Writing is putting thoughts and arguments that are usually complex and multidimensional into a single-dimensional flow. Thoughts about anything of intellectual interest usually go in several directions at once; but writing always has to string one word and one sentence after another. All writing problems boil down to this: choosing which words in which order. You have to make it flow through a single sequence in time.


Structuring is figuring out the sequence in which ideas will emerge. There are two main levels of structure: the micro-structure of word by word within sentences; and the overall macro-structure of the entire argument. The larger structure is determining over what is within it, as structure tends to be set from the outside in. Knowing what the larger structure of your argument is will help you with micro-structuring sentences, but the reverse is not usually true.


The easiest way to work out the structure is to make an outline. This is often the intellectually hardest part of writing, but there are routines you can use to force your way ahead, even when it is very difficult.

Collect your notes in one place. Wherever you jot your ideas—on margins in books, scraps of paper, summaries of things you've read—try to boil these down into major points and get them on one piece of paper. (Or as many pages as necessary; but then you need to re-summarize them, as many times as you need until you can see the whole thing at once.) Then go through it, for instance with a colored pen, and try various sequences (A,B,C; 1,2,3, 3a, 3b, 3c, whatever) until you get one that conveys your argument in a reasonable sequence. You will find yourself collecting similar points together, but also wrestling with contradictions and gaps. This may seem mechanical, but it also is a time that can produce creative ideas.

If you can't find an ideal sequence, or even decide which point to start with, be arbitrary: pick a starting point and go from there. If you can't decide which point comes next, be arbitrary: pick one and string the others after it. All that means is that no one point obviously leads into or dominates the others, and they all have to be treated on the same level. Don't waste time over such quandaries; a crucial factor is getting your argument flowing, and you can't do that by being stalled at choice-points.


When you are actually writing, you will sometimes get stuck. You can't figure out what to say next; the words won't come; you can't decide on the next sentence or the right term. If you are seriously stuck, it is because of a problem in the macro-structure of your argument, not in the surface structure of the sentence itself. You don't know where your argument is going. If I go back and look at the outline—sometimes I'll make a new outline based on what I've written so far—I always find that I got stuck at the point where my outline no longer told me where to go next. Do your revising on the outline. It saves a lot of trouble in rewriting; it's a lot easier to tear up unsatisfactory outlines and start again than to have to do it with unsatisfactory drafts of your paper or book.


If you really know what you're going to say, a written outline may be unnecessary. If you feel that potential flow of sentences ready to pour out, go ahead and write it. But you'll have the outline in your head, and essentially will be saying to yourself: first I'll say this, then that, then.... Of course, the process of writing itself can be creative; you can get new ideas on the fly. But they will be effective ideas only if a structure of argument emerges as you go along. When it happens, consider yourself lucky. When it doesn't (or when it stops), go back to outlining.


An outline may go through several phases: a messy collection of notes and ideas; various ways of getting them in order; and a final outline page reduced to a set of headings in sequence. Some of this will get explicitly transferred into the paper or book itself. But good writing is that which is not cluttered by a lot of pedantic-looking

"I. IA. IA.1 Ia.2 etc." Some of this is useful to guide the reader through (see TRAFFIC below). But it is far better for the structure to be implicit in the writing, than overlaid with these markers. (If you want to see an example of a clumsy use of such markers, look at Oliver Williamson's Markets and Hierarchies: an important and intelligent book, but very unpleasant to read.) These order-markers were useful in getting your argument in sequence, but now the sequence will generally carry itself. Get rid of most of the numbers, letters, and hierarchies of sub-headings. To the extent that such markers are still useful, disguise them as vivid and apt titles for sections of your argument. Retain numbers and letters only in places where you genuinely have to list a series of points that are arbitrarily collected, that have no intrinsic order.


Headings and sub-headings are a good place to make sure the reader gets your point. Don’t waste them on conventional labels like “Introduction” “Summary” “Discussion.” Write like Nietzsche, not like a bureaucrat.


Sections of your argument stand out better when surrounded by some empty space. This is a visual trick to influence the reader's mind. More than that, writing generally improves by adding space within it. Break up your paragraphs when they get too long. Usually this is not hard to do, even if you have to do it arbitrarily. You may think that the thread of a complex argument is being kept together by having it all in the same paragraph, but the effect on the reader tends to be to bury it.

The same thing holds for sentences. When they get too long and complicated, it is almost always better to break it into several sentences. This is not hard to do, even if you have to repeat a subject noun to do it. Bertrand Russell (who was a wonderfully lucid writer) gave this as his one piece of advice on writing: whenever you have to convey something complicated in a sentence, put at least part of it in a separate sentence.

A negative example is Pierre Bourdieu. He has even explicitly defended himself (in the Preface to Distinction) for his inordinately long sentences and paragraphs; he claimed that the complexity and subtlety of his ideas, and all the qualifications they involved, required this form of writing. Don't believe it. Russell, Wittgenstein, and other philosophers have dealt with matters of equal subtlety in lucid (and well-spaced) prose;* and one could certainly rewrite Bourdieu to good intellectual as well as stylistic advantage. Habermas is just as complex as Bourdieu, and better organized; if his writing seems heavy, it is for other reasons, more on the level of the way he expresses his concepts. Erving Goffman was not only well-organized, but also had a light and elegant touch.

* Russell was obviously very good at structure. When you have that skill, you don't have to rewrite much. Russell once quipped: "I have only rewritten once in my life, and the result was so much worse than the first time that I resolved never to do it again."


Writing sentences is not so difficult if you follow the above advice: get the overall structure so you know where you're going in each part of the argument; break up long involved sentences (which will also give you an easier syntax). Inside particular sentences, these points help:

> Use the active voice more than the passive. This is old advice, but still good. But no need to be rigid about it. Do whatever sounds right. Whatever is easiest to write, usually turns out to be easiest to read. (Sartre had a terrible time writing his Critique of Dialectical Reason, and it shows.)

> Try to keep the parts of a verb phrase together, where possible. Get the main action of the sentence into the reader's attention early on, and move the qualifications to the end. This isn't necessarily the way it will first come out in your head, or on paper. Don't worry about it; just get the sentence out and then engage in "word-processing", moving things around to where they fit best. With more experience, rearranging sentences happens faster and faster, and eventually will occur almost before you have it written down.

> The most important thing in getting writing done is the flow. If you notice your sentences need to be broken up, reorganized, etc., but it seems tedious and a side-track to do it now, then don't. You can always do it later, as long as you know what it is you have to do. The hardest part of writing is getting that first draft on paper. Once it's there, you can always fix it.

> If you can't decide which of two words, or two expressions, you want to use, don't get bogged down over the decision. There is no such thing as "the perfect word". When I'm in this situation, I just write both words (both expressions) down, one above (or alongside) the other, and later come back and cross one out. If they're both about the same, then it doesn't make any difference which to choose, so just be arbitrary. Again, with writing experience, the choices happen faster and more easily. Do anything to keep up the flow.


My father-in-law, who was a newspaper editor and columnist, gave this as his one piece of writing advice, and it has always worked. When revising, or just plodding along deciding how to say things, it almost never hurts to cut. If you can't decide whether or not to cut a word, phrase, or paragraph, cut it.


There are two kinds of sentences: substantive sentences, and traffic sentences. Because complex arguments do not necessarily flow in a single sequence of ideas, it is sometimes necessary to stop and explain the order in which you are giving them: in other words, directing verbal traffic. One of the major differences between good and bad writing is that the former uses traffic sentences forthrightly, while the latter avoids them. If there is a problem with the complexity of your exposition, be up-front about it. Let the reader in on the problem: "This topic is complicated because... To unravel it, we have to pull apart these features... I'll take them in the following order..." (An example is Habermas, Theory of Communicative Action. Although Habermas is not a stirring writer, he is clear on the structural level, and he uses traffic sentences well. Student papers, on the other hand, often get balled up for lack of this.)

Summaries are a kind of traffic sentence, but coming at the end of the argument and looking back to where it has gone. This can be useful, but you have to use your judgment as to when the summary is really useful to keep things straight (and especially when it leads into the following part of the argument). Summaries which are too mechanical, or which break up the flow of the argument, really fall under the category of "scaffolding" which should have been taken down.

Questions (including rhetorical questions) can be a graceful way of setting up the flow of what is to come next, or acting as local traffic sentences. Immanuel Wallerstein, who is quite a good writer, organizes a lot of dense material this way. ("The other element involved in banditry was part of the nobility, but again which ones?..." The Modern World System, volume 1, p. 142. "What does our argument add up to so far?..." etc.) Questions tend to give a nice flow to the sentences, and to lighten up heavy indicative exposition.


The use of special technical vocabulary is a matter of taste. Often things can be said more directly without it. Lord Kelvin, the physicist, said that if a theory really has something to say, it should be possible to explain it in words your bartender can understand. That may be exaggerated, but it is usually true for sociology. Some writers, like Bertrand Russell, got a lot of malicious pleasure out of deflating jargon, by defining its meaning in simple terms. (C. Wright Mills once did this in a famous passage on Talcott Parsons.)

But if you write without jargon, bear in mind you are taking a risk. Technical language shows off one's membership in a particular linguistic community, and people who are committed to a professional specialty tend to automatically put down people who don't use their jargon. Whether or not to use jargon is more of a social decision than a stylistic one; it represents different strategies toward the intellectual field. But let’s at least be honest with ourselves about what we are doing. Contemporary neo-Marxists (which includes most of the post-modernist/ post-structuralist/ post-colonialist/ liberationist movements) tend to be bad offenders here. They are among the most jargon-ridden of today's intellectuals, which reflects the fact that they write esoterically for an elitist group of intellectuals. This is a remarkable example of self-deception for movements which regard themselves as anti-elitist and liberating. Marx himself was one hell of a lot better writer (probably because he hung around with Heinrich Heine, the liveliest of all German poets—and because he genuinely wanted people to get the point).


A lot of otherwise competent writers in sociology are flat, because they give us a steady diet of abstractions: heavy nouns and verbs which are really nouns with verb endings. If you have any good metaphors, and any good colloquial turns of speech, this is where they are most needed. But it has to come naturally; artificial metaphors (or old clichés) have the opposite effect from what they are intended to do. If you can't write vividly, too bad; don't try to force it.


One tendency of mediocre writers is to try to be extremely impersonal, never using the word "I". Good style, on the contrary, is quite willing to say "I will come to this later...." Or the word "we:" "So far we have found...." First person pronouns are pretty much necessary in traffic sentences. It is foolish and clumsy to try to avoid personal pronouns when they are the most direct way of making your point. Over-formality is a mark of the semi-literate. (Unfortunately, we find a lot of this among copy-editors and journal reviewers. Dealing with these kind of people is an occupational hazard.)

Exception: starting sentences with “I think…” or “I believe that…” is usually just extra verbiage that will annoy the reader. Go ahead and say what you have to say. If you need to write that way to get the flow of words on paper, okay, but come back at the end and cut out the unnecessary words.

A related problem, common in abstract social science, is to write so as to avoid any active agent in one's sentences at all. For example, George Herbert Mead, Philosophy of the Act, p. lxiv: "The undertaking is to work back from the accepted organization of human perspectives in society to the organization of perspectives in the physical world out of which society arose." Mead was a poor stylist for this and other reasons. Max Weber, by comparison, is structurally a much better writer, even when he is being very abstract.


The difference between mediocre writing and good writing is often just taking the time when you are finished to go back over what you have written, and making corrections. This is the difference between a memorable article or book, and a turgid one; or between an A+ paper and a B+ paper.

Personally, I enjoy re-reading what I have written (unless I’ve done a terrible job of it, which means a lot of work hasn’t been done yet). Even in good writing, there are a lot of things to clean up: typos (leaving them in shows you don’t care what your writing looks like); places where I can say it more sharply with fewer words, or sometimes where something has to be better explained; good ideas I’d like to add, or sometimes where too many things are being said and it’s better to save some of them for another piece. Get rid of anything that sounds like a cliché, unless you’re being sarcastic.

A harder task is places that repeat what I’ve said somewhere else in the same paper. The question then is: do I need to make this point in different contexts? If so, OK. If not, it’s annoying to the reader to read the repetitions. So I flag them all and make a list of where I said this; then figure out where is the best place to introduce it, and cut the others. Some very good writers I know are sloppy in this respect; but this kind of sloppiness can make it hard to get your stuff published. – So why do I enjoy this? As your text gets sharper, it acquires more rhythm. It feels right.


"Writer's block" is a common complaint: you just can't get yourself to start writing. It's basically a matter of getting into the rhythm. If you write every day, it's easier the next day. A lot of writers start off by going over what they wrote the previous day, or their outline, or reading something you want to argue with. But still, you may feel you are starting cold. Just get it going, no matter how. The first words on paper aren't important; you can always come back and cross them out later. Good flow by the writer is the key to good reception by the reader. So keep plugging until you get into a good rhythm, and then throw out the stuff that isn't good.

When I can’t get myself going at the keyboard, I take a pad of paper and scribble out the easiest parts of the argument I can think of, as fast as I can. This is a trick to tell myself, it doesn’t really matter, this is just a preliminary draft. After a while it starts to flow (assuming I don’t have a macro-structure problem where I don’t know what I want to say). Then you’re home free—sort of. After it’s on paper I tell myself, now the rest is just typing it up. That’s not really true, but whenever you get momentum, better expressions and new ideas come easily. Sometimes this will get me into a prolonged writing binge: I feel like I ought to take a break, get a drink of water or something—but, when you’re riding that horse you want to go as far as you can with it.


"Lucidity, force, and ease:" Edmund Wilson, a wonderfully competent writer and critic, singled these out as the great virtues of classic prose. Lucidity is what all the advice about structure is supposed to produce; if you can get the hang of it, and develop a sense of verbal rhythm, the ease will be there too. Where does force come from? The energy of one's writing comes mainly from having something to say. The argument drives itself along, because it is going somewhere.

How do you have something forceful to say? Mainly, by being involved in the intellectual discourse of your field, knowing where the arguments are, and trying to move it forward. Set some high standards for yourself, so that you know in what direction to move.


The way you acquire intellectual and stylistic standards is by being exposed to the best in the people you read. One reason sociologists are often bad writers is because they read so many other sociologists (or philosophers, or economists or statisticians) who are bad writers. For style, don't confine yourself to reading social science. Personally, I think the secret is to read poets—Yeats, Dylan Thomas, whoever you like—in order to pick up your own sense of rhythm. Of course you can't always just read good stylists; often you need to read for content. But whoever you are reading, ask yourself if they are writing well or not. Either way, notice how they are doing it.

A book that breathes the sheer energy of writing is D.H. Lawrence, Studies in Classic American Literature. Lawrence was always an impassioned writer and this book, though really a work of literary criticism, is even more impassioned about writing than he was about sex. Which is saying a lot. You can get a contact high just from reading Lawrence—his energy is contagious.

That’s what you want to aim for: make your energy contagious.