I
As I look at a horse coming slowly towards me I am presented, in experience, with that elegant animal. Yet seeing something face to face is not the only way for it to appear to us. I may look at a black and white phot ograph of a horse by the sea, observe Whistler’s vivid painting, dream that I am edging ahead at the Grand National, or simply visualize a stallion with its thick hair caressed by the light wind. Looking at pictures, having a dream or just imagining something are all phenomena too ordinary for any theory of human experience to attempt to analyse. A proper analysis of such phenomena, though, encounters a serious puzzle: how is it possible that we are presented with something that is not physically present?
Several philosophers have outlined a story of how to deal with that perplexing aspect of human experience. They usually approach the phenomena as deliverances of a supposedly peculiar faculty whose role is to populate one’s consciousness with immaterial entities we call images. Imagination is thus conceived as a failed perception, a weak or defective attempt to launch into reality, that effectively locks oneself inside one’s mind. This approach treats perception and imagination on a par, rendering the former a particularly vivid version of the latter, turning the sharp distinction between looking at something, and visualizing it, into a dubious difference of degree.
Sartre sees imagination in a very different light. For him, clarity of philosophical vision brings respect for sharpness of conceptual distinctions. And the concept of “the percept” (what is perceived), is very different from the concept of “the imaginary” (what is imagined). The title of the book we shall examine, The Imaginary, is aptly chosen by Sartre for drawing our attention to that which we focus on when we imagine. He thus achieves the strategic goal of making the reader switch into the phenomenological mode of examining the object of consciousness first (how it appears to us, what it is to be an object for consciousness), before we start theorizing about “psychological faculties” or other notions of the traditional metaphysics of mind.
It was not until Sartre’s extensive research in the field that a philosophical analysis of imagination was grounded on the phenomena it aims to explain. His treatise on the subject was to become the standard reference for theoretical discussions in this area, and is one of the most influential books of recent times on the nature of imagining.1 In discussing imagination, Sartre is at his phenomenological best: he brings to light unduly neglected aspects of imagining experience, he dwells on cases that resist a neat categorization, he invokes psychological–experimental evidence that upsets ordinary preconceptions, and he gracefully acknowledges the limits of philosophical discourse on aspects of human experience that seem to escape or, rather, precede conceptualization.
From the wealth of topics addressed in Sartre’s treatise, we shall discuss here the fundamental issue of visual imagination. The analysis of visualizing will help us address a series of core theoretical issues concerning the methodology appropriate for a philosophical enquiry into imaginary phenomena, the difference between perceptual and imagining consciousness, the distinction between the content and the object of imagining experience, and, after I support Sartre’s dissatisfaction with some standard accounts of the “analogical representative” involved in mental images, I shall offer some critical remarks that tell against the success of Sartre’s own account of imagination.
II
Whatever else it might involve, when I imagine my father boarding a train I can easily tell whether what I do is imagining him rather than, say, touching his arm or seeing him in front of me. Yet it is quite puzzling why that is the case. What is it about imag ining that makes it so easy to tell that it is an image I now have of my father rather than a perception of him? The puzzle can be resolved if we attend to those features of imagining that constitute its “essence”, that is, those characteristics of imagining that mark it as a distinctive type of conscious activity (IPPI 4). In the first, and most influential, part of The Imaginary, Sartre purports to give us an account of that essence. Before we look at the account, let me raise some methodological issues.
The discussion in the first part is conducted at the level of “The Certain”. By that term Sartre intends to capture all those features that make themselves present to consciousness when consciousness turns its attention on its own activity. The problem is that a cursory reading of that part generates the puzzle that some of the features identified by Sartre as evident have been hardly obvious to thinkers who have reflected on such matters. Indeed, a major point made by Sartre is that important thinkers of the past have fallen victims to the “illusion of immanence”, according to which in imagining something one is supposedly looking at a picture hanging inside (and thus “immanent to”) one’s mind (IPPI 5, 6, 18, 19, 59). This view might indeed be an illusion, but it is baffling why so many philosophers have endorsed that view if a moment’s attention to one’s conscious activity presented it as “certain” that imagining has nothing to do with looking at mental pictures.
A response to this problem could appeal to a division of labour: we should distinguish clearly the phenomenological description of conscious experience from the explanation of the phenomena under consideration. Sartre’s attack on the illusion of immanence may draw on what is given as “certain” to reflection, but concerns a controversial issue in the philosophical debate over the best explanation of those phenomena. However, that response can be thought to entail that the aim of phenomenological description is simply describing, in no particular order, what is given to consciousness. For Sartre, though, “the first task of [descriptive] psychology is to make [the content of imaginative experience] explicit, describe it, fix it” (IPPI 4).2 Therefore, the overall aim of “The Certain” is not to reiterate what passes through one’s mind but to identify the essential features of conscious experience on which a philosophical account of imagination should be based (IPPI 5).3
A second methodological point concerns the process by which the imaginary material is supplied for phenomenological inspection. Sartre invites us to “produce images in ourselves, reflect on these images, describe them” (IPPI 5). The proposal sounds philosophically neutral, but it can be objected that it is a process that privileges a particular type of imaginative experience, wherein images are produced through an act of volition.4 There is, though, a rather different kind of imagining experience, in which images come unbidden. Those phenomena range from brief spells of daydreaming to the obsessive recurrence of unwelcome images in a subject who is fully awake. Such experiences appear to fall outside the province of deliberate imaginings and, thus, Sartre’s method is prejudiced against a significant part of the phenomena of imagination.
The objection is important, but it can be partly met by the following considerations. First, Sartre does examine several cases where the subject feels passive towards the occurrence of mental images, including instances where the subject asserts that she imagines something despite herself, such as in the philosophically hard cases of psychaesthenia, schizophrenia and visual and auditory hallucination. However, that examination takes place in the penultimate part of the book, after the phenomenological description of the standard cases has delivered sufficient material on which to base hypotheses for the adequate interpretation of the relevant phenomena (IPPI pt. IV). Second, the Sartrean approach deals with three moments of imaginative experience: the constitution of the image, the appearance of the object thus constituted, and the responses or attitudes taken towards the image. “The Certain” deals with the first two moments of imagining, giving particular emphasis to the second. That preference is, in my view, motivated by Sartre’s overall suspicion of the traditional view of mind as creating out of its own stuff an immaterial universe for consciousness to watch. If there is no such mental stuff, and no immaterial qualities that consciousness carries up its sleeve and bestows on its objects, then it is not in the so-called mental faculties that one should look for the characteristics of the imagined object, but in the way the object itself appears to consciousness when it is imagined. Therefore, Sartre may feel justified in commencing his phenomenological analysis by focusing on that aspect of imagination – namely, the object imagined – whose description can proceed independently of traditional philosophical doctrines, and whose understanding is a precondition for an adequate interpretation of imagining experience, whatever form that experience might take, unbidden, compulsive or deliberate.
Nevertheless, there is a remaining worry that is hardly addressed by Sartre or his commentators.5 The Sartrean approach bets on the hypothesis that, in its essential features, how an imagined object appears to consciousness is independent of whether its imagining was willed or unbidden. Sartre, I think, wins on his bet. However, that does not justify the assumption that the activity of producing images by an act of volition leaves unaffected the phenomenology of what is thus imagined. And that is precisely the methodological point I have raised.
Finally, there is a methodological problem regarding the very possibility of attending to imagining experience. Under normal circumstances, in imagining something one is simply absorbed in what one imagines. Trying to keep one’s reflective gaze fixed on consciousness as it is imagining something might thus produce theoretical cross-eyedness. Sartre notes this problem in his discussion of the elusiveness of imagination (IPPI ch. 2). We may attempt to overcome this problem by exploiting the resources of instant memory, producing an image of, say, a chair, and then constructing from memory how the chair as imaged appeared to consciousness in the moment that has passed. It might be thought that the non-coincidence of the memory with what is remembered creates sufficient distance for consciousness to fix on its imagining activities without itself being involved in the activity that has already passed. The difficulty with this proposal is that, within the phenomenological tradition at least, the memory of something involves not the observation of immaterial copies of past objects or events, but the reviving of the conscious activity that intended those objects or events.6 However, if that reviving includes the actual replay of the original imaginative act, then the initial problem of double theoretical vision would be reiterated (only further complicated by the introduction of the third item: that of remembering the conscious act on which reflection is directed).
It seems to me that Sartre surmounts this problem via a different route, that is, by bringing to the open the awareness consciousness has of itself while it is imagining something. As we have noted several times already, consciousness is non-positionally aware of itself being positionally conscious (in thought, perception or imagination) of its intentional object. Sartre’s standard procedure in the first part of his book is to make explicit the awareness implicit in the very act of imagining.7 As far as the analysis of imaginary phenomena is concerned, it is an approach that seems to deal effectively with a problem that is not exclusive to Sartre’s project, but affects any attempt to build a theory of imagination on the evidence supplied by the agent’s own experience. A non-negligible virtue of Sartre’s approach is its methodological modesty. Having spent around ten pages identifying and lucidly describing some essential characteristics of the experience of imagining, Sartre asserts that it would be wrong to proceed further without recourse to broader philosophical considerations, comparative enquiries and empirical (behavioural, psychological, linguistic and physiological) data: “Simple reflection, it seems to us, has delivered all that it can” (IPPI 15).8
What reflection delivered in Sartre’s case is the material for one of the most insightful analyses of imagining. Sartre’s text resists a simple review and the interested reader is strongly advised to look at the text itself for a subtle and illuminating narrative of imaginative experience. What I propose to offer instead is a systematization of his various remarks on how imagining consciousness relates to its objects. To that end, I shall focus on the distinguishing features of the intentional act and its object when something is imagined, and perceived, so as to gain a better understanding of the Sartrean approach to “the two great irreducible attitudes of consciousness” (IPPI 120).
III
Look at the open book in front of you. Then close your eyes, and imagine it. What was it that you just imagined? Well, obviously, you imagined the book that lies in front of you. The object at which your imagining is directed is the book that you now have started read ing again. This all might sound trivial but it is of great philosophical import. That way of recounting your experience implies that what you imagined was (not a picture, or a copy, or a photo, or an image of a book, but) the book: to imagine something is not to see an image of it (IPPI 52). Imagining does not involve three entities: yourself, an image and an object. Rather, it is a distinctive way or “attitude” through which consciousness intends a particular object: to have an image of something is neither more nor less than to have an imaginative consciousness of that thing. An image, in other words, is not the positional object of consciousness but a way in which consciousness intends its object.9
We could extend this line of reasoning to the point of denying that there are such things as mental images. Sartre could be thus interpreted as an eliminativist, anticipating a number of recent philosophers who consider talk of mental images as devoid of reference.10 That eliminativist reading, though, is resisted by Sartre’s text. In considering the option of denying “the existence of mental images”, Sartre asserts that such a “radical solution is contradicted by the data of introspection”, since we can “at will, imagine a horse, a tree, a house” (IPPI 6).11 It seems to me that what Sartre is really opposed to here is the doctrine that denies the reality of the experience of images.12 He would thus decline the characterization of “eliminativist” if it committed him to the view that consciousness was somehow systematically misled into believing that it forms images of objects. However, his wholesale attack on the “classical theories” of imagination makes the retention of the term “mental image” problematic; it might be acceptable to hold on to that term only if it is employed as shorthand for the phrase “imaging consciousness of an object”.13
As far as the reality of “the experience of images” is concerned, it would be good to ascertain whether it refers to the alleged experience of looking at an immaterial entity commonly referred to as a “mental image”, or whether it denotes the phenomenon of experiencing something as of an image, whereby one is implicitly aware of attending to something as imaged – in contrast, for instance, to being perceived. Sartre would deny the former option, since he denies that there are such peculiar entities for us to have an experience of them. The latter phenomenon should therefore be the focus of our analysis. Let us proceed by considering more closely our example of imaging a particular book.
With your eyes closed, consciousness intended the book as its intentional target, and that intending was experienced as, in crucial respects, different from perceiving that same item. But in what exactly does that difference consist? It cannot simply be a matter of whether your eyes were open or not: although having your eyes open is necessary for vision, it is not per se excluding the activity of imagination. First of all, imagining can be, and most of the time is, actually done with one’s eyes open. Second, the movement of our eyes might be relevant to certain forms of visualizing, but has no part in sensory imagination that operates in a non-visual mode, such as imagining the beginning of a melody, dialogue in your radio play, the roughness of sandpaper, or the smell of beef stew.14 There is, finally, a vast domain of imaginative experience in which we form, test or elaborate hypotheses without involving any sensory modality. We may subsume this type of experience under the category of “imagining-that” something is the case: sometimes also referred to as “propositional imagining”, in contrast to “perceptual imagining”.15 However, I shall not dwell here on this type of imagination, for two reasons. First, Sartre focuses on forms of imagining consciousness that match our perceptual modalities (especially sight and hearing), and for which he employs the term “imaging”, so as to highlight that it is the presentation of objects under their sensible aspects he is interested in;16 second, even when he discusses cases of imagining a situation or a series of events extended through time, Sartre’s analysis is cast in terms of the imaging activities of visualizing and audializing (IPPI IV, III).
IV
A better way to approach the difference between imaging and perceiving is to reflect on how the object appears when perceived and when imaged.
An imaged object shows itself orientated in a certain way, in virtue of which only a particular aspect of the object appears to u s. The object thus seems to have a “sensible opacity”, yet it appears to lack any real depth (IPPI 10). That impression is reinforced by the peculiarly tight relation that holds between the imagined object and space: as the object moves so does its space. In fact, the space of the imaged object is rather like a halo that surrounds the object in its imaginary adventures. The exact limits of that halo are hard to determine; its margins are essentially vague, neither fixed by the contours of the object, nor extending far beyond it. In contrast to perceived objects that occupy and move in and through space, imaged objects seem not to traverse any distance so much as to carry their own space with them. No wonder relations among imaged objects seem impossible to determine; to measure any two items they must occupy a common space, but shared space is a feature of perceived, not imaged, objects. Note that the same indeterminacy applies in cases where one images a scene with two or more objects. The exact distance between a horse, say, and a palm tree is not something that can be revealed to the subject, however hard one “observes” the imaged scene. And if you decide that the imaged distance is, for instance, four metres, there is still no criterion of correctness for your judgement except your decreeing that for a distance to be four metres, it must be how far the one object appears to you to be from the other.
That lack of fixed measures is indicative of the overall lack of determinations characterizing imaged objects. There are no set distances between any two points of an imaged object. Even in cases where one imagines an object along given specifications (e.g. by following the instruction of forming the image of a terracotta vase whose right handle is twelve centimetres from its left) one is not able to check that the object imaged fits the specifications, for example by bringing forwards the imaged vase, so as to apply your mental tape measure to it. The distance or any other relation among features of an object is not something that you discover but something that you, in the very act of imagining, make so: the imaged vase cannot check your judgement, since what and how a thing is imagined co-varies with your imagining it.
An important implication of the above considerations is that, contrary to popular – and philosophical – preconceptions, the world of imagination might not be as exciting as one might initially think. In comparing the world of perception with that of imagination, Sartre’s preferences for the former are hard to mistake (IPPI 8–11, 128–30). A perceived object always has more to give you than a quick glance reveals to you. The longer you observe it, the more details about itself will become evident to you: its surroundings; its relations, matches and contrasts with other objects. In turning an object around, or in making a tour of it, new aspects will come into view, and the previous aspects will acquire all sorts of practical, aesthetic or other dimensions that might not have been obvious before. An object encountered is, as Sartre puts it, overflowing our experience; neither our perceptual takes on it nor our descriptive accounts can exhaust all that is ever possible to experience of, or to say about, that thing (IPPI 10).
The imaged object, by contrast, is characterized by an “essential poverty” (IPPI 9). It is not only a lack of detail in the colour, texture, shape or outline of the imaged object that accounts for its poverty in sensible content. It is also the striking absence of relations with other objects or features of the world. Every perceptible object is in a large, if not infinite, number of relations with other objects in its environment; indeed, a wealth of relations to other objects goes hand in hand with the object’s individuality. It is precisely because an object is exactly what it is and not another thing that it finds itself in innumerable relations with other things constituting the world of sense experience. An imaged object, on the other hand, enjoys a very limited set of relations both internally and externally: “The different elements of an image maintain no relations with the rest of the world and maintain only two or three relations between themselves: those, for example, that I could note, or those that is presently important to retain” (IPPI 9). The paucity of internal and external relations and the lack of various kinds of determination combine in the arising of the feeling that attending to an imaged object makes at best for “quasi-observation”. Observing an imaged object does not promise any revelation, since however hard I attend to it, “I will never find anything there but what I put there” (IPPI 9).17
The last remark, although obvious, should be treated with some care. To me it does not sound like a phenomenological claim, but an explanatory note on the nature of imagining experience. It might therefore be worth specifying the phenomenological grounds for Sartre’s claim that imaged objects appear as they do because we somehow make them to appear thus and so.
One reason that can be offered in support of this claim is that we experience no time lag between the imaged object and the imaging consciousness (IPPI 11). The problem, though, is that this fact can be subject to different interpretations. The simultaneity of imaging consciousness and imaged object means that the latter does not precede the former; hence the object does not exist as imaged in advance of our imagining it in one way or another. However, it can be argued that for an object to owe its constitution – or, at least, its manner of appearing – to some other thing, that other thing should precede the object that it determines. But the simultaneity of consciousness and object appears to preclude the scenario of the object being what it is because the imagining consciousness occurs first, and a moment or two later the object, as an independent product, makes its appearance. Furthermore, that claim would be falsified by ordinary experience, and Sartre certainly admits as much.
The problem, perhaps, is generated by the assumption that we have so far been concerned with the temporal relation between two “things”. But consciousness is not a thing, it is an intentional “aiming at”, a “directedness toward” things. In imagining a vase, our previously acquired knowledge (about vases) and a current intention (to form an image of a specimen) are “indissolubly linked” in the synthetic act of imagining the object. Neither the imagined object nor the consciousness intending it can exist separately, since for an object to appear as imaged is just consciousness intending that object in the image form: “the intention reveals itself at the same time as it realizes itself, in and by its realization” (IPPI 11).
The indissoluble link between knowledge and intention may further illuminate why the Sartrean realist is not in a hurry to leave the world of perception for the universe of imagination. The fact that no imaging intention can be formed without some knowledge about what is imaged and about how the thing imaged looks (sounds, feels), entails that in imagining there is “not a second of surprise” (IPPI 11). To be sure, one’s imaginings can be accompanied by a feeling of surprise about oneself: more precisely, about one’s ability to follow through the instructions in imaging a cube, changing its colour and rotating it 180 degrees, or about one’s ability to give a really adequate description of what an imaged object looks like. But the surprise does not come from the part of the imaged object itself. We do not often hear someone say, “I was really surprised that the brown horse that I imagined turned out to be white”. To imagine something with different colours is not a process of getting to know more about a fixed thing, but of imagining different things.
V
We have so far focused on the differences between the appearance of an object in imagination and its appearance in perception. The way something looks – regarding its shape, volume, depth, relation to other things, the location in and movement through space – is sharply different when it is perceived and when it is imagined. Those differences seem to remove in large part the mystery that initially surrounded the question of why it seems so simple to tell whether I now imagine – rather than perceive – a horse galloping by the sea: a horse imaged just looks so differently from a horse perceived.
It is not only the objects, though, of the acts of imagination and perception that appear very different, it is also those acts themselves. Their difference consists in the way each of those acts posits – sets before itself – its object. The positing of perceptual consciousness involves the affirmation of the object’s current existence in the subject’s perceptual field. Perception, in other words, comes with a belief in the presence of its intended object. Imagination, on the other hand, denies existence of its objects in four possible ways: by directly positing the object as nonexistent, by positing it as not currently present or as existing elsewhere, or by withholding any commitment as to its existence (IPPI 12).
The lack of existence is not an extra feature “superimposed” on the image after it is formed. The positing of an object as absent or nonexistent “is constitutive of the image consciousness” (IPPI 13). To think otherwise would lead us to the twin errors of idealist metaphysics and false phenomenology. Sartre is rather brief on this issue, so let me spell out what I think is his underlying reasoning on this point.
The “superimposition” thesis would imply that there is an object fully formed inside one’s mind for us to watch and that two of our mental faculties – perception and imagination – vie to possess it: if existence is added to that internal object, the object passes to perception; if existence is denied it, the object goes to imagination. A first problem with that account is that it underestimates the radical difference in nature between the imaginary and the perceived. As we saw in the previous section, how something appears to consciousness when it is imaged is very different from how it appears when it is perceived. Therefore, the claim that there is a fixed object that appears in an identical manner to both perception and imagination is falsified by the facts of ordinary conscious experience.
A second problem with the superimposition thesis is the lack of phenomenological evidence in its support. We just do not seem to first imagine something and then subtract presence from it. I do not, for instance, form the image of Pierre having a stroll in East Berlin, and then consider whether or not what I intend is actually in front of me. Hence, in imagination, consciousness does not “reduce” the reality of the scene after it has intended that scene: to imagine a scene is already to intend it as not presently occurring. Sartre illustrates this point with an apposite remark:
If the image of a dead loved one appears to me abruptly, there is no need for a “reduction” to feel the ache in my heart: it is part of the image, it is the direct consequence of the fact that the image gives its object as a nothingness of being.
(IPPI 13)
The third and, in a sense, the most fundamental problem with the superimposition thesis is that it is premised on the assumption that what is aimed at by consciousness is an internal immaterial object that is made available to either perception or imagination. Sartre, though, believes that no such internal object exists: in imagination, no less than in perception, whatever consciousness is directed at is outside consciousness (IPPI 53). When I imagine Pierre walking by the Brandenburg Gate, neither that stone edifice nor my friend are inside my consciousness: whatever that “inside” is taken to mean. If it is taken in the physical sense, the material entities (Pierre and the Brandenburg Gate) cannot possibly be inside my consciousness, which is not even a three-dimensional thing. If it is taken in the intentional sense, the entities imaged (Pierre and the Brandenburg Gate) are not immanent to consciousness but, as Sartre puts, it, transcendent to it: they are those objects towards which consciousness is directed. Either way, the claim that what is imaged is an “internal” object seriously misrepresents the intentional structure of conscious activity.
VI
The phenomenological source of the puzzle of imagination is that the object as perceived looks different from how it looks as imaged. The traditional response to the puzzle is that the difference in “looks” is really a difference in objects; in conscious activity we deal with two different kinds of object, one being the object imaged, the other the object encountered in perception. Sartre, on the contrary, states that it is one and the same object intended in two different ways. The world does not contain things imagined located next to, beneath or above things perceived. There is only one reality intended in two different ways.
Imagining is a way in which consciousness intends its object. The notion of “object”, though, is itself in need of some clarification. Consider first the kinds of thing we imagine: they can be an entity (a vase), an event (the vase falling on the floor) or a state of affairs (that the vase is no longer on the shelf). When Sartre enquires about the object of the imaging act, he is focusing on single entities (a chair, the hidden arabesques of a wall tapestry, or Pierre; see e.g. IPPI 3, 7, 12, 119, 120). He is interested, in other words, in entities as imaged. We may refer to such entities in response to the question “What is it that you image?” An answer to that question, though, might refer ambiguously to both the object and the content of a mental act. And this raises a crucial issue in philosophy of mind. Which of those two things, the content or the object, is “before our mind” in an imaging act?
According to one line of reasoning, imaging consciousness cannot be intentionally related to an object. A relation relates things that exist. The alleged object of imagination is non-existent. Hence, there cannot be an intentional relation between imaging and an object. Imaging, however, is not devoid of content: we can always give some answer to the question “What is it that you image?”; we cannot image without imaging something, whatever that might be. Therefore, imaging is an intentional relation, not to an object, but to a content.
This line of reasoning is bold, and comes with a strong philosophical pedigree.18 Its conclusion, though, is problematic. The source of the problem is its view of intentional content. That view makes content dependent on an intentional object. A mental act intends its object always in a particular way; the object appears from a particular angle, or at a certain distance, showing itself under this or that particular aspect. Call this the “aspectual shape” of the object. To talk about the “content” of the mental act is then understood as talking about “the intentional object with its aspectual shape” (Searle 1992: 155; cf. Crane 2001: 38). However, if content is defined in terms of object, it is not clear to me how there can be intentional acts that lack the latter but retain the former.
We can express the same worry in the form of a dilemma. Imaging is either contentful or it is not. If it is not, then it cannot consist in a relation to an intentional content. But if it is, it cannot involve a relation to a content without intending the object in virtue of which such content exists in the first place. The dilemma shows that we cannot bypass the issue of the intentional object by postulating content as the focus of imaging intention.
The intentional object holds a prominent place in the Sartrean analysis of imaging consciousness. The opening and closing remarks of The Imaginary are devoted to a clarification of what exactly the intentional object of an imaging act is. However, the Sartrean approach appears to suffer from the opposite problem to that which affected the previous theory. Whereas that theory of imaging banished the intentional object from its account, the Sartrean emphasis on object seems to work against the reality of content. The paucity of references to “mental content”, and the absence of a section dedicated to that notion, might give the impression that the concept of mental content has been excluded from the Sartrean system. The impression is reinforced if we recall the attack on content launched in Sartre’s early paper “Intentionality”. A commitment to mental content was there taken as the trademark of “digestive” philosophies, with their idealist predilection for an all-consuming mind, which swallows material things, turns them into immaterial entities, stores them in a spiritual container, and constructs out of them a simulacrum of reality.
Sartre’s early attack on digestive philosophy was gradually developed into a detailed critique of philosophical doctrines – common to both idealist and empiricist schools of thought – that give a distorted picture of human experience (see esp. Ion chs 1–2). However, as his critique of the traditional view of mind became subtler, so his attack on mental content became qualified. By the time of the publication of his second work on imagination, the notion of mental content seems fully rehabilitated.19 I would maintain that if The Imaginary includes no section headed “mental content”, that is not because content is kept in exile, but because it is a notion underlying much of the discussion in each chapter, making it necessary to employ a variety of terms that would specify which particular aspect of content is under consideration in the course of each discussion. Thus Sartre talks about “matter”, “hyle”, “analogon” and “psychic datum” (see e.g. IPPI 50, 51, 53). The most interesting among those notions, and the one that has attracted most philosophical attention, is the notion of the “analogon”. Let me explicate this notion, and to explain why its use generates problems that Sartre’s theory appears unable to resolve.
Consider the case of seeing a portrait. A painting hangs on a gallery wall, and by looking at it, you see the face of, say, Jacqueline Du Pré. The painted canvas counts as a portrait because you see a face in it. The material thing in front of you is so marked, coloured and, in a word, “manipulated” that the face of someone absent is presented to you. Your consciousness intends that person by responding to the painting in a particular way. Hence, there are at least three elements involved in the seeing of something in a painting: an intention that aims at that thing; a matter that the intention transforms; and the thing aimed at that is not present.
To be sure, seeing, in this case, does not come on its own. There is background knowledge (acquired perhaps by seeing in the past the face of the famous cellist on CD covers), knowledge that helps you identify what you see in the painting as the face of that particular person; and there is an affective response to features accentuated or otherwise emphasized in the painting, which make you feel in the presence of the uniquely gifted yet tragic figure of recent music history.
The painting supplies the material that our visual intention animates so as to let something that is not actually in front us become present.20 The painted canvas is not experienced as an indifferent object but as a portrait, that is, as something that solicits our seeing in it the face of the absent person. In the normal run of things, the reception of something as a portrait is the intended result of the act of drawing, marking, colouring or, in general, creating a marked surface whose visible aspects, especially its outline, shape and composition, resemble the person pictured (IPPI 24).21 The resemblance of the portrait to the person portrayed is one way in which something perceived functions as the analogical representative of the pictured object. “Analogon” is the term Sartre coins for denoting any type of content, awareness of which presents us with an absent object.
Putting the above points together gives us a definition of image as “an act that aims in its corporeality at an absent or non-existent object, through a physical or psychic content that is given not as itself but in the capacity of ‘analogical representative’ of the object aimed at” (IPPI 20). When the content is “physical”, the image may include the perception of a photo, a film screen, a painting, a sketch, a line drawing or anything else (such as a mime performance) that solicits the appearance of some absent person, object or scene. If the content, though, that our intention animates is “psychic” then there occurs a “mental” image.
To assess the justifiability of this scheme, it would help to consider three questions. Why talk about an “analogon” at all in the case of mental images? What is the analogon of a mental image according to standard accounts of imagination? What is put forward as the analogon of a mental image by Sartre himself? I shall answer those questions in a way that highlights the difficulties Sartre invites by subscribing to the idea of the “analogon”.
VIII
The analogon is not something revealed by phenomenological analysis. When I imagine Pierre, there is no in-between matter, no psychic canvas that I animate with a view to reaching an absent object: the imaged Pierre is directly presented to me as absent. Sartre does not deny that the experience of imaging someone is that of being directly intending that person, without first having to watch a psychic intermediary of mental stuff (the way one has to see the physical intermediary of the painted canvas if one is to see something in it). Yet he seems to hold on to the existence of a psychic intermediary endowed with a somehow lesser existence than their physical equivalents, material pictures. He admits, for instance, that, contrary to the painted surface of a canvas, which can be perceived in a number of different ways (visually, observing its patches of colour; tactilely, running the tip of our fingers through its corners; even olfactorily, checking, for example, for the presence of varnish or other preservatives), and which can, as an object, exist well before and after our visit to the art gallery, “the matter [of a mental image] has so little independence that it appears with the image and disappears with it” (IPPI 24). But “little independence” is already too much: in imaging Pierre, consciousness seems aware of no thing awaiting observation so as to get to Pierre by way of seeing that thing; nor is there any matter of the image left after consciousness ceased imaging, available for consciousness to examine in its own right: “when consciousness is annihilated, its … content is annihilated with it … there remains no residue that can be described” (IPPI 53).
Note that the perfect simultaneity between imaging an object and the existence of the analogon of the object imaged may not on its own tell against the reality of the analogon. After all, we could perhaps conceive of a universe where portraits sprang into existence the very moment one turned one’s gaze at them, and ceased to exist as soon as one stopped looking at them. Even in that universe we could in principle observe and examine the physical analogon on its own, for its various sensible and compositional properties vis-à-vis its role as a material representative of the pictured object. However, such an independent examination of the analogon is not available in the case of mental images. The simultaneous existence of imaging and of the psychic analogon is a source of philosophical worry because it indicates that we have no independent hold on the notion of the analogon, except for its postulation as a psychic stand-in of the imaged object.
IX
Recall that, according to Sartre’s own phenomenological enquiry, to form an image of Pierre is not to imagine an image (be it mental or physical) of Pierre. Imagining is an act directed not at images, but at objects themselves when absent from the agent’s perceptual field; an image is “a certain way that an object has of being absent within its very presence” (IPPI 85).
The puzzlement that is understandably generated by this phenomenon has led theorists of the past to think of images as substitute entities. On that account, what consciousness is directed at is a thing in miniature, which inhabits the mind, and which carries the exact qualities of the thing it represents. The psychic miniature represents the real thing precisely because it shares with it those qualities that lead imaging consciousness from the immanent copy to the transcendent original. Alas, by transferring to the psychic content the qualities of the imaged object, philosophers of the past fall prey to the illusion of immanence, or so Sartre contends (IPPI 87–8). Before we discuss the Sartrean way out of this problem, let me explain why, in my view, Sartre is right to attack past theories of this matter as philosophically implausible.
There are two kinds of reason why the miniature story is unlikely to be true. On the one hand, it is hard to think what might fill the role of such a miniature. If it is to be a tiny material object, whose ability to represent something is due to its sharing with the original thing various sensible properties (shape, colour, etc.), then that material miniature does not match our ordinary conception of a mental image. To use a simple example, an image of a ripe yellow banana does not appear itself to be curved, yellow and ready to eat. We cannot literally change the colour of that image (e.g. by applying red dye to it), and even if we switch from the image of a yellow banana to that of a red banana, that is not because we have in any way altered the colour of the yellow banana, but because we have now started imagining something different. If, though, the miniature entity is to be immaterial, then it becomes even harder to understand how a non-material object can bear sensible properties such as shape, colour and so on.22
The other kind of reason for not accepting a miniature entity as the analogon that stands in for the imaged object is that the structure of imagining is not the one of seeing one object by looking at something else. That is by now our familiar fact that the phenomenology of mental imaging does not exemplify the indirect structure of perceiving a portrait. In imagining, we are directly presented with Pierre as imaged, and not with some other thing (a miniature, a canvas, a piece of printed paper) in which, or through which, we can make out Pierre.
Can Sartre improve on the standard accounts of the “matter” of mental images? We may answer this question by considering his suggested candidate for the role of the analogon.
X
In his discussion of the analogon, extended through seven chapters of The Imaginary, Sartre identifies three elements that come together to form the imaging attitude: “knowledge”, “affectivity” and “kinaesthetic sensations”. Let us briefly examine each one in turn.
“Knowledge” refers to the cognitive content that informs our intention of imaging someone or something as a particular entity or, at least, as an instance of a particular kind of entity. I can, for instance, respond successfully to the instruction “form the image of a cube” because I possess knowledge about the shape of that geometrical object. Similarly, I can recognize an image that came to my mind unbidden as an image of the Eiffel Tower, seen from the exit of Trocadéro Métro station, because I have relevant background knowledge that goes into the formation of that image.
Knowledge, however, does not come on its own; it brings with it – we might even say, in it – a felt quality that corresponds to the affective properties of the object or of the scene imaged. To stay with our Paris example: imaging a metal edifice from that particular viewpoint is, for me, to intend the Eiffel Tower, because to that viewpoint there are attached feelings (e.g. of longing, of travel fatigue, of the peculiar numbness following a cancelled rendezvous), which give an affective texture to the experience. It is therefore the imaging not of any old metal construction, but of the Eiffel Tower, as seen by someone rushing out of Trocadéro Métro station.
What about the third element of the Sartrean account, the “kinaesthetic sensations”? To get a taste of how this notion works in Sartre’s system, try to image the Eiffel Tower – taking in, in your imagination, the whole of it from bottom to top – without moving your retinas upwards and downwards; or try to image a moving pendulum without letting your eyes swing from right to left, back and fro. Sartre bets that it cannot be done, and he offers this fact as presenting an instance of the general rule that in imaging something our conscious intention animates sensations of our body, dressing them, one might say, with the shapes of the image world (IPPI 78–83).23
If that approach to imagination could be adequately developed, its advantages over the traditional competitors would be considerable. It does not present the analogon as a miniature object watched by consciousness, and thus avoids the above-mentioned difficulties involved in explicating what type of object (material or immaterial) could supposedly intervene between consciousness and the imaged object. Indeed, the Sartrean approach appears to do away with the indirect structure of seeing one object in another (taken over from the case of seeing something in a physical picture), and to preserve the autonomy of imagination as one of “the irreducible attitudes of consciousness” (IPPI 120).
I believe, though, that Sartre’s approach faces problems of its own. The first is what we might call the problem of discrimination. The retinal movement (or, for that matter, any other bodily movement whose awareness is supposedly involved) in imaging the Eiffel Tower, from the ground to the top, might not be very different from the retinal movement in imaging the Empire State Building from its ground entrance to its spire at the top. The same problem applies to the cases of imaging a garden swing and a steel pendulum, or indeed, any two or more objects whose pattern of movement, or outline shape, is more or less identical. Yet, we have no difficulty telling that what we imagine is a swing hanging from two trees or a small pendulum decorating an office desk. Therefore, invoking the Sartrean analogon of kinaesthetic sensations cannot help explain the obvious differences in the things imaged, when those things are re presented (in their outline shape or pattern of movement) by the same kinaesthetic sensations.
The second problem is what might be referred to as the problem of comprehensiveness. Awareness of bodily movement (be it of the retina, of our index finger, or of any other moving part; IPPI 74–8, 80–81, 134) is not a sufficient ground for all the sensible properties of things imaged. Imagine a transparent cube, and then imagine a cube of the same size but of red colour, or of yellow, or of purple. The colour of the thing imaged will in each case be different; yet the outline shape, which is (supposedly) grounded on the feeling of retinal movement, is clearly the same. Therefore, the Sartrean hypothesis of the kineasthetic analogon leaves out properties – such as colours – that are an indispensable part of many an imagining experience.
It might be worth clarifying the nature of the critical remarks I made against the Sartrean view of the analogon. It was granted that Sartre discusses the analogon as part of an explanatory story of the phenomena of imagination. References to the analogon do not form part of his phenomenological description of how consciousness intends the imaged object; on the contrary, Sartre is keen to underline that the analogon of a mental image, even though it is not appearing to unreflective consciousness, is something that should exist, given the structure of imaging experience in general (whether, for instance, we intend a painting, a printed image or a mental image). Hence, one might respond to my critical remarks by stating that they target the phenomenological shortcomings of the Sartrean theory of the analogon, whereas that theory presents an explanatory hypothesis, not a phenomenological account of imaginary experience.
However, that is not how my criticism works. The worries I expressed concern the ability of the Sartrean theory to explain the phenomena under consideration. As an explanatory hypothesis, the idea of the kinaesthetic analogon is shown inadequate to the twin tasks of (i) telling apart experiences of imaged objects that are clearly different, and (ii) covering an essential aspect of any visualization that intends its object in its various sensible (including its colour) qualities. I have not brought into the discussion considerations external to Sartre’s way of thinking about imagination,24 nor have I sought to stress implicit inconsistencies between some of his phenomenological statements.25 My point has been that, even if we take Sartre at his own words, the philosophical task of giving a satisfactory analysis of the “analogon” of mental image remains ahead of us.
XI
We began this chapter by posing the rather common question of how it is possible for us, in imagination, to somehow be presented with things not physically present. The philosophical sense of that question concerns the conditions of the possibility of imagination. It aims, in other words, to an explication of what should be the case for consciousness to be able to intend things as being otherwise than they really are.
If the above question sounds philosophically pressing, if, in uttering it, we seem to express a serious puzzlement as to the very possibility of the phenomena of imagining, that might be due to an underlying assumption about what consciousness is, and how it supposedly works. We seem, to put it differently, to come to the question of the alleged “peculiarity” of imagination, having already decided (if we ever thought about it) how consciousness works in the “normal” run of things. It might, I think, be more fruitful and, from a phenomenological standpoint, less prejudicial, to reverse the order of explication by asking what consciousness is, given its ability to imagine. That ability is not a mere, or a rare disposition: it is something ordinarily and effortlessly realized, in our daily transaction with the physical and social world.
Accordingly, in the final chapter of his treatise, Sartre asks: “what are the characteristics that can be attributed to consciousness on the basis of the fact that it is consciousness capable of imagining?” (IPPI 179). That question is soon substituted by a different one: “is the function of imagining a contingent … specification of the essence ‘consciousness’ or should it rather be described as a constitutive structure of this essence?” (IPPI 179). The former question invites the derivation of other characteristics of consciousness on the basis of its ability to imagine; the latter question stays with that ability, and probes the ability’s relation to the essence of consciousness. Recall that, according to Sartre, consciousness is necessarily consciousness of something, or, as he puts it in that chapter: “it is in the very nature of consciousness to be intentional and a consciousness that ceased to be consciousness of something would thereby cease to exist” (IPPI 183). Intentionality, being the essential characteristic of conscious experience, will set up a train of reasoning that will arrive at the conclusion that “imagination, far from appearing as an accidental characteristic of consciousness, is disclosed as an essential and transcendental condition of consciousness” (IPPI 188). My aim in this section is to reconstruct that piece of Sartre’s reasoning. Here is how, I think, it goes.
To imagine is to posit an object as “irreal”. We can posit something as irreal only on the background of what is experienced as real. The imaged object is something that consciousness constitutes through isolating that thing from its actual environment, and annihilating it, that is, intending that object as not being part of or in the world. Negating the worldly status of an object, however, is only possible if one has a sense of there being a world. In perception, objects are posited as real because they are presented against the background of the synthetic totality of spatiotemporal, physical, practical and other relations that constitute the world. Having a sense of the world, though, presupposes the ability to stand back, and take in what is occurring in one’s experience as being an event in the world, to intend what is encountered as being located in the world, to engage in practical activities as unfolding in the world and so on. The constitution of the world implies the ability of consciousness to stand back and intend the world as the meaningful synthesis of the items that make up reality. This “standing back” of consciousness, its ability to withdraw from the given, so that it may then be able to intend it in its multiple significations, is nothing other than the ability required to set something before oneself as irreal.
“We must bear in mind”, Sartre writes, “that the act of positing the world as a synthetic totality and the act of ‘standing back’ from the world are one the same act” (IPPI 184). The puzzle of imagination was generated by the worry about how it is possible for consciousness to withdraw from the real and to constitute objects as irreal. The key to the puzzle is that without the ability of consciousness to stand back, and intend things as being an integral part of a network of objects and events that far exceeds the individual item on which consciousness is focused, we would have no sense of reality, no awareness of there being a world. It is because consciousness is not exhausted in the individual item (side, aspect or part of a thing) currently perceived, that it can intend the whole object – any perceptible object – as real. But that is also what is required for the activity of imagination, that is, for intending objects in a way different from the exact way they are given to us. It is because consciousness is not overdetermined by what is immediately given to it that it can intend whole objects as being what they are (in perceiving) and as being different from what they are (in imagining). The ability to transcend the given is the fundamental precondition for both perception and imagination; and – to complete that section of the Sartrean reasoning – since we think there is nothing extraordinary in realizing that ability in the case of perception, we should think the same in the case of imagination.
That is a truly interesting argument, but it is not clear that it serves the goal set by Sartre himself. What Sartre was setting out to establish was that imagination is nothing less than a “constitutive structure of the essence” of consciousness. Instead, what he appears to have shown, so far, is that a major activity of consciousness – that of being directed at perceptible objects – requires the very same ability activated in the case of intending objects as imaged. More simply expressed, my point is that what the Sartrean argument has delivered (if successful) is that an essential ability of consciousness is employed also in imagining things, not that imagining things is an essential ability of consciousness. Can we move from the former claim to the latter?
The move is, I think, effected in a complex manner, as it reaches completion in several steps. The first step capitalizes on the ability of consciousness to stand back from the instantaneous given so as to synthesize things into a meaningful whole. Sartre gives this ability a name no less potent than that of “freedom”: “to posit reality as a synthetic whole is enough to posit oneself as free from it and this surpassing is freedom itself since it could not be effected were consciousness not free” (IPPI 184). It is worth noting how straightforward and specific the introduction of “freedom itself” is in the Sartrean system; popular misconceptions to the contrary, there are no political, moral, humanistic, theistic or secular overtones in Sartre’s primary use of the notion of freedom. Instead, it is a notion employed to convey something revealed by the analysis of conscious activity.
The “surpassing” mentioned in the above quote is the next important concept for our argument. Surpassing never happens in a void. It involves, instead, three specific dimensions: there is something that gets surpassed, from a certain point of view, towards something that was not given to consciousness in advance of that activity. In the simplest case, the facing side of a cube is surpassed from my spatiotemporal standpoint towards the totality of its sides, so that perceptual awareness of a whole object is duly formed. However, surpassing is not limited to the perceptual case, and its function is not exclusively representational. In a side remark, that, in my view, is crucial for bringing imagination to the heart of this argument, Sartre notes that “there are in fact, for consciousness, many other ways to surpass the real in order to make a world of it: the surpassing can and should be made made first by affectivity or by action” (IPPI 185). To appreciate the significance of that remark, think of what is involved in attending to the world as an agent: someone acts in the light of how things are (perceived by him to be), from a particular (spatiotemporal, practical, emotional or other) standpoint, towards realizing his view of how things ought to be.26 Being able to intend things as being otherwise than they are is a necessary condition for acting so as to effect changes that would make the real world fit one’s “irreal view” of it. Within that view, I “negate” certain aspects of a real object (some of its current qualities, or dimensions, or location, or relations to other things) and intend that object, under its sensible aspect, in different ways (having different qualities, dimensions, location or factual relations) than the ones perceived. But what is that ability but imagination itself?
Three qualifications of the above account are in order. First, our reconstruction sharpens the distinction between different conscious activities that are not, in practice, experienced as clearly segregated, or separable from each other. Second, the passage from the real to the irreal need not always be materialized; it suffices, according to Sartre, that the perception of something on the background of the world involves, on the part of consciousness, the possibility of a double negation: that of denying that the thing belongs to the real, or that of denying the real in positing that thing (IPPI 183, 186). Third, the negation is never general, or arbitrary, in the sense that it is aimed at a particular thing (object, event or state-affairs) of which one negates particular aspects. Consciousness, for Sartre, does not hover over reality; it is always necessarily “situated in the world”, being the directedness towards things (objects, events or states of affairs) from a particular (spatiotemporal, historical, moral, political, aesthetic or affective) standpoint (IPPI 185–7). The situatedness of consciousness means that in imagining, one “nihilates” particular things by negating specific aspects of them, forming a view of those things as being in certain respects different from how they are perceived. “Thus,” Sartre insists, “the situation of consciousness must appear not as a pure and abstract condition of possibility for all of the imaginary, but as a concrete and precise motivation for the appearance of a certain particular imaginary” (IPPI 185).
Let us retrace the steps of the Sartrean argument so far. To imagine is to posit something as irreal; that is possible only against the background of things appearing at first as parts of the real world that is itself not a bare given, but the synthetic totality of things; the world can be constituted by consciousness because consciousness is free, that is, it is able to “stand back” and “surpass” the immediately given towards a meaningful whole, in perception, in action or in affection. When, in surpassing the given, one intends something as being different from how it is perceived as being, when one negates certain of its aspects, posited as real, one brings forth an “irreal” object: an object “withdrawn from reality”. That “nihilation” of the object is always specific because it is motivated by the situation in which consciousness finds itself. And thus we reach the final important step in this reasoning.
“Constitution” and “nihilation” are the two essential moments of imagining activity. But, according to Sartre, who on this point remains an unrepentant Cartesian, constitution and nihilation are also the crucial moments of the activity through which consciousness reveals itself to itself:
Is not the very first condition of the cogito doubt, which is to say the constitution of the real as a world at the same time as its nihilation from its same point of view, and does not the reflective grasp of doubt as doubt coincide with the apodictic intuition of freedom?
(IPPI 186)
The question is rhetorical, and its objective is none other than to highlight the central place of imagination in the life of consciousness. Imagining is not an optional extra, it is not a rare charisma, or a contingent addition to our conscious engagement with the world. For Sartre, “imagination … is the whole of consciousness as it realises its freedom” (IPPI 186).
In offering an account of the phenomena of imagination, Sartre (and I, along with him) appealed freely to a basic understanding of notions such as reality, essence and existence. It is time to look more closely at Sartre’s mature analysis of those issues. The text that our discussion refers to will be the opening sections of Sartre’s major philosophical treatise, Being and Nothingness.