108 : CHAPTER SIX
I don't insist that we aspire to this level of realism in telling users about society. But not to do that is a choice. We could do it if We thought it was important enough, and being aware of this possibility makes us realize that every choice of what to include or leave out is in fact, a choice, not a necessity forced on us by theoretical or practical impossibilities. (I'll take a longer look at the possibilities of dramatic representations in chapter 12.)
Why Do We Believe It?
I twice cotaught a course with Dwight Conquergood at Northwestern University called "Performing Social Science." We wanted to explore the possibilities of communicating social science ideas via public performance (other than the routinized performances of a scholarly "talk"). Twenty of our students came from Dwight's department, Performance Studies, and its neighboring department Theater, and half from social science, mostly sociology. Neither Dwight nor I had much idea about how to perform social science, and we counted on the students' inventiveness to give us something to work with. We gave them a simple assignment: perform something that might, under a very loose interpretation of the term, be called "social science."
The students' inventiveness exceeded our hopes. I had the feeling that they had all done the only thing they could think of—but no two did anything similar. One performance raised the problem of this chapter in an acute and interesting way.
The class had argued at length about the importance of the truth of the material you presented in your performance. Did it matter if it was something that had really happened? What if you tarted up the details to make it more "dramatic"? Or presented a result that had been disproved? Not surprisingly, the social science students insisted that the material performed had to be true; if it wasn't, how could you call it social science? And the students from theater and performance thought the truth of the material didn't matter if people responded to the piece as an aesthetic work. The arguments were heated. A traitor to my own people, I said that the truth might not matter.
110 : CHAPTER SEVEN
For many of the students' performances, this question didn't arise. One student simply read from an article in the American Sociological Review reporting correlations between expenditure on education, race, and income in some school system. He did something simple but effective: he read the article "with feeling." When the text said that there was "fully a twelve per cent difference" between educational expenditures on blacks and whites, he said: "There is fully! a twelve percent! difference," his voice rising in high-pitched outrage as he indicted the variables "correlated with" (for which the author had clearly meant us to understand "were to blame for") the discriminatory results. The emotional reading exposed the ideological subtext of. the sober scholarly report. Most interestingly, while the student's recitation sounded a little silly, it didn't sound "wrong." He hadn't misplaced the emotion, nor did he misrepresent what the author intended; he had just brought it to the surface and made it evident. No one questioned the truth of the findings or of his assertion that an article in the journal had actually contained the words he spoke.
But some performances did raise the question of truth. Tom, an ingenious and mischievous theater major, came into the room and passed out 3X5 cards to everyone. Each card had a woman's name written on it. He told everyone to look at their card and then to ask him whatever they liked. "Who is Mary Jones?" "She was my first-grade teacher." "Who is Betsy Smith?" "She's the first girl I ever kissed." "Who's Sarah Garfield?" "She's my aunt. She's married to my mother's brother." And, after a beat, and in the same conversational tone, "She and my father have been having an affair for the last five years." Someone immediately asked, "Is that true?" Tom considered this judiciously, then said, "I don't think I'll answer that question," and grinned. The room exploded. And, oddly enough, the theater and performance students insisted, much more than the social scientists, that the truth, goddamnit, did matter. They insisted that he tell them, and he wouldn't. These were the same people who, two days earlier, had said the truth didn't matter.
I pointed out the inconsistency and insisted that these defenders of "it doesn't matter whether it's true" had just proved to us that it does matter, even for an aesthetic work, and that we'd better devote our-
REALITY AESTHETICS : ill
selves to understanding how it matters and how to communicate the truth of what we performed.
Users care about the truth of what they're told, even when the message comes in an artistic genre, and certainly when it's science, and makers incorporate reasons for users to accept what they present as true in their work. But all these terms are ambiguous.
Truth: Questions and Answers
Is it true? The question, so filled with philosophical traps, becomes more tractable if we put it more simply as a problem of questions and answers. I'll use the example of documentary photography to keep the discussion down to earth, and I'll begin with these premises:
1. Every photograph can be interpreted as the answer to one or more questions.
2. We care whether the answer the photograph gives to our questions is true.
3. Every question we ask of a photograph can be put, and therefore answered, in more than one way.
4. Different questions are not the right or wrong way to ask (or answer); they are just different.
Saying that we can interpret photographs as answers to questions doesn't mean that we always do, only that we often do, in principle we always can, and that is a useful way to think about photographs. We can ask simple descriptive questions: What does Yosemite look like? What does the Republican candidate for president look like? How did our family and friends look in 1957? Or historical and cultural questions : How did people make photographs in 1905? How do the Yoruba make them? What did the battlefield at Gettysburg look like? Sometimes we ask scientific questions: Is this lung tuberculous? What happens when I bombard an atomic nucleus this way? Or psychological ones: What is the true character of the Republican candidate for president? Sometimes we ask for an abstraction: Tell me the essence of virginal innocence, or Mexican peasant life, or the urban experience.
Different people can ask different questions of the same photo-
112 : CHAPTER SEVEN
REALITY AESTHETICS 1 113
graph, not always the question the photographer had in mind. Some questions interest many people, who ask it in the same way. News photographs answer common questions about current events. Scientific photographs answer questions arising from the common concerns of a narrower professional community. Its members ask the same questions and find the same answers in photographs offered as evidence.
Other questions interest a very small circle, because they ask about personal relations and personally experienced events of no concern to most people. A photograph of me in front of the Eiffel Tower interests only me and mine. But pictures that once had only personal interest can. years later, answer questions interesting to a larger audience: childhood snapshots of people who later become famous or photographs of places in which events of general interest later took place.
We care that photographs that tell us about society give credible answers to our questions. Different people ask different questions of the same photograph. (Chapter 11 shows how this can be done from the perspectives of documentary photography, photojournalism, and visual sociology.) So there is no general answer to "Is it true?" We can say only that its answer to a particular question is more or less believable.
When we interpret a photograph as saying something about some social phenomenon, we suggest an answer to a question that might have a different answer. That raises the problem of truth. Because questions about society involve interests and emotions, people can disagree about the answers, often suggesting that they're not credible because the photographs are biased, misleading, subjective, or an unfair sampling.
Many problems arise over this ambiguity: A series of photographs suggests X is true; we don't deny it but think Y is true too. Do the photographs suggest that X and only X is true, or allow for the possibility that, while X is true, Y is true too? Specifically: many people think Robert Frank's book The Americans (1969; a sequence of eighty-four photographs made all over the United States in the 1950s) tells us that American life is bleak, nasty, uncultured, and materialistic—and that that's all it is. Without becoming an apologist for "the American Way of Life," it is possible to cite images by other photographers giv-
jfla a different view. Does Frank's book suggest that this is all there jS to American life? The book's length, long enough to allow for the inclusion of a greater variety of imagery, prompts that interpretation. If it does suggest that this is all there is, you could say the statement is wrong, because other kinds of evidence exist. (Photoessays can be seen as a kind of specified generalization. See my discussion of John gerger and Jean Mohr's A Seventh Man [1975/1982] in Becker 2002.)
Skepticism and the "Plenty Good Enough" Criterion
Suppose we do believe some of what we are told. Some skeptics won't accept that and will call attention to the shakiness of all knowledge about society, reminding us that all statements purporting to communicate such knowledge rests on a basis of "facts" that have been selected and interpreted in ways so hopelessly biasing the results that you just can't believe anything. If so, there's nothing to talk about and we can call the rest of this off.
People who talk that way don't really mean that they don't believe any representation of social reality. Do they believe, for instance, in the telephone directory, which presents itself as a more or less accurate listing of who is at the other end when you call a number? Skeptics might point to the inevitable errors committed by the clerical workers who type in the original information, or to the errors that arise because of changes that occur between the time of information gathering and printing, and between that time and the time you consult the list, or to the refusal of some people to be listed or to be listed under their real name. But those skeptics probably, like the rest of us, use that list and its numbers, for lack of anything better. The data aren't accurate, but they are "plenty good enough" for the purpose we will put them to, which is to call someone.
The same thing is probably true of a map of city streets, which purports to tell you how to get from here to there by using surface streets with names and numbers. With all its inaccuracies and omissions, it's good enough for most people's purposes. When a cabdriver turns on the overhead light and looks in a street guide for an address, the ad-
114 : CHAPTER SEVEN
dress is probably there and the way to get to it is probably more or less clear. If I want to drive from Seattle to San Francisco, and then to a particular address in San Francisco, a few state maps and a city map will show me the way. The maps will not show where the hills are in the city (though they will indicate the height of various mountains and mountain passes the highway goes through), but they will get me where I'm going. "Plenty good enough," knowledge good enough for what I want to do with it.
What about the US. Census? This is more complicated, because many people use the census for many purposes, and while it's good enough for some people and some purposes, it's not good enough for others. It wasn't good enough for several purposes when the i960 enumeration seriously undercounted young black men by as much as 20 percent. That miscount wasn't good enough for the constitutionally required apportionment of seats in the House of Representatives and of electoral votes. It wasn't good enough for the calculation of crime rates, because undercounting the denominator of a fraction like the crime rate inflates the rate over its true value. If you don't count all the people in a particular population category, such as "young, black, and male," but you count all the criminals who fall in that category, the resulting ratio will be larger than it would be if you had an accurate count of the denominator. This undercount had political consequences, as well as mucking up social science thinking and research with faulty data.
Such flawed findings might once have been good enough, at least for people who were in a position to make that judgment in an effective way. But now new people began making their own assessments, and it wasn't good enough for them. Accepting a number that affects congressional representation because it's "plenty good enough" has a political component.
Which is not to say that science is "all politics" or that all episte-mological questions can be settled by political means. It does mean that when you look at even so scientific an operation as the census, some of what's done has no "scientific" warrant but rests on an agreement among interested parties to treat something as good enough for some purpose, flaws and all. Users accept the resulting description
REALITY AESTHETICS : 115
n0t because it has an incontrovertible epistemological basis but because it's better than nothing for something they want to do.
So we all believe some of these representations all, or most, of the time, and some of us believe some of what we are told some of the time. No one disbelieves all of it all the time. Even with all these troubles, users treat representations as "essentially correct," which is the way physicians talk about laboratory findings that, they know perfectly well, have many errors built into them but are "plenty good enough" for what they will use them for.
But user communities ask different questions and use the answers for different purposes, and what's good enough for one won't be good enough for another. My map doesn't have to be accurate to the nearest foot, because I'm just using it to get to my friend's house. If I were using it to settle a property dispute, I'd need a different kind of geographic knowledge expressed in a different way. The two uses and the two question-answer pairs aren't competing with one another to see which is most accurate or "best"; they are different animals in a different environment.
As an epistemological judgment, "plenty good enough" has no philosophical justification. It's a social agreement based on another kind of justification. That doesn't make all knowledge totally relative, though. Once users enter into that agreement, they can and do arrive at reliable conclusions by following the agreed-on rules of evidence.
The Social Agreement to Believe
What justification does "social agreement," as a way of creating social knowledge that is "plenty good enough," have? For one thing, everyone accepts these agreements, and much work in the particular field has been based on them with no apparent ill effects. Latoufs (1987, 21-77) parable of the doubter who questions a scientific result explains this mechanism. The doubter arrives in the scientist's laboratory demanding evidence for what everyone else in the lab accepts, refusing to "believe" in what is well attested to in the literature and by the use of accepted tools and techniques—and his questions become so ludicrous that no one takes him seriously and he finally slinks
116 : CHAPTER SEVEN
ignominiously out. Which leads to Latour's rule of method: believe in scientific results just as much as the scientists do, but no more than they do.
That's not an epistemological judgment either, it's the practical judgment that if you start doubting what everyone else believes, it's likely that you'll be eliminated from the dialogue altogether as a nut. But you can doubt what others will accept as possibly dubious.
Further, the social agreement allows scientific work for any kind of collective activity) to proceed, which is no small thing. Thomas Kuhn (1970, chap. 3) made the point in connection with episodes of scientific progress: the only way any science ever gets done in is that workers in a field agree to concentrate on one or a few related problems, which they all approach the same way The premises of the approach may be false, but work can proceed when everyone agrees and can't when everyone is working on different, idiosyncratically defined problems. Agreement on a paradigm lets researchers do collectively whatever they're going to do.
More generally, we could say that people who make and use a particular kind of representation (a film or table or novel or mathematical model) have come to some agreement as to what will be "plenty good enough" for their purposes. Plenty good enough for the purposes of the makers, whoever they are and whatever their interests are, and plenty good enough for the users, whoever they are and whatever their interests are. Not perfect, not as good as everyone would like, but good enough, given the circumstances, to rely on for guidance.
Participants in a representational world agree on an object that everyone involved knows how to make, read, use, interpret, discount. It's what John Hersey, as we will see in the next chapter, claims about journalism, when he says that of course journalists leave relevant facts out of their stories but since everyone knows they do that, no one minds. Readers just discount for that source of error as they read.
Where such an agreement exists, we believe the statements made by an object that bears the marks of living up to it. If it shows, in its presentation, that it was made the way users and makers have agreed on as the way to make things like that, then the results will be plenty good enough for the agreed-on purposes. If it's a documentary film,
REALITY AESTHETICS : 117
there's no fiction in it. If it's a statistical table, it follows agreed-on procedures guaranteeing a user that appropriate safeguards have been taken and potentially misleading practices have been avoided (e.g.,
area of the bars in a histogram are proportional to the numbers the bars represent). If it's a "realistic" novel, it doesn't include factual stuff that, if you look into it, isn't factual.
Is my characterization of representational activity itself true? Do representational worlds work that way? All the time? Some of the time? Now and then? The answer isn't "all the time," because every trade that produces reports about society is usually rocking with some land of conflict over exactly what I've described, afew paragraphs ago, as matters of peaceful agreement and harmonious consensus.
Criteria of Believability
Whether to believe what you're told and why are matters of agreement. Fair enough. But what criteria of believability, specifically, do people accept and use in everyday life?
We frequently compare what we are told with our own experience of life. We all have plenty of that and are usually unwilling to believe anything we're told that runs counter to it, at least until we are given pretty good reasons to change our mind. If what we are told resembles our own experience, we accept it. People who have experienced recreational drug intoxication typically dismiss die exotic fairytales people with no firsthand experience believe. Their own experience tells them that smoking marijuana has not made them crazy.
We evaluate what we are told in the light of other knowledge we have of a more academic or secondhand sort. If we have read a lot about Russia and what we read here is congruent with that, then OK, we'll believe this too.
We imagine the method the maker probably used to get what we're being told and then criticize that method. We don't believe what someone with no firsthand knowledge of an event or activity says about it.
Since makers don't always give this information, users reconstruct it, if necessary, from fragments. A friend complained to me about David Remnick's descriptions of Russian politics and, by extension,
i
118 : CHAPTER SEVEN
about the New Yorker school of reportage, which he described thus-"They just go in with a tape recorder and write down everything they're told, which is aimed at an American audience, and then string it together." I disagreed, feeling sure that, for instance, Remnick spoke' Russian fluently, though 1 couldn't say why I thought that, and that it counted toward the believabifity of what he wrote that he seemed knowledgeable about Russian literature and history.
We also reconstruct the methods that make a report believable from our understanding of what someone would have to do to get "good stuff." We are suspicious of people who have visited some place = for a few days, don't speak the language, and have an explanation for everything. The proverbial Life photographer, parachuted into wherever-it-is for a few days and then airlifted out, is not a believable documenter, for some of us, of that place's way of life.
We judge believability from the consistency of what we see and hear. Anna Deveare Smith described riot situations in Brooklyn and Los Angeles on the basis of long interviews with participants, which she reenacted for an audience (Smith 1992,1993). We put together a picture of the chaotic event from the fragments she gives us, the little bits of testimony offered by many different participants. Gradually," we acquire enough knowledge to crosscheck, however crudely, remembering that if this one said that it happened this way, then there's a conflict with what thaf one said and we'd better be wary. (The playwright Caryl Churchill, as we'll see in chapter 12, uses a similar method to create a theatrical report on a major political event from interview fragments.)
Passing Tests
In all these procedures, users compare the representation to something else they already believe and sees how it stacks up: is it congruent with what I aheady know and believe? The representation has to prove itself in competition with what's already on the accepted list. That's a version of a process that Latour speaks of as undergoing "trials of strength" (1987, 53-56, 74~79> 87-94). ln this way: if the representation suggests a conclusion or fact that isn't congruent with what I
REALITY AESTHETICS : 119
j-oow or t>e'ieve> it na5 to Pass a 'ot of tests and find allies in other reports and sources before I will believe it.
Many representations do convince us to accept facts we didn't accept before. So this feat can be accomplished, just as, despite the obstacles to acceptance that Latour describes, new scientific facts do get accepted. But a maker doesn't achieve that just by announcing the new idea or fact or interpretation. Skeptical users insist on tests.
Makers can construct representations to produce the effect of obstacles overcome and tests passed. The standard journal article does this by providing all tire facts conventionally required in a standardized format, allowing skeptics to convince themselves that all the potential sources of error have been avoided and all the potential sources of information investigated. The idea that investigators should guard against "threats to the validity of their hypotheses," formulated and propagated by Donald Campbell and his colleagues (Campbell and Stanley 1963; Cook and Campbell 1979), is a systematic way of listing what has to be dealt with.
You can also produce the effect of obstacles overcome and convincing proof produced by using data so widely accepted that the possibility they could be false doesn't arise. Hans Haacke's "Guggenheim Project," mentioned earlier, rests on easily verified facts about the trustees of New York's Guggenheim Museum, including their names and addresses, their family connections (they are mostly all Guggen-heims, whatever their names), and the other organizations on whose boards they sit (large multinational mining corporations). Finally, we learn that Chile's soon-to-be-dead (assassinated or by his own hand) president Salvador Allende had made the mistake of confiscating properties belonging to one of these companies.
There's nothing to argue about factually. Any user can easily check everything stated in these panels in any well-furnished library or by using Google. But the user needn't check them, because it's obvious that, were the facts not as Haacke states them, someone would have said so. People who found Haacke's work distasteful (there were plenty) would have been glad to dispute any disputable facts. The reliance on what is publicly available disarms mistrust and suspicion. You can criticize the reasoning, but that's all. Here Haacke uses the
120 : CHAPTER SEVEN
REALITY AESTHETICS I 121
ploy discussed earlier, leaving all the reasoning and conclusion draw- . ing to users, who do the rhetorical work, convincing themselves that the conclusion is justified. •-•>••.
Natural scientists get very upset when they are told that they just"; "believe" things rather than having "discovered the truth." They thinJc this means that their proofs have no real epistemological justification, " j that anyone can in principle believe anything they want to, and that • i such a stance, opening the door to rampant mysticism, is the death of ] real science.
George Polya, a mathematician, argued that compelling proof, the kind scientists like to think is provided when Nature speaks (Latour 1987,94-100), is available only in the fields of logic and mathematics, V which make no reference to the empirical world and whose truth rests on demonstrative logic; what is true is so by definition and by _ logical deduction from those definitions (Polya 1954,140-41). Everywhere else in the empirical sciences, and not just the weak social sci-ences but the strong natural sciences as well, we can only estimate degrees of credibility or believability, and those only roughly.
Polya illustrates how scientific conclusions are contingent on evidence with a small detective story. A yacht explodes. We discover that ~~~~ the owner's son-in-law, with whom he does not get along, bought some dynamite a week earlier, and so we think it likely that he "did it." But then we discover that the son-in-law used all the purchased dynamite to blast out a tree stump in his backyard. That makes it less likely that he did it. And so on: each new bit of evidence changes our assessment of his guilt.
Empirical science, Polya says, works like that. No matter how well proved a statement may seem, new facts can always cause us to reconsider our belie! Latour calls well-established conclusions "black boxes," like the ones in computer science, whose workings we no longer inquire into but just accept their outputs (derived from our inputs in ways we don't inspect and may not understand at all) as reliable bases for further work (Latour 1987, 2,131).
Instead of definitive knowledge, empirical science gives us degrees of credibility and procedures for agreeing on them. Polya provides a collection of diagrams (1954, 3-37) showing how different empirical
results produce different degrees of credibility. When you inspect them, you see that they codify the practices of your own reasoning. Scientists needn't get upset about this, because the analysis of degrees 0f credibility shows that scientists can use these procedures just as they use the criterion of truth. Nothing changes in the day-to-day world of scientific work if you do that; in fact, that is how scientists work, talking about "truth" in order to deal with the rest of the world gXid convince nonscientists that science is, after all, worth supporting.
So when we discuss alternative formats for the presentation of social science ideas, conclusions, and research findings, we should look tor the procedures people use that lead them to find different kinds of reports more or less credible.
Aesthetics
It's not clear what "aesthetic" means when we consider representations as objects conveying information and ideas about society. From a purely "artistic" point of view, it might refer to what are usually spoken of as the formal aspects of the object: the harmony or balance displayed by the relations between its parts. We often just use vague words like beautiful, by which we mean things like gorgeous sunsets or natural landscapes that appeal to us in so obvious a way as not to require explanation; it's enough to point and say "Wow!" to register our response. Others will know what we mean. That kind of judgment would never pass muster among serious students of aesthetics, who require a more philosophically defensible justification of our responses and criteria of judgment (Becker 1982,131-64).
Lets consider the criteria makers and users apply to representations of social reality that could in some way be taken as "aesthetic."
Even the most relentlessly realistic representation, we've seen, results from the selection and reduction of material to be represented, the translation of the raw material of experience into the language of the maker's medium, and the subsequent arrangement of the translated items. Are there better and worse ways to perform these operations? Is there craft involved, and if there is, is it being done as well as it could and should be? That's the kind of tiling people discuss as the
122 : CHAPTER SEVEN
aesthetic issues of representing society. Let's consider some common criteria of aesthetic worth.
Craft standards influence judgments, perhaps more among makers than among users, and certainly more than anyone would like to admit. Does the author write properly elegant prose? Critics disapproved of Theodore Dreiser's novels of urban America for his failure to meet that standard. Are the photographs in focus and printed appropriately? Photo critics of the 1950s complained that the images in Robert Frank's The Americans ([1959] 1969) didn't exhibit a full tonal range, from a pure black to a pure white with as many shades of gray as could be managed in between—an aesthetic criterion that Ansel Adams, the photographer of Yosemite, had successfully installed in the world of high art photography. Did the journal article fail to cite "relevant literature"? Many studies fail to be published in the best journals only to reach their users via book form, which doesn't require meeting that particularly restrictive set of craft standards (I speak from experience). You can find similar examples in every artistic and scholarly form.
This problem arises in an illuminating way in the difficulties of making films in a verite style. Verite requires the filmmaker to avoid pushing the people being filmed around too much and instead to let them do whatever they would do in the kind of situation being filmed if the filmmakers weren't there and to shoot what happens as best you can given those constraints. But the resulting footage is often badly lit, out of focus, and in other ways not up to "professional standards "
Editors, the people who have to make cinematic sense of documentary films, complain that the results of verity shooting do not give them the material they need to create the sense of continuity, or continuous intelligible flow of the action, that marks a properly edited film. Because the filmmakers used just one camera or because they did not anticipate what might be needed, the editor may have no material for a "cutaway," in which you break up a continuous shot of someone talking, for instance, so that you can use just a few parts of it, by going to a shot of something else that covers the discontinuity the jump from one part to another might create. Or the editor may not have an available "establishing" shot, which prevents viewer confu-
REALITV AESTHETICS : 133
sion by showing them where the action is taking place. Charlotte Zwerin. who edited Salesman, a classic documentary about a Bible salesman made by David and Al Maysles, explains to an interviewer:
As I began editing I found I needed certain establishing shots, and Al went back and got them for me. I think these shots included things like the exterior of the motel in Boston, and some stuff around the Florida motel.
[I'm interested in the question of continuity in editing verite style films. Did you find this much of a problem in Sa/es»ian?]
Sure, it was murderous. Al goes into a place and has so many things to think about—lighting, reasonable camera angles, how to shift position without falling all over everybody—that, consequently, he can't really consider how it's all going to edit smoothly. . . .
[Can you give me an example of a sequence that was hard to edit?]
One of the funniest but also most difficult scenes, was when Charlie and the Rabbit go in to sell a Bible to this old Irish lady and her daughter in Boston. The two women were marvelous characters and very amusing, but the sequence drove me mad for a couple of months because Charlie and the Rabbit kept shifting around from the piano, to the coffee table, to the door; they went all over the place and Al didn't do a thing about it. He obviously couldn't tell them to sit or stay rooted to one spot, but every time Al cut away it looked as if Charlie and the Rabbit had gone to another house. The lighting of the room was also maddening and didn't help the cutting, The salesmen were in one end of the room which was black, and were wearing dark clothes, while the two women were sitting on the couch wearing very light clothes in a very light situation. There was no room to get back and get an establishing shot, and after looking at the rushes I was left with the feeling that the two groups weren't even in the same room. (Zwerin 1971, 90)
Because verite produces these results, the very "imperfections" become guarantees of the "authenticity" of the film and convince viewers to take the film as "true," even when the blurs and shakes did not actually result from the conditions of shooting. Gillo Pontecorvo's 1966 Battle 0/Algiers was entirely a movie fiction, artfully rehearsed
124 : CHAPTER SEVEN
"events" with paid actors and extras, but it mimicked the imperfe tions of documentary footage so perfectly that audiences had a hard time accepting that they were not seeing newsreel footage of real in
cidents from the uprising that led the French to leave Algeria. ---
Criticism based on craft standards permeates the making of almost all representations of society. Most representations are made by people who belong to some craft community. That community main- '; tains standards of acceptable craft, and its members criticize products that do not meet them. Makers of representations accept those standards and apply them to their own products, knowing that their craft peers will criticize any failure to meet them. And they will share that' -judgment. They try to live up to those standards, even at the expense of some other value they want to maximize, like "documentary truth " Users who want truths about society, not just an entertaining movie worry that filmmakers may sacrifice truth, however defined, to craft standards.
How do these considerations affect the work's truth value? Does making a scene dramatically effective simultaneously make it impossible to have it tell some kind of truth?
When we consider representations that, at least in part, describe—1 social life and social events—in the realm of "art," that includes photographs, films, novels, and plays—we find ourselves dealing with a criterion different from the perfection of formal relations mentioned earlier. In these cases, we're interested in the relation of what the work depicts to the "real world," in the truth or accuracy of what the work tells us about social reality. We take the work seriously, in part, because it claims to tell us something we didn't know before about some aspect of society.
Dickens's prose is magnificent, his plots complicated and engrossing, his characters memorable. But an important part of the effect of ... his later novels rests on our belief that they tell us the truth, however caricatured, about the social and economic institutions of Victorian England. Imagine, as an experiment, that historians, working with masses of court records, discover that lawsuits did not drag on for years, like jarndyce v. Jarndyce in Bleak House, until the lawyers had gobbled up all the money involved in fees. We would feel differently
REALITY AESTHETICS : 1K5
about the novel, regard it as a fantasy rather than a realistic account 0f events that might well have happened, and probably judge it a smaller achievement. We could not take what we read there as fact of some kind on which to base a response to social conditions, we could not answer questions about Victorian institutions credibly, and we vvould not find the plot and characters so affecting. It would be a different book, even though the words were all the same.
Dickens apparently feared that some, refusing to believe that British courts could behave so badly, might think he had invented his story. Proud of the accuracy of his reporting (he had, after all, been a journalist), he defended himself, in a preface to Bleak House, against such a charge and insisted on the story's substantial truth, which he evidently thought necessary to the book's aesthetic success:
Everything set forth in these pages concerning the Court of Chancery is substantially true, and within the truth. The case of Gridley is in no essential altered from one of actual occurrence, made public by a disinterested person who was professionally acquainted with the whole of the monstrous wrong from beginning to end. At the present moment (August, 1853) there is a suit before the court which was commenced nearly twenty years ago, in which from thirty to forty counsel have been known to appear at one time, in which costs have been incurred to the amount of seventy thousand pounds, which is A FRIENDLY SUIT, and which is (E am assured) no nearer to its termination now than when it was begun. There is another well-known suit in Chancery, not yet decided, which was commenced before the close of the last century and in which more than double the amount of seventy thousand pounds has been swallowed up in costs. If I wanted other authorities for Jarndyce and Jarndyce, I could rain them on these pages, to the shame of a parsimonious public.
Adam Hochschild deals with Joseph Conrad's Heart of Darkness, a classic exploration of the relations between Europeans and "others," that is, the indigenous people of countries that once were European colonies. In the story Mr. Kurtz, an agent of a European trading outfit, has gone wild and established a personal fiefdom along the course of the Congo River in what was then the Belgian Congo, eventually became Zaire, and is, as 1 write this, the Democratic Republic of
126 : CHAPTER SEVEN
REALITY AESTHETICS : 137
the Congo. Hochschild recalls a particularly gruesome image in th, novel:
Something we especially remember is the scene of Marlow [the narrator] on the steamboat looking through his binoculars at what he thinks are ornamental knobs on top of the fence posts near Kurtz's house and finding that each is "black, dried, sunken, with closed eyelids—a head that seemed to sleep at the top of that pole, and with the shrunken dry lips showing a narrow line of white teeth." Even many people who haven't read the novel remember the severed heads, because Francis Ford Coppola included a few when he transferred "Heart of Darkness" to the screen in "Apocalypse Now." (Hochschild 1997,40-41)
This bothers Hochschild because the book is now routinely read as being about almost anything but the reality of Africa, a reality Conrad had firsthand knowledge of, as Hochschild painstakingly documents.
Writers and academics have looked at the novel in terms of Freud and Jung and Nietzsche, of Victorian innocence and original sin, of patriarchy and Gnosticism, of postmodernism and post-colonialism and post-structuralism. But, as hundreds of monographs and Ph.D. theses pour out, with titles like "The Eye and the Gaze in 'Heart of Darkness': A Symptomological Reading," it is easy to forget that the novel was closely based on a real place and time. It is also easy to overlook, as almost all Conrad's many biographers have done, certain real people: several likely models for the novel's central figure, who is one of the twentieth century's most notorious literary villains — Mr. Kurtz. . . .
When scholars talk about the more bloodthirsty aspects of Kurtz, they often assume that Conrad made these up, or borrowed them from indigenous practices in the region. . . . Norman Sherry writes, "As for the shrunken heads on poles around Kurtz's house, these might be a macabre transference by Conrad of the fate of Hodister • [a Belgian active in the ivory trade at that time, who was massacred by rival traders who beheaded him] and his men."
We know from other witnesses that the local warlords along the river at this time did indeed display the severed heads of their victims. But did Conrad have to make a "macabre transference" to imagine
Kurtz doing the same? Sherry and others have chosen to ignore several other prototypes who share a feature of Kurtz that the critics prefer to think of as phantasmagoric: they were white men who collected African heads. (1997, 40-41)
Hochschild does not accept this as just the way academics go on about anything. He sees it as having a submerged political motivation:
Europeans and Americans have long been reluctant to regard the conquest of Africa as having been on the same genocidal scale as the deeds of Hitler and Stalin. For this reason, we find it more comfortable to think of Kurtz's head-collecting as a "macabre transference" and to locate the sources of this murderousness in Conrad's imagination. We have eagerly pulled "Heart of Darkness" loose from its historical moorings and turned it into a universal parable. The most macabre transference of all is our insistence on moving the novel out of Africa. [He cites film versions set in Spain and Vietnam.] Would we not think it strangely evasive if a director filmed Solzhenitsyn's "One Day in the Life of Ivan Denosovich" but didn't set it in the Soviet Union, or filmed Elie Wiesel's "Night" but didn't set it in Auschwitz? (1997- 46)
What's at stake here? Hochschild wants to see the book as descriptive, telling the truth about a particularly cruel practice the central European character engages in. He produces evidence to substantiate his claim and explains why others have ignored this crucial aspect of the book. He's taken something usually thought of as an aesthetically motivated invention and turned it into the simple factual report of something the author saw. And though he doesn't say this, you could take that as an element in the aesthetic experience of a reader who knows it—it shocks you by telling you that this is a way our kind of people really did behave when they had the chance and no one whose opinion they cared about wTas watching.
Hochschild's complaint suggests a general point. Many more works of art than we ordinarily so understand can be taken to be, and their makers very likely meant them to be, literal descriptions of some social fact, a verifiable description of a particular social organization at a particular time and place.
126 : CHAPTER SEVEN
We can go further. The presumed truth of an artistic representa tion of a social fact is an essential element in our appreciation of t{ work as art. That is, art and truth do not work at cross-purposes go-that you can have one or the other but not both. In a lot of works, y0u~ can only have both, or neither: no art without truth. The truth of the work's assertions about social reality contribute to its aesthetic effect ' That's why the class got so angry at Tom. If the story about his aunt arid-father was true, it moved and upset us. If not, it was just a silly j0ke No truth, no art.
8
The Morality of Representation
Representing society raises moral questions for participants, for makers and users. These come in several varieties: misrepresentation as a moral wrong; the way common techniques shape our moral judgments; the related questions of assigning praise and blame for the results of action and of casting participants in social action as heroes and villains.
"Misrepresentation"
Sociologists in my tradition routinely seek understanding of social organizations by looking for trouble, for situations in which people complain that things aren't as they should be. You can easily discover the rules and understandings governing social relations when you hear people complain about their violation. Fields of representational activity undergo periodic violent, heavily moralistic debates over the making and use of their characteristic products. The cries of "It's not fair" and "He cheated" would sound like the games of five-year-olds were the stakes not so much higher and the matters dealt with so much more serious. The problem of misrepresentation invites us to begin our analysis by looking for these conflicts.
Anthropology students at the University of Papua New Guinea complained, in the Nova program "Papua New Guinea: Anthropology on Trial" (Nova 1983), that Margaret Mead's Growing Up in New Guinea was "unfair" because it repeated the derogatory stories her informants had told about the students' ancestors, for whom the informants'