Abstract
It is standard to suppose that, whether or not they are actually instantiated in our environment, colors are properties. Presumably those who are convinced of this thesis are convinced because they think that’s how we see colors--how visual experience represents them. I argue, in contrast, that there are cases of illusory color perception in which it is more plausible to suppose colors are represented as kinds of stuff or substance rather than as properties. I then show how to extend this result to support the conclusion that colors are always represented in vision as kinds of stuff rather than as properties. In a concluding section, I consider further ways to explore and test the “stuff theory” of visual color representation. I also extract a moral from this investigation about our ability to draw accurate conclusions about our conscious visual experience.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
A useful survey of these issues appears in Maund (2012).
The proposal I will defend here has antecedents. Levinson (1978) criticizes the thesis that colors are properties and defends an account of them as what he calls qualities, which he takes to be kinds of stuff or at least stuff-like. Levinson’s thesis is an ontological one about what colors actually are, but presumably he would claim that if colors are a kind of stuff then they are also represented in that way in visual experience. Additionally, Byrne (2003), while he endorses a property theory of visual color, indicates that our visual experience of colors, as well as our thinking about them, has stuff-like features.
My proposal, if correct, does have some effect on the search for what, in our environments, colors might be, albeit a defeasible one. If colors are kinds of stuff, then a principle of interpretive charity suggests that, in looking for what colors are, one should, other things being equal, give preference to proposals that they can be identified with certain independently specifiable environmental substances. Other things may not be equal, and, indeed, there may be no such candidate substances.
I have based the category of concrete substance on the category of concrete mass noun used in Steen (2016). Steen describes “data” and “information” as abstract mass nouns and “pain” and “admiration” as psychological mass nouns.
Boring (1946) cites a spirited disagreement at a meeting of the Society of Experimental Psychologists earlier in the twentieth century between E.B. Holt and Edward Titchener about whether green could be determined by visual inspection to contain both blue and yellow. (Holt thought it could. Titchener disagreed.) It is tempting to think that the dispute was at least abetted by confusion between the mixture of two kinds of substance, color pigments and the color substance argued for here.
For the purposes of this discussion, I am assuming that luminous colors are properties of light and that light is a physical substance.
One indication of just how entrenched the view is that colors are properties can be found in the Stanford Encyclopedia entry on properties. In this entry, colors are mentioned fourteen times as examples of properties (Orilia and Swoyer 2016). By the way, that’s fourteen mentions of the category colors. Individual colors are mentioned as examples on additional occasions.
A similar characterization of “stuff” or “material” has been used in computer vision research as part of a strategy for using the relationship between objects, such as cars, and regions of stuff, such as pavement, to identify those objects in high-resolution satellite images and elsewhere (Heitz and Koller 2008).
Of course if the king, or the dog for that matter, is a (mythical) shape-shifter, then they’re an exception to what’s typical.
Perhaps we can attribute the ability to reason in parallel here about ontological matters and representational matters to a version of the principle of interpretational charity. The fact is, however, that both inferences seem plausible on their own merits.
Critchley describes a wide variety of illusory colors phenomena. For present purposes, I am focusing on those cases where the colors have been visually dislocated from actual environmental objects rather than invented whole cloth by the visual system.
There are other experimentally induced phenomena in which, apparently as a result of binding errors, the wrong color gets attached to the wrong existing object. Wu et al. (2004), for instance, describe a kind of case in which moving objects of different colors appear to trade colors with each other. Because, in these cases, colors appear as moored to objects, they seem less germane to the question for whether or not colors are represented as substances.
A fuller account of these researchers’ usage may be helpful:
-
(1)
Hong and Blake (2009), describing various dislocated color phenomena write that “color can spread beyond its figural boundaries into neighboring regions of the visual field…” (p. 403). This description characterizes the color itself as what spreads, rather than an illusory surface or substance that exhibits the color, and, in so doing, attributes to the color the property of spreading, a property ordinarily reserved for substances. These authors also say, of the dislocated colors produced by binocular rivalry, that they are” experienced as a diffuse, somewhat faint cloud” (p. 403) and refer to this phenomenon as involving a “formless cloud of color” (p. 406). These locutions seem to imply that the kind of substance the cloud is composed of is the color itself rather than that the cloud is composed of a kind of substance that possesses the color as a property.
-
(2)
Bressan et al. (1997), throughout their review of research on neon color spreading, attribute to colors the characteristics of flowing (p. 2), undergoing dilution (p. 3), bleeding (p. 7), being absorbed (p. 8), undergoing diffusion (p. 8), and, of course, spreading. These are verbs that describe substances. (Page numbers cited here are from the version of Bressnan et al. retrieved from http://cns.bu.edu/Profiles/Mingolla.html/cnsftp/pdf/Bressan-et-al-97neon.pdf.)
-
(3)
Critchley’s (1965) descriptions less consistently suggest his construal of colors as stuff or substance, but, for instance, he refers to cases in which epileptic patients see “spectral colors flooding the field of vision” (p. 712).
There are a couple of differences between Critchley (1965) and the other researchers cited here that may account for the differences in their usage. First, Critchley reports cases of dislocated colors that were experienced by individuals with brain damage or who had been subject to techniques that Critchley seems not to have undergone. Thus, he is apparently not reporting the results of his own visual inspection. By contrast, Hong and Blake (2009) and Bressan et al. (1997) are reporting not only the results others’ visual inspection of such cases, but their own as well. Second, Critchley’s article is not exclusively devoted to cases of dislocated colors and related phenomena but to other unrelated color anomalies that are produced by brain damage. Thus the more uniform focus on dislocated colors in Hong and Blake (2009) and Bressan et al. (1997) may also at least partly explain the differences in usage.
-
(1)
Because PR will be used here to defend the stuff theory of color, it is worth stressing that this argument, though it may seem to beg the question, does not. Notice that this argument doesn’t assume the accuracy of the impression that dislocated colors occur apart from any color bearer. For it may be that dislocated color experiences merely occur apart from the most obvious bearer of color and are mistakenly thought, thereby, to be occurring apart from any bearer whatsoever. And yet, such inaccurate assessments can still explain, in accord with PR, the impression that dislocated colors are substances. Notice too that this argument for PR doesn’t assume everyone has the impression that dislocated colors are substances. Indeed, a reviewer for Philosophia reports a contrary intuition. It would, of course, be good for a supporter of the stuff theory, like me, to have an independently plausible explanation for such contrary intuitions, and I do not. I can hazard the guess that the sheer entrenchment of the philosophical thesis that colors are properties is responsible for the contrary intuition. But I have no explanation for why I am not also subject to that intuition, for I was, until recently, also convinced that colors are visually represented as properties.
Sekuler and Palmer (1992) quote a passage from Kurt Koffka’s The Principles of Gestalt Psychology that serves to highlight the apparent paradox:
Counsel: “Where was the book?”
Witness: “On the table, sir.”
C: “And what was underneath the book?”
W: “The table, sir.”
C: “How do you know?”
W: “But I saw it, sir.”
C: “Are you willing to testify under oath that under the book there was no opening in the table through which a revolver might be dropped?”
W: “Certainly not, sir.”
C: “Why not?”
W: “Because I could not see it, the book being where it was.”
C: “And yet you say that you saw the table was under the book? Thank you.”
Somewhere James Gibson proposes a striking way to demonstrate the fact that environmental features beyond our visual field are perceptually represented. He asks us to imagine that we are standing on the edge of a cliff and facing away from it and to notice the result.
It would be reasonable to say, in the case described here that you have inferred the presence of exhaust fumes from the black swirl and other evidence. What I am suggesting is also true is that your inference results in a hybrid representation of that exhaust fumes in your visual field, one that is partly perceptual and partly not, and that we are happy enough, speaking colloquially, to call this seeing the exhaust fumes.
Bressan et al. (1997) report a variety of examples of illusory neon colors in motion. They all seem to be cases of luminous color, but it isn’t clear that achromatic, non-luminous cases couldn’t also be produced. Additionally, it would not be hard to produce a computer animation of non-luminous color blobs that change size and shape and location and that behave in the visual experience of a viewer like unmoored colors. Those blobs would actually be colors on the surface of a computer screen, of course, but they would also be readily dislocated from that surface, and, instead, seen as unmoored colors floating in a shallow space somewhere behind the screen.
I am not claiming that visual experience cannot represent non-instantiated properties. Indeed, anything can represent anything, with the right preparation, so there is no chance that visual experience can’t represent non-instantiated properties. All that I am claiming is that, in the first instance (so to speak), visual experience represents particulars (such instantiated properties) and that these particulars might then be appropriated to represent other things (such as non-instantiated properties).
I am indebted to an anonymous referee for Philosophia who suggested this account.
A referee is concerned that the argument of this section, for the conclusion that dislocated colors are unmoored, and hence are substances, is an argument that overgeneralizes. It would be an unfortunate consequence of the argument if, for instance, it committed one to the conclusion that the three-dimensional shapes of physical objects and substances were, themselves, substances and not properties. It would also be unfortunate if the argument committed one to the conclusion that surface textures such as being rough, or spiky, or cratered, or undulating were substances and not properties. However, in these cases, it seems to me, there is no danger of overgeneralization. A superficial reason for being assured no such overgeneralization occurs is that the Property Rule, a crucial ingredient in the argument of this section, is only defined for colors, not for other perceptible features. But one may think anyone who endorses the Property Rule must be ready to generalize that rule. I think such a generalization is reasonable. But I also think one should only generalize to other features that can genuinely appear in visual experience as dislocated from the objects that possess them. Significantly, while the shapes of objects can appear as unbound from other features of physical objects and physical substances (their colors for instance), the location of a shape seems to me to determine the location of the object or substance whose shape it is, and, in that respect, shapes aren’t subject to dislocation—not from their objects or substances. This makes shapes different from colors and thus not subject to a reasonable generalization of the Property Rule. Something similar, I suggest, holds for the kinds of textures mentioned above. Textures such as being rough, or spiky, or cratered, or undulating, are aspects of the three-dimensional structure of a surface, and, wherever those textural features are exhibited, their surfaces are found. In that respect, such structural features can’t be dislocated from their surfaces even if the surface can be dislocated from its object. However, at least some perceptible features of objects and substances, such as their shading, do seem to be capable of dislocation from the surfaces that exhibit them, for shading and shadows are arguably superimposed on, and hence separate substances from, the substances whose shading they are (cf. Hering 1964, on shadows). The issues mooted in this footnote, unfortunately, need much more discussion than I can offer here.
Where liquids, clouds, and crystals are concerned, we can count on detailed explanations of just how the constraints in question operate. Should we expect the same to be true where stuff theory is concerned? That depends. Whether we need to say how the constraints operating on color substances are enforced depends on whether colors actually exist and whether, if they do, they are actually substances. If they don’t exist, or if they aren’t substances, we need say no more than has already been said. The upshot is this: When it is used to accommodate the case of moored colors, a stuff theory is not saddled with a significant explanatory burden except in cases in which we can expect that burden to be met.
The limits on the expressive completeness of vision are, at least to some extent, recognized in Fodor’s (1983) claims that perceptual modules are domain specific and that they have shallow outputs.
The use of phosphenes to support a stuff theory of visual color representations is complicated by the fact that Block (1996) and others have construed phosphenes as non-intentional color qualia. I suspect that, at least in part, Block reaches that conclusion because he fails to consider the possibility that phosphenes are representations of (illusory) color substances located in space somewhere beyond the observer’s point of view.
The colors in question are, of course, moored to the surface of the paintings, but, at least often, they count as dislocated by virtue of the fact that they are seen as occupying a shallow, fictitious space behind the painting’s surface (cf. Note 19).
References
Block, N. (1996). Mental paint and mental latex. Philosophical Issues, 7, 19–49.
Boring, E. (1946). Mind and mechanism. The American Journal of Psychology, 59(2), 173–192.
Bressan, P., Mingolla, E., Spillmann, L., & Watanabe, T. (1997). Neon color spreading: A review. Perception, 26(11), 1353–1366 Retrieved from http://cns.bu.edu/Profiles/Mingolla.html/cnsftp/pdf/Bressan-et-al-97neon.pdf.
Byrne, A. (2003). Color and similarity. Philosophy and Phenomenological Research, 66(3), 641–665.
Byrne, A., & Hilbert, D. (2003). Color realism and color science. Behavioral and Brain Sciences, 26(1), 3–21.
Critchley, M. (1965). Acquired anomalies of colour perception of central origin. Brain, 88(4), 711–724.
Feldman, J. (2013). The neural binding problem(s). Cognitive Neurodynamics, 7(1), 1–11.
Fodor, J. (1983). The modularity of mind. Cambridge: MIT Press.
Heitz, G., & Koller, D. (2008). Learning spatial context: Using stuff to find things. In D. Forsyth, P. Torr, & A. Zisserman (Eds.), Computer vision – ECCV 2008: 10th European conference on computer vision, proceedings, part I. Berlin: Springer.
Hering, E. (1964). Outlines of a theory of the light sense, trans. L. Hurvich and D. Jameson, Cambridge: Harvard University Press. (Original work published 1920.)
Hong, S. W., & Blake, R. (2009). Interocular suppression differentially affects achromatic and chromatic mechanisms. Attention, Perception, & Psychophysics, 71(2), 403–411.
Jensen, M. S., Yao, R., Street, W. N., & Simons, D. J. (2011). Change blindness and inattentional blindness. Wiley Interdisciplinary Reviews: Cognitive Science, 2(5), 529–546.
Kanisza, G., & Gerbino, W. (1982). Amodal completion: Seeing or thinking? In J. Beck (Ed.), Organization and representation in perception. Hillsdale: Erlbaum.
Levinson, J. (1978). Properties and related entities. Philosophy and Phenomenological Research, 39(1), 1–22.
Maund, B. (2012). “Color”, The Stanford Encyclopedia of Philosophy (Winter 2012 Edition), Edward N. Zalta (ed.), URL =<http://plato.stanford.edu/archives/win2012/entries/color/>.
Orilia, F. & Swoyer, C. (2016). “Properties”, The Stanford Encyclopedia of Philosophy (Spring 2016 Edition), Edward N. Zalta (ed.), URL = <http://plato.stanford.edu/archives/spr2016/entries/properties/>.
Robertson, L. C. (2003). Binding, spatial attention and perceptual awareness. Nature Reviews Neuroscience, 4(2), 93–102.
Sekuler, A. B., & Palmer, S. E. (1992). Perception of partly occluded objects: A microgenetic analysis. Journal of Experimental Psychology: General, 121(1), 95.
Steen, M. (2016). The metaphysics of mass expressions. The Stanford Encyclopedia of Philosophy (spring 2016 Edition), Edward N. Zalta (ed.), URL = <http://plato.stanford.edu/archives/spr2016/entries/metaphysics-massexpress/>.
Sternheim, C. E., & Boynton, R. M. (1966). Uniqueness of perceived hues investigated with a continuous judgmental technique. Journal of Experimental Psychology, 72(5), 770–776.
Wu, D. A., Kanai, R., & Shimojo, S. (2004). Vision: Steady-state misbinding of colour and motion. Nature, 429(6989), 262–262.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Montgomery, R. Colors and Stuff: Exploring the Visual Representation of Color. Philosophia 45, 1283–1298 (2017). https://doi.org/10.1007/s11406-017-9853-9
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11406-017-9853-9