What Writers Mean by “Flow”


We all have our pet peeves. One of mine is the word flow. In my three decades as a creative writing teacher, I’ve heard it literally thousands of times. It’s a rare class in which I don’t hear “It flows” or “It doesn’t flow” offered as an explanation of what’s good or bad about a story we’re discussing. What bothers me about the word—beyond the fact that I hear it so often—is that my students generally don’t seem to understand what they mean by it. They intuitively recognize flowing prose when they read it, but they’re not sure what constitutes it. If I ask them what makes a particular sentence or story “flow,” they’ll answer with semi-synonyms that are equally vague: “It’s the rhythm,” they’ll say, or “the pace,” “the style.” They can’t really define it.

I’m afraid I can’t either, at least not adequately. My response to flow is undoubtedly as intuitive as theirs, for when we talk about flow we’re talking about an element of writing that is more music than meaning and thus beyond rational explanation—perhaps even beyond language itself. Hence it’s extremely difficult to discuss, much less define or teach.

Difficult, but not impossible. While there is much about the flow of prose that will inevitably remain instinctual, there are some aspects of it that can be discussed, understood, and even practiced. The principal purpose of this essay is to try to make our unconscious understanding of flow conscious, so that those of us who don’t instinctively write flowing prose can practice the skills and strategies involved until they become so habitual they are, for all practical purposes, instinctive.

Let’s begin by looking at a paragraph that—my students and I agree—flows extremely well. It’s the opening paragraph of a story submitted to Ford Madox Ford in 1909, when he was editor of the English Review. According to Ford, the story was sent to him by a schoolteacher from Nottingham who informed him that it had been written by a young, unpublished author who was “too shy to send his work to editors.” Ford didn’t expect the story to amount to much, of course, but the moment he finished reading the first paragraph, he laid the story in the basket reserved for accepted manuscripts and announced to his secretary that he had discovered a literary genius—indeed, “a big one.” And that night, he told his dinner companion H.G. Wells the same thing, and Wells passed the word on to people seated at a nearby table. Before the night was out, two publishers had asked Ford for first refusal rights to the young author’s first book. All of this happened before the author even knew his work had been submitted to an editor, and it all resulted from a single paragraph. What was it about this paragraph that impressed Ford so much that, without reading a single word further, he accepted the story and judged its unknown author a genius? In his explanation of his decision he points out many of the paragraph’s virtues, but he stresses two in particular that convinced him he could trust the author “for the rest” of the story: The author employs “the right cadence,” Ford says, and “He knows how to construct a paragraph.” In my opinion, cadence and paragraph construction are two of the principal things we talk about when we talk about flow. If I’m right, the paragraph’s flow is a major reason—perhaps even the principal reason—Ford recognized genius in it.

Lest this turn into an essay on how to create suspense, let me say now that the then-unknown author of this paragraph is D.H. Lawrence and that it is the opening of “Odour of Chrysanthemums,” his first published story. Here’s the paragraph:

The small locomotive engine, Number 4, came clanking, stumbling down from Selston with seven full wagons. It appeared round the corner with loud threats of speed, but the colt that it startled from among the gorse, which still flickered indistinctly in the raw afternoon, out-distanced it at a canter. A woman, walking up the railway line to Underwood, drew back into the hedge, held her basket aside, and watched the footplate of the engine advancing. The trucks thumped heavily past, one by one, with slow inevitable movement, as she stood insignificantly trapped between the jolting black wagons and the hedge; then they curved away towards the coppice where the withered oak leaves dropped noiselessly, while the birds, pulling at the scarlet hips beside the track, made off into the dusk that had already crept into the spinney. In the open, the smoke from the engine sank and cleaved to the rough grass. The fields were dreary and forsaken, and in the marshy strip that led to the whimsey, a reedy pit-pond, the fowls had already abandoned their run among the alders, to roost in the tarred fowl-house. The pit-bank loomed up beyond the pond, flames like red sores licking its ashy sides, in the afternoon’s stagnant light. Just beyond rose the tapering chimneys and the clumsy black headstocks of Brinsley Colliery. The two wheels were spinning fast up against the sky, and the winding engine rapped out its little spasms. The miners were being turned up.

When I show this paragraph to my students, they invariably praise its flow. Even those who complain that the prose is too “descriptive” or “old-fashioned” (words that many students consider synonymous these days, alas) find the flow of this overly descriptive, old-fashioned prose to their liking. When I press them for an explanation of what makes the passage flow, however, I rarely get more than the verbal equivalent of shrugged shoulders. To help clarify for them, and me, what makes Lawrence’s paragraph flow, I offer them a revision that, we all agree, does not flow. I won’t subject you to the entire revision; my point should be painfully obvious after you see how I’ve butchered Lawrence’s first two sentences.

The small locomotive engine came down from Selston. It was Number 4. It clanked and stumbled. It had seven full wagons. It appeared round the corner. It made loud threats of speed. It startled a colt from among the gorse. The gorse still flickered indistinctly in the raw afternoon. The colt out-distanced the train at a canter.

Awful, isn’t it? But why? My sentences contain the same content as Lawrence’s, and that content is presented in essentially the same order, yet the passage is as stagnant as the afternoon light Lawrence describes. So clearly neither content nor order determines flow. (For further evidence, take a look at Raymond Queneau’s Exercises in Style, in which he tells the same brief incident ninety-nine times, keeping its content and order intact and changing only the style and, therefore, the flow.) Nor does ease of reading determine flow, since the revision is significantly easier to read than the original—even a grade-schooler could follow it. So what is the essential difference between the two versions? Nothing more, or less, than variety of sentence structure. That sentence structure is related to flow is an obvious point, no doubt, but if there’s one thing I’ve learned as a writer and a teacher, it’s that when something is obvious, we tend not to pay it sufficient attention. So let’s pay closer attention to the relationship of sentence structure and flow in Lawrence’s paragraph.

There are, of course, four basic types of sentence structure—simple, compound, complex, and compound-complex. But within these four general categories, there are many different types of structure, as the grammarian Virginia Tufte has demonstrated so superbly. In her book Grammar as Style, Tufte defines—and illustrates—innumerable ways to structure sentences, using left-, mid-, and right-branching modifiers, balance, repetition, coordination, inversion, apposition, and a vast array of other techniques. Significantly, Lawrence uses all four sentence types in his paragraph, not to mention many of the structural techniques Tufte describes. More importantly, seven of his ten sentences are either complex or compound-complex, the two types that permit most variation in structure. For example, both the fourth and seventh sentences are complex, but one contains five dependent clauses and the other only one.

Because of the variety of sentence structure in the paragraph, Lawrence’s sentences range from 6 to 62 words. I use only the simple sentence pattern in my revision, however, and so my sentences range—if they can be said to “range” at all—from 4 to 9 words. According to Tufte, “The better the writer, … the more he tends to vary his sentence length. And he does it as dramatically as possible.” Since variation of sentence length results from varying sentence structure, ultimately it’s our syntax that determines whether our prose flows or not. As Stephen Dobyns tells us, syntax is like a landscape: If it’s too uniform, as in my revision, our prose will look more like Nebraska than Switzerland. A variety of sentence structure—and therefore of sentence length—will give our prose a more flowing, and appealing, landscape.

But because we don’t think enough about syntax when we read, we don’t think enough about it when we write, either. As a result, our work—my own, as well as my students’—tends to rely far too heavily on the two most basic sentence structures, the simple and compound. There’s nothing inherently wrong with either, of course. In fact, the simple sentence is the base structure, the ground note of all prose. We can’t, and shouldn’t, do without it. But it is also the structure with the least possibility for variation in syntax and length since there are no other clauses, dependent or independent, attached to its single independent clause. The compound sentence structure is only slightly more complicated since it merely connects simple sentences with a conjunction. Because these two sentence types so dominate our writing, they prevent our prose from achieving that flowing cadence that marks the best fiction. As Robie Macauley and George Lanning have said, the simple, minimalist style “has its Spartan virtues but it also has its Spartan vices.” And chief among those vices is a lack of flow.

Why are the simple and compound sentence types so dominant in our prose today? I asked my students and colleagues this question, and virtually everyone gave me the same answer: It all goes back, they confidently asserted, to the influence of Hemingway. But I disagree: Hemingway’s simplicity is far more a matter of diction than of syntax. Like Lawrence, Hemingway knew how to vary sentence structure so his paragraphs flow. If you look at random paragraphs from his work, you’ll notice how the simplicity of his diction exists within the context of complex syntax. The opening paragraph of “A Clean, Well-Lighted Place” is a good example.

It was late and every one had left the café except an old man who sat in the shadow the leaves of the tree made against the electric light. In the daytime the street was dusty, but at night the dew settled the dust and the old man liked to sit late because he was deaf and now at night it was quiet and he felt the difference. The two waiters inside the café knew that the old man was a little drunk, and while he was a good client they knew that if he became too drunk he would leave without paying, so they kept watch on him. 

The prose here is admirably straightforward and clear, but its syntax is by no means simple. All three of these sentences are compound-complex, and no two share the same structure. The number and placement of dependent and independent clauses in each varies significantly; the sentences have two, five, and three independent clauses, respectively, and one, four, and three dependent clauses. And the placement of the dependent clauses varies widely, too: The one in the first sentence follows an independent clause whereas three of the four in the second sentence precede independent clauses. And in the third sentence, two dependent clauses are embedded in the middle of independent clauses. Flaubert once said that “The sentences in a book must quiver like the leaves in a forest, all dissimilar in their similarity,” and these sentences do exactly that.

I don’t believe for a millisecond that Hemingway was thinking consciously about varying the placement of dependent clauses in these sentences—at least not when he first drafted them. No doubt he was responding to an instinctive sense of what would make the paragraph flow. We, too, should do our best to follow the ebb and flow of our rhythmic instincts, but to ensure that we have the skills needed to follow our instincts, we should also practice varying the structures and lengths of our sentences as rigorously as concert pianists practice scales. 

While I don’t think Hemingway can be held accountable for the current dominance of simple sentence patterns, I do think it’s true that many of his followers have tended to use syntax as simple as their master’s diction. This is certainly true of Raymond Carver—or, at least, of Raymond Carver as edited by Gordon Lish (as D.T. Max has revealed, Carver’s hyper-minimalist style was due largely to Lish’s drastic editing)—and it is also true of many of the writers who were influenced by the stories in Will You Please Be Quiet, Please? and What We Talk About When We Talk About Love. But the best of Hemingway’s followers use syntax nearly as complexly. Even Carver, once he no long-er allowed Lish to edit his work, varied his sentence structure and length considerably more than many of Hemingway’s other disciples (not to mention Carver’s own devotees). Witness the opening paragraph of “Menudo,” whose four sentences use three different structures and vary in length from 4 words to 35.

I can’t sleep, but when I’m sure my wife Vicky is asleep, I get up and look through our bedroom window, across the street, at Oliver and Amanda’s house. Oliver has been gone for three days, but his wife Amanda is awake. She can’t sleep either. It’s four in the morning, and there’s not a sound outside—no wind, no cars, no moon even—just Oliver and Amanda’s place with the lights on, leaves heaped up under the front windows.

There’s nothing wrong with simplicity, in short, if it’s only apparent, not actual. The best simple writing is, at its deepest level, the level of structure, complex.

So if we can’t blame the current tendency toward simplicity of syntax on Hemingway’s example, or even on Carver’s, why is it so dominant? It’s not, I’m sure, because we lack the linguistic skills to write more complexly (provided, of course, that we practice those skills). And it’s not, I hope and pray, because we agree with Robert Bly’s ludicrous assertion that “The use of subordinate clauses in sentences reveals the writer’s tendency to fascism.” One reason simple syntax dominates our writing, I believe, is that such sentences are just plain easier to write. They take less effort, less thought. Plus, there’s less risk of grammatical mistakes or—a worse crime in these dumbed-down times—of appearing pretentious. To some of us, it seems, writing a compound-complex sentence is about as embarrassing as wearing an ascot to a hoedown.

But I suspect the most important reason we overuse simple structures is that we’re excessively afraid of not writing clearly. Often, in the struggle to express a complicated, only half-understood idea or emotion, we sacrifice the truth we’re trying to convey in order to write simply and clearly. As Wright Morris has said, “When we give up what is vague in order to be clear, we may have given up the motive for writing.” Donald Barthelme also questions the value, even the possibility, of creating art that is simple and clear. “However much the writer might long to be, in his work, simple, honest, and straightforward,” he says, “these virtues are no longer available to him. He discovers that in being simple, honest, and straightforward … he speaks the speakable, whereas what we are looking for is the as-yet unspeakable, the as-yet unspoken.”

So am I—or Morris or Barthelme—advocating the overthrow of English grammar and the production of vague, convoluted prose? Hardly. What we are advocating, however, is a conscious struggle against our natural inclination to simplify, for the sake of clarity and ease of reading, the complex, uncertain ideas and emotions that constitute our experience. And the best way to struggle against this inclination is to struggle against our tendency toward simplicity in syntax. The more we experiment with syntax, then, the more opportunities we give ourselves to discover our thoughts and express what would otherwise either remain vague or be sacrificed in the name of clarity.

Thus, altering our syntax does more than help us write flowing prose; it allows us to get our thoughts off the normal track on which they run. Syntax is nothing if not the very structure of our thought, so if we change the way we think, we can sometimes change what we think. But don’t take my word for it; take Yeats’s. In the introduction to his collected plays, he wrote, “As I altered my syntax I altered my intellect.” Morris also believes that changing our syntax changes the way we think. According to him, “syntax shapes the mind … and does our thinking for us. If the words are rearranged, the workings of the mind are modified.” And if the words are rearranged, the rhythm of those words is modified, too, of course. According to Robert Hass, it’s this alteration in rhythm, more than the alteration in meaning, that changes our intellect. “New rhythms,” he has said, “are new perceptions.” In any case, the more we concentrate on altering our syntax, the more we free ourselves to discover other modes of thought. I’m not sure I’d go as far as Yeats, Morris, and Hass do, though, and assert that changing our syntax actually changes our intellect. Rather, I believe that as we alter our syntax, we discover our intellect—i.e., we find ways to say what we always knew but never knew we knew, our deepest beliefs and feelings. And it just may be that we discover not only the self but also the world. Bertrand Russell certainly believed syntax revealed the nature of outer as well as inner reality, for he concludes his An Inquiry Into Meaning and Truth with these words: “For my part, I believe that, partly by means of study of syntax, we can arrive at considerable knowledge concerning the structure of the world.”

Given this relationship between syntax, thought, and discovery of both self and world, it shouldn’t be so surprising that some of our greatest writers blossomed when they abandoned their native languages to write their work. As Morris says, “In this release from the overfamiliar, the apparently exhausted, and immersion into new resources, we may understand better than we did in the past the flowering of a talent like Conrad’s. The new and strange language is part of a new consciousness.” Nabokov is another example. He was so dissatisfied with his original Russian version of Lolita that he destroyed it. Only when he began to rewrite the novel in English, he says, did he find the syntax appropriate for the book, the syntax that made the book conform to what he calls “its prefigured contour and color.”

We may not alter our own intellects when we alter our syntax, but by discovering and expressing them, we just may alter our readers’ intellects. Indeed, it’s possible that one of the things we talk about when we talk about flow is the feeling that the writer’s syntax is altering our consciousness, making us think—and therefore feel—in new ways.


You might also like:

  • No Related Posts