3 Is the concept of phenomenal precision incoherent?
According to Fink, if there is no solely generic phenomenology (i.e., generic without specific phenomenology) then the concept of phenomenal precision is threatened by incoherence. What is Fink’s argument for this conclusion? Suppose there is no solely generic phenomenology. Then, according to Fink, “…the p-precision of an experience is either contradictory, generally minimal, or generally maximal, which trivializes the notion.” (p. 9) And why is that? Because, according to Fink, if you experience the color of his Figure 2 as cayenne66, then if you also experience it as red, then there will be no unique precision to the experience. For red has a much wider precision range (i.e., lower precision) than cayenne66. His solution is to allow for experiencing it as red without experiencing it as any specific shade: generic without specific phenomenology.[6]
Let us approach this issue by asking what the representationist should say by way of response to Fink’s concern that there will be no unique visual precision. Then we can ask whether some version of that response is available to me.
Recall that representationists must acknowledge phenomenal precision (assuming they acknowledge representational precision) since on their view, if the representational precision of one conscious perceptual representation is greater than the representational precision of another conscious perceptual representation, then the phenomenal precisions must follow suit. Phenomenal precision—on their view—is just the shadow of representational precision. But when we see a cayenne66 object as cayenne66, do we thereby also see it as red? It is often supposed that this is some sort of necessity (Confession: I once thought that). To his credit, Fink points out that this is false. He says (footnote 17):
Conceptual or nomological relations do not necessarily transfer to the realm of experiences. Imagine seeing an animal as a mouse. One does not thereby see it as an owner of a heart, or as a member of the phylum chordata even though all mice belong to each category necessarily. [7]
Certainly Fink is right that seeing something as a mouse does not require seeing it as a chordate. However, he thinks any experience of cayenne66 is “likely” to be an experience of red. He doesn’t say how he knows this.
Here is a tempting but wrong view that I believe may stand behind what Fink says (and is also exemplified I believe in Begby 2011 and in a more complex form in Siegel 2010). Look at the cayenne66 patch in Fink’s Figure 2. I know what a red thing looks like and I can tell from looking that it is red because…well…it looks red. So I visually represent it as red. Similarly, it looks colored. And a baseball bat looks like a baseball bat, so I visually represent it as a baseball bat.
However, I also know what a 1969 Chevrolet Camaro looks like, as well as what a 1961 Jaguar E-type looks like. Do I thereby visually represent the property of being a 1969 Camaro or a 1961 E-type? I know what my wife looks like. Do I thereby have a singular visual representation that represents her? Perhaps what I am really visually representing in each of these cases is just constellations of low level properties that are recognitionally equivalent to the property of being a 1961 Jaguar E-type or to the singular property of being my wife.
I have argued that the extent of seeing-as in the sense of visual representation is not a matter for the armchair (Block 2014c). From the armchair one does not know whether something’s looking like a 1961 Jaguar E-type is a matter of representation of constellations of colors, shapes, textures, illumination, motion and other low level properties as opposed to an actual representation of the property of being a 1961 Jaguar E-type.
For example, I give evidence that we can visually represent facial expressions (high level property) and in addition constellations of colors, shapes, textures, etc. (low-level properties). The evidence is that there are distinct “adaptation” effects for both the low and high-level properties. (Adaptation is the neural “fatigue” effect underlying afterimages.) For example, if you vary the low level properties but keep the face identity (or expression or just faceness) constant, you get smaller adaptation effects, showing an extent of low level perception. And the fact that there is a residual face adaptation effect is one of many items of evidence favoring face-specific perception.
You can experience such an adaptation effect for yourself. Stare at the picture on the right for 1 minute, covering the two pictures on the left with something. Then very briefly look at the center picture asking yourself whether it looks more fearful or more angry. Now cover the two pictures on the right and stare at the picture on the left for one minute. Now look at the center picture very briefly again. It will appear to have a different expression. The center picture is a morph of a fearful face and an angry face. When you adapt to the fearful expression you are more likely to see the morph as angry-looking and conversely for adapting to the angry expression. This doesn’t prove that there is an adaptation effect for facial expression over and above adaptation effects for constellations of low level properties. The best one can do is form hypotheses about what those low level properties might be and vary those properties keeping expressions constant.
In addition, one can look for other signs of visual representation of faces or facial expressions. For example, faces show “visual popout”. Since typically “conjunctive” properties do not show visual popout, that fact suggests that visual representations of faces are not “conjunctive” properties and hence not conjunctions of low level features. The upshot of this and other work I cannot describe here (Block 2014c) is that it is very likely that there are representations of face-attributes such as facial expressions in addition to representations of low level properties.[8]
Figure 1: From Butleret al. (2008) with permission of Elsevier
The upshot of all this is that a single visual experience can represent both low level properties and high level properties. So: there can be distinct precisions for the different representations. For example, the precision of the experiential representation of fearfulness could be ascertained by investigating how much variation in the percentage of fearfulness in a morph like the middle one in the figure above is compatible with exactly the same visual representation of fearfulness. And similar methods could be used to ascertain precisions for the low level properties that are represented. There is no reason to expect these precisions to be the same.
An experience that represents cayenne66 could also represent red and there could be distinct precisions for each of these representations. And what goes for representational precision also works for phenomenal precision. If more than one property is genuinely present in phenomenology then there can be distinct precisions for the distinct properties. So the solution for the representationist works even if representationism is false.
So why is there supposed to be a problem concerning unique precisions? Fink argues as follows
You might think that this color experience has two p-precision values: The first value is for being experienced as red, and the second for being experienced as cayenne66. But this seems contradictory: why should one and the same experience of a color have two p-precision values, but only one for r-precision? And for that matter, why not three values for p-precision? You likely experience the color not only as cayenne66 and as red, but also as a color? Why not four, then, if you experience it as a visual experience? Or five, if you experience it as something? Or even six, if you experience it as phenomenal? [NB: p-precision is phenomenal precision; r-precision is representational precision]
The argument is not spelled out but one can guess that it depends on the idea that there is incoherence because there is no end to the number of properties that are present in experience. (Fink seems to suppose that there are not multiple representational precisions but does not say why.) We don’t need to see exactly what the argument is supposed to be to see that this premise is wrong. There is absolutely no evidence that experiences of colors present (or represent) colors as colors or as something or as phenomenal. These presentations and representations cannot be simply postulated. The reason that I went through the example of fearfulness was to give the reader a sense of how much work has to be done to show representation of a high level property. The problem in Fink’s argument is the assumption that you “likely” experience his Figure 2 not only as cayenne66 but as a color and the insinuation—not explicitly stated—that you experience it as something and as phenomenal. There is simply no reason to believe this.
On my view, color experience—like all perceptual experience—is non-conceptual. But the point is even stronger if color experience is conceptual since then the concept of color and the concept of something would be required to see the cayenne66 patch as colored and as something. Ask yourself whether an animal that can visually represent the color patch in Figure 2 as cayenne66 must also represent it as red or as colored. Must the animal be able to attend to or notice the redness or the coloredness as well as the specific shade? Or consider 4 month old human babies whose color perception is known to be good but who do not appear to notice colors to the extent of being able to use color information to judge whether there is one or two items. Even two year old children are so bad at conceptualizing color that a term was coined in the early 20th Century, “farbendummeit” (color stupidity), to describe their cluelessness. Darwin thought his own children were color-blind because they were so poor at learning color names (Bornstein 1985; Campbell 2014).
To conclude this section: uniqueness of precision is not required for coherence. The representationist can reasonably hold that to the extent that there is more than one representational content, there is more than one precision: precision of representation depends on what representation is in question. And the same can be said of what properties are presented in perception as opposed to represented in perception, even if as I argue, representationism is false.