Review - The Philosophy of Expertise
by Evan Selinger and Robert P. Crease (Editors)
Columbia University Press, 2006
Review by Isabelle Peschard, Ph.D.
Jan 15th 2008 (Volume 12, Issue 3)

This collection of fifteen essays constitutes a new and much needed contribution to the philosophical landscape. Neither the subject of expertise nor the questions it raises are new. However, by cutting across disciplinary boundaries to bring together isolated perspectives, this volume makes expertise visible as a specific subject of philosophical inquiry. The diversity of approaches, in epistemology, political philosophy, legal theory, social studies of science, constitutes expertise into a complex and fascinating object of reflection.

Though the essays were all previously individually published, the possibility of confrontation or articulation that the collection offers is clearly a precious additional asset for each of them. It reveals hidden dimensions in what is at issue in each of them and provides different platforms for interpretation and evaluation of the arguments, but in addition, by presenting them as elements of a multi-dimensional and specific domain of investigation, it bestows on them a new significance, pointing beyond the individual contributions.

The general introduction by the editors includes a concise presentation of the organization of the volume and of its rationale. The volume is organized in three Parts, around three kinds of issues about experts and expertise: the relation between expert and non-experts, expertise as particular ability, and the social embedment of expertise.

The questions broached respectively in the first and third Parts are not always easy to separate. For instance, in the first Part, one will find Goldman's essay addressing the question of whether it possible for non-experts to make an epistemically grounded choice between competing experts, and Turner's discussion of whether deference to experts is compatible with liberal democracy. In the third Part, one will find Hardwig's answer to the question whether it is rational at all to believe what experts say when, as non-experts, we can have no evidence for what they say, as well as Fuller's conception of expertise as socially constituted.

But the partial overlap of these two Parts is not really problematic; it reveals, rather, the complexity of the subject and the intricacy of the questions it raises and it fights the tendency to isolate epistemological and socio-political discussions from one another. 

What is more questionable is whether the second Part really succeeds in illuminating what expert abilities consist in. Whereas Dreyfus' conception of expertise and its normative implications are the object of a thorough and insightful critical examination by Selinger and Crease, Dreyfus' essay itself is not much more than a brief restatement of his conception of the process of acquisition of expertise. Annas' presentation of ancient conceptions of expertise introduces an original perspective on expert ability, but this account figures only as a stipulation in service of her argument for a certain conception of moral knowledge. The discussion by Selinger and Mix of the concept of interactional expertise, developed by Collins elsewhere and mentioned in the contribution by Collins and Evans in the first Part, is especially interesting. But its target is very specific, the embodiment of linguistic abilities, and the reader may have the impression of witnessing one side of a well-advanced and rather technical dialogue.

The remainder of this review will focus successively on each Part of the volume. The quality and diversity of the contributions makes it impossible to do justice to the richness of the material that is presented and analyzed or to the refinement of the arguments that are developed. My reading will necessarily be selective, focusing on certain essays, by which I express no disapproval by omission, and on only few of the different ideas weaved together in these contributions.

The first Part contains contributions from Alvin Goldman, Harry Collins and Robert Evans, Scott Brewer, Stephen Turner, Peter Singer.

          Goldman and Brewer are both concerned with the situation of non-experts confronting conflicting experts and having to make a choice about which to believe. Both agree that a direct justification of experts' claims would require some epistemic competence that, by definition, the non-expert doesn't have, but might there be sources of indirect evidence available to the non-expert to guide and justify his or her choice? Goldman is rather optimistic as to the possibility of relying under certain specific conditions on some indirect indicators of epistemic competence, such as dialectic competence or past track records of cognitive success. The problem with Goldman's proposal is that it is difficult to imagine a concrete situation in which these indicators would be indeed available. For instance, assessing dialectic competence would require that all evidence and arguments relevant to each expert's position be presented, criticized, and defended in a face to face confrontations attended by the novice. The kind of situation closest to this ideal might be the one specifically addressed by Brewer, where in the context of legal decision procedures judges and jurors have to evaluate conflicting scientific expert testimonies.

According to Brewer no indirect source of evidence can constitute a basis for epistemic justification. For instance, Brewer rejects not only demeanor but also rational evidentiary support as source of epistemic justification. And even though there might be more to dialectic competence than demeanor and rational evidentiary support, these are surely part of it. Brewer calls into question also the possibility of relying on the credentials of the competing experts, with reasons that would apply as well to what Goldman calls the past track records of cognitive success. Reading the two papers opens up new questions for each of them. For instance, with Goldman's paper in mind, one cannot help wondering whether the particular stakes of the legal procedure that Brewer focuses on makes his conception of epistemic justification particularly demanding, whereas reading Brewer makes one wonder to what extent Goldman would defend the epistemic character of the form of justification he sees available to non-experts.

It is interesting also to confront Brewer's proposal that only scientifically trained judges and jurors should evaluate scientific experts to the reflection developed by Collins and Evans on the different forms that expertise can take. With their contention that only those who have expertise should take part in technical decision making, Collins and Evans appear to be no less demanding than Brewer with respect to the requirement of epistemic competence. But they argue against reducing this competence to that of scientists. They distinguish different forms of expertise required by the process of technical decision making, especially contributory expertise and interactional expertise, and argue that being scientist is neither sufficient nor necessary to count in a particular case as an expert. It is not sufficient, for the epistemic competence has to be specific to the technical problem at issue. Nor is it necessary, for nonscientists may qualify as expert on the basis of their experience, as shown by their case studies of farmers' expertise regarding the ecology of their farmland.

Turner's contribution examines the political dimension of expertise by asking whether and how deference to experts is compatible with liberal democracy. Turner elicits and explores two sources of tension: the conflict between the ideal of equality and the restricted 'possession' of expertise, and the conflict between the ideal of the state's neutrality and the reliance on expertise as a particular sort of knowledge. Whereas Collins and Evans categorized expertise in terms of epistemic competence, Turner lays out a categorization of expertise in terms of the audience of the experts, those who accept their cognitive authority, and of how this audience is constituted.

Though both categorizations do justice to important aspects of the identification of experts, it is not obvious that they can always be reconciled. In particular, it is not clear whether and where the experience-based expertise and the interactional expertise brought forward by Collins and Evans would find a place in Turner's categorial schema. As we will see, an unexpected answer to this question may be suggested by Evan Selinger and John Mix in their critical examination of the concept of interactional expertise in the second Part.

The second Part of the volume contains essays by Hubert Dreyfus, Evan Selinger and Robert Crease, Helene Mialet, Julia Annas, Evan Selinger and John Mix.

Dreyfus addresses the question whether with the development of Internet, distance learning might replace face to face education so that ”a first education will be available to everyone”.

Dreyfus' answer, on the basis of his earlier analysis of the process of acquisition of expertise, is that without the encounter with a teacher ”incarnating and encouraging involvement” the student may be unable to go beyond the first phases of this process. The main interest of the paper may in fact lie in the thorough critical examination the editors, Selinger and Crease, provide of Dreyfus' description of expertise and of its normative implications. According to Dreyfus, expertise is a deeply embodied competence which draws ”on prior concrete experience in a manner that defies explanation” and justification. Consequently, requiring from experts that they provide a rational justification of their judgments could only lead to a misrepresentation of their intuitive process of decision making. As Selinger and Crease show, however, Dreyfus' account is problematic in that it admits too much, in treating ordinary driving as a case of expertise, and not enough in precluding the possibility of being an expert in music without being musician. In the first example, the specificity of the professional driver is lost while the competence of the ordinary driver is idealized; in the second, it is the possibility of being 'expert in x' without being an 'expert x' that is lost and with it, the diversity of forms of expertise. Thus Dreyfus' model suffers from its generalist ambition of treating all embodied competence as a case of expertise and all expertise as a case of embodied competence.

As is aptly argued by Selinger and Crease, ”each real life controversy involving expertise takes the form of jockeying between those who advance claims of expertise to advance their authority and those seeking the right authority to whom to defer”. And to understand such a situation ”requires moving beyond the practical expert's point of view”.

The last paper of the second Part, by Selinger and Mix, is a critical review of Collins' concept of interactional expertise. Interactional expertise is characterized in terms of a certain amount of knowledge which enables someone to converse with expert practitioners of a certain field, but not to contribute to the science of the field. However, the authors object, even though AIDS activists would be considered by Collins as interactional experts, they did, contrary to what Collins' model would imply, make contributions in the field. Moreover, their being considered as interactional experts may in fact result from this productive interaction rather than be a prior characteristic.

It seems to me however that Collins could easily answer that it is their experience-based expertise which, together with the interactional ability they developed, made the AIDS activists' contribution possible. Their second point is more troubling for Collins' model. The identification of what counts as contribution to the field is much too diffuse and multiform for being reduced to Collins' two forms of contributory expertise. Some feminist philosophies of science, for instance, are clear examples of interactional expertise which aims to, and I think does, affect the development of science. A promising suggestion to deal with these cases is the distinction proposed by Selinger and Mix between direct and indirect contributions.

The last part of the collection contains papers by John Hardwig, Steve Fuller, Paul Feyerabend, Edward W. Said and Don Ihde.
          Endless is the list of the beliefs we have, for the truth of which we have no evidence. Is that irrational? In a provocative paper, Hardwig argues that ”rationality sometimes consists in refusing to think for oneself” and, consequently, that our conception of rationality has to be reexamined. It is not that the layman cannot sometimes inquire into the reliability of expert judgments, but in the best epistemic case, that will also amount ultimately, according to Hardwig, to relying on the opinion of an expert.

Even though the epistemic dependence he displays hurts our ideal of intellectual autonomy, Hardwig's argument has a compellingly realistic tone. However, it overlooks the intellectual resources one could demonstrate when crucially concerned with an issue, as AIDS activists are, as well as the crucial relevance specific experiences may bear to dealing with certain problems, as in the example Collins and Evans discuss of farmers regarding radioactive contamination of their environment.

These two cases undermine also Fuller's contention that the distinctive character of expertise is that experts' performances are somewhat self-vindicating. If the client's problem is solved, Fuller says, the experts takes the credit; if it is not, the expert can claim that the problem lies outside his or her range of expertise, suggesting that the failure was the client for not locating the right expert. But the development of procedures -- such as 'participatory research' -- in which 'laypeople', on the basis of direct experience of certain situations, interact with official experts testifies to the public acknowledgement of the general insufficiency of official expertise to address certain issues. Even though, admittedly, the still exceptional character of these procedures testifies to the entrenched resistance that this acknowledgement encounters.

The collection ends with a refreshing contribution by Ihde on the contrast between art or literary criticism and 'technoscience criticism' where, in contrast to the latter, the former, even though it can be harsh, is rarely considered as a general condemnation of art or literature. His diagnosis is that whereas art critics and literary critics are part of art and literary institutions, critics of science or technology are only regarded as outsiders.  His proposal to the philosophers comes very close, it seems to me, to advocating an 'interactional expertise', suggesting the possibility thereby of a contribution, even though maybe indirect.  But to interact, just as to dance, one needs a well-disposed partner. 

© 2008 Isabelle Peschard

Isabelle Peschard, University of Twente, (NL) and  San Francisco State University.