TWOFOLDNESS AND THREE-LAYEREDNESS IN PICTORIAL REPRESENTATION

paintings have only representational content, while figurative paintings have both.46 In my account, however, what I call figurative content is the only kind of content that corresponds to the item that one sees in a picture, while what I have called pictorial content is instead not ‘seen in’ at all. Rather, it structures what the picture is about. Thus, these contents respectively contribute to articulate what in a Husserlian framework is the second and the third pictorial layer. Twofoldness and Three-Layeredness in Pictorial Representation 102 Estetika: The Central European Journal of Aesthetics, LV/XI, 2018, No. 1, 00–00 45 See Wollheim, Art and Its Objects, 206. 46 See Richard Wollheim, ‘On Formalism and Pictorial Organization’, Journal of Aesthetics and Art Criticism 59 (2001): 131. Zlom1_2018_Sestava 1 23.3.18 11:39 Stránka 102


I. INTRODUCTION
A neo-Husserlian way of attacking Richard Wollheim's account of pictorial experience as a sui generis twofold perception says that picture perception is not two-, but threefold. For in it, one is perceptually linked not only and directly both to (1) the picture's vehicle and (2)  In this essay, I defend a Wollheimian account of a twofold picture perception.
While I agree with Wollheim's objectors that a picture involves the three layers that qualify a picture in its complexity -its vehicle, what is seen in it, and its subject -, I argue that the third layer does not involve perception, even indirectly: what is seen in a picture constrains its subject (that is, what a picture is about) to be a subject of a certain kind, yet it does not force the latter to be pictorially perceived, not even indirectly. So, even if a picture is three-layered, pictorial experience remains a twofold experience, as Wollheim claimed. Neither the proponents of threefoldness nor Wollheim himself, however, have convincingly explained how the experience really is a perceptual experience.
My Wollheimian account thus aims to reconceive the pictorial experience in properly perceptual terms.

II. A STANDARD PROBLEM WITH SEEING-IN AND THE THREEFOLDNESS ACCOUNT
As is well known, for Wollheim seeing-in is the distinctive pictorial experience, once it is conceived as a sui generis twofold perceptual experience that, typically but not exclusively, suitable spectators of pictures entertain. 1 Seeing-in is indeed made of two folds, the configurational fold (CF) in which the picture's vehicle (the picture's physical basis) is grasped, and the recognitional fold (RF), in which what the picture presents, which Wollheim originally tended to equate with the picture's subject, that is, what the picture is about (erroneously, as we will see), is grasped. According to Wollheim, first of all, such folds are inseparable, for neither is identical with the corresponding experience taken in isolation. Seeing the picture's vehicle qua picture's vehicle is not the same as having a perceptual experience of it qua mere physical object in the world, while grasping that what the picture presents is not the same as perceiving that thing face-to-face. 2 Moreover, one may say that the second fold depends on the first fold, at least in the sense that the former could not exist if the latter did not exist as well. 3 A standard criticism of Wollheim's thesis says that it is unclear how the RF can have a perceptual character, hence how the whole seeing-in experience can be a perceptual experience. True enough, the CF may be taken as perceptual insofar as it is an experience of something, the picture's vehicle, which is over there, that is, at a certain (though approximate) distance from the picture perceiver. Yet, typically, what the picture presents is not where the picture's vehicle is. Granted, sometimes that what the picture presents is in the actual world just as the picture's vehicle (as is surely the case with portraits and photographs), but typically it is not in a part of the world that is available to the perceiver's view.
So, while one can perceive the picture's vehicle, how can one perceive what the picture presents? 4

Twofoldness and Three-Layeredness in Pictorial Representation
Here defenders of the idea that pictorial experience is threefold rather than twofold arrive on the scene. For, they say, in order to solve this problem, one must first introduce an intermediate item between the picture's vehicle and the picture's subject. The latter is indeed what the picture is about but it is not what the picture presents, which turns out to be an intermediate item between the picture's vehicle and the picture's subject. The picture's experiencer properly perceives this item over and above the picture's vehicle. Granted, what this intermediate item amounts to is spelled out differently by the different threefoldness theorists. Following Husserl (2006), some tend to label this item the image-object. 5 In conformity with Solveig Aasen, some others may characterize it as a universal. 6 Yet in conformity with Robert Briscoe, some others may conceive it as a virtual model. 7 Finally, Bence Nanay holds that it is a visually encoded object. 8 However different such conceptions are, metaphysically speaking, this intermediate item plays exactly the same role for all -namely, it is what is present to the perceiver's mind over and above the picture's vehicle. Thus, one can perceive the intermediate item just as the vehicle, that is, directly.
To be sure, Edmund Husserl limited himself to stating that what now amounts to the picture's subject is just a third layer of the picture, over and above the picture's vehicle and the intermediate item. 9 Thus, for Husserl a picture is three-layered, in the sense that those three elements -the vehicle, the intermediate item, the subject -must be added one to another in order to understand what a picture overall amounts to. 10  different folds, but so too does the picture's subject itself involve (typically at least) 11 a third -indirect -perceptual fold in that experience. It is given either through seeing-in again -so just as the intermediate item is seen in the vehicle, the subject is seen in the intermediate item 12 -or through mental imagery, 13 or at least through a combination of such factors. 14 To my mind, however, as to the perception of the picture's first two layers, proponents of threefoldness hardly account for the problem Wollheim himself unsuccessfully addressed -namely, how seeing-in can be a properly perceptual experience, as we will see in Section III. In Section IV, I will offer an alternative account that is able to solve this problem. This account accepts the idea that a picture involves three layers: the picture's vehicle, what is seen in it, and what the picture is about. Yet the account not only reconceives of how we have a properly perceptual experience of the first two layers, it also, as we will see in Section V, claims that, pace proponents of threefoldness, a third fold for the pictorial perception is not needed. For the third layer of picture perception does not involve perception, even indirectly. Consequently, this account may still be taken to be Wollheimian. In Section VI, before concluding the essay, I consider and ultimately reject some objections to this account.

III. THE PROBLEMS OF THE THREEFOLDNESS ACCOUNT
The threefoldness account faces two connected problems. First, the 'relation' problem: how are the picture's vehicle and the intermediate item related, in the sense of how does the latter emerge out of the former? Second, the 'presentation' problem: how can that item be present to the picture experiencer, so that, in virtue of that presentness, her or his experience, like any other perceptual experience, counts as a pictorial perception, as the defender of that account believes together with Wollheim?
As we have seen, the 'relation' problem basically consists in wondering how the intermediate item, which in any of its metaphysical interpretations is not a physical thing out there, emerges from perceiving the picture's vehicle. What makes it the case that, in virtue of perceiving the picture's vehicle, we also somehow apprehend the intermediate item?
Let us now consider the extant solutions of the first problem on behalf of defenders of threefoldness. First, a proponent of threefoldness may endorse Yet appleness is not instantiated by that picture's vehicle: no part of it is an apple.
In the vehicle, universals of colours and shapes are instantiated, insofar as it contains, say, a red round spot.
Third, a proponent of threefoldness may exploit Briscoe's claim that the perception of the vehicle merely causes not only the imagining of the intermediate item, but also a perception of it. 20 Yet one is then implausibly forced to allow for hallucinatory pictures, that is, to take as pictorial perceptions cases in which, by looking at a thing, one hallucinates something else. For instance, by looking at a canvas, one may hallucinate an apple; yet it does not seem that the canvas becomes a picture of an apple, or at least acquires that figurative value, in virtue of such a hallucination. 21 the fact that the perception of the vehicle suitably changes allows for something else to be discerned in it.
The 'presentation' problem is related to the 'relation' one. In order for something to be present in a perceptually relevant sense, the property of being out there, out-there-ness in short, must be attributed to it with respect to the perceiver. In the experience of virtual depth and 3D structure in the right way' , that is, by being 'systematically guided by sources of optical information in the light reflected (or emitted) by S to the eye' (Briscoe,'Depiction,Pictorial Experience' ,63). Yet even adding this constraint does not seem to provide a sufficient condition for something to be a picture. For -as Nelson Goodman originally suggested -a twin of X may cause the experience of virtual depth and 3D structure in the right way and still is not a picture of X. See Nelson Goodman, Languages of Art: An Approach to a Theory of Symbols (Indianapolis: Bobbs-Merrill, 1968). 22 Nanay, Aesthetics; 'Threefoldness' . 23 '[W]hat makes pictures of this kind special is that before you get to see the dog, you do not see these illusory contours -you see them only once you see the dog in the picture' (Nanay,Aesthetics,53). '[T]he perceptual phenomenology still changes as a result of seeing something in the picture […] the dog is part of our perceptual phenomenology and it is the perceptual experience of the Dalmatian that makes it possible for us to perceptually experience the illusory contours.
[…] If this argument shows that the "dog" […] is perceived, this is to be understood as claims […] about the three-dimensional object visually encoded in the surface: the argument shows that we perceive the three-dimensional object visually encoded in the surface' (ibid., 54). 24 It is fair to say that even when one sees a star by focusing on a certain dot in the vault of the sky, one sees the star to be present, although it is disputable whether one feels it as present. See  if the item is a universal, 28 then, whatever one's conception of universals, outthere-ness is not attributed to it. If one has a Platonic conception, universals are not out there, but are beyond out-there-ness. If one has an Aristotelian conception, there certainly are universals that are instantiated out there but, as we have seen, they are not the relevant universals (when we have a pictorial perception of an apple, appleness is not even instantiated out there). Thus, a universal cannot be present in the relevant perceptual sense.
Third, if the item in question is hallucinated rather than imagined, a possibility compatible with Briscoe's idea of a virtual model, 29 out-there-ness may be attributed to it, hence it counts as present in the relevant sense. Yet again, even perception-based hallucinations, or other experiences of non-pictures, would implausibly count as picture perceptions. Fourth, if the intermediate item is visually encoded, 30 as we have seen, out-there-ness is even more scarcely ascribed to it.
It thus seems that no actual solution to these two problems, the 'relation' problem and the 'presentation' problem, which the threefoldness account may exploit, is satisfying. Consequently, the idea that the threefoldness account has managed to provide a solution to the original problem from which we started,  27 Eldridge, 'Depicting and Seeing-In' . 28 Aasen, 'Pictures, Presence and Visibility' . 29 Briscoe, 'Depiction, Pictorial Experience' . 30 Nanay, Aesthetics; 'Threefoldness' .
pace Wollheim -how can the RF, hence seeing-in as a whole, have a perceptual character -seems ungrounded. But there is an alternative solution to the original problem. This account is also able to solve the two aforementioned related problems that proponents of threefoldness have unsuccessfully faced. Yet this account remains in the (general) spirit of Wollheim's account, as we will see.

IV. AN ALTERNATIVE SOLUTION
The 'relation' problem consists in how something can emerge out of the perception of the picture's vehicle, with the hypothesis that this something is an intermediate item lying between the picture's vehicle and the picture's subject, where the latter now means what the picture is about.
Here is a new solution to this problem, which conforms with Wollheim's twofoldness account of picture perception. Suppose, first of all, that one again splits, à la Husserl, what the picture presents from what the picture is about, which is now assumed to be the picture's subject, and therefore takes what the picture presents to be an intermediate item between the picture's vehicle and what the picture is about, that is, the picture's subject. Yet one may then again take picture perception to be, à la Wollheim, a twofold experience that is respectively addressed to the picture's vehicle in its CF and to the intermediate item in its RF, once this is no longer equated with the picture's subject, as Wollheim tended to assume.
Moreover, suppose also that the perception of the picture's vehicle in the CF has a content that is enriched with respect to the content that it has when the vehicle is perceived in isolation, as a mere physical object among many others.
This enriched content enables the CF to be properly related to the RF: in virtue of grasping such an enrichment in the CF, the RF arises with its own emerging content, which is exactly the intermediate item we are looking for.
How does this enrichment enable that emergence to obtain? To begin with, in the CF, one grasps not only the vehicle's colours and forms, but also its grouping properties, the properties for its elements of being ordered in a certain direction in a dimension of the publicly available three-dimensional space. 31 Plausibly enough, such properties are perceptual. To be sure, they are perceptually higher- Rather, in that RF she or he sees that item as emerging out of the patches in virtue of their having been opportunely grouped in the CF. 35 To be sure, thinking of  35 Lopes makes a similar mistake while denying that grouping properties are genuine design properties of the vehicle, that is, properties that are responsible for the fact that a certain item is 'seen in' it. See Dominic McIver Lopes, Sight and Sensibility: Evaluating Pictures (Oxford: Oxford University Press, 2005), 41-44. As I have pointed out elsewhere, grouping properties are special design properties for they are responsible both for that fact and for the emergence of that item. See Voltolini, Syncretistic Theory. the intermediate item may enable one to prompt that grouping operation, a sort of weak cognitive penetration in Macpherson's terms: 36 such thoughts limit themselves to penetrate the phenomenal character that qualifies the CF, but not its content. Yet it is not perceiving that item that induces such an operation, for things go precisely the other way around. Now, 3D groupings are direction-oriented groupings just as 2D groupings, those that are involved, for example, in the Gestalt switch affecting the Mach figure: all such groupings involve a movement towards dimensional 'poles' . Yet, unlike the dimensions that 2D groupings involve, the fundamental dimension on which groupings in pictures operate, depth, is projected upon the vehicle, which is basically a 2D entity. 37 As a matter of fact, this allows one to talk of the vehicle's depth as apparent or illusory. 38 This feature of illusiveness is very important, for it enables one to see how this new solution to the first problem that the threefoldness account raises may allow one also, in the context of Wollheim's twofoldness account of picture perception, to solve the second problem of the threefoldness account, the 'presentation' problem. As we have seen, this second problem consists in how the intermediate item can be present in a perceptually relevant sense involving the ascription of out-there-ness to that item.
To begin with, the illusion of depth that occurs in the CF allows the RF to occur as well as a sort of illusory perception, the perception of the picture's vehicle as the picture's vehicle so modified, see the picture's vehicle as a Dalmatian, without seeing that the vehicle is a Dalmatian, for of course there is no such fact.
Moreover, this illusory perception is a special kind of non-delusory illusion. For it is an illusory perception known as such, since the picture perceiver knows that the picture's vehicle is not the picture's subject. To be sure, many standard perceptual illusions are non-delusory forms of illusions, that is, illusions that are known as such. Consider, for example, the well-known Müller-Lyer illusion, in which one sees a certain segment as being longer than another, even though one knows that they are of the same length. Yet, unlike such cases, the picture perceiver knows that the RF is an illusory perception because she or he simultaneously knowingly perceives, in the same sensory modality (typically the visual one), 40 the picture's vehicle in the CF.
Armed with these reflections, the new solution can now positively address the 'presentation' problem. By being an illusory perception of an item different from the picture's vehicle, out-there-ness is attributed to the 'seen-in' item in the RF's content, just as in a veridical face-to-face perception of it: that item is represented to be out there. Thus, presentness in the relevant perceptual sense is attributed to that item. The attribution is, however, simply erroneous, for the item is not out there; the vehicle is there. True enough, that attribution of presentness in the RF's content is a specific one, for it is not accompanied by the feeling of presence that surrounds ordinary perceptual experiences. In a nutshell, unlike ordinary perceptual experiences, that illusory perception has no 'thetic' mode. 41 Indeed, the fact that such an attribution is known as being erroneous because the picture perceiver already knowingly perceives the picture's vehicle in the CF, means that in the RF that item is not felt as being out there. In sum, being (falsely) perceived as being out there accounts for the 'seen-in' item's being present in a perceptually relevant sense. Yet not being felt to be out there is what accounts for the peculiarity of such a presentness. 42 Once things are stated in this way, the original problem for Wollheim, which prompted proponents of threefoldness to articulate their account against the Wollheimian account, can be solved as well. As we have seen, this is the problem of how an experience of seeing-in can really be a perceptual experience, by letting its two folds genuinely count as having a perceptual character. According to the new solution that I have provided, seeing-in is a properly perceptual experience. Not only its CF, but also its RF, is such. For although the latter, unlike the former, is not a veridical perception, it is still an illusory perception of the picture's vehicle as what it is seen in the picture, a given intermediate item lying between the picture's vehicle and what the picture is about, its subject. To be sure, some people question the notion that illusory perceptions are metaphysically the same kind of mental states as veridical perceptions, 43 yet nobody seriously doubts that both are perceptual experiences.

Proponents of threefoldness also claim that, although the intermediate item is
what is properly perceived in picture perception, there is a sense, albeit indirect, in which the third pictorial layer, the picture's subject, is also perceived in that experience, thereby making picture perception threefold rather than twofold. Yet, once this new solution is available, it can be further meant as a three-layered account of what a picture is, which, however, remains Wollheimian, since it still holds that picture perception is twofold. For it does not appeal, even indirectly, to perception with respect to the third layer.
To begin with, this account comports with that of proponents of threefoldness -namely, that a picture involves not only (i) a vehicle and (ii) an intermediate item seen in it, but also (iii) a picture's subject, conceived of as what that picture is about. Indeed, we definitely need a distinction not only between (i) and (ii), but also between (ii) and (iii), because the second and the third layer clearly differ.
For one and the same picture may be, or is indeed, about different things, even if what is seen in it remains the same. As we shall soon see, there are indeed cases in which one goes on seeing the very same item in one and same picture, even if (possibly or even actually) the picture's subject is each time different: the picture is, or may be, now about a certain particular, now about another. Pace Husserl, however, this constancy in pictorial perception with a change in the picture's subject also occurs in pictures whose subject is general, not particular: while still seeing the same intermediate item in a certain picture, that picture may be either about some F or about some G. Thus, the distinction between (ii) and (iii) is not one between a general item and a particular subject, but between a general item and a subject that is either particular or merely less general than that item. about Leonardo da Vinci, who according to many was the model that Raphael had in mind while painting his masterpiece, even if one still sees in it the same long-bearded old man. Second, one sees in a certain snapshot of Woody Allen's Zelig (1983) a bespectacled middle-aged man. When watching the movie, the snapshot is clearly of Zelig, the main character of the movie. Yet while seeing that snapshot outside the movie, the shot counts as a picture of Allen himself, even if one still sees in it the same bespectacled middle-aged man. 44 In both examples, the picture's subjects that match the same intermediate item are particular individuals. Yet there may also be cases in which those subjects are simply less general items than the intermediate item. For example, when looking at a drawing of a certain elegant female silhouette, that drawing may count as a picture either of a businesswoman or of some female professor; the drawing therefore has as its subject not a particular individual, but an item that is simply less general than the one seen in it.
One might wonder why for pictorial purposes a third layer is needed over and above the second. The answer to this concern, however, is, I contend, rather obvious. While the intermediate item accounts for the picture's figurativity, that is, its being a depictive representation, the picture's subject accounts for the picture's intentionality, that is, its being a representation, by being what the picture is of. To be sure, the second and the third layer are not independent of each other, because the former constrains the latter. What a picture is about -let's now call it the pictorial content of a picture since it accounts for its intentionality -may change, yet not indefinitely. Whatever the picture is about must comport with the features that characterize what is seen in the picture. Let's call it the figurative content, since it accounts for its figurativity.
All the above examples demonstrate this point. Both Plato and Leonardo are long-bearded old men; both Zelig and Allen are bespectacled middle-aged men; both the businesswoman, whoever she is, and the professor, whoever she is, share the same elegant female silhouette. Indeed, if a picture were about something that did not respect the constraint its figurative content poses, allowing its intentionality to float free from its figurativity, it would not be about that something as a picture, but simply as any representative sign. To see this point, consider the difference between an ordinary picture of Jesus and the fish-like For this example, see Wiesing,Artificial Presence,[46][47] logo early Christians used to mean Jesus. The former is a picture of Jesus, because in it one sees an adult man. Yet the latter is just a symbol of Jesus, for in it one sees a fish, something whose features Jesus does not obviously share.
I wish to reiterate that distinguishing the third pictorial layer from the second sounds rather un-Wollheimian. This clearly stems from the fact that cases like those we have sketched out above were treated differently by Wollheim. For instance, he said that a picture of Henry VIII which someone may take as a picture of the British actor Charles Laughton prompts a different distinction between different experiences of seeing-in: a correct experience -which provides a picture with its right subject, that is to say, what the picture is about, Henry VIII in this case -and an incorrect experience -which provides a picture with its wrong subject, the actor in this case. 45

Twofoldness and Three-Layeredness in Pictorial Representation
However, although Wollheim's distinction is not exactly the same as the distinction between the second and the third pictorial level that I have appealed to here, the present account, concerning picture perception, remains broadly Wollheimian. For it holds that, pace proponents of threefoldness, the third layer involves no appeal, even indirect, to perception, although it is constrained by the second layer in the aforementioned way. In other words, even if a picture is threelayered, its perception remains twofold, as Wollheim claimed.
This point becomes clear if we consider the difference between a merely In order for a perceptual change to occur when a change in aboutness occurs as well, one must pass from an MRA picture to a PA picture, a picture affected by a Gestalt switch: for example, the famous duck-rabbit picture, the Rubin vase picture, the young girl-old woman picture, and so on. Yet note, first, that in all such cases the perceptual difference at stake primarily affects the CFs of the relevant seeing-in experiences, because of the difference in the grouping operations that concern what remains the very same vehicle with respect to its colours and shapes. In the duck-rabbit picture, for instance, the Gestalt switch induces two different seeing-in experiences, primarily because in the respective CFs of such experiences one grasps either (by means of a certain grouping operation) a certain duck-like arrangement (in front of a background), or (by means of another grouping operation) a certain rabbit-like arrangement (in front of a background). Moreover, once that perceptually relevant shift in the CFs of such experiences occurs, then another perceptually relevant shift in their RFs occurs as well. For instance, in the duck-rabbit picture, one sees either an anatid or a leporid. Yet, beyond the switch in what is 'seen in' in those RFs, no other perceptual factor is involved in the third pictorial layer. After one fixes what is seen in the PA picture in a certain way, then a certain intermediate item (an anatid, for example) is seen in the RF of the picture perception as providing a figurative content for it. One may further take that PA picture as being about a certain subject (a duck), or about another subject (a goose), without any further perceptually relevant change occurring. After one fixes what is seen in the PA picture in another way, then another intermediate item (a leporid, for example) is seen in the different RF of the picture perception as providing a different figurative content for the PA picture. One may further take that picture as being either about a certain subject (a rabbit) or about another subject (a hare) for that matter, without any further perceptually relevant change occurring.
If this is the case, proponents of threefoldness are wrong in claiming that the third pictorial level involves a further, however indirect, perceptual fold in picture perception. According to some of them, one sees the third layer, the picture's subject, in the second layer, the intermediate item. 47 Yet, as we saw before, seeing-in occurs only when the second layer is perceptually mobilized in virtue of the way the first level is so mobilized by means of a certain grouping operation. Passing to the third layer requires no further grouping operation, hence no further perceptually relevant process affects the picture's subject by involving the intermediate item. Hence, speaking of seeing the picture's subject in the intermediate item is plainly metaphorical. According to Nanay, mental imagery is typically involved in passing from the second to the third layer. 48 Now, first of all, a pictorial perception may certainly prompt a mental image of the picture's subject. Yet even if mental imagery has a quasi-perceptual nature, as Nanay holds, this merely causal way for that subject to be involved does not force it to be perceptually apprehended, even indirectly, in that pictorial perception.
Certainly, moreover, mentally imagining that a picture has a certain subject -in particular, so imagining that an MRA picture has one or another subject -has an import as to the overall phenomenal character of the pictorial experience involved.
In this respect, one of the two subjects of such a picture may, for example, seem more familiar than the other. As Nanay himself has emphasized, this may also happen in a case of Dantian indistinguishability between a picture and its copy,  In 'Depicting and Seeing-In' , Eldridge combines both proposals, by saying that in the third fold of picture perception both the consciousness of the picture's subject is quasi-perceptual and we see the subject in the intermediate item. Obviously, that combination fares no better, for it inherits the problems of each proposal. With reference to Husserl, Phantasy, Image Consciousness, he also adds that the intermediate item exhibits the subject. Hopp claims that the (experienced) resemblance between VI. SOME OBJECTIONS AND REPLIES To be sure, Nanay puts forward some arguments in favour of the idea that grasping the picture's subject amounts to a quasi-perceptual fold of the pictorial perception. His main argument centres on the fact that there is a perceptual change when one recognizes that a certain picture is a picture of a certain subject.
This typically happens with caricatures. As soon as one recognizes that a certain caricature is a caricature, say, of Mick Jagger, one's experience of that caricature definitely changes. Moreover, that change affects the perceptual phenomenology of that experience, insofar as it amounts to the quasi-perceptual recognition of This kind of perceptual change obtains not only in pictorial, but also in ordinary perception, as some maintain; see René Jagnow, 'Can We See Natural Kind Properties?' , Epistemology and Philosophy of Science 14 (2015): 183-203. By looking at a pine, one refines one's perception insofar as one's recognitional abilities are sophisticated enough to grasp a certain pine-Gestalt. Yet no further perceptual change would occur if what one is facing were not a pine, but a Twin Earth pine, that is, a different natural kind yet sharing its observational properties with a pine. One would still merely perceive the same Gestalt.
In the example of Jagger's caricature, the recognitional experience that perceptually sophisticates what is seen in that caricature would remain exactly the same even if it were a caricature of Mick's twin rather than of Mick himself.
Clearly, knowing that the picture is a caricature of Mick may prompt such a recognitional experience. Yet again, the perceptual content of that experience does not involve Mick, since that experience would remain the same even if it turned out that the picture is a caricature of Mick's twin.
Note that such a situation occurs in many other cases that have nothing to do with caricatures. We may make discoveries in pictures, especially in those that Once things are put in this way with regard to a recognitional change concerning a picture, one may similarly assess another argument Nanay provides in favour of threefoldness. For Nanay, aesthetic appreciation of a picture may also consist in attending to both the intermediate item and the picture's subject, or even their relationship. The best explanation of such attention again appeals to quasi-perception. See Nanay, Aesthetics, 58; 'Threefoldness' . Yet once recognitional changes are explained as involving a sophistication of the intermediate item but not the picture's subject, there is no need to explain the appreciation of that subject in quasi-perceptual terms. To return to the Dantian example that Nanay himself considers, one may appreciate the difference between a picture and its accidental copy by the fact that, unlike the latter, the former is a representation of a certain subject. Yet no quasi-perceptual experience of that subject must be involved in such an appreciation. 54 Husserl, Phantasy, Image Consciousness. Yet following Wollheim and Wittgenstein, I disagree with Nanay's interpretation of those cases. The Matisse painting is obviously of Mme Matisse; yet we do not see a green-striped alien in it; rather, the intermediate item we see in it is a flesh-coloured woman. 56 Green is perceptually captured in the apprehension of that picture's vehicle, in the CF of the relevant seeing-in experience. If that vehicle suffered a physical alteration that changed the hue of green that is painted on it, we would still see in it the very same flesh-coloured woman. This is even more evident with respect to black-and-white pictures. In a black-and-white photo of a boy, we do not see a black-and-white exotic kind of individual, but rather a normal flesh-coloured human being. 57 Black and white are perceptually captured in the apprehension of that photo's vehicle, in the CF of the relevant seeing-in experience. If due to a physical process the photo turned into a sepia one, we would still see in it the very same flesh-coloured human. One may also put it the other way round: if while watching a football match on TV our device suddenly became unable to transmit colours, we would still see in the match normally flesh-coloured human beings, not exotic black-and-white individuals.
Moreover, let me suppose with Nanay that there are people who do not recognize coloured individuals in black-and-white photos. I agree with Nanay that if those people finally happened to have such a recognition, there would be a perceptual change in their overall pictorial perceptions. Yet just as in the Jagger caricature, that change would primarily affect the CF of such experiences. For the CF of such experiences would become more sophisticated by grasping a new objective look for the figures those people group in that fold. Instead of grasping a similarity in appearance for those figures to paradigmatic looking-greyish objects, they would pass on grasping a similarity in appearance for those figures to paradigmatic looking-coloured objects. 58 As a result, the RF of such experiences would change as well: those people would pass from seeing in those photos black-and-white kinds of individuals to seeing coloured kinds of individuals in them. Yet once again, no perceptual or quasi-perceptual change would affect There is one last objection to my account, which I would assess. Suppose one were to apply to pictorial perception phenomenal disjunctivism, the doctrine that (in a minimal version of it) says that a mental state of veridical perception belongs to a different metaphysical kind than a (possibly phenomenally indistinguishable) state of hallucination. One might then say that not only veridical perceptions, but also pictorial perceptions put us in contact with what they are about, that is, their subjects. Could one then not say that seeing a picture as a picture of subject M (say, Mick Jagger) and seeing another, yet phenomenologically indistinguishable, picture as a picture of subject N (say, Mick's twin, Nick), or even the same picture as an MRA picture of N rather than of M, are perceptually different perceptions since they present different things as the pictures' subjects? 60 For argument's sake, let me concede to my possible, disjunctivist-inspired, objector both that presentation is a perceptual, not a merely consciential, relation, between a perceptual state and its object, and that presentations of different objects, even when they are phenomenologically indistinguishable, make a perceptual difference in the perceptions involved. Yet it remains true that the difference affecting the different pictures in question, or the two interpretations of an MRA picture, is a difference not in presentation, but in picture aboutness. To see this point more clearly, consider again an image by chance, that is, an object in which one sees something naturally, without any artefactual intention of letting one see anything in it. Now, from the perceptual point of view, nothing changes once a certain representational value is added to an image by chance, by becoming a picture of something. For example, one can see a face in a Martian rock. Ufologists say that it is a picture of a particular face, for that is how Martians depicted the rock. Believe them or not, nothing changes perceptually. The image still perceptually presents something -namely, what is seen in it. If you believe them, then an image-by-chance of a face becomes a picture of a given face. 61  61 Nanay also takes these examples into account, yet he acknowledges that they are cases of pictures involving just a twofold, not a threefold, perception. See Nanay, 'Threefoldness' .

VII. CONCLUSION
I think it is right to say that in order to analyse what a picture is, one must appeal to three layers: the picture's vehicle, what is seen in it, that is, the picture's figurative content, and the picture's subject, that is, what the picture is about, its pictorial content. Yet I also think that this three-layered account of pictures does not force one to endorse an analogous account of picture perception, which holds that such an experience is a threefold rather than a twofold perception. For, as Wollheim himself repeatedly stressed, its perceptual character, though distinctive, is exhausted by simultaneously and interlockedly grasping the first two pictorial layers, the picture's vehicle and what is seen in it. Simply, in order to solve the original problem that Wollheim left open, that is, of how that character may be properly perceptual, one must both properly reconsider what the characters of the CF and those of the RF respectively consist in and overall make pictorial perception a properly, though distinctive, perceptual experience. One must take the CF as a veridical perception that includes the vehicle's grouping properties and the RF as the knowingly illusory perception of that vehicle as what is seen in it, which is (knowingly illusorily) taken to be out there. This account will also solve the two problems that proponents of twofoldness have been unable to address -namely, what is the relationship between the CF and the RF and what confers 'presentativity' to what is seen in a picture.