Qualia realists: explain radically deceptive experiences
A call to action!
(This piece is half personal experience, half Dennett’s Quining Qualia, and half Derek Shiller’s Hidden Qualia. That’s three halves, so you know this was a fun one to write!)
Radically deceptive experiences
An ordinary deceptive experience strongly inclines you to believe false things about objective reality.
Looking at the Müller-Lyer illusion, depicted above, induces an ordinary deceptive experience. The two lines are, in reality, the same length. But even if you get a ruler and check (as I have), they really look like they are different lengths. I am aware, in this case, that my experience is deceiving me; I believe that they are the same length. But that doesn’t change the visual experience itself, nor the intuitive pull that they must be different lengths.
Phantom limb pain is another kind of ordinary deceptive experience. Some amputees report feeling as though their missing limb is still there and hurting, even when they can see that the limb isn’t there. I doubt they believe or even suspect that the limb is really there, only invisible; nonetheless, the experience strongly pulls one towards that belief.
These are unproblematic cases for phenomenal realists. Nobody said qualia have to accurately portray objective reality. More problematic are what I call radically deceptive experiences.
A radically deceptive experience strongly inclines you to believe false things about the contents of your own experiences. So, while the ordinary Müller-Lyer illusion inclines you to believe that the top line is longer, when in reality it is not, a radical Müller-Lyer illusion inclines you to believe the top line looks longer to you, when in fact it does not look longer to you.
The very concept seems incoherent to a lot of people, my past self included, at first glance. How can it seem to you that it seems to you that p, if it does not seem to you that p? My definition seems to rest on a fundamental confusion about what experience is! But I promise you, I am going somewhere with this: radically deceptive experiences do exist, and they do spell trouble for phenomenal realism.
Imagine a patient steps into your office with rod monochromacy: their cones (the photoreceptors responsible for detecting color) are completely unresponsive, but their rods (which only detect brightness/darkness) are intact. If they are aware of their condition, they are having a kind of ordinary deceptive experience: red and blue swatches of equal luminance will appear to be the same color. The phenomenal realist’s picture is still clear: the colored qualia are gone, replaced with greyscale qualia, just as though you or I were looking through a black-and-white filter. Simple enough.
But suppose this patient, to your surprise, insists that they can see in full-spectrum color. Show them many common images, and they can accurately identify their color: the grass is green, the banana is yellow, the sky is blue. But show them a novel stimulus, like a swatch of fabric they’ve never seen before, and their accuracy plummets, no better than chance. They insist the blue coat is green, the green marble is yellow, and the yellow paint is turquoise. You can even show them the same stimulus twice, and they’ll give a different identification each time. But if you try to tell them this, they are flummoxed, insistent that those things really do look colored to them. You can find no signs that they are lying, or that their verbal processes have “gone rogue”, separate from what the rest of the brain “believes.”
What exactly is happening, according to the phenomenal realist? I can think of three possibilities:
The patient’s qualia are in fact greyscale; they are not experiencing color at all, but they think that they are. This is a radically deceptive experience: they think they are having an experience that they are not having.
The patient’s qualia are colored, but they are more or less random outside of familiar cases and can shift whenever an object leaves their field of view. This is an ordinary deceptive experience: they see colors that lead them to believe that objects have pigments that they don’t really have.
The patient experiences some other qualia when they look at objects which they think is what other people mean by “color;” maybe objects commonly labeled as “green” by others have a certain “texture” to the patient, and they think it is this property that “green” refers to. But this property does not in fact align with ordinary color in many cases, even on re-presentation of the same object, and they are confused by the incongruity.
The first trouble for the phenomenal realist is that there is no way for the patient to tell by introspection which of these is the case. While inverted-spectrum fans might admit ambiguity between #2 and #3, we are supposed to know what our qualia are like by direct acquaintance. There is a strong temptation, when I am in a realist mood, to think that if only I were in this patient’s position, I would see which was the right case; whether the “movie” was in color or not, and that I would know this deep down, even if I couldn’t verbalize it. But how could I? If qualia are something over and above dispositional, functional states, the type of states that incline me to believe I am having a certain experience, then what’s stopping my beliefs from being radically out of line with my qualia?
Still, for a long time I couldn’t escape the feeling that there was some dirty trick being played in case #1; something there was incoherent or confused.
Then, I tried a card trick!
I lost my faith in introspection
Reading Daniel Dennett’s Intuition Pumps and Other Tools for Thinking, I came across the following demonstration:
Here’s how to do it, according to Daniel Dennett:
Sit in front of a mirror so you can monitor your own compliance with the directions, which are to stare intently into your own eyes, fixating on them as a target instead of letting your eyes get attracted to peripheral goings-on. Now, without looking, take a card from the middle of a well-shuffled deck of cards and hold it, face-side toward you, at arm’s length just outside the boundaries of your peripheral vision. Wiggle the card. You will know you are doing it, but you won’t see it, of course. Start moving the card into your field of view, wiggling it as you do so. First you can see motion (the wiggling) but no color! You can’t tell whether it’s a red card or a black card … When you are finally able to identify the card, it’s almost directly in front of you.
My experience, the first time, was rather different from how Dennett described it, and it was initially very disappointing. The card immediately and obviously looked black to me… until I glanced down and saw it was in fact red. Surprising! But so far, a very ordinary deceptive experience: the card’s color quale is red, when the card’s pigment is in fact black. I repeated the experiment: the next card looked red, but was in fact black. Still ordinary deceptive, still theoretically unproblematic.
But after doing this a couple more times, I began to have a peculiar experience: whenever I focused on the cards, they began to look neither red nor black to me, but something indeterminate or unclear. But whenever I wasn’t trying to discriminate some new stimulus, my peripheral vision really looked like it was in full color. That is, I didn’t perceive my visual experience as getting less and less colorful towards the edges of my field of view. But if you asked me to actually name any of those colors in a new environment, I would come up empty, and be very surprised that I was coming up empty.
What happened to me? I see three possibilities, similar to but distinct from the possibilities we considered in the last case.
Jack does not experience fully detailed color qualia in his peripheral vision, even though when he introspects, it really seems to him like he does.
Jack does experience fully detailed color qualia in his peripheral vision, but is unable to report on them accurately; he is unaware of what he is experiencing.
Jack does experience fully detailed color qualia in his peripheral vision when he is not concentrating, but those qualia vanish or change whenever he attempts focused discrimination, shifting to an indeterminate character.
As before, I really can’t discriminate between these. #1 and #2 are both radically deceptive experiences: I do not, in fact, have accurate access to my own qualia. #3 is not much better. It suggests that I am experiencing red, but as soon as I try to name whether it’s red or black, the experience of red (and my memory of it, I guess?) disappears and is replaced by something indeterminate. But what grounds could I really have to believe that this is the case? Does it really explain my experience any better than #1 or #2? It doesn’t rescue reliable introspection either way.
Another example, from Pete Mandik on a podcast episode with Keith Frankish:
Another experience I had, also on LSD, … there was a film on the TV, and the film was in German, and there were no subtitles. And I had this incredibly intense experience that I understood what the people were saying. By the way, I don’t speak German. … No one, including myself, thought to ask this question. So what are they saying? If you can understand them, go ahead, tell us in English what they just said. (emphasis mine)
If Pete thought they were saying “I am German, hahaha, look at me with my pretzels and lederhosen,” this would be an ordinary deceptive experience. He would be wrong that his experience is the experience of understanding the film, because his experience is out of sync with reality. But, to the phenomenal realist, he is right about his qualia.
But Pete never even thought about what they were actually saying. He believed he was having the experience of understanding language. But how can you have the experience of understanding language if you have no idea what the language actually means? That contradicts what “understanding” means! Instead, it sounds like Pete was having a radically deceptive experience. To the phenomenal realist, Pete was wrong about the contents of his own qualia, not just how his qualia relate to the outside world.
And, then, of course, there are the famous examples from “Quining Qualia”:
“The patient failed in all tasks in which he was required to match the seen color with its spoken name. Thus, the patient failed to give the names of colors and failed to choose a color in response to its name. By contrast, he succeeded on all tasks where the matching was either purely verbal or purely nonverbal. Thus, he could give verbally the names of colors corresponding to named objects and vice versa. He could match seen colors to each other and to pictures of objects and could sort colors without error.” (Geschwind and Fusillo (1966) This patient was quite unaware of any deficit. He "never replied with a simple 'I don't know'" to the demand for naming a color." (Geschwind and Fusillo, 1966, p. 140)
Phenomenal realists often assert that we will never be truly able to know what another person’s qualia are like; inverted-spectrum cases are always theoretically possible. But phenomenal realists should be able to assign at least one coherent picture of what a human is experiencing. So if you’re skeptical about radically deceptive experiences, try it: describe one internally coherent account of this patient’s qualia which does not require that they have a radically deceptive experience.
What’s the big deal?
The eliminativist often says: how do you know you have private, intrinsic, ineffable qualia? How do you know those are phenomenal states and not just functional states? And some realists say: I know what my experience is like by direct acquaintance. There is no mediation by which I can get it wrong; whatever it seems like to me, that is my experience.1
But since encountering radically deceptive experiences, I don’t think that’s true. I think you can be wrong about your own experience, even the supposedly “intrinsic” elements of it. Color is paradigmatically intrinsic for most definitions of qualia, and I can think I am experiencing a full field of color when in fact I am not.
To me, this undermines a great deal of the reason for believing that one has qualia. The illusionist perspective is just that all experience that you think is phenomenal is radically deceptive in that way. Now we know that radically deceptive experiences do in fact exist. So what makes this a bad explanation?
This is not a rhetorical question! I am inviting all phenomenal realists to try and provide an account. Importantly, if you have not thought of an answer so far, try to think of one before consulting the literature. I know that several phenomenal realists in the literature do offer explanations, which I am reading now! But if you are unaware of them then you must admit that these answers come posterior to your reasons for believing in qualia. So if you are unable to think of an answer on your own, I would suggest that the reasons you had for believing in qualia in the past were not that strong.
Personally, I am pretty uncertain. I lean towards illusionism & qualia quietism,2 but some of the realist explanations I am starting to survey look promising. But that’s a significant shift from my position that realism is near-certain, and I hope more of you will join me in this uncertain space until we work it out!
I don’t believe all realists think this. Nevertheless, I bet some do, because I used to be a committed qualia realist, and this was the answer I gave.
I’m about 50/50 on whether illusionism or QQ is the better explanation.
All your examples seem like errors of judgements about our experiences, not experiences that are misrepresented to the subject. They’re all mistaken beliefs about the experience or the limits of introspection as a way of giving reports of our experiences, not mistakes about the existence of the experience.
The monochrome example, the subject has some kind of quale, shades of black and white with various intensities. Those shades are associated with learned colour associations, and so their mistake is a conceptual error, not an error about their qualia.
The peripheral cards is a mistake about the detail in your peripheral vision, not that you were mistaken about the quale, but your interpretation of the quale. You assumed your peripheral vision had the same level of detail as normal vision, but in fact it’s low resolution. Notice you know it must be red or black, so those details are filled in. But when you actually try and determine the colour you can’t. Not because you’re not having some quale or other, but because your peripheral vision doesn’t have the detail you’re expecting it to have. And you’re surprised because your theory about the resolution of peripheral vision is wrong.
And you discovered peripheral vision is low resolution by introspection because you end up saying the quale was “something indeterminate or unclear”. That sounds like introspection actually led you to an accurate description of peripheral vision, not the mistaken belief that everything in our visual field is the same level of detail.
Reports of experience are always interpretative. Qualia realists don’t require infallible introspective precision, they only require there to be some kind of qualia available to introspection.
And just to up the ante, your entire argument relies on realism. The lines seemed a certain length the card colour was a quale you describe as indeterminate colour, the monochrome is experiencing some kind of quale they then report incorrectly. All of which requires qualia to exist to be misreported.
We are not very much not qualia realists, but the set-up here seems to slip into a false assumption, namely, that a qualia realist is committed to saying that directly knowing that you have a phenomenal experience is directly knowing the correct concept to associate with it. All qualia realists will deny this (it is one of the things that distinguished qualia theory from versions of the older sense-data theory for which qualia theory was proposed as an alternative). If someone comes to you claiming that things look colored to them, you check this by simply testing whether they can make color judgments correctly (and ruling out that they are not doing so due to faulty testing set-up, etc.). This is because identifying the qualia as 'looking colored' is a *conclusion* that they are drawing, and it is tested like any other conclusion. They are drawing the conclusion (the qualia realist will generally say) on the basis of how the perception is located in their experience (the conditions associated with it, the qualia-patterns of which it is part, what other people say about the situation), which leads them to *associate* the immediate experience with concepts like 'color'.
In other words: How do you know that you have qualia *of some kind*? The qualia realist will say, you have immediate experience; the qualia are just those things in experience that are immediate. (Depending on the qualia realist, you may be told that you can know this by introspecting that you are having an immediate experience, although not all qualia realists put the same importance on introspection.) How do you know that qualia are *of this or that particular kind*? This requires a very different answer. The standard qualia realist answer has usually been that, since they are non-conceptual, you can only indirectly classify them by locating them in experience; this is why one usually uses circumlocutions ('what it's like to see something as colored') to talk about them rather than just labeling them directly. Recognizing that these qualia are 'what it's like to see red' just means that we reasonably and consistently group them with a bunch of experiences that we associate with the concept and label 'red'. Qualia realists don't generally think you directly see that this quale is what it's like to see red (if any do, they are heretics); all of your qualia could completely change, and this has nothing at all to do with how you classify it *unless the change also changed how your experiences were groupable*.
Thus on at least a standard qualia realist view, we call deceptive experiences 'deceptive' because of conceptual problems, despite the fact that they are obviously real experiences, which is why we call them 'experiences'. The existence of qualia for a qualia realist is what makes them real experiences; what makes them deceptive is that we are misfiring, in some way, in how we are associating them with concepts. Deceptiveness just *is* a conceptual matter, not a matter of immediate experience. The qualia realist would take your card experience as an example of why the qualia and the concepts have to be kept distinct.