agencies like the FDA and is also very popular in the assessment of international development aid projects. The third strategy is 'inclusion', which emphasises participation of lay actors and transparent procedures, and is epitomised by the now-common inclusion of patients' groups in the deliberation of medical advisory bodies. The fourth strategy, 'outsourcing', advocates for the promotion of independent, self-governing organisations that are inclusive, have limited aims (mainly establishing scientific consensus) and emphasise uncertainty and precaution over decisions and prevention. The Intergovernmental Panel on Climate Change illustrates this approach, while the theoretical equivalent is the idea of coproduction of science and society.
While the author's loose preference for this latter strategy (and aversion to the exclusive reliance on RCT) is clear, the normative import of the book is broader than the mere assessment and ranking of these different approaches. Eyal argues that reacting to the crisis of expertise by denouncing the deliberate illiteracy of the general public and blaming the Internet and social media will backfire by widening the gap between experts and laypeople, reinforcing mutual distrust and deepening the legitimacy crisis. Expert organisations should instead accept to be challenged by social movements and laypeople by making their procedures transparent and making possible for outsiders (including maverick scientists and suspicious activists) to contribute to the scientific process. Although some controversies might be properly speaking manufactured, Eyal believes that events like 'Climategate' in the long run end up reinforcing science by removing really existing biases and by attracting increased scrutiny and public attention. As a telling illustration, Eyal offers the example of the Surface Stations project, a citizen science initiative aimed at assessing the condition of US weather station promoted by climate change deniers which ended up showing that the real temperature was higher than reported by many stations.
Eyal concludes with a plea for embracing the ambivalence inherent in scientific activity: the vocation of modern science requires experts to dance between confidence that the best knowledge at our disposal should inform the decisions that shape our society, and recognition that even our best knowledge is limited and uncertain. But truly legitimate policy decisions based on sound expert advice will require a new institutional configuration. Eyal tells us that which we do not want: we do not want a 'republic of science', because without people keeping the experts in check fraud and violations will happen. We do not want an 'agora', because a strong backbone of professionalised experts is necessary. And we do not want a simple 'political debate', characterised by strong adversarial procedures, because this is the status quo and it is not working.
The Crisis of Expertise is an entertaining read and never fails to astound for its richness and range. Eyal spans multiple disciplines, from the philosophy of Artificial Intelligence to the sociology of risk, and, although he mainly tells an Anglo-American story, he draws on a wide range of examples to develop his argument, from the 1923 Frye ruling to the 1985 Bodmer report, and from the 1962 Kefauver Harris Amendment to the No Child Left Behind Act of 2002. Unlike other recent books (Brown, 2009;Collins & Evans, 2017;Moore, 2017), Eyal's is not a political theory of the rightful place of science in society. As noted above, it is mainly a reconstruction and interpretation of the debate about expertise, which maps the space of possibilities rather than identifying the appropriate route. Political theories, and not only those about scientific decisionmaking, often have the unfortunate tendency to draw a stark contrast between two equally dogmatic positions (such as 'technocracy' and 'populism'), to make the case for a balanced, sensible, intermediate position, and, in doing so, to push their opponents into the two pitfalls. The issue with this kind of theorising is that few would accept the way the extreme positions of the debates are depicted, and most would end up agreeing with the intermediate position, stripped of any controversial, and therefore interesting, traits. Collins and Evans arguably commit this sort of centrist repositioning when they write that what their approach (which they call 'elective modernism') argues 'is simply that technological decision-making in the public domain should be approached case-bycase and not driven by a pre-determined preference for a particular kind of outcome' (2017, p. 124), an uncontroversial qualification if there ever was one. Eyal's book is not completely immune to this tendency, but, on the whole, he successfully identifies the advantages and weaknesses of different responses to the crisis of expertise, opening up spaces for discussion about the appropriate strategy to be taken in a given set of circumstances. Even when he makes explicit normative claims, such as when he suggests that as a rule of thumb mistrustful actors and potential 'merchants of doubt' should be included rather than excluded, this view is clearly formulated in contrast to plainly existing positions in the debate about expertise, so that it retains its controversial and thoughtprovoking qualities.
In sum, in a debate that is becoming increasingly interdisciplinary, international and polarised, The Crisis of Expertise is a breath of fresh, clarifying air.

Federico Brandmayr
Centre for Research in the Arts, Social Sciences and Humanities, University of Cambridge, UK Email: fb446@cam.ac.uk