Desire, imagination, and the perceptual analogy

ABSTRACT

According to the guise of the good, a desire for P represents P as good in some respect. ‘Perceptualism’ further claims that desires involve an awareness of value analogous to perception. Perceptualism explains why desires justify actions and how desires can end the regress of practical justification. However, perception paradigmatically represents the actual environment, while desires paradigmatically represent prospective states. An experience E is an awareness of O when the nature of E depends on the nature of O. How could desires depend on the merely possible? Extant perceptualist accounts have not adequately addressed this question. I propose a novel account of how desires can be an awareness of value. An awareness of value involves the successful exercise of a capacity to discriminate value out of non-evaluative representations. The resulting content and phenomenology of such a desire depends in the right way on the value properties of the desired state. An agent requires the right view of the non-evaluative features of a state to discriminate its evaluative features. I argue that imaginings are uniquely able to provide such a view, and so enable value discriminations. My account retains the epistemological attractions of perceptualism despite the disanalogy between desire and perception.

KEYWORDS:

• Desire
• the guise of the good
• imagination
• perception
• awareness
• moral perception

Acknowledgements

I wish to thank Markos Valaris, Karyn Lai, Waldemar Brys, D’Arcy Blaxell, and Stephen Hetherington for their valuable feedback on earlier drafts. I also wish to thank the anonymous referees from Philosophical Explorations for their careful, attentive, and generous feedback which help me clarify a number of important points in the article.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 None of the previously cited advocates of the guise of the good have discussed the significance of this disanalogy. For instance, the book-length Oddie (2005) does not discuss the problem presented by the prospective character of desires. The famous Stampe (1987) does not directly address the disanalogy, although it presents an account of the indirect causal networking of desires and value which could partly address the issue. I am not aware of any explicit, detailed treatment of the problems generated by this disanalogy from an advocate of the guise of the good. McBrayer (2010) discusses the problem posed to moral perception by the observation that direct causal connections with value may not be possible, depending on one’s background meta-ethical assumptions. Milona and Naar (2020) discuss a similar problem in the context of the emotions.

2 The proposed account is inspired by capacity-based accounts of perception, especially, by Schellenberg (2017, 2018), McDowell (2011, 2013), and Millar (2019). The account applies much of Schellenberg’s account of perceptual discrimination to the context of desire, see Schellenberg (2017, 2018).

3 I follow Johnston (2001) in arguing that desires present evaluative truthmakers.

4 It can also include lower-order evaluative representations. I make the simplifying assumption that its content is non-evaluative.

5 I wish to clarify the distinction between my terminology of ‘representational bases’ and the popular term ‘cognitive bases.’ Deonna and Teroni (2012) introduced the notion of ‘cognitive bases’ to the literature on emotion. The terminology has seen widespread adoption and is used in more expansive or restrictive ways. For example, Milona and Naar (2020, 3072) define cognitive bases as ‘the mental states to which an emotion responds.’ However, Deonna and Teroni’s original usage was more restrictive, including primarily beliefs and perception. Deonna and Teroni (2012, 60) also coined the term ‘motivational base’ to include the affective and dispositional grounds for an emotion. To further complicate matters, the term ‘cognitive’ is naturally construed in doxastic terms. To avoid ambiguity, then, I will use the all-encompassing terminology of ‘representational base,’ which is defined simply as the representational grounds for a desire. My usage of ‘representational base’ includes affective states and other representational states that are not conceptually laden, such as perception and mental imagery. Thanks to an anonymous referee who urged greater clarity on these points.

6 Mark Johnston (2001, 211–212) argues that desires ‘discern’ value and this process depends on imaginings. However, Johnston (2001, 212) claims that imaginings can be a source of evaluative knowledge, while I will claim that imaginings enable the evaluative knowledge. Johnston (2001) does not discuss the process of ‘discernment’ in great detail nor the problem posed by non-actual states discussed here. Nevertheless, Johnston’s view is an important precedent for the account developed here. Jessica Moss (2012) has argued that Aristotle understood desires as based on appearance-states called phantasia. Phantasia is similar in many ways to the modern conception of imaginings. As such, Aristotle is an important historical precedent for the view that desires depend on imaginings. It is not entirely obvious whether Aristotle’s notion of phantasia perfectly maps onto the modern conception of imaginings. Furthermore, Moss (2012) argues that phantasia presents value via pleasurable or painful representations. Here, I work under the assumption that imaginings do not themselves have any valence (are not themselves pleasurable or painful) but rather precipitate other states that possess may contingently possess valence. I also do not assume that there is an essential connection between value representation and valence, which is another key part of Moss’s account of Aristotle.

7 I exclusively focus on phenomenal desires, that is, desires with an experiential, affective character. ‘Desires’ that lack an affective character are taken to be as dispositions to have a felt desire, motivational dispositions, or normative beliefs.

8 The guise of the good is naturally allied with value realism. However, value realism is not mandatory. It is possible to be an error theorist, subjectivist, quasi-realist, and so on, while endorsing the guise of the good. I assume value realism as part of the standard view under discussion. I thank an anonymous referee for urging greater clarity on this point.

9 Thanks to an anonymous referee for urging me to rethink the presentation of this point.

10 This distinction between causal and epistemic dependence follows Pryor (2013, especially 218–219 endnote 5).

11 Döring (2009) argues that emotional recalcitrance supports the claim that emotions provide non-inferential justification.

12 My discussion in this section is partly inspired by Elijah Chudnoff’s discussion of intellectual intuition as a form of awareness (2013a, 2013c, 208, esp. 205–225). This type of problem in the context of emotions is discussed in Milona and Naar (2020, 3090–3094). See McBrayer (2010) and Milona (2018, 207–208 ff.) for a similar problem about moral perception.

13 The first option is endorsed by almost every contemporary account of perception, except for some naïve realist or disjunctivist views. For the second option, see Brewer (2017), Campbell (2002), Fish (2009), Johnston (2011), and Martin (2004). Naïve realists and disjunctivists can accept a causal constraint on perceptual awareness. They deny, however, that the satisfaction of a causal constraint is what perceptual awareness consists in.

14 Thanks to an anonymous referee for prompting me to tease these two problems apart.

15 See Dretske (1999) for a taxonomy of forms of awareness.

16 As will become clear, this account is in essence an application of Susanna Schellenberg’s account of perception, see Schellenberg (2017, 2018). It is also influenced by Chundoff’s theory of intuitive awareness, see Chudnoff (2013a, 2013c).

17 Schafer (2013) and Schroeder (2008) have argued that the task of explaining the origin of evaluative content of desires has been neglected by perceptualists. The present account should, to some extent, assuage the worry that perceptualists cannot explain the origin of evaluative content.

18 For discussion of perceptual discrimination, see Siegel (2006), Schellenberg (2017, 2018). Siegel (2006, 434–437) proposes a phenomenology-based discrimination condition, and Schellenberg (2017, 2018) proposes a capacity-based discrimination condition. Chudnoff (2013a, 2013c, 173–203) also discusses perceptual discrimination.

19 As mentioned above, the following account draws on Schellenberg (2017, 2018). Schellenberg’s discrimination-based account of perception has been generalised to experiential awareness as such, then applied to the case of desire. Chudnoff (2013a, 2013c) employs a similar strategy for the case of intuitions and draws on somewhat different accounts of perceptual experience.

20 Schellenberg (2017, 2018) motivates her capacity-based view of perception as a ‘middle ground’ between active attributional views of perception and passive relational views. Here, I am drawing out an analogous benefit from applying her framework for perception to the case of desire.

21 The idea for a phenomenological differentiation condition for perception comes from Siegel (2006, 434–437), see also Schellenberg (2017, 4–5, 2018, 13–30), Brogaard and Chudnoff (2016, 62), and Chudnoff (2013c, 174–180). For an analogous condition for intuitive awareness, see Chudnoff (2013a, 710–712, 716–720, 2013c, 217–223).

22 Although, for the idea that perception involves an exercise of mental agency, see Schellenberg (2017) and McDowell (2011) and Millar (2019). The active, agential character of the process that underwrites a desire does not undermine the perceptual analogy insofar as the desire itself has a quasi-perceptual phenomenology and performs a similar epistemic function.

23 On this point I follow the ‘infallibilist’ understanding of exercising a capacity of Millar (2008, 2019) rather than Schellenberg (2018), see Schellenberg (2018, 43–45). Millar (2008, 2019) talks of recognitional abilities but there is considerable overlap between such abilities and capacities for discrimination.

24 Following Chudnoff (2013a, 715–716, 2013c, 215–217) we might call this ‘formal’ rather than material constitution. Schellenberg (2018, 15–16) argues that perception is constituted by particular objects in the sense that perception is (at least partly) grounded in particulars in a way that does not imply material constitution. There are important differences between these accounts but, for my purposes, the central idea is that there is a non-material form of constitution. I will work with the thinner, and correspondingly more neutral concept of essential parthood.

25 Chudnoff (2013a, 715–716, 2013c, 215–217) argues that intuitions counterfactually depend on abstract objects because it is abstract objects form part of the essences of intuition experiences. The idea here is similar. An awareness of value essentially depends on the reality of the relevant value since if that value would not obtain were P brought about, then the agent’s capacity to discriminate value would have been unsuccessfully exercised.

26 For discussion of perspectival conditioning of value representations in desires, see Oddie (2005, 60–64, 219–227, 2009, 137–140, 2015, 72–74, 2016, 98–100, 2018, 245–249).

27 Thanks to an anonymous referee who suggested that the account might be better formulated as an ‘imaginationist’ account and urged clarification on this important issue.

28 For example, we may want further fundamental types of imaginative content. Perhaps there is a fundamental type of content that is intention-like or action-relevant content. I do not have the space to explore these further complications.

Additional information

Funding

The research presented in this article was supported by an Australian Government Research and Training Program (RTP) scholarship.

Notes on contributors

Kael McCormack

Kael McCormack in his research focuses on the nature and epistemic role of affective experience. His current research develops an epistemological account of desire.