How to Write a Sentence

by Stanley Fish

  CHAPTER 3

  It’s Not The Thought That Counts

  Notice that it doesn’t matter which three-word sentence you use as a base. It doesn’t matter what the sentences you practice with say; it doesn’t matter what their content is. In fact, the less interesting the sentences are in their own right the more useful they are as vehicles of instruction, because, as you work with them, you will not be tempted to focus on their content and you will be able to pay attention to the structural relationships that make content—any content—possible. The conventional wisdom is that content comes first—“you have write about something” is the usual commonplace—but if what you want to do is learn how to compose sentences, content must take a backseat to a mastery of the forms without which you can’t say anything in the first place.

  To be sure, your eventual goal is to be able to write forcefully about issues that matter to you, but if you begin with those issues uppermost in your mind, you will never get to the point where you can do verbal justice to them. It may sound paradoxical, but verbal fluency is the product of hours spent writing about nothing, just as musical fluency is the product of hours spent repeating scales. For the purposes of becoming a facile (in the positive sense) writer of sentences, the sentences you practice with should have as little meaning as possible. Indeed, nonsense sentences—sentences that display a logical arrangement of components, but are without a readily discernible message—may be the best materials. The linguist Noam Chomsky famously offered the sequence “colorless green ideas sleep furiously” as an example of a verbal unit that is perfectly grammatical but semantically nonsensical. It is grammatical because the doer (ideas), and the doing (sleep), along with the manner of the doing (furiously), are in the appropriate structural slots. It is nonsensical (or so it is said; were it a line in a Wallace Stevens poem, sense would be attributed to it in a heartbeat) because ideas do not sleep, and sleep is a quiescent activity rather than an activity that can be performed furiously. It is a well-formed structure without meaning.

  Chomsky contrasts “colorless green ideas sleep furiously” with “furiously sleep ideas green colorless,” which, because it exhibits no logical relationships whatsoever, is a list of the kind I presented in my first example. You can work with “colorless green ideas sleep furiously” by treating it as a formal structure with “slots” that can be filled with alternative words (e.g., “mystical white filaments exfoliate silently”), and you can learn something about what makes a sentence a sentence by maintaining that structure in the face of serial nonsense. You can’t do anything with “furiously sleep ideas green colorless,” not because it is without meaning, but because it is without form. It is true that you can’t get from form to content, but it is also true that without form, content cannot emerge. When it comes to formulating a proposition, form comes first; forms are generative not of specific meanings, but of the very possibility of meaning. Despite the familiar proverb, it’s not the thought that counts. Form, form, form, and only form is the road to what the classical theorists called “invention,” the art of coming up with something to say. It follows that familiarizing yourself with form independent of any content you might want to elaborate later is the way to learn how to write a sentence.

  Here’s another exercise that will illustrate the point. Begin with the first stanza of Lewis Carroll’s “Jabberwocky”:

  ’Twas brillig, and the slithy toves

  Did gyre and gimble in the wabe;

  All mimsy were the borogoves,

  And the mome raths outgrabe.

  Now replace the nonsense words with good English words in a way that leads to a meaningful sequence. The exercise is useful because there is no content guiding your performance of it. All you have are forms, but they are enough. If you are a speaker of English you know—although it is the kind of knowledge you may never have articulated—that there are only certain classes of words that can follow “ ’Twas.” “Exciting,” “evening,” “unfinished,” “urban,” “Texas,” or “hilarious” would be okay, but “did”—“ ’Twas did”—would not (unless you were e.e. cummings). You know, to be a bit more technical, that a “linking verb” such as “was” introduces a state or a condition or a location but does not introduce a verbal auxiliary such as “did.” That knowledge, which, again, is formal, constrains what can replace “brillig,” but within that constraint the possibilities are vast. Formal knowledge also tells you that the phrase “the slithy toves” will be composed of a noun and an adjective (“the whispering breezes,” “the agitated monkeys,” “the unhappy suburbanites,” “the beautiful lilies”) and not of two nouns (“the whispers breezes”) or two adjectives (“the agitated beautiful,” as opposed to “the agitated beauty,” which is fine). In exactly the same way, you know that “did” accompanies and specifies the tense or time frame of verbs, so that “gyre and gimble” can be replaced by “sway and bounce” or “breath and shine” or a thousand other combinations, but not by “high and low” or “bright and cheery.”

  Making these substitutions is the easy part. Explaining how you knew how to do it is harder, because it requires bringing to the surface of your analytical consciousness inferences you have made without thinking about them. But it is that extra step that will put you into meaningful contact with the world of forms and help you begin the journey that will end with your being a fully enfranchised and informed citizen of that world.

  Let me say again that by “forms” I do not mean parts of speech or any other bit of abstract machinery. I mean structures of logic and rhetoric within which and by means of which meanings—lots of them—can be generated. The logical structures are the ones we have already met: the structure of relationships between actor, actions, and the objects acted upon. The rhetorical structures are structures of argument (that is what rhetoric is, the art of argument); they too are formal—abstract, contentless—but rather than being the forms that make random words into propositions (sentences), they are forms that link propositions together in more complex units. Relationships are also central to their operation, but they are relationships among statements, not the relationships that must be in place if there are to be statements at all.

  Gerald Graff and Cathy Birkenstein call these forms of argument “templates.” The title of their book, They Say/I Say (2006), identifies one of them. Speakers and writers who can deploy this template know how to summarize conventional wisdom on a topic on the way to disagreeing with it. “They say that money talks, but I say money corrupts.” You might think that disagreeing is a “natural” act, requiring no formal skills. But disagreeing is a learned activity; it is more than just saying no; it involves giving reasons for your rejection of what others have said. As Graff and Birkenstein put it, “they say/I say” enables writers not only to make claims, but also to “map those claims relative to the claims of others.” Organizing a discourse around views you oppose is an achievement of artifice, and it is made possible by a form you have to acquire through practice.

  Although the forms of argument are more numerous than the forms of sentence structure, their number is limited; they can be catalogued (Graff and Birkenstein do it), they can be added to your argumentative repertoire, and you can use them not simply to arrange thoughts but also to create thoughts. Creativity is often contrasted with forms to the latter’s detriment, but the truth is that forms are the engines of creativity. “Our templates,” say Graff and Birkenstein, “have a generative quality, prompting students to make moves in their writing they might not otherwise make or even know they should make” or (I would add) even know they could make. “When we ask students to write sentences using the form ‘at this point you probably object that’—they invariably come up with objections—content—that had never occurred to them and they would never have written on their own.”

  The same thing will happen if you give yourself the assignment of writing a sentence in which three or even four time zones—past perfect, past, present, future—are structured into an a
ccount of related actions. You can start with any simple proposition, say, “We ate the pizza”; and then imagine a prior event, “After we had finished the job”; and then imagine an event in the present, “and now we’re getting ready to go home”; and finally move on to the future, “where we will finish the day eating ice cream.” Or you can start with a contrary-to-fact construction like “Had I been there,” and give yourself the task of completing the sentence using at least two additional tenses. “Had I been there, I would have prevented them from doing it and now I will have to clean up their mess.” You can come to this exercise from any direction—from a future tense assertion from which you have to move backward in time: “I will do this, although long before you proposed it, I had already made up my mind”; from a present assertion: “I reject the idea” to prior and future events: “because you didn’t consult with me, I reject the idea and will consider other options.” There is no end to the variations you can run on this exercise, and in any of the variations, the point will not be the content but the formal requirement—“write a sentence with the following features”—that produces it.

  You can get even more basic. Give yourself the assignment of completing a sentence that begins “Had I.” Examples might be “Had I known you were coming I would have baked a cake,” “Had I read the book I would be able to answer the question,” and “Had I not been there, I wouldn’t be so afraid.” Now ask yourself how many contents are there that could fill this form? The answer is an infinite number. How many forms are there? Only one. Then the hard question: What is the form? What exactly is a “Had I” sentence? What do you know about a sentence the moment you hear or read the words “Had I”? It will take some time to come up with the answer, but in the end you’ll get it, and it will be something like this: In a “Had I” sentence, an action taken or not taken in the past will be related causally to an action you did or did not take at a later point or in the present. This very abstract account is an account of form; as a form, it is empty, but precisely because it is empty—not hostage to any particular content—it serves as a mold into which innumerable contents can be poured. There is no limit to the forms you can practice in this way: “Even though,” “Were I to,” “Notwithstanding that,” “Depending on whether,” “In the event that.” Each of these forms exists to make available a certain content, and apart from them, that content could not be briskly produced. An “Even though” sentence—“Even though I was exhausted, I watched another episode of Law & Order”—makes a complicated statement: it specifies two actions that follow one another but, in the usual course of things, shouldn’t (if I’m exhausted I should turn off the TV), and it signals awareness of the oddness of the sequence even before we know exactly what its specifics are. Without the “Even though” at the beginning of the sentence, you’d have two propositions in tension; with it, the tension is acknowledged before the propositions emerge. Just as you know what words can fill the slots in Carroll’s “Jabberwocky” simply by attending to the poem’s skeletal form, so do you know what kind of relationship between propositions will follow when the first two words of a sentence are “Even though.”

  A famous sonnet by William Wordsworth begins, “Nuns fret not at their convent’s narrow room; / And hermits are contented with their cells; / and students with their pensive citadels.” Wordsworth’s point is that what nuns, hermits, and students do is facilitated rather than hindered by the confines of the formal structures they inhabit; because those structures constrain freedom (they remove, says Wordsworth, “the weight of too much liberty”), they enable movements in a defined space. If the moves you can perform are prescribed and limited—if, for example, every line in your poem must have ten syllables and rhyme according to a predetermined pattern—each move can carry a precise significance. If, on the other hand, there are an infinite number of moves to perform, the significance of any one of them may be difficult to discern. (This is one of the insights of information theory.) That is why Wordsworth reports himself happy “to be bound / Within the Sonnet’s scanty plot of ground.” It is a scanty plot because it is bounded, and because it is bounded, it can be the generator of boundless meanings.

  This, then, is my theology: You shall tie yourself to forms and the forms shall set you free. I call this the Karate Kid method of learning how to write. In the 1984 cult movie (recently remade), the title figure is being trained to perform in a match, but rather than being instructed in a match’s rhythms and demands, he is asked by his teacher to practice polishing cars (“wax on, wax off”) and painting fences. Although the kid thinks he isn’t learning anything, he is learning everything; he is learning the formal motions that, when actual combat occurs, will come to him naturally. Like the verbal forms that enable thought and meaning, these physical forms enable action in a sequence, even though they are essentially static and abstract. Know what makes a sentence more than a random list, practice constructing sentences and explaining what you have done, and you will know how to make sentences forever and you will know too when what you are writing doesn’t make the grade because it has degenerated into a mere pile of discrete items.

  CHAPTER 4

  What Is a Good Sentence?

  But even if you follow my advice and become adept at producing well-formed sentences and at diagnosing ill-formed ones, you are only halfway home. You may be able to write a sentence; you still have to learn how to write a good sentence, and before you can do that, you have to know what a good sentence is. And this is where content, banished from the discussion so far, comes roaring back in. Content, the communication in a thrilling and effective way of ideas and passions, is finally what sentences are for. But just as you can’t produce a sophisticated meal without a thorough knowledge of ingredients, seasonings, sauces, temperatures, utensils, pots, pans, and much more, so you can’t produce powerful content in the shape of sentences that take your readers by storm without having a command of the devices—formal devices—that are at once content’s vehicles and generators. Those devices make the enterprise go, but they are not, except for linguists and grammarians and sentence nuts like me, ends in themselves. The end, the goal, the aspiration is to say something, and the something you want to say will be the measure of whether you have written a sentence that is not only coherent but good.

  “Good,” however, is a measure that cannot be reduced to a set of features or a taxonomy. You can’t produce a good sentence—a sentence not only well formed but memorable—by consulting a recipe. Indeed, even a bad sentence—a sentence that creaks or is clumsy or is overblown or is impossibly loose—can be a good sentence. There’s an NPR program called The Annoying Music Show, and when the remastered set of Beatles albums came out in 2009, an episode was built around it. The cuts played included Tiny Tim singing “Hey Jude” and Telly (Kojak) Savalas singing “Something in the Way She Walks.” These performances were truly bad and they were truly good. They were bad because in both instances the voice, the pace, and the intonation were all wrong for the song; hearing the Beatles’ version in the background of these dismal renditions was painful. But the performances were good for the same reason. The point of playing the cuts—not the point of recording them originally—was to provoke just such a counterpoint in the listener’s mind. That’s what makes the whole thing funny; and if it’s funny, it’s good, at least in the context of The Annoying Music Show.

  The important word in the previous sentence is “context” and it should be paired with another, “purpose.” People write or speak sentences in order to produce an effect, and the success of a sentence is measured by the degree to which the desired effect has been achieved. That is why the prescriptive advice you often get in books like Strunk and White’s The Elements of Style—write short sentences, be direct, don’t get lost in a maze of piled-up clauses, avoid the passive voice, place yourself in the background, employ figures of speech sparingly—is useful only in relation to some purposes, and unfortunate in relation to others. The first thing to ask when writing a
sentence is “What am I trying to do?”

  The answers are innumerable (that is why examples, not rules, are what learning to write requires), but one of them won’t be “to tell it like it is.” It is often said that the job of language is to report or reflect or mirror reality, but the power of language is greater and more dangerous than that; it shapes reality, not of course in a literal sense—the world is one thing, words another—but in the sense that the order imposed on a piece of the world by a sentence is only one among innumerable possible orders. Think about what you do when you revise a sentence: You add something, you delete something, you substitute one tense for another, you rearrange clauses and phrases; and with each change, the “reality” offered to your readers changes. An attempt to delineate in words even the smallest moment—a greeting in the street, the drinking of a cup of coffee, the opening of a window—necessarily leaves out more than it includes, whether you write a sentence of twenty words or two thousand. There is always another detail or an alternative perspective or a different emphasis that might have been brought in and, by being brought in, altered the snapshot of reality you are presenting. This does not mean that your sentences are always incomplete and that you should strive to cram them full of everything in the universe on the model of what Molly Bloom does in the monologue from Joyce’s Ulysses. Sentence writers are not copyists; they are selectors. It is impossible not to select when you are making an assertion. The goal is not to be comprehensive, to say everything that could possibly be said to the extent that no one could say anything else; if that were the goal, no sentence could ever be finished. The goal is to communicate forcefully whatever perspective or emphasis or hierarchy of concerns attaches to your present purposes.

  In his great book How to Do Things with Words (1962),

  J. L. Austin considers the apparently simple sentence “France is hexagonal.” He asks if this is true or false, a question that makes perfect sense if the job of a sentence is to be faithful to the world. His answer is that it depends. If you are a general contemplating a coming battle, saying that France is hexagonal might help you assess various military options of defense and attack; it would be a good sentence. But if you are a geographer charged with the task of mapping France’s contours, saying that France is hexagonal might cost you your union card; a greater degree of detail and fineness of scale is required of mapmakers. “France is hexagonal,” Austin explains, is true “for certain intents and purposes” and false or inadequate or even nonsensical for others. It is, he says, a matter of the “dimension of assessment”—that is, a matter of what is the “right or proper thing to say as opposed to a wrong thing in these circumstances, to this audience, for these purposes and with these intentions.”