Self-Knowledge for Humans

Quassim Cassam laments the focus that philosophers place on so-called trivial self-knowledge, e.g. knowing that one believes it is raining or knowing that one wants salmon for dinner. ‘Substantial’ self-knowledge, by contrast, includes knowledge of one's values, character traits, aptitudes, emotions, abilities, etc. Cassam claims that while the latter is of great interest to non-philosophers, contemporary philosophers largely ignore it. A major and admirable aim of the book is to correct this state of affairs.

Cassam hypothesizes that the cause of this is the belief that trivial, but not substantial, self-knowledge is epistemically special in that it is not based on evidence. Even granting this alleged asymmetry, Cassam denies that it warrants philosophy's neglect of substantial self-knowledge, for ‘[t]here is more to philosophy than epistemology’ (p. viii). In fact, however, Cassam denies the asymmetry, defending what he calls ‘inferentialism’.

Inferentialism claims that a key source of knowledge of one's standing (i.e. non-occurrent) mental states is inference from behavioural evidence, other standing states, and occurrent internal promptings, such as visual imagery, inner speech, feelings, etc. For example, you might come to know that you intend to vote for Candidate A on the basis of your having donated to her campaign (behaviour), your belief that she is the best candidate (standing state), and your visual image of voting for her on the upcoming election day (internal prompting). Such inference is mediated by a background theory of mind and need not be conscious.

Cassam claims that one can infer standing states from internal promptings only if one has ‘access to’ those promptings, and he describes as ‘lazy’ the inferentialist who is silent regarding how such access is achieved. Cassam's answer is (again) inference: we gain access to internal promptings by inferring them from behaviour, standing states, and other internal promptings.

To illustrate, you might infer that a certain unpleasant feeling (internal prompting) is jealousy of a friend (as opposed to, say, anxiety about your future) on the basis of your having just finished talking with your friend about his new house (behaviour), your desire to have much of what he has (standing state), and your expression in inner speech that you wish you were more like him (internal prompting). Because your access to these mental states/occurrences is inferential too, the structure of self-knowledge is highly holistic on Cassam's view.

When discussing lazy inferentialism, Cassam writes: ‘[a] much more serious charge is that lazy inferentialism is incomplete because it presupposes other self-knowledge which it doesn’t seek to explain’ (p. 168, original emphasis). I worry, however, that Cassam is to some extent guilty of this very charge. Although he claims that our access to internal promptings is inferential and interpretive, he is silent regarding how we gain access to the internal data that is subsequently interpreted.

For example, even if one infers that a given feeling is one of jealousy (as opposed to anxiety), this does not explain how one initially gains access to that feeling. Cassam seems to take for granted our access to our visual imagery, inner speech, feelings, and so on, whose precise nature, on his account, we subsequently infer. Moreover, the correct explanation of such access/knowledge might have negative implications regarding inferentialism's plausibility.

The negative part of Cassam's book primarily focuses on the ‘transparency’ approach to self-knowledge (though he also criticizes ‘inner sense’ in ch. 10). This approach originates with Evans (1982), who claimed (roughly) that one can answer the inwardly directed question ‘do I believe that P?’ by answering the outwardly directed question ‘is P true?’ In order for this approach to generalize beyond the attitude of belief, Cassam, following a suggestion by Finkelstein (2012), interprets the transparency theorist as recasting her proposal in terms of rationality: for any given attitude type, ϕ, when considering the inwardly directed question ‘do I ϕ that P?’, one can instead consider the outwardly directed question ‘ought I rationally to ϕ that P?’ Cassam calls this version of Evans’ proposal ‘rationalism’.

Rationalism is criticized often and forcefully throughout the book. One criticism alleges that determining whether one has a particular attitude is often easier than determining whether one ought rationally to have that attitude, which suggests that we do not primarily determine the former by determining the latter. Cassam's most significant criticism, though, concerns ‘the Disparity’, i.e. the alleged significant mismatch between our actual attitudes and the attitudes that we ought rationally to have. He claims that this mismatch—which he supports with a combination of empirical data and folk psychology—prevents us from knowing our attitudes by reflecting on which attitudes we ought rationally to have. This is, in part, why Cassam has written a book on self-knowledge suited for actual human beings (not homo philosophicus). The Disparity and its relation to rationalism constitute the bulk of the book's first half.

Cassam makes a very strong case against rationalism (so understood). I worry, though, that he has overlooked an important nearby view best represented by the work of Alex Byrne (e.g. 2011, 2012), who unfortunately receives little of Cassam's attention. Byrne's view, like rationalism, is outwardly directed. Unlike rationalism, however, it does not concern rationality.

Consider Byrne's treatment of knowledge of desire and intention. Oversimplifying a bit, Byrne (2012) claims that we can know what we desire by considering what is desirable (i.e. having the qualities which cause a thing to be desired). Regarding intention, Byrne (2011) claims that we can know what we intend by considering (without relying on evidence) what we will do. He elsewhere offers similar suggestions for knowing what we see and think. Because these outwardly directed considerations do not concern rationality, the Disparity is not obviously relevant.

The issues here are of course complex, and so there might ultimately be good reasons to reject Byrne's approach. But Cassam's decision to not engage with it is unfortunate, for neither of the aforementioned objections to rationalism straightforwardly applies to it. First, determining whether P is desirable is not obviously more difficult in general than determining whether one desires that P. Secondly, the Disparity does not make improbable the possibility that one's desires reliably correlate with (or ‘match’) one's (fallible) judgements of desirability. These points apply, mutatis mutandis, to Byrne's treatment of intention, seeing, and thinking. An important alternative to Cassam's brand of inferentialism thus remains standing.

Despite these concerns, I highly recommend Cassam's book. It provides a deep, yet admirably accessible study of self-knowledge and surrounding issues, and is full of helpful and often surprising insights. Finally, although I regret not having the space to say more about Cassam's fascinating discussion of substantial self-knowledge, he certainly succeeds in demonstrating that it merits greater philosophical attention.¹

Footnotes

REFERENCES