Naïve realism, sensory colors, and the argument from phenomenological constancies

ABSTRACT

The sensory colors that figure in visual perceptual experience are either properties of the object of consciousness (naïve realism, sense-data theory), or properties of the subject of consciousness (adverbialism) (Section 1). I consider an argument suggested by the work of A. D. Smith that the existence of certain kinds of perceptual constancies shows that adverbialism is correct, for only adverbialism can account for such constancies (Section 3). I respond on behalf of the naïve realist that naïve realism is compatible with the existence of such constancies, so long as naïve realism adopts the view that sensory colors are relational properties of physical objects, not intrinsic properties (Section 4). In other words, the naïve realist should adopt the theory of appearing (Section 5).

KEYWORDS:

• Adverbialism
• colors
• intentionalism
• naïve realism
• perceptual constancies

Acknowledgements

Thanks to two anonymous referees for this journal for helpful comments.

Disclosure statement

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

Notes

1 I take the term ‘sensory character’ from Smith (2002). According to Smith, sensory character is that ‘whereby perceptual consciousness differs from mere thought or imagination’ (2002, 43); he also notes that ‘a veridical and a matching illusory experience have a shared sensory character’ (40, emphasis in original).

2 Color primitivists include Campbell (1993) and Westphal (2005). For a critical discussion of color primitivism, see Byrne and Hilbert (2007). I say that ‘many’ primitivist views take colors to be sensory colors because it is not clear to me that they all do. Gert (2017), for example, is a primitivitist about color, but although he takes colors to be primitive properties, it is not clear to me that he takes colors to be sensory properties, properties ‘with a distinctive sensuous nature’ (Chalmers 2010, 398). I will not address the details of Gert’s position here.

3 I take primitivist colors to be primitive and simple in the sense that their distinctive, sui generis, sensuous natures cannot be described. But the primitivist, in holding that colors are simple, is not denying that they can be compared with one another along more than one dimension. In particular, they can be compared with respect to hue, saturation, and lightness.

4 Perhaps a fourth view is that sensory colors are properties of the visual field (Peacocke 2008). Given my purposes in this paper, I will not address this view here.

5 Naïve realists who recognize the existence of sensory colors include Allen (2016) and Campbell (2002; 2005). Other naïve realists do not seem to recognize the existence of sensory colors; Fish (2009, especially chapter 6), for example, has a physicalist account of color, and does not otherwise mention sensory colors.

6 The locus classicus of adverbialism is Ducasse (1942).

7 Compare Smith (2002, 47), who claims that ‘any account of perception that cannot fully acknowledge the sensuous character of perceptual experience can be simply dismissed.’ Such accounts include ‘all forms of eliminativism [and] reductionism’ (47).

8 See also Butchvarov (1980, 272) and Crane (2006, 142–143), who notes that ‘there are familiar reasons why the adverbial theory is indefensible … and their source can be traced back to the theory’s failure to accommodate even the apparent relationality of perception.’

9 Thus having provided his account of perceptual intentionality, Smith concludes that the ‘phenomenological objection to the … “adverbial” account of sense-experience has been implicitly answered’ (2002, 308, n. 29). Kriegel (2007) explicitly combines (non-reductive) intentionalism with adverbialism; Hatfield (2016) does so a little less explicitly. Butchvarov (1980, 272) expresses the phenomenological objection to adverbialism by saying that the ‘adverbial theory is incapable of doing justice to the most obvious and indeed essential phenomenological fact about perceptual consciousness (perhaps all consciousness), namely, its intentionality, its object-directedness.’ But my point is that there is no reason the adverbialist cannot supplement his theory with the claim that perceptual consciousness has intentionality. Some might resist this supplementation of adverbialism with intentionalism on the grounds that adverbialism and intentionalism are incompatible views. But one can hold that adverbialism and intentionalism are incompatible views only if one holds that ‘adverbial theory … attempt[s] to explain all features of what it is like to have an experience in terms of intrinsic non-intentional qualities of experience’ (Crane 2006, 142, emphasis in original). Philosophers such as Smith and Kriegel, who do, as noted, combine adverbialism with intentionalism, would surely dispute this characterization of adverbialism. They might hold that adverbialism attempts to explain only the sensory features of the phenomenal character of experience in terms of non-intentional qualities, thus leaving open the possibility that intentional qualities are needed to explain non-sensory features of the phenomenal character of experience (such as its apparent relationality). Relatedly, I should note that combining adverbialism with intentionalism does not yield mere intentionalism; the adverbialism is still playing a role in the resulting view. Not all versions of intentionalism are adverbial; see Bourget’s distinction between adverbial (non-relational) versions of intentionalism and relational versions of intentionalism (Bourget 2020).

10 According to Allen (2016, 35–42), we are aware of two distinct kinds of sensory colors in visual experience: colors, which are intrinsic properties of the objects of experience, and apparent colors, which are relational properties of the objects of experience. But whereas I hold that sensory colors are mind-dependent relational properties of objects of experience (since the subject is one of the relata), Allen holds that apparent colors are mind-independent relational properties of objects of experience: they are ‘determined by the colour of the object and the nature of the illuminant and the colour of the background,’ but they are ‘independent of the psychological responses of the perceiver’ (36). I reject Allen’s account for two reasons. First, I think it gets the phenomenology wrong to say that we are aware of two distinct sensory color properties of an object in visual experience; we are aware only of one (see Gert 2010, 673). But put that issue aside; why do I think the sensory colors we are aware of are mind-dependent properties rather than mind-independent properties? Recall the case of size constancy. Suppose there are five people in a room looking at an object from different distances; also suppose that for each of the five people, the object occupies a differently-sized region of their visual field. According to Allen’s view, there will be five distinct instances of (sensory) apparent color in the room, because each of the apparent-color instances occupies a differently sized region of a visual field. I agree with Allen about that. But on Allen’s view, because these five instances of sensory apparent color (all associated with the same object) are mind-independent, they will continue to exist even after the subjects, say, close their eyes. I find that incredible; surely such a view is both ‘ontologically mysterious’ and ‘ontologically profligate’, despite Allen’s insistence to the contrary (Allen 2016, 37).

11 Similar accounts can be given to explain other phenomenological constancies, such as color constancy. In explaining color constancy, the first point to note is that sensory colors are not merely the way that an object appears to a subject (from a certain spatial perspective), they are the way that a certain property of the object appears to a subject, the object’s non-sensory color. Sensory colors are the way that an object’s non-sensory colors appear to a subject. Second, although I described sensory colors as a relation with at least three relata, it is a fourth relatum that is relevant for explaining color constancy: the ambient light. We can experience color constancies because a qualitative change in sensory color need not appear as a change in the object’s non-sensory color; rather, it can appear as a change in one of the other relata, and, in particular, it can appear as a change in the ambient light (Smith 2002, 176). For a defense of the view that both sensory and non-sensory colors are presented in experience, see Langsam (2000). I will remain neutral here as to exactly what properties non-sensory colors are, but for a defense of a view of non-sensory color with which I am sympathetic and which can be employed to explain color constancies, see Gert (2013).

12 I would submit that the fact that the object is the only relatum that is present in the experience in a sensory way explains the so-called transparency of experience, the fact that a subject introspecting her experience can attend only to the object of the experience (and not, say, to herself, the subject of the experience). For canonical discussions of transparency, see, for example, Harman (1990) and Tye (2000, chapter 3).

13 Insofar as my relational account is an account of sensory colors, and sensory colors are categorical properties, my account is very different from the relational accounts of color of such philosophers as Averill (1992), Cohen (2009, 2012), and Levin (2000), for their relational accounts are all versions of dispositionalism about color, and dispositional properties are very different from categorical properties. These three philosophers are not giving accounts of sensory colors; on the contrary, Averill (1992, 567–569) and Levin (2000, 153, 164, fn. 44) deny the existence of sensory colors, whereas Cohen’s view (2012, 292) is neutral about the existence of sensory colors.

14 For defenses of the theory of appearing, see Alston (1999) and Langsam (1997). For a different understanding of what the view is, see Pautz (2010).

Additional information

Notes on contributors

Harold Langsam

Harold Langsam is Professor of Philosophy at the University of Virginia. His main areas of research are the nature of conscious states (especially perceptual experiences) and their role in epistemic justification. He also occasionally writes about Nietzsche.