If I say of a piece of Schubert’s that it is melancholy, that is like giving it a face. —Ludwig Wittgenstein

Expression is the suffering countenance of artworks. —Theodor Adorno

To experience the aura of a phenomenon means to invest it with the capability of returning the gaze. —Walter Benjamin

Aesthetic judgments and ethical choices take place as face-to-face experiences. —Susan Stewart1

Wittgenstein was clearly fascinated by faces and face perception. But he seems to have regarded the face not only as something important to think about, but also, one might say, something profoundly useful to think with. For Wittgenstein, that is, the face was not only a special object of philosophical inquiry, but also a concept that functioned in a metaphilosophical register, having to do with philosophical methodology and intellectual vision: what we might call physiognomic vision. Taking to heart Wittgenstein’s dual interest in the face as both privileged object and methodological guide, I argue in this essay that Wittgenstein’s numerous remarks about faces and face perception hold the key to a better understanding of the common, but still theoretically puzzling, phenomenon of aesthetic expression. Moreover, I suggest that by seeing the connections between aesthetic perception and the way we perceive faces, we can better appreciate the deeper stakes of ongoing theoretical disputes about the concept of aesthetic expression: especially debates about whether the expressive qualities of artworks are real or merely due to the projections of aesthetic beholders. What’s ultimately at stake in such disputes, I will suggest, is the proper acknowledgement (or denial) of the expressiveness of the human body. Philosophical debates about the expressivity of artworks, that is, serve as proxies for debates about the ontologies—and, in particular, the expressive nature—of human beings as such.

I will begin with a brief characterization of the theoretical impasses that trouble current discussions and debates about aesthetic expression (among both philosophers of art and literary theorists), before turning to an exploration of what Wittgenstein’s remarks about the human face can help us to see in the “faces” or “physiognomies” of expressive works of art.

I. Expression and Art

Understanding what it means to say that aesthetic artifacts—like poems, paintings, or melodies—can, or for that matter cannot, be “expressive,” especially expressive of human emotions, has long been a vexing theoretical problem for the humanities. There are at least two major problems, or puzzles, that continue to trouble our understanding of the very idea of aesthetic expression. As will become clear, the two problems I have in mind are quite different, and indeed, many will think them diametrically opposed, and so I doubt that any one person will find both equally compelling. Yet between the two, I think they capture a good deal of why we find ourselves still having to ask fundamental questions about the nature and even, at times, the very possibility, of aesthetic expression.

The first of the two problems is one that will trouble only those who accept that what we call aesthetic expression is “real”: that artworks do, in fact, express what we call feelings, affects, or emotions. The central problem faced by such thinkers, who assume aesthetic expression to be real, is to account for how it could possibly work: so I will call this the “how does it work?” problem. When a work of art involves an actual living human being, as in dance or song, the problem seems relatively simple, because the expressiveness of the performer’s body would seem naturally to underwrite the expressiveness of whatever artistic composition is being performed. Whether that intuition is correct or not (and, of course, it’s a contentious one) analytic philosophers of art who have tried to tackle the “how does it work?” problem have usually assumed it to be right, and have therefore set their sights on what they consider the much harder part of the problem of explaining aesthetic expression: explaining, that is, what they consider the secondary or derivative expressiveness of self-standing artworks (like paintings, poems, or symphonies), which do not benefit from the central presence of a naturally expressive, living human body. Moreover, those philosophers (like Jerrold Levinson, Stephen Davies, and Peter Kivy) who have worked especially hard on this particular problem, have focused on art forms that are non-verbal and non-figurative, and so, somewhat unsurprisingly, instrumental (or absolute) music has become for them a privileged object of philosophical inquiry. 2 The understandable assumption is that if we can explain the emotional expressiveness of music that does not have lyrics or in any way feature the human voice or body, then explaining it elsewhere will be a relative piece of cake.

But anyone who has even dipped into this literature will know that agreement about the solution to this harder problem has been elusive, to say the least. One popular type of solution, which Stephen Davies calls “appearance emotionalism” posits a resemblance between musical properties (like rhythm or timbre) and the appearance of emotions as they are expressed by actual human bodies. Davies describes the link like this: “…music is expressive in recalling the gait, attitude, air, carriage, posture, and comportment of the human body. Just as someone who is stooped over, dragging, faltering, subdued, and slow in his or her movements cuts a sad figure, so music that is slow, quiet, with heavy or thick harmonic bass features, with underlying patterns of unresolved tension, with dark timbres, and a recurrently downward impetus sounds sad.” 3 Notice, in this typical formulation, the strict focus on immediately perceptible surfaces or appearances, both those of the music itself and also of the bodily behaviors such musical patterns are said to resemble.

To understand the significance of this characteristic emphasis on aesthetic surfaces by recent philosophers of art, one needs to know something about the failed history of prior attempts to explain the expressiveness of artworks, attempts which often tried to plumb the inner depths of artists’ minds. Earlier expression-theorists, like Tolstoy and Croce, had assumed that the emotions expressed by a work of art ultimately refer to emotions actually felt by (and so within) the creative artist, which could be externalized and then transmitted via the artwork to another person. As Tolstoy put it, in What is Art?: “Art is a human activity consisting in this, that one man consciously, by means of certain external signs, hands on to others feelings he has lived through, and that other people are infected by these feelings and also experience them.”4

There are, of course, numerous difficulties with a position like this, including the logical problem that the argument would seem to commit a genetic fallacy in positing a causal link between the appearance of the finished artwork and the mental state of the artist said to have motivated its creation. Nevertheless, Tolstoy’s expressivist position remains as popular as it is conceptually problematic, and similar difficulties have troubled the numerous subsequent attempts (such as by Croce and Collingwood) to somehow ground the emotional meaning of works of art in the mental states of their creators.

It is in reaction to these earlier failed attempts to reach beneath the surface of a work of art in order to explain aesthetic expression that more recent philosophers of art (like Davies, Levinson, and Kivy) have turned their attention squarely to that aesthetic surface itself, looking at the resemblances between the appearance of artworks and the appearance of human emotions that I mentioned a bit earlier. But those more recent attempts, while apparently more promising than Romantic quests into the deep subjectivity of the artist, have had conceptual difficulties of their own. Their strength lies, as I suggested, in the way they methodically restrict their investigations to the surface features of aesthetic objects. However, here also is the very source of their greatest weakness, since they regard that aesthetic surface as intrinsically expressionless. The emotional expressiveness of absolute music, for nearly all the major contemporary philosophers of art, is not intrinsic but secondary or derivative to the primary expressiveness of actual human beings. For, as Davies puts it, “music is nonsentient, and only sentient creatures can be literally sad or happy.” 5 So, as Peter Kivy has suggested, listeners must somehow “animate” the sounds in a certain manner, giving musical sound-patterns an expressive life they themselves do not (or rather cannot) possess. 6 Given the basically “projective” picture of the relationship between aesthetic surface and expressive import that is operative in most current philosophical debates about this aesthetic issue, we can see why various philosophers have had trouble convincing each other that they have found the one, true account of how it is we actually perceive the expressiveness of a work of art. If projection of some sort does indeed play a central role in the perception of aesthetic expression, disagreements about the accuracy of various accounts of such perception will necessarily remain irresolvable. Once we admit that there’s no “fact of the matter” to the phenomenon of aesthetic expression, then who’s to say whose aesthetic projection is the right one? And so, even for those who are convinced that the experience of aesthetic expression is real, how we can justifiably call it so remains very much an open question.

Now let me turn briefly to the second of the two major problems faced by the concept of aesthetic expression. Unlike the first (“how does it work?”) problem, which has more powerfully gripped aesthetic theorists who work in departments of philosophy, this second problem will, I think, be much more familiar to those who work in departments of literature, where it has, in a variety of ways, exercised a profound influence on literary theory and criticism over the past few decades. This second way of regarding aesthetic expression as theoretically problematic does not (unlike the first) take expression’s reality at face value, and then try to explain how it works. On the contrary, it “deconstructs” or otherwise dismantles the very picture of the human “subject” that is taken to underlie all traditional expression-theories (like those of Tolstoy) and so radically alters what the very concept of aesthetic expression could possibly refer to. If, after the poststructuralist “death of the subject,” there is no longer any unified inner self, and so, in turn, no unified field of cognitive or affective experience, then of course this will have profound consequences for the meaning, indeed the very possibility, of expression in the various arts. The problem now is to explain what aesthetic expression could possibly mean once the supposedly expressive subject has been philosophically deconstructed, radically de-centered, or simply reduced to a social construct. So I call this the “no-subject” problem for the concept of aesthetic expression: without the concept of subjectivity, whither the concept of aesthetic expression?

However, the ultimate theoretical and critical consequences of the so-called “death” of the expressive subject are not obvious, and have not, in fact, played out in any single, predictable disciplinary direction. For some literary critics, it has simply meant that what we traditionally call “expression” is, in truth, unreal, and the more philosophically inclined among them (like Paul de Man) have thus tended to regard the experience of expression as a phenomenal illusion, as an unstable “effect” (as subjectivity itself is now understood to be an “effect”) of the anonymous play of language. For others, the philosophical death of the subject has opened the entire field of aesthetic expression to powerful forms of ideology critique: for if there is, in fact, no subject that can “express” itself in art, then it makes sense to ask what ideological commitments might motivate the continuing investment of particular artists, not to mention our culture as a whole, in that (now unmasked) idea? The theoretical critiques leveled by Language Poets (and their critical allies) against the expressive “I” (or voice) of traditional lyric poetry is a good example of the now widespread approach to self-expression I’m thinking of.7

But then again, there has also been an entirely different conclusion about the fate of expression that has sometimes been drawn from the dismantling of the Cartesian subject: that far from signaling the end of emotion, not to mention emotional expression in art, the deconstruction of the Cartesian “I” means, in fact, the radical liberation of emotion or affect from traditional (and restrictive) philosophical and psychological categories. As philosophers like Gilles Deleuze and literary theorists like Rei Terada have each, in his or her own way, suggested, perhaps the death of the subject only means the end of certain mistaken ways of thinking about affective experience: now, that is, we can see that emotions and all the various affects are not mere qualities or properties of bounded individual subjects, but have a robust and free-floating reality (even force) of their very own.8 So the affective intensities of Olivier Messiaen’s music or Francis Bacon’s paintings should not in any way be regarded as “derivative” or somehow “secondary” to the emotions of actual human beings, since of course, on this view, human beings themselves no longer “possess” emotional experiences in any literally “primary” sense. But these thinkers, who would unmoor (or as is it is sometimes put, “deterritorialize”) the emotions from the psychological experiences of human individuals, have yet to explain something that even Nietzsche could not quite work out: how to justify the use of concepts (like “the will,” or in this case the “emotions”) that have their home in the way we ordinarily talk and think about persons, while jettisoning, or radically decentering, the concept of the person altogether. Though I have my doubts, this may of course be possible, but it remains to be done, and so, once again, as with the “how does it work?” problem, we are left with more questions than answers when it comes to the role of emotional expression in the arts.

II. Expression, Art, and Faces

That’s where Wittgenstein comes in. Wittgenstein attributed profound philosophical importance to the concept of expression, and his later writings are filled with probing reflections on how properly to understand expression’s pervasive role in human life and culture. He clearly recognized that such understanding was not easy to attain, for a great many of his remarks about expression concern common ways we tend to misunderstand how it works, and thus to misunderstand what it means to do such ordinary, everyday things as express pain, use words, recognize facial expressions, or express emotions by means of bodily gestures. If, as Wittgenstein everywhere suggests, there is a deep difficulty in understanding such apparently simple things as what it means to say “ouch,” or to curse someone with a rude flick of one’s hand, then is it any wonder we have had so much difficulty understanding the expressiveness of artworks, which are certainly no less complex than the complex beings who make them.

Deeply concerned as he was with the psychological concept of expression, Wittgenstein’s later writings have a great deal to contribute to present-day discussions of aesthetic expression. So I would like to consider what Wittgenstein can show us about the expressiveness of aesthetic objects, like poems, paintings, and melodies. As I said at the outset, however, I will do so by focusing on his numerous remarks about faces, face perception, and physiognomy. For it is there that his most important insights for contemporary aesthetics are to be found, rather than in his very few explicit remarks about art and aesthetics, a subject about which he said famously little.

Anyone who has read widely in Wittgenstein’s later writings will already know how intrigued he was by conceptual questions raised by the way we see and understand human faces. Not only did he spend a great deal of time thinking about face perception itself (in a few hundred remarks scattered throughout the later manuscripts), but significantly for us, he seems also to have been fascinated by how the way we understand the human face might be conceptually related to how we understand other meaningful phenomena, which he often tellingly spoke of as possessing faces or physiognomies of their own. As he remarked in Part II of the Investigations, for example, a familiar word can often strike us as possessing a “face [Gesicht],” as if the word had become a “likeness” of its meaning (PPF §294).9 And he recognized faces in a wide variety of other cultural phenomena, too, from the game of chess to the musical compositions of Bruckner, Haydn, and Schubert. During a discussion of the rules of chess in the Investigations, Wittgenstein even suggested, quite simply, that “meaning” is “a physiognomy” (PI §568), and he once also remarked that he found the “face” of Bruckner’s music rounder and fuller than the long and narrow one he associated with Haydn’s.10 This physiognomic way of thinking about aesthetic phenomena may explain why Wittgenstein would at one point liken the lack of a musical ear to the inability to recognize the emotional expressions of a human face.11For if understanding a work of music is somehow like understanding a face’s expressions, then being deaf to the expressiveness of one may indeed have something to do with being blind to the expressions of the other.

This criss-crossing of concepts and ideas between traditionally distinct regions of philosophy—here between the philosophy of mind and the philosophy of art—is of course characteristic of Wittgenstein’s approach to philosophical problems. By seeing (and drawing) connections between apparently disparate phenomena (between faces and artworks, for instance), Wittgenstein was often able to perceive entirely new ways to understand, and thus sometimes to resolve, difficult philosophical questions, including those raised by the expressiveness of melodies or lyric poems.

Though it might seem highly peculiar (and even off-putting) to some that Wittgenstein would think of aesthetic phenomena as possessing faces or physiognomies, he was, in fact, far from alone among modern theorists of the arts in making recourse to talk about the human face in order to describe and understand the meaning and expressiveness of aesthetic objects. It is, in fact, rather striking how many philosophers and theorists have turned—if only in passing, but often at key moments in their writings—to metaphors associated with faces and face perception when grappling with the complex phenomenology of artworks. The epigraphs to this essay offer three representative examples, from Theodor Adorno, Walter Benjamin, and Susan Stewart, but I could also have cited remarks by Béla Balázs, Allen Grossman, Paul de Man, Emmanuel Levinas, Gilles Deleuze, and Michael Fried (among many others).12 It is as if, when thinking deeply about the meaning of a work of art, theorists like these can sometimes do no better than describe the experience of looking into another’s face and seeing that other face looking back.

I want to suggest that what underlies this surprisingly common connection between artworks and faces is a widely shared (if usually non-explicit) intuition that the expressiveness of aesthetic objects is in some way conceptually related to our understanding of the expressiveness of the human body. And furthermore, I want to suggest that the way we understand the expressiveness of one can significantly determine our understanding of the expressiveness of the other. There is, so to speak, an internal relation between our theories of psychological expression and our theories of aesthetic expression. If we therefore want to hold on to the thought that art is, in fact, expressive, a great deal will depend on how we understand the expressiveness of the human figure. Yet at the same time, if we are convinced that art is not expressive (for whatever theoretical reasons), then that may, in turn, influence the way we see the human body itself: perhaps draining not only works of art, but the human body too, of their expressive powers. This, at least, suggests one reason why debates among critics and artists about the place of emotional expression in the various arts can themselves often feel so emotionally charged. Though ostensibly such discussions are “merely” about art, we can now see how they might also function as proxy debates about our own ontologies as human beings. For to ask about the expressiveness of art is to ask about nothing less, I would suggest, than the expressiveness of ourselves as embodied sentient beings.

This underlying link between the ontology of artworks and human beings explains why, as Michael Fischer has noted, analogies between literary texts and human bodies featured so prominently in the debates that swept through U.S. literature departments in the late 1960’s, following the advent of poststructuralist theory.13 For “rear-guard” inheritors of the phenomenological and New Critical traditions, like Georges Poulet and Murray Krieger, the literary text was understood to “benefit from the intelligibility of the mind, that is, the mind’s capacity to express itself… in words [and] gestures.”14 For their avant-garde rivals, like Paul de Man and J. Hillis Miller, however, “Instead of a living body infused with meaning and consciousness, the text” was more often regarded as “a hollow shell or [even a] corpse,” to invoke two very popular metaphors for the literary work, which came to be seen as intrinsically “lifeless.”15 And as I suggested in the previous section, this new attitude towards the text would, in turn, rebound upon the human body itself, transforming it, for de Man and for many others, from a primary site of expressive meaningfulness into just one more inscrutable—endlessly interpretable—text among others. As de Man put it in Blindness and Insight, even the simplest of utterances necessarily requires unending interpretation: “The simplest of wishes cannot express itself without hiding behind a screen of language…. The interpretation of everyday language is [therefore] a Sisyphean task, a task without end and without progress.”16

It is hardly accidental, then, that poststructuralist theorists like de Man would take a special interest in the concept of the face, since the face is commonly regarded as a paradigmatically expressive surface, the most expressive surface on the human body. But rather than see the expressiveness of the human face as mirroring the expressiveness of aesthetic objects (like Wittgenstein), de Man claimed that the experience of the face is actually but an epiphenomenon of the anonymous (which is to say, non-human) play of language. As de Man put it in “Autobiography as Defacement”: “Man can address and face other men… because he has a face, but he has a face only because he partakes of a mode of discourse that is neither entirely natural nor entirely human.”17 And in his essay on Shelley’s The Triumph of Life, de Man went on to claim that the experience of the face is inherently unstable, the result of a dizzying rhetorical process of face-giving and defacement that he famously called “prosopopoiea.” For de Man, there is no end to the figurative emergence and eventual disfiguration of the face, for it is part of the very “madness of words,” and “no degree of knowledge” can stop it. Incurable, the only therapy possible for this condition is the ironic insight that there is no point celebrating or denouncing it, since we human subjects are, in fact, “its product rather than its agent.”18 So for de Man, texts clearly cannot benefit from the primary expressiveness of actual human beings, since even that so-called “primary” form of expressiveness turns out to be a derivative, and linguistically produced, phenomenal illusion. What is truly primary is not the expressive human body, but language, and language understood as not even “entirely human.”

That de Man would recognize the importance of the face to the explanation of crucial issues in the philosophy of art (like that of expression) only further confirms the broader importance I have been attributing to the concept of the face for modern aesthetics in general. However, that he would take a position on the significance of the face so radically different from that suggested by a thinker like Wittgenstein also suggests something equally important: that the status (or ultimate meaning) of the face in, and for, modern aesthetics is as highly problematic and contestable as it is apparently philosophically crucial. This is surely so because of course it will be hard to know what the significance of the face could (or should) be for the philosophy of art, when we late- and post- modern individuals have so much trouble being sure about the meaning (or meaningfulness) of the flesh-and-blood human face itself.

Consider, for instance, the following questions about the face, which were posed by J.M. Coetzee in his foreword to Cecile Pineda’s debut novel, Face, which tells the story of a man with a horribly disfigured countenance:

What is this thing, this structure of skin and bone and gristle and muscle, that we are condemned to carry around with us wherever we go? Where does it begin, where does it end? And why does everyone see it rather than seeing me? Or – turning the questions on their head: Who is this I that dares to think of itself as concealed behind its face, other than its face, so that its face is not it?19

As with everything that Coetzee writes, these queries are profoundly ambivalent, and I wouldn’t want too hastily to settle their meaning or intent. Does Coetzee mean to suggest that we are right to “dare to think” of our selves as concealed behind our faces, our inner subjectivity radically distinct from the outer appearances we present to the world? Or, put another way, is our inner self, rather, to be understood as private (and to that extent, free) with respect to the categories the social world insists on imposing upon us? Or, on the other hand, is Coetzee rhetorically suggesting that the very idea that there exists a split or a seam between the inner and the outer (where the face begins or where it ends) is a conceptually mistaken picture? Luckily, I don’t need to settle the matter at the moment, because what really matters is that such questions about the face (and I mean all of them) make sense. Deep uncertainty, even outright skepticism: are these not the intellectual moods that for a long time now have characterized our (and our culture’s) thoughts about faces and what they may or may not reveal about ourselves and other persons?

And this, of course, poses a deep problem for my attempt to argue that Wittgenstein’s remarks about face perception have something to teach us about aesthetic expression and experience. For what are we to make of Wittgenstein’s suggestion that familiar words have meaningful faces, or that the hearing of musical expressiveness is like the recognition of facial expressions, when there seems to be such widespread uncertainty about whether faces themselves are actually meaningful in the first place?

For Wittgenstein, meaning is to a word as mind is to a face; but he can liken a word to a face, in this way, only because he already regards faces as naturally expressive of mind. Consider, for example, this typical remark from Zettel:

“We see emotion.”—We do not see facial contortions and make inferences from them (like a doctor framing a diagnosis) to joy, grief, boredom. We describe a face immediately as sad, radiant, bored, even when we are unable to give any other description of the features.—Grief, one would like to say, is personified in the face. 20

If Wittgenstein is wrong about faces, that will surely compromise his physiognomic approach to aesthetic expression, even if he is right that there is a conceptual connection between the two—leaving us, perhaps, with a theoretical position not unlike de Man’s.

And, indeed, I would not be surprised if some of my readers—especially my colleagues in literary studies—find it hard to believe what Wittgenstein seems so clearly to be saying. The direct perception of mental states? Emotion, personified in the face? Of course, in our everyday lives, many of us act and feel as if we could in fact see other people’s feelings and emotions (and sometimes even their thoughts) written on their very faces, so to speak, but if we ask ourselves whether and how this can be, theoretical reflection will likely make it difficult to keep our grip on this ordinary intersubjective experience. Indeed, upon further thought, such immediate perceptual experiences of other minds will likely begin to seem only apparently immediate and direct, but in truth, we will think, they must involve some (perhaps tacit, or maybe lightning-fast) form of inference or interpretation. Otherwise, it would seem we are committing a category mistake of sorts: claiming to see mind (and mental states) where clearly there is only the body (and its overt behaviors).

The human face seems wonderfully expressive of mental states, but (so this argument might continue) this is only because it has evolved so superbly well as a readout of inner affective states. The perceptible appearance of the face’s flesh, though, is certainly not itself imbued with intrinsic psychological content, whatever that could mean (at least this is what I might think if I continued reflecting along these lines): instead, perhaps we should think of the face as a complex signaling system, as evolutionary psychologists like Alan Fridlund have argued (Deleuze and Guattari, perhaps entertaining a similar thought, liked to call the face a “machine”).21

All the way from cognitive psychologists to poststructuralist theorists, a key underlying assumption that unifies so much contemporary thinking about the face (and indeed about the human body as a whole) is that its surface is without intrinsic psychological significance. As the developmental psychologists Annette Karmiloff-Smith and James Russell very succinctly put it: there is “nothing specifically mental about human faces.”22 What they mean is that the face can, by altering its spatial configuration, indicate, refer to, or signal psychological content, but the face as such is assumed to be without any mental meaning of its own: which is why it is presumed to need interpretation in order to be understood.

The roots of this attitude towards the face’s appearance go deep and very far back in Western culture, but Descartes’ fateful re-interpretation of human beings as composite entities made up of minds, on the one hand, and bodies on the other is clearly of special historical importance. After Descartes, the very idea of human expression suffers a deep inner division, very like that suffered by the overarching concept of the human being: on the one hand, there is now that which is expressed (the feeling or emotion) and on the other, there is the perceptible bodily expression (a gesture or an utterance). Unsurprisingly, how one connects or re-connects the outer to the inner becomes a very troubling philosophical problem: the modern problem of other minds.

So though we do, of course, still speak of psychological expression (of “expressing” our thoughts, “expressing” our feelings) that is now, in truth, but a manner of speaking: our thoughts or emotions are not actually expressed in our expressions, but rather our so-called “expressions” are simply ways of referring, or pointing, to that within which passes show, as Hamlet once said. Or, as the cognitive psychologist Alan Leslie put it more recently: “Because the mental states of others (and indeed of ourselves) are completely hidden from the senses, they can only ever be inferred.”23 So we do not see actual grief, but only “the dejected havior of the visage”; not happiness itself, but only an upward curvature of the lips. But as Hamlet knew all too well, “one may smile and smile and be a villain,” so we should take care lest we take the face at face value.

And so it now seems to many (like Leslie) simply self-evident that interpretation and inference are necessary in order to understand the linguistic and behavioral expressions of other human beings. But anyone who has traveled this far down this particular philosophical road, will need to acknowledge that interpretation or inference may not now be enough to re-connect the outer to the inner. Having rent that expressive connection apart, is there a way to weld them back together?

If this picture of the intrinsic inexpressiveness of the outer were in fact true, we could never be sure that our inferences or interpretations about others’ mental states are correct. De Man would have a very good point, then, when he claimed that interpreting everyday interactions is a “Sisyphean task.” And one does not have to be a literary theorist to recognize, given this way of understanding the surface of the body, that our access to the mental states of others will be uncertain or indeterminate, at best. Indeed, this is why the analytic philosopher Daniel Dennett embraces what he calls the “radical indeterminacy” of our mental state attributions, though, given his dual commitments to Darwin and pragmatism, he is (unlike de Man) happy to emphasize how we understand each other just fine for the purposes of everyday life.24

III. “The Face is the Soul of the Body”

Wittgenstein’s later writings are, in large measure, a response to the epistemological uncertainty, even skepticism, that naturally follow from the Cartesian picture of the inner-outer split, the very sort we see shared by modern thinkers as otherwise different as Dennett and de Man. Since I can’t provide a full accounting here of Wittgenstein’s response to the Cartesian picture of mind, let me outline just two key aspects of it, with an eye to our overarching focus on the face itself and the so-called “face” of the work of art. I want to start with what I see as Wittgenstein’s basic refusal of the Cartesian picture of the mind.

Unlike the better-known French critiques of the Cartesian subject, which provisionally accept, and only then deconstruct, Descartes’ binary of mind and body, Wittgenstein simply reminds us that we need not ever accept, in the first place, Descartes’ deeply flawed picture of who and what we are.25 For Wittgenstein, what makes no sense about the Cartesian picture is its famous conclusion that humans are composite entities. We are not minds and bodies, we are simply human beings. The twin concepts of mind and body can, of course, be theoretically abstracted from the primary concept of the human being, but the concept of the human being cannot be re-constituted once one regards mind and body (as did Descartes) as the true primary substances.

In numerous remarks, Wittgenstein gently reminds us of the primacy of the category of the human being in our understanding of ourselves and of other persons. Consider this well-known remark, for example, from the Investigations:

… only of a living human being and what resembles (behaves like) a living human being can one say: it has sensations; it sees; is blind; hears; is deaf; is conscious or unconscious. (PI §281)

Here Wittgenstein suggests that we do not see overt behaviors, interpret them, and then conclude that they are the behaviors of a human being, whom we can infer to be in this or that mental state. On the contrary, we first recognize the living presence of a human being, and so see that human body as immediately expressive of psychological life. Wittgenstein touched on this important point in one of the very late manuscripts, when he wrote: “It’s always presupposed that the one who smiles is a human being and not just that what smiles is a human body…. I react immediately to someone else’s behavior. I presuppose the inner insofar as I presuppose a human being.”26

This is what he meant when, in a well-known remark in the Investigations, he claimed that we see other humans as minded creatures not because of some inference or explicit belief about the concept of the human, but because of a much more fundamental attitude we naturally take towards their appearance and behaviors: “My attitude towards him is an attitude towards a soul. I am not of the opinion that he has a soul” (PPF §22).

If we talk about a person’s body or about a person’s mind, we are not talking about two distinct things, moving between the visible and the invisible: rather, we are referring to two aspects of one, single, undivided entity: the human being, him or herself. The body is not wholly body, as conceived by the scientific tradition that is the most powerful of Descartes’ modern inheritors, but rather the body is always already expressive of psychological meaning. To think of the body as merely material substance (mere matter, mere flesh) is, in fact, to inflict upon it a violent theoretical reduction. And so, Wittgenstein can succinctly formulate his alternative to the Cartesian picture of the mind in one justly famous sentence: “The human body is the best picture of the human soul” (PPF §25).

The second important aspect of Wittgenstein’s response to the inner-outer problem follows naturally from his basic refusal of the Cartesian picture, and is already very clearly suggested by it. The surface of the human body, that is, need no longer be conceived of as inexpressive, as without intrinsic psychological content, as we’ve seen is the case for so many different domains of modern thought, from philosophy to literary theory to psychology. The evolutionary psychologists Leda Cosmides and John Tooby simply echo modern scientific doxa when they write: “Normal humans everywhere not only ‘paint’ their world with color, they also ‘paint’ beliefs, intentions, feelings, hopes, desires, and pretenses onto agents in their social world. They do this despite the fact that no human has ever seen a thought, a belief, or an intention.”27

But only if we insist on thinking of the body and its overt behaviors as somehow distinct from the mind and its mental states will we feel compelled by this characteristically modern view. There is, in fact, no conceptual problem with the ordinary belief that we can see mental states themselves in the very movements of the human face or body.

Wittgenstein was, in part, so deeply fascinated by the human face because its movements are often so complex and subtle that there is frequently no way to provide a purely spatial description of its appearance. Indeed, there is often no better way to inform another person of how someone else’s face happened to look (or change) than by means of imitation: he made this face, we might say, or his face altered like this, while we imitate the expression with our own face. But what fascinated Wittgenstein is that this poses no problem at all when it comes to understanding the face’s expressions, as it would if we needed to infer psychological meaning from purely spatial information. In the following remark, Wittgenstein attributes great importance to this simple (but telling) fact about face perception: “One may note an alteration in a face and describe it by saying that the face assumed a harder expression – and yet not be able to describe the alteration in spatial terms. This is enormously important.”28 Since we often cannot directly describe a face’s appearance in purely spatial or quantitative terms, why insist that the mental states we attribute to faces are the indirect outcome of an interpretive process? What sense does the idea of indirect psychological interpretation have if there is not necessarily a direct spatial perception we can contrast it with?

Of course, that we can immediately see mental states in the appearance of a human face does not mean that it may not be very difficult, in a given situation, to see or to figure out what someone else is thinking or feeling, but that is an ordinary human difficulty, not a deep philosophical problem. And it poses no trouble at all for the idea that the human body is intrinsically expressive of mind.

Wittgenstein was of course concerned with the expressiveness of the human body as a whole, but I believe he saw the human face as somehow special, and the bearer of particularly important lessons about expression for those who would wade into the murky waters of the philosophy of mind. And so though he would memorably call the human body the best picture of the human soul, he would also feel moved to write: “The face is the soul of the body.”29

But as I said earlier, Wittgenstein saw many more things than just the human body as possessing faces or face-like qualities: things such as words, games, and melodies. These examples show clearly that when Wittgenstein attributes a physiognomy to a non-human phenomenon, he is not thinking of some merely phenomenal or perceptual resemblance. The family resemblance, if one exists, must be purely conceptual in nature. What, then, might Wittgenstein have been getting at when he saw words as possessing faces, or the game of chess as having a physiognomy, or the music of Schubert as having a countenance? In one of his rare lectures on aesthetics, he told his students this: “If I say of a piece of Schubert’s that it is melancholy, that is like giving it a face (I don’t express approval or disapproval). I could instead use gestures or dancing. In fact, if we want to be exact, we do use a gesture or a facial expression.”30 Indeed, Wittgenstein’s radical re-consideration of the very meaning of the concept of a “surface” when it comes to our perception of the human body, ramified quite naturally into his thinking about cultural (and especially aesthetic) phenomena. What Wittgenstein realized was that just as the surface of the human body is intrinsically expressive of psychological states, so too, can be the “merely” material surfaces of human-made cultural artifacts: lyric poems, melodies, and pictures.

The text known as The Brown Book contains a fairly lengthy discussion of the concept of expression, and in it Wittgenstein seems to be especially concerned with the temptation we may feel to think of what is expressed and the expression itself as distinguishable. At one point in this discussion, Wittgenstein reproduces the following drawing of a face:

And he asks us to look at the face-picture and notice how, despite how primitively it is drawn, it can nonetheless strike us as possessing a distinct, particular expression. A familiar experience, of course, and not problematic in the least. But as Wittgenstein goes on to point out, we are often, upon reflection, tempted to draw a further (mistaken) conclusion about the relationship between the image and the particular expression we see expressed in it:

And yet one feels that what one calls the expression of the face is something that can be detached from the drawing of the face. It is as though we could say: “This face has a particular expression: namely this” (pointing to something). But if I had to point to anything in this place it would have to be the drawing I am looking at. (We are, as it were, under an optical delusion which by some sort of reflection makes us think that there are two objects where there is only one. The delusion is assisted by our using the verb “to have,” saying “The face has a particular expression.” Things look different when, instead of this, we say: “This is a peculiar face.” What a thing is, we mean, is bound up with it; what it has can be separated from it.)31

And so, when we say that a face, or image, or anything else, for that matter, has a particular expressive quality, it is all too easy to conclude that we are referring to some sort of conceptual “relation” between two distinguishable terms: what is expressed, and the expression’s perceptible surface. In this way, the grammar of the verb “to have” may bewitch us into accepting a Cartesian-like distinction, the very one that has convinced so many modern thinkers that actual facial expressions stand in a relation to some distinct (often inner) mental state.

Wittgenstein’s key point, though, and one that is absolutely crucial if we are to regain our grip on the reality of emotional expression in the arts, is that the concept of expression does not name a “relation” at all: expression does not connect inner to outer, mind to body, or emotions to their expressions.32 What is expressed is present in, and as, the expression itself. To think otherwise will be to create the very problem of relationship (the “how does it work?” problem) that the idea of expression will then never be able to solve: the concept of human expression, once rent asunder, cannot be put back together again. Wittgenstein saw this temptation to break apart expressive phenomena into theoretically distinguishable components as a basic (and tragic) consequence of the Cartesian picture of the inner and the outer. It will be a major concern throughout his later manuscripts on the philosophy of psychology, and the issue shows up particularly clearly in his many remarks on the topic of seeing aspects, some of which are reproduced as Part II, section xi of the Investigations.33

What concerns Wittgenstein, in his discussion of aspect-seeing, is our inclination to regard the perception of expressive qualities as indirect rather than immediate, a belief that follows naturally from the assumption that expressive phenomena themselves are instantiations of relations between multiple (and distinguishable) terms. If expression is in fact a relation, then of course the surface of an expressive phenomenon (like a melancholy melody) will need to be interpreted if the sadness is to be perceived in the music’s sound patterns, which would be all we could directly or immediately hear (as theorists like Davies, Levinson, and Kivy assume). But in remark after remark, Wittgenstein asks us why we are so tempted to think of experiences of expression (or of meaning in general) as indirect, as if all we can see or hear or feel are what the natural sciences say our senses can, and that anything more that we “perceive” must be the indirect result of interpretation or projection?

Wittgenstein provides no knock-down arguments to refute once and for all such ways of thinking about perception. Instead, he simply wants us to see that we need not choose to think of our sensory experiences in that way, but of course, nothing can stop us from doing so if we insist on regarding the physical world as void of qualitative meaning. That stance, after all, is consistent with the scientific world-view, which claims that even colors do not have the same degree of reality as the primary qualities of spatial extension and mass. But we no more need to “paint” our world with colors than we need to “paint” our fellow humans with mental states; nor, I’d now like to add, do we need to “paint” artworks with expressive import.

Wittgenstein knew this last claim (just like the first two) would strike many of his readers as less than obvious, as the following discussion makes clear:

Just think of the expression “I heard a plaintive melody”! And now the question is: “Does he hear the plaint?”
And if I reply: “No, he doesn’t hear it, he merely senses it” — where does that get us? One cannot even specify a sense-organ for this ‘sensing.’
Some would now like to reply: “Of course I hear it!” — Others: “I don’t really hear it.” (PPF §§229-230)

We are by now familiar with what is at stake in this back-and-forth. Notice, though, that Wittgenstein does not decisively settle the issue. He simply gives voice to both sides and then seems to ask: which voice is yours?

But what is especially telling, I think, is that these remarks about musical expression are immediately followed by a set of remarks about seeing (and not seeing) the psychological expressions of a human face. The subsequent remark about face perception reads like this:

We react to a [timid] facial expression differently from someone who does not recognize it as timid (in the full sense of the word). — But I do not want to say here that we feel this reaction in our muscles and joints, and that this is the “sensing”. — No, what we have here is a modified concept of sensing.
One might say of someone that he was blind to the expression of a face. Would his eyesight on that account be defective?
But this, of course, is not simply a question for physiology. Here the physiological is a symbol of the logical. (PPF §§231-2)

I consider this to be one of the most important and dense of all the remarks in Part II of the Investigations, so I cannot hope to do justice here to its many facets. But I will say that I take it to be primarily about the fact that though we can often immediately see emotional expressions in a face, that ability is not merely biological in nature (not simply “hard-wired” into us) but a kind of skill or technique that therefore requires enculturation (the full mastery of concepts like “timidity”). Whether or not we are fully familiar with a given state of mind will be shown by how we react to its appearance in others. Someone who is blind to a person’s timid expression might, for example, treat him with indifference, rather than the sympathetic concern he wants or needs.

The numerous and profound implications of the idea that one might be conceptually, rather than physiologically, blind to the expressions of another person have been most thoroughly explored by Stanley Cavell, in the beautiful passages in The Claim of Reason about soul-blindness, for instance.34 And as Cavell makes clear there, to fail or to refuse to acknowledge the soul or humanity of another person is, in fact, a standing and quintessentially human possibility, even temptation. That we can be blind to the humanity of another is a logical possibility that, in part, constitutes what it means to be a human being in the first place.

What I am intrigued by, for the moment, is the fact that Wittgenstein’s remark about face-blindness so closely follows his discussion of musical perception, indeed coming right after we hear a voice exclaiming “I didn’t really hear the plaint.” It follows so closely, I suggest, because Wittgenstein regards blindness to the face’s expressions as internally related to the blindness we might display towards emotional expressiveness in art. It follows as well, though, that such blindness to aesthetic expression will be no less a standing possibility, and even temptation, than the possibility of our being soul-blind with respect to one another. There is nothing to stop us from regarding artworks as void of intrinsic expressive life, just as there is nothing to stop us from seeing the human face as without intrinsic psychological expressiveness. Indeed, “expression-blindness,” to give this condition a name, appears to be a widespread and particularly powerful temptation in modern culture and philosophy. It is attested to no less by the numerous modern artworks about faces that portray them as masks, screens, or opaque enigmas,35 than it is by so many of the theoretical discussions of the face I’ve canvassed in this essay.

But just like skepticism with respect to other minds, as Cavell has long argued, neither of the two forms of expression-blindness I am talking about can be philosophical refuted. Cavell’s important discovery about skepticism was that far from simply being an intellectual error in need of correction, the skeptic’s position expressed an important philosophical truth: that there is no absolute ground for the meaningfulness of our lives together (like a framework of concepts or rules), only the fragile attunements we ourselves maintain by means of our continuing investment in, and care for, our shared sense-making practices. There is thus nothing to stop any of us from withdrawing our acknowledgment of those attunements, fragile as they are, which is of course the skeptic’s tragic choice. And just so, there is nothing to stop any of us from withdrawing our mutually attuned acknowledgments (fragile as they are) of the expressive meaningfulness of our very bodies, or of the artworks we make, enjoy, and study. The aesthetic expressiveness of art will indeed be but a fiction—and artworks will be dead: mere sounds, images, and dead letters—in so far as we choose (as we always can) to see them in that way. Indeed, as I think my essay has made clear, quite a few modern thinkers have already made that very choice.

But what I would like to suggest, by way of conclusion, is that that choice need not be one we ourselves feel compelled to make, as if it were somehow philosophically truer and less theoretically naïve to see the emotional expressiveness of artworks (as of ourselves) as something that’s not really there, but rather some sort of interpretive projection, an animating fiction, or what have you. That, I think, is what Wittgenstein and others, like Adorno, Benjamin, and Susan Stewart, are suggesting, when they figuratively endow artworks with faces, and even sometimes with the face’s power to return our gaze. Such physiognomically-inflected remarks are reminders that the very stuff, the materiality, of our aesthetic artifacts, no less than the flesh of our finite bodies, has the power to express mind and meaning. And so it may be that what is ultimately at stake in our continuing re-thinking of the category of aesthetic expression is not simply how best to understand art, but also how truly to see—that is, understand—our very selves.