DOUBLE VISION, PHOSPHENES AND AFTERIMAGES: NON-ENDORSED REPRESENTATIONS RATHER THAN NON-REPRESENTATIONAL QUALIA Download PDF
Işık Sarıhan (Central European University)
© 2020 Işık Sarıhan
DOI: 10.31820/ejap.16.1.1
Original scientific article – Received: 13/08/2019 Accepted: 24/1/2020
ABSTRACT
Pure representationalism or intentionalism for phenomenal experience is the theory that all introspectible qualitative aspects of a conscious experience can be analyzed as qualities that the experience non-conceptually represents the world to have. Some philosophers have argued that experiences such as afterimages, phosphenes and double vision are counterexamples to the representationalist theory, claiming that they are non-representational states or have non-representational aspects, and they are better explained in a qualia-theoretical framework. I argue that these states are fully representational states of a certain kind, which I call “automatically non-endorsed representations”, experiential states the veridicality of which we are almost never committed to, and which do not trigger explicit belief or disbelief in the mind of the subject. By investigating descriptive accounts of afterimages by two qualia theorists, I speculate that the mistaken claims of some anti-representationalists might be rooted in confusing two senses of the term “seeming”.
Keywords: Perception; representationalism; qualia; non-conceptual content; afterimages; double vision
1. Introduction
This paper deals with the question of the nature of experiential phenomena such as afterimages, double vision and phosphenes. When you close your eyes and apply pressure on them, you can experience a swirling array of colored dots or patches. Is this a representational mental state? Is it possible to say that there is a way that the world visually appears to you while having this experience, that is, does the experience consist in a visual appearance that can be judged to be truthful or misleading about the visible environment? Does the experience, as it were, tells you anything true or false about reality, does it have a truth-evaluable content? When you introspect, do you become aware of this experience by being aware of how the world qualitatively appears to be, a way which it may or may not be, or do you become aware of it by becoming aware of something else, say, some actual qualities of the mental state itself that cannot be spelled out as the qualities that the world appears to have? Or consider the visual phenomenon of double vision: Is the difference between “single” and double vision a difference about the representational content of these experiences, that is, a difference about how things look? If not, what is the difference?
One might think that the answer is straightforward: The difference between double and ordinary vision is a difference between how things visually appear to be. In double vision, objects visually look doubled in an unusual way, even though we do not take them to be doubled in the great majority of cases. In the case where I rub my eyes and undergo a phosphene experience, my vision suggests me a two-dimensional ghostly world of chaotic colors that exist slightly behind my eyes, even though I do not believe that there are such colors somewhere in there or out there. Some reasonable philosophers disagree with this account. They claim that such experiences, or some aspects of these experiences, cannot be spelled out in terms of appearance or representation.[1]
Representationalism or intentionalism for phenomenal experience, in its stronger variety, is the theory that all introspectible qualitative aspects of a conscious experience are qualities that the experience non-conceptually represents the world to have.[2] Some philosophers have argued, on the basis of introspective examination, that there are certain experiential states which are not fully representational, with afterimage experiences being the paradigmatic example of such states (Boghossian and Velleman 1989; Block 1996; Kind 2008). According to the anti-representationalist line of analysis, such states, or some aspects of these states, do not represent the world to be in a certain way. They do not have truth-evaluable content or intentional objects. It cannot be said that they are “about something” or that they are true or false. And as we can be aware of these mental states by introspection, the properties in virtue of which we are aware of them should be non-representational or non-intentional properties, and then the point about these special experiences are generalized to all experiences by some argumentative move or other. Therefore, the argument goes, representationalism about experience is false: There is more to experience than representation, and this can be revealed by an introspective analysis of the phenomenology of certain experiences. Introspection allegedly reveals “qualia” besides the representational facts about these experiences.[3] I will assume that the reader has some acquaintance with the concept of a quale and its place in contemporary philosophy of mind. I will just leave it at mentioning that here we are dealing with qualia in the restricted sense of the term that applies to mental qualities[4] that are concretely instantiated while undergoing an experience, and not with the term “qualia” that is sometimes used as a placeholder for whatever accounts for the qualitative aspect of experience.[5]
In this paper, I argue that the experiential states that are brought forward as examples in such arguments for qualia are not really non-representational states, but they are a sub-class of fully representational states. These are representational states the contents of which we do not endorse at the cognitive level, and their qualitative phenomenology is exhausted by the facts about how things appear, that is, by facts about what qualities show up in the intentional or representational content of the experience. By examining the claims of two defenders of the qualia view, Ned Block (1996) and Amy Kind (2008), I will try to show that the invalid argument against representationalism rests on confusing non-representationality with what I want to call “non-endorsement of mental representations”. That is, these authors themselves describe such states as if they were non-endorsed representations, before making the logically illegitimate move that they are non-representational. I will also speculate about the causes of the confusion. I will suggest that the content of the experiential states deployed in such arguments differ from other non-endorsed representations in being either obviously non-veridical or being impossible, and so they lose their belief-inducing function, becoming automatically non-endorsed states. This makes it harder to see how they could be true or false in the first place, as we almost never have environment-directed beliefs, or even explicit disbeliefs, triggered by these states. When this is coupled with an ambiguity in our appearance-talk, some of us are lead to the confused conclusion that such states are non-representational. When we are helped out of the confusion, we see that there is no need to postulate qualia to analyze such states. There are, for sure, various arguments in defense of qualia, and the appeal to afterimages and the like is only one such argument. Qualia theorists, in general, can be, and often are, fine with the idea that there are representational aspects of experience and there can be a close relation between qualia and the representational content. But arguments that appeal to afterimages and the like rest on a specific phenomenological analysis of these experiences which aim to show that these experiences are (partly) non-representational. So these particular arguments for qualia shall be rendered unmotivated with the help of a full representationalist analysis of these experiences.
Before we move on, it should be noted that to demonstrate that a mental state is representational, one doesn’t need to demonstrate that the represented qualities are represented as environmental, material or mind-independent, or external to the body or to the subject. It is possible that these states represent something as immaterial, mind-dependent or mental (whatever “being represented as mental” may amount to), or as internal to the body or to the subject. They can represent objects as internal to the head, like right behind the eyes or on the eyes, in the ear, or somewhere in the head (as in the experience of internal speech and inserted thoughts). Representation is not only a matter of representing things in the external physical environment. As Byrne puts is:
“The subject [of an afterimage experiment] attends to the world as it appears to her, just as she did in the initial experiment [involving colored chips.] … According to some, it appears to the subject that a red filmy thing is some indeterminate distance from her eyes. According to others, it doesn’t appear to the subject that the object is in her physical environment at all: Instead it appears that the object is in some inner realm. However, this does not affect the point that the subject can only discover the phenomenal character of her experience by attending to the world (either external or internal) as her experience represents it.” (Byrne 2001, 13)
Also, it needn’t be the case that these representational states represent objects in the ordinary sense of the term “object”. It can be that they represent properties like color and shape without these properties being bound together into what some would call proper objects, so the properties end up being represented as properties of “areas” or “patches”.[6] These states would still have intentional objects understood in the sense of “object” expressed by the term as it is used in a representational context, while these intentional objects may not be proper “objects” in the other sense of the term[7] – they can belong to ontological categories such as properties and events.[8] In any case, I will be arguing for the position that, whatever the other details about these representational states are, the properties being represented in visual phenomena like phosphene and afterimage experiences are those like color, shape and location which we also find in more ordinary visual states.
One last terminological point I should note is that the term “representation” that I rely on throughout this paper does not necessarily rest on an externalist or relational understanding of the term found in accounts that analyze experiential representation as requiring some causal-informational relation to a concrete entity that is the represented object or property.
The discussions in this paper concern a dispute between representationalists and qualia theorists, without touching the disagreement between representationalist and anti-representationalists like direct realists and sense-data theorists. Some of my assumptions will not be argued for and will be just taken for granted, and that is because it is an uncontroversial point as far as the debate between the representationalists and the qualia theorists is concerned, even though it might be found controversial by the direct realists and friends of sense-data. Nothing said here is aimed at persuading them, though perhaps the representational analysis of afterimage experiences and the like can provide them examples to get a grip on how experience in general can coherently be analyzed as a representational phenomenon.
In the next section, I elucidate the concept of a non-endorsed representation, and in section three, I introduce the concept of an automatically non-endorsed representation. In section four I offer an analysis of double vision as an automatically non-endorsed representation, and in section five I do the same for afterimages and phosphenes. In section six I borrow an argument from David Bourget (2015) to help demonstrate further the representational aspect of the aforementioned visual experience types. In section seven, I analyze passages from Block (1996) and Kind (2008), and try to show that their anti-representational arguments fail and that the failure is due to a confusion. Section eight investigates the psychological and linguistic roots of this confusion, pointing to an ambiguity in our appearance-talk and to special features of afterimages and the like that make them different from other illusory experiences that are not deployed for anti-representationalist arguments. The ninth section concludes the discussion, briefly hinting at similar strategies to resolve the debate about other allegedly non-representational states like pain and orgasm experiences.
2. Non-Endorsed Representations
Hereby I introduce the term “non-endorsed representational content”: A mental state has a non-endorsed representational content if it has truth-evaluable content but the subject doesn’t take the content as true on a higher, cognitive level.[9] This doesn’t mean that the subject necessarily takes the content to be false. As we will see, there are cases of non-endorsed representations which are neither taken as true nor taken as false, states that have completely lost their automatic belief (or disbelief) inducing role and also probably have lost most of their other causal roles that produce intentional behavior. Maybe it would be better to coin two different terms for these two different phenomena, to capture the difference between, on the one hand, a subject explicitly disbelieving the content of the experience, and, on the other hand, neither explicitly believing nor disbelieving it, but my terminological aims will be less ambitious. And perhaps there can be a non-conceptual feature of phenomenal experience that is something like a pre-cognitive analogue of non-endorsing, but for the purpose of this paper, we need not enter that territory.
Let me make clearer what I mean by “endorsement”. There is a type of endorsement which is directed at the representation itself, which requires the subject to understand that she is in a representational state, and to hold the belief that this state represents the world veridically.[10] This kind of endorsement is presumably limited to adult human beings and perhaps some animals that have the concept of representation or the concept of an appearance-reality distinction. However, there is a more basic type of endorsement, which can also be applied to non-human animals and human infants. This kind of endorsement can be simply defined by a way that the world looks to a subject, and given no disturbance from an appearance-reality distinction at the cognitive level, what appears to exist is the same as what is taken to exist.
I will leave it open whether creatures without higher-order beliefs directed at their own mental representations can have non-endorsed mental states or not, that is, whether there can be a purely non-conceptual state of non-endorsement, but as I remarked above, nothing I will say in this paper hinges on this. Also, I will not be concerned with the question of whether all mental states need a higher-order state to become endorsed (which would imply that some mental states will always end up non-endorsed), and I will also not be concerned with the question of whether some states, like belief, carry a “built-in” feature of endorsement and are automatically endorsed unless challenged by a higher-order doubt. In the following discussion, I am mainly concerned with non-cognitive states such as vision that lose their function of inducing beliefs about things that they ordinarily do induce beliefs about, and that will be what I am primarily trying to capture with the term “non-endorsement”.
The Müller-Lyer illusion is an often presented example for what I have called a non-endorsed representation: Two lines seem to be of different length due to the different direction of the arrowheads at the ends of the lines, but they are indeed equal. As soon as you learn that this is an illusion, you stop believing what is represented. But the phenomenal state still keeps representing the world in the way it did, it retains its visual representational content. This applies to all illusions the illusoriness of which we are aware of while undergoing them, and to pseudo-hallucinations like those had by subjects with Charles Bonnet syndrome. Lucid dreams and some hallucinogen-induced experiences are among other examples. In all these cases the world is represented to be in a certain way, it phenomenally seems to a subject that things are some way or other: That one line is longer than the other, that the clouds are getting twisted, or that there is a purple monkey in the room. However, such representations are not endorsed by the subject. We can ordinarily talk of such states in a way of saying “that looks thus and so to my eyes, but I know that it isn’t thus and so”. In these states of non-endorsement there is a clash between what is phenomenally represented and what is believed, or to put it idiomatically, one’s mind or senses are playing tricks on one while one is aware of the tricks.
There are also some states, like imagination, which, by their nature, do not make us committed to endorsing their content. When you imagine a yellow lemon, you do not believe that there is a yellow lemon, nor you disbelieve that there is a yellow lemon. Imagination does not induce true or false beliefs, at least not the kind of belief that is targeted at something in the perceptible environment that should match the content of the act of imagination, just like a painting of a yellow lemon is not in the business of making you believe that there exists or doesn’t exist a yellow lemon. Still, like the painting, the state has an intentional object: A yellow lemon. It represents a yellow lemon. By visually imagining, one forms in one’s mind a contentful visual state, representing objects and their properties as viewed from a certain perspective.[11] One can find other examples, from standard experiences or clinical data, where the appearance content stays the same, but the feeling of presence, or the feeling of reality, and therefore the strength of endorsement, differs.
The notion of a non-endorsed phenomenal representation can be elucidated further with similar states that are the non-sensory, cognitive counterparts of these representations: Sometimes we can entertain a thought, or have a feeling that something is the case, without believing that the content of these cognitive states are true (or false). We can understand someone else’s thoughts through her utterances without believing (or disbelieving) their content. By understanding the proposition expressed, and before judging it, we represent the world in a certain way that we don’t necessarily take to be true (or false).
So far, I have not said anything that contradicts the qualia theory. Many defenders of qualia will readily acknowledge the existence of what I have called non-endorsed phenomenal states, as there doesn’t seem be a reason to deny them. And it is important to note that arguments for qualia do not rely on states like the Müller-Lyer illusion, at least not in a way that depends on an alleged non-representationality.[12] But beginning with these phenomena will help us clarify things as we move on to more philosophically problematic phenomena, as I will later investigate why some philosophers have problems with the representational analysis of phenomena like double vision or afterimages, while they would grant that other illusions can be non-endorsed representations. Before doing that, I shall demonstrate that afterimages, phosphenes and double vision are indeed a type of non-endorsed representation, the non-endorsement of which results automatically rather than with theoretical reflection unlike in the case of the Müller-Lyer illusion.
3. Automatically Non-Endorsed Phenomenal Representations
In the previous section, I described non-endorsed representational states. These are representational states that represent the world in certain ways, but the veridicality of which the subject is not committed to. Now I want to turn to a special class of non-endorsement, a non-endorsement of an automatic sort which does not require conscious deliberation on the side of the subject regarding the illusoriness of the experience in question. My primary aim is to analyze the representational features of these states, and my remarks about how they become automatically non-endorsed should be taken as speculative psychology, serving here the function of helping elucidate the analyzed phenomena. The question of what causal mechanisms make these states non-endorsed is an empirical question, not a philosophical one. My suggestions regarding these mechanisms might be, and most probably are, wrong or incomplete, but these suggestions should serve as illustrative possibilities that shall help the reader understand in what way the analyzed experiential states are representational states, and what, in general, it means for a representational state to become non-endorsed.
Afterimages, phosphenes, tinnitus, and psychoactive-induced experiences all represent the world as being in weird ways, a very fundamental weirdness that contradicts the basic facts we know about the physical world and how it interacts with our sense organs.[13] Now, the content of other non-endorsed representations can also be weird, but I am trying to point to a more fundamental weirdness that is found in automatically non-endorsed representations. I am talking about experiences the contents of which are so weird that we don’t even take these experiences as information-carrying states, and we do not even attempt to develop a proper vocabulary to talk about how the world looks while we are undergoing these states, unless a philosophical disagreement about them prompts us to.
In the previous section, I mentioned some non-endorsed representations like the Müller-Lyer illusion, which the defenders of qualia do not deploy for their arguments. Those states differ from afterimages and the like in that they represent the world in rather ordinary ways. The content of such states can be “extraordinary” in some sense, like when a subject with Charles Bonnet syndrome hallucinates a bright yellow monkey in the room, but still, the content is ordinary when it comes to the general properties represented: There appears to be a stable middle-sized object with a certain color, shape and location in the visible environment. The extraordinariness is due to some background contextual knowledge: We do not live in a world where bright yellow monkeys pop up in one’s room out of the blue. If a subject would have a strong background belief that monkeys with crazy colors were the kind of things that popped up in one’s room every now and then, then there wouldn’t be anything extraordinary about the content of such experiences.
Similarly, when we look at cases of non-endorsement as in the Müller-Lyer illusion, imagination, or hallucinogen-induced experiences, we see that what makes these states non-endorsed is not the basic physical features represented by the experience. Rather, the reason for the non-endorsement is due either to other acquired beliefs (e.g. we know by our own or someone else’s measurement that the Müller-Lyer lines are actually equal) or the way the experience is caused (e.g. we know that we are merely imagining, or that we have taken hallucinogenic substances, etc.). But when it comes to cases like afterimages and double vision, the extraordinariness is due to some very fundamental physical properties things are represented to have, which clashes with our very basic knowledge of the perceptible environment and how we relate to it through our sensory organs. As we will shortly see, a lot of the extraordinariness seems to have something to do with the represented spatial features.
Among the representational phenomena I will describe, I leave it to the reader to decide whether what is found in their content is merely weird or impossible. Perhaps what makes these representations automatically non-endorsed is not our beliefs regarding the basic physical structure of the environment. But an incoherency within the visual content of these experiences that makes it the case that things just can’t be the way they look, that is, these representations would be non-endorsed representations regardless of what we believe about the environment. Whether phenomenal states can represent impossibilities is a controversial matter, but in any case, nothing much hinges on this for the purposes of this paper, as the extraordinariness of the content, whether impossible or not, suffices to elucidate the nature of the experiences to be discussed. However, the possibility that these states might be representing impossibilities should be kept in mind as an option.[14]
4. Double Vision as an Automatically Non-Endorsed Representation
Double vision represents things as double. That is, presumably, why it is called “double vision”. We understand what is meant when people ask each other “how many fingers do you see?” to find out how drunk they are. This representational aspect of double vision, however, has been disputed, and it was claimed that double vision phenomenologically differs from ordinary vision without differing in representational content. Boghossian and Velleman, for instance, have claimed that:
“If you press the side of one eyeball, you can see this line of type twice without seeing the page as bearing two identical lines of type. Indeed, you cannot even force the resulting experience into representing the existence of two lines, even if you try. Similarly, you can see nearby objects double by focusing on distant objects behind them, and yet you cannot get yourself to see the number of nearby objects as doubling.” (Boghossian and Velleman 1989, 94)
What I will aim to do in this section is to convince the reader that the phenomenology described by Boghossian and Velleman above is better analyzed as a case of non-endorsed representation of doubleness.[15]
We can begin to demonstrate the representational aspect of double vision with the help of anecdotal examples where the illusion actually works: Once I was standing by the side of a dark road, and saw the headlights of a vehicle approaching. There were, apparently, two headlights, and they were the only visible parts of the vehicle. Given that vehicles with two headlights are cars or trucks, I thought it was a car or truck approaching. However, I was subject to alcohol-induced double vision. There was in reality only one headlight, but it visually seemed to me that there were two. As the vehicle drove closer, the two apparent headlights collapsed into a single headlight, and I realized that it was not a car or a truck, but a motorbike.
This is a rare example where the representation of there being two things is endorsed by the subject, as the overall visual state in this example doesn’t have some of the peculiarities of most double vision experiences. The only represented objects in the whole visual field are two headlights surrounded by darkness, while in most double vision experiences in better light we are presented with a wholly doubled world, with things spatially overlapping with each other in a strange way, while we can identify the two apparent instances of the doubled thing as the same thing. In such cases, it is easy to see one reason why the representation is never endorsed, as there is something logically awkward about one thing being two: Two things are conceptually identified as the same thing in many occurrences of double vision, possibly because we know that there is only one thing, and unlike in the headlights case above, we see the “two things” as somehow superimposed. When we see something double, we see one thing (on the cognitive or conceptual level of representation) as double (on the visual level of representation.) There are also spatial peculiarities about how the two things are represented: Their locations can be identified in relation to each other, say, one being on the left, but still, it is not like the two things are taking up space in the same three-dimensional spatial field. They are usually partly superimposed in an unusual, ghostly way. That is quite different from how we ordinarily experience the world to be and how we think it actually is.
Perhaps the causal history of how we get these experiences also contributes to the automatic non-endorsement of these representations. Most of the time, we know that causes of double vision come from inside and not from the world being really that way; it is either due to consumption of alcohol or some other substance, or due to a failure to focus. However, our knowledge regarding the causes of the experience might not be a necessary factor for the non-endorsement; we can speculate that the spatial way things are represented is most probably sufficient. Because of the peculiarities of the content we find it difficult to see how double vision could be endorsed in the first place, how to even exactly describe the spatial properties objects should have for this appearance to be truthful. The disregard is so automatic that the relevant feature of the representation loses its belief-triggering function at the cognitive, non-experiential level (except for occasionally triggering beliefs about the contents of one’s visual experiences). Still, what separates double vision from ordinary vision is a representational fact, no matter how weird that represented world may be, and how hard it might be to describe the truth conditions of this representation.[16]
5. Phosphenes and Afterimages as Automatically Non-Endorsed Representations
Phosphenes and various other closed-eye visuals are examples of visual phenomena which are sometimes described in English via phrases like “the play of shadow and light”, or “booming-buzzing confusion” for more anarchic cases. A world of unstable, two-dimensional chaotic colored patches is much different than the world we are accustomed to seeing, which is a world of stable, middle-sized, three-dimensional objects that behave in broadly predictable ways. Some might be tempted to say that these chaotic visual experiences are not “world-suggestive”, but this would only amount to saying that they do not suggest a world that we are familiar with, and not that these experiences do not suggest a world at all. There could be such worlds, after all. Some artists try to depict them, and some photographers find little worlds in nature that look like psychedelic closed-eye visuals and draw our attention to the similarity, like natural patterns on a stone, or the northern lights. Lycan similarly suggests that “[…] given any visual experience […] there is some technological means of producing a veridical qualitative equivalent—e.g. a psychedelic movie shown to a subject in a small theater” (Lycan 1987, 90). However, as Schroer (2004, 536) also notes, we don’t need to put things this way in order to point to the representational nature of these experiences. It might be the case that there is no technological means to produce a veridical equivalent of these experiences, just like there is no technological means to produce the veridical equivalent of a belief in a magical wand, a round square, or the world of double vision. What is represented might be a nomological or logical impossibility.[17]
As in the case of double vision, there might be factors other than the content that might be playing a role in the automatic non-endorsement of this kind of experiences. We have experiences of phosphenes and afterimages from childhood on. We are not surprised by such experiences and we do not ever wonder if they are truthful. We know how to get them, and how we get them is not by orienting ourselves towards some perceptible external object to be explored by sight, but by playing around with our visual apparatus, say, by closing our eyes and rubbing them. Moreover, the spatial properties of these experienced objects make it impossible for them to be objects that can be explored by ordinary visual means – at least for phosphenes, which are not represented to be in front of the eyes, but rather on something like, as it were, an internal screen which has some of the characteristics of the ordinary visible environment, like color, shape and location.[18]
Afterimages require a more complicated analysis when it comes to their experienced location, as they are usually experienced with eyes open, but still linger when you close your eyes. They can switch between being “out there” and being on an “internal screen”. Afterimages also seem to occlude other things in the visual field (which, by the way, is another fact that should make it clear that having an afterimage is a modification of how the perceptible environment is represented.) These represented spatial properties probably constitute the most important factor for these experience’s being automatically non-endorsed.
In order to point out that these experiences are representational or intentional, some philosophers have pointed at the fact that we can talk about them by using the vocabulary we use to describe ordinary phenomena perceptible by vision, like color, size and location. Tye claims that a red afterimage backed by a yellow background is “similar perhaps to that of viewing (in dim lighting) a bloodstain on a transparent sheet of glass suspended between oneself and a yellow background surface” (Tye 2000, 85), while Schroer provides an anecdotal example of an endorsed afterimage experience he had while fixing a lightbulb, where he
“immediately took the object to be a red beanbag. […] It did not occur to me that there really was no beanbag and that I was merely having an afterimage experience until I looked away from the table and the object in question suddenly moved and changed in several very unexpected ways.” (Schroer 2004, 543)
Phillips, meanwhile, proposes that afterimage experiences are best described as “illusory presentations of light phenomena often apparently projected from the subject’s point of view” (2013, 433).
The above examples help demonstrate that these experiences are representational, but we should keep in mind that these examples are not strictly necessary: It is a possibility that the content of a representational experience may not be similar to anything that can be veridically represented, or for some experiences it can be the case that there are absolutely no cases where the experience is taken to be veridical. Also, if our aim is to provide anecdotal examples to make a case for the contentfulness of these experiences, we don’t need to search for cases where the experience is taken to be veridical, as it will do equally good to find cases where it is taken to be illusory, since the claim of the anti-representationalist is that these states are neither veridical nor illusory, as these are, allegedly, not contentful states at all.
Another way to demonstrate the representational nature of these experiences is to think of situations where an automatically non-endorsed representation regains its belief-inducing role. Floater experiences provide us with an example to use as a nice springboard for this purpose. Floaters are specks in one’s eyes which are normally transparent, but may become visible due to degeneration in the eyes, which makes subjects experience these specks as floating in front of, or inside, one’s eye. Floaters are especially vivid against a white or monochromatic background, and they move along with one’s gaze. Some of the spatial properties of floaters make them similar to other visual noise like afterimages. Indeed, when I first learned about them, I was surprised to hear that experiencing a floater is a type of veridical perception, as I have always thought that they were a species of visual illusion created by the condition of my eye. When I acquired this information, floater experiences regained their object-targeted belief-inducing function (replacing the occasionally induced belief that there is something wrong with my visual system because I am experiencing illusory patches).
Floater experiences initially lose their belief-inducing function most probably because floaters are spatially similar to the patches experienced in other visual illusions like afterimages (having a strange location, moving along with the eye, etc.), and because one cannot behaviorally interact with a floater any more than one can interact with an afterimage, besides moving them along with one’s gaze. One cannot touch or hold a floater, or visually investigate it by further usual means.
Floaters are not the kind of phenomena that are brought up in defense of the qualia theory, so why did I bring up this example? Because floaters constitute a real-world example to think of how other non-endorsed representations could regain their belief-inducing function. Think of a scenario where visual scientists discover that, like floaters, phosphenes are actual colored patches that temporarily exist in your eye, or think of a naïve person getting fooled into believing that in phosphene experiences, she experiences colored things in a parallel reality inaccessible to others,[19] or rare real-life cases like retinal detachment (discussed in Gow 2019 in the context of the representationalism-qualia dispute) where subjects undergo phosphene experiences with eyes open and initially take the phosphenes to be actual flashes of color and light. The fact that we can coherently imagine these scenarios should help us understand that these experiences have visual representational content of a non-endorsed type.[20]
6. Bourget’s Argument from the Loss of Visual Information
Before moving on, I will briefly look at an argument by Bourget (2015) where he similarly argues for representationalism for cases which he classifies as cases of visual distortion, including double vision and blur. Bourget invites us to consider cases of ordinary vision, such as seeing a square, and then think of a series of cases where the visual effect (blur, double vision, etc.) is gradually amplified, and draws our attention to the fact that in cases of higher distortion, it is obvious that there is an impoverishment in the representational content of the experience (e.g. we cannot see a square as a square anymore as the blur or double vision gets extremely amplified.) The cases of extreme distortion are not everyday cases of perceptual distortion, but Bourget asks: If, the phenomenal character is independent from the content in mild cases of distortion as anti-representationalists argue, then at what point does the content get intertwined with the phenomenal character of a distorted experience as the distortion gets amplified? Bourget concludes that all possible answers to this question are implausible and/or ad hoc.
I do think that this presents an interesting case against anti-representationalism for double vision and blur. The primary reason I bring up Bourget’s argument, however, is that it provides an interesting phenomenological exercise for the qualia theorist that could help her understand in what way the allegedly non-representational aspects of experience are actually representational, and help her have a better grasp of what it means for a state to be automatically non-endorsed. Imagining a gradual transition from ordinary to extreme cases can help one see that ordinary cases also involve a loss of visual information, even though the information that is lost might be little. The exercise should also help us see why we do not prefer to have these distorted experiences, that is, why we avoid them. If what makes an experience an experience of double vision or an afterimage had nothing to do with the representational content, then our preference regarding whether to have these experiences or not would merely be a cosmetic preference, it would be a preference regarding which “mental paint” we would like our experiences to be painted with: Whether we would strive to be or not to be in such visual states would depend, perhaps, on which phenomenology we enjoy more, and not on whether we prefer to see the world as it is or see it in a distorted manner. One can, of course, from time to time, prefer to have these experiences, mainly for phenomenological curiosity or merely to enjoy a certain experience. Some children, philosophers, cognitive scientists, and recreational users of psychedelic substances do occasionally strive to have experiences of visual distortion. But for the sake of seeing the visible environment clearly and as it is (rather than blurrily or doubly), these experiences are experiences that we prefer to avoid, because their representational content is distorted. We, of course, rarely complain about ordinary experiences of afterimages or double vision, because they are temporary and mild. But imagining extreme cases of these visual distortions should make it more obvious why it is not preferable to have these experiences for epistemic reasons, and therefore make it obvious that they have representational content.[21]
7. Confusing Non-Representationality with Non-Endorsement
I hope I have so far established that double vision, phosphenes and afterimages can be analyzed as non-endorsed representational states, and this non-endorsement is an automatic one in great majority of cases. Now I wish to demonstrate that some defenders of qualia confuse non-endorsement with non-representationality when talking about these experiences.
In his Mental Paint and Mental Latex, Ned Block suggests that “[Afterimages] don’t look as if they are really objects or as if they are really red. They look … illusory” (Block 1996, 32; ellipsis in original). He uses this observation in support of non-representationality of such experiences, and therefore the existence of qualia: In such experiences, it cannot be the representational content that is responsible for the qualitative phenomenology we are introspectively aware of, so it should be a non-representational property such as a quale (or “mental paint”) which constitutes that phenomenology, such as qualities instantiated in experiences.[22] Interestingly, Block argues for the anti-representationalist conclusion via mentioning illusoriness, which is a feature of representations.
I take it that the ordinary understanding of visual illusion is the following: Something visually seems to be a certain way, but it is not really that way. But what would it mean to say that afterimages “look illusory”, or that “afterimages don’t look as if they are really red”? It is easy to understand what it means to say that something does not look red, but what about something’s “not looking really red”? Whatever Block means by that, he seems to talk about something that looks or does not look some way or other, which means that the experience has an intentional object that is represented to be some way or other. The statement about an object’s “not looking really red” is a statement about how it looks, how it appears to be. In more technical terms, that is how an object is represented by the experience.
And what about “looking illusory”? Does Block mean that there is a brute represented feature of “unrealness” or “illusoriness” attached to these represented objects or properties at a pre-cognitive level?[23] If we could make sense of this, and if it were indeed the case, the state would still be a representational one: Redness is represented, and it is represented as “unreal”. Or does Block mean that something looks illusory in the sense that the way that thing is represented makes it unlikely that it is real, because of the weird or contradictory properties it is represented to have, so it “looks illusory” at a judgmental level? If we follow this interpretation, we should take Block as implying that the object of the experience is represented in such a way that it cannot be real, or we automatically know that it is not real. But neither interpretation can help one reach an anti-representationalist conclusion about afterimage experiences, because in both interpretations “illusoriness” is understood as a representational phenomenon.
Perhaps, one might think, in order to interpret Block properly, we should turn to what an illusion is supposed to be according to qualia theory. Qualia can be thought of as vehicles of representation that do the representing. A certain quale’s instantiation makes the mind represent a certain property. How this representational feature is to be understood depends on the details of the theory. According to externalist theories, a quale represents something in virtue of the instantiation of that quale’s being in some causal-informational relation with instances of the represented property. A “red quale” represents redness (or some other property of red-looking objects, if one is an anti-realist about colors) in virtue of being reliably triggered by redness (and by fulfilling some other conditions depending on the details of the theory). According to internalist theories of representation that rely on qualia, a “red quale” represents redness either in a projectivist manner, where subjects mistakenly take these experiential qualities to be the qualities of external objects (and form a useful but erroneous representation of the world with the help of qualia and the act of projection), or somehow end up representing red-looking objects if the experiences, somehow, acquire a representational aspect which signals that these experiences are caused by external things. In all of these theories, illusion is a state where a quale is instantiated, but the thing it is supposed to represent is not there. If the qualia theorist wants to claim that an afterimage experience is the bare awareness of a quale-complex, without any representation going on, “illusoriness” is not the concept to be alluded to in order to demonstrate that. Such a state would be neither veridical, nor illusory. A qualia-theoretical analysis of an afterimage experience can be given without appealing to illusoriness, by saying that “when we introspect an afterimage experience, we find a mentally-instantiated quality, which does not represent anything”. This formulation would not face the problems that the illusion formulation faces, but it faces other problems: It has no appeal for people who, introspecting an afterimage experience, find no qualities other than those like colors that objects appear to have also in uncontroversially representational experiences, which makes it natural to think of afterimages as a special type of misrepresentation.
Is there perhaps some reason, that introspection can provide us, which would urge us to conceptualize afterimages differently, in a way that involves qualia? Let’s turn to Amy Kind, who, in her article How to Believe in Qualia, gives us instructions to find qualia in afterimage experiences:
“In general, afterimages occur subsequent to the removal of some original (usually intense) stimulus. When a camera flash goes off, you might experience an afterimage in front of the photographer’s face. If you stare intently at a bright light for a little while and then close your eyes, there will be a lingering glow in the darkness. And if you stare at a green dot for half a minute and then shift your attention to a bright white piece of paper, you will visually experience a red dot similar in size and location to the green dot you had been staring at. But in none of these cases does it seem as if the afterimage represents something that is really there. When you close your eyes after looking at the bright light, for example, you don’t take the lingering glow to be on the inner surface of your eyelids. When you see the red afterimage against the white page, you don’t take the redness to suggest the existence of a red dot on the page.” (Kind 2008, 289)
Kind puts the bottom line with the aforementioned quote by Block, where he mentions the illusoriness of afterimages. Note that Kind talks about these allegedly non-representational experiences by mentioning the same kind of experienced phenomena found also in ordinary experiences: Both before and after looking at the bright light and then turning away, we experience brightness, referred to as a “glow” by Kind; after staring at the green dot, we see a “red dot similar in size and location to the green dot” we have been staring at. Kind holds that these experiences are not representational: Unlike the experience we have when we are looking at a green dot, which represents a green dot, the corresponding afterimage does not represent a red dot. But isn’t this claim of non-representationality contradictory to what Kind says, that we experience a red dot? Isn’t the red dot the intentional object of the experience? Can we coherently assert that we “experience a red dot” while it is the case that a red dot is not represented, that it doesn’t visually seem to us that there is a dot which is red? I think not. If we experience a red dot, it follows that it seems to us that there is a red dot. “Seeming” is of course used here in the pre-cognitive (or “non-conceptual”) sense, in a way that allows us to say “it seems to me [in the first sense of “seeming”: it seems to my eyes] that there is a red dot, but it doesn’t seem to me [in the second sense of “seeming”: I don’t judge] that it is real”.
I think what is going on here is a confusion, taking non-endorsed representations as non-representational. Saying that a non-representational state is illusory, as Block does, or saying that we experience something without that something being the object of experience, as Kind does, is a contradictory way of talking. In some experiences, of course it can appear to us that the experienced thing doesn’t exist, if things can “look illusory” in some sense, but that would also create no motivation to postulate the existence of mental qualities to account for the experience, as those experienced qualities don’t seem to exist. Merely intentional qualities of merely intentional objects that figure in the representational content would do the job, or to put it another way, the fact that things seem a certain way, that there seems to be an “illusory redness”, can fully account for the qualitative phenomenology.
8. Possible Causes of the Confusion
In the previous section, I made an attempt to show that Block and Kind confuse two different phenomena, an experience’s being non-representational and its being a non-endorsed representation. Now I want to investigate the psychological and linguistic causes of this confusion. What makes some of us prone to making this logically invalid move? I speculate that one determining factor here might be a verbal ambiguity in some expressions related to mental representation, expressions in the form of “it seems to me that X” or “it appears to me that X”. As I have mentioned earlier, one can talk of seeming in a cognitive way, as opposed to a visual/experiential way, as in phrases like “it seems to me that…” followed by a clause that expresses a proposition associated with a belief or belief-like state. So when you are in a visual state where the world in front of you seems to you to have such and such objects and properties, it is still possible that it doesn’t seem to you, in the cognitive sense, that there are such objects and properties: Your mind is in two conflicting representational states at two different levels, and you endorse the higher, cognitive one. Even though at the end of the day the world doesn’t include a red patch according to you, the cognizing subject, there is a red patch in front of you according to your visual system.[24] The logical fallacy lies in the following move: Starting with the premise that the world doesn’t seem to be like this or that to you at the cognitive level, and moving to the conclusion that it doesn’t seem to be like this or that to you at all.
But why don’t we come across a similar fallacy when it comes to other non-endorsed visual states, like the Müller-Lyer illusion? I suggest that the answer can be found in the difference between how the visual state at hand becomes non-endorsed. If the illusoriness of the visual state is not so phenomenally obvious, and when we reflect on these experiences, we form higher-order beliefs that take as data the visual state already conceptualized as a representational state, and when we come to know about the illusoriness, we are in a state of explicit disbelief. To put it another way, when we are looking at the Müller-Lyer lines, understanding that we are in an illusory visual state involves, first of all, understanding that we are in a representational visual state, a state with a truth-evaluable content, and understanding that the content is false. So, no one would argue for qualia for a state like the Müller-Lyer illusion on the grounds of non-representationality, as this state causes explicit belief or disbelief in us, and having explicit belief or disbelief that takes visual appearances as data means that we are aware of the visual representational state as a representational state. (It is called an “illusion”, after all.) But we rarely have explicit belief or disbelief in cases of afterimages or phosphenes. The factor that would prevent a similar fallacy in the Müller-Lyer illusion case is missing from the cases of automatic non-endorsement.
One could argue that I am not being charitable here to Kind and Block, and say that even though they might be mistaken, they are not confused: Maybe I shouldn’t take their descriptions of these states at face value. Let’s follow this suggestion and try to read Block and Kind in a way that wouldn’t make them contradict themselves.
As for Block’s way of putting things, it is hard to come up with a charitable reading. How can a talk of “illusoriness” be translated into non-representational talk? I do not know. However, there can be prospects for charity for some aspects of Kind’s description of afterimages. When talking about qualia, it is common that theorists use terms that are normally used to talk about represented properties, like when some philosophers say things like “a red-quale”. By this, they don’t mean that a certain quale is itself red, or that it necessarily represents redness. What is meant here is that the quale is the sort of quale we find in experiences that ordinarily represent redness. Also, it could be that when these philosophers talk of experience in a way where the grammar suggests an act-object structure, we shouldn’t accuse them of failing to recognize that if a state such as an afterimage experience is supposed to be non-representational then it cannot have an act-object structure. Maybe when Kind says that “you might experience an afterimage in front of the photographer’s face”, she uses “experience” in the same way when we say that we “experience joy”, where joy is not (or not so obviously is) the intentional object of the experience.
If this charitable reading is correct, then all of what Kind says should in principle be translatable to non-intentional qualia talk. According to this translation, in the above sentence, “afterimage” doesn’t refer to the intentional object of the experience. Rather, the whole phrase “experiencing an afterimage” refers to a non-intentional mental state, as in “experiencing joy” (conceptualized for the sake of the argument as a non-intentional mental state), even though the “an” in “an afterimage” sets off some alarms: We can experience two afterimages at the same time, but can we experience “two joys” at the same time? Perhaps, and perhaps not, but in any case the parallelism already seems suspicious.
In any case, even if a non-intentional reading of “experiencing an afterimage” is possible, what to do with “an afterimage in front of the photographer’s face”? My mental state is not in front of the photographer’s face. Well, if “an afterimage” refers to an experience, and mental state tokens are identical to neural state tokens, and I am in front of the photographer, then perhaps my experience is in front of the photographer’s face in a sense, but this is obviously not what is meant by the phrase. Maybe we should understand it in the same way we are supposed to understand red-talk about qualia: Just like there can be red-afterimages in the sense that they instantiate a quale that we also find in red-representing states, there can be in-front-of-the-photographer’s-face-afterimages, where we find the type of qualia, presumably shape and location related qualia, that our minds are supposed to instantiate when looking at faces or face-shaped objects. This translation doesn’t seem impossible, but nevertheless, as Kind is describing an environmental situation that causes us to have afterimages, and our experience of the photographer’s face, unlike the afterimage, is uncontroversially representational, it is much more plausible to read “in front of the photographer’s face” as referring to nothing other than a spatial property related to the photographer’s face as it is represented by the experience. So, after all, there visually appears to be something in front of the photographer’s face.
Kind can insist that this natural reading is not correct, and the qualia-theoretical analysis is the correct description of the situation. But we should remember that Kind’s article is titled “How to Believe in Qualia”, and it is aimed at the reader who doesn’t believe in qualia yet, or may not even have the concept of a quale, but who will be made to believe in qualia after following some instructions. So the initial steps of the description of the mental state and the description of the environmental factors that trigger it should not have references to qualia until the reader stumbles upon the relevant property by following the introspective steps. It would be surprising if Kind was expecting the uninitiated reader to translate this in-front-of-the-photographer’s-face-talk into qualia-talk. More plausibly, what she is doing is just describing a non-endorsed representational state, with ordinary language that refers to represented objects and properties, such as the photographer’s face and an illusory glow in front of it.
Perhaps there is a way that Kind could provide us a with a complete qualia-theoretical interpretation of the way she describes the afterimage experience that could get around the above objections, so maybe I am mistaken in attributing her a confusion. But that still wouldn’t make the qualia theory more appealing. There is, after all, no apparent reason why we should interpret the ordinary talk of afterimages in such a complicated fashion and prefer the qualia view for a philosophical analysis of the mind. We have a simpler analysis which treats these states as illusions of an automatically non-endorsed sort, an analysis that applies to ordinary talk of afterimages, a talk which sounds linguistically the same as Kind’s description of afterimages, but still makes sense without being translated into something very complicated that allegedly lies beneath the surface linguistic form and that relies on a property called qualia that we have independent reasons to be suspicious of.
9. Concluding Remarks
There are several other kinds of experiential phenomena, experiences with affective aspects like pain and orgasm, which are also among the armaments of the anti-representationalist. I want to briefly point to the similarity of these phenomena to the cases discussed above. These states might also be among the states which sometimes lose their belief-inducing function regarding the veridicality of their contents, not because they represent impossibilities or oddities, but because we care more about the very existence of these experiential states than their veridicality, as feeling pain on a real arm is as undesirable as feeling pain on a phantom arm.[25] I will not provide a defense of representationalism about pain and orgasm along these lines here as it is beyond this paper’s scope, and leave it at pointing at hints for this strategy found in the literature (see the discussion in Aydede 2009, sections 3.2 and 4.2). But if this initial line of thought is on the right track, in the future it can provide us with an explanation of the confusion of the anti-representationalist in the case of affective states as well.
In any case, I hope what I have argued above makes a case for representationalism for afterimage-like experiences. These experiences are automatically non-endorsed representations confusedly taken to be non-representational states by some defenders of qualia, and when we examine the way they define these states, we see that they are actually talking about these experiences as if these states were a type of illusion, that is, a type of mental representation with truth-evaluable appearance content. And their representational nature makes the postulation of qualia unnecessary. They represent the world in certain ways, and if the represented qualities are not in the world, then they need not be instantiated in the mind or anywhere else. They simply do not exist. Afterimages and the like do not trigger explicit belief or disbelief at the cognitive level, but the lack of a representation at the cognitive level is not an obstacle to a representationalist analysis at the non-conceptual visual level. Compare with visual imagination: Imagination is also a representational state, and visually imagining a red patch does not make you believe or disbelieve that there is a red patch, while still being a representational state which has a phenomenal character that is exhausted by the qualities of what it is an imagination of. The same applies to experiences of red afterimages.
Acknowledgments
Thanks to Katalin Farkas, Howard Robinson, Alex Byrne, Amy Kind, Gianfranco Soldati, Martine Nida-Rümelin, and an anonymous reviewer for helpful comments on earlier versions of this paper.
REFERENCES
Ali, R. 2018. Does hallucinating involve perceiving? Philosophical Studies 175: 601-627.
Aydede, M. 2009. Pain. In Stanford Encyclopedia of Philosophy (Winter 2012 Edition), ed. E. N. Zalta, http://plato.stanford.edu/entries/pain/ Accessed 16 May 2020.
Ben-Yami, H. Unpublished..Afterimages and related phenomena.
Block, N. J. 1996. Mental paint and mental latex. In Philosophical Issues, vol. 7: Perception, ed. E. Villanueva, 18-49. Atascadero: Calif.: Ridgeview.
Block, N. J. 2010. Attention and mental paint. In Philosophical Issues, 20: Philosophy of Mind, eds. E. Sosa and E. Villanueva, 23-65. Oxford: Wiley-Blackwell.
Boghossian, P., and D. Velleman. 1989. Color as secondary quality. Mind 98: 81-103.
Bourget, D. 2015. Representationalism, perceptual distortion and the limits of phenomenal concepts. Canadian Journal of Philosophy 45: 16-36.
Byrne, A. 2001. Intentionalism defended. Philosophical Review 110: 199-240.
Chalmers, D. J. 2004. The representational character of experience. In The Future for Philosophy, ed. B. Leiter, 153-181. Oxford: Oxford University Press.
Crane, T. 1988a. The waterfall illusion. Analysis 48: 142-147.
Crane, T. 1988b. Concepts in perception. Analysis 48: 150-153.
Davis, F. A., D. Bergen, C. Schauf, I. McDonald, and W. Deutsch. 1976. Movement phosphenes in optic neuritis: a new clinical sign. Neurology, 26: 1100–1104.
Gendler, T. 2008. Alief and belief. Journal of Philosophy 105: 634-663.
Gow, L. 2019. Everything is clear: All perceptual experiences are transparent. European Journal of Philosophy 27: 412–425.
Jackson, F. 1977. Perception: A Representative Theory. Cambridge: Cambridge University Press.
Kind, A. 2008. How to believe in qualia. In The Case for Qualia, ed. E. Wright, 285-298. Cambridge MA: The MIT Press.
Lycan, W. 1987. Consciousness. Cambridge MA: The MIT Press.
Mellor, D. H. 1988. Crane’s waterfall illusion. Analysis 48: 147-150.
Moore, G. E. 1939. Proof of an external world. Proceedings of the British Academy 25: 273–300.
Peacocke, C. 1983. Sense and Content. Oxford: Oxford University Press.
Phillips, I. 2013. Afterimages and sensation. Philosophy and Phenomenological Research. 87: 417-453.
Salari, V., F. Scholkmann, R. L. Pandey Vimal, N. Császár, M. Aslani, and I. Bókkon. 2017. Phosphenes, retinal discrete dark noise, negative afterimages and retinogeniculate projections: A new explanatory framework based on endogenous ocular luminescence. Progress in Retinal and Eye Research 60: 101–119.
Siegel, S. 2010. Do visual experiences have contents?. In Perceiving the World, ed. B. Nanay, 333-638. Oxford: Oxford University Press.
Schroer, R. 2004. Environmental representationalists on afterimages and phosphenes: Putting our best foot forward. Southern Journal of Philosophy 42: 531-546.
Thau, M. 2002. Consciousness and Cognition. Oxford: Oxford University Press.
Tye, M. 2000. Consciousness, Color, and Content. Cambridge MA: The MIT Press.
Tye, M. 2003. Blurry images, double vision, and other oddities: New problems for representationalism?. In Consciousness: New Philosophical Perspectives, eds. Q. Smith and A. Jokic, 7-32. Oxford: Oxford University Press.
Endnotes
- In philosophical literature, the term “appears” and its cognates have also taken on a separate sense which is distinct from the sense expressed by the term’s use in the context of an appearance-reality distinction. Such uses of the term do not necessarily imply the experiential state’s having truth-evaluable content, rather, the term is used to capture “phenomenal character”. Throughout this paper, I use “appear” and similar terms in the sense that implies a truth-evaluable state, as in “that box appears white to Agnes” (“X appears to have the property Y to subject Z”). Perhaps there is a coherent understanding of phenomenal character distinct from content, even though I doubt that there is, but in any case, my preference regarding the usage of the term confined to its narrow sense shall be taken as a mere terminological preference rather than a theoretical attitude towards how the term should be used. But for an argument that phenomenal character logically implies content, see Siegel (2010). ↑
- For seminal examples of representationalism, see Byrne (2001), Thau (2002), and Tye (2000). Representationalists of the “impure” sort (see Chalmers 2004 for the distinction) hold that the qualitativeness of a phenomenal experience is not only determined by the content of a representation but also by its “mode” or “manner”. The final verdict on pure-versus-impure representationalism has no bearing on the debate dealt within this paper. ↑
- In the times before there was a debate between qualia theory and representationalism, these revealed properties would be classified as “sensations” or “properties of a sensory field” (Peacocke 1983) or “sense-data” (Moore 1939; Jackson 1977). ↑
- It has been suggested by an anonymous reviewer that “mental qualities that are concretely instantiated while undergoing an experience” is too weak to be rejected by representationalists. This would be the case if this characterization mentioned the instantiation of “properties” rather than “qualities”. According to representationalism, experience instantiates properties for sure, such as the property of representing something. The term “quality” is chosen here particularly to narrow down the range of properties instantiated in an experience that are acceptable by the representationalist, as non-representationalist theories rely on the actual instantiations of a quality to explain the qualitative character of experiences, such as qualia, or qualities of sense-data, or qualities of the objects in the actual environment. According to representationalism, the mere appearance of such qualities, without any actual quality concretely existing, explains the qualitativeness of experience. ↑
- According to this second meaning of “qualia”, qualia exist according to all theories of experience such as representationalism, direct realism and sense-data theory, though in such cases the term might constitute a case of terminological inflation, as talking of “experienced qualities” would most probably suffice. ↑
- Also see the related discussion in Schroer (2004). ↑
- These intentional objects might even be qualities instantiated in an experience. So even if qualia theorists were right in analyzing experiences as states where we seem to find qualities instantiated in experience, there would still be some theoretical space to say that experiences are representational states, where our experience represent an experience as having a certain quality (see Thau 2002 for a similar argument). And it would require a separate argument for the qualia theorist to demonstrate that we should reject the representationalist analysis of such states in favor of an analysis that relies on concretely instantiated mental qualities that constitute such experiences. ↑
- See Phillips for a list of features of afterimages which, according to anti-representationalists, makes afterimages “incompatible with […] being apparent presentations of public objects” (2013, 418). ↑
- The concept of a non-endorsed representation has some parallels with the concept of an alief (Gendler 2008), but aims to capture a phenomenon at the experiential level. ↑
- Of course, a subject who is not a philosopher or a cognitive scientist by profession does not think about her experiences via technical terms like “representation”, “content”, or “veridicality”, but can nevertheless think of appearances and think about whether things are as they appear or not. ↑
- Imagination, of course, can have some element of endorsement, when, for instance, a subject imagines something to remember how it looks, and a painting, similarly, can be made for the purposes of inducing beliefs about how something looks, but these are not the necessary features of imaginings or paintings. ↑
- The argument known as “the argument from illusion”, a species of “argument from error”, can rely on the Müller-Lyer illusion, or any other illusion or hallucination, where the property encountered in experience doesn’t exist in the environment, and is therefore allegedly instantiated in the mind. That debate is outside the scope of this paper, and it is often dealt with a separate counter-argument which relies on considerations related to the intentionality of mental states. ↑
- Here, I use “weird” not as a technical term that is supposed to capture a clearly defined common feature of these experiences, but merely as a descriptive prelude to the examples that follow. ↑
- For an early discussion of experiential phenomena that might represent impossibilities, see Crane (1988a; 1988b) and Mellor (1988). ↑
- Also see Tye (2003) for a counter-argument to Boghossian and Velleman. ↑
- Bourget (2015) suggests that the difficulty we might have with very clearly articulating the contents of such experiences might be due to the limitations of our concepts relating to their phenomenology, and why we have these limitations might be due to the fact that we are not interested in communicating them, as the representational contents that make these experiences the type of distorted experiences they are do not correspond to anything in the actual world. ↑
- Indeed, Lycan is probably wrong in suggesting that a veridical equivalent of these experiences can be produced in a movie theater, because that perceived scene would be in front of the subject and experienced as such, while some of these experiences represent their objects as being somewhat located inside the eyes of the subject, while at the same time representing properties we see with our eyes. ↑
- An auditory analogue of these visual experiences might be tinnitus, or “ringing-in-the-ear”, though it is a somewhat weaker case of automatically non-endorsed representation, as they are more commonly endorsed, especially if the tinnitus experience is a novel experience for the subject and she genuinely believes that there is a source of sound somewhere inside her ear. Some tinnitus experiences, after all, are veridical experiences of a sound produced in the ear. ↑
- Several philosophers have recently suggested that some phosphene experiences are instances of veridical perception. Based on empirical work by Salari et al. (2017), among others, Ben-Yami (unpublished) suggests that in some cases “the phosphenes perceived are due to ultraweak bioluminescent photon emission of cells in the visual system”, and Ali (2018) makes a similar claim that some phosphenes are generated internally, basing this claim on Davis et al. (1976). Perhaps these findings could also be interpreted as the electromagnetism produced by the eye causing illusory experiences rather than us veridically perceiving electromagnetism produced by the eye, but I will not discuss this here, for the examples discussed by the above authors involve only some types of phosphenes (such as those experienced in a dark room), and if it is indeed true that we veridically experience something in these experiences, this can be accommodated within a representationalist framework: They are veridical representations. The qualia theorist, on the other hand, should provide a strong argument to the effect that the veridical experience claim is confused: That the scientific evidence cannot be interpreted in such a way because, according to the qualia theory, we cannot make sense of phosphene experiences being veridical, as phenomenological reflection allegedly shows that these experiences lack intentional objects or representational contents. ↑
- An anonymous reviewer has suggested that even if the endorsement/non-endorsement framework is on the right track, there is still the possibility that the difference between endorsement and non-endorsement is due to a change in phenomenal character without a change in content. This, however, is not possible, given that the notions of endorsement and non-endorsement are, by stipulation, cognitive notions related to the formation of beliefs or belief-like states. Perhaps my analyses in this paper are wrong and there is nothing in the mental realm that can be captured by the notions of endorsement and non-endorsement, but if there are, then by definition endorsement and non-endorsement refer to content-related differences. At best, there might be a phenomenal difference between endorsement and non-endorsement, alongside representational and functional differences, if there is a phenomenology specific to various cognitive states. The question of whether that phenomenological difference can be captured by representational terms or not would be a general question regarding the cognitive phenomenology of all experiences, because all experiences are either endorsed or non-endorsed, and the question would be out of the scope of this paper, as this paper takes up issue with the claim that particularly afterimage experiences and the like are resistant to a representationalist analysis for reasons specific to these type of experiences. ↑
- The phenomenon of blurry vision is outside the scope of this paper, but I will just mention in passing that the information-degrading nature of experiences of blur should be pretty obvious even without considering extreme cases. It is part of the daily visual experience of many people with eye problems, and a rather undesirable one. ↑
- This definition of qualia or mental paint as non-representational might seem to go against the definitions of some qualia theorists that define these properties as representational, in the sense that these properties are vehicles of representation, as when Block says mental paint is the “mental properties of the experience that represent the redness of a tomato” (Block 1996, 29). Mental paint can be a “representational” property in the sense that it can serve representational functions, though it can also be instantiated without representing anything. When I define these properties as “non-representational”, what I mean is that these properties are neither represented properties, nor the property-of-representing-something. If there is no mental paint, introspection can only reveal which property is represented, along with the fact that the subject is in a representational state (and perhaps, if impure representationalism is true, introspection can reveal some other properties like representational “modes”), but it does not reveal properties that do the representation in the way that one looks at a painting and becomes aware of the paint. Block claims that what is found by introspection in afterimage experiences is neither the represented property, nor the property of something being represented, but only mental paint, and in this sense mental paint is non-representational. Contrary to Block, I claim that analyzing an afterimage experience gives us only representational properties: Introspection reveals the fact that one is undergoing a representational state along with the content of this representation, but no mental paint. ↑
- An anonymous reviewer has suggested that this is possibly the correct interpretation, as Block describes afterimages (more particularly, an afterimage’s changing size) as “somehow unreal or unobjective” in his more recent work (2010, 54). This suggestion might be on the right track, but for the purposes of this paper, it is useful to explore all the possible interpretations of Block’s original remark, and follow where they theoretically go. ↑
- Here I don’t mean to say that your visual system is an independently conscious homunculus. It is just that in cases of non-endorsement you distance yourself from how things look to you as represented by your visual system, and endorse your cognitive, judgmental representations about the environment. ↑
- By “caring about the very existence of these experiential states rather than their veridicality” I do not mean that we care about some non-representational or non-intentional aspects of these states. We care about pain experiences because of the object of experience, the felt pain quality. ↑
