‘Philosophy’, wrote Bertrand Russell at the beginning of The Problems of Philosophy, ‘is merely the attempt to answer…ultimate questions, not carelessly and dogmatically, as we do in ordinary life…but critically…after realising all the vagueness and confusion that underlie our ordinary ideas’, And he goes on, of course, to question whether he really knows that there is a table in front of him. When interviewing candidates I sometimes got them to read this chapter and asked them, first, if they found it difficult (I hoped they did), second, if dull (I hoped not), and, third, if daft. While I regarded a positive answer here as something of a disqualification, Stephen Boulter would seemingly think otherwise. His book is a sustained defence of common sense against the ‘truly bizarre claims’ with which ‘the history of philosophy is littered’ (xii), and to that end he presents a very different account of what philosophy is from Russell's.

The first half of Boulter's book is, indeed, an altogether too rare essay in metaphilosophy. Philosophy principally consists, he believes, in the attempt to provide a ‘Big Picture’ whose materials come from the sciences, the humanities and, of course, common sense. Philosophy's task is to coordinate such data and, especially, to resolve the aporiai to which they give rise. Here common sense beliefs play a special role for philosophers of Boulter's persuasion, since their preferred resolution of an aporia is one that preserves them intact, for ‘one should not abandon a common sense belief on the strength of ‘mere’ philosophical arguments' (23), which provide no new data. Rather, common sense beliefs function as default positions, shifting the burden of proof on to their critics. It is, in fact, to the data of evolutionary biology that Boulter somewhat surprisingly turns to support this latter claim. Our common sense beliefs are, he suggests, likely to be true because having them will have been favoured by natural selection. By contrast, ‘there was no selective process, natural or sexual, for philosophical ability in the ancestral environment’ (44).

Boulter next moves on to consider why philosophers are so prone to the sort of error illustrated by their ‘truly bizarre claims’. Here he identifies the difficulty of second order thought as well as a lack of agreed methods or criteria for correctness in the subject and a proneness to fall under the grip of a particular theory. But why do philosophers then take their claims to be discoveries rather than reductiones ad absurdum? Not only are various intellectual vices cited in answer but also ‘the lack of economic or social constraint’ arising from the fact that ‘the philosopher can afford …to cut his or her links to the real world as no scientist can’ (75). And in a later chapter Boulter argues that assumptions carried over from other disciplines may lead us into error, suggesting that modern denials of natural necessity stem from medieval theological doctrines.

Having completed the metaphilosophical part of his work Boulter turns to what an uncharitable critic might regard as the over-ambitious project of disposing in some twenty pages each of five theses contrary to common sense: the rejection of metaphysical realism, semantic anti-realism, eliminative materialism, the denial of free will, and moral non-cognitivism. It would be more generous, however, to view these chapters as illustrating the methods of common sense philosophy which he advocates. Thus Boulter defends metaphysical realism against John McDowell and other neo-Kantians on the grounds that ‘a pre-condition of visual perception itself is precisely what the common sense philosopher maintains, namely, the mind-independent existence of a featured, or pre-packaged world’ (98) to which human beings have access. And he argues for this once more from the findings of science – cognitive psychology and neurophysiology – which explain what evolutionary biology would lead us to expect, that, just like other animals, we can track features of our environment.

Similar considerations crop up elsewhere; for example, in defending free will Boulter argues both that homo sapiens' sociality demands a distinction between voluntary and involuntary acts and that the adaptability of the species to different environments relies upon the development of decoupled representations which make deliberation possible. (Here he might have mentioned the good fortune of our Pleistocene ancestors in having experienced the rapid climate changes that led to fluctuating environments.) Furthermore in arguing for moral realism Boulter adverts to the biological advantage of tracking moral facts as conducive to ‘the distribution of items of non-moral value among as many people as possible’ (186).

Indeed there are so many arguments of this character that it is not clear that Boulter really depends enough upon his avowed programme of treating common sense as the default position, which only powerful counter-arguments could unsettle, and then showing that these counter-arguments are not forthcoming because erroneous. For although Boulter maintains that once the errors of opponents have been identified the common sense conclusion ‘wins by default…and so requires no independent proof’ (177), independent proof is what his sociobiological arguments seem to offer. Boulter goes so far as to suggest that ‘a belief ought to be counted as a common sense belief and treated as a default position if that belief has ecological and social fitness value, (159). Here, though, a distinction needs to be made. For sometimes Boulter appeals to such considerations to argue that a belief – in free will, for example – falls into the common sense category, at others that a common sense belief – in moral facts, say – is true. The connection is presumably that such beliefs would not normally have conferred fitness value and thus entered common sense unless they were true.

A view seems to emerge here of common sense, unless confuted by reliable science, as constituting our Big Picture because imprinted by evolution. It turns out to be remarkably similar to Russell's story of instinctive beliefs which we have no reason for rejecting so long as they can be seen to harmonise into a larger system. The difference is that, unlike Boulter, Russell does not regard instinct as evidence of truth, only, as Boulter ostensibly holds, of providing the default position for philosophy.

It is worth turning back, then, to examine what Boulter's common sense beliefs really are. He exemplifies them with lists drawn from Thomas Reid and G. E. Moore, remarking that they may not be consciously formulated by ordinary people – as, for instance, ‘Mental acts are ontologically dependent upon bodies’ (29) presumably is not – but that they are presupposed in people's everyday beliefs. This qualification rules out ‘just any belief which happens to be accepted pre-theoretically by virtually anyone’ and restricts common sense to those beliefs which constitute ‘the fundamental elements, principles or cornerstones of the conceptual scheme lying behind the views and actions of the ordinary person' (27). It is this scheme which is, apparently, the basis of the philosopher's Big Picture. But now a very different method of arriving at it is suggested from the one adopted by Boulter, and certainly not one that needs the support of socio-biology. Rather, what seems to be needed is a transcendental method for discovering what is presupposed by our ‘views and actions’ – the sort of project that Wittgenstein undertakes in On Certainty, which, disappointingly, is given no extended discussion in Boulter's book.

This omission is especially surprising in view of Boulter's anti-McDowellite theme in his defence of metaphysical realism. For just as Boulter, rightly to my mind, rejects a view of perception as intrinsically conceptualised, because tied to more primitive actions, so he could have related his analogous view of common sense beliefs to Wittgenstein's denial of a genuinely propositional status to those that form the ‘foundation’ or ‘bedrock’ of our thought. Had he done this, however, he would have had to reconsider two of his claims. The first is that our common sense beliefs can in principle be overthrown by scientific discoveries. The second, and more important, is that common sense delivers facts about the world, in the way in which Moore supposed they did, rather than functioning as rules, as Wittgenstein argues. But then Boulter's whole conception of philosophy as aiming at a Big Picture which systematises such facts would be called into question, especially if we think of our lives, as Wittgenstein did, as made up of a variety of practices with their associated diverse language games.

Boulter would no doubt hold that this position itself runs counter to common sense, for common sense is, on his view nothing if not realistic. Yet is this right? It is surely one thing to identify a presupposition of thought and action such as ‘human beings have minds’ (29). It is quite another to impute an analysis of this to ordinary folk. Perhaps they would say that this is a fact. But remember that we must not treat as common sense ‘just any belief that happens to be accepted pre-theoretically by virtually anyone’, like the belief that human mentality is a fact. We need to ask if such a realistic analysis is really presupposed in people's lives. So far as I can see Boulter's arguments for realistic analyses of common sense beliefs mostly turn on evolutionary considerations and the thought that if the beliefs so conceived were false then our success would be mysterious. But falsity is not the only alternative here. Indeed arguably it is not an alternative at all if the falsity of common sense beliefs is unthinkable because of their status as presuppositions of all our thought. Boulter does not, I think, seriously consider what might be involved here.

The worry one might have, then, about this style of philosophy is, of course Russell's. For him philosophy is not to be studied for definite answers but to counter dogmatic assurance. There is plenty of speculation in Boulter's book, but it is not philosophical speculation designed to criticise established views by setting contrary ones against them. Arguably Russell is right to see this critical task as essential to philosophy rather than Boulter's constructive one. To suppose otherwise is to risk the kind of intellectual presumption that allows philosophers to think they can deliver conclusions, on ethics, for example, that only practitioners are qualified to do. Thus Aristotle, whom Boulter questionably regards as the founding father of common sense philosophy, holds the practically good man, not the philosopher per se, to be the source of true moral judgments. And the endoxa which Aristotle seeks to preserve are not, as Boulter supposes, just common opinions, but those of the wise which take precedence over the ordinary person's.

Despite these strictures Boulter has performed a valuable service in writing on a much neglected topic. Metaphilosophy is surely a part of philosophy but thankfully one on which common sense has nothing to say. To that extent Boulter's position is unsupported by his preferred underpinnings. But Boulter may be right to think that his line runs counter to that of established views, and so it serves a valuable critical function. His book is clearly written, though unfortunately it presupposes too much specialist knowledge to be intelligible to ‘the philosophical laity’ (19) on whose judgments he sets such store. The book is also full of interesting material on the substantive issues consideration of which is beyond the scope of this review. I particularly like his discussion of the separability thesis that led Hume to deny natural necessity and Moore to deny moral naturalism. This error he traces back to the Bishop of Paris's condemnations in 1277 of any view that limited God's power by rendering Him unable to do whatever was logically possible. Without this theological basis the principle lacks credibility. The example illustrates how much more learned this book is than one might expect from a defender of common sense. Though reverencing some naïve opinions, Boulter's style is not, unlike Moore's, even seemingly naïve.