r Persona and Authority 42 it, by the time they achieve editorial positions; however, the receptiveness of editors may be a moot point since most journals are refereed. Perhaps some referees are receptive to this writing style, but perhaps most are not. Articles that are verbose, pretentious, and dull still abound in sociology. . . . We question the wisdom of advising students and faculty just entering the world of "publish or perish" to abandon the ponderous, rigid style of the discipline. .. . Currently, and in the probable future, graduate students . . . will "learn" to write by reading what is written. They generally find dull, verbose, pretentious writing, perpetuating the problem and suggesting that most referees expect such a stilted style (Hummel and Foster 1984, 429-31 [my emphasis]). KJ cholarly writers have to organize their material, express an argument clearly enough that readers can follow the reasoning and accept the conclusions. They make this job harder than it need be when they think that there is only One Right Way to do it, that each paper they write has a preordained structure they must find. They simplify their work, on the other hand, when they recognize that there are many effective ways to say something and that their job is only to choose one and execute it so that readers will know what they are doing. I have a lot of trouble with students (and not just students) when I go over their papers and suggest revisions. They get tongue-tied and act ashamed and upset when I say that this is a good start, all you have to do is this, that, and the other and it will be in good shape. Why do they think there is something wrong with changing what they have written? Why are they so leery of rewriting? 43 One Right Way 44 One Bight Way 45 It might be laziness. You might decide (chapter 9 discusses this) that it is physically too much trouble to do it again. You just don't feel like retyping a page or cutting-and-pasting any more. More often, students and scholars balk at rewriting because they are subordinates in a hierarchical organization, usually a school. The master-servant or boss-worker relationship characteristic of schools gives people a lot of reasons for not wanting to rewrite, many of them quite sensible. Teachers and administrators intend their schools' systems of reward to encourage learning. But those systems usually teach undergraduates, instead, to earn grades rather than to be interested in the subjects they study or to do a really good job. (This discussion is based on the research reported in Becker, Geer, and Hughes 1968.) Students try to find out, by interrogating instructors and relying on the experience of other students, exactly what they have to do to get good grades. When they find out, they do what they have learned is necessary, and no more. Few students learn (and here we can rely on our own memories as students and teachers) that they have to rewrite or revise anything. On the contrary, they learn that a really smart student does a paper once, making it as good as possible in one pass. If you really don't care very much about the work you are doing—if it is just a chore to be done for a course, and you have calculated that it is worth only so much effort and no more—then you might reasonably do it once and to hell with it. You have better ways to spend your time. Schools also teach students to think of writing as a kind of test: the teacher hands you the problem, and you try to answer it, then go on to the next problem. One shot per problem. Going over it is, somehow, "cheating," especially when you have had the benefit of someone else's coaching after your first try. It's somehow no longer a fair test of your own abilities, You can hear your sixth grade teacher saying, "Is this all your own work?" What a student might think of as coaching and cheating, of course, is what more experienced people think of as getting some critical response from informed readers. Joseph Williams suggested to me that students, being young, simply don't have the experience of life that would let them use their imaginations to get out of their own egocentric worlds. They thus cannot imagine an audience's response or the possibility of a text other than the one they have already produced. That may be true. But the lack of experience may result less from youth than from the way schools infantilize young people. Graduate students certainly appreciate the need for rewriting more keenly when, contemplating reading their paper at a professional meeting, they envision total strangers assaulting their logic, evidence, and prose. Such reasons might explain why people don't rewrite, but not the shame and embarrassment they feel at the thought of doing it. These feelings also originate in schools. No one connected with schools, neither teachers nor administrators, tells students how the writing they read—textbooks or their own teachers' research reports, for instance—actually gets done. In fact, as I said earlier (citing Latour, Shaughnessy, and others), the separation of scholarly work from teaching in almost all schools hides the process from students. (Just as, according to Thomas Kuhn, histories of science hide all the false turns and mistakes in the research programs that produced the successes they celebrate.) Students don't know, never seeing their teacher, let alone textbook authors, at work, that all these people do things more than once, rather than treating their professional work as a quasi-test. Students don't know that journal editors routinely send papers back for revision, that publishers hire editors to improve the prose of books to be published. They don't know that revising and editing happen to everyone, and are not emergency procedures undertaken only in cases of scandalously unprofessional incompetence. One Bight Way 48 Students think of their teachers, and the textbook authors their teachers stand for, as authorities for another obvious reason: these people stand above them in the school hierarchy. They are the bosses who give the grades and judge whether students' work is good enough. Unless students decide that the educational institutions they attend are frauds (and surprisingly few do, considering the evidence available to them), they will accept the implicit organizational proposition that the people who run schools know what they are doing. Not only, then, do their academic superiors—as far as they can see—never rewrite anything, they also get what they write "right" the first time. So students learn and really believe, at least for a while, that "real writers" {or "professionals" or "smart people") get it right the first time. Only dummies have to do it over and over. This might be another version of the test mentality: the ability to do it right the first time shows superior ability. This, too, is hierarchy, full-blown, at its worst: subordinates taking such evaluations as grades and teachers' comments, which are legitimated by the stratification of schools and scholarship, as ultimate and not-to-be-questioned evaluations of their own personal worth. (Becker, Geer and Hughes 1968, 116-128, detail the evidence far this interpretation.) All these ideas—about not rewriting, about the school paper as a sign of worth—rest on the fallacious premise that there is a "right answer," a "best way" to do things. Some readers will think I have invented a strawman, that serious students and scholars know there is no One Right Way. But students and scholars do believe in One Right Way, because the institutions they work in embody that idea. The ideas of the right answer and a best way find their natural home in hierarchy. Most people believe that the higher-ups in hierarchical organizations know more and know better than the people lower down. They don't. Studies of organizations show that superiors may know more about some things, but usually know a good deal less I One Rignt Way 47 i I about many others. They even know less about the L organization's central business, which you might sup- | pose they would know better. But the official theory of I the organization, and usually of its environing society, ignores such results, holding that higher-ups really do know better. What they know is, in fact, by definition the "right answer." 4 No matter that real authorities on any subject know [ that there is never one right answer, just a lot of \ provisional answers competing for attention and ac- J ceptance. Students, undergraduates particularly, don't ! like such talk. Why bother learning something that isn't \ true only to have to learn something else in its place [ tomorrow? Nor do true-believing scholars like it, I whether they have discovered the truth themselves, or i are only followers of the discoverers. The leaders of the I field must know. What they know is what's in the book. [ That is real hierarchy, seen most clearly when a chem- I istry experiment performed in class fails to produce the | "correct" result and the teacher tells students what t should have happened and what, therefore, they I should write in their notebooks. (Yes, that does hap- l pen.) [ If there is one right answer, and you believe that the £ authorities who run the institution you work in know I it, then you know that your job is to find out the right | answer and reproduce it when required, thus showing j, that you deserve to be rewarded, maybe even to become j one of the guardians yourself. That is the undergradu- \ ate version. A slightly more sophisticated version af- t flicts graduate students and professionals. Since what j? you are writing is something new, the One Right Way } does not exist, but its Platonic ideal exists somewhere \ and it is up to you to discover it and put it down on |' paper. I suppose that many of us would like readers to feel that we have found such a preordained right way to say what we say, one that looks as though it could only be that way. But serious writers discover that perfect form (that is, some form that does what they want done, One Eight Way 48 One Bight Way 49 even though not the only possible one} after lengthy exploration, not the first time. Harvey Molotch put the point like this in a note to me: A problem that writing people have is the idea in their heads that a given sentence, paragraph or paper must be the right one. Their training in a land of "facts," in the celebration of "right answers"—including the "right" way to approach their Chem lab book or English theme—immobilizes them at the typewriter keyboard. Their problem is that there are many right sentences, many right structures for an essay. . . . We have to free ourselves from the idea that there is only one CORRECT way. When we don't, the contradiction with reality absolutely stifles us since no sentence, paragraph or paper is demonstrable [to ourselves) as clearly the right one. Students watch their words come out, but of course these words— in first draft—are not even meeting the test of "OK," much less CORRECT and PERFECT ESSENCE OF CORRECT. Not having a vision of tentativeness, of first-draft, of n-draft, they can only feel frustration at the sight of failure. After a while, one sees the first tentative thoughts of a paragraph or paper as obviously failing this test— and so one doesn't even start: writer's cramp. The fear of failure is an accurate fear, because nobody could pass this self-imposed test of getting the one correct version, and the failure to do so is especially (and distressingly) evident at the point of first-draft. Some very common, quite specific writing difficulties have their origins in this attitude: the problem of getting started and the problem of "which way to organize it." Neither one has a unique solution to be discovered. Whatever you do will be a compromise between conflicting possibilities. That doesn't mean that you can't arrive at workable solutions, only that you can't count on finding the one perfect one that was there all along waiting to be found. f Most writers, even professionals, have trouble get- ting started. They start over and over again, destroying if reams of paper, working over the first sentence or i paragraph again and again as they find each successive ■fc. try unsatisfactory in some new way. They start that way r because they believe that there is One Right Way. If { they can only find the Right Way to begin, they think, [ everything else will take care of itself, all the other .<:. problems that they fear are lurking ahead of them will t; disappear. They set themselves up to fail, r Suppose I am reporting on my study of Chicago I schoolteachers. (I have immodestly used this ancient | document, my own Ph.D. thesis, as an example because ; I know it well, and because the problems it exemplifies ? still bother students, who find the solutions I discuss i helpful.) The study dealt with, loosely speaking, race, class, professional culture, and institutional organiza-■ f tion. How shall I begin? I could say: "Schoolteacher I culture defines lower-class, and especially black, stu- dents as difficult to work with. As a result, teachers avoid those schools, transferring to higher-class i schools as soon as their seniority makes it possible, and * that in turn means that lower-class schools are always I staffed by new, inexperienced teachers." Even though I am talking about a thesis completed and accepted in 1951,1 still have trouble writing a concise introductory sentence. (Imagine me trying to do it in 1951, when I | still wasn't sure what the thesis was about.) When I i look at the sentence I just typed, I might think, "Wait a c minute, do I really want to say 'schoolteacher culture'? After all, it's not exactly culture in the strict anthropo-| logical sense, is it? I mean, they don't pass it on from f generation to generation, and it doesn't cover all as- | pects of life, isn't really a 'design for living.' If I call it * culture, I'm sure to get in trouble, and I'll deserve it, | because I will be saying something I might not mean." < So I put that sheet in the wastebasket, and try again. One Right Way 50 I might substitute "shared beliefs" for "culture" and feel happier with that. But then I would see that I was talking about class and remember what a tangle of implications surrounds every one of the many ways sociologists talk about class. Whose version would I mean? W. Lloyd Warner's? Karl Marx's? I might decide to go back over the literature on class again before using such an expression. So I would put another sheet in the typewriter. But now I might notice that I had said "As a result of something teachers something-or-other." That is a pretty direct causal statement. Do I really think that social causality works like that? Shouldn't I use some less committing expression? In short, every way to say it would start me down some path I hadn't fully explored and might not want to take if I really understood what it would commit me to. The simplest remarks would have implications I might not like, and I wouldn't even know I was implying them. (Curious readers can see what I actually did by consulting Becker 1980.) That is why people make outlines. Maybe working the whole puzzle out in outline will show you where you are going, help you catch all the implications, evade all the traps, and get it all to come out right. You will find the One Right Way. An outline can help you get started, even if it won't find the Way, but only if it is so detailed as to be the actual paper whose skeleton it pretends to be. That just gives you the same problem in a slightly different form. Introductions raise the problem of unwanted implications in a specially difficult way. Everett Hughes told me, when I was still in graduate school, to write introductions last. "Introductions are supposed to introduce. How can you introduce something you haven't written yet? You don't know what it is. Get it written and then you can introduce it." If I do that, I discover that I have a variety of possible introductions available, each one right in some way, each giving a slightly different twist to my thought. I don't have to find the One Bight Way 61 One Right Way to say what I want to say; I have to find out what I want to say. But I can do that more easily after I have said it all and know pretty much what I mean than when I am writing the first sentence. If I write my introductory sentences after I finish the body of my text' tne Problem of the One Right Way is less compelling. Fearing commitment to the implications of an initial formulation also accounts for people beginning with the vacuous sentences and paragraphs so common in scholarly writing. "This study deals with the problem of careers" or "Race, class, professional culture, and institutional organization all affect the problem of public education." Those sentences employ a typical evasive maneuver, pointing to something without saying anything, or anything much, about it. What about careers? How do all those things affect public education? People who make outlines do the same thing by making topic rather than sentence outlines. The minute you turn the topic headings into nonvacuous sentences, the problems the outline solved return. Many social scientists, however, think they are actually doing a good thing by beginning evasively. They reveal items of evidence one at a time, like clues in a detective story, expecting readers to keep everything straight until they triumphantly produce the dramatic concluding paragraph that summarizes argument and evidence at once. They may do this out of a scientific prudery which forbids stating a conclusion before laying out all the evidence (which ignores the excellent example of mathematical proofs that begin by stating the proposition to be proved). Investigators frequently report survey research results this way. A table shows, for example, that class and racial prejudice are directly related. The next table shows that that is true only when you hold education constant. Further tables showing the effect of age or ethnicity complicate matters further, and so on down a long road of items before One Right Way 53 whatever conclusion the assemblage warrants finally appears. I often suggest to these would-be Conan Doyles that they simply put their last triumphant paragraph first, j. telling readers where the argument is going and what all this material will finally demonstrate. That flushes out the other reason for this caginess: "If I give the ending away at the beginning, no one will read the rest of what I've written." But scientific papers seldom deal with material suspenseful enough to warrant the for- ! mat. If you put the paragraph that gives the secret away at the beginning, you can then go back and say explicitly what each section of your work contributes to reaching that result, instead of having to hide its function in noncommittal prose. Suppose you are reporting, as Prudence Rains (1971) did, the results of a study of unwed mothers. You could, in classical evasive style, begin your book like this: "This study investigates the experiences of unwed mothers, with special attention to their careers, moral I-aspects of their situations, and the influence of social agencies." Giving nothing at all away, that beginning would leave the reader with a collection of unrelated tokens to be exchanged later in the book (if the author delivers on the I.O.U.) for sentences asserting real relationships between real entities. Fortunately, Rains did not do that. She wrote, instead, a model introduction, which explains exactly what the rest of the book then analyzes in detail. I quote it at length: Becoming an unwed mother is the outcome of a particular sequence of events that begins with forays into intimacy and sexuality, results in pregnancy, and terminates in the birth of an illegitimate child. Many girls do not have sexual rela- £ tions before marriage. Many who do, do not get pregnant. And most girls who get pregnant while unmarried do not end up as unwed mothers. Girls One Bight Way 53 who become unwed mothers, in this sense, share a common career that consists of the steps by which they came to be unwed mothers rather than brides, the clients of abortionists, contraceptively prepared lovers, or virtuous young ladies. The most significant aspects of this career are moral ones, for sexuality, pregnancy, and motherhood are matters closely linked to conceptions of feminine respectability and intimately connected to women's conceptions of themselves. Becoming an unwed mother is not simply a private and practical trouble; it is the kind of trouble that forces public accounting, raises retrospective questions, and, above all, calls into question the kind of person the unwed mother was and is. The moral career of an unwed mother is, in this sense, like the moral careers of other persons whose acts are treated as deviant, and whose selves become publicly implicated. Important, if not central, to the moral career of such a person are the social agencies with which he may come into contact as a result of his situation. Social agencies and institutions, whether geared to rehabilitation, incarceration, help, or punishment, provide and enforce interpretations of the person's current situation, of the past that led to it, and of the possibilities that lie ahead (Rains 1971, 1-2.). That introduction, laying out the map of the trip the author is going to take them on, lets readers connect any part of the argument with its overall structure. Readers with such a map seldom get confused or lost. Evasive vacuous sentences, however, are actually good ways to begin early drafts. They give you some leeway at a time when you don't want or need to be committed, and most important, they let you start. Write one down and you can go ahead without worrying that you have put your foot on a wrong path, because you haven't really taken a step yet. You just have to remember, when you have written the rest of One Right Way 54 what you have to say, to go back and replace these placeholders with real sentences that say what you mean. Suppose I take this advice and start somewhere else. If I don't begin at the beginning, where do I begin? What do I write first? Won't anything I write commit me as much as a first sentence? Doesn't every sentence somehow contain in itself, at least by implication, the whole argument? Sure. So what? Remember that any sentence can be changed, rewritten, thrown out or contradicted. That lets you write anything at all. No sentence commits, not because it doesn't prefigure your argument in just the way people fear, but because nothing bad will happen if it is wrong. You can write utter nonsense, things that turn out not to be what you think at all, and nothing will happen. Try it. Once you know that writing a sentence down won't hurt you, know it because you have tried it, you can do what I usually ask people to try: write whatever comes into your head, as fast as you can type, without reference to outlines, notes, data, books or any other aids. The abject is to find out what you would like to say, what all your earlier work on the topic or project has already led you to believe. (1 here "invented", as I mentioned earlier, the device known to teachers of composition as "freewriting," which is described fully in Elbow 1981, 13-19.) If you can bring yourself to do this [Pamela Richards discusses the reasons for not doing it in chapter 6], you will make some interesting discoveries. If you follow the directions and write whatever comes into your head, you will find that you do not have the bewildering variety of choices you feared. You can see, once you have your work on paper, that most of it consists of slight variations on a very few themes. You do know what you want to say and, once you have the different versions before you, you can easily see how trivial the differences are. Or if there are real differences (though One Eight Wsy 55 there seldom are), you now know what your choices are. (The same trick helps students who get hung up trying to frame a dissertation topic. I ask them to write down, in no more than one or two sentences, one hundred different thesis ideas. Few people get past twenty or twenty-five before they see that they only have two or three ideas, which are almost always variations on a common theme.) If you write this way, you usually find out, by the time you get to the end of your draft, what you have in mind. Your last paragraph reveals to you what the introduction ought to contain, and you can go back and put it in and then make the minor changes in other paragraphs your new-found focus requires. In short, by the time we come to write something, we have done a lot of thinking. We have an investment in everything we have already worked out that commits us to a point of view and a way of handling the problem. We probably couldn't, even if we wanted to, handle the problem any differently from the way we will end up handling it. We are committed, not by the choice of a word, but by the analysis we have already done. That's why it makes no difference how we begin. We chose our path and destination long before. Writing an unthought-out, unplanned draft (what Joy Charlton once inelegantly but accurately called a "spew" draft) demonstrates something else. You can't deal with the welter of thoughts that flash through your head when you sit at your keyboard trying to think where to begin. No one can. The fear of that chaos is one reason for the rituals the students in my seminar described. First one thing, then another, comes into your head. By the time you have thought the fourth thought, the first one is gone. For all you know, the fifth thought is the same as the first. In a short time, certainly, you have gone through your whole repertoire. How many thoughts can we have on one topic? One Right Way 5Q One Bight Way 57 Trying to evaluate, elaborate, and xelate all that we know on a given topic can easily overload the capacity of our working memory. Trying to com pose even a single sentence can have the same effect, as we try to juggle grammatical and syntactical alternatives plus all the possibilities of tone, nuance, and rhythm even a simple sentence offers. Composing, then, is a cognitive activity tha; constantly threatens to overload short-term mem ory. (Flower 1979, 36} That's why it is so important to write a draft rather than to keep on preparing and thinking about what you will write when you do start. (Joseph Williams suggests reserving the word draft for the first version that aims at coherence, to emphasize that freewriting produces a collection of working notes that shouldn't be mistaken for something more organized.) You need to give the thoughts a physical embodiment, to put them down on paper. A thought written down (and not immediately thrown into the wastebasket) is stubborn, doesn't change its shape, can be compared with the other thoughts that come after it. You can only learn how few thoughts you really have if you write them all down, sel them side by side and compare them. That's one reason why dictating an early draft onto tape, even if you do the transcription yourself, is useful. You can't throw away a page of a tape very easily; you can still erase a foolish thought, but it is a lot of trouble, and most people find it easier just to keep talking and make changes on a typed version. Making the words physically real, then, does not commit you to dangerous positions. Just the opposite. It makes sorting out your thoughts easier. It makes writing the first sentences easier by letting you see what you want to say. Using the language of cognitive psychology, Flower and Hayes 1979, describe a similar process of working back from written materials to a plan and then forward to another piece of writing. The paper deals with a much smaller project—writing a short theme over the course of a few minutes, rather than a scholarly paper or book over a period of months or years—but the discussion of how writers create elaborate networks of goals and sub-goals and change their high-level goals in the light of what they have learned by writing is relevant to our discussion. A problem as insoluble as how to begin—another version of it, in fact—is how to organize what you have to say. Students often complain that they can't decide how to organize their material, whether to say this or that first, whether to use this idea as an organizing principle or that one. The theory of One Right Way to do things causes mischief here too. Another example from my thesis will provide material for the analysis. I had simple results to report. Schoolteachers evaluated a number of aspects of their job: their relationships with the students they taught, the students' parents, the principal they worked for, and the other teachers they worked with. They liked those people in each category who made their work easier, disliked those who made it harder. In their view, schools varied most importantly in the social class of their students. They found slum children difficult to teach; they found upper-middle class students difficult too, smart but not respectful enough of the teacher's age and authority. Most teachers preferred working-class children, who could do ordinary schoolwork but were docile and thus easy to handle. They also preferred working-class parents, who were most helpful in controlling their own children. Residential segregation made distinguishing schools by students' social class easy to do. Most schools were predominantly one or another class. That analysis gave me a simple choice of ways of organizing my material (which came from sixty interviews with teachers.) I could analyze, in turn, the relations teachers had with students, parents, principals, and other teachers, describing under each heading how those relations varied depending on the social One Right Way 38 OneBigntWay 59 class of the school. Or I could write in turn about slum schools, working-class schools, and upper-middle-class schools, explaining the particular constellation of teachers' relations with those four groups that characterized schools of each class. How did I choose? I couldn't see that it made anv difference, at least with respect to the bulk of the writing I had to do. Whichever way I chose, I would have to describe teachers and working-class kids, teachers and slum school colleagues, teachers and the principals of middle-class schools, and all the other combinations of relations and school types created by cross-classifying relation and class. My smallest descriptive units, analyzing those combinations, would be the same. The opening and closing sentences, relating the smaller units to the whole, would be different, as would the final arguments I made. But I would be able to use whatever I wrote, however I finally put the material together. Either way, I would report the same results (although in a different order) and arrive at essentially the same conclusions (though the terms they were put in and their emphases would differ). What I said about the implications for social science theory and social policy would differ, naturally. If I used my results to answer different questions, the answers would look different. But none of that would affect the work that lay immediately ahead of me as I began writing my thesis. Why worry about it? I worried about it—everyone worries about it—because the problem, while very important, can't be solved rationally. Whichever way I chose, I found myself wanting to talk about, or talking about, something I hadn't mentioned or explained yet. I could start by talking about slum schools, but only if I talked about the four groups and teachers' relations with them. But I couldn't talk about those relations without explaining the theoretical issues involved. I would have to explain, for instance, that service workers, like teachers, typically judge people they work with on the basis of hnw easy or difficult those people make it to get through a day's work. If I did that, I would be starting vrith the relations. But I couldn't say anything sensible about the relations without first explaining social class and its bearing on children's ability to learn school materials and to behave in ways acceptable to teachers, and on parents' willingness and ability to help teachers keep children in line. You can see where that leads. It once led my colleague Blanche Geer to wish for a Way to write what she had to say on the surface of a sphere, so that nothing would have to come first. That would shift the problem of what to take up first to the reader. The image of writing on a sphere exactly captures the insoluble nature of the problem, as people usually define it. You can't talk about everything at once, no matter how much you want to, no matter how much it seems to be the only way. You can, of course, solve the problem. Everyone eventually does. You do it by taking up, for instance, the relations between the teachers and other groups and saying that there is also this other way of looking at it, and in due time you will explain that too. This is not so much a placeholder as an I.O.U. Writers find the question of which-way-to-organize-it a problem, again, because they imagine that one of the ways is Right. They don't let themselves see that each of the several ways they can think of has something to recommend it, that none are perfect. Believers in Platonic perfection don't like pragmatic compromises and accept them only when reality—the need to finish a paper or thesis, for instance—compels it. But writers have more immediate reasons to worry than not knowing the One Right Way. They don't even know, at the beginning, what those smallest units are, the fragments out of which the final result will be made. Another is that they don't have much idea about the alternate ways they might be put together. They don't, for instance, know that they can choose between organizing their discussion around kinds of schools or One Right Way 80 kinds of work relations. They have vague notions that one thing might lead to another, that one idea might stand in a causal relation of some sort to another, that one idea is a specific version of another more general one. But they might be wrong. Those ideas might contradict something they read in Dürkheim or Weber, conflict with the results of someone else's research, or even be belied by their own data. People hope to solve these problems by making outlines. Outlines can help, but not if you begin with them. If you begin, instead, by writing down everything, by spewing out your ideas as fast as you can type, you will discover the answer to the first question: the fragments you have to work with are the various things you have just written. These fragments will be at every level of generality or should be. Some will be specific observations: teachers hate kids who talk dirty. Some will be more general: teachers can't stand anyone challenging their authority in the classroom. Some will relate to the scholarly literature: Max Weber says that bureaucracy is a rule of secret sessions. Some will be about social organization: slum schools have unstable teacher populations, while upper-middle class schools, because teachers seldom leave them, have more stable teaching staffs. Some will be about careers and individual experience: teachers who, for whatever reason, have spent several years in a slum school, no longer want to leave it. Once you have the fragments, you can see how disparate they are, how they range from the general to the particular and don't seem to stick to any Dne way of thinking about your topic. Now you have to arrange them so that they at least seem to move logically from point to point in what a reader would recognize as a reasonable argument. How can you do that? People solve this problem in a variety of ways. I use this principle to choose among possible solutions: Do whatever is easiest first. Write the part that is easiest to write, do simple housekeeping chores like sorting your One Right Way 61 papers out. (A contradictory approach regards any task that is easy suspect and tries, rather, to start with what is hardest. I don't recommend that kind of Puritanism.) Here is one easy way to discover how to organize your materials. Its greatest virtue (and this is a corollary of the principle of doing easy things first) is that it transforms a difficult mental task into a largely physical, and therefore easier, one. Begin by taking notes on what you have written, putting each idea on a file card. Don't discard any of the ideas in your draft. They may come in handy, even if you can't see how at the moment; your subconscious knows things you don't. Now sort your stack of cards into piles. Put the ones that seem to go together in one pile. "Seem to go together?" Yes, and don't look too closely, for the moment, for what they have in common. Follow your intuition. When you have assembled these piles, make a card to go on top of each one, a card that summarizes what all the cards in the pile say, generalizing their particulars. For the first time you can begin to be critical of what you have done. If you can't think of a statement that covers all the cards in the pile, take out the ones that don't fit and make new piles for them, with their own summary cards. Now lay your generalization cards out on a table or on the floor, or pin them up on the wall (I got the pinning-on-the-wall habit from working with photographs, which photographers ordinarily inspect by leaving them pinned up for a week or two). Lay them out in some order, any order. Maybe you can make a linear order in which one idea leads to another. Maybe you can lay some of them out in a column, one under the other, which would physically indicate a relation of specific example or sub-argument to more general statement. You will soon see that there is more than one way, but not very many more, to make your case. The ways are not identical, because they emphasize different parts of your analysis. If I organize my analysis of schoolteachers around kinds of schools, I will empha- One Right Way 6S size the local social organization of the school and to some degree lose the comparative emphasis on professional problems that an analysis focused on the relations would emphasize. This way of experimenting with the organization of ideas has been somewhat formalized in the idea of the flow chart. Walter Buckley provided a good example in his formalization of Thomas Scheff's theory of mental illness. The chart, reproduced here as figure 1, comes from Buckley {1966}. You needn't know the theory involved to see how this device clarifies an argument. Doing all these things, by the way, helps solve another common "minor" problem. Social scientists reporting empirical research always include a descriptive section, telling something about the country, town, or organization they did their research in. What should such sections include? Researchers vaguely intend them to give readers "a feel for the place," and fill them from a commonly accepted list of things every reader would presumably need to know, a melange of geography, demography, history, and organization charts. Writing enough to know what your argument is helps you make the choice more rationally. The facts about places, people, and organization do more than give readers a general familiarity. Social organizations work the way the research report says they do only with the right kinds of people and in the right kinds of places. So preliminary descriptive materials set down some of the basic premises on which the report's argument rests. If our book (Becker, Geer and Hughes 1968, 15 ff.) describes a student culture which profoundly affects student lives and perspectives, the reader needs to know that the college we are talking about is, for instance, large and that it is, in fact, the dominant institution in a small midwestern town, and that a large number of the students come from smaller, less cosmopolitan places. I find one further way of dealing with organizational problems interesting. Instead of trying to solve the 6 B Si o .(0 •8 to I o 0 fa H DO fa « AND 1 IRMEt fiat One Bight We^r 64 One Highway 65 insoluble, you can talk about it. You can explain to readers why whatever it is is a problem, what ways of solving it you have thought about, why you chose the less-than-perfect solution you actually chose, and what it all means. The what-it-all-means will be interesting because you wouldn't be having the problem if it didn't embody some interesting dilemma in the work you are doing—for instance, the way problems of class and professional structure intersect in concrete organizations so that you can't talk about class without talking about teachers' shared perspectives on their professional relations, and can't talk about those without talking about class. You have trouble only if you insist that, in principle, they have to be discussable separately. Talking about them, instead of trying to wish them away, solves all sorts of scientific problems, not just those of writing. When anthropologists and sociologists do field research, for instance, they typically have problems establishing and maintaining those relationships with people that will let them observe what they want to over a long period of time. Delays and obstructions while you negotiate these arrangements can be discouraging. But experienced fieldworkers know that the difficulties provide valuable clues to the social organization they want to understand. How people respond to a stranger who wants to study them tells something about how they live and are organized. If the poor people in a city neighborhood you want to study are suspicious and won't talk to you, that is a real problem. You may eventually discover that they are standoffish because they think you might be an investigator trying to catch violators of the welfare regulations. The trouble, personally painful, will have taught you something worth knowing. Similarly, experimental social psychologists got upset when Rosenthal and others demonstrated that an experimenter's seemingly extraneous and irrelevant actions affected the results of experiments indepen- dently of the variables supposed to be at work. They shouldn't have. As Rosenthal showed (1966), while psychologists thus lost the illusion of total control over experimental situations, they gained a new and interesting area of study: social influence in small groups. That is gained by talking about the insoluble problem rather than ignoring it. It's the same with writing. When you can't find the One Right Way to say it, talk about why you can't. Bennett Berger adopted this solution in The Survival of a Counterculture (1981), which reported his study of hippie communes in northern California. He was interested in Utopian experiments. He felt personally close to the hippie culture and ethos. He wanted to study how communards dealt with the inevitable gap between what they professed and how they behaved as they adapted their beliefs to the circumstances of their lives. He called the methods people used to deal with gaps like that "ideological work" and conceived of studying such work as a microsociology of knowledge. But he had trouble writing about what he found: I delayed writing this book for several years because I couldn't find an interpretive frame to put around the social life I observed. Without that frame, I wasn't sure that I understood the meaning of what I was seeing. Without that understanding, I had no posture toward the data, and that reduced my motivation to write. And when that understanding emerged, I didn't like the "cynical" posture it invited me to take. He described the problem of the cynical posture, which bothered him deeply, as it affected the study he had done in the commune: [It is] the tendency of the sociology of knowledge to impugn, weaken, or undermine ideas when analysis of them reveals their self- and group-serving functions. ... If the idea of urban One Bight Way 68 apocalypse serves the interests of survival-equipped communards, is that sufficient reason for casting a cold and skeptical eye an it? If the idea of equal rights for children serves the purpose of those adults who initially had neither the time nor the inclination to be middle-class parents, is that sufficient reason for being cynical about their motives? If the affirmation of "authenticity" in interpersonal relationships serves the interests of people so situated that their dense interactional textures make them ill-able to afford emotional disguises, isn't that reason to regard [their belief in] "openness 'n honesty" as simply another self-serving element of ideology (like belief in cultural pluralism by ethnic minorities or in low taxes by the wealthy]? Or on the other hand, when groups are caught in contradictions between the ideas they profess to believe in and their day-to-day behavior, is their hurried ideological repair work best understood in an ironic, contemptuous, and cynical manner? My answer to these questions is no, at least insofar as the [people he studied] dealt with them. But the answers provided by the major tradition of the sociology of knowledge would seem to be a resounding YES—in part because one of the major motives informing the sociology of knowledge as an intellectual enterprise has been the desire to "unmask" or "demystify" ideas by revealing the "real" interests or functions they serve. (168-69) "It's easy to see how such a problem can paralyze you: It has taken me a long time to gain the perspective on beliefs and circumstances adopted in this book, and my failure to apprehend it earlier has functioned as a kind of bit in my mouth, preventing me from speaking clearly (223)." Berger wanted to discuss the social bases of what communards believed without making fun of them. Until he could figure out how to do that, he couldn't write his book. I don't want to pursue his argument further (although it deserves to be read in One Bight Way 67 full) because I am citing it as a solution to another kind of problem. Not Berger's problem of how to avoid making fun of what he was studying, but the even more common difficulty of not being able to write because you haven't found the One Right Way to handle that or some other problem. Berger doesn't say how to avoid that fruitless search for the One Right Way, but he demonstrates how. Write about it. Make it the focus of your analysis. He devoted a sizeable part of his book to just that task. In so doing, he found a way to write his book as well as a large subject to embed the story of his research in: the intellectual vice of explanation as a putdown. Taking readers into your confidence about your troubles requires admitting that you had them and, therefore, that you are not the paragon who always knows the Right Way and executes it flawlessly. I don't think that difficult, since no such paragons exist, but some people don't like to make such admissions. The remedy is to try it and prove to yourself that it doesn't hurt.