Index

The difficult part of talking about prose style organization, or discourse of any kind, is asking the question: Why is this clear? We might say, 'Well, it's clear because it's straightforward, it's concrete, it's not abstract, it's relatively concise.' But the problem is when you ask yourself what do the words "clear," "concise," "direct" and so on refer to. They don't refer to anything that's on the page; what they refer to is what's going on in your head. Because if you say "It's clear," you're saying you can understand. Almost all the words that we use that seem to describe style in fact describe how we feel about style. So the problem that we've got is that we're talking either about somebody else's style or more particularly our own style. We've got a real problem because if we use those kinds of words, we're only dealing with impressions. And the problem with impressions is that they're even more deceitful in some ways.

Researchers at Carnegie Mellon University created a passage of prose on some technical subject, with abstraction, complexity and so on. They gave the passage to two groups of readers and asked each simply to pick where in the passage they had a hard time understanding. The difference between the two groups was that one group had received some background reading on the subject, so they knew something about the passage before they read it, and the other group was just reading cold. Which group do you think was better able to pick out where these deliberately inserted problems were? The people that did not have the background knowledge were better able to locate the rhetorical problems. That's because most reading is just reminding yourself of what you already know. Most reading is just remembering knowledge.

When we look at our own prose, who knows more about what we've written than we do? When we read our own stuff, all we're doing is reminding ourselves of what we wanted it to mean when we wrote it. That means that we are our own worst editors. We are constitutionally incapacitated from looking at our own writing the way others will read it. And that is just a flat fact. What that means is that when you're looking at your own writing, everything is going to look just fine. It's all going to be coherent. It's all going to hang together because everything has a cohesive thought. You know you've been working on it for so long you almost could recite it if you had to.

So the problem that any writer has is to find a way to look at what is on the page, what he or she has actually put down on the page, that would tell objectively how a reader is going to react.

That's the problem both when looking at your own writing and also for talking about other people's writing. It's not enough just to say to a young reporter, "Well, this is not very clear, this is not very focused, this doesn't hang together. Go away and do something about it." They go away and stare at their computer for a while, then come back, and it's not much different. You have to be able to point to certain things that will cause a reader problem, and those features have to be easily, mechanically identified. Life is too short to count words and do long linguistic analysis of everything you write. It's got to be quick and dirty.

So you can talk about writing in two ways. You can do very fine-grained analysis in milliseconds that is absolutely crucial for the work that Morton Gernsbacher is doing on cognitive processes. And then there is the kind of research that is sort of cruder and much more clunky in lots of ways, the kind of research that asks what is it that ordinary writers will be able to understand and use and act on that is not so complicated that they have to be linguists to be able to apply these principles. That's a complicated problem of research, but it can be done quickly and reliably. You're asking writers to learn about treacheries of writing, to make distinctions and to try and make these revisions. And after a lot of trial and error, we can give people ways of making distinctions. First, recognizing the distinctions, then making predictions about how readers are going to take their prose and then, third, making the revisions. Then it's kind of trial-and-error research.

I referred to a 10% brainpower test that working writers had to be able to apply to their work to make quick and dirty decisions about where they ought to revise and where they ought not to bother revising it. The 10% brainpower test is simply to look at subjects of sentences, at the first four, five, six or seven words of every sentence. Ignore short little introductory phrases like "in the beginning" and "for the most part," transitional phrases and so on. To the degree that you see the names of characters as subjects immediately followed by a verb with some kind of specific action, your readers are going to judge your style to be relatively clear, relatively direct, more direct certainly than if you see sentence after sentence after sentence beginning not with the names of characters but rather with long strings of abstractions.

Now that can be generalized into a higher and more general principle. I suspect I could ground this principle in a lot of the psychological literature, though I have not yet tried to do it. It would go under the term of initial wellboundedness, a fancy term that means only that, at the beginning of a sentence, a reader can see a relatively wellbounded subject. Which is to say, at the syntactic level, something short, something concise; at the conceptual level, something relatively concrete; at the larger semantic level, something whose meaning is also relatively compacted. Then to that degree, a reader is going to be judging one's prose as relatively clear, relatively direct, relatively straightforward. So the key to a clear prose style really resides in the first six or seven words of every sentence.

You can make a character out of anything you can make the subject of a series of sentences. The concept of characters as necessarily flesh and blood is much too simple-minded. You've got to expand the notion of character to include anything that you can make a subject and a verb and that you can repeat as the subjects of several verbs. The overarching principle is the sense of well-boundedness, the sense of beginning a sentence or a series of sentences with elements like a grammatical subject that is relatively clear and quickly graspable, semantic concepts that are graspable. It means that the units at the beginnings of sentences are short, well defined, well bounded. This principle in fact applies to more than sentences; it applies to units of discourse of all lengths. What readers count as clear, direct, straightforward, readable writing can be defined by this notion of relative distinct beginnings in short well-bounded concepts that introduce longer, more complicated subjects.

Please read the following four paragraphs:

1. Writing is revising, because few are expert enough to draft perfectly. To shorten, erase, but if you mis-arranged parts, cut and re-order; you can restore them. Even great writers revise, so surgery signals no weakness.

2. Writing includes revising, for no writer is so expert that he always writes a perfect first draft. If your draft just needs shortening, use a pencil, but if you have mis-arranged its parts, save yourself labor by cutting it up and re-ordering it. Slash it to ribbons if necessary; you can always restore it tomorrow. If your manuscript needs major surgery, that does not signal weakness or defeat. The best writers have to revise.

3. Revising is part of writing. Few writers are so expert that they can produce what they are after on the first try. Quite often the writer will discover, on examining the completed work, that there are serious flaws in the arrangement of the material, calling for transpositions. When this is the case, he can save himself much labor and time by using scissors on his manuscript, cutting it to pieces and fitting the pieces together. If the work merely needs shortening, a pencil is the most useful tool; but if it needs rearranging, or stirring up, scissors should be brought into play. Do not be afraid to seize whatever you have written and cut it to ribbons; it can always be restored to its original condition in the morning, if that course seems best. Remember, it is no sign of weakness or defeat that your manuscript ends up in need of major surgery. This is a common occurrence in all writing, and among the best writers.

4. Making large and small changes in a piece of prose is included in the process of producing a written document. Not very many of those who write are so highly skilled from long experience that they can produce a draft the first time that has no flaws in it at all. On many occasions, the writer of a document will find out that when he looks at the completed work carefully, that there are in the text serious errors in the way in which elements of the material have been ordered, calling for a rearrangement of them. When this situation turns out to be the case, the writer can save himself hard work and much time if he uses scissors to cut up his manuscript, cutting it into several pieces and putting the pieces back together in an order . . .

Here's the point that I'm trying to get at. One should defend a piece of prose based on what is on the page. Every word can be defended. If you had picked No. 2, then No. 3 would now seem to be kind of redundant. But that would have contradicted your first judgment if you had heard that it was E.B. White.

If you had "Elements of Style" in front of you, would you have been willing to make that judgment about E.B. White's prose? Knowing that he is the icon of advice about English prose style, I would bet that most of us would have just gone with our judgments of "Oh, well, you can't improve E.B. White's stuff." E.B. White's style is after all E.B. White's style. And we would've simply taken it for granted that that was the best he could have done.

The bigger point that I'm making here is that, that's the defense a lot of the public intellectuals will make of their writing style: that this is the only way. Don't believe it. Don't believe that that's the only way it can be written. The only way that you can know whether or not it can be written any better is to try.

Title page

Introduction: what editors discovered by meeting with language experts

Structure, goals and comprehensibility revisited

Hierarchy of abstraction key concept in powerful writing

If language is instinctual, how should we write and teach?

First mention drives how people read and comprehend language

The ivory tower of Babel and the myth of a general literacy