Since around the turn of the twenty-first century, philosophers have become increasingly occupied with using experimental methods to answer philosophical questions. Specifically, experimental philosophers have drawn upon the design, tools and findings of experimental psychology and the behavioural sciences in general in order to shed light on the phenomena that affect actual judgments about philosophical cases. Do folk think that George should shove the portly man onto the track in in order to save ten scrawny strangers? Do they think the farmer knew that his prized cow was in the field all along? Do they consider the hermetically-sealed symbol shifter to understand Chinese? Furthermore, do responses to such cases vary between different cultures, ages and genders or when presented in different ways?

          In his new book, Édouard Machery, a long-time proponent of the experimental philosophy (‘X-Phi’) movement, claims that the cases with which many philosophers and non-philosophers are familiar have been deployed from the armchair in order to answer questions concerning the modal facts bearing on knowledge about what is necessary and what is possible. Is it the case that necessarily someone knows that p if and only if they have a justified true belief that p? Is it the case that necessarily the extension of natural kind terms is determined by empirical properties? Is it the case that necessarily descriptivism about the reference of proper names is false?

Since the inception of X-Phi, experimental philosophers have provided evidence that judgments about cases are subject to demographic and presentational effects. On that basis, the so-called method of cases, which, according to Machery, aims to determine what facts hold and to distinguish competing philosophical theories, should be restricted. Machery’s latest monograph is an innovative and substantial contribution to the project of ‘negative X-Phi’ in that it argues for a ‘radical restriction’ on the method of cases as the basis for modal claims.

He starts by arguing that the judgments we make in philosophical cases are no different to the beliefs we hold in response to everyday situations. There is nothing special about the judgments elicited by philosophical cases; these judgments do not rely on irreducible intuitions or conceptual mastery or a priori judgment (chapter 1). Machery goes on to provide a systematic review of the literature in experimental philosophy in order to support the claim that judgments elicited in response to the most well-known philosophical cases vary according to demographic factors, the order in which cases are presented and the way in which questions about them are framed (chapter 2).

The results of the literature review provide the inductive basis for two arguments that support his radical restrictionism. Firstly, because judgments elicited by philosophical cases are influenced to a large enough degree by demographic and presentation variables, these judgments are unreliable. Furthermore, since the method of cases aims to distinguish between competing philosophical theories, and since, according to Machery, philosophers from different camps tend to agree when responding to mundane situations, it is not an accident that philosophical cases feature ‘disturbing characteristics’. For Machery, philosophical cases express unusual situations, adopt irrelevant and tendentious narrative elements and pull apart properties that go together in everyday life (physical violence from doing-more-harm-than-good; truth and justification from getting-things-right-by-luck, and so on). On the basis that the disturbing characteristics of philosophical cases elicit judgments that we would not be expected to make in everyday life, these judgments are fundamentally unreliable (chapter 3).

The second argument is that since many proponents of the method of cases do not take seriously the variations in judgments about philosophical cases, the premises of their arguments should not be taken seriously. The point is that either there is a disagreement between philosophers and non-philosophers or they talk about different things when making judgments in philosophical cases. In either case, Machery argues that the method of cases is more likely to enshrine philosophical prejudices than articulate modal facts (chapter 4). If peers are likely to disagree about a philosophical case or if judgments about a case are plausibly severely unreliable, then we ought to suspend judgment about answers elicited by a particular case. Indeed, according to Machery, the consequent of that conditional applies to all philosophical cases (unless certain cases are known to elicit either a reliable judgment or agreement amongst peers). Alternatively, if it is the case that peers do not disagree but refer to different properties (epistemic, moral, semantic, and so on) when they respond to philosophical cases, then, as Machery sees it, philosophers should abstain from the method of cases, ‘reorient their philosophical interests and stop theorizing about the philosophically familiar’ (p. 148).

Machery articulates what a suitable ‘reorientation’ might look like. After responding to several common arguments for the method of cases (chapter 5), he sets out his own particular vision for ‘positive X-Phi’. For Machery, the method of cases is part-and-parcel of the modal immodesty of contemporary analytic philosophy. Although the purported authority of modal claims involving metaphysical necessity or possibility allows philosophy to establish distinct research projects on which scientific practices cannot supervene, the fact that the main philosophical method for identifying modal facts cannot facilitate such knowledge means that we should adopt more modally skeptical positions. Machery’s radical restrictionism concerning the method of cases undermines modally immodest philosophical enterprises, especially in less ‘naturalistic’ areas of philosophy. For Machery, standard philosophical issues, specifically, issues that preclude investigation by means of scientific methods, are beyond our epistemic reach (Chapter 6). Advocating philosophical modesty, he calls for a ‘naturalised’ approach to conceptual analysis (chapter 7). If the concepts actually possessed by folk are interpreted as psychological entities – subsets of people’s belief-like states (which Machery calls ‘bliefs’ (p. 210)) – and thereby difficult to access for those that possess them, then, according to Machery, we can make progress in analyzing concepts by using the designs, tools and findings of experimental psychology to make the implicit structures underpinning our belief-like states explicit. If the first goal of a naturalized conceptual analysis is descriptive, then, for Machery, the second goal is prescriptive. If we assume that concepts are psychological entities, then, as is the case with our beliefs, they can contain false information. According to Machery, naturalized conceptual analysis would not only assess the validity of inferences that concepts dispose us to draw by exposing the ‘bliefs’ that constitute them and the relation of those ‘bliefs’ to empirical knowledge about the world, it would set about revising those inferences that are in some way deficient. In response, the proponent of modally immodest philosophical enterprises might argue that in order to assess the validity of our inferences drawn from concepts like ‘knowledge’, ‘justification’ and ‘truth’, we would be required to gain knowledge about knowledge, justification and truth, whereby the processes for accessing such knowledge would be tantamount to the practice of making modal claims.

This leads us to consider whether Machery’s overarching modal scepticism is reasonably justified. One can accept his case against the method of cases and yet still remain open to the possibility that knowledge of metaphysical modality is not a fruitless non-naturalistic goal. Even though Machery considers some alternative routes to modal knowledge in chapter 6, it would be difficult to reasonably claim that he has justified the inductive leap from the modally-defunct method of cases to a totalizing modal skepticism. Indeed, there are other inferentialist, essentialist, explanationist and principle-based approaches to modal knowledge that do not rely on the method of cases and (when taken at face value) do not seem to fall neatly under the categories of alternative approaches discussed by Machery in chapter 6. I accede that this is only a minor contention. After all, neither the validity of Machery’s case against the method of cases nor his proposals for a naturalized conceptual analysis depend on him demonstrating that modal skepticism is true.

Nevertheless, the case against the method of cases does depend on the reliability and validity of the experimental findings that Machery uses as the inductive basis for his arguments against philosophical dogmatism and parochialism as well as for his claims regarding the reliability of judgments in philosophical cases. It also depends on the reliability and validity of the inductive leap from the unreliability of judgments elicited by those cases that have been analyzed by experimental philosophers to the fundamental unreliability of judgments elicited by philosophical cases in general. Regarding the findings of experimental studies that Machery discusses in chapter 2, he acknowledges that the X-Phi movement has examined only a limited number of cases, of which few have been systematically studied, and many have not been subjected to repeated draws (p. 85). He also mentions deficiencies with the experimental design of these studies, though he does not go into detail about what these deficiencies are. Consequently, Machery claims that ‘the arguments based on this empirical record…must assume that it is representative of the phenomena that characterize the judgments elicited by philosophical cases in general’ (p, 85). This assumption may be too much of a ‘leap of faith’ for some experimental researchers, that is, without further details of the limitations of the experimental designs and without repeated trials. That said, repeated draws come with their own risks; primarily, an increase in the likelihood of reactive behavior on the part of the study participants. Consequently, as is commonly understood, an experimental design with repetition does not automatically guarantee or improve the validity of a study. Furthermore, what is important about experimental practice is not so much observational results and the products of experiments, but the design and implementation of experimental setups that reveal or produce causal relationships in a reliable manner. Consequently, even when Machery argues that ‘judgments elicited by typical philosophical cases are similar to experimental artefacts – outcomes of experimental manipulations that are not due to the phenomena experimentally investigated, but to the (often otherwise reliable) experimental tools used to investigate them’ (p. 90), it will not take an experimentally-oriented researcher long to suggest that unless the implementation of the experimental design is successful, the same holds true for the judgments elicited by X-Phi surveys.

As previously mentioned, Machery’s inductive leap from the unreliability of some of the specific judgments of studied cases to the fundamental unreliability of judgments elicited by philosophical cases in general appeals to the similarity of philosophical cases in the sense that they exhibit disturbing characteristics (pp. 111-120). Firstly, it is not clear how the presence of such characteristics relates to the inductive leap. Secondly, Machery does not investigate whether demographic and presentation variables influence judgments in usual, everyday situations. He briefly considers this latter issue but quickly dismisses it (p.129-30), claiming, with no evidence, that ‘people do actually agree about everyday situations eliciting the application of concepts of philosophical interest’ (p. 130). Much rides on the distinction between unusual judgments elicited by philosophical cases and judgments in ‘everyday life’. If it were found that people do not, in fact, agree in their philosophical judgments elicited by everyday situations, the case against the method of cases would lose some of its potency in the sense that one might have reason to suggest that the method of cases is not epistemically inferior to other philosophical methods employed to determine which facts hold in certain situations.

Assuming that the principle aim of experimental methodologies is to make reliable causal inferences about the effects of some particular causal relationship, one way in which Machery could respond to these two interrelated problems is by explaining the causal factors that operate between specific disturbing characteristics and (what are otherwise considered to be) reliable processes of judgment-making. Many philosophers of the social and behavioral sciences suggest that the causal inferences within an experiment are reliable if they are the results of causes that happen under the governance of a causal relationship, which, in turn, results from an underlying causal structure. To warrant the belief that a causal inference can be extrapolated from a particular experiment or small set of experiments to situations involving philosophical cases in general, proposed techniques of causal inference stress that what is required is knowledge of the nature and stability of the causal factors involved as well as knowledge of the background causal structures that operate in judgments elicited by philosophical cases. Such knowledge cannot be underwritten by any particular experiment or set of experiments. As result, both experimental and non-naturalistic philosophers may have reason to question whether the inductive leap has been established.

Overall, it seems that what Machery presents is a manifesto – a tightly-and-powerfully-argued, eminently readable, innovative manifesto – for the X-Phi movement rather than a full-blown evidence-based policy. However, unlike the majority of contemporary political manifestos, it deserves to be taken seriously. By consolidating findings drawn from individual experimental studies, by incorporating various approaches to experimental philosophy and by advocating a particular picture of what the future of ‘positive-X-Phi’ could look like, this book serves not only as the foundations on which experimental philosophers can build, but as a provocative challenge to more common approaches to theorising in the tradition of analytic philosophy.

© 2018 Jonathan Lewis

Jonathan Lewis, Dublin City University