[an error occurred while processing this directive] [an error occurred while processing this directive]

November 07, 2022

BooksArticlesReviewsPapers presented

Rob van Gerwen


Depiction and the Intimation of Experience.



[Aesthetic attributes] furnish an aesthetic idea, ... with the proper function ... of animating the mind by opening out for it a prospect into a field of kindred representations stretching beyond its ken. [The fine arts] drive the soul that animates their work wholly from the aesthetic attributes of the objects-attributes which go hand in hand with the logical, and give the imagination an impetus to bring more thought into play in the matter, though in an undeveloped manner, than allows of being brought within the embrace of a concept, or, therefore, of being definitely formulated in language. (Kant, Critique of Judgement, section 49:7, B195).

1. INTRODUCTION.

Usually, analytical aestheticians understand the riddle of pictorial representation as the problem of how to account for the functionality of pictures in making present the absent. Naturalists explain pictorial functionality in terms of similarity, resemblance, or imitation, and they refer to our natural powers and propensities to recognize the like and the unlike. Conventionalists tend to dismiss the arguments presented in defence of these notions one after the other, and contend that it is conventions which explain our so-called natural propensities and faculties, instead of the other way around. Evidently, the approach here diverges from the merely epistemological concern with the (im)possibility of establishing whether our representations are true or not, a concern, that is, with the idea of a God's eye point of view enabling us to establish such truth. Starting then from the aesthetic approach to the matter of representation, I further limit my concerns in this paper to pictures, photographic or filmic, that represent persons, their facial expressions, their actions and reactions, and, more narrowly, to those among these pictures that also profess to render the experience of these persons. I am interested here in pictures then that pretend to inform the spectator about what it is like to go through a certain experience. Normally, such making present of the experiential is not taken as a case of representation, but either as the evocation of a relevant emotion in the spectator, or as expression, which is taken to function in a reversed direction when compared to representation.

The latter two notions, evocation, and expression, however, are connected with the problematic idea that we can and should separate the affective impact of a representation from its representational achievement.[1] But can we? This separation of representation from its affective impact has not been questioned adequately in analytical aesthetics, due to the neat ways in which it is laid out. The emotional is private, intrinsic, and subjective and this might seem to imply that we cannot represent it in the first place, or that its representation is possible unproblematically through representing its publicly accessible outlook; we are supposed to be able, then, to analyze the problem of representation without reference to the affective. I agree that we can account for representation without reference to the emotional response evoked, and also, that a picture of a woman can be said to represent the woman irrespective of the question of the exact expression involved. Nevertheless, in this paper I shall comment on the separation of the affective from representation, and shall argue that the affective is more narrowly connected with the representational if the representation is of the experiential.

There exists a peculiar way to represent the experiential and it is not at all unproblematic. In terms of intimation I shall propose an account of such representation of the experiential, and shall determine its singular contribution to our identification of the problem of representation.[2]

2. SHOW AND TELL.

Wittgenstein's remark in the Tractatus that "What can be shown, cannot be said" can be seen as the utterance of an exclusive disjunction, referring to the general difference between language on the one hand, and pictures or the gestural, real life, presentation of things on the other: either you put something into words or, if you cannot do this, you show it.[3] My proposal here is, that we also distinguish what I call `intimation' of what we cannot say or show literally: the experience of the antagonist person figuring within the described or depicted (fictional) world. Intimation is the recurrence within the spectator of the experience supposedly being lived through by the represented person. Thus an object or event intimates an experiential aspect if and only if it makes us experience this aspect for ourselves. To achieve such a recurrence, it is my thesis, the spectator must actively bring in certain relevant, personal, memories, and therefore intimation is a side-effect of more literally shown or described events or states of affairs. Pictorial intimation then is not a case of showing, but of not-showing instead.

Let me give two telling examples: I once saw a news footage on television about a racial riot in South-Africa. People were stoning a young man. While watching this I was fairly confident that he would escape: people were merely throwing stones at him, and they wouldn't want to go on throwing them as soon as they would realize that he was wounded, et cetera. Next, other events were shown and I had already forgotten the young man when, out of the blue, one single shot was shown with the young man lying on the deserted street: he had died. Devastatingly, the news-reader didn't even notice. Had they shown the man being stoned to death this wouldn't have made the difference that the not-showing did. Now, however, his dying nearly grew into an experience of my own. That this profound effect was not merely a function of the reality of the event may become clear from the other - fictional - example, of a scene taken from a moving picture by Robert Bresson, L'Argent. In this film, a criminal hides out in a shack of a farm owned and run by an elderly couple. The wife takes care of the criminal, the husband thinks he should be taken to prison. In the relevant scene the wife, on her way to bring the criminal a cup of coffee, is stopped short by her husband. A few irritated glances are exchanged, and, just when the husband strikes out to slap his wife in the face, the camera moves downward, to show the impact in the sudden movement of the trembling cup, and the coffee spilling over. Instead of being shown the man's hand hit the woman's face we are being shown the dancing cup and saucer. The shock occasioned derives not so much from watching a man slap his wife, but from the sheer moral depth of this event.[4]

Now, in both cases this not-showing is more impressive than showing might have been. Because the slapping is not depicted the situation becomes more intimate, we are more concerned with what it would be like to be this elderly woman being slapped in the face by this husband with which you have had -to say the least- a fairly regular marriage up to now. That is, you are forced to imagine what it would mean to go through this very event if you were this very person or persons; not just any man and woman having a fight, but these two persons with their unique history, having this fight. This is why the not-showing morally deepens the event; it intimates the event's moral depth. How to understand this phenomenon? Evidently, this moral depth is not depicted, yet it is an effect of what is depicted. An account in terms of expression though, as suggestive as it may appear to be, seems to me to give way under the load of what it is supposed to explain. A dancing cup of coffee is shown and intimates the moral depth of an event in two person's lives. The cup can never achieve this in itself; it cannot possibly be expressive of such complexities, although it may express the force of the blow. Yet another possible answer might be given in terms of evocation: the whole sequence in the film, that which is shown, and that which isn't, evokes in us a response to our awareness that what we are witnessing is of great implication for these people. However, evocation would have to be explained in terms of our response to something comprehended, but it does not show how we got to comprehending it. It does not explain, in other words, why not-showing the event would be significantly different from showing it.

My thesis regards the merits and limitations of pictorial representation of experience. Thus, I am concerned with the aesthetic problem of representation. My thesis is twofold: first, the experiential, which is invested with the moral depth relating it to the life of the person having the experience, cannot be depicted, but, secondly, it can be represented. The way in which it can be represented, is by intimation.

We cannot fully explain the difference between depiction and intimation as long as we adhere to the exclusive disjunction between representation and expression, because this would leave us with the sole possibility for explaining artistic communication of the experiential in terms of expression, in terms, that is, of a reference relation somehow antipodal to the one of representation. Indeed, we cannot communicate the experiential by any `literal' depiction, but this does not preclude its being representable altogether. To dissipate the traditional separation of representation from expression I propose we distinguish three kinds of representation: telling (description of matters of fact), showing (depiction of visual aspects of the world), and intimation (occasioning associations which enable one to relive the involved experiential aspect). Before I elaborate on the issue of expression, there are some things to be said about the measure of conventionality of discursive and, respectively, pictorial representation.

3. PICTURES DO NOT STATE, THEY EXEMPLIFY.

It is of great importance to realize how depiction differs from assertive description: pictures cannot state. They cannot assert knowledge regarding matters of fact at all, even though they do show certain perceptual aspects of them. A picture never asserts: "this is the case, and this, not that": one single photograph would be conveying uncountably many facts. They are just as much overdetermining our descriptions as reality itself is, as I shall further argue below. Donald Davidson once put this rather clearly:

How many facts or propositions are conveyed by a photograph? ... Bad question. ... Words are the wrong currency to exchange for a picture. (Davidson, 1984, p. 263).

If I were to ask three persons to describe a certain situation, three different descriptions would be the result of their endeavours. If we would be trying to decide which of these descriptions is true we might come up with a photograph of the scene to compare the descriptions with. However, this could never function as a criterion, because the photograph merely repeats certain visual aspects of the situation, and these underdetermine the descriptions. Contrary to the descriptions, the photograph is not similarly underdetermined, but is instead determined fully by the situation it depicts. Therefore, it provides no alternative for an assertive description. This is illustrated by the fact that no pictures can deny that something is the case.

Evidently this relates to the distinction between telling and showing. Pictures show forth certain perceivable aspects of a situation. Because they do this by a reduction in dimensionality, specific conventions will be required to thus render some three-dimensional, moving event into a flat, still, photograph. However, this does not reduce pictorial representation to mere conventionality: some parts of what we see in the picture can be seen in the depicted scene as well. Questions concerning (epistemic) truth, as are quite normal for discursive assertions, are not to the point in a pictorial context. This difference between the pictorial and the discursive relates to the measure of thoroughness of the conventions involved. In section 5 I will return to this difference. So no picture states that in this scene X is the case rather than Y, although we can use a picture to provide by way of exemplification the material necessary for understanding and verifying a statement of such intent.[5] We also can arrange whatever is exemplified in a picture by introducing subtitles or interpretations, but this does not change the picture into a statement. Nor do pictures of themselves denote what they depict. They merely exemplify it.

To be sure, exemplification is what samples do: they are examples of whatever it is that explains your peculiar interest in it. For example, an interior decorator might show you some samples in order to suggest the textile that you might want your new curtains to be made of. We all know how a sample works, but how should we account for it? According to Nelson Goodman, who reassessed the notion of exemplification, in the end the sample refers to the curtain you want to have tailor-made, but it does so in a complex way.[6] In fact, Goodman thinks, exemplificatory reference is the inverse of denotation. Now denotation is the reference of labels to things. Exemplification, then, involves a symbol, the sample, which refers to certain labels, which denote certain properties possessed by the sample, and, in our example, by the curtain. Goodman thinks, however, that `properties' is merely short for `being denoted by some label', or `belonging to the extension class of certain labels'. Therefore, according to Goodman, a sample only exemplifies labels, not the properties that these labels denote. Exemplification also is conventional, in that it is the conventions which decide which of the sample's `properties' supposedly exemplify: the texture and the colour, but not the width, or the absolute weight of the sample. Now Goodman is certainly right in claiming that exemplification is not like denoting in that it does not directly refer to things, primarily, because the thing the sample is a sample of, does not exist yet. This doesn't make exemplification merely fictional, because even though the curtain does not exist yet, it will soon enough. To short circuit these complexities, then, Goodman elegantly suggests that we take the exemplificatory reference relation to be aimed at the label which denotes both the sample, and the curtain. For several reasons, however, this account is not satisfactory.

First, the conventions involved are strictly formal. They regulate which species of properties should be relevant, but they do not regulate the exact token instance of these properties. Even though we will know that only texture and colour are significant, this does not tell us which colour or texture the sample exhibits. Showing forth a specific shade of colour clearly cannot be reduced to convention: the sample itself will have to exhibit the properties. To say, secondly, that talk of properties is merely short for talk of extensions of labels is not going to help explain exemplification, because the exemplificatory reference supposedly starts with the sample, not with the labels denoting the sample's properties: it will be the properties of the sample which make one decide whether the sample is going to be of use to us. Thirdly, the elegance of Goodman's proposal is illusory. We can describe a sample in many different ways, but we won't normally have the exact terms at our disposal, even though we know fairly well how to pick the right sample, and use it. Instead, then, of first referring to hard-to-find labels, it is exactly the function of the sample to make the use of descriptions obsolete. A sample rather is a conglomerate of secondary qualities,[7] and that is exactly where its strength lies. Put differently: all labels on offer for a description of a sample will be underdetermined. This underdetermination with regard to exemplification will help provide an explanation of the remarks just made on behalf of the citation from Davidson. Fourthly, Goodman introduces his account in terms of a reversed reference to labels to sustain his rejection of similarity as a necessary condition of exemplification on the one hand, and on the other, to sustain his rigid conventionalism. However, it is precisely because of the underdetermination of the labels which supposedly describe what is being shown by the sample, that the similarity of the sample with the curtain is going to be of paramount importance, instead of obsolete. Goodman is right in that the principles of exemplification will be relative to the context in which a sample is used, but he is wrong about the conventionalism involved in the proper meaning of some singular exemplification, which instead stands in need of a naturalistic account. Exemplification, then, is not a case of a reversed reference to a label.[8] A sample's functionality depends, instead, on a projected recurrence of properties: the properties in the sample are supposed to be the same as, and not merely similar to, those of the curtain. The conventionality involved, then, is relative to property species, not to specific properties: to colour in general, for example, and not to a specific colour. Exemplificatory functionality at large is subject to conventions, but the question of exactly what is exemplified in a concrete instance must be accounted for naturalistically, in terms of a recurrence of properties.

I propose we understand representation as a kind of exemplification.[9] Now, the idea that the exemplified object exhibits the same properties that we find in the sample is an anticipatory claim, a projection based on the general significance of the referential medium of exemplification. The same goes for depiction: we anticipate that the properties exhibited in the picture will also be found in the depicted subject as it may exist in the real world. In short, a picture's representational properties are those properties which are projected onto the depicted object. The difference between exemplification and depiction then lies in the ontological status of the object referred to. With depiction the object may be fictional, with a sample the fictionality of its `object' would make the exemplification dysfunctional. The possible fictionality of a represented object, however, does not change the representational functionality. Because pictures at first merely suggest to refer to some real event, the ontological problem of fictional entities evaporates. Indeed, of its essence a representation is a case of exemplification, it does not denote, nor state matters of fact.[10] Establishment of the reality of a picture's referent lies beyond the reach of the picture itself. The crucial part of what makes a thing a representation, then, lies in the exemplificatory recurrence involved, which sometimes is merely anticipated; and to understand this we need a naturalistic account.

Another major contribution of Goodman lies in his thesis that what he calls the `syntactical' properties of the pictorial `system' are lacking in distinct characters.[11] If sometimes we appear to be able to distinguish such characters (the pictorial symbol of a house has this and that parts) it will become clear soon enough that there is no possibility whatsoever to differentiate the `characters' involved in a finite way, which we are able to do in the case of letters and words.[12] In Goodman's words, the pictorial is syntactically dense and relatively replete; the discursive is syntactically differentiated and not replete. This is consequential for the differences between the semantic peculiarities of discourse and those of the pictorial. Ambiguity, and semantic density appear to be paradigmatic of discourse, but they are not so with regard to pictures. On the contrary: a photograph of my mother not only unambiguously depicts my mother, it does so exclusively: this photograph depicts no other person. Not that every picture will represent its object unambiguously, but, ideally the pictorial has this potency, whereas the discursive does not. Description will always be underdetermined by its subject, whereas pictorial representation will be as overdetermining as its subject matter is.[13] We can and should explain this naturalistically, by taking depiction as a form of exemplification. In itself a pictorial representation - if referring to real events at all - refers to one singular event, by (anticipated) recurrence of its perceptual aspects, and does not state anything about the event.[14] Our thoughts about the event depicted in a photograph, then, derive from without the picture.

4. REAL LIFE EXPRESSION AND EXPERIENCE.

Representation, then, depends on noticed (anticipated) recurrence. How does this relate to intimation? The crucial difference between pictorial and intimated representation is that in the case of depiction the spectator's activity is one of noticing what is there in the picture, whereas in the case of intimation the spectator's activity lies in spontaneously supplying the associations needed to fill in what is not there to be noticed. To the question of how these associations come about, and where we should look for their standards of correctness, the most illustrative answer provided by analytic aestheticians is in terms of expression.[15] The notion of artistic expression apparently is based upon real life expression, and the representation of a face expressing some state of mind illustrates this point.

First, there is an obvious difference between the `literal' expression of a real life face, and the `expression' on a represented face: the success of the latter is far more subject to conventions than the former is. People endowed with normal cognitive powers will have a natural propensity to perceive things and their natural properties when confronted with them. Apparently they also have a propensity to look through the merely perceptually available outlook of a person, straight into the mental life expressed in it. We do not require specific interpretative activities for such recognition. I am not implying that we always fully recognize a person's feelings from his gestural and facial expressions, but in principle we find little difficulties in it. Lastly, real life expression is directly and causally related to the mental life in question, which means that whatever emotional state we ascribe to a person his subsequent actions will reveal whether we were right and to what measure.[16] However this is, the experiential awareness one has of the mental event one is in, is irreducible to one's outlook, and cannot therefore be fully represented by repeating such an outlook. This is not merely a consequence of the fact that the experiential awareness of one's mental life would lie hidden `behind' its visual symptoms, but also of this awareness being bound up with a moral depth embedded in a complex personal history.

Now then, if depiction equals exemplification, and if this implies that only a situation's visual aspects recur in a picture, then this would dismiss the capacity of the pictorial to represent an event's experiential aspect. The ease involved in mere depiction will be at odds with the gravity and moral depth of the experience involved. This is, first, because in real life we are being served with more clues which are causally connected to the experience; secondly, because in real life we are less tended to think that some singular `image' will settle the question as to the exact nature of the experience involved, i.e. the causal and contextual complexities of the experience form part of our everyday recognitions. In representation, however, because of the absence of these complexities, ellipsis, narrative allusion, and discursive information will have to fill the gap. These devices, however, are not pictorial, and for an understanding of just how they fill in the gap more than a keen analysis of the merely pictorial is required.

To depict a crying face is merely one way to direct our associations, it certainly does not automatically depict the grievous experience in full depth.[17] We know that cubist painters have tried to fill in the gap by providing more perspectives on a face or event, and so did cinema in furnishing moving images, but to little avail, as it is not just a matter of quantity of information, but of how to induce the spectator to empathically fill in the emotion. Instead, the problem of the representation of experience is a matter of the individuality of the represented. To convey this individuality only empathy might be successful. It is the associations of the spectator that must come to the aid of the visual if the representation is successfully to encompass an experiential aspect. By way of these personal associations the spectator fills in the gaps in the visual with his own experiences, past or present. This is to say that the experiential aspect can only be intimated. Cinematic montage does provide an answer, as we saw already in the example with the coffee cup taken from Bresson.

How exactly should we understand the hiddenness of an individual's experience? For the most part there is a difference between real life expression and a representation of an experience, because the person experiencing the state nevertheless is somehow in a privileged position. This thesis may appear to collapse under the weight of its presupposing that I know better than anyone else what it is that I am experiencing. But that is not my point. I agree that my understanding of my own experience to a large extent converges with insights provided by other people: their insights in why I am sad, and even their acknowledgement of the fact that I am sad might even outdo mine at times. On top, my own insights derive mostly from what I have learned to think about mental states from other people's actions. So I am not stating that the person having the experience is better equipped in a cognitive manner. However, he does possess a privilege deriving from his having the experience and from his, again, experiential acquaintance with his own individual personal history. Evidently, to understand an experience is a far cry from having it.[18] And to have it is an act more solitary than might be concluded from its public accessibility. To represent experience then is more difficult than the existence of real life expression might induce us to think. In fact, representation of the experiential depends on the success of making the spectator associate with what is depicted specific memories of feelings and events, so that he can revivify the represented experiential aspect within his own mental life. This subjective supplement is the exemplificatory recurrence requisite for the intimation of the experiential to be a case of representation.[19]

5. INTIMATION AND CONVENTION.

Discursive reference to things and events involves a symbol system that is highly conventional and possesses a syntax of characters that can be differentiated in finite manner: letters, words, and sentences. The relation between a word and its object is arbitrary; it does not involve similarity (not normally at least; onomatopoeias are not paradigmatic), nor recurrence. The reference works its way due to the semantics related to the characters used; a functionality based exclusively on convention. We must distinguish between two aspects of this conventionality. First, the discursive medium at large is conventional, amounting to something like `with things of these types (words, sentences) we can refer to things of those types (things, events, et cetera)'. Then there is a more specific conventionality relating singular words and sentences, to specific kinds of things and events. Now it is because discourse is conventional in both ways that it takes such great pains to learn a language: we must learn from each single word what it stands for. And it is this thorough conventionality which generates the question of truth to the facts, which generates, in other words, discourse's potency to state matters of fact in ways underdetermined by the data.[20] Contrary to this a pictorial representation does presuppose regularities between the symbol (if that is what pictures are) and what it stands for. Certain similarities between some surface parts of the picture and some surface parts of the depicted found the possibility of the pictorial reference relation. We have already analyzed this in terms of exemplification, and exemplification is conventional only as a medium, not in its token singular instances. To sum up, pictorial conventions relate to the medium only, they do not pertain to the token-instances, singular pictures. This explains why we have less trouble in learning to decipher pictorial representations: once we have grasped one, we will know how to grasp most others.[21] With regard to the measure of conventionality involved intimation resembles discourse more than depiction.

Moreover, there is an asymmetry between depiction and intimation, in our recognition of what is represented. Recognition of a merely visual aspect equals experiential awareness, which, clearly, is what makes them secondary qualities.[22] Here, perception equals experience. However, to recognize someone expressing a mental state does not automatically involve an experiential recurrence of this person's mental life. As a consequence, representational intimation of an experience will not take place automatically whenever its visual expression recurs pictorially, and such intimation, therefore, will be subject to token-conventionality also. In order to draw the spectator towards a re-experiencing of a mental event a representation must lure the spectator into providing spontaneously a subsidiary subjective activity on top of his perceptions. These token-conventions, however, because they induce the shown to occasion something which is not shown, appear to be of a `negative', dependent kind: they would rule over when not to depict, i.e., over where to leave the open spots in the picture so that the spectator can fill them in empathically with his own associations.[23] Because intimation of the experiential originates in not-showing, it can only be conveyed by some significant transgression of available conventions: there are no direct ways available. Thus intimation presupposes and motivates creativity.[24]

Above, we have been meaning to oppose `pictorial' intimation with pictorial showing, which we took as kinds of recurrence of experiential and, respectively, perceptual aspects. We have described this opposition in terms of the difference between the respective conventions. In discursive language, however, due to its thorough conventionality, we find a continuity between the paradigmatic use of the medium, which is statement, and the non-stating involved in discursive intimation. This seems to suggest that discursive representation of an experience does not involve any special problems. The analogy with the pictorial, however, points in a different direction, namely, that intimation in discourse is achieved by metaphorical use of terms. I cannot follow up on this in the space allotted to this paper. Instead I will offer one last thought regarding the criteria of success germane to intimation.

6. INTIMATION AND AESTHETIC EXCELLENCE.

We must distinguish between the psychology of evocation and the aesthetics of intimation. It is common knowledge that with respect to evocation the mental aspects involved in a work often diverge from our response to them: the anger or sadness of the main character in a scene in a film may arouse a feeling of pity. The question, however, whether this scene also succeeds in intimating the experiential aspect of the main character's anger or sadness is a different question altogether: it is a question about representation, not one of psychology. So for an appropriate ascription of intimation to a work we must distinguish whether our experience is merely a psychological reaction to the represented, or whether it is some recurrence of this experience.

Secondly, a revivification of some experiential aspect relates to aesthetic evaluation in a very substantial manner, as is illustrated by the citation from Kant with which this paper started. Kant describes aesthetic ideas as the addition of aesthetic attributes to what logically belongs to a certain concept. (I would say: an addition to what is literally described or shown). He ascribes to this addition the functionality of animating the mind. Re-experiencing a mental life evidently animates the free play of one's mental faculties. If this free play provides us with an adequate understanding of certain aesthetic evaluations, as I think it does, intimating efficacy will be an essential part of it.

Lastly, we must ask if experiencing an emotion can form a sufficient justification for attributing this emotion to the work's intimation. Evidently, it is part of my thesis that for such attribution experiencing the emotion is a necessary condition: the only way for us to really know if a work has represented some experience is by going through the experience. On top of this, however, we must also believe that it was the work that made us go through this experience, and we must be right about this. To establish this latter rightness it will not do to take issue with some spectator's claim that he did not have the specific experience. Instead we must try to find out whether the ontological implications are viewed correctly. It is a consequence of the point of view defended in this paper that standards regarding this issue are an ideal at best, because the experiential aspect is being rebuilt within the mental life of the individual spectator by way of his own associations. These associations will be guided by what is present -shown, or told - in the representation, but these discursive and pictorial parts in themselves do not provide the decisive criteria for the way in which we filled in the supposedly absent experience. Therefore, with regard to the intimation of the experiential aspect we must be non-realist, because the very success, and even existence, of intimation depends upon our personal associations. Now this non-realism regarding intimation may be problematic from the point of view of universal validity claims. From the point of view of the individual and the way he lives his life it appears not to be defective at all. Intimation makes us compare our own past experiences with a new experience based upon these past ones but nevertheless not reducible to them. The impact of the intimated associations provides the explanation of the impressing grip which a representation of people may have on us. Here the ethical import of (representational) art is at its deepest, and to ask for the truth of our ascriptions may be inappropriate for that very reason.

© Rob van Gerwen