Hume's Dispositional Account of the Self

ABSTRACT

This paper will argue that Hume's notion of the self in Book 2 of the Treatise seems subject to two constraints. First, it should be a succession of perceptions [THN 2.2.1.2, 2.1.2.3]. Second, it should be durable in virtue of the roles that it plays with regard to pride and humility, as well as to normativity. However, I argue that these two constraints are in tension, since our perceptions are too transient to play these roles. I argue that this notion of self should be characterized as a bundle of dispositions to our perceptions, such that these dispositions are durable and counterfactual-supporting. I argue that Hume confused his ‘philosophical’ notion of dispositions, as nothing above and beyond their effects, with the thicker notion of dispositions to which the passions respond—which explains his mistaken commitment to the durability constraint.

KEYWORDS:

• Hume
• self
• normativity
• dispositions

Notes

¹ ‘THN’ refers to the Treatise of Human Nature, ‘EHU’ to the Enquiry Concerning Human Understanding, ‘EPM’ to the Enquiry Concerning the Principles of Morals, and ‘EMPL’ to Essays Moral, Political, and Literary. Arabic numerals refer to section and paragraph numbers (EHU and EPM); to book, part, section, and paragraph numbers (THN); or to page numbers in the Miller edition of the Essays (EMPL).

² One might take ‘bundle of perceptions’ to refer to the synchronic self and ‘succession of perceptions’ to refer to the diachronic self.

³ This is noted by Penelhum [1976: 9], Baier [1979: 235], and McIntyre [1989: 545].

⁴ Moral sentiments are outside the domain of reason, but would not constitute ‘facts’.

⁵ I say more about this in Qu [2012].

⁶ See McIntyre [1989], Baier [1991: 130], Garrett [1997: 168], Rawls [2000: 87], Carlson [2009], Schmitter [2009: 235], and Alanen [2014].

⁷ Hume also notes our ‘intimate … consciousness’ of the self [THN 2.1.2.2]. As I argue in Qu [forthcoming a], Hume's usage of ‘conscious’ and its cognates often relates to that immediate awareness that we have of our perceptions.

⁸ One seeming difference between ST and SP is that the latter is sometimes described as encompassing body [e.g. THN 2.1.9.1]; Penelhum [1976: 10] takes Hume to be inconsistent in this respect. In contrast, I think that Hume does not explicitly refer to body in THN 1.4.6 because of the sceptical context of this section, occurring just a few sections after a discussion of scepticism about the external world [THN 1.4.2]. Thus, he probably treats body as he does any other external object—that is, as a series of perceptions. In Book 2 Hume is less strict about this issue, and so he sometimes mentions body as distinct from a series of perceptions. Another possibility (suggested by an anonymous referee) is that Hume's notion of the imagination is borrowed from Malebranche [Wright 1983; Kail 2008], and thus involves the effect of the motion of animal spirits in the brain: given that ST is the self ‘as it regards our thought or imagination’, it might be seen to incorporate some notion of body. For disagreement, see Capaldi [1989: 168] and Alanen [2014: 13], who take body to be crucial and novel to SP.

⁹ Also puzzling is why Hume speaks of ‘the idea, or rather impression of ourselves’ [THN 2.1.11.4], since THN 1.4.6 argues that we have no such impression. Garrett [1997: 169] argues that the idea of a self is an abstract idea, and any impression we have could serve as this abstract idea. Carlson [2009: 177] argues that the ordinary person's vague idea of themselves as their perceptions might be sufficiently enlivened that it approaches the status of an impression.

¹⁰ Thanks to Jennifer Marušić for pressing this point.

¹¹ Thanks to an anonymous referee for pressing me on this.

¹² Of course, if the self comprises durable qualities, the self will correspondingly be durable.

¹³ In response, Pitson [2002: 90] objects that passions do not have ‘genuine duration’.

¹⁴ Loeb [2002] sees vivacity as fundamentally a dispositional rather than experiential notion. See Marušić [2010] for some excellent criticism of this view.

¹⁵ Shaw [1989], Smith [1994: 112], and Stroud [1997: 164–5] claim that the calm passions are completely insensible; I argue against this in Qu [forthcoming b].

¹⁶ A further problem for McIntyre's account is that it struggles to account for natural abilities, which Hume considers ‘on the same footing’ [THN 3.3.4.1] to moral virtues, but many of which have no obvious relation to the passions. In a recent talk, McIntyre [2015] states that she now believes Humean character to encompass other mental qualities, such as beliefs and natural abilities.

¹⁷ Thanks to an anonymous referee for raising this crucial point.

¹⁸ Of course, if the effect of the disposition were durable, then the disposition would be durable itself, since the two are identical. I will argue that the relevant effects are perceptions, which are transient.

¹⁹ The notion of possibility to which Hume appeals in THN 2.1.10.6–7 is nothing like metaphysical or even nomic possibility, which would be too weak. What Hume means is, rather, a loose and everyday sense of ‘possible’ as having a non-negligible chance of occurring—e.g. ‘it is possible that the medication will have side-effects’, or ‘it is impossible for Leicester to win the Premiership’. The circumstances pertinent to the relevant counterfactual will correspond to this form of possibility. To say that someone has a disposition to X is to say that X would occur under circumstances that are possible in this sense.

²⁰ Notably, Hume claims in THN 2.1.10.7 that a person who did not actually harm me ‘philosophically speaking’ had no power to do so even if this person could have harmed me. This clearly indicates that Hume's official notion of powers is not counterfactual-supporting.

²¹ Purviance [1997: 208] argues that a bundle theory of character is problematic, because it would face the same problems that ST faces in the Appendix. I do not find this objection compelling, since Hume finds that ‘this difficulty is too hard for my understanding’ [THN App.21]. It would be surprising if Hume did have a quick and easy resolution to this difficulty in Book 2; it seems that he simply brackets the issue and moves on with life. Ainslie [2005: 152–3] argues that one must give an account of why Hume's second thoughts in the Appendix are written as directed only at his Book 1 account of the self. I think that this is simply because the problem directly arises from the theses explored in Book 1, which is not to say that it does not have deeper ramifications.

²² Strictly speaking, Hume thinks that certain qualities must be possessed to an uncommon degree for us to be proud or humble of them [THN 2.1.6.4]. Nevertheless, if benevolence to an uncommon degree is a virtue and is a part of SP, then benevolence to even a common degree should also be part of SP.

²³ See Endnote H to EHU 9.5, where Hume discusses how people may differ in their powers of causal reasoning.

²⁴ Hume points out in THN 3.3.4.13 that the memory does not particularly affect our attributions of virtue and vice. Nevertheless, he only says that memory ‘has the least virtue or vice in its several degrees’, not that it has none: he notes that, when unusually good or bad, it still merits praise and blame.

²⁵ Thanks to Jennifer Marušić for pressing me on this point.

²⁶ Thanks to an anonymous referee for discussion of this.

²⁷ I would like to thank Jennifer Marušić and Ramona Winter for some excellent and detailed comments on the paper. A debt of gratitude is also owed to the NUS reading group, which was instrumental in improving the paper—Michael Pelczar, Tang Weng Hong, Ben Blumson, Neil Mehta, and Neil Sinhababu. Thanks also to an audience from the 4^th International Oxford Brookes Conference for some very useful comments, and also to an audience from Kyoto University for the same. I am also very grateful to three anonymous referees for this journal for their probing and helpful comments.