Anthony Lombardy

A notable trend in recent American poetry, the return to the use of traditional metrical forms and rhyme, which some time ago accumulated enough momentum and self-consciousness that it became sensible as well as convenient to speak of this trend as a movement, a "new formalism," has not lacked a vivid theoretical account of itself. A very ambitious program of theoretical justification for the new formalism is the "organicism" of Frederick Turner, which purports to show that the verse line is organically related to the human manner of retaining and processing information as well as to more fundamental physical traits like the rhythms of our hearts and respiration. Such accounts have long been familiar in American pedagogy (cf. John Frederick Nims's admirable text, Western Wind: An Introduction to Poetry.) One could even say, as is usually acknowledged, that ultimately these claims go back to Aristotle's observation that iambic rhythm most closely resembles natural speech.

Organicism is very helpful as a framework for thinking about verse, since it promises new avenues for investigation and makes claims that are both explanatory and potentially falsifiable, but it does have one important inadequacy as the theoretical justification for a new formalism: it fails so far to account for the success of non-metrical poetry, which, if metrical verse is "natural," must be seen as unnatural or anomalous or an aberration. Nor can organicism account for dramatic changes within the tradition of metrical verse or for the separation of poetry from song.

Readers who observe the seasons of poetry and criticism don't fail to associate claims for metrical versus non-metrical poetry with other changes in literary fashion, and may anticipate that theoretical justifications for literary practices are headed the way of other self-justifying literary programs and theses. In the end, one may well think, it is only the poetry that matters. Certainly the lack of an all-encompassing theory is not necessarily a bad thing, if literary movements prosper more in the absence of a theory than in its hectoring presence, but there is in the present case a serious disadvantage to the theoretical void. A consequence of the failure of the neo-formalists to account theoretically for the success of non-metrical verse is that a dismissive tone sometimes creeps into their practical criticism, a tendency to ascribe the success of non-metrical poetry to transient social and political causes. This tone echoes, unfortunately, much of the mainstream American poetry establishment's dismissive attitude toward the new versification. An unfortunate dynamic has thus been created that sours the tone of critical discourse in this country. This tendency is most pronounced when the poetry is read as a token for political attitudes. At its worst, the polemic of the formalist puts on the armor of conservative neo-classicism and sends that re-animated fossil into imaginary battles, while the polemic of the free verse zealot is still enflamed by whiffs of tear gas from the Chicago of August 1968.

The re-invigoration of traditional verse forms is beginning in recent years to receive the interest it merits, but what is needed, clearly, is an account which does justice both to metrical and to non-metrical poetry and which tries to motivate the processes of change within poetic traditions. In the following paragraphs I shall sketch a few notes for such an undertaking. Doing so will involve us in questions about the nature of poetic form, poetic schools, movements, revolutions, and why they occur.

Discussions of poetic form frequently end up involving claims about what poetry is, has been, or ought to be. Such claims are hard to vindicate in light of the absence of a specific differentia for poetry, a defining feature that every poem has to have, bad poems potentially and good poems actually. I won't waste time trying to find such a feature for poetry, since there is no reason to think that one exists. Yet, although I accept the negative premise that poetry doesn't have to be concrete, beautiful, dense, imagistic, or formal, poems that survive the politics and mores of their own generation do seem to have something more in common than felicity of thought and expression, and bad poems have something in common with good poems, and in this paper I do want to assert what that is and why it is so elusive. To do so, we need a concept from linguistics, the concept of "marking."

Roman Jakobson defines marking as follows: "The general meaning of a marked category states the presence of a certain (whether positive or negative) property A; the general meaning of the corresponding unmarked category states nothing about the presence of A." Gregory Nagy continues this definition (Pindar's Homer 5): "The unmarked category is the general category, which can include the marked category, whereas the reverse situation cannot hold. For example, in an opposition of the English words long and short, the unmarked member of the opposition is long, because it can be used not only as the opposite of short ("I am reading a long book, not a short one.") but also as the general category ("How long is this book?")...The marked member is defined in terms of the unmarked member--and not the other way around." Since all linguistic phenomena can be described as marked or unmarked, and what we call "poetry" is marked in a great variety of ways, we are looking for nothing like a specific differentia, but we are going to find it useful to talk of some types of linguistic marking as "poetic form."

The notion of marking entails the notion of generality. The unmarked term is the more general. I would like now to emphasize a claim crucial to an understanding of poetic (or any other) fashion: generality is generality in a domain. So the category that is less general and marked in one domain can be the more general/unmarked in another. For example, in an opposition between rhythmical order/disorder (I mean not the words "order/disorder" but the rhythmical phenomena), disorder (in the relevant sense) is more general hence unmarked in the domain of everyday speech, while in the domain of metrical verse order is more general, hence unmarked. Since generality is a function of the category's occurrence in a domain, change in or of the domain can reverse the valence of marked/unmarked categories. For example, in the world of everyday speech "night" is marked, "day" unmarked. We say things like "It's a beautiful day!" or "Those were the good, old days!" The latter usage clearly includes "night" in the more general, unmarked category. If we move into the domain of hotelkeeping, however, valences reverse. The desk clerk routinely says something like, "How many nights will you be with us?"

In a narrow and formal sense a domain is the set of substituends which can be the value of the variables in a given discourse. So the domain of students at a certain university would have, say, ten thousand members, the domain of English vowel sounds, 60 or so members. The domain of metrical combinations possible in English prose or free verse an indefinite but not infinite number of members. Every element of poetry that can be talked about has been conceived as such a domain, e.g., diction, syntax, rhythm, alliteration, metaphor. When we cite some feature of a poem because it is unusually felicitous or striking in one of these aspects, we are naming a marked element. When we find, on the other hand, cliches and metrical tedium, these elements are unmarked. Not every marked feature is felicitous nor every unmarked feature lacking in felicity, since sometimes the familiar is apt and the unfamiliar inept. I remember as a student of Robert Fitzgerald having my eyes opened by his claim that poems have a certain "cliche quotient," since the language could not and should not always be new. But you'd better not go over your quotient!

Before a poem begins to be read or heard, the domain of the poem is configured by the performative, that is, by the genre or by the functional expectations of a community. We do not expect an elegy to sound like or to treat the same topics as a love poem or a hymn. The setting helps to frame the expectations of an audience, as would the church at a memorial service, or the literary magazine, and the title also serves to orient the audience, sometimes with a clear genre indicator, such as the word "elegy." Once we are within the poem, those expectations are modified as they are informed by the poem's progression. Genre, on this account, is an engine for generating the expectations which configure some aspects of the domain, especially those which pertain to the poem's argument, subject, and attitudes, as well as, sometimes, its metrical form. Roman elegy, for example, is defined by its metrical form (the elegiac couplet) as well as by the subject and attitudes of the genre. The genre also contains signals, topics, and construction types adapted to discharge its function as well as the rhetorical intentions of the poet. For example, the classical hymn typically includes an invocation, hypomnesis (a recollection of past services between the petitioner and the deity), and a request. An audience expects to encounter these elements, and those expectations, however variable with the audience, and impossible to capture precisely, constitute the "unmarked" template from which the poet diverges and to which he returns.

Marking, as we've said, means minority status in a domain. What poets do, primarily, is not generate ideas or arguments, but attune themselves to the complex and elusive status of the domains from which a poem claims its phonological, lexical, syntactical and semantic elements. I shall define poetic revolutionaries as those who are the first either to recognize a reversal of valence between marked/unmarked categories or as those who successfully reconfigure the domain of a poetic utterance. Such success is not necessarily a grand thing if it is achieved at the cost of narrowing the domain. Such poets are then writing for other poets. In the most extreme versions of this phenomenon the poet is writing only for himself, a poetry whose marking is intelligible only in the domain of his own experience. Some confessional poetry is like this.

So much for the theory of marking. As for poetic marking, I agree with Johnson, Benveniste, and Nagy that poetry is performative and that "an utterance is performative insofar as it names the act performed," that is, insofar as it names itself and the occasion of its utterance. The "occasion" of poetic utterance in primitive societies is, perhaps, a ritual or mythic re-enactment. In later societies the "occasion" is, on one level, the literary genre, which, whether or not it recalls a mythic "occasion," is the domain of poetic discourse. More concretely, the occasion of the poem is the poetry reading or literary magazine, each of which identifies itself as a performative or epideictic occasion. Among the most notable features of postmodern American poetry are the absence of genre indicators and the working assumption among critics that each work of art must be taken as something sui generis. This phenomenon probably owes more to Romantic notions of originality and the cult of the individual than to deconstruction or philosophical solipsism.

Our notion of poetry is, then, a nest of several variables. It is speech designated as a performance or speech object and marked phonologically, semantically, and syntactically. Examples of semantic marking include rhetorical figures like metaphor and simile (as opposed to literal expressions) as well as unusual collocations and associations. Syntactic marking would range from the Latinate syntax of Milton to the deliberate use of a series of four vs an unmarked tri-colon. Phonological marking might be any unusual regularity not characteristic of speech, such as rhyme, assonance, or alliteration.

If a poetic tradition is successful enough for long enough its forms and topics become familiar and normative. It becomes difficult to distinguish oneself; "originality" becomes a conspicuous value. The marked terms and categories of the tradition have, because of their success, lost their marking, and the first poets to recognize this change in valence are poetic "revolutionaries." In the twentieth century, the poetic revolutionary is one who in Pound's words is "making it new, " and this innovation, this "newness" is inevitably connected in twentieth century criticism with the loosening of poetic meters begun by the modernists and completed by their postmodern successors. But should it be?

The association of innovation with a freeing from traditional forms may seem intuitively obvious, even compelling to an audience reared in a late Romantic (or late Hellenistic) sensibility that highly values feeling, freedom, and the idiosyncratic, but the loosening of poetic form is sometimes associated with the conservation of traditional language rather than with poetic innovation. This association is obvious at the very beginning of the Western tradition. Gregory Nagy has demonstrated how the dactylic hexameter of Homer, with its free substitution of spondees for dactyls in the first four feet of the hexameter line is a loosening of a more rigid and isosyllabic line, called the Aeolian third pheracratean, a loosening motivated by the desire to conserve inherited poetic formulas in the face of changes in the Greek language, like the loss of intervocalic sigma and upsilon. Because of these changes, short syllables were harder to come by in the language of Homer's contemporaries than in the language of his oral predecessors. Loosening the metrical rules made it possible to conserve the inherited language of the oral tradition. If one believes, as I do, that the English pentameter owes its most conspicuous debt to a Latin line, the hendecasyllable, then the English epic line also represents a loosening from a more rigid, isosyllabic predecessor.

In traditions where linguistic originality is highly valued, as in the Hellenistic period, stretching 500 years from Callimachus and Apollonius of Rhodes through silver age Latin, meter tends to get tighter not looser, and this is true across literary genres. So the fifth syllable of the Sapphic pherecratean, that could be long or short in Sappho, Alcaeus, or even in Catullus, is always long in Horace. In epic, the line with a spondee in the fifth foot, one in eighteen in Homer, all but ceases to exist in Vergil. Poets like Ovid and Propertius admit hiatus and elision more carefully and sparingly than their "classical" predecessors or their less original contemporaries.

These examples should, I hope, suggest the error of thinking that we will ever do away with literary fashion, that poetic forms are above such changes, or, on the other hand, that they represent merely the rear guard action of a dying aesthetic. If one looks seriously at the diction of 20th century American poetry in terms of Modernist claims for verbal innovation, it's paradoxical, on the one hand, that Pound (in long stretches of the Cantos) is cultivating, if not a Homeric kunstsprache, then certainly an archaizing diction, and that Williams, on the other, in most of his corpus is hewing doggedly to quotidian, apparently "unmarked" diction. Followers of Williams seem quite aware of this fact when they programmatically dismiss from their poetry "poetic," i.e., marked diction, in favor of everyday speech. In general, the most innovative diction in twentieth century poetry up until, say, 1960, can be found in the more traditional forms, while the lexicon of free verse in our era has been, with some interesting exceptions, profoundly conservative. The two figures most frequently recruited to represent this dichotomy are Williams and Stevens. After 1960, the eclipse of traditional forms and the advent of post modern influences on American poetry coming from the Continent and Latin America make it difficult to sustain such a comparison, though I believe the validity of this association is still suggested in the work of poets like Lowell and Roethke who have more and less formal work to be compared. The point of this little survey is not that those who use rigid metrical forms are more original or inventive in their diction or phrasing, but merely that there is no necessary connection between originality of diction or phrasing and metrical freedom. Moreover, an exception seems to be Robert Frost, with whom everyday diction and strict forms go together.

I would argue that Frost's diction is a function of his literary style and his poetic persona. In discussing Frost we would have to distinguish sharply between everyday diction and plain style. Plain style entails everyday diction, but the entailment is not mutual, since diction is but one element of style and everyday diction can be used to generate an elevated rhetoric, if the appropriate auxetic words, expanding cola, and the other elements of style are in place.

A progression in metrical regularity may simply be a sign of an advanced tradition. The comparative evidence is considerable: In the Indic tradition metrical regularity indicating late composition (Nagy Comparative Studies in Greek and Indic Meter35); the progression of the Greek glyconic to greater rigidity (Nagy 36); and the examples already cited from Hellenistic versification. It is the job of practical criticism to say whether greater metrical regularity should be taken as a sign of maturity or decadence; theory is no help with such evaluations.

One very important motive for the restriction or tightening of metrical forms is the desire to avoid rhythmically misleading or distracting an audience. For example, Richard Heinze's classic study of Horace's versification concludes that Horace's normalization of anceps syllables, syllables that can be long or short, is motivated by Horace's desire to avoid the impression of iambic, trochaic, and dactylic rhythms (Heinze Die Lyrischen Verse Des Horaz 91-2). A similarly motivated restriction is that the one form prohibited in the "free" first half of the Greek glyconic is the choriamb (long, short, short, long) that comes at the close of the line (Nagy, Comparative Studies 37). So, too, twentieth century free verse practitioners at times feel themselves constrained to avoid the impression of iambic rhythm in their verse.

A very widely distributed feature of indo-european versification is greater metrical regularity at the end of a line. This is true in the Rig-Vedic meters, in dactylic hexameter, the elegiac pentameter, in most Greek lyric meters, as well as in English iambic pentameter. In the domain of metrical verse, we can say, then, that the beginning of the line tends to be more heavily marked than the close. I would be inclined to generalize this principle yet further by noting that in poems where the meter is regular, as in, say, the first stanza of Frost's "Stopping by Woods on a Snowy Evening," the stanza unit is syntactically more heavily marked in its first half and unmarked in the second:

This claim observes the fact that there are several more normal syntactic realizations of the first two lines: "Whose woods these are I think I know/His house is in the village though." vs., say, "I think I know whose woods these are/But his house is in the village." or "I think I know who owns these woods/His house is in the village." Lines three and four, however, are syntactically normalized: "He will not see me stopping here/To watch his woods fill up with snow." Just as it is normal to see the language more heavily marked with respect to rhythm and syntax early in each phase of composition, so, too, in rhyming verse it is typical to see the semantically deviant word come first, the unmarked usage second. Naturally, this is a tendency much more obvious in poets of moderate skill. Poetasters don't bother to disguise, even if they have recognized, their own verbal embarrassments, while the most skillful poets, like Frost, tend to check this inevitable tendency. It is necessary to consult the stanza rather than the line, because the stanza, though not "the" unit of composition, is certainly an internally organized phase of composition, and the idealization we call, say, iambic pentameter, has different capacities in different stanzas and in different places within a stanza. Compare, for example, the highly "regular" lines seven and eight of Yeats's ottava rima :"Live lips upon a plummet-measured face" or "Gave women dreams and dreams their looking glass." with lines one and two: "Pythagoras planned it. Why did the people stare?" or "No! Greater than Pythagoras, for the men..." While we know that it is possible for a generative metrics to give an elegant explanation of almost any metrical combination, and it may at times be useful to rise to that level of generality, what is useful about traditional metrical notions is precisely that they reference the aural expectations of an audience and thereby make intelligible an audience's response to the fulfillment or disappointment of those expectations. With respect to phonological and rhythmical marking, the technical mastery and extreme popular success of Victorian poets, especially Tennyson, Swinburne, Longfellow, Lanier, and the more astringent realizations of Hardy and Houseman, perhaps made possible a reversal of the valence of phonological and rhythmical marking among English and American poets. Abetting the change was the fact that English is syntactically rigid and rhyme poor, a language in which the opportunities presented by such a reversal would be most alluring.

The revival of interest in formal verse in recent years must contend with the difficulty that free verse has not been dominant enough for long enough to permit the wholesale reversal of valence that characterizes poetic "revolutions." Although free verse has dominated the journals and recent anthologies, it has not fully displaced traditional verse in the consciousness of a wider public, and just as the path of a Modernist revolution in English was made easier by its accommodation to the intractable realities of our language, the formalist revival is impeded by tendencies of our culture, most notably the suspicion of techne or artifice inherited from the Romantics.

A formalism is misguided that gives excessive importance to any particular mode of marking, though I do find plausible claims that phonological and rhythmical features of verse have a basis in human physiology and that such poetry exhibits an organicism not present in unmarked speech. Although poetry has no specific differentia, the poetry that survives has always been heavily marked, and the less marked has generally not survived, though profound in thought and felicitous in expression.

Since the apprehension of the domains, and, hence, the valences of a poet's language requires familiarity with works of the same period, genre, sensibility, or ideology, this sketch vindicates the collaboration of many different schools of criticism as they sift through rival claims about the factors conditioning the poet's work. The notion of marking represented here is, thus, only minimally proscriptive, insofar as it describes what already goes on among literary critics who are attentive to genre, history, culture, and especially to the text, and to the practical sorting of these domains according to their usefulness at any particular moment of a poem's interpretation.

Following Nagy, I define poetic form as that marking of performative language associated with the divergence of poetry from song. Poetic revolutions are poetic performances motivated by a realization that the valence of marked/unmarked categories has reversed within the performative domain. Some consequences of these definitions are that poetic form is a variable, not to be "pinned down" outside of certain well defined genres. Moreover, poetic revolutions or abrupt changes in the practices of poets are essentially without philosophical or ideological content. The coincidence between poetic revolutions and the emergence of certain social and political ideas must generally be treated as fortuitous.

In practical terms, one hopes that the young formalists will remember Aristotle's caution that the meter of Empedokles does not make him a poet, nor does the mixture of all meters by Chaeremon make him less of a poet. Salutary models lie before us in the work of poets like Justice and Hecht. The subtlety and formal variety of their verse appeals, in part, because it suggests a metrical domain that includes so much of our tradition, not excluding the contemporary. And what is true of the formal characteristics of their verse is also true of its ethical and psychological dimensions. If a reader takes these remarks to justify a certain aloofness from literary theory, from literary fashions, and especially from the ideological commitments which are often the hallmark of literary movements, then he has gotten my meaning exactly. While we may continue to think that great changes in poetic practices are forged by the mighty, individual geniuses of a tradition, and it is nice to think so, in any event, poetic revolutions, from this point of view, are cognitively inevitable. Like linguistic innovation, poetic innovations spread from the cultural or linguistic center, which is not a geographical center, but whatever corner of the planet holds the most cultivated minds, where the words, topics, and rhetorical practices of the period are most thoroughly familiar, hence least able to succeed in the performative roles that marked language plays, least able to succeed, that is, as poetry.

LINGUISTIC MARKING AND THE NEW FORMALISM

Poetry

Fiction

Essays

Reviews

~~~