"As a record of Pantera's appearance the painting could hardly be surpassed". This ostensibly innocuous statement in The Portrait in the Renaissance by John Pope-Hennessy concerns a striking picture by Moroni, a painter active in Northern Italy in the sixteenth century. But how could Pope-Hennessy know how good a record it was? Isn't likeness measured by a comparison between the portrait and what the sitter looks like? The sitter has been dead for hundreds of years and there is no other record of his appearance. On examination this statement sounds totally absurd.

How could a famous scholar and museum director like John Pope-Hennessy, hardly a stupid man, allow himself to write something so silly? What is there in Moroni's painting that prompted him to do so? It must be, above all, the wealth of details and the abundance of precise information about the appearance of the man represented. The liveliness of the personage is also arresting; the painter has found a trick to suggest that the sitter has just turned his head around to face us: he has left the long beard in the position it would normally have had if the figure was seen from the side. In brief, the painting itself, through what one might call its visual rhetoric, convinces us that it is an exact likeness without our having to turn to the model. This is what I shall call the likeness effect.

This does not mean that the referential function of likeness does not exist, nor that it is without importance. Ever since the Renaissance, portraits have been made to record the appearance of actual persons, and this is the reason for the success of the genre: it does the job well. We feel confident that those who knew Pantera would have recognized him in this painting, or, vice versa, that having seen the portrait one could have spotted the man.

Indeed, the real test of likeness is recognition rather than a comparison between the image and the model. The two things are distinct. What does recognition consist in? First of all, it is the fact of realizing that a person who comes before our eyes is known to us, that it is So-and-so. There are exceptional cases, mostly uncomfortable when experienced, where we recognize a person in the sense that we realize it is someone we know, but we are unable to say who it is. Identification is a question separate from recognition. More upsetting still, we sometimes think we recognize a person without being able to confirm the impression or deny it either. Such an experience is not unlike the effect of so called "déjà vu" of which psychologists have given different interpretations.

If a person leaves our field of vision only to reappear shortly after, we do not say we recognize her. In order to speak of recognition, there has to be a sufficient hiatus to create some suspension in the continuity of our memory. Something has to click inside us: "Ah, that's so-and-so!" Recognition in the full sense of the word is an interior event, the resolution of a tension, however slight. For this reason, this kind of recognition experienced in our everyday encounter with people, as well as in front of portraits, is not unrelated to the theatrical recognition as defined by Aristotle in the Poetics, the resolution of a dramatic situation.

Let us now turn to images and how we recognize people there. A charming cartoon by Frasconi goes to the heart of the matter. "Mrs. Hammond", says the school teacher, "I would have recognized you anywhere from little Billy's portrait of you." Indeed, the woman who enters unquestionably resembles a child's drawing! The point, of course, is that recognition, and consequently portrait likeness, are associated with some specific modes of representation and incompatible with others. In addition, "natural" or "realistic" representation, a necessary condition for recognition to occur, is also conventional: in the cartoon the teacher is drawn in a manner which is not that much more "realistic" than Billy's mother, but within the genre of the cartoon, this kind of representation is the standard idiom and it is coded "natural" or normal as opposed to the "naïve" childlike drawing. Portrait recognition functions fully only within what is known as "realistic" or "illusionistic" representation, that is, within a set of conventions which developed during specific periods, especially Greco-Roman antiquity and from the 14th century to the present. The portrait thus conceived keeps the memory of the portrayed individual alive. In the long run, in fact, it tends to fuse with our actual memory and to substitute itself for it.

There is no question that recognition on the basis of an image, the situation humorously evoked by the cartoon, does occur. The use of mug shots by the police to locate delinquents depends on it. Yet, how good a likeness does it have to be? A case in point is that of description based, "composite" portraits used when no photograph is available. These are made by constructing an image from the description provided by the witness, then having the witness criticize the image in order to correct it, a process of making and matching where the matching is not against an actual visual experience, but against the impression left in memory. Rather surprisingly, this method does produce results and criminals have been apprehended thanks to the publication of such crude images.

It may be possible to spot someone on the basis of such pictures, but they are useless for formal (forensic) identification; even photographs are problematic for that purpose. Identification is different from recognition since, as we have seen, we can recognize someone without knowing who it is. Both can apply to actual persons as well as to their representation. Recognition is an almost automatic or preconscious reaction whereas identification is a deliberate thought process. In addition, identification is not always based on likeness the way recognition is. For instance, the portrait by Moroni that I discussed at the beginning has been identified as that of Pantera not because it looks like him (his physical appearance is otherwise unknown) but because the sitter holds a book entitled Monarchia del Nostro Signore Gesu Christo which Pantera is known to have written. This is why we assume that this is his portrait without in fact being certain of it. Given this portrait of Pantera, and assuming that the identification is correct, we might be able to recognize him in other unidentified portraits. Conversely, should another portrait of Pantera be discovered it might confirm, or invalidate, the identification of Moroni's sitter on the basis of likeness.

One often gets the impression that likeness is the necessary and sufficient condition of the portrait. But it is not. On the one hand, likeness is not a sufficient condition for an image to be a portrait. Consider this simple observation. [Fig. 3] will easily pass for a portrait, an informal, familiar portrait, no doubt, but portrait all the same. Here, we assume likeness, and the possibility of recognizing the sitter, because of the photographic method which was used to produce the image, but it is not the only thing that makes us decide it is a portrait. As it happens, I have extracted this "portrait" from another image. And in front of the complete image we categorically reject the idea of the portrait, not because there are several characters (the group portrait is a well established genre), but because the function or purpose of the image is blatantly foreign to that of the portrait and elicits an entirely different form of attention. We are asked to attend not to the physiognomy of the young man but to his apparel. The likeness, however, is the same.

On the other hand, is likeness necessary for a portrait? Let us take the extreme case of an English ceramic plate from the late seventeenth century with a figure drawn in a manner which, like the child's drawing encountered in Frasconi's cartoon, definitely excludes any possibility of physiognomic likeness. Yet the royal crown and the initials C R (for Carolus Rex) designate King Charles II. Can we call this a portrait of King Charles? Yes, if the identification of the individual represented is considered as the essential criterion. No, if one insists on physiognomic resemblance. It is simply a matter of definition and the pursuit of an ontological, universally valid, definition of the portrait is illusory. There is no such thing as "the portrait" outside of history, but only what a portrait is at a given time, in a given society. This may be something of a truism by now, but it is still necessary to remember it clearly.

Let us return to the question of likeness. When we look at a portrait in the fullest sense of the term as it has been used since the Renaissance, for instance the portrait by Moroni we examined at the beginning, we do not decide that we are looking at a portrait because there is a physiognomic likeness with an actual person (as we have seen, the identity of the sitter is not even established with certainty), but on the contrary we assume this resemblance because we are aware that it is a portrait. Two main factors enter into play here: the presentation of a single figure, disengaged from any action, without any allegorical attributes identifying it as a mythical, symbolic, or legendary being, signal the painting as belonging to the category portrait. In addition, there is what I called the "likeness effect", that is, the devices used in the representation to convince us that this is an actual and exact likeness: the differentiation and slight irregularities of features, the subtle asymmetry of the face, the abundance of details.

It is important to understand the interplay between the internal rhetoric of representation and the referential function of portraits. In order to be effective, the likeness effect depends on the hypothesis of a possible reference, on our assuming that such a person does or did exist, on the eventual test of recognition. But in practice, this referential test is most often excluded, the referent - the sitter - being inaccessible, dead, or even unknown. The result is a tendency to integrate likeness as much as possible inside the system of representation. Likeness then becomes essentially a matter of the presence or absence of the likeness effect. As a result, we are no longer dealing really with a greater or lesser resemblance between an image and a person, but with the degree of difference between the image and an "ideal", an interiorized generalized image or "model", or imaginary idea of the perfect man, the ultimate housewife, the baby- which is always largely a social convention rather than a personal creation. Going back to the commercial in figs. 3 and 4, while the excerpt can pass as a portrait because of the photographic likeness, the moment we look at the whole picture we no longer see a person but a model, not a man but a manikin (mannequin is French for a fashion model).

In fact, recognition, which we have so far discussed mostly as the test for the reference of the image to the outside world, functions very often inside the realm of representation. The portrait of Homer is a case in point. Within Greek art (and again since the Renaissance), the portraits of Homer are readily recognizable. This cannot be through reference to a biological individual whose very existence has often been questioned and who would have lived at a time when the record of appearance was not an option. Recognition here is made possible by reference to a physiognomy constructed, established, and perpetuated by artists. It is striking also that these portraits of Homer can be very different from one another while still remaining recognizable. In classical Greece, recognizability was possible in spite of a low degree of "likeness effect" and a fairly wide margin of variation around an established portrait type, like Homer, because portraiture was restricted to a relatively small number of individuals, a select group of role models. In modern times, after the Renaissance, having a portrait was extended to ever larger sections of society and it became necessary to specify the subject's appearance, to "individualize" the face more and more in order to recognize the subject.

As it became increasingly common, the likeness effect was devalued; inflation, so to say, set in. And with photography, where it becomes an automatic consequence of the very method of producing the image, its meaning changes. The signal "this is somebody", which is an integral part of the portrait, is transformed into "this is anybody." It is in these circumstances that such works could appear as the famous photograph by Paul Strand entitled Portrait, New York or Walker Evans's Portraits, New York, 1931. Yes, these are portraits, recognizable as such by the artifice of their deliberate pose, portraits where the photographic method guarantees us physiognomic likeness to an actual sitter; but paradoxical portraits because they are of nobody in particular. The title indicates that the author deliberately detached the "portrait" from the context where the recognition and identification of the individuals could play their part. We see a specimen of the category "individual", a face in the crowd

-----

Enters Andy Warhol- with Marcel Duchamp, the most intractable, the most ironic, Sphinx-like artist, about whom anything one says or writes always seems somewhat beside the point. The following remarks should be read with this caveat in mind.

The portrait plays a major part in Warhol's work, and this is not surprising since he was fascinated, one might say obsessed, by people: people as in People magazine, but also just people who came in and sometimes out of his life. The drawings exhibited here are not the most characteristic or familiar part of this large output. Let us try and find their place in it.

Warhol emerged suddenly, meteorically, into the art world with the 1960s. Pop Art. The New York public first encountered Warhol in 1962 at Sidney Janis's "The New Realists" exhibition and with his own show at the Stable Gallery. Portraits were prominent here, but not the ordinary kind of portrait. The most striking were certainly those of Marilyn Monroe, who had just tragically ended her life; a popular icon if ever there was one. A brilliant young critic, Michael Fried, had immediately spotted "Warhol's beautiful, vulgar, heart-breaking icons of Marilyn Monroe" as "the most successful pieces in the show" at the Stable. Gold Marilyn Monroe, given to the Museum of Modern Art by Philip Johnson in 1962, did more than any other work to establish the artist and he would never surpass it. In other paintings, the image of Marilyn was repeated, multiplied in rows, a little like the celluloid strips of films, the image changing slightly, not because of motion as in actual film strips, but in nuances of printing and coloring.

For these paintings Warhol used an unprecedented technique: a photographic image, a publicity still for the film Niagara, was silk-screened onto the canvas. The canvass had been prepared with more or less flat areas of colors corresponding to the areas of hair, lips, eye lids, as well as background, emulating a crudely colored image.

As in the case of Greek portraits of Homer, the image of Marilyn was a cultural icon, recognizable as a "star", and not meant to inform us about a human being, her inner self and what she might have been like in life. What matters here is explicitly what Marilyn stands for, her myth. The fixity of that myth is signaled by the fact that in his numerous variations on Marilyn, from the year of her death in 1962 until his own in 1986, Warhol never deviated from the one still initially selected.

Fried did not find the paintings of Troy Donahue nearly as successful "because the fact remains that Marilyn is one of the overriding myths of our time while Donahue is not, and there is a consequent element of subjectivity that enters into the choice of the latter and mars the effect. (Epic poets", Fried epigrammatically adds, "and pop artists have to work with the mythical material as it is given: their art is necessarily impersonal, and there is barely any room for personal predilection.)"

Had Warhol painted himself into a corner? Significantly, soon after his breakthrough, Warhol accepted two portrait commissions: the wife of a New York cab tycoon and a high level, but obscure, insurance executive were not the stuff that myth is made of. Warhol's approach to portraiture is very different in the two commissions, although in both cases he adopted the film strip, multiple image format. Ethel Scull Thirty-six Times is as close as Warhol got to a traditional view of the portrait as a psychological account of an individual. Warhol used 36 (4 x 9) different photobooth images of Mrs. Scull in different moods and poses. Each one was silk-screened on a separate canvas with a different color preparation; mounted together, they formed a large variegated composition. Not only is likeness guaranteed by the photographic process, but recognition functions within the painting, from one "frame" to the next. This allows Warhol to evoke the personality of the sitter by building up her identity in the multiplicity of attitudes. This is as close as Warhol got in his mature years to the standard ideology of the portrait.

For Watson Powell, Warhol chose the most banal, "official", three-quarters view and repeated it thirty-two times (4 x 8) on small canvases assembled in a large upright composition, changing the ground only slightly. The title, The American Male - Watson Powell, points to something epic. The picture was also known as "Mr. Nobody", a kind of ironic corporate version of Everyman.

Warhol's only- and, as it turned out, very ephemeral- monumental project, also involved portraits. Thirteen Most Wanted Men of 1964 as a series of masonite panels with enlarged silkscreened portraits, most of them taken from police mug shots of criminals, which was made as an outside decoration for Philip Johnson's New York State Pavilion at the New York World Fair. The project was found politically objectionable and the images were obliterated with aluminum paint, but the work still exists in a dispersed series of canvases of slightly different formats. Here, the physiognomies are not jazzed up in any way with color patches or even a painted background; the rawness of the police document remains intact. The photographic likeness is allowed to speak loud and clear and the enlargement gives a powerful sense of human encounter. Here, the portrait functions as a document of human individuality. These men, especially since the ambiguity of the title, Most Wanted Men, inescapable for an openly homosexual artist, eliminates the possibility of a straightforward, pious social message.

Around 1970 Warhol began to use a painterly handling in his portraits but this method was mostly developed in 1972, with the portraits of Mao Tse-Tung. Here the "personal" touch of brush-work works against the silkscreened, "impersonal" photographic image. This device would become the basic equipment of the "travelling portrait artist" as Warhol once described himself.

In his naïve/cynical fashion Warhol established a standard protocol for producing portraits. He would make a photographic likeness with a Polaroid camera (one or a few shots being selected from the many taken in quick succession and enlarged onto silkscreens). He added the "style" or "art" with brush work and sometimes texturing the paint with his finger in the case of canvases. Henry Geldzahler tells how Warhol having presented him with his portrait, he said: "But Andy, I don't want it, it's nothing but a blow up version of the Polaroid. You've left out the art!". "Oh", Andy said, "I knew I forgot something".

The "art", the "style" in Warhol's portraits works against the photo likeness in a dialectic way. Warhol's sitters, whether famous or not, are "stars" whose recognizbility is so high that it comes through a thick layer of "art". That this worked for Marilyn is not surprising, but could it function for others? For the obscure rich who commissioned their portrait? "In the future, everybody will be world famous for fifteen minutes."

Warhol's trump card (one of them, at least) was to make his passivity, his ostensible lack of involvement and impersonality, highly distinctive. Leonardo played the artist as engineer, Velasquez the artist as courtier, Warhol became the artist as "star". He designed an idiosyncratic appearance for himself, which conferred instant media recognizability to his person. He became an icon. And the instant recognizibility of his pictures, the Warhol "look", was part of his stardom. The people he portrays are famous for a moment, as he portrays them, in the limelight of Warhol's radiance. The very existence of the Warhol portrait becomes a presumption of recognizability, just like the photographic method confers a presumption of likeness.

---

DRAWING. Like all artists, Andy Warhol drew and his drawings bring out most strikingly his subversive relation to the artistic tradition. As with many artists, beginning with Durer, one of Warhol's earliest known works is a self portrait in the mirror. This intensely expressive self investigation, drawn in 1942 when the artist was fourteen years old, betrays not only exceptional gifts, but considerable sophistication. This is not the naïve attempt of an untrained hand or eye. The pose already is savvy: rather than facing the mirror squarely, the boy has his head turned slightly to the left; this makes the description of the features easier while it increases the expressive tenseness of the gaze. The meticulous observation of appearance - what produces the likeness effect - is contained within a strong geometric understanding of the face's planes and volumes combined with a fine sense of the decorative arabesque, in the manner of such artists as Grant Wood or Thomas Hart Benton.

While studying at Carnegie Tech in Pittsburgh (now called Carnegie-Mellon University) Warhol perfected a method of indirect drawing known as "blotted line", a kind of monotype or hybrid between drawing and printing. This idiosyncratic technique was central to Warhol's practice throughout the 1950s, when he made a living as a commercial artist in New York. It consisted essentially in drawing with pencil on a sheet of non-absorbent paper. It is kind of rudimentary printmaking, that distances the image from the direct "touch" of the artist's hand. It was not only the particular look that appealed to Warhol, but also the possibility of inking the master sheet again to obtain the same image several times, although with slight variations.

It was also during this days at Carnegie Tech that Warhol took to tracing from photographs or other material like magazine reproductions or earlier art. Thus the master sheets that the blotted line works were made from were usually not free hand drawings but already tracings, tracings of already printed images. It is important to insist upon this, because free hand drawing holds such an important place in the traditional ideology of art. "Drawing is the probity of art" Ingres had insisted. The use of photography or any other device that interrupted the direct flow from eye to hand to paper had always been considered as highly suspect, or even illicit (although artists quietly used such devices for convenience sake). From the beginning, the draughtsman Warhol used anti-artistic means. And when he made free hand drawings, he often used a generally disdained tool, a ballpoint pen, which hardly registers their variations of pressure from the hand and produces an even, impersonal line. Many of the elegant portraits of Warhol's early years, and even those of Gene Swenson made in 1962, are in that technique.

In the years 1960-62, as he was developing his new "Pop" imagery, Warhol made a very different kind of drawings. They are mostly in pencil. He traced the outlines from various commercial sources and filled in the shaded areas roughly. Unlike the elegantly attractive, finished and often signed drawings of the previous years, these are ostensibly crude. They were, for the most part at least, real working drawings with paintings in mind. Thus at the time when he was getting ready to break into the "serious" art world, Warhol used drawing in its traditional function as a way of developing ideas on paper with another medium in mind. Then, after 1962 drawing all but disappears from his practice for ten years.

Warhol returned to drawing, quite massively, with his work on Mao in 1972-73, following Nixon's historic visit to China. The official photograph of the Chinese leader, the frontispiece of the Little Red Book, was the basis of a large series of painting, prints and drawings. Some of the drawings are not unlike those of 1962, the shaded areas being filled in with hatching, while others return to a more linear system. But even the shaded drawings are different from those of 1962 because the hatching is more organized and not so rugged or "ugly". It is clear that these drawings, which are signed and dated on the back, are not preparatory, certainly not for the paintings, but not for the silkscreen prints either. In the prints, a line drawing is superimposed on the photographic silkscreen and this may have been the original incentive, but the number of these drawings, some forty of them, and their wild variation in sizes unrelated to the prints, make it clear that they took on a momentum of their own and constitute a separate entity. By tracing the drawing on the photograph - no longer by transparency but now by projecting the image onto the paper, which conveniently allows for variations in size- Warhol secures fidelity to the source material, and likeness, but each drawing reinterprets the portrait. Some of the drawings, by outlining the areas of light and dark make it look a little like the solarized photographs of Man Ray, producing a ghostlike appearance. Others insist on the purity of contour. One senses an obvious delight in this inventive graphic effusiveness. And the recognizable likeness of Mao is never lost. More than that, it is a specific interpretation of the great leader's appearance. The large wart on Mao's chin is edited out as well as the pockets under his eyes; the eye brows are much heavier. The line also is much thicker and assertive. By comparison, this is a more forceful, rather brutal image.

This intensive investigation of the possibilities of graphic variation is exceptional in the work of Warhol, and he would not make such an extensive series of drawings again. But the regular practice of drawing stayed with the artist until the end and would accompany the majority of his many portraits. More than that, drawing became increasingly important for Warhol. In the seventies, it generally appears in the prints in a dialectic with the photographic image. Later, in the eighties, it even invades painting.

With the portrait drawings of the 1970s and '80s Warhol seems to revisit the manner of his commercial years: with few exceptions, they are treated in a strictly linear manner. But the mature drawings are different; the obvious recollections of Matisse and Ben Shan, pointed out by Rainer Crone and others in the early drawings, have been distanced. Warhol favors a thick rather rough line which shuns the somewhat affected elegance of the early work. There is also a great deal of wit in Warhol's art; in the pre-Pop drawings it is usually attached to the subject matter, but in the later ones it is in the graphic invention, the way for instance he indicates a shadow with strictly linear means (Princess Caroline), binds figure and framing device (Carter Burden), or syncopates the outline (Einstein). In his last years, Warhol used something like a carpenters pencil, a tool so wide that the line is more like a ribbon and each stroke has a shape of its own.

Obtaining a recognizable likeness was not an issue for Warhol. It was a given: the camera provided it, whether he or someone else clicked it. Warhol was an avid and brilliant photographer as well as film maker. A curious document in the Warhol archive is emblematic of the relation between photography and drawing: it is a photograph of Andy on which he has drawn to make his face more regular, with a less bulbous nose. This is not a retouched photograph, not a make believe, but overt and deliberate day dreaming. Just like his wig, which became the outstanding feature of the late self portraits, a rug with no pretense of being real hair but still very much Andy's own, his fantastic self.

By tracing a projected photograph, Warhol extracted the generic in the individual, evacuating most of the details that make up individuation. In the portrait drawings, the dialectic between photograph and drawing remains, but, so to say, in absentia. Sometimes a universally known face, like Gerard Depardieu's, becomes almost unrecognizable, but the drawing still makes sense because of whom it is, of the absent traced photograph. More often, one has a feeling that the sitter is recognizable with almost no traditional likeness effect: Michael Jackson or Einstein are unmistakable. As for the drag queens Ladies and Gentlemen, we are not expected to recognize them, except from one image to the other.

What strikes me most, perhaps, in these large and bold drawings is that, although they are very similar in format and execution, one can say deliberately standardized, they vary enormously in effect. On one hand, they fall into distinct categories of sitters (even if the categories can sometimes overlap): glamorous ladies, dashing male stars, the powerful and rich, cultural heroes. On the other hand, within the categories the emotional feeling varies drastically, and one senses that in spite of his attempt to distance himself, to remain a passive observer, Warhol's feelings for his subjects come through. John Lenon exudes warmth and sympathy, while Aretha Franklin is all dazzle. In the end, for all their stripping down of inessentials, for all their ironic play with the genre's conventions, Warhol's drawings still do the job of portraits.

This essay was included in the exhibition catalog Andy Warhol: Portrait Drawings, published on the occasion of the related Brooke Alexander exhibition in 1998.

The essay is copyright 1998 Henri Zerner, and is reproduced here for educational purposes only.