Read an Excerpt
The Princeton Handbook of Poetic Terms
By Roland Greene, Stephen Cushman PRINCETON UNIVERSITY PRESS
Copyright © 2016 Princeton University Press
All rights reserved.
ISBN: 978-1-4008-8064-5
CHAPTER 1
A
ACCENT. In Eng., accent is the auditory prominence perceived in one syllable as compared with others in its vicinity. Accent and stress are often treated as synonymous, though some literary scholars and linguists distinguish the two terms according to a variety of criteria. Disagreements persist about the source and acoustical nature of syllabic prominence — loudness, volume, pitch, duration, or some combination of factors — but they are arguably of peripheral relevance to the understanding of accent within Eng. poetics.
The phenomena of accent vary among langs. and the poetics associated with them. The Eng. lexical contrast between convict as noun and as verb has no parallel in Fr. (Sp. resembles Eng. in this regard, while Finnish resembles Fr.) For Fr. speakers, stress contours are perceived on the level of the phrase or clause, and learning Eng. entails acquiring the ability to hear contrastive accent in words, just as a Japanese speaker learning Eng. must acquire the distinction between the liquids l and r. A consequence is that, while Fr. meters count only syllables, Eng. meters conventionally also govern the number and distribution of accents.
In Eng. speech, accent operates in various ways on scales from the word (convict) through the sentence. As the units grow larger, accent becomes increasingly available to choice and conscious use for rhetorical emphasis. One step beyond the accents recorded in dicts. is the difference between "Spanish teacher" as a compound (a person who teaches Sp.) and as a phrase (a teacher from Spain). Eng. phonology enjoins stronger accent on "Spanish" in the compound and "teacher" in the phrase.
These lexical accents and differences in accent between compounds and phrases are "hardwired" into the Eng. lang. Beyond those, speakers exercise more deliberate choice when they employ contrasting accent to create rhetorical or logical emphases that are intimately entwined with semantic context. In the opposition Chicago White Sox vs. Chicago Cubs, it is the variable rather than the fixed element that receives the accent. Consequently, the question "Are you a fan of the Chicago Cubs?" accords with what we know about the world of baseball, while "Are you a fan of the Chicago Cubs?" implies a Cubs team from some other city. This kind of contrastive stress, so dynamic in Eng. speech, also plays a variety of important roles in the poetic manipulation of lang., perhaps esp. in how written poetry contrives to convey the rhetorical and intonational contours of speech. When a line break, for instance, encourages the reader to place an accent on some word where it would not normally be expected, the emphasis may suggest an unanticipated logical contrast. This foregrounding of accent may have rhetorical implications: "The art of losing isn't hard to master; / so many things seem filled with the intent / to be lost that their loss is no disaster" (emphasis inferred; Elizabeth Bishop, "One Art"); "The sound of horns and motors, which shall bring / Sweeney to Mrs. Porter [not Actaeon to Diana] in the spring" (T. S. Eliot, The Waste Land).
Within the specific realm of traditional Eng. metrical verse, words are treated as bearing an accent if they are short polysyllables (whose stress can be looked up in a dict.) or monosyllables that belong to an open class (noun, verb, adjective, adverb, interjection). Other syllables tend to be unstressed. Yet several factors can alter this perception. One is the kind of rhetorical force created by contrastive stress, esp. in the volatile case of pronouns. Another, more pervasive influence arises from the complex interaction between the abstract, narrowly constrained pattern of meter and the concrete, highly contingent rhythm of the spoken words. This fundamental distinction — meter and rhythm are related similarly to "the human face" and "a human's face" — crucially conditions how we perceive accent; it accounts for some difficulties that an unpracticed reader of metrical verse, though a native of Eng. speech, may have in locating the accents in a line.
Some of the confusion surrounding the term may be reduced if we recognize that accent names phenomena on two different levels of abstraction, the acoustical and the metrical. There is an analogy with phonemes. Speakers of Eng. unconsciously insert a puff of air after the p in pan, but not in span. The difference can be detected by using acoustic instruments or by holding a palm in front of the mouth, yet is not detected by speakers in the absence of exceptional attention. The p in both cases represents the same phoneme, the same distinctive feature in the Eng. phonetic system — a system that does not merely divide the continuous acoustic stream of speech but abstracts from it a small set of three or four dozen discrete items. Similarly, various acoustical phenomena (pitch, loudness, etc.) give rise to an indefinitely large number of degrees and perhaps even kinds of accent; yet within a metrical context, the accustomed reader — analogous to a native speaker — reduces this continuum to an abstraction of (usually) two opposed values, stressed vs. unstressed. (The analogy fails to capture how the reader is simultaneously aware of a continuum of stress weights in the speech rhythm and a binary feature in the metrical pattern, both embodied in a single set of words.)
Differences of accent between compounds and phrases, or introduced for the sake of rhetorical contrast, which operate prominently within the larger manifold of rhythm, make no difference on the level of meter. "Spanish teacher" in either sense would be scanned as two trochees, and the stronger stress on one word or the other has no specifically metrical effect. The four degrees of stress adopted by Chatman and others from Trager and Smith, while useful in the phonological analysis of Eng. and in the poetic understanding of rhythm, are unnecessary in the specifically metrical treatment of accent. The "four levels" represent an intermediate abstraction, as does the more traditional compromise of secondary stress or the hovering accent of Brooks and Warren. "Trager and Smith ... demonstrated that stress and pitch are much more complex and variable phenomena than could be accounted for by the binary unstress-stress relation of traditional prosody" (Bradford 1994) — but this important truth should not mislead us into trying to weld speech rhythm and metrical pattern into an unwieldy whole, rather than hearing their interplay.
Readers are sensitive to a far wider range of rhythmic phenomena in poetry than those that are encoded within a metrical system. The nuanced stress patterns of speech, though they are foregrounded in nonmetrical or free verse, do not disappear from the reader's awareness in metrical verse with its two-valued feature of accent. Rather, the give and take between the claims of meter and rhythm become a major source of auditory richness. Syllables may be heard as stressed either because of their prominence in speech or because of their position within the metrical line.
Any of the kinds of speech accent — lexical, phrasal, rhetorical — may coincide with a stressed position within the metrical line (as in the even-numbered positions within an iambic pentameter); or the speech and metrical accents may be momentarily out of phase. Within the accentual-syllabic system of Eng. metrics, these possibilities give rise to a repertoire of more or less common or striking variations. When speech accents occur in metrically unstressed positions, they give rise to metrical substitutions of one foot for another, such as the trochee or the spondee for the iamb:
Singest of summer in full-throated ease
When metrical accents occur where no speech accent is available to embody them, the syllable receives "promoted" stress. The conjunction in the middle of W. B. Yeats's line, "We loved each other and were ignorant," which might pass unstressed in speech, exhibits this kind of promoted accent. It may render the verse line different from and semantically richer than its speech equivalent. The metrical expectation of accent in this position in the line is presumably the initial cause of the promotion; whether the rhetorical point — that love and ignorance are not at odds as one might think, but inextricable — is an effect or another kind of cause would be difficult to decide.
The phonological and metrical understandings of accent can sometimes even be directly at odds. In a compound word like townsman, the second syllable is not unstressed (its vowel is not reduced to schwa). Phonologically, then, the syllable sequence "townsman of" presents three descending levels of stress. In A. E. Housman's line, however, "Townsman of a stiller town," the reader hears "of" with an accent created or promoted by the underlying metrical pattern of iambic tetrameter; and in comparison, the syllable "-man" is heard as unstressed. The case is complicated by the copresence of other details: because the line is headless, e.g., we know not to scan the initial compound word as a spondee only once we get the following syllables ("a still-"); the unambiguous accent on the last of those syllables (confirmed by the final alternation, "-er town") anchors the whole iambic matrix and retrospectively clarifies the metrical role of "Townsman of." Complications of this kind are typical in the interaction between metrical pattern and speech rhythms and constitute a primary reason for apparent ambiguities of accent in lines of Eng. verse.
See DEMOTION.
G. L. Trager and H. L. Smith Jr., An Outline of English Structure (1951); W. K. Wimsatt and M. C. Beardsley, "The Concept of Meter: An Exercise in Abstraction," PMLA 74 (1959); Brooks and Warren; Chatman, ch. 3, 4, appendix; N. Chomsky and M. Halle, The Sound Pattern of English (1968); M. Halle and S. J. Keyser, English Stress (1971); R. Vanderslice and P. Ladefoged, "Binary Suprasegmental Features and Transformational Word-Accentuation Rules," Lang 48 (1972); P. Kiparsky, "Stress, Syntax, and Meter," Lang 51 (1975); M. Liberman and A. S. Prince, "On Stress and Linguistic Rhythm," LingI 8 (1977); E. O. Selkirk, Phonology and Syntax (1984); B. Hayes, "The Prosodic Hierarchy in Meter," Phonetics and Phonology, ed. P. Kiparsky and G. Youmans (1989); R. Bradford, Roman Jakobson (1994).
C. O. HARTMAN
ACCENTUAL-SYLLABIC VERSE. In Eng. poetry that is not written in free verse, the most common and traditional metrical system is called "accentual-syllabic" because it combines a count of syllables per line with rules for the number and position of accents in the line.
Metrical systems in different langs. measure lines by various linguistic elements and combinations of elements. Some systems are based on counting a single kind of element (syllables in Fr., monosyllabic words in Chinese). Others — such as the Lat. quantitative meters and the accentual-syllabic system used in mod. Eng., mod. Gr., Ger., and many other langs. — coordinate two different measures, such as syllables and their length or syllables and their stress. Cl. theorists provided a method of analyzing these meters by dividing them into feet, small units defined by various permutations of the two kinds of elements. Thus, the iamb orders a slack syllable and a stressed one (in Eng.) or a short syllable and a long one (in Lat.), while the trochee reverses the iamb's order and the anapest extends the iamb by doubling the initial weak syllable position.
Two-element meters tend to maintain the stability of the lines' measure without as much rigidity in adherence to the meter's defining rules as single-element meters require. Consequently, much of the richness of variation in a two-element meter such as the accentual-syllabic can be captured at least crudely by analysis of substituted feet, metrical inversions, promoted stresses, and similar concepts that build on the notion of the foot. This variety can be sketched or roughly diagrammed by scansion of the lines.
Particular meters in such a two-element system are conventionally named by combining an adjectival form of the name of the base foot (iambic, dactylic, trochaic, etc.) with a noun made of a Lat. number word plus "-meter": trimeter, tetrameter, pentameter, hexameter, and so on.
See DEMOTION, PROSODY.
C. O. HARTMAN
ACCENTUAL VERSE. Verse organized by count of stresses, not by count of syllables. Many prosodists of the 18th, 19th, and early 20th cs. looked upon most med. verse, a large amount of Ren. verse, and all popular verse down to the present as loose, rough, or irregular in number of syllables and in placement of stresses; from this assumption (not demonstration), they concluded that regulated count of stresses was the only criterion of the meter. Schipper, e.g., effectually views most ME and mod. Eng. four-stress verse as descended from OE alliterative verse. But in this, he misconceived the nature of OE prosody. Other prosodists, too, have lumped together verse of very different rhythmic textures drawn from widely different social registers and textual contexts, treating all of them under the general rubric of accentual verse. This generalization masks differences that should probably be characterized in metrical (and not merely more broadly rhythmic) terms. Ideally, we would isolate the similarities among species of accentual verse and then identify either features or gradations in strictness of form that differentiate them. No clear theoretical foundation has been established to accomplish this.
Several varieties of accentual verse have been proposed in the Western langs.: (1) folk verse as opposed to art verse, i.e., the large class of popular (e.g., greeting card) verse, nursery rhymes, college cheers and chants, slogans, logos, and jingles — both Malof and Attridge rightly insist on the centrality of the four-stress line here; (2) ballad and hymn meter, specifically the meter of the Eng. and Scottish popular ballads and of the metrical psalters in the Sternhold-Hopkins line; (3) popular song — an extremely large class; (4) literary imitations of genuine ballad meter such as the Christabel meter; (5) genuine oral poetry, which indeed seems to show a fixed number of stresses per line but, in fact, is constructed by lexico-metrical formulaic phrases; (6) simple doggerel, i.e., lines that hardly scan at all except for stress count, whether because of authorial ineptitude, scribal misprision, textual corruption, or reader misperception — there are many scraps of late med. verse that seem to be so; (7) literary verse (often stichic) that is less regular than accentual-syllabic principles would demand but clearly not entirely free, e.g., the four-stress lines that Helen Gardner has pointed out in T. S. Eliot's Four Quartets; (8) Ger. knittelvers, both in a freer, late med. variety subsequently revived for literary and dramatic purposes by J. W. Goethe and Bertholt Brecht, and in a stricter, 16th-c. variety (Hans-Sachs verse) in octosyllabic couplets; and (9) Rus. dol'nik verse, a 20th-c. meter popularized by Alexander Blok, mainly in three-stress lines (interestingly, this form devolved from literary verse, not folk verse as in Eng. and Ger.). In all the preceding, there has been an assumption that accentual verse is isoaccentual; if one defines it more broadly (organized only on stresses but not always the same number per line), one would then admit Ger. freie Rhythmen and freie Verse and possibly Fr. 19th-c. vers libéré. But these verge on free verse.
When Robert Bridges, the Eng. poet and prosodist, studied accentual verse at the turn of the 20th c., he believed he discovered a paradox in claims that accentual verse works by counting the natural stresses in the line. For example, despite S. T. Coleridge's claims that "Christabel" is in a "new" meter and that every line in it will be found to have exactly four stresses, the poem actually contains a number of problematic lines, like "How drowsily it crew," which cannot by any reasonable standard carry four natural accents. Of this line Bridges remarks: "In stress-verse this line can have only two accents ... but judging from other lines in the poem, it was almost certainly meant to have three, and if so, the second of these is a conventional accent; it does not occur in the speech but in the metre, and has to be imagined because the metre suggests or requires it; and it is plain that if the stress is to be the rule of the metre, the metre cannot be called on to provide the stress" (italics added). For Bridges, the definition of true "accentual verse" is that it operates on only two principles: "the stress governs the rhythm" and "the stresses must all be true speech-stresses" (Bridges set these sentences in all capitals). This two-part definition, he asserts, strictly distinguishes accentual verse from accentual-syllabic verse, in which it is the function of the meter to establish and preserve in the mind's ear a paradigm, an abstract pattern, such that, if the line itself does not supply the requisite number of accents, the pattern shall supply them mentally. But Bridges's view is complicated by the dominance of accentual-syllabism in so much mod. Eng. verse that poets and readers may hear promoted stresses even in a mostly accentual context. In many poems by Elizabeth Bishop, such as "The Fish" (three-stress), e.g., the near regularity of accent counts encourages us to fill out some shorter lines with stresses that, while strong relative to surrounding syllables, would not be heard in the speech rhythm of the line; analyzing the verse as either strictly accentual or as consistently but roughly accentual-syllabic requires acknowledging exceptions, but calling the result "free verse" ignores important gestures toward regularity.
In Eng. as in other Teutonic langs., accentual verse has some claim to being fundamental; it is "the simplest, oldest, and most natural poetic measure in English" (Gioia). It lies near the root of poetry, as it were, and near the point where poetry and music diverge (and reconverge in song). It comes closer than accentual-syllabic meters to manifesting the beat that defines the term meter in the musical sense of that word. Perhaps partly for this reason, versions of accentual verse (measuring lines in four, three, or even two stresses) have been popular alongside mod. free verse, sometimes only notionally distinct from it.
(Continues...)
Excerpted from The Princeton Handbook of Poetic Terms by Roland Greene, Stephen Cushman. Copyright © 2016 Princeton University Press. Excerpted by permission of PRINCETON UNIVERSITY PRESS.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
