We are now in a position to see the integral relationship between expression and representation, and thereby, the inadequacy of their traditional separation. The traditional view assumes that an object can be perceived and described just as it is in itself independently of any human interaction with it. As a result, any human interaction that does creep into the description is said to be biased and therefore ought to be removed in the name of objectivity. On the other hand, the traditional view insists that emotion can also exist by itself independently of objective reality. This leads to the idea of a pure abstract emotionalism in art without recourse to objective representation – an ideal strengthened by the example of music.

Both assumptions, we can now see, are mistaken. You can’t have a description of objects without some human interaction and hence subjective interest; that is, there is no representation without emotional expression, As Richter and his friends discovered, each of them saw and portrayed the scene differently, depending in part on their individual subjective outlooks. The melancholy painter, for instance, "straightened the exuberant contours and emphasized the blue tinges." Sartre also tried in his early fiction, such as Nausea, to remove all subjective bias and describe the "naked Reality." But as Robbe–Grillet points out, Sartre fails to remove a tragic, defiant, romantic attitude toward this brute material reality, as is clear from the following description of a city park in Nausea,

I sank down on the bench, stupefied, stunned by this profusion of beings without origin: everywhere blossoming, hatching out.... It was repugnant ... grotesque, headstrong, gigantic....

For Robbe–Grillet, this tragic sense of meaninglessness is just as biased as the ordinary sense of a world of meaningful objects understood in the usual way. "The world is neither meaningful nor meaningless. It just is!" But Robbe–Grillet suffers the same fate as Sartre when he tries to describe things just as they are in themselves. In all his descriptions of things he projects a cold, hard–edged, geometrical, purely visual aloofness, as in the following description (which tells us as much about Robbe–Grillet as it does about the pier).

The stone rim – an oblique, sharp edge formed by two intersecting perpendicular planes: the vertical embankment perpendicular to the quay and the ramp leading to the top of the pier – was continued along its upper side at the top of the pier by a horizontal line extending straight toward the quay. (Le Voyeur)

Similarly, there is no pure emotion outside a humanly interpreted environment of objective, public entities, as we saw in the last chapter in our discussion of Hegel, Rilke, and Eliot’s "objective correlative."

Yet historically, objective reality and human emotion were considered two independent artistic goals to be achieved. These two impossible but perennial dreams have dominated the history of art and ideas. Fortunately, artists do not always practice what they preach. Leonardo says he is merely holding up a mirror to nature, Zola, that he is simply observing and recording the facts. Fortunately for us, what they actually did was something different and more than they promised. Abstract Expressionists claimed a totally nonrepresentational expression of emotion, but in fact the emotional effects depend in every case on the emotional physiognomy of features of objects outside of art. It is true that by concentrating on the general features of objects, such as shape and color, their art is abstract and nonfigurative, but it is not and cannot be totally nonrepresentational. In each case the art is superb, though the idea of it is exaggerated and confused.

Now we can begin to see why. The representational content is internal to the artwork by its dependence on the particular way in which it is done in the specific artwork. But this "way" or "manner" is precisely how our human interest in the object (called "emotion") is expressed. You must choose one way of doing something rather than some other way, and this choice will inevitably involve some human interpretation. In this sense everything we do expresses our attitude, or stance toward the world – the way we dress, the way we walk, or furnish our apartment. As an experiment it is interesting to try doing something or describing something in a completely neutral way. Most people find it simply can’t be done. Our action or description need not be an "emotional" interpretation in the ordinary sense of the word, limited to extreme states of mind like fear, sadness, etc., but only in the sense that it conveys how you look at the object. And this in turn, coming full circle, modifies the object. There are not then two things, an object to which an emotion is added, but only one, the object–as–seen–in–this–light. You can’t really separate the two, you can only talk about them separately; there are not two things but only two ways of speaking about the same thing. Representation is one of the elements that makes up the total context and this context is the creation of a new human interpretation or point of view (called "emotion" in the traditional aesthetic jargon).

If we now understand the "form" of a work of art as this total contextual arrangement of elements, we will be able to extend our analysis to include form as a part of an indissoluble union with representation and expression. Just as you cannot separate the experience of things that are organized from their organization, so you cannot separate form from representational and expressive elements organized in a work of art. But this organized whole is precisely what we mean, or ought to mean, by the "form" of a work of art. Expression and representation are in this sense simply two of the elements integrated within the artwork.

But "formalism" in aesthetics has not always been understood in this way, and indeed "form" has been used to support many different theories of art. Formalism was originally motivated by the attempt to put the emphasis squarely on the artwork itself (autonomy), rather than on anything outside it (heteronomy). This primary sense of form we might call the "principle of autonomy," and as such it is no stranger to us. In discussing the problem of style and imitation in Chapter Two, the inadequacy of the concept of imitation which loomed largest was its tendency to subordinate aesthetic to non–aesthetic standards. In particular, it appeared to establish the ordinary objects of the everyday world, apprehended in the usual way, as the models both for art and for non–art. What the concept of imitation seemed to deny, in other words, was any domain peculiar to aesthetics, any standards, features or virtues uniquely applicable to works of art. Similarly, for all its admitted differences, the same problem haunts the theory of expression – the subordination of aesthetic criteria to extra–aesthetic considerations, in this case, the human psyche. The late nineteenth and early twentieth century, however, witnessed a reaction to this domination by non–aesthetic standards, and an important liberation movement of sorts was begun under the banner, "art for art’s sake." Tracing its roots back to the original conception of "aesthetic" in the eighteenth century as an autonomous domain, formalism nonetheless came into prominence in support of the abstract art of the Post Impressionist period early in the twentieth century and nonobjective art in the United States during the Abstract Expressionist period of the 1940s and 1950s. The leaders of this school of thought have tried to isolate that which differentiates our enjoyment and understanding of works of art from everything else. "Form," in the beginning, was simply the name for the intended object of this search. Modern formalism, then, is based on the principle that whatever is proper to ascribe to works of art should be something which is not ascribed to anything else, and whatever does have some extraneous reference outside of art is not a legitimate response to a work of art – what Roger Fry called an impurity irrelevant to "pure" art...