[This paper was given at a conference on the sociology of art which took place in Grenoble in November, 1999. The theme of the conference was "L'œuvre elle-même," "the work [of art] itself."]

The Work Itself

What does it mean to speak of "the work of art itself"? What are we distinguishing it from by adding the qualifier "itself"? I suppose that when we say this we are pointing to the difference between the work of art seen merely as a signifier of something else, which is not art, and the work of art as a thing to be appreciated in itself and for itself, for what it is just by existing. Sociologists, from this point of view, might be said to "reduce" the work of art to something less than itself, to something which merely reflects or contains something else. The "something else," again presumably, is some thing sociologists are more at home with: social classes or races or organizations or institutions or "worlds of art." (The word "merely," in the formula I stated above, indicates, as it usually does in philosophical and aesthetic discourse, that if that were what a work of art was, it would be something less than what it might be or actually is when properly seen.)

When we ask ourselves, as sociologists, to speak about "the work itself," we ask ourselves to tread on ground that, in a conventional academic division of labor, belongs to other disciplines. Musicologists (some of them, sometimes) devote themselves to the analysis of works of music "in themselves." Art critics and literary analysts (some of them, sometimes) devote themselves to works of visual and literary art "in themselves." And so on. Every art has its dedicated specialists, some of whom sometimes explain, interpret, and analyze the "work itself."

Analyses of art works typically take one of two forms. On the one hand, analysts and critics look to the internal structure of a work, to the nature of its constituent elements and their interrelations, looking for patterns of harmony, tension, and resolution which are interesting, engaging (perhaps even engrossing), and emotionally moving. On the other hand, they look to the way the work explores questions of emotional and philosophical significance. Very often, of course, they are concerned with both at once: with the way the formal elements of a work and their arrangement lead the user (consumer? reader? listener? viewer?) to experience, in a way that is simultaneously intellectual and emotional) certain emotional and philosophical truths.

When we ask ourselves to speak, as sociologists, about the "work itself," what do we have in mind that is not already done by the adepts of some other field of scholarship? I suppose that what is meant is that sociologists too should explore just these questions of formal pattern and of emotional and philosophical significance, and add something analysts from other disciplines have left out, something which sociological concepts and ways of thinking produce which has hitherto been lacking. What could that be?

Sociology can certainly do some preliminary work in this area. A first job is to make clear that the very idea of "the work itself" is empirically suspect, for many well-known reasons. The authors of the works make many versions of the same thing. Sometimes these are treated as separate works, as when a visual artist makes many sketches and paintings of what is "essentially" the same subject. But very often they are treated as variants from which the one "authentic" work must be extracted, as when editors try to create the definitive text of Proust's work or of an opera which Verdi had revised many times for performances in various houses or for performances by different singers.

It is, of course, worse than that, and most clearly in the performing arts. What constitutes "the work itself" in the case of a musical composition? Is it the score as prepared by the composer and, perhaps, vouched for by scholars as being the authentic real work as the composer intended it? Or is it the work as created in performance by a player or singer or some combination of such performers? And, if the latter, which performance is the work itself or is every performance to be taken separately as a work itself? And similarly with the performances of works for the theater, where stagings of the same play differ widely, as do performances of the same staging.

I could multiply examples endlessly, but it isn't necessary. I mean only to indicate the empirical reality that lies behind what we could call the Principle of the Fundamental Indeterminacy of the Art Work. That is, it impossible, in principle, for sociologists or anyone else to speak of the "work itself" because there is no such thing. There are only the many occasions on which a work appears or is performed or is viewed, each of which can be different from all the others.

Many people would say that this is sophistry. After all, these various versions are "fundamentally" or "essentially" the same thing, the same physical (or abstract) object, even though there may be "slight" variations from time to time. This could best be argued, perhaps, in the case of such physical objects as paintings or sculptures. Well-known counter-examples lie in wait for the person who makes this argument: the Greek sculptures which once were painted but are now just white marble. The Renaissance paintings whose layers of varnish have darkened so that the paintings are quite different in color and mood than they once were. The profession of art restorer rests on the recognition that physical objects change and can even be said to deteriorate. I put that tentatively because in some cases the physical changes are intended, as when a sculptor uses a metal which will acquire a patina that is different from its original color. So objects are never the same, any more than performances or editions are. They vary.

Whether the variation in objects or performances is to be taken seriously is another matter. Are the differences from moment to moment, version to version, "fundamental" or only "slight" and "immaterial"? This is not merely a philosophical conundrum or a sociological caprice. If the art work is fundamentally indeterminate, people will have problems dealing with it. They will not be sure when it exists and when it doesn't, won't know what its form and nature are, won't be able to talk about it (after all, which version are we discussing?). They (and therefore we) can only distinguish "the work itself" by invoking some convention as to what--which of the many forms it takes from moment to moment--counts as the "real," "basic" work and which kinds of variation don't matter, don't interfere with the "fundamental" or "essential" character of "the work itself. This is the kind of shared understanding which social scientists recognize under a variety of names: culture, norm, shared understanding, etc.

Sociology tells us that the choice of such conventions is not arbitrary. Individual actors choose them on particular occasions, but they do not choose them arbitrarily or individually. Rather, they choose in context, in the course of their participation in the social organization of the world in which works of that kind are made. Typically, the conventions competent members of an art world use to decide when an art work is "the same" and when it is "different" are collective, the ones other competent participants in that world use.

How does such an organized world of art choose conventions in general,, as opposed to an individual choosing them for an occasion? It is a mistake, of course, to imagine that the world, acting as a collectivity, necessarily makes a deliberate choice of the conventions which make Version A the "real" or "authentic" one and other versions not. But not always a mistake, since the choice of conventions sometimes is made just that way, as when the ultimate holders of power in such a place as Hollywood decide what the "final cut" of a film will be. Others may not like their decision but recognize that when the studio that has ultimate control, or the director whose contract gives him that power, says this is the final version, that is the final version, though other versions may exist in fact or potentially.

Control of the choice of what constitutes "the work itself" is seldom monopolized this way. More typically, many people have something to say about the decision; in principle, all of the people who participate in its making of the work have some effect on the final choice of "the work itself." Everyone who participates has, depending on the circumstances, some influence, and the general choice of the convention by which works will be recognized results from a political process which is continuous, and never settled for good.

The consumers of the work also share in its production. The work has no effect unless people see it or hear it or read it and they do that in various ways, again depending on the social organization of the world in which the work is made. People with different training will read and experience a work differently. People at different stages of their own lives will read and experience a work differently. They will see or hear or read it under varying circumstances, which will affect what the work "is." So a symphonic work will sound different in different halls on different nights, a painting will look different in different settings and under different light, the effect of a literary work will be influenced by the typography and paper of which it is composed. And they will all be different for people of different ages, genders, classes, emotional states. The "work itself" may not be different, but the work the viewer takes in may well be.

All this is well known. But it is also easily forgotten when we start theorizing. So the first contribution of sociology to an understanding of "the work itself" is the recognition, which should be incorporated into everything else that the analyst says, that the work takes many forms and that the "work itself" is isolated only by virtue of a collective act of definition. Which means that what the work is, while in no way arbitrary, is subject to great variation and can never be settled definitively, in some way that is dictated by its physical nature. This is the full meaning of the Principle of the Fundamental Indeterminacy of the Art Work I announced earlier.

The recognition of the indeterminacy of the work is a contribution sociology can make to the understanding of art works but it is, of course, a negative contribution. It doesn't really get to the heart of the work itself, of what matter it deals with and how it deals with it, of what emotions and ideas are provoked by contemplation of the work–in short, it doesn't get to what we would want to get to if we analyzed the work itself instead of talking about its setting.

Here I think a sociology of art, conceived at least as I have conceived it, can contribute an analytic approach that goes to the heart of what a work of art is. You could call it a genetic approach, since it focuses on how the work is made (and, of course, remade), on the process by which it takes shape, on a step-by-step understanding of how the work came to the form it has when the analysis is undertaken (including an appreciation of all the various forms the work has taken and might yet take).

Such a genetic approach is not unique to sociology, of course. Musicologists prize, as sources of insight into the structure and meaning of a musical composition, those scores which bear the marks of the composer's compositional process: the sketches, alternative versions, revisions, and so on. Literary analysts similarly prize the manuscripts of novels, stories, and plays which show--in what has been changed, deleted, added--the various ways the author had considered constructing the plot, the alternative descriptions of the same person or place that had been considered, the varying placement of the same narrative elements. They particularly love the marked up proofs the author went over and sent back to the printer, and especially those from the days when publishers allowed authors to rewrite their books extensively in proof.

Analysts love such evidences of how the work was done, because the choice the author or composer finally made from the field of possibilities such documents reveal can be taken to indicate what intention motivated the work's making. And that tells us, in the conventions accepted in many art worlds, what the work's ultimate meaning is, what the "work itself" is.

And here I have finally mentioned the word I would make central to a sociological analysis of artistic works: "choice." I first learned to appreciate this work and the idea behind it when I took classes in photography thirty years ago. The teacher who most influenced me, Philip Perkiss, refused to let us treat anything in our picture as beyond our control. Instead, he pointed out that we had always had the choice to do something differently: stand somewhere else, wait until another moment, use a different lens, a different aperture, a different shutter speed, to name just a few of the most obvious choices photographers make, wittingly or not, every time they make a picture. And let's remember, of choices made by the people in the image too (whether and how they would react to being photographed), and by the people who made the film (and thus limited what could be done) and the paper on which the image is printed, and so on. The resulting picture is the consequence of all those choices.

I learned the same lesson when I studied theater communities in the United States and spent a lot of time with theater people. Actors, directors, and others often spoke, approvingly or disapprovingly, of the "choices" someone had made: how he had chosen to read a line, how she had chosen to walk across the stage, the pauses the director had called for, or the details of scenery and costume. The show playgoers see results from all these people making all these choices, millions of tiny choices which add up to the total effect on playgoers.

Any work of art can thus profitably be seen as a series of choices: the choice of which word or note to put next in my poem or song; the choice of which way the plot of the story should go; the choice of a color to load the paintbrush with and where to put that next spot of color; the choice of an inflection in the reading of a line in a play; the choice of the phrasing of a musical idea. As someone composes a work or makes a painting, as they perform a work on stage or in the concert hall, everything they do constitutes a choice. The choice could always have been made differently and everyone who works in these trades knows what the range of possibilities was and what might have motivated the particular choice that was finally made. Even if many, or most, of the choices are made in a conventional or routine way, they are still choices. Making routine choices gives a work a certain character, making unusual choices another character. Almost all works of art and performances of them and readings of them are some combination of routine and unusual choices among available possibilities.

This opens the way to a kind of sociological analysis of "the work itself," even the fundamental indeterminacy of the work I have mentioned already makes clear that we can do that only by virtue of an arbitrary choice of a moment and state of the work that could itself have been made differently. A full understanding of any work means understanding what choices were made and from what range of possibilities they were made, the knowledge I have just suggested is relatively common among practitioners of an art. These choices are made in a complicated social context, in an organized world of artistic activity which constrains the range of choices and provides motives for making one or another of them. Sociological analysis of that context is well-equipped to explain the constitution of the range of possibilities and the conditions that surrounded and thus might explain the actual choices made.

Imagine an artist who is free to choose anything at all: a composer who could choose any kind of sound that it is possible to make with any kind of instrument or voice, any arrangement of any number of tones between the octaves (from, say, five to forty-two), any kind of composition of any conceivable length, and so on. Keep that extreme case in mind and now consider an actual work created by a real person or persons, who work under all the constraints of an existing art world. Although the sociological analysis of that world will not reveal all the secrets of a particular composition or performance, it will tell us a lot about what the performer or composer did, why they did it that way, and what the likely effect on audiences with a particular training and background might be.

Sociologists could do this sort of analysis of particular works, but in general don't. That does not mean that such analyses cannot be done. But the failure to do them requires explanation. I had originally thought, when I planned this paper, to conclude with a sociological analysis of a famous jazz work, Lester Young's well-known improvised solo on the chords of "Lady Be Good," recorded with the Count Basie Orchestra on October 9, 1936. I had the recording of this solo, so that I could play it for you, and a notated transcriptions of it to hand out so that you could follow the analysis.

I almost immediately abandoned this plan. Why? To make such an analysis requires an enormously detailed knowledge of the work and of the organized context in which it was made. I quickly realized that I did not know enough to make the analysis I believed was in principle possible to make. I knew enough practical keyboard harmony, and had enough experience playing in such settings, to understand the musical innovations Young had made in these solos, which are famous exactly because what Young did in them harmonically and rhythmically innovations marked a new era in jazz improvisation. If I had not, I could have relied on the analyses of Gunther Schuller (1989, 230-35), who pointed out, among other things, Young's increased use of sixths and ninths as important melodic elements, and his effort to break up the regularity of two and four bar phrases with motifs that lasted three or five bars and began at unusual places in the song's structure.

What I did not know, well enough to create a full analysis (with, I should say, the six months or so warning that I was to give this paper at this colloque), was the full historical and social context of the development of Young's style and of the conditions of production of this particular solo. Keep in mind that jazz is in some ways a particularly fugitive art. Jazz players play, night after night, in a variety of places, the same tunes over and over, experimenting endlessly with choices of notes, of rhythms, of architectures of the solo. No one makes a careful documentation of the conditions of performance in the various clubs and dance halls in which the Count Basie band, with whom Young then played, appeared. There are no well-preserved scores, with alternate possibilities notated in the performer's hand.

These conditions and experiments can be recovered, but only by a painstaking process of interviewing and archival research, looking for the people who were there and can talk about what went on and for the ephemeral recordings of such events, more of which exist than one might imagine. Thus the great interest in alternate "takes" of the same tune from the same recording session. It is possible to acquire that kind of detailed knowledge, and thus to make a sociological analysis of particular works "in themselves," as several recent works of jazz scholarship demonstrate.

I will say only a few words about these exemplary works, indicating what I might have done but did not in fact do. Paul Berliner, in his Thinking About Jazz (1994), describes (on the basis of detailed interviews, intensive observation, and musical analysis of recorded improvisations) the moment to moment construction of particular solos and the relation of that work to the recorded and unrecorded tradition of jazz soloing. Though Young is no longer alive, I might have relied on the many published interviews the jazz literature makes available to a dedicated scholar for at least some hints.

Scott Deveaux's The Birth of Bebop (1997), on the basis of similar evidence, shows how the bebop style, and particular examples of it, arose from the intensive interaction of a small group of players who were, for historical and organizational reasons, thrown together in New York over a considerable period to time and thus could produce not only new individual ideas, but a new style, collectively recognized and performed. As one instance, he describes how Dizzy Gillespie taught what he had discovered from his experimentation with the half diminished chord to a widening circle of players, so that the chord became a standard resource for jazz players generally (I can remember my own excitement at learning of this "invention"). I could conceivably have done something similar with the conditions surrounding Young's inventions, though most of the players from that time are no ;longer around for interviewing and reminiscing.

Ingrid Monson's Saying Something (1996) gives a very detailed account of the way a jazz rhythm section constructs the background for a soloist's improvisation, including even an explanation of what happens when one of the players "gets lost," does not know where he is in relation to the structure of the song that provides the basis for the work, and I might have done something similar with the work of the famous Basie rhythm section which accompanied Young on "Lady Be Good."

But I did none of those things. That is not what sociologists do. And, to be truthful, it is not what we want to do. We almost never attempt to explain a single instance of some kind of behavior. No, we look for an explanation of a kind of behavior which happens repeatedly, to which we give the name "social structure" or something similar, which connotes regularity, durability, and predictability. And we do not customarily. Nor do we consider it worthwhile to gather the kind of highly detailed accounts of the genesis of particular works that are what specialists in fields like musicology or art history or literature routinely collect. We would rather learn a little about many instances of something, or at least much less about any particular piece and more about many of them. Which doesn’t mean that, if we gathered that kind of information, we could not provide a sociological understanding of "the work itself" that would be coordinate with those provided by other varieties of analysis, formal, emotional, philosophical.

Well, what of the examples I gave a moment ago? Berliner? Deveaux? Monson? Now I will tell those of you who are not experts in the very small field of the sociology of jazz an unpleasant truth: none of the authors whose analyses of particular works I have just cited as exemplary is a sociologist. All three are musicologists and/or ethnomusicologists. But that does not mean that what they have done is not sociological analysis. A sociological analysis does not have to have been made by a card-carrying sociologist, a dues paying member of the American or French (or some other) Sociological Association. I am very imperialistic in this regard, and recommend this policy generally. I always count as sociologists people who make analyses of the kind I've suggested, even though their disciplinary affiliation may be "musicologist" or "art historian." Some of the best sociology of the arts, as is well known, has been done by such non-sociologists as Michael Baxandall and Barbara Herrnstein Smith.

So, in a way, the premise of this colloque--that sociologists do not make analyses of l'œuvre elle-même--is false. We do. Some people who do may not even know that they are sociologists, but the rest of us who are officially sociologists, if we are clever, do know and will be proud of our colleagues and profit from their example.
 
 

References

Berliner, Paul F. 1994. Thinking About Jazz: The Infinite Art of Improvisation. Chicago: University of Chicago Press.

Deveaux, Scott. 1997. The Birth of Bebop: A Social and Musical History. Berkeley: University of California Press.

Monson, Ingrid. 1996. Saying Something: Jazz Improvisation and Interaction. Chicago: University of Chicago Press.

Schuller, Gunther. 1989. The Swing Era: The Development of Jazz, 1930-1945. New York: Oxford University Press.



El diseño de la página y las imágenes son
© 1996-2000 Grupo "mmm"
Comentarios a: © Dr.Vicent Fores
València  15th September 2000