The HyperTexts

Poetic Meter in English: Roots and Possibilities by Richard Moore

Introductory Note:  The following essay is an attempt to provide a background to the meter controversy which has appeared and reappeared in American poetry and poetics since Whitman's day and which has come into fresh prominence with the emergence of the so-called "New Formalists" in recent years. Such a background—that is, such a redefinition of traditional meter—is necessary, I believe, because considerable uncertainty has developed in the decades of free verse ascendancy about what meter in English was, is, and can be. Indeed, one of my conclusions is that the narrowing and rigidification of English meter, particularly in America, has been intimately associated with its periodic abandonment in its New World surroundings. As with other kinds of custom, a simplification, a dogmatism, and a lack of flexibility in traditional meter have been signs of weakness rather than strength and have been the forerunners of its total rejection. In such a situation, a deeper understanding of one's roots becomes crucial, so that a more firmly based line of growth can be found.

The American experience has developed from many cultures. From the beginning, alien presences modified the English settlers and helped them become American; and this process has become profounder and more various through the years. But poetry, after all, is a language art; and even if our English heritage is an evil genie for some of us, it must be understood if it is to be properly exorcised. And there is no aspect of poetry more deeply dependent on the language in which it is composed than its rhythms. I have therefore restricted myself almost entirely to a consideration of English literary history because that is what seems relevant to the topic under discussion. I talk about our English past in order to understand our American present.

Poetic Meter in English: Roots and Possibilities

It may be that one's attitude toward meter will often come down to one's attitude to poetry's monuments—to Milton, say, or Shakespeare. If you tend to think of those poets as meaningless excrescences of a past essentially boring or monstrous and the sooner forgotten, the better, then your verse will likely be the freest of the free, insouciant of the suffocating rules they seem to imagine helped them to harmony and life. But if you see them as poets who fashioned still relevant masterpieces, and as climbers to heights unlikely to be scaled again in any foreseeable future, then you will be inclined to study them in the hope of gaining some understanding of their secrets; you will see them and their age as possible repositories of a lost wisdom, detectable even in the music of their verse.

It is important, I think, to be aware of this historical dimension to the meter controversy. Some reference to it is implicit in the two frequent replies to the proposition, Poets should return to regular meter. "Did we ever leave it, then?" those for whom the past is a living presence will ask; and, "What is there to return to?" the skeptics will enquire. Indeed, beginning with this last question, it doesn't take much discussion of regular meter with various contemporary practitioners or an extensive scanning of their lines to reach the conclusion that there is only the vaguest and roughest commonly held notion of what regular iambic meter is. Perhaps there is, in fact, nothing to return to. Is—was—regular meter, then, so complicated? After all, don't we all know what an iamb is?

In fact, the first rule about iambic verse appears to be that just iambs are not enough. When Marlowe's Doctor Faustus with his devilish magic summons the apparition of Helen of Troy from the shades, he greets her with the famous line,

            Was this the face that launched a thousand ships . . .

Great, certainly memorable, but—when I went to school, at least—a textbook example of how not to write iambic. Those five absolutely even and identical stresses are just too monotonous—so runs the argument, begging questions at every breath. If monotony—which is, after all, little more than a pejorative word for regularity—is undesirable, then meter itself, which is regular and therefore monotonous, is to be avoided, and we enlightened moderns can sing psalms of gratitude that after all those benighted centuries, we are free of the curse at last. The problem is real: who decides or what determines what is too regular, what is sufficient variety, and what is too much variety—variety unto chaos? Some lovers of Wagner's music find Mozart's tediously regular; and the young Keats judged the English Augustans of Mozart's century similarly:

            with a puling infant's force
            They sway'd about upon a rocking horse,
            And thought it Pegasus.

—though he evidently revised this judgment later, when he wrote Lamia in couplets modeled on Dryden's.

And the implication that Marlowe was a poor incompetent forerunner of Shakespeare is inaccurate. He was a marvelous poet, the man who first demonstrated that powerful dramatic poetry could be written in the new blank verse and so, perhaps more then any other, was the founder of Elizabethan drama, and even he at the very beginning, when blank verse was at its most regular, knew that the pure iambs of the line quoted were not enough. The lines that follow in the same speech demonstrate his understanding of the medium:

            And burnt the topless towers of Ilium?
            Sweet Helen, make me immortal with a kiss.     [Kisses her]
            Her lips suck forth my soul; see where it flies!

And a little further on, most memorably,

            O, thou art fairer than the evening air
            Clad in the beauty of a thousand stars;
            Brighter art thou than flaming Jupiter
            When he appeared to hapless Semele;
            More lovely than the monarch of the sky . . .

Even in this brief typical quotation we may see clearly why that opening "thousand ships" line with five full iambs sounds unnatural to us: iambic pentameter doesn't have five full stresses; it only has four—and sometimes for another kind of special effect, as in the last line here, only three. In almost every one of Marlowe's "mighty lines," one of the five accents called for by the metric pattern is weaker than the others and tends to disappear in recitation—as the second accent in "Ilium" in the first line above, "with" in the line following, and, in the next group, "than," "of," and the second accents in "Jupiter" and "Semele." The line, "Her lips suck forth my soul; see where it flies!" would seem to be another five stress exception. The line is one of the most crucial in the play. Faustus has just sealed his damnation by having carnal relations with an apparition from Hell, and the meter makes us feel the event. The repeated "s" sound forces a pause between "lips" and "suck" which, in turn, makes us feel "suck forth my soul" as one agonizingly long beat, recording the precise moment of Faustus' doom.

Thus, even apparent exceptions are often most meaningfully seen as variations, not of the regular iambic pattern, but of the equally regular, simultaneously occurring line of four main accents which divides typically into two two-beat half lines separated by a slight pause or caesura. The persistence of this four-beat accentual pattern from the alliterative line of Old English has often been noted, but its continuing significance in the actual practice of the great English poets has drawn less attention. That iambic pentameter verse in English is not a single system but the elaborate and constantly varying enmeshment of two separate systems of prosody, each enriching the other, goes a long way toward explaining its extraordinary persistence in our language—in contrast, for example, to the comparatively minor role it has played in German, the major European language whose rhythms most closely resemble ours.

In English, as in other stress-accent languages, the accents occur most frequently on every other—sometimes on every third—or, more rarely still, every fourth syllable; and when rhyme as a structural device in verse made its European appearance in the early Middle Ages, more-or-less systematic accentual-syllabic lines appeared along with it. One senses in Chaucer's metric his effort—aided, no doubt, by "feedback" from his court circle—to find rhythms in his native English which would ally it to the rest of European literature. In his early "French period," we find him beginning with a four-beat, generally eight-syllable line; but later, having become familiar with Boccaccio and Dante, he changes to the longer line of his "mature" poems. One wants to call it iambic pentameter—which indeed it is much of the time—but there are important differences. Take the first line of The Canterbury Prologue, for example:

            Whan that Aprill(e) with his shoures soote.

When the nineteenth century discovery of syllabic final e's in Chaucer's language finally made metric sense out of his lines, there was a scholarly tendency also to credit some mere scribal curlicues with syllabic value and make him too regular. Thus, if "Aprill(e)" above is three syllables accented on the second, the line works out nicely as an iambic pentameter with an opening trochee and a feminine ending. But later scholars have decided that that particular e was only a curlicue. "Aprill," therefore, is a trochee, and the first line of The Canterbury Tales is, strictly speaking, in trochaic pentameter. Some save appearances by calling such lines "headless" iambic pentameters. What, if anything, Chaucer called them has not been recorded.

Such uncertainties, together with a lackluster century of poetic and linguistic chaos during the Wars of the Roses following Chaucer's death, left little in the way of detailed metrical procedure for the Tudor and early Elizabethan poets to go on; and in the 1560's there was ferment on the subject. There seems to have been general agreement with the view, first expressed in Roger Ascham's Schoolmaster, that English poetry would have to imitate that of classical antiquity, reinstate the ancient system of quantitative measures, and hopefully dispense with "barbarous rhyme." But there was debate about whether the ancient rules for short [u] and long [—] syllables should be literally applied or whether a long syllable in English could be taken to mean an accented syllable. This latter system won acceptance as more natural and produced results roughly similar to Chaucer's, as rediscovered three centuries later. But the classical terminology made possible a more precise understanding of effects and led to a marvelous variety and individuality in the following centuries.

The system is simple, and the almost naive way that it is based on ancient prosody is an important aspect of it. The line used in the spoken part of Greek tragedy is the iambic trimeter—which has six iambs. The Greeks thought of iambs in groups of two [u—/u—] —the "dipody." (The reason for this, apparently, is that in the analogy of walking, a long syllable [—] stands for a step and is equivalent to two short syllables [u u]. This is why in epic meter a dactyl [— u u] can be freely replaced by a spondee [— —]. In iambic, therefore, it takes two iambs for the steps to "come out even.")

In Greek, the first iamb in the dipody may be a spondee, but not the second (which suggests that in iambic, as opposed to dactylic, the substitution of a spondee is felt to be a break in the rhythm, an actual syncopation—as the walking analogy would also imply). The rule is a specific answer to a question which always comes up in rhythmical schemes: how much or for how long can the rhythm be altered without being lost? The sixteenth century founders of English prosody, judged by their practice and that of the poets who followed, had a rule clearly modeled on the Greek: in English iambic, the accent in any iamb may be moved or removed if the accent in the following iamb remains in place. Thus, spondees may be substituted anywhere; and, in addition, u— u— may become —u u— (substitution of a trochee [— u] for an iamb), u u u— (substitution of a pyrrhic [u u] for an iamb; but in this case the second of the light syllables tends to receive a light accent because of the prevailing rhythm), or u u— — (a "pyrrhic spondee"). This last is rare in Marlowe and in "conservative" iambic generally, but Doctor Faustus, xv, 119,

            Yet, for Christ's sake, whose blood hath ransomed me,

would seem to qualify.

One cannot always be absolutely sure of a particular scansion in this type of accentual verse, since there are often choices about placement of accent —and even the number of syllables—that have to be made by the performer (for whom the meter itself sometimes serves, or should serve, as a guide).

The variations possible in this simple system are endless, and to note them in exhaustive detail would be unrewarding. If I have not covered all conceivable instances, let me quote my Indian cookbook: "These recipes are not immutable formulae, but invitations to improvise." An undue fussiness on the part of official metrists may be one of the reasons for meter having fallen into disuse. The only other writers I know to have mentioned the rule "that two successive accents cannot be suppressed or displaced without destroying the underlying pattern" (as they phrase it) were W. H. Auden and Norman Holmes Pearson in the introduction to Volume I of their anthology, Poets of the English Language. They give two examples of lines that will pass as iambic pentameters:

            I want to be a genuine success

and

            Give me your hand; promise you'll still be true

and two that for corresponding reasons will not:

            I want to be in an excited state

and

            Lay your knife and your fork across your plate.

But there is another matter about which Auden and Pearson seem a little unclear: "Single dactyls [— u u] and anapests [u u —] often appear through an inversion of an iamb or a trochee, but as a metrical base they have played only a minor role." If dactyls and anapests occur in iambic verse only by trochaic substitutions [— u u —], then in fact they do not occur at all since they are metrical terms and metrically the pattern, — u u —, in this type of verse is a trochee followed by an iamb. But this raises the question whether anapests occur through the simple addition of extra light syllables—as in the line, "Sweet Helen, make me immortal with a kiss." Are the syllables "me immor-" an anapest or are the open vowels "me im-" elided, as they would be in French or Italian? It is perilous to use the word "never" in matters of this kind, but in the bulk of the drama, all of Paradise Lost (except for Raphael's description of The Creation, where Milton departs from iambic in order to use Biblical phrasing), and all of the Augustan period, it is, I believe, impossible to find an anapest that cannot be made into an iamb by elision. These ghostly anapestic presences are a wonderfully subtle source of rhythmic excitement throughout this period. The inclusion of real anapests would have spoiled the game.

The very strictness of the system imposed by the classically derived rules served to enhance the individualities of the poets using it. Milton, who, I think, metrically owes more to Marlowe than to Shakespeare, clearly frowned on Marlowe's tendency to place the weak accent last in the line: that's part of Milton's particular music, together with the avoidance of feminine line endings, and, as indicated by their metric contexts, a specifically Miltonic pronunciation of certain words. (My personal favorite is "spirit" consistently used as a monosyllable. How did Milton say it?—with a kind of slow drawl—"spir't"—or did it have a clipped, Scottish rolled r—"sp'rit"? A contemporary remarked that Milton pronounced his r's somewhat harshly, so it must have been the latter—and I can't imagine Milton drawling anything.)

But individual reactions to the metrical system not only distinguish different poets from each other; they also separate stages—incarnations, if you will—of the same poet in cases where the career is extended and complex. Every one of Milton's major poems has its individual metric within the system, and it is well-known that one can determine the order of Shakespeare's plays with considerable accuracy by tuning in on the evolution of his blank verse. The sense of the four-stress alternate system underlying the precise syllabic rules is important in making such distinctions. As Shakespeare's development proceeds, the phrasal rhythms of the older, more native line seem to assert themselves more clearly in an ever more lilting counterpoint peculiar to him and immediately recognizable. That the old four-beat line played a conscious role in Milton's thinking is beautifully evident in Paradise Lost where one, and only one, stress is weaker than the others in almost every line. It has been remarked that Milton carefully varies the position of the caesura from line to line; but he varies with the same apparent deliberation the position of the weak accent among the first four. (The fifth, as I mentioned, is always strong.) This regularity makes the departures from it more powerfully expressive. When Eve, for example, tells the Serpent about God's command, she betrays her simplicity and uncertainty in a sing-song line of monosyllables with only three strong accents:

            But of this Tree we may not taste nor touch; (IX, 651)

but when she wants to register anger in her arguments with Adam, she can cram the line with five full stresses as well as Faustus:

            Nay, didst permit, approve, and fair dismiss. (IX, 1159)

And finally, before I apply these observations to the present situation, let me give a revealing example from a lesser-known writer. John Webster, the author of two excellent tragedies and a comedy, all clearly under Shakespeare's influence, was also—quite on his own—a brilliant metrist who showed some bold ways to deal with the traditional line. In The Duchess of Malfi, the increasingly maddened Duke Ferdinand, who, we suspect, has an incestuous passion for his sister and who orders her death out of apparent mad jealousy, says upon seeing the corpse,

            Cover her face; mine eyes dazzle: she died young.

—a regular iambic pentameter with a feminine ending—except that the final light syllable is heavy, giving the line an air of metrical chaos at the end, perhaps reflecting the chaos in the Duke's own mind. The violent trochaic substitution, "dazzle," suggests that the actor register something close to a scream, and the lame, emotionally alienated "she died young" suggests a near mumble—without, however, losing sight of the Duke's lust.

A few lines later, there are other remarkable effects, when Ferdinand says to the hired murderer,

                                                    Let me see her face
            Again. Why didst not thou pity her? What
            An excellent honest man mightst thou have been . . .

The spondee-spondee-trochee in the middle of the second line again suggests to the performer (or the involved reader) mounting anguish and rage, leading up to an extremely heavy accent on "pity;" and the seemingly awkward "What" as the final stress in the line has a fine expressiveness as a transition to the nasty deranged irony of the third line ("excellent," of course, scans as two syllables to avoid the anapest.)

During the later seventeenth and the eighteenth century there was a general simplification and regularizing of the system in harmony with the neoclassical aims of achieving greater clarity and elegance. Variations as extreme as Webster's were excluded and, as a result, the verse—at its worst, if not at its best—opened itself to Keats' accusation that it resembled "a puling infant" on a "rocking horse." But Keats himself—like his contemporaries in this—did little to revive the old Baroque variety, and basic iambic became, if anything, more regular during the nineteenth century. As the decades wore on, it began to seem that relief from mechanical rhythms lay in "interesting new metrical systems" entirely—as in the gradually increasing use of anapestic rhythms, old ballad and other "popular" meters, and the far-reaching innovations of Coleridge in "Christabel" and later those of Hopkins, Swinburne, Whitman, Dickinson, and Hardy.

A description of the modern situation ought to begin, I think, with Yeats' metrical experiments and in particular his development of the Renaissance-Baroque pentameter, which has had few imitators but represents most powerfully the continuing presence of traditional meter.

In his volume of 1904, among "Celtic Twilight" poems in very regular lyrical Pre-Raphaelite rhythms, enlivened only by a free use of anapests derived from ballad meters, the solitary poem "Adam's Curse" stands out starkly in its illusion of spoken language, its use of the pentameter couplet (in another return, like that of Keats, to the Augustans), and its reliance on strict iambic rhythms in a way which is new to Yeats and, I believe, new to poetry in English. Here is the opening verse paragraph:

            We sat together at one summer's end,
            That beautiful mild woman, your close friend,
            and you and I, and talked of poetry.
            I said, 'A line will take us hours maybe;
            Yet if it does not seem a moment's thought,
            Our stitching and unstitching has been naught.
            Better go down upon your marrow-bones
            And scrub a kitchen pavement, or break stones
            Like an old pauper, in all kinds of weather;
            For to articulate sweet sounds together
            Is to work harder than all these, and yet
            Be thought an idler by the noisy set
            Of bankers, schoolmasters, and clergymen
            The martyrs call the world.'

From the Renaissance beginning, there has been an ambiguity about the basic attitude of the poet and his audience toward metric substitution: exactly how free was it to be? In the hexameter line of the Greek and Roman epic, as I have remarked, spondees are substituted for dactyls in the first four feet with perfect freedom. The only reason for calling the meter dactylic rather then spondaic, apparently, is that the fifth foot in Greek epic hexameters is customarily a dactyl. (In Latin the rule is obligatory; that is evidently because Latin runs so persistently to spondees that the dactylic presence tends to get lost.)

Substitution in English iambic, as defined by the rules, never achieved this freedom. Trochees, for example, were fine in the first foot, acceptable in the second, third, and fourth if they followed a caesura, but almost any poet much before or much after Webster would have found his "dazzle" in the line quoted above awkward and unmusical. Similarly one comes to feel that a pyrrhic-spondee ought perhaps be alternatively pronounceable (with something of a tuneful lilt) as two iambs—as is clearly the case in the third and fourth feet of Shakespeare's line,

            Thou art more lovely and more temperate.

Perhaps pyrrhic-spondees are like anapests: perhaps there really aren't any.

But in "Adam's Curse," all this is fundamentally different. In the poem's opening line, the third and fourth feet are an unambiguous pyrrhic-spondee. (There is no other way to say them.) At the same time, one doesn't feel them as an exception because they help establish the feeling of easy colloquial speech. Any doubts about this are dispelled in the second line—a single iamb followed by two pyrrhic-spondees in the same easy, casual tone: a classically correct iambic pentameter with only one iamb (there are others in the poem). It's difficult to imagine what Dryden, Pope, or Samuel Johnson would have thought of this; but I suspect that Homer would have been delighted. "Don't let me see you," his Agamemnon says to the old priest, "either [in spondees] hanging around here now or [in dactyls] coming back again later."

But Yeats is his own best illustration here, where almost every line quoted displays some inspired touch. Placing "maybe" in rhyme with "poetry," forcing it to become a spondee, suggests the difficulty of the action described, like the almost tongue-twisting "stitching and unstitching." The rapid light syllables, "Yet if it does not," with just enough of an accent on "does" to adhere to the rule, give a virtuoso demonstration of the ease that poetry should seem to have, as does the similarly constructed, marvelously melodious line, "For to articulate sweet sounds together." The reader can find other felicities—and not to miss the brilliant line,

            Of bankers, schoolmasters, and clergymen,

where a trochee, "-masters," makes its flippant appearance, not just without a caesura, but in the middle of a word.

(But I am wrong to imply that Yeats invented all this. There is exactly the same device in a well-known poem of his fellow Protestant Irishman, Jonathan Swift:

            And could he be indeed so old
            As by the news-papers we're told?)

In a way there is something about all this that can strike a reader as almost decadent. One can think of Yeats as using his fine wit and superb craftsmanship to turn the whole metric system upside down—to corrupt its values from within, as it were. In any case, he himself never wrote another poem with exactly this casual freedom within the rules of the English pentameter. It remains a continuing presence, however, in the beautifully varied iambic of his later poetry. And it also represents a continuing presence, I would like to think, for contemporary poets.

It was certainly a presence and a challenge for Robert Frost, sojourning in England and attending Yeats' soirees. Frost, who apparently never related comfortably to his betters or his equals, seems to have been scandalized by Yeats' social and superstitious oddities, but the lessons, particularly the metric lessons, he learned from Yeats are evident in his poems. "'Out, Out—'" in his 1916 volume is a worthy metrical successor to "Adam's Curse." It opens:

            The buzz saw snarled and rattled in the yard
            And made dust and dropped stove-length sticks of wood,
            Sweet-scented stuff when the breeze drew across it.
            And from there those that lifted eyes could count
            Five mountain ranges one behind the other
            Under the sunset far into Vermont.
            And the saw snarled and rattled, snarled and rattled,
            As it ran light, or had to bear a load.
            And nothing happened: day was all but done.

Students have to be told that the "buzz saw"—the principal character in this horrifying poem—was not a chain saw, which hadn't been invented in 1916. Frost is clearly describing one of those large circular saws run by a belt from a tractor. They are fiendishly dangerous because they are usually rickety, have no guards or fences to protect the body, and because they cannot be immediately turned off when something goes wrong. It is the perfect symbol for Frost's sense of the brutality that coexists with the beauty on this particular corner of earth, and meter plays a large role in bringing it to life.

The first pentameter, like the line in Homer where Agamemnon threatens the priest, is in two rhythms: spondaic for the saw when it is cutting wood—"The buzz saw snarled"—and lightly tripping for when it is running free—"and rattled in the yard." The second line (iamb-trochee-spondee-spondee-iamb) is musically brilliant in its thumping clumsiness. (The wood falling to the ground is "stove-length:" short pieces as opposed to the longer fireplace length.) But the most expressive effect in this passage is the way in which the first three halting end-stopped lines which describe the work, contrast with the easy run-on rhythms of the next three, which open the view into the countryside and its excitement of color and depth at sunset. The whole theme and effect of the poem is prefigured in this simultaneous juxtaposition of rhythm and imagery. Only "those that lifted eyes" could see the beauty; but to lift one's eyes, to let one's attention wander for the smallest instant from the work at hand, is the very thing that one must never do when operating such machinery. This one subtle concrete detail says more forcefully and memorably than a whole choir of Rilke's angels that beauty is dangerous. How many times does one have to read the poem, fascinated by its rhythms, before one realizes that the boy loses his hand and his life because he looked up? When the saw and its rhythms returns, it has already become a thing of terror. (And don't overlook "As it ran light, or had to bear a load"—another line in contrasting rhythms.)

With examples like these to inspire, it is surprising that more recent productions in regular iambic have so frequently tended to sound like speeches from Gorboduc, that first blank verse play in English and inexhaustible storehouse of sterile pentameters:

            O king, the greatest grief that ever prince did hear,
            That ever woeful messenger did tell,
            That ever wretched land hath seen before,
            I bring to you: Porrex, your younger son,
            With sudden force invaded hath the land
            That you to Ferrex did allot to rule,
            And with his own most bloody hand he hath
            His brother slain, and doth possess his realm.

If the contemporary effort to write strict iambic has so frequently resulted in rhythms that sound like that, then the possibility should at least be considered that after four centuries strict iambic is indeed dead and ought to be replaced by something else. (The problem in part may be that the lines of the iambic / free verse controversy were first drawn in Whitman's time, when iambic had already lost much of its early music. In consequence, the verse of the metric conservatives, even to this day, partakes of a tradition, starting with Longfellow and Colonel Higginson, which, like A. E. Housman at his worst, valued excessive regularity and suggested to poets like William Carlos Williams the stultifying proprieties of Victorian times.)

What, then, might be the replacement? Or what thoroughgoing modification might suffice? One change, already mentioned, made with increasing persistence and seriousness since early Romantic times, has had great effect and is quite simple: add anapests. Frost was fond of remarking that there are "virtually but two" meters in English: "strict iambic and loose iambic." By the last he meant iambic with anapests: when a student at a writer's conference asked him what strict anapestic would be (which has been written in English, usually as light verse, at least as early as Matthew Prior at the beginning of the eighteenth century), "that," he replied solemnly, would be "strictly loose iambic."

A problem for all metrical innovation is that of recognition. The reader encounters the new metric with conventionalized expectations and tends consciously or unconsciously to fit new meters into old ones. Frost himself, aside from trimeters, seldom used his loose iambic. A well-known exception to that statement, "Mowing" in his first book, shows why. Of the fourteen pentameters, eleven have only one anapest or none at all. Even so, the lines sound long, largely because the extra syllables tend to result in five full stresses. They sound, therefore, despite their variety in syllable count, a little like Marlowe's "thousand ships" line. Allowing a few anapests has had the paradoxical effect of making the meter sound more regular.

This suggests that anapests might work better in a four stress line, as Coleridge tried in "Christabel" and Frost in the rather obscure and strangely flat narrative poem "The Discovery of the Madeiras." The problem here, I think, is that the lines tend to oscillate in the reader's understanding between the familiar lyrical / satiric octosyllabic line with no anapests on the one hand and, with too many anapests, the line of "Twas the Night Before Christmas" on the other. It's very difficult for a poet—any poet—to stay in control in a situation like that.

But the effect is entirely different when anapests are added to a trimeter iambic line. Yeats did this with extraordinary results in an abab quatrain having muted rhymes—most notably in the great poem, "Easter 1916." The line creates an impression much like Swift's octosyllabic: easy, colloquial, with potentialities both for the lyrical and the burlesque but without the octosyllabic's ever-present tendency to sound sing-songy and mechanical. Frost took up this meter in couplets with burlesque rhymes to great effect in superb comic poems like "Departmental" and "A Drumlin Woodchuck." The reason for the success of this line, I think, is that it exists already in the reader's verse experience as the three long lines of a limerick. Frost's poems in this meter with their rollicking rhymes are brilliant disproofs of Eliot's well-known statement that pronounced meter and colloquial speech are at odds with each other.

In order to find a long, epic or "serious" line to replace the pentameter, we have to make more far-reaching changes based on the underlying four stress pattern in the pentameter itself. The "Death by Water" section of The Waste land points the way:

            Phlebas the Phoenician, a fortnight dead,
            Forgot the cry of gulls, and the deep sea swell
            And the profit and loss.
                                                        A current under sea
            Picked his bones in whispers. As he rose and fell
            He passed the stages of his age and youth
            Entering the whirlpool.
                                                        Gentile or Jew
            O you who turn the wheel and look to windward,
            Consider Phlebas, who was once handsome and tall as you.

Clearly the persisting pattern here is the accentual four-stress line from Old English, falling into two half lines divided by a caesura. The first line even has the Old English alliteration on the first and third stressed syllables. More important, it unambiguously establishes with the first three words that three light syllables between stresses are to be a regular alternative. This prepares the way for the second line, where "Forgot the cry of gulls" is not three iambs but two stresses, "-got" and "gulls," again reinforced by the alliteration, in which "the cry of gulls" is felt as another "augmented" anapest. Lines that scan in the old system as perfectly regular pentameters can occur in this looser, more various, four-stress scheme—as, for instance, the fifth and seventh lines here—but they are not felt as pentameters any more. Or, if one wills, they are felt as pentameters in the four-stress way that good pentameters have always been felt. It is noteworthy that this pivotal and metrically important passage is a translation from the close of Eliot's French poem, "Dans le Restaurant," where the lines suggest the four-accented, strongly caesuraed line of French drama. Who would have thought that the way from modern English back to its Old English roots lay through the French classical alexandrine?

Much of the twentieth century "free verse" or "loosely cadenced verse" in English follows these principles, more or less knowledgeably, either with four, three, or two stresses. There are disadvantages in that much of the subtlety and fine tuning of the old system have been lost. Perhaps they have been lost in any case; and there are compensations in the greater freedom for a skilled poet to provide sound effects of his or her own. Unfortunately the most frequent result has been a kind of non-verse forever lapsing into prose and forever forcing itself to sound like poetry by distorting its syntax and making itself otherwise incomprehensible to the general reader. Yet superb poetry has been written in this scheme, and it may well be the most viable metric available to contemporary poets.

Finally there is one more alternative to consider: the introduction from other languages of new verse patterns that may be adapted to the peculiarities of English. The most frequent source today, as in the Renaissance, has been the fine array of poetic meters in classical antiquity. Such meters have been a tradition and an accepted procedure in German since the eighteenth century, and Rilke has carried it on in modern times. In German, as in English, the only workable way to proceed is to translate quantity into stress and substitute modern accented for ancient long syllables. In both languages this makes the pattern slower and heavier than it almost certainly was in the now imperfectly understood originals. The experimenters with such meters in English—Longfellow most visibly, and Auden—have gone along with the German practice in not trying to bring all the ancient long into modern accented syllables. Spondees are very common in ancient metrical schemes and practically impossible to come by in German. This may not be so in English, however. Consider, for example, this stanza from Swinburne's "Sapphics" [The first three lines of the sapphic stanza each scan as follows: trochee-spondee-dactyl-trochee-trochee (or spondee) and the fourth as: dactyl-trochee (or spondee)]:

            By the gray seaside, unassuaged, unheard of,
            Unbeloved, unseen in the ebb of twilight,
            Ghosts of outcast women return lamenting,
            Purged not in Lethe.

I have picked what struck me as the most regular of the twenty stanzas. The others are not as consistent about keeping the three consecutive accented syllables called for by the meter. And in the first two lines quoted it takes a little pushing to read "By the" and "Unbe-" as trochees. One of the advantages of using such detailed and precise metric patterns, when they are strictly adhered to, is that the pattern itself often resolves an ambiguity about how the words should be accented (as it instructs us in the fourth line here that "not" is to be an unaccented syllable). One feels that the language has been carefully choreographed. The major disadvantage is that, without any considerable tradition for writing verse in this way, the whole procedure comes to seem arbitrary and pedantic. Who cares, after all, whether this or that syllable fits some unfamiliar abstract pattern? At one time I was quite taken with the possibilities of writing elegiac couplets in English, but when the resulting poems appeared in magazines, the manner in which they were printed sometimes suggested strongly that the editors had understood them to be free verse.

Reading Swinburne's "Sapphics" makes the reader aware of how much Ezra Pound—the lyrical, celebratory Pound that seems to be the most enduring—owes to it. It is as though Pound realized what Swinburne had accomplished in this poem better than Swinburne did himself; but when Pound imitates "Sapphics" in that decisive poem for him, "The Return," the ancient metric pattern is only hinted at in a free verse context:

            See, they return, one, and by one,
            With fear, as half awakened;
            As if the snow should hesitate
            And murmur in the wind,
            and half turn back.

The ancient pattern is lost, but its cluster of three long syllables remains in alien surroundings as a haunting presence. Realizing the ingredients involved (or some of them, at least), one wonders, as so often with Pound, whether we have a marvelous new subtlety or an easy exploitive sloppiness —or both alternatives at once.

Maybe too, there really is some kind of magical power in the classical metrical patterns, even when they are reproduced in a more cumbersome medium. Maybe they affect the reader semiconsciously and induce editors to publish poems which they would be horrified to learn were in a regular meter. The situation is further complicated by poems, which a blurb writer will proclaim to be "expertly composed" in this or that ancient meter, when the poem itself displays no hint of it. Such a strong faith in the ignorance of the poetry-reading public does not bode well for the future.

There is, indeed, such a chaos and cacophony of voices and views in even our limited poetry world, that it is difficult to imagine any consensus about metrics emerging today, as one did in the sixteenth century. On the other hand, the very chaos is an invitation to virtuosos who can, and therefore must, try everything. This, in turn, becomes an invitation to empty virtuosity. And that, then, is balanced by the severest temptation of all: to write, without using any noticeable sound pattern, so memorably that the reader longs to remember the writing word-for-word. This struggle between serious intent and more-or-less blatant metric effects, between the important adult, if you will, and the frivolous child, has been going on for a long time. A friend who knows a vast store of odd rhymes for children and enjoys putting on a hillbilly accent said recently, turning a widely held critical view these days upside down, "If it don't sound like 'Hickory, dickory, dock,' it ain't poetry."

"Shakespeare and Milton don't sound like that," I said.

"O, they was so good, they didn't have to."

We may play and experiment as we will, but we had better not ignore, I think, the underlying four-stress pattern that has been working in English for more than a thousand years now. Even those lovely, lost quantitative meters from ancient Greece can feed into them, as Pound demonstrated. Above all, each of us had better make up his or her own mind about the matter—as Chaucer did in the linguistic uncertainty of his day, unequivocally choosing for himself the uncouth native dialect that had only been used at Court for a century or two. His friend, John Gower, was less certain and, like a good business man hedging his bets, wrote three major poems, one in each of the contending languages, English, French, and Latin. All three of them have been forgotten.

The HyperTexts