1 Cf. Hollis, (1977) 136–8.
2 Cf. Parfit, (1984) 73–4.
3 Cf. Graham, (1986b) 21–2.
4 Cf. Mackie, (1977) 107–14, Lukes, (1985) 32, Warnock, (1971) 26.
5 For a more patient discussion of intrinsically irrational desires, see Parfit, (1984) 120–6.
7 Williams, (1981a) 106–9.
10 Cf. Nagel, (1986) 152, Parfit, (1984) 143.
11 Nagel, (1986) 159–60, 171–2.
12 For example, Nozick, (1974) 33.
13 Graham, (1986a) 95–116.
14 Graham, (1986b) 29–31.
15 Graham, (1986a) 110–14.
16 Even F. H. Bradley, in his enthusiasm for the individual's active identification with the collective, recognized that I might find myself in a collective which was in a ‘confused or rotten condition’ (Bradley, (1962) 203). The grounds for active identification with its actions would in those circumstances be weak to vanishing point.
17 The analogy is not perfect, and more than one phenomenon might be placed under the heading of invisible collectives. A number of people might be entirely unaware that they constituted a collective; or they might be aware of this, but entirely ignorant of the nature of the acts performed by the collective; or ignorant of some of those acts; or ignorant of some possible descriptions of those acts.
18 This was enunciated by Olson (1965) and discussed more recently by Buchanan (1982) and Elster (1985).
19 Cf. Buchanan, (1982) 89.
24 Discussions of the free-rider problem tend to be a mixture of theory intended to explain people's behaviour and theory intended to establish how people ought to behave. The two are connected, because we presumably believe that people sometimes behave rationally; but they are distinct. The emphasis in my own remarks is on understanding the conceptual equipment potentially available to us on such occasions, rather than on understanding why people currently behave as they do. It is explanatory individualism that Elster sponsors (Elster (1985) 5, 359), and he is critical of that methodological collectivism which ‘assumes that there are supra-individual entities that are prior to individuals in the explanatory order’ (ibid. 6). Though I might disagree with Elster about this, the claims I make here do not commit me to doing so. My claim is that there are prior collective entities, whose significance is not properly understood, in the explananda rather than in the explanans. Indeed, for reasons mentioned at the beginning of section III, the lack of assimilation of the considerations I am outlining makes an explanation of people's actual behaviour in individualist terms perhaps more plausible. People behave as they do partly because they think in individualist terms. I believe that they think in those terms where they should not, and therefore sometimes behave as they should not. But I do not underestimate the difficulty of changing these habits. See further note 31.
25 MacIntyre, (1981) 146. But don't all people share an interest, say, in the preservation of the earth as a safe habitat, and don't they then constitute a potential collective in the way described earlier? In a sense, yes. But if some of the claims Marx makes about the nature of class-divided society are correct (if, for example, people have irreconcilably different views and interests in the matter of the form in which the earth should be preserved), then calls to actualize this potential collective are premature.
26 See, e.g., Lukes, (1985), Miller, (1984).
27 Marx, and Engels, (1967).
31 Buchanan entertains the possibility that the proletarian might ask not ‘What should I do?’ but ‘What should we do?’ and thus avoid the free-rider problem. But he objects that this assumes what Marx must establish: that this new way of conceiving decisions could be arrived at in the isolating, egoistic environment of capitalism (Buchanan (1982) 191, n27). That seems to me to put the problem in exactly the right place. Even if there is a theoretical resolution available, will it have any practical effect? One part of the answer (and one way of making sense of Marx's own life) is that Marx saw a role for his own theories in helping to shape and change proletarian consciousness. In that respect he was optimistic: he could not be expected to foresee the dire consequences of his posthumous Leninization (cf. Graham (1986a) 204–30).
32 In a number of respects the position I adopt coincides with that adopted by Miller, (1984), though his discussion is mostly couched in terms of individual human beings and their different class locations rather than corporate entities themselves.
33 For helpful written comments on an earlier draft I am extremely grateful to David Archard, G. A. Cohen and Martin Hollis.