the Fallacy of Imitative Form Revisited
from Southwest Review, Volume 99, Number 4 2014
The rhythms of economic history show us that bull markets, even when they do not plunge headlong into their bearish counterparts, are subject to periodic corrections. These corrections, painful as they may feel in the short run, are a necessary part of the process in capitalist economies. Although the analogy will not please everyone, the rhythms of literary criticism often appear to move in comparable ways. With respect to poetry, interest in the formal patterns of verse has been enjoying a bullish stretch for perhaps a decade now, and it is time for a correction. In this case, what needs correction are some of the ways that readers of verse try to connect their observations about the formal patterns of that verse with what they think it is trying to say to them. The need for correction is, from one point of view, a happy one because it follows from a rediscovery of the formal patterns of verse by readers who twenty years ago might well have dismissed any concern with those patterns as trivial, frivolous, self-indulgent, bourgeois, or even pernicious, when that concern threatened to steal attention from more important matters, usually social or political. During this stretch of renewed attention to poetic form, the term "new formalism" has emerged to label a grab-bag of critical approaches and confuse itself with the New Formalism of some poets in the United States during the 1980s.
The correction I have in mind does involve a "should"—how could it not and be a correction—but it has only one "should" and it waives the right to pronounce on any others. There is no statement here about what poems should say or how they should say it; about whether "I" should be anchoring a poem or whether all anchors should be weighed in favor of decentered subjectivity, about whether poems should produce the illusion of coherence or whether they should incline toward indeterminacy and the thwarting of coherence; about whether they should engage politics explicitly or whether they should avoid politics altogether, to the extent that doing so is ever possible; about whether they should appear in lines of print or whether they should use new media to explode conventional habits of reading; whether they should be funny and playful or earnest and sincere; whether they should confess the pain within or witness the pain without. In fact, there is no statement here, and no assumption, that a reader of poems should pay attention to the formal patterns of verse at all. There is not even a statement or an assumption that reading verse is something everybody or anybody should do. As far as the argument here is concerned, those who write poems or read them may do so in their own ways, same as they eat or kiss.
What they may not do, or should not do, is to speak to others in lazy, inaccurate ways about what they have read or what they have written. The reason for this "should" is simple: there is enough lazy inaccuracy already, and we do not need any more of it. Furthermore, lazy inaccuracy misleads others, and misleading others, knowingly or unknowingly, is not acceptable, except perhaps in the case of a white lie or in the case of omitting information that cannot help people and could well hurt them, neither of which is relevant here. This last sentence directs itself to those who teach or who are being taught to teach, whether that teaching takes place in classrooms or in pieces of writing aimed at showing others something. The way you write poems or read them on your own time is your business; the way you teach people who are depending on you for right guidance is mine and everybody else's in the fellowship of literate people. When it comes to talking about poetry, the stakes of misleading and being misled are both very low and very high. They are very low in comparison with misleading or being misled about how to perform an emergency tracheotomy; when it comes to talking about poetry, laziness and inaccuracy are unlikely to kill anyone. But they are high precisely because relatively few people care about how to talk about poetry, placing a much greater burden on those few who do. The people you teach, whether in person or in writing, may never have the benefit of another teacher who can clean up the mess you have made in their minds. What is worse, they will pass your mess on.
Truth be told, "you" includes me, for I, too, have been guilty of making and passing along messes, both in the classroom and in print, and to the extent of that guilt, these remarks constitute a palinode. The imagined scene of mess-making goes something like this: Having said sharp and persuasive things about what they think particular poems are trying to say to them (as though poems only try to say things but cannot actually do so), and having noticed that there is a trochaic inversion in the fourth line of a normatively iambic poem, or a string of alliterating /s/ sounds, or a thumping series of accented monosyllables in a list, or a stretch of particularly gnarled enjambments after a smooth run of endstopped lines, or a novel twist in the paradigmatic pattern of a Petrarchan sonnet or a common-meter ballad or a sestina, people in need of the correction I hope to describe will then say something on the order of "And what is going on in the form of this poem reflects what is going on in the content." They may not use the humdrum words" form" and" content"; they may opt instead for "the performed or interiorized sound stream" and" the lexical realm of discourse and predication"; they may not use "reflects," opting instead for "represents," "imitates," "mimes," "mimics," "mirrors," "signifies," "echoes," "enacts," "dramatizes," "realizes," or "embodies."
But the differences in nomenclature do not alter the basic assumption that in talking to others about what particular poems are doing, the important move is to connect all the insightful things one has to say about the stuff that can be comprehended only by someone who understands the language in which the poem has been composed with the stuff that can be heard by anyone whose ears are open to the lapping of sound waves. (In the case of someone deaf or reading silently, the possibility of perceiving some interiorized sound shape in the mind's ear will depend on whether the silent reader knows enough to be able to convert the alphabetic signs of a language into phonemic equivalents.) In the phrasing Virginia Woolf used to describe the challenge faced by Lily Briscoe as she works on her painting throughout To the Lighthouse, the problem is how to connect the mass on the right with that on the left. In the case of talking to others about poems, one mass consists of feelings, thoughts, expressions, ideas, concepts, statements, arguments, questions, commands, appeals, hymns, prayers, instructions, descriptions, monologues, dialogues, flatteries, satires, complaints, laments, elegies, memorials, charms, invitations, seductions, riddles, curses, invectives, jeremiads, valedictions, narratives, chronicles, dramas, catalogues, addresses, interjections, expletives; the other of sounds and, where writing or printing is involved, visible features of format, some of which may have implied auditory consequences, such as italic font, some of which may not, such as capital letters at the beginnings of lines, some of which are ambiguous, such as not-so-radical enjambments in verse in which enjambment is the norm.
Of course we want to connect these two masses, those of us who talk to others about poems or who listen to others talking about poems or who read their writing about poems, especially now, when it is possible and acceptable to attend both to big things in the world that poems partake of and to little things poets do when they arrange the sounds and sights of language. It is not the desire to connect the two masses that needs correction; it is the way of doing it. For too many of us the default setting is to think and say that the mass consisting of formal patterns is the shadow and servant of the mass consisting of the paraphrasable gist. But most of the time it is not, and when it is, we are in the presence of notable, often celebrated, exceptions.
There are important precedents in Western poetics for treating form as the lackey of content, or if not the lackey, then the deferential and subordinate helpmate. A full genealogy would have to include Plato's Socratic frettings, in books two, three, and ten of the Republic, about the degraded status and potentially nasty side effects of mimesis; Aristotle's rehabilitation of mimesis in chapter four of the Poetics as one of the two pleasure-giving instincts from which poetry springs, the other being the instinct for harmony and rhythm; Horace's advocacy for verisimilitude in Ars Poetica; and Boileau's imitation of Horace in L'Art Poetique (1674), translated into English by Dryden and imitated in turn by Pope in Essay on Criticism (written 1709; published 1711) . Pope's essay gave us the handy, aphoristic formulation that continues to exert such pervasive influence, whether consciously recognized and embraced or unconsciously absorbed and transmitted: "'Tis not enough no harshness gives offence, / The sound must seem an Echo to the sense." The second half of the couplet provided Laurence Perrine with the title of his widely used textbook Sound and Sense: An Introduction to Poetry, first published in 1956 and now available, having been edited by others, in a fourteenth edition, which still sells briskly on Amazon.com, even for $124.39. Selling almost as well, for only $56.81, is Paul Fussell's Poetic Meter and Poetic Form (1967; rev. 1979), which in the final paragraph of the chapter devoted to free verse delivers the preemptive dictum, wholly appropriate for a rationalist student of eighteenth-century prosody (subject of his doctoral dissertation), "The principle [of excellence] is that every technical gesture in a poem must justify itself in meaning."
Anyone who has had to endure, either in person or in print, someone else's tortured justifications of every small formal gesture in a poem into meaningful interpretation must feel the dubiousness of this pronouncement. One might like to think of it charitably as a stimulating hyperbole, but there is no evidence that Fussell thought of himself as exaggerating. Nor is there evidence that he and those who solemnly repeat Pope's maxim feel the playful nudge of "seem," which receives what Halle and Keyser would have called a stress maximum in "must seem an Echo," as though seem were italicized and the line were admitting, ''It isn't really an echo, but in the aesthetics of neoclassicism, for the sake of harmony and concord, one should strive to feign that it is, in the same way that one strives to feign ease and naturalness in the midst of artifice when dancing a minuet." Then there is the whole matter of what the sound of verse should be trying to seem, which is not merely an echo, with lower-case "e," but because of the ambiguities latent in Pope's habits of capitalization also an Echo, the nymph herself with all her narrative and mythological possibilities. True, she could only speak when spoken to, but she got even in the end.
It would be both unfair and mistaken to suggest that only those under the influence of Augustan neoclassicism could see form as secondary and subordinate to content, the former silently reflecting the latter's bright light like a dead moon. Thanks to an oversimplified misreading of Emerson's statement in "The Poet" (mostly written in 1842 but published in 1844) that "it is not meters, but a meter-making argument that makes a poem" (Emerson was arguing that meter by itself could not make a poem, not that meters should be discarded), twentieth-century fictions of organic form based on free verse tended to play out various versions of the all-caps commandment attributed to Robert Creeley by Charles Olson in the essay "Projective Verse" (1950): "FORM IS NEVER MORE THAN AN EXTENSION OF CONTENT." Here Creeley's "never" renders his statement about as reliable as Fussell's "every." Hyperbole characterizes the rhetoric of Black Mountain more often than it does that of Poetic Meter and Poetic Form, but even in a less polemical version, which substitutes "usually" or "sometimes" for "NEVER MORE THAN," the hierarchy is clear: content comes first, form comes second, and devil take the hindmost.
Before turning to the work of correction—in the double sense of removing errors and of adjusting to meet a different standard, as when one speaks of glasses correcting for near-sightedness—we should practice what we preach and avoid the overstatements of Fussell's "every" and Creeley's "NEVER" by acknowledging instances when content does seem to call the shots and form does seem to follow its orders, as in a game of Simon Says. Here is a famous example from the flamboyantly self-referential passage in Pope's Essay on Criticism, a couplet much loved and discussed by the late John Hollander:
When Ajax strives some rock's vast weight to throw,
The line too labours, and the words move slow.
Only a stubborn curmudgeon could maintain that Pope was not trying here to stage a little piece of flashy verbal theater in which the rhythmic shape of the iambic pentameter couplet appears to take its cue from Homer's narration of Ajax and the giant rock he throws at Hector (Odyssey, Book 14). In this moment, content sings lead, form sings back-up.
The problem comes when one tries to talk about the form as "imitating" content. Following his Horatian model, filtered to him by Boileau, Pope assigned Nature priority over Art in a passage that appears almost three hundred lines before the Ajax couplet:
First follow Nature, and your judgment frame
By her just standard, which is still the same:
Unerring NATURE, still divinely bright,
One clear, unchanged, and universal light,
Life, force, and beauty, must to all impart,
At once the source, and end, and test of Art.
According to the fiction conjured by these lines (which are guilty of the wholly predictable rhyming of "bright" with "light," every bit as "expected" as the rhyming of "breeze" and "trees" Pope criticized in the later passage that includes the Ajax couplet), Nature is always the content, playing the leading role, and Art is always the form, playing an imitative or reflecting one. But in the Ajax couplet, there is no Nature. Ajax never hefted the big rock at Hector except in the realms of myth and art. If the rhythmic form of the Ajax couplet imitates anything, it imitates an earlier piece of art, something wholly imagined. If the sound imitates sense here, then it is an imitation of an imitation.
What is more, the supposed imitation of content by form is not the production of a likeness of an original but of an abstraction from the original, an auditory abstraction that resembles the original about as closely as one of Picasso's more extreme cubist experiments resembles anything recorded by a surveillance camera on an actual street in broad daylight. Even if we turn from the representation of something imagined in Greek mythology to the representation of something human ears might actually hear in the natural world around them (a turn from the meontic to the mimetic in Thomas McFarland's formulation), as in Tennyson's often-quoted lines from The Princess (1847),
The moan of doves in immemorial elms
And murmuring of innumerable bees [,]
there is still a problem with speaking of the momentary aggregation of the English phonemes /m/, /n/, /d/, /v/, /z/, /r/, /b/, and /1/ as doing anything like imitating the sounds of doves and bees in a way at all comparable to the imitation of a wild turkey call by a hunter wanting to draw a bird in close for the purpose of shooting it. In the case of the hunter, the likeness of the imitation is everything; turkey for dinner depends on the imitation being indistinguishable from the original. In the case of Tennyson's lines, likeness is tenuous at best and really not the point at all. No one hearing Tennyson chant his lines would mistake them for the sounds of doves and bees. Instead, what happens to the sound stream of Tennyson's verse is that it enters a patch of Class 2 or 3 rapids (Class 4 is reserved for Hopkins), and suddenly there is a bracing event for the kayaking reader or listener to negotiate, an event from which the sense of the lines benefits, in much the same way that the sense of the song "Happy Birthday to You" benefits when the words of the song are sung with gusto rather than droned in a monotone. In the case of both the Tennyson lines and "Happy Birthday to You" when sung, the form confers power on the content, the sound on the sense, which would otherwise be rather banal: here are some doves and bees, and by the way you're a year older. The content or sense has not suddenly manifested in awful majesty, which the loyal form or sound cannot help but herald with worshipful hosannas.
Hovering here is a version of the post hoc ergo propter hoc fallacy in logic. The rooster's crowing does not cause the sunrise simply because it precedes it. The intensified repetition of sounds in lines of verse does not have to follow from the sense of those lines. If anything follows anything, it is the temporary prominence of the sense that follows the sudden intensification of sound patterning. Any successful advertiser knows the trick. The trick for a poet, or a poet of Pope's aesthetic persuasion, is to make it seem that the auditory razzle-dazzle follows from the prominence of the subject matter, as if by fiat of the Muse, rather than the other way around. When we celebrate instances such as Pope's Ajax couplet or Tennyson's doves and bees, what we are doing is celebrating a high degree of success in performing this trick. There is nothing wrong with celebrating high degrees of success; there is something wrong with generalizing from these exceptional moments to statements that they characterize, or should characterize, the relationship between form and content, sound and sense, in the writing, reading, and discussion of all other poems.
There is another kind of relation between form and content that deserves more than a passing nod because in fact it does characterize most verse. (I admit the temptation to say "all," but though I have not been able to come up with exceptions, common sense says there must be some.) A useful example, because an extreme one, is Milton's bravura flourish about the topographic features of many a Hellish "Region dolorous" in Book 2 of Paradise Lost (1667; 1674):
Rocks, caves, lakes, fens, bogs, dens, and shades of death.
Put a gun, at three o'clock in the morning, to the sleepy head of a gifted, nimble interpreter and say that the trigger gets squeezed unless she or he can come up with an account of how sound echoes sense here, delivering up a plausible justification for this anomalous string of ten monosyllables, eight of which have claims on rhythmic stress, while only the five even-positioned ones hold tickets for the metrically prominent seats, and impressive things may happen.
You may hear talk of spondees (the existence of which in English-language verse some question, arguing that the contours of contrastive stress will always promote one monosyllable above its neighbors, but that argument becomes hard to sustain in the case of an itemized list or enumeratio, the monosyllabic elements of which are separated by actual or implied punctuation) and of the heaviness of spondees, echoing the dolorousness of the region or the laboriousness of demonic travel across it. You may hear how the enumeration of features echoes the imagined rhetorical shape of the divine curse, referred to in the next line: This, this, this, this, this, this, and this be cursed. You may hear that the suddenly dense consonantal mesh of glottal stops and terminal spirants, along with heavy assonance (three soundings each of so-called long "a" and short "e," depending on whether the vowel of "death" had finished its shortening from Chaucer's longer pronunciation by the time it came to Milton's ear, along with perhaps two of the vowel sound in "Rocks" and "Bogs," if he heard them as identical) and the rhyme of "Fens" and "Dens" echo the hideously garbled soundtrack of demonic gasps of despair at this first glimpse of the nether "Universe of death."
What you are unlikely to hear is that in his enumeration of otherwise ordinary topographic features of landscapes in the British Isles, Milton's iambic pentameter line was imitating a line of classical dactylic hexameter from Homer or Virgil, who, along with Dante, pioneered literary explorations of the underworld in Western poetry before him, and in particular he was imitating hexameter lines that, by the rules of quantitative substitution in classical prosody, allowed spondees to replace dactyls in the first two-thirds of a line. Moreover, in a stroke of double virtuosity, under the spelling rules observed by sixteenth- and seventeenth-century experimenters with classical quantitative meters in English, he made each of the ten syllables "long" (every vowel not long by "nature" as in "Lakes" or "Caves" becomes long by spelling convention when followed by two consonants), even as he allowed the two concluding iambs to imitate the return to regularity in the final two feet of the hexameter line, the first of which was usually a dactyl, the second of which seldom was. In other words, at exactly the moment when the conventions of classical prosody tended away from metrical substitutions in a line, Milton abandoned them, too.
It is not important for the argument under construction here that a reader follow or savor every detail in the last paragraph. What is important is this: the sound of Milton's line does not echo the sense, with due respect and apologies to the inventive interpreter rudely awakened at three in the morning; it echoes earlier lines of verse and becomes, in turn, a line that later lines echo. Although apparently he could not bring himself to ignore his own critique in the Ajax passage and permit "ten low words" to "creep in one dull line," Pope crafted two lines for the Rape of the Lock (1712-1714) that nod to Milton's: "Puffs, Powders, Patches, Bibles, Billet-doux" appears in Book I, and (note the inconsistency in his capitalization of nouns) "Men, monkeys, lapdogs, parrots, perish all!" in Book 4. In his footnotes to The Dunciad (1728-1743) Pope explicitly called attention to lines he patterned after Milton's in sections he headed "Imitations," kindly pointed out to me by my colleague David Vander Meulen. Well, of course, one might say. The Augustans prized imitation of prior models as a worthy aesthetic undertaking, but Romanticism brought the sea-change of valuing originality over imitation. Perhaps, but then how do we explain the third line of "Hamatreya" (1846) by Emerson, high priest of intuitive self-reliance, who declared, "Imitation is suicide": "Hay, corn, roots, hemp, flax, apples, wool and wood." This line—also an enumeration of ordinary natural phenomena, in its case the products of the land around Concord, Massachusetts—follows the Miltonic blueprint exactly, right down to the return to iambic regularity in the last two feet.
How do we explain such a line? We explain it by saying that although in occasional lines of verse, such as Pope's Ajax couplet, sound may seem to echo sense, what most lines of verse do (down" all," down) is to echo other lines of verse, and the Milton-Pope-Emerson example, though perhaps extreme in being so obvious, is also typical. Lines of iambic pentameter echo and imitate, but also vary and revise, other lines of iambic pentameter, without which they would never exist; short lines of free verse with lots of enjambment echo and imitate, but also vary and revise, other short lines of free verse with lots of enjambment, without which they would never exist. In other words, the sounds of verse may seem to be echoes of its sense, but really they are echoes of the sounds of earlier verse. To say so is not to banish all possibility that art can imitate non-art or ever be "about" anything but itself. The Quattrocento Florentine painter who aspired to modeling his Madonna and Child after nature may have been looking at, and imitating, a real-life woman or a real-life baby as a model. But with or without a real-life model, he was also imitating hundreds of other paintings of Madonna and Child already in his mind, the same as a twenty-first-century blues harmonica-player, who lives and plays the harmonica in the thick of a city, is much more likely to model a "train," the long sustained note meant to suggest the sound of a train horn and a touchstone convention of blues harmonica-playing, on previous musical images of train horns, rather than on the train horn itself, mostly likely unheard in the vicinities of Penn Station or Grand Central.
The subtotal of the above is that when we teach or write about poetry, we give our students or readers a much fuller and more accurate sense of the relation between form and content when we help them hear the sounds of verse in relation to other verse instead of encouraging them toward an interpretive contortionism that insists on twisting sound into an auditory allegory of sense. Where sound does remarkable things, sense may seem more remarkable also, and it is fine to recognize and cherish such moments. But it is also possible that sound may come on so strong that one loses the sense of sense altogether, as at times in Hopkins's poems, when any fiction of form echoing content must be dropped, as one struggles to swim against the powerful currents of an overwhelming sound stream to retrieve the bobbing flotsam and jetsam of comprehensible predication.
It is not to be expected that every student or reader will thank you for this correction. The last time I suggested to a class that lines of verse echo or imitate lines of verse, making unmediated and original self-expression in verse very difficult and rather unlikely, a student said she was offended. But the offense taken is understandable. In the ongoing afterglow of Romantic individualism, which the digital age may burnish to new luster or snuff altogether or leave pretty much as it is, few of us like to be told that originality, as we now conceive it, is an endangered species, even among those experimentalists who think of themselves as dismissing originality as an aesthetic value. (It is worth noting that the word "experimentalism" was coined by Coleridge and bears the ineradicable fingerprint of Romanticism, as does the aesthetic mode or modes it has come to signify.) Some applaud Kenneth Goldsmith's Day (2003) for copying verbatim an issue of the New York Times as an act of "uncreativity" that has no time for anything as old-fashioned as originality, but few would applaud anyone who uncreatively copied verbatim Goldsmith's verbatim copy, least of all the people who hold his copyright. Furthermore, if lines of verse echo or imitate other lines of verse, then to hear the resonances in lines of verse one has to read a lot of other verse, much of it necessarily older than the last ten or twenty years, and few people, especially beginners, whether young or old, want to be told any such thing.
So the correction is a correction of another fallacy, a fallacy of imitative form. The phrase "fallacy of imitative form" recalls Yvor Winters's Primitivism and Decadence: A Study of American Experimental Poetry (1937), where it means something different from the fallacy under examination here. For Winters "the fallacy of expressive, or imitative, form" was the fallacy of justifying the use of "disintegrating form in order to express a feeling of disintegration," a fallacy Winters dismissed as "merely a sophistical justification for bad poetry." More than three-quarters of a century later, with many other precepts of New Criticism now shelved and forgotten, this one still has sharp teeth to it, although Winters himself had a blind spot, or an acoustic shadow, when it came to hearing and valuing the integrities of Whitman's verse (as well as those of Joyce's prose). For Winters the fallacy of imitative form was that in itself the formal intent to imitate some inward or outward phenomenon is an insufficient justification for, and guarantee of, aesthetic value. Whitman, for example, lapsed into the fallacy by "trying to express a loose America by writing loose poetry." To paraphrase in older language, holding the mirror of art up to a mess in nature, including human nature, may result in nothing but a reflective mess. Or, to extend the thinking into our own moment, experimental poems justifying themselves as imitations of distracting digital noise may produce nothing more or less valuable than distracting digital noise.
For all its corrective bite, however, Winters's exposure of one kind of imitative fallacy left unchallenged the assumption that it is the job of form to imitate in the first place, to play second fiddle to something more important than itself, some content or sense or meaning it altruistically enhances or promotes. His sense of hierarchy was and is quite clear: "Form is expressive invariably of the state of mind of the author." With his "invariably" anticipating Creeley's "NEVER MORE THAN" and Fussell's "every," Winters set the formalist bar even higher by specifying that what form expresses invariably should be understood as the moral and spiritual discipline of the poet. Following the lead of Earl Miner's Comparative Poetics: An Intercultural Essay on Theories of Literature (1990), we might use the term "mimetic-expressive" to describe Winters's aesthetics here. Although he focuses on poetic form as an image of the poet's interior world, rather than on the exterior world of doves and bees, he still assumes that the function of poetic form is to imitate, represent, reflect, or embody that interior world in a formal image that serves content. In working to challenge and correct exactly the assumption, I am simply trying to keep before us the insights of some others and to add a small contribution of my own. The effort cannot help making certain theoretical assumptions and having certain theoretical implications, but it is primarily an effort in practical criticism, one resulting from years of trying to help newer readers and writers face what they think of as the divide between poetic form and poetic content. Many of these newer readers and writers have been or are quite good ones, and some are truly outstanding. But the problem of how to connect the mass on the left with that on the right still dogs most of them and often muddies otherwise excellent work.
Forty years after the publication of Pope's Essay on Criticism, Samuel Johnson wrote his ninety-fourth number of The Rambler, dated Saturday, February 9, 1751. This short essay abundantly repays reading or rereading by anyone interested in the questions at hand. Formulations such as "Sound can resemble nothing but sound" are potential safeguards against a host of errors. The essay is an instance of judiciously skeptical neoclassical thinking Fussell seems to have neglected or dismissed. Johnson named the subject of his meditation in the respective first sentences of his first two paragraphs, "The resemblance of poetick numbers, to the subject which they mention or describe," and "The general resemblance of the sound to the sense." By "numbers," he meant meters, and for our purposes it will do to hear instead "prosody" or "form" or even just "sound" in the formulation most often quoted from the essay, which opens the third paragraph:
"It is scarcely to be doubted, that on many occasions we make the musick which we imagine ourselves to hear, that we modulate the poem by our own disposition, and ascribe to the numbers the effects of the sense."
Here is the imitative fallacy of much practical criticism of poetry in a nutshell: we ascribe to the numbers the effects of the sense. We project onto the sound or form our own preoccupations with the sense or content. (Johnson's "we" refers to readers of poems, but readers of prose fiction and non-fiction can project this way as well.) If we read poems under the signs of desire, possession, jealousy, loss, we are likely to hear them in the sounds of a poem; if under the signs of race, class, gender, politics, transnationalism, and hybridity, ditto. As for the value of moments when sound does seem to echo sense, form to imitate—or "resemble" in Johnson's language—content, his designation of these momentary congruencies as a "species of embellishment" hardly suggests the rock on which to build an entire poetics. Milton gets most of the attention in Rambler 94, and the concluding sentence of the essay puts Johnson's periodic style on full display in the final assessment of Milton's echoing embellishments, albeit with a slight misquotation of the opening invocation of Paradise Lost: "He had, indeed, a greater and nobler work to perform; a single sentiment of moral or religious truth, a single image of life or nature, would have been cheaply lost for a thousand echoes of the cadence to the sense; and he who had undertaken to vindicate the ways of God to man, might have been accused of neglecting his cause, had he lavished much of his attention on syllables and sounds."
Cheaply lost? One single sentiment of moral or religious truth is worth a thousand echoes of sense by sound? Not every poem is or aspires to be Paradise Lost, although perhaps Winters would counter that every poem aspires, or should aspire, to the same degree of moral integrity and spiritual discipline as those of Milton's epic, but in the Johnsonian economy the exchange rate could not be clearer. By all means, enjoy and point out to others notable convergences of sound and sense, but do not inflate their value into a dominant poetic principle of imitative form, and do not assume, in Johnson's phrasing, that they are "studied" rather than "casual," that they are exemplary nodes of aesthetic ideology rather than happy accidents of what Hart Crane called "felicitous juggling."
But we cannot leave the correction there, with sound so thoroughly devalued in relation to sense, especially during a period when the percentage of people associating poems with moral or spiritual truth, despite the arguments of Winters, is much smaller than it was among readers of enlightenment Europe and their counterparts across the Atlantic. It is not hard to say what sound does not do, or what it rarely does, in relation to sense; it is harder to say what it does do, and yet this is the direction the argument needs to go if we are to offer others a richer sense of why sound might matter at all. One bold and innovative attempt to move beyond what she calls "time-worn principles, such as 'the sound must be [sic] an echo of the sense'" is Veronica Forrest-Thomson's Poetic Artifice: A Theory of Twentieth-Century Poetry (1978). In her preface Forrest-Thomson argues against "Naturalisation" or the "attempt to reduce the strangeness of poetic language and poetic organization by making it intelligible, by translating it into a statement about the non-verbal external world, by making the Artifice appear natural." Forrest-Thomson's recommended alternative to Naturalisation, schematized neatly a few pages later, has a polemical edge to it, as she charges William Empson, for example, with failing "to accept that the future of poetry lies in the exploitation of non-meaningful levels of language."
But the argument here is that the sounds of poetry have been exploiting non-meaningful levels of language all along, and it is we, or those of us who subscribe to some belief in the sense-echoing obligations of sound, who have missed or ignored them.
To jump the argument several notches from Johnson's corrective devaluation, and to provide some descriptive underpinning for Forrest-Thomson's prescriptive vision of the future of poetry, we can turn to Julia Kristeva's La revolution du langage poétique (1974), abridged and translated into English as Revolution in Poetic Language (1985). Kristeva's title points to the revolutionary activities of poetic language, particularly its rhythms, but her own argument is no less revolutionary and paradigm-shifting. Most important for the discussion here is Kristeva's claim that poetic rhythm should be identified with an infant's prelinguistic connection to its mother (the realm of the "semiotic"), a connection that must be repressed during the acquisition and development of language in order for the infant to say things (the realm of the "symbolic"). According to Kristeva, when translated into the language of Pope, sound threatens to disrupt sense by tending toward "autonomy of meaning" and "sonorous distinctiveness." The sound stream of verse and the rhythms it generates reactivate the repressed instinctual, maternal element of language, which revolts against the symbolic order of language in pursuit of its own pleasures and tantrums.
Particular readers may find various aspects of Kristeva's thinking not to their taste or liking. Some might object to the neatly conceived antithesis, or constructed binary in the prevailing parlance of today, of semiotic versus symbolic. One does not need an advanced certificate in deconstructionist methods to suspect that the boundary between these two categories is somewhat flimsy and porous. Some might be distracted or alienated by the dominance of a Lacanian psychoanalytic idiom in Kristeva's theoretical fable. Some might question the relevance or validity of assigning genders to the two sides of the antithesis, especially when even casual observation of men and animals could lead one to the impression that testosterone can also manifest itself in disruptive, transgressive, libidinal, pre- or non-linguistic ways that don't give a hoot for the repressive functions of the symbolic and are happy to rejoice in their own sonorous autonomy. The history and appeal of rock music, for example, would be very different if it were not so. With the accession of more and more women, since the publication of Kristeva's book, to the thrones of patriarchy in universities, corporations, or governments, in time the coding of this primal unruliness as feminine may come to seem as quaint as describing the United Kingdom as post-colonial because the Romans once controlled Britannia.
Whatever one's comfort or discomfort with the trappings of Kristeva's argument, now forty years old, her core insights about poetic rhythm boldly suggest necessary correctives for the fiction that verse form should imitate content, sound should echo sense. The reason that form and sound do no such things most of the time is that most of the time they are operating autonomously and with no necessary, and certainly with no necessarily subordinate, relation to content and sense. Here the objection might be that there is no separating sound from sense when it comes to the operations of language, as is suggested by Frost's coining of the phrase "the sound of sense" to refer, for example, to the speech contour of rising inflection that ends a question, or as in the case of the dactylic sloping of the Latinate word "copulate," as opposed to the monosyllabic slap of any of its Germanic synonyms. The speaker or writer who opts for the former over the latter could be doing so for a host of reasons, some of them social but some of them not, and the content of any word chosen will include the social, political, and linguistic history of the word, the last of which cannot help but have rhythmic implications for the form of the word because, in this case, the fossilized inflections of Latinate words borrowed into English, especially verbs, tend to lengthen them and thus render them rhythmically distinct from their shorter Anglo-Saxon counterparts.
Fair enough. But to say that the sound of verse operates autonomously with respect to the sense is not to say that sound and sense are always separated by a chasm. It is to say that although a particular word may package sound with sense in a particular way, the sense of a line or stanza or poem is not the lawgiver to its sound. The historical template of a Shakespearean sonnet has one hundred-forty syllables with even ones stressed, seven rhymes, and a final couplet; the words in it will combine the forty-something phonemes of English in various ways. Because of these constraints certain sound events are likely and others are unlikely, if not impossible. These are the laws that govern the sound of the verse, that make it autonomous. In addition to arousing various associations with the history of the sonnet-associations that may not appear to be as unruly as those of local sound-patterning but that are not necessarily tamely, and invariably, subordinate to the sense or content of a particular sonnet either-particular sonneteers may create or discover moments when certain sound events parallel certain sense events, and fruitful, memorable, sometimes jaw-dropping things may happen at such moments. But these moments of convergence or congruence are moments of temporary alliance between sovereign states, not the manipulations of a puppet regime by its overlord.
Johnson labored to get us not to overestimate the significance or prevalence of sense-echoing sounds. Kristeva gave us the chance to recognize sound as operating distinctly, autonomously, and often disruptively with respect to sense. Turning briefly to recent research in neuroscience, we can complete the correction of the imitative fallacy by considering that the way sound affects us is not only independent of the effects of sense (much of the time, if not most of the time) but prior to the effects of sense. In this final turn, there is no aspiration to make a fetish of science or to give it the last word, which it will not have here. Predictably, the recent flood of books and articles explaining this or that aspect of human life in terms of electrodes and brain scans has produced a healthy, skeptical reaction. But the accumulated weight of the fallacy to be thrown is vast, and if a little neuroscience can help us budge it, the assistance is welcome.
In a volume entitled Beyond the Finite: The Sublime in Art and Science (Oxford, 2011), editors Roald Hoffmann (a poet and a Nobel laureate in chemistry) and Ian Boyd Whyte (an architectural historian) have gathered essays that offer various perspectives on the subject of the subtitle. In his contribution to the volume, "Affective Foundations of Creativity, Language, Music, and Mental Life," psychobiologist Jaak Panksepp has provided an account that in many ways parallels Kristeva's, although he has cast his narrative in the nomenclature of neurobiology rather than in that of a psychoanalytic feminism. In Panksepp's narrative, the neocortex occupies the position held by the symbolic in Kristeva's. But rather than signifying patriarchy, the neoocortex is an evolutionary Johnny-come-lately that suddenly expanded on the tops of their brains when our forerunners discovered cooking, which made obsolete the massive muscles needed to grind raw food and "allowed our cranial plates to expand." Rapid cortical expansion followed, multiplying structures of "three thousand neurons or so-called cortical columns" that "endowed us with a better capacity for symbolic thinking and communication than any other animal and a greater capacity to manipulate the world. It made us smart."
What Pope called "sense" clearly emanates from and addresses the neocortical thinking cap of the human brain. What he called" sound" emanates from another realm altogether, the ancient subcortical basement of the brain, which the brains of other creatures also have, enabling them to experience their basic emotions—"playful joy, fearfulness, anger, lust, and many other desires," for instance—just as intensely as we do, even though they do not play instruments or paint or write. Panksepp's argument contains many more fascinating details and formulations, including "the brain molecules that resemble the elixir of the poppy" and "can reduce all painful feelings, especially the sting of loneliness"; laughter in rats who are tickled; the dependence of the emotional power of music "on the arousal of subcortically situated emotional systems"; and the shivers or chills or goose bumps of "skin orgasms" which result from the ability of certain sounds to resemble the separation cries of babies and touch "the evolutionary roots of social pain." For our purposes here, what matters most is the basic mapping of sense in the top of the brain, where it is relatively new and where it distinguishes us from other animals, and sound in the lower part of the brain, where it is relatively old, very old, and links us to other animals.
With this map before us, it becomes easy to see why any poetics describing or prescribing poetry in which sound echoes sense or in which form is the lackey of content cannot stand or long endure. The sounds of animal utterance, including those of our two-legged ancestors, have been going about their independent business for much longer than people have been making sense with rhythmically arranged words. In the particular kind of verbal rodeo we call poetry, it may be possible for a particularly skilled, hard-working poet to get up on the back of the large bull of sound and, under carefully controlled and regulated conditions, "seem" to show some mastery by staying mounted on the bucking animal for at least eight seconds, but it would be laughable to assign the bull a secondary or subordinate position. The magic trick or con-game performed by Milton or Pope or Tennyson or anybody else is to get listeners or readers to narrow attention to those eight seconds and evaluate the performance accordingly. For the space of a couplet Pope may be able to make it seem he has tamed sound to let his sense ride it, but in fact the sonorous properties of the couplet may be and probably are arousing all kinds of activity in subcortical emotional systems that could not care less who Ajax might be. To say that activity in the subcortical emotional system is taking its cue from the sense-making operations of the neocortex would be like saying that the ethyl alcohol imbibed in a beer or glass of wine or extra dry martini is taking its cue from the subject matter of a conversation, which seems to grow more and more profound as the blood-alcohol content rises.
Johnson's Rambler essay makes the crucial and checkmating point that Milton used the same sound embellishments in different lines that share no sense, thereby exploding the claim that particular sounds arise as a result of particular sense. What Panksepp's thinking helps us understand is that they do not need to share sense in order for us to enjoy the illusion in the case of each line that the sound fits the sense as though it were tailor-made for it, when in fact it was bought off the rack. The very activity of hearing and interpreting sound in terms of sense is a neocortical activity; sound arouses autonomous subcortical responses, but the neocortex jumps to read them as related to the sense the neocortex itself produces and understands. What is surprising about that? The neocortex is the Narcissus that wants to hear all sound as Echo in love with him. To switch the metaphor, in most of our practical criticism of poetry the neocortex behaves as a small modern colonizer appearing to bring sense and civilization to the much larger, uncharted continent of sound and primitive savagery.
This last formulation in turn enables us to grasp that a poetics describing or prescribing sound as the echo of sense may happen to have emerged, as part of a larger aesthetics of imitation, at a particular moment in a particular part of the world, as in the case of the Mediterranean rim and then Western Europe, but other poetics could emerge at other moments in other parts of the world. And in fact they have, as some time spent with Miner's Comparative Poetics (1990), which argues that the Western poetics of imitation is a minority of one among the many poetics of the world, or the newest edition of the Princeton Encyclopedia of Poetry and Poetics (2012) makes clear. Let us move toward closure with brief looks at two examples, the first of which is Sanskrit. As Western poetics evolved from Plato to Aristotle to Horace to Boileau to Pope and an aesthetics of imitation and echoing, something different was happening a few thousand miles to the east in Sanksrit poetics, the influence of which reached eventually as far as southeast Asia. At the center of Sanskrit poets lies an aesthetics of affect, not one of representation. In considering the ornamentation of a poem (alamkara), which would include but not be limited to its euphony or sonorousness, what matters is the arousal of an emotional taste or flavor (rasa; literally, "sap" or "juice") in the hearer or listener.
When it comes to metrics, for example, the point of an ancient Sanskrit meter and its rhythmic variations is not to imitate or echo but to produce an effect, which in this case is an affect, in its purest, least contingent form, a form that does not depend on or limit itself to the sense of the words the poet has arranged metrically. As the title of Guy L. Beck's Sonic Theology: Hinduism and Sacred Sound (1995) suggests, sound in Sanskrit verse, especially in the chanting of metrical mantras, fuses aesthetics with theology, cosmology, liturgy, and ritual. (See Beck's first chapter, "Vedic Sound," for example). In the opening verses of the Chandogya Upanishad, for instance, it is clear that sound is not the servant of sense; it is the source and essence of the creation:
Let us meditate on OM the imperishable, the beginning of prayer. For as the earth comes from the waters, plants from the earth, and man from plants, so man is speech, and speech is OM. Of all speech the essence is the Rig Veda; but Sama is the essence of Rig, and of Sama the essence is OM, the Udgitha .... What is rig, what is sama? As rig is speech, so sama is song, and the imperishable OM is the Udgitha. Speech and breath, Sama and Rig, are couples, and in the imperishable OM they come together to fulfill each other's desire. For those who, knowing this, meditate on the imperishable OM, all desires are fulfilled. (The Upanishads, trans. Eknath Easwaran )
Whatever a Western reader's level of familiarity and comfort with the erotically charged metaphysics informing this passage, one cannot help but feel the ridiculousness of assigning sound, as it is conceived here, the menial task of imitating doves and bees. To do so would be to repeat the bathos imagined by Emerson in the "Language" section of Nature (1836): "We are like travellers using the cinders of a volcano to roast their eggs."
Here the perfectly reasonable rebuttal might be, Doesn't representing or imitating an emotion produce the emotion, and doesn't Aristotle's notion of catharsis point to the importance of affect in his poetics as well? The answer to the former would be perhaps (not all readers or watchers of King Lear will experience the same emotions as a rash and dispossessed old king), to the latter, yes indeed. One could also argue that any aesthetic arrangement designed to produce a particular emotion (one thinks here of Eliot's formulation of the objective correlative) could also be considered a representation of that emotion. All reasonable arguments, but in Aristotle's poetics, which is founded after all on a theory of drama, the road to affect, in the form of catharsis, necessarily leads through the watching of mimetic imitation of human actions. For Aristotle there is no affect without imitation, whereas in the Sanskrit understanding of sound imitation is irrelevant. OM imitates nothing larger or prior to itself. The important point here is that the Sanskrit theory of the rasa is closer to Panksepp's understanding of how sounds arouse subcortical emotional activity in a listener than it is to Pope's fiction that sounds echo neocortical sense. The rasa produced by the ornamentations of a poem is analogous to the harmonic vibrations produced by rubbing or striking a Tibetan or Himalayan singing bowl with a mallet. The vibrations generate a continuous hum that can support or enhance meditation, religious ritual, and healing. What the continuous hum does not do, or what it is not valued for, pace Winters, is representing or expressing the mind or spirit of the person who hammered the bowl into shape.
Then there is the classical poetics of China. A few centuries after Plato was narrating Socratic doubts about poetic copies of ideal forms, writers of the Han dynasty (after early third century B.C.) formulated a poetics in the Great Preface of the Shijing, the anthology of poems compiled, according to anachronistic tradition, by Confucius. The Great Preface prescribed the proper role of the Shijing, as teaching the reader to act rightly and ethically in concentric circles of government, from self-government at the innermost to government of the state at the outermost. In the opening note (titled "Procedure") to his pungent translation of the Confucian Analects (1951), Ezra Pound quoted this summary of poetics by Confucius, or Kung, the Master: "There are 300 Odes and their meaning can be gathered into one sentence: Have no twisty thoughts." Unlike the poetics of Aristotle, based on drama, classical Chinese poetics emerged from theories of music, and traditional Chinese poetics had no separate term for "poetry"; shi was the closest approximation, and originally it referred to the combination of the ancient musical repertoire and the lyrics sung to the music. The purpose of shi was the formation of personality in the literate, governing class and, consequently, of social harmony. In the formulation Miner adopts, Chinese poetics is "expressive-affective," meaning that it (" it" serving here as shorthand for an historically wide-ranging and diverse body of thinking, as Haun Saussy has patiently reminded me) focuses on conveying to the listener interior states of order and harmony, not on producing recognizable images of exterior phenomena. Eventually, when the old musical settings were forgotten, shi came to refer most commonly to the written texts of the songs they accompanied. To equate shi-as-music with form and shi-as-lyrics with content, on the model of Pope's formulation of sound and sense, would be to import a foreign duality and impose it where it has no relevance.
To those who object that bringing principles from Sanskrit or Chinese poetics into Western classrooms looking at Western English-language poems betokens an unenlightened critical Orientalism, and is every bit as misguided and inappropriate as resorting to the form-content duality in talking about Chinese shi, one response is that the best pedagogical use of comparative poetics is not necessarily to substitute the greener grass of a seemingly exotic paradigm for the burnt-out lawn of our familiar one. When trying to help readers think more deeply about poetic form and content, sound and sense, the most helpful use of comparative poetics is as an invitation and an encouragement to move beyond reductive, deadening habits of reading that all too often are driven by compulsive obsession with representation, with reducing form or sound to a representation of content or sense.
But what is there apart from representation? Doesn't everything represent something apart from or beyond itself? Once again, a few minutes spent considering another model, that of the prohibition against representation in traditional Islamic art, might help with loosening interpretive fetters. Instead of insisting on hearing and understanding sound as a representation of sense, what might begin to happen to our sense of sound if we began to treat it as the auditory analogue of the non-representational geometric designs of zellige tilework in Moroccan architecture? We might become even more acutely aware of the schematic intricacies of auditory design as ends in themselves, as well as of the imminent sublimity of design that consciously, willfully renounces representation of what cannot and must not be represented, design that is complete in itself and yet also part and parcel of a great and ultimate hush. We might take a baby step beyond mimesis, beyond the assumed fissure between ideal and copy.
Many a reader or student may not be able to make this last step, or even a baby version of it, right away. But that reader or student probably can follow, and quickly benefit from, a corrective reminder that the sounds of poems never took an oath to uphold the making of sense; they never took an oath or signed a contract binding them to an instrumentalist understanding of poetry as necessarily a form of communication, an end to which all parts of a poem must contribute dutifully as means. Much of the time we use vocalized sounds or written images of vocalized sounds to signify or express. But we do not always do so, and we do not always receive vocalized sounds or written images of those sounds as messages neatly addressed to the nearest neocortex.
"Only the lull I like, the hum of your valved voice." Whitman felt so committed to the auditory shape of this line from the opening of the 1855 "Song of Myself" that even after he subsequently adopted the convention of substituting apostrophes for es in past-tense inflections (for example, "form'd" in the sixth line of the Death Bed edition), he not only retained the e of "valved"; he left nothing to chance and marked it with an accent grave: "valvèd." This is a small moment, and a surprising one where metrical regularity is not the norm, but it testifies to pleasure, and to the orthographically enforced prolonging of pleasure, one can take in the lulling hum, whether of voices or verses, pleasure independent of messages or communications. And there are other examples of voiced sound unharnessed to instrumental utility. What is one doing when one hums tunelessly to oneself? What does acousticophilia—the tendency to be sexually stimulated by sounds, whether verbal or non-verbal—have to do with instrumental utility? How is the man or women muttering or shouting to no one in particular on a city street advancing communication? What about autistic echolalia or Tourette Syndrome? Poems can behave, and people can respond to them, in ways analogous to each of these examples. That the sounds of verse can also accompany memorable messages, sometimes seeming even to carry those messages, often profound ones sent decades or centuries ago, confirms their versatility and capacity to cooperate with the demands of sense-making. But they do not have to cooperate to be marvelous, and we have much to gain by learning to marvel, and helping others to marvel, when they do not do so.
* * *
About the Author
Stephen Cushman's newest books are The Red List: A Poem (LSU, 2014) and Belligerent Muse: Five Northern Writers and How They Shaped Our Understanding of the Civil War (UNC, 2014). He is general editor of the fourth edition of the Princeton Encyclopedia of Poetry and Poetics (2012) and Robert C. Taylor Professor of English at the University of Virginia.
Southern Methodist University
Editor-in-Chief: Willard Spiegelman
Senior Editor: Jennifer Cranfill