The Music of Poetry

Most western music since the Renaissance is organized by a particular 
key, such as C major; musicians use the word tonal to describe such music. Some non-western languages, such as Mandarin Chinese, are tonal in a different sense: the Mandarin word ma, depending on whether the voice rises or remains level when uttering it, may mean either horse or mother. Although syllable stress may determine the meaning of English words, allowing us to hear the difference between contract and contract or between minute and minute, English is not a tonal language. And while English syllables may be uttered at different pitches relative to one another, neither is the sonic life of the English language tonal in the musical sense. What then do poets, in contrast to linguists or musicians, mean by the word tone? A poem’s diction, rhythm, or syntax is palpably describable, but asking a poet to produce a poem with an interesting tone is like asking a chef to produce a meal that tastes good: if successful, the chef will be thinking about the manipulation of particular ingredients. You can’t reach into the pantry for a cup of tone.

Consider Ezra Pound’s “In a Station of the Metro.”

The apparition of these faces in the crowd;

Petals on a wet, black bough.

In the first line, the multisyllabic word apparition is necessarily pronounced apparition — two unstressed syllables followed by a stressed syllable and an unstressed syllable. The subsequent function words (of, these, in, the) aren’t generally stressed in an English sentence unless something directs us to do so. Tension between syntax and line may produce that direction — 

The apparition of these faces in

The crowd

 — thereby throwing emphasis on an otherwise unstressed preposition; but Pound’s poem offers no such direction, leaving us to stress the more semantically charged nouns, as the function words fall away.

The apparition of these faces in the crowd;

Since the nine unstressed syllables in this line are broken into three groups of three, each group followed by a stressed syllable, it seems the poem may possibly be establishing a regular rhythmic pattern, one that might be repeated — ti ti ti tum ti ti ti tum ti ti ti tum. We won’t know if that’s true, however, until we listen to the second line. Pound is famous for being one of the inventors of free verse in English, but he also wrote artfully metrical verse throughout his entire career.

In the second line, the multisyllabic word petals is necessarily 
pronounced petals, and, once again, the function words following this noun remain unstressed (“Petals on a”). This little string of unstressed syllables throws us forward into the more semantically charged adjectives and noun concluding the line (“wet, black bough”), which turns out to sound nothing like the first line.

Petals on a wet, black bough.

This density of stressed syllables feels emphatic in itself, and the density is reinforced by the rhyme of “wet” with the first syllable of 
“petals,” the alliteration of “black” with “bough,” and, most importantly, by the way in which the second line’s irregular rhythm (tum ti ti ti tum tum tum) disrupts the regularity of the first (ti ti ti tum ti ti ti tum ti ti ti tum). The whole poem delivers us into the concluding triplet of stresses, transforming what would otherwise be a disposable observation into a thrillingly repeatable event; we don’t reread “In a Station of the Metro” to remember what color the bough is.

Now consider the second quatrain of  Shakespeare’s twelfth sonnet.

When lofty trees I see barren of leaves,

Which erst from heat did canopy the herd,

And summer’s green all girded up in sheaves

Borne on the bier with white and bristly beard    ...    

The syllables at the ends of these iambic pentameter lines sound like one another, though leaves sounds closer to sheaves than herd does to beard, at least to twenty-first-century ears. At the same time, syllables echo one another within the lines. In the line “When lofty trees I see barren of leaves,” three of the five stressed syllables share a vowel (trees, see, leaves), while two of them share an initial consonant (leaves, loft); the first syllable of barren stands alone. In the line “Borne on the bier with white and bristly beard,” the stressed syllable beard shares an initial consonant with borne and bristly and, more forcefully, beard shares an initial consonant and vowel sound with bier; the syllable white stands alone.

An abundance of echo tends to reduce sense to nonsense, discourse to incantation, and while a poet may covet this effect, as in these lines from Stevens’s “Bantams in Pine-Woods” — 

Chieftain Iffucan of Azcan in caftan

Of tan with henna hackles, halt!

— or this line from Swinburne’s “Nephelidia” — 

From the depth of the dreamy decline of the dawn through a notable nimbus of nebulous noonshine

— great lines of poetry more often embrace and resist the allure of echo simultaneously. If all five stressed syllables alliterated with one another in Shakespeare’s line “Borne on the bier with white and bristly beard,” the line would lose its rhythmic vitality: “Borne on the bier with bright and bristly beard.” If the glut of echo reinforcing the preponderance of stressed syllables in the second line of “In a Station of the Metro” (“Petals on a wet, black bough”) also distinguished the first line (“The apparition of these faces in the crowd”), the poem would have nowhere to go.

The sonic effect of every poem, whether we call it formal or free, traditional or innovative, depends on a concerted relationship between stressed and unstressed syllables, between syllables that may or may not echo one another: these effects are the material result of the poet’s manipulation of the medium. If the source of a poem’s tone seems more elusive, it is largely because tone has not adequately been described in relationship to the syllables that produce it. We might easily sense the tone of Pound’s or Shakespeare’s poem, but how precisely might we reproduce it, using the basic elements of the English language?

Ancient Greek rhetoricians used the word tonos (meaning, literally, a tightening or stretching, as a string on a lute might be stretched) to characterize the quality of an orator’s performance of a speech; they were interested in describing the sounds produced by a particular human throat on a particular occasion. Predictably, our now more common sense of tone as the quality of a written text, rather than the quality of a particular oratorical performance, began to prevail with the rise of print culture in the seventeenth century, and in the early years of the twentieth century the influential literary critic I.A. Richards defined poetic tone as a speaker’s “attitude to his listener.” But it’s important to remember that Richards uses the words speaker and listener metaphorically; he’s talking about a written text, not an oratorical performance. What he refers to (equally metaphorically) as a poem’s tone is generated by the material characteristics of rhythm and echo that I’ve been describing so far.

Consider a simple sentence, three words, three syllables: You said that. Depending on how we understand the syntax of this sentence, it might be a statement (You said that), a question (You said that?), or an exclamation (You said that!), and each one of these three sentences sounds a little different from the other two. One feels one’s voice rising in the final syllable of the question, falling in the final syllable of the statement, and falling more dramatically in the final syllable of the exclamation.

Using the word intonation differently from the way musicians do, linguists would say that the statement, question, and exclamation are distinguished by three different intonations of the simple sentence. But the possibilities don’t end there, for each of these three intonations will in turn sound different from itself, depending on which of the three syllables is stressed; linguists refer to the emphasized syllable as the tonic syllable of an intonational unit (which is often but not necessarily a grammatical clause or phrase).

You said that?

You said that?

You said that?

In real life or on the stage, the context of an ongoing conversation might indicate these differences instantly, though often in real life we make mistakes about context, and sometimes we intentionally violate the decorum of a particular context; a child who asks “You said that?” at the wrong moment might be told sharply to “watch your tone” — a reprimand that deploys a visual metaphor to assert control over a sonic quality. How do you watch a tone?

The child might not know what to do if he were enjoined to shift the stress from the second syllable of the question to the first, but such adjustments are more precisely what we’re responding to when we talk about tone: given the three available syntactical options (statement, question, exclamation), along with the three available placements of the tonic syllable (first, second, or third syllable emphasized), we could say that the sentence “you said that” might be uttered in nine different ways. Reading a poem on the page, how do we know that we’re supposed to understand a line as “You said that” and not as “You said that”?

The most obvious answer is typography, to which I’ve been resorting throughout this discussion; by highlighting a particular syllable or word, italic or bold type can signal the intonation of a particular word or phrase. Especially in his earlier poems, such as “A Coin for Joe, with the Image of a Horse; c. 350–325 BC,” Frank Bidart does this with bracing accuracy — 

you          chip of the incommensurate

closed world         A n g e l

— but when deployed without Bidart’s precision, typographical 
indications of rhythm may inadvertently highlight a poem’s lack of 
intrinsic rhythmic vitality.

Below the surface-stream, shallow and light,

Of what we say we feel — below the stream,

As light, of what we think we feel — there flows

With noiseless current strong, obscure and deep,

The central stream of what we feel indeed.

Here, Matthew Arnold wants us to feel that, among the five stressed syllables in the line “As light, of what we think we feel — there flows,” the greatest emphasis falls on think, making it the tonic syllable: “As light, of what we think we feel — there flows.” Arnold’s use of italics threatens to seem like compensation for an inability to control the intonation of his lines by poetic means, as if the strategic variation of the syntax within an ongoing metrical pattern were not enough.

Such control is precisely what we’ve heard Shakespeare muster in the lines from his twelfth sonnet:

Borne on the bier with white and bristly beard.

Four of these stressed syllables are linked by similar sounds (borne, bier, bristly, beard). But because white stands apart, we’re liable to hear this syllable as the most emphatically emphasized of the five, giving the entire line a particular intonation: not “Borne on the bier with white and bristly beard” or “Borne on the bier with white and bristly beard” but

Borne on the bier with white and bristly beard.

Similarly, because four of the five stressed syllables in the following line share similar sounds (loft, trees, see, leaves) — 

When lofty trees I see barren of leaves

 — we’re liable to hear the first syllable of barren as the most emphatically emphasized of the five, especially because this intrusion of an alien phoneme is reinforced by rhythmical variation. The line scans easily as an iambic pentameter, but the fourth iamb is flipped into a trochee (barren), throwing extra emphasis on the syllable that already stands apart, so we hear the line not as “When lofty trees I see barren of leaves” or as “When lofty trees I see barren of leaves” but as

When lofty trees I see barren of leaves.

Working together in this way, rhythm and echo determine the intonation of innumerable metrical lines, and there are as many possible variations of these procedures as there are lines to enact them.

How would a free-verse poem exert such control?

They taste good to her

This sentence from William Carlos Williams’s “To a Poor Old Woman” is, like the sentence “you said that,” flat on the page. Linguists might say that there is no indication of the tonic syllable in this intonational unit; neither is there a metrical pattern that might be varied in order to throw emphasis on any particular syllable. Every poem chooses 
either explicitly or implicitly to do something at the expense of something else, however, and having relinquished the power of meter 
in this free-verse poem, Williams capitalizes on the power of line as a way to control the intonation of his syntax.

munching a plum on

the street a paper bag

of them in her hand

They taste good to her

They taste good

to her. They taste

good to her

You can see it by

the way she gives herself

to the one half

sucked out in her hand

— From To a Poor Old Woman

Following the first stanza’s run-on syntax and enjambed lines, the second stanza’s first line (“They taste good to her”) feels whole and complete, a moment of stability. Subsequent lines animate this sentence by reintroducing enjambment to the poem’s lineation, asking us first to hear the sentence as “they taste good to her” and then as “they taste good to her.” Each of these shifts in the placement of the tonic syllable adjusts the meaning of the sentence, but ultimately more provocative than the individual adjustments is the sequence of adjustments, which makes the poem feel purposefully thoughtful: it suggests that no one way of hearing the sentence will do complete justice to the rich experience of savoring the plum. “You can see it,” says the poem’s next line, but actually we’ve heard it, not seen it, the poem’s lineation creating a sequence of rhythms that in turn generate what we call the poem’s tone — the quietly rapt attentiveness it brings to even the most ordinary human pleasure.

Consider the lineation of Ellen Bryant Voigt’s free-verse poem “Sleep.” This is not the way Voigt lineates her syntax:

I flung myself into the car

I drove like a fiend to the nearest store

I asked unthinking for unfiltered Luckies

oh brand of my girlhood

I paid the price

I took my prize to the car

I slit the cellophane

I tapped out one perfect white cylinder

I brought to my face the smell of the barns

This, in contrast, is how Voigt actually lineates her syntax:

                             I flung myself into the car I drove like a fiend

to the nearest store I asked unthinking for unfiltered Luckies oh

brand of my girlhood I paid the price I took my prize to the car 
I slit

the cellophane I tapped out one perfect white cylinder I brought to my face

the smell of the barns the fires cooking it golden brown smell of my father

my uncles my grandfather’s tin of  loose tobacco his packet of delicate paper

the deliberate way he rolled and licked and tapped and lit and drew in

and relished it the smell of the wild girls behind the gym the boys

in pickup trucks I sat in my car as the other cars crept by

I looked like a pervert it was perverse

a Lucky under my nose

What precisely do we mean if we say that these two lineations of the poem’s syntax have very different tones? Why, given that the words are in each case identical, does the second arrangement sound fiercely self-aware, rueful yet comical, while the first sounds self-satisfied?

Most of Voigt’s clauses are linked paratactically, laid side-by-side without any subordination, and most of the clauses are simply declarative. My imposed lineation preserves the integrity of these clauses, resulting in a sequence of lines beginning with the first-person pronoun and a monosyllabic verb (I flung, I drove, I asked, I paid, I took, I slit, I tapped, I brought), and this repetition encourages us to hear the tonic syllable in the same position in each line, 
producing a consistent rhythm: “I flung myself into the car, I drove like a fiend, I asked unthinking.” As a result, the poem seems like a record of considered thought (I came, I saw, I conquered), rather than a volatile act of thinking.

In contrast, Voigt’s actual lineation of “Sleep” avoids any consistent alignment of syntax and line (“I slit / the cellophane”), violating the strong grammatical integrity of the clauses with enjambment and thereby disrupting the regular rhythm of the parallel syntax. At the same time, her lineation emphasizes the weaker grammatical links between the clauses by running them together (“I looked like a pervert it was perverse”), thereby creating variable rhythms within the lines.

It is not the lack of punctuation alone that produces this rhythmic vitality, for whether punctuated or not — 

                              I flung myself into the car, I drove like a fiend

to the nearest store, I asked unthinking for unfiltered Luckies, oh

brand of my girlhood, I paid the price, I took my prize to the car, 
I slit

the cellophane, I tapped out one perfect white cylinder

— each of Voigt’s brief clauses is easily understood, and the lack of punctuation accentuates the work that the syntax is already doing in relationship to the lines: the poem refuses to allow us to experience syntactical closure at the same point at which we experience linear closure, and these disjunctions between brief clauses and long lines deflate the already weak organizational power of the parataxis. The poem sounds like the work not of the reliable witness but of the raconteur, someone who doesn’t realize “it was perverse” until the words “I looked like a pervert” provoke the insight. The tone is not predictably consistent, as it is in my relineated version of the poem, but rather consistently shifting, as it is in Williams’s little poem.

Once, lyric poems were songs: “the words must of course determine the music,” said Plato of what we’ve come to call lyric poetry. But Plato himself never used the word lyric; he called such poems melic. The word lyric was not employed until after the third century B.C.E., when the scholars of the Alexandrian library began collecting and organizing the remaining poems of ancient Greek poets such as Sappho and Anacreon, and by this time, the musical settings of these poems had been lost. As the classicist Glenn W. Most reminds us, the word lyric was coined to refer not to the experience of a sung poem but to the experience of a poem on the page. The music of poetry could not be other than a metaphor.

But the metaphorical status of the words music and tone does not render them useless to the discussion of lyric poems, unless they’re deployed so generally that we postulate an implausible coincidence of the artistic media from which “Lycidas” and a Chopin nocturne are made. In order to describe or produce a poem’s tone, a poet can’t think profitably about tone; a poet must think about the linguistic elements that produce a poem’s tone, most prominently rhythm and echo, just as a painter who wants to achieve a certain quality of light must think more precisely about the nature of paint. The medium of the language is our given, but the variety of poems that might be made from it is an unpredictable gift. “The movement of poetry,” said the author of “In a Station of the Metro,” “is limited only by the nature of syllables.”