Projection and Realism in Hume's Philosophy

This book is an outstanding contribution to the question of Hume's realism. Kail argues that projection and realism are not mutually-exclusive notions, and thereby finds middle ground between projectivist and realist interpretations of Hume's philosophy. The interpretation will, however, be more gratifying to those of a realist tendency, since it amounts to incorporating projectivist elements into an overarching realism.

At the centre of Kail's project is a distinction between feature projection and explanatory projection. The two are not mutually exclusive, but the difference lies in the former involving the projection of features of mentality onto objects, whereas the latter offers a feature of mentality as an explanation for why someone takes the world to be a certain way. Thus, the explanation of religious belief in terms of fear of the unknown is not feature projection, because it does not project that fear into the world as a real property in it; but it is explanatorily projective because it explains the origin of the belief "non-detectively," in terms of a feature of mentality rather than of the world. The distinction matters because "projectivism" is commonly understood in terms of feature projection, and hence as essentially anti-realist; but explanatory projection is quite compatible with realism. A belief in invisible intelligent power may derive from fear of the unknown, a fear which is in no way a form of detecting such power—but there may be invisible intelligent powers nonetheless. The issue is not settled by offering an account which is explanatorily projective.

The religious example is apt here, because the first of the book's three sections is devoted to an examination of Hume's explanatorily projective accounts of religious belief and of the external world. The question is, are these beliefs both non-detective? Kail brings out the parallels between Hume's handling of these two issues—his aim in both cases is causal explanation of the belief in which the removal of psychological discomfort plays a key role, and in which vulgar versions of the belief are readily seen to be false—but nevertheless is able to offer an answer in the negative. Both types of belief are threatened by both semantic and justificatory objections, but, while these are fatal for religious belief, they are not for belief in the external world. Belief in the reality of the external world is able to survive the semantic threats because the supposition of continued external objects that resemble perceptions is coherent, and gains authority from the practical consequences of accepting what our senses present to us; and because the instability of the distinction between primary and secondary qualities does not resolve itself into a Berkeleian reduction of the former to the latter. And it is able to survive the justificatory threat by appeal to the category of natural belief. Religious belief, in contrast, succumbs to both threats. So, although both accounts are explanatorily projective, the former belief, in the reality of the external world, is not ruled out by Hume.

The second section is concerned with causal realism. Here also Kail identifies semantic and justificatory threats to realism. The semantic threat—that causal talk is inevitably empty, because Hume's denial of any impression of power reduces talk of power to that of a something we know not what—is met by what he calls the "Bare Thought." This is Hume's minimal characterisation of causal power: in Kail's words, "that feature that, were we acquainted with it, would yield a priori inference and render it inconceivable that the cause not be followed by its effect" (84). This is indeed minimal, but not empty: in looking for an idea of causal power, we do know what we are looking for. The justificatory threat is more delicately balanced. Hume's account of our idea of power involves both feature and explanatory projection, and, although this does not rule out realism, Kail argues that Hume's various remarks bearing directly on the issue are too ambivalent to settle things one way or the other. However, in a strikingly original discussion of Hume's second thoughts about personal identity in the Treatise's Appendix, Kail argues that Hume's worries there fall into place against a background of realism about necessary connection. The balance is thereby tipped away from agnosticism, firmly to favour causal realism.

The third section addresses moral realism. Kail's strategy here differs from the first two, in that the three chapters of this section are independent but mutually-supporting. The relation between them is expressed in their titles: "Gilding" (which addresses the nature of the projection involved, in particular the error theory implicit in Hume's appeal to the secondary qualities model of moral value); "The Gold" (the essential goodness of pleasure and badness of pain); and "The Golden" (the relational values of the useful and agreeable as the foundation of our moral sentiments). Again, Hume's account is explained as explanatorily projective (and incorporating elements of feature projection), but such as to license a realist interpretation. Kail thus opposes both the non-cognitivist interpretation of Hume on valuation and motivation, and its close relative, the purely dispositional account of secondary qualities. Against the former he argues that the projective character of Hume's argument requires that beliefs are involved in valuing, not merely desires—just as projective accounts of colour experience explain beliefs about objects' colours (176–7); and that, for Hume, reason cannot motivate not because it merely concerns belief, but because it cannot produce the essentially motivating ideas of good and evil (192). Against the latter he argues, in the same spirit, that Hume remains sufficiently in the debt of Hutcheson to hold that moral sentiments are responsive to objective features of the world, and it is for this reason that they can be corrected by reason—just as our sensory perspectives on the world can be rationally corrected (235–6). But this is a mitigated moral realism when compared to Hutcheson's because the objective features in question are the relational values of the useful and agreeable, informed by mechanisms of sympathy, rather than Hutcheson's providential teleology (240).

This is a densely-argued work, and as such not for the faint-hearted. But the effort is matched by the rewards, not least in those discussions which will provoke disagreement. It displays an impressive mastery of the Treatise, which is its primary focus, but also makes very effective use of Hume's other relevant works, including often-neglected works like the Natural History of Religion. It shows an equally impressive grip on Hume's immediate intellectual context, making informed appeal at relevant stages to Hobbes, Hutcheson, Leibniz, and (especially) Malebranche, and thereby arriving at more nuanced conclusions than afforded by a limited diet of Descartes, Locke, and Berkeley. Thirdly, and certainly not least, it makes very effective use of concepts, distinctions and examples from contemporary philosophy, wherever they can fruitfully be brought to bear. It is, moreover, very well written, with a fine feeling for metaphor. There is no doubt that it is a very impressive piece of work by a fine young philosopher.

If we leave aside niggling complaints about details, there is only one general criticism worth making. It is that the book lacks a conclusion which ties together the threads from the three sections. This would not have been the most difficult of tasks, since the raw materials are in evidence along the way: the treatment of the mental equipment from which we derive our picture of the world brings out Hume's sense of our imagination-grounded continuity with the animal world; and the account of action and morality is in the same spirit, emphasising the hedonism which, for creatures thus conceived, is the motivating force. (The wedge Kail drives between psychological and metaphysical hedonism—the latter embedding a necessity claim which the former lacks—is, from this point of view, less happy, since Hume's own account of the origin of the idea of necessity in habitual associations offers a bridge between the two.) These and other passing comments are building-blocks for a naturalistic conception of a creature afflicted by projective errors, but capable nonetheless of a reasonable grip on its real circumstances. It is a shame these thoughts are not brought together to unify the account offered—but this is in no way to deny that Projection and Realism in Hume's Philosophy is a major contribution to the interpretation of Hume's philosophy.