Some Observations on the Line



By Miller Williams

To pay attention to the line is not to suggest that a poem exists on the page. It doesn't. A poem comes into existence when the imagination of a writer and the imagination of a reader confront one another inside an act of language. The writer and the reader bring to that confrontation different imaginations structured by different associations. The poem in print is the ground on which the meeting takes place.

The line is the structural and functional unit of the printed poem, as decidedly as the paragraph is the unit of both structure and function in exposition and as the scene is in fiction. This is not to say that the line is necessarily a unit of sense, of course, but a poem doesn't work as a poem when the lines don't work as lines. The question is, what does it mean for a line to work? The answer is, to borrow a phrase of Auden's, simple and hard. It means that at the end of a line the reader feels rhythmically pleased but expectant. (In the case of the last line the expectation will be for nothing to follow, which is the most radical change in the poem.) It is this expectation that creates much of a good poem's sense of forward motion and, in the fulfillment of it, the sense of pleasure.
The anticipation created by the line's ending may be for the completion of a rhythmical pattern.

Shave and a haircut
Six bits

It may be for the completion of a statement, which is to say, simply, that the line is enjambed. In verse with a pattern of rhyme, the expectation is partly for the completion of the pattern each time a rhyme is begun. To speak of a pattern of rhyme is still to speak of the line, because rhyme (assuming that it's terminal) is a form of line break and thus is a function of the line.

Frequently, especially in freer verse-where there is no dependable pattern of rhyme and therefore no expectation-the end of a line will raise a question in the reader's mind, and the beginning of the next line will provide an answer. The following are some examples of how the reader is involved in this kind of juncture. From Louis Simpson's "American Poetry":

Whatever it is, it must have
A stomach that can digest
Rubber, coal, uranium, moons, poems.

From Archibald MacLeish's "Winter Is Another Country":

if this would end
I could endure the absence in the night
The hands beyond the reach of time, the name
Called out.

This kind of juncture is also important in more formal verse, such as William Meredith's "The Open Sea":

Nor does it signify, that people who stay
Very long, bereaved or not, at the edge of the sea
Hear the drowned folk call.

At the line breaks in these brief passages, a reader can ask-a good reader will ask-"What?" or "Then what?" or "How long?" The asking and the move to the answer not only propel the reader through the poem but heighten the reader's sense of participation in the poem. .Every time a word is added to a poem, the poet has made a decision about a line's relative and absolute length; every time a line is ended with a rhyme word, the poet makes a decision about the quality of sound setting up a rhyme set or decides whether to answer the first sound with true or slant rhyme. In the case of topographical or spatial poems like many of Cummings , and Ferlinghetti's, a decision is made about the placement of the line on the page. It is by all these means that a poet makes the line, and the poem, a more effective construction so that it becomes a more convincing illusion of conversation.

If a reader is to take part in the experience of a poem, the poem must be credibly of this world. We trust minds that talk our talk, and we are excited by minds that talk it with energy and leave as much as possible still to be said-minds, that is, that invite us into dialogue in our own language about the phenomenological world we live in. What is not always recognized is the intriguing relationship between reality and illusion.

Plain talk doesn't make for conversationality in poetry. The very fact of a poem is theatrical. We know that the realm of Prufrock is not the "real" world; it's an illusion of the world, or a part of it, as the stage is in a theater.

Reality in a framework of illusion (an actual living-room conversation in a play) or illusion in a context of reality (Lear's mad scene at a bus stop) gives a sense of the grotesque. Reality in a context of the "actual," phenomenological world or illusion in a context of a posited world gives a sense of the real.

This obviously invites a discussion of diction, but it is something that wants saying-I think it insists on being said-in an examination of any aspect of poetry, and it bears directly, as I mean to show, on the study of the line. The poet engineers the line and contrives line breaks so that the reader-in a context of illusion not unlike that represented by the proscenium arch-believes that the lines are natural things for a human to utter; the reader decides to believe it, that is, wants to and does, as we believe Hamlet when he is on the stage but not when he is on the sidewalk.
Poetry, like all art, is ritual, and ritual doesn't want conversation. The poet balances the two demands, plays conversationally against form, and finds in this tension much of the energy that means life to a poem.

Some of the most effective means of holding this balance, of heightening the sense of ritual or increasing the sense of spontaneity, are discovered in the way the lines end. Enjambment is the termination of a line at a point other than at the end of a phrase; it tends to increase the feeling of conversation. End-stopping is the cotermination of a line and a phrase or sentence; it tends to increase the feeling that the reader is involved in a ritual act.

Enjambment increases the sense of the lyric and compromises the ritualistic effect of rhyme. It was especially popular with the Elizabethan poets but-understandably-was not much favored by the neoclassicists. The romantic poets returned to the practice as an important aspect of their release from the formal strictures of the poetry of their recent past. The French, German, and Spanish histories have not been much different. Attitudes toward enjambment-from disapproval to tolerance to preference-have changed as the larger critical and aesthetic views of the society's literature have shifted.

Slant rhyme-as a form of line break-makes a poem more conversational than true rhyme, less conversational than no rhyme.

The metrics of the line break requires finer distinctions, but before turning to them, I feel compelled to make a statement correcting a general misapprehension about metrics: the nature of the foot at the end of a line is as relevant to nonpatterned poetry as it is to what we call formal poetry. No matter how purely accentual a line may be until the end, it is in the nature of the language that the last syllables in the line are going to be recognizable and that the reader is going to hear them as accented or not. This is the same thing as saying that the reader will recognize the terminal foot. The poet has to take this into account or lose some control over what the poem is doing.

An iambic or anapestic ending conveys a greater sense of formality than a trochaic or dactylic ending, or-to put it in less traditional terms-a final stress suggests greater seriousness than when the last stress is on the penultimate or an earlier syllable. The frequent use of the unstressed ending by the Elizabethan and Jacobean verse dramatists accounts for much of the apparent conversationality that marks their work.

Midway between these options is what is usually called the weak ending, an anapestic or iambic ending with less than full stress on the accent. This also tends to carry a suggestion of conversationality, but without risking the sense of lightness sometimes created by completely unaccented endings:


u / u / u / u /
I thought you sent the money in

This was a favorite device of Marianne Moore's; it is-with all her syllabics­a purely metrical consideration.

The omission of what would be a final unaccented syllable in a regular metrical line (called truncation or catalexis) is a useful device available to the poet writing in trochaic feet, because it helps to avoid the monotony that the trochaic line is likely to produce.


/ u / u / u /
What am I supposed to say?


rather than:


/ u / u / u / u
What am I supposed to tell him?


Another option, obviously, is the length of the line itself. Most generalizations about this run into immediate contradictions. It is often said that as the line becomes shorter the poem becomes lighter, but then we have to take into account the fine small poem by Donald justice, "Poem to Be Read at 3A.m.," a moving and contemplative piece with one to three stresses per line, and most of Ogden Nash's comic poems, with lines like old fence rails.


What is probably safe to say about line length is that a line longer than six stresses is likely to be broken down in the reader's mind into smaller units, 5-2, 4-3, 5-3, or 4-4. It is also probably safe to say that the five-stress line is the most flexible line in English, the one to which most readers come and with fewest suspicions.


The last contact the poet has with the reader's imagination is in the poem's resolution; because of this, it's a highly important and sensitive moment in the poem's life. There are at least nineteen or twenty means by which a poet can change the reader's expectations so that the ending of a poem seems to be in the natural order of things; some of these function through the line break and changes in line length.


Here, as in all talk about poems, there are no hard rules. But there are principles that, when heeded, result most of the time in a more effective poem. This is all that rules can mean in poetry. I doubt if they mean more in any other art, but this makes the understanding of these principles and the use of them by the poet no less essential.


Three of the most important principles relating to closure as a function of the line are the following: A reader will tend to feel that it's right that the poem end 1) when a line is noticeably shorter or longer than the established line in the poem; 2) when an established pattern of terminal rhyme is modified or when a pattern is introduced where none had been; 3) when there is a shift from run-over to end-stopped lines. Examples of these types of closure can easily be found in most any anthology of poems from any period.


Much of the difference between verse and prose is found in the line and what it does. Prose "broken down" so that it has an uneven right margin does not become verse, because the length and ending of a line of verse are not arbitrary; they are contributing parts of an organism. And when a verse poem is put into paragraph form, something of what it does as a poem is lost.


The fact that good works sometimes lie uncertainly in more than one genre does not mean that the genres are not real. They are real, and the separate natures of the paragraph, the scene, and the line lie at the root of the distinctions between them. Our understanding of what a poem is starts with such an awareness.