119 92 3MB
English Pages 240 [238] Year 2008
# : % & 4 * ( /
#:%&4*(/ * / 5 & / 5 * 0 / * / 1 0 & 5 3 : " / / & ' & 3 3 :
4UBOGPSE6OJWFSTJUZ1SFTT
4UBOGPSE $BMJGPSOJB
Stanford University Press Stanford, California © by the Board of Trustees of the Leland Stanford Junior University. All rights reserved. No part of this book may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying and recording, or in any information storage or retrieval system without the prior written permission of Stanford University Press. Printed in the United States of America on acid-free, archival-quality paper Library of Congress Cataloging-in-Publication Data Ferry, Anne. By design : intention in poetry / Anne Ferry. p. cm. Includes bibliographical references and index. >H7C ---- (alk. paper) . English poetry—History and criticism. . Intention in literature. . Selfconsciousness (Awareness) in literature. . Poetics—History. I. Title. EG >7 '.—dc
Previously published material includes 8=6EI:G “Love Rhymes with Of,” Modernism/ Modernity , no. (): –. © The Johns Hopkins University Press. Reprinted by permission of The Johns Hopkins University Press. 8=6EI:G “The Sense of Rhyme,” Literary Imagination , no. (): –. © Association of Literary Scholars and Critics. Reprinted by permission of the Oxford University Press. 8=6EI:G “Titles in George Herbert’s ‘Little Book,’ ” English Literary Renaissance , no. (Spring ): –. © English Literary Renaissance Inc. Reprinted by permission of Wiley-Blackwell Publishing Ltd. 8=6EI:G “Revisions of Visions: Wordsworth and His Inheritors,” Raritan , no. (), pp. –. © by Raritan: A Quarterly Review. Reprinted by permission of Raritan, Rutgers University. 8=6EI:G “Frost’s ‘Obvious’ Titles,” chap. in Reading in an Age of Theory, ed. Bridget Gellert Lyons. © Rutgers University Press. Reprinted by permission of the Rutgers University Press. 8=6EI:G “Frost’s Design,” Literary Imagination , no. (): –. © Association of Literary Scholars and Critics. Reprinted by permission of the Oxford University Press. 8=6EI:G “Anonymity: The Literary History of a Word,” New Literary History (): –. © The Johns Hopkins University Press. Reprinted by permission of The Johns Hopkins University Press. Designed by Bruce Lundquist Typeset at Stanford University Press in / Minion
For her students
$0/5&/54
Introduction Love Rhymes with Of
The Sense of Rhyme: Sidney and Shakespeare
Titles in George Herbert’s “Little Book”
Thing and Things: Wordsworth, Stevens, Ashbery
Optical Illusions: Wordsworth and His Inheritors
William Carlos Williams’s Redesigning
Frost’s “Obvious” Titles
Frost’s Design
Anonymity: The Literary History of a Word
Notes
Index
*/530 % 6$ 5 * 0 / A plan or scheme conceived in the mind and intended for subsequent execution . . . Purpose, aim, intention . . . Contrivance in accordance with a preconceived plan: adaptation of means to ends; pre-arranged purpose . . . . . . by . . . design : on purpose, purposely, intentionally . . . A plan in art . . . after which the actual structure or texture is to be completed; a delineation, a pattern. . . . The artistic idea as executed . . . “Design,” Oxford English Dictionary
in the workings of conscious intention, for which the evidences are elements of design. Conscious intention is there in almost every definition of the term design. The book intends to demonstrate, through a series of examples, how the ever-changing vitality of literary history, of poems acting within the constraints, pressures, and urgings of history, is demonstrated by particularly self-aware, critically adept activities of poets: in significant features of the design of poems in their books, in the design of the books themselves, and in the relations of the poets to, their designs on, their targeted audiences. The examples range from the sixteenth century to the twentieth. In every instance, as I hope will be shown, something startlingly and significantly new has taken place, new things happening in the detailed construction of new poems because of the high degree of conscious awareness and self-awareness of these artists at work in successive historical phases, reading one another, exploring
5 )*4 # 0 0 , *4 "4 5 6 %:
*/530%6$5*0/
one another’s vocabularies and ways of working, reinterpreting one another, and in some cases themselves, in order to carry out their own designs. These chapters attempt the sharpness of focus that results from concentration, in each case, on single exemplary features of the work the designing artists in question are doing: changing practices of rhyming; the titling of poems; the changing uses of a key word, thing; the entrance into literary history of another key word, and concept, anonymity, and the ways it entered into the design of certain important poems of the modernist period; the occurrence of a new figure, “optical illusion,” as it enters into poetry of the Romantic period, and after; the self-presentation of William Carlos Williams, in the design of his books, by the deliberate juxtaposition, often side by side, of different versions of the same poem; one poem, Frost’s “Design,” as a complex instance of his self-presentation, his attitude toward the design of his books, and of his career. The first chapter, “Love Rhymes with Of,” is appropriately first in the book, because it looks at a series of new events in rhyming practices across the periods, from the sixteenth century to the present. The chapter begins with the sixteenth-century ideal of harmony and decorum, rhyme pairings alike in grammar as in sound, and alike also in harmonizing meanings, words with like dignity, belonging to the same or similar lexical levels. These preferred forms characterize English rhyming practices, though in widely varying ways, from the mid-sixteenth century through the reign of Alexander Pope, when poets, otherwise different in their designs, began sometimes to experiment with surprisingly combined rhyme-sounds: Blake’s tomb–comb, Keats’s patiently–sigh, and Longfellow’s wood–solitude are examples. Experimenting with new modes of rhyming culminated in the radical rhyming of Marianne Moore, whose iconoclastic rhymes bring into sharp focus the unbridgeable gap between words and the phenomena that they point to as if at random, her rhymes decidedly not echoing some universal harmony. The second chapter, “The Sense of Rhyme: Sidney and Shakespeare,” is a study of how Shakespeare’s rhyming radically departs from Sir Philip Sidney’s practice, where characteristically the most active function of the end rhymes is not their signifying capacity but the usefulness of their phonetic effects for the definitions of relations among rhythm, meter, and syntax. The relative freedom from connotation of most of Sidney’s words at line ends allows their agreement in sound to be heard with special distinctness. Shakespeare, who learned so much from Sidney, nevertheless works very differently, where, in
*/530%6$5*0/
sonnet after sonnet, the end words of rhymed lines are charged with the dramatic energy of semantic as well as phonic relationship. This is to be seen most plainly in those sonnets of Sidney and those of Shakespeare that are most closely comparable in other ways, and where it is most clear that Shakespeare was an intent and opportunistic reader of his mentoring poet. Finally, a consideration of likeness between Shakespeare’s working with end words in his sonnets and his working with end words in a passage of blank verse from Hamlet reveals another astonishing dimension of his dramatic designs. The third chapter, “Titles in George Herbert’s ‘Little Book,’ ” explores how George Herbert, almost (after Ben Jonson) the first author to be responsible for his own titling of poems, seized the opportunity to make a book, and how the deliberate associations of Herbert’s poems with collections of religious commonplace books and the Psalms as a “harmony of holy passions” were suggested by his carefully designed and consistently followed principles of titling. By submitting his poems to that self-imposed discipline, he achieved for them a freedom to be at once anonymous and intensely personal, detached from self and yet autobiographical, representative but private, gathering particulars into simplicity. These paradoxes are something understood in the poems. The likening of his collection to a “harmony” for use in private spiritual exercises may also have granted him another kind of freedom. It may have made it possible for him to imagine arranging his poems in a “little Book” that might one day be “made publick,” a plan that would otherwise most likely have been distasteful to a poet who struggled for a language untangled from self and from human invention. Such preparation of his own poems in a volume for publication would have been unacceptable unless, perhaps, he could think “it may turn to the advantage of any dejected poor Soul” who would use it as a devotional commonplace book modeled on the Psalter. The fourth and fifth chapters are explorations of poetry associated most directly with Wordsworth and his inheritors. “Thing and Things: Wordsworth, Stevens, Ashbery” considers the way, for some poets, the chameleon nature of the word thing has served their peculiar conceptual and imaginative needs, especially those of Wordsworth, followed by those of Wallace Stevens and John Ashbery. To begin with, there is the simple, elemental immediacy the words share with, say, bird and stone. The word’s immediacy gives thing the suggestion of an embodied presence like bird and stone: it can stand for either. But in its long linguistic history, thing has also been commonly understood to signify what we can only conceptually imagine, believe in; what is not confirmable by
*/530%6$5*0/
the senses; not utterable in the logic of grammar: invisible things. The noun thing and its plural form things are dominant in the personal idioms of these poets when they have tried to distance their language, in radically different degrees and ways, from the vocabulary of transcendence used most grandly by Milton. Thing is a key word in Wordsworth’s visionary vocabulary. He discovered in this noun a simultaneity of meanings suggesting mysterious power that might make it a space for the meeting if not the deep interfusion of external nature with the eternal, but his use of it expressed his sense of the baffling tenuousness of that meeting. The chapter studies the virtuosic skepticism of Wallace Stevens in his obsessive play with the word thing, and beyond that, how John Ashbery used this paradox of a simultaneously overloaded and empty linguistic figure as a device for expressing the mysterious coexistence and disconnection of transcendent things and everyday things. In “Optical Illusions: Wordsworth and His Inheritors” the cultural situation that pertains directly to the comparatively microscopic focus of the discussion can be summarized in this combination of circumstances: that some implicit uneasiness about supernatural visions as matter for poems of dominant kinds, particularly lyric and descriptive, coincided not coincidentally with the spread of scientific and philosophical theories about the nature of bodily vision; that these theories by the early eighteenth century began to interest a wide audience, including poets who made very specific use of them in their poems; that those poems, in turn, directly prepared the ground for the work of later poets where questioning, undermining, denying, or ignoring the possibility of supernatural visions, or compensating for them, is the explicit focus of the poem. A consequence of this cultural situation is the figure or described circumstance where an entirely natural event occurs, as if in a supernatural vision, completely explicable but not immediately perceived as being so. Such figures or situations, under Wordsworth’s potent influence, have become a convention that has informed, entered into the design of, much of the best work of such later poets as Stevens, Eliot, Frost, Elizabeth Bishop, James Merrill, and Seamus Heaney. Four closing chapters are concerned with what is usually called modernist poetry. Three of them concentrate on single poets; one is, through the study of a single word, anonymity, and the concepts associated with it, a more expansive history of modernism, which reached its climax in the early poems of Eliot. “William Carlos Williams’s Redesigning” is a study of what was his simplest and most immediate way of making his intentions discernible, by show-
*/530%6$5*0/
ing his poems in juxtaposed alternate states. He does this to make readers perceive his activities as a writer as cases of process rather than fixity and to present himself as simultaneously a hyperrealist poet and a linguistic experimenter. In a number of significant instances he published alternative versions of the same poem, or the same material, side by side, following, perhaps, the example of painters with whom he was associated, abstracting the concrete descriptive terms from a realist poem in order to produce a new, made, independent object, concrete in an utterly different sense. Many of the revisions in this and other ways constitute radical assaults on the poems they are revising, designing “new forms” (new associations of the elemental particles), “breaking down everything” “to get at the essential,” “escaping” “forms of the poem.” “Frost’s ‘Obvious’ Titles” is a further inquiry into the titling of poems as elements of the poems’ designs. It is the seeming straightforwardness of his titles that most immediately divides them from a quintessentially modernist poem, which assaults the reader with a title that predicts the poem’s obscurity by its own. But the chapter demonstrates how the appearance his poems have of being accessible is itself a form of ulteriority: “A poem would be no good that hadn’t doors. I wouldn’t leave them open though.”1 “Frost’s Design” is a study of revision and self-presentation that explores the poet’s extraordinary treatment of the sonnet first called “In White,” which he promptly but radically revised yet did not allow in any of his volumes of poetry for twenty-four years. The chapter is a consideration, first, of reasons for the detailed revisions that produced the great dark poem “Design.” Second, it is a consideration of reasons for the delay in including it in a book, Frost’s own cautious sense of how to present himself to his audience, and his own acknowledged anxieties about the attitudes expressed in such a poem. Third, it considers the manner and effect of its placement in the book in which it finally appeared, as an instance of Frost’s bookmaking. And fourth, it examines the relation of such a poem as this to the difficult dark works of the modernists around him, and Frost’s response to the redefining of his reputation, at a late stage of his career, by such critics as Randall Jarrell and Lionel Trilling. The chapter “Anonymity: The Literary History of a Word” considers the origin and history of the noun anonymity as a model of the sorts of energetic transactions that task place over time among words, poems, other writings, and the pressures in the culture that produces and is produced by them. The line of argument in the chapter traces the literary history of a noun—together
*/530%6$5*0/
with its parent adjective—that has by now become so packed with presuppositions and preoccupations that we can often sense the structure of feelings associated with it to be present even in contexts where the word is not explicitly used. Poems are the spaces where this model can best be demonstrated, because in them the accumulated force of its concentration of meanings can be felt most powerfully. The chapter brings the story of this word, first occurring in English in the late sixteenth century, up to the s, when it attached itself to the aesthetic of impersonality that sought to distance the personality of the author from the work; simultaneously anonymity became the name for the misery of mass urban living. Eliot’s modernist poems amalgamated these two distinct meanings as an aesthetic ideal and as a cultural condition. In this chapter, and in the others in this book, I have been studying the sense of themselves and of their work, of poets who knew, or were engaged in finding out, who they were and how they were situated in their own locations in history and in relation to other poets as well as to themselves; and I have been studying how their self-knowledge or their self-discovery enters into their designs on their audience, the designing of their books, and of the poems within their books.
$IBQUFS
-0 7& 3 ): . & 4 8 * 5 ) 0 '
“Love Rhymes with Shove,” it would probably make us smile. One reason would be that the second rhyme word is unexpected: we would be likely to predict a more conventional, decorous choice of echo for love in dove or above. At the same time, love–shove is a familiar type of bathetic rhyme designed to raise a smile in all sorts of humorous poetry, where absurd pairings of like-sounding words are the traditional vehicle for expressing what W. H. Auden called “comic contradiction.”1 By contrast, love–of does not belong to a category of rhyme we would find in poetry we are typically used to. Such combinations virtually never appear in English poetry except in rare humorous verses until the nineteenth century, and even then they were seldom used, and only for comic effect or for a formal reason such as to encourage enjambment. Rhymes like love–of have signifying possibilities that were first taken seriously in the early twentieth century because they satisfied the intentions of the modernist aesthetic. The premise of this discussion, that “Love rhymes with of,” works in unique ways. The statement makes us laugh, not only because we had not thought of the rhyme before but because the paired words are mismatched in a different, unfamiliar, and unsettling way from either love–dove or love–shove. Although the statement is instantly recognizable as true, it is true in a way hard to know how to think about; it is outside our everyday patterns of thought and language. How it comes to have these incongruent effects is both a conceptual and a historical question that this chapter will explore. *' 5 )& 5 * 5 - & 0 ' 5 )*4 $ ) " 1 5 &3 8 &3 &
-07&3):.&48*5)0'
One clear signal the title gives is that what follows will be about poetry and not prose, even though “Love rhymes with of ” is a prose sentence that uses a rhyme to prove its statement. We tend to associate rhyme with poetry as distinct from prose because we do not, except in extraordinary instances, hear or think of like-sounding words in prose speech or writing as rhymes: for instance, in familiar phrases like the love of God or the god of love. The reason for this difference in response is that prose is not systematically arranged to emphasize similarity of final sounds by their accent and syntactical position, the way almost all rhymed poetry in English was structured roughly from Chaucer to Tennyson. Simply, rhyme in English poetry—described by T. V. F. Brogan as “historically its single most consistent characteristic”—is the bringing together into related positions in the verse of words that are partially like in sound but differ in meaning.2 It is a specialized form of repetition since the linked words are not identical. Linguists describe the relationship between words that rhyme by various terms, among them consequence, consonance, convergence, equivalence, parallelism, symmetry. Rhyme works rhetorically like a simile, because both kinds of verbal patterning make a comparison in which perceived likeness at the same time calls attention to difference. In the paradigmatic instance of simile, the beloved in Shakespeare’s Sonnet is like a summer’s day and also is not. Structurally, a rhyme word placed at the end of the line works to mark off the line as a unit while binding together different lines by their like end sound. The prominence granted by their final position strengthens the double power of such rhymes to liken and distinguish, as does the traditional practice, recommended in theory until the nineteenth century, of having the metrical accent fall on the final syllable. End rhymes, empowered by their position to engage the reader most actively in recognizing both likeness and difference, are the subject of this discussion. “It is in the nature of rhyme to bring differences together and to reveal the differences in likeness,” according to Jurij Lotman’s definition, 3 but the balance of the comparing and distinguishing effects of rhyme can tilt in either direction between pairs of words, between whole poems, or between phases of an author’s work. The greater emphasis of poets on the closeness or distance between rhyme pairs has varied within the same period—particularly with respect to different genres—and from one period to another. Even so, there are signs that there has been a growing interest among poets writing in English
-07&3):.&48*5)0'
over more than five centuries in widening the phonetic, structural, semantic, and grammatical gaps between words linked by resemblance of sound. This discussion proposes that the most radical phase of this history, the practice revealing the most fundamental transformation of attitudes toward the ordering of thought in the language of rhymed poetry, is the acceptance by poets in the twentieth century of rhymes like love–of. Their common appearance is witness to wide recognition of truth in the paradoxical proposition that there are legitimate conceptual grounds for such agrammatical rhymes in spite of their inherent conceptual incompatibility. Every reformation in English poetry has involved shifts in attitudes toward rhyming, in the practices of it, and in the rules for its proper conduct. A starting point for looking at this slow and unsteady movement is in the sixteenth century. Then English writers—some of them theorists as well as practitioners of rhyme—first explored the state of their language as it was emerging from what Sir Philip Sidney called “that mistie time” of the greatest poet England had yet boasted, Chaucer.4 The challenge these writers faced was to find support for their argument that poetry in English could achieve its own excellence without the inflectional grammar that had supported the unrhymed system of ancient verse and had left traces in the rhymed poetry of Chaucer and his followers. Chaucer himself had long ago complained of a “skarsete” of rhymes in the English he had to work with.5 Inevitably, the sixteenth-century defenders of rhyme turned to the new mode of endings that George Puttenham, in The Arte of English Poetrie, called “the cheife grace of our vulgar Poesie,” which “consisteth in the Symphonie,” a synonym for rhyme that meant at once likeness of sound and metrical accent in particular, agreement or harmony in general.6 The “most auncient” vocabulary from “our naturall Saxon English” with its wealth of monosyllables supplied a high proportion of words with like sounds that could substitute for inflectional endings.7 They were preferred for end rhymes over words of more than one syllable because, as Puttenham said, the “monosillables of our English Saxons . . . do naturally and indifferently receive any accent.”8 This meant that at the end of the line they accommodated themselves inevitably to the pattern of masculine rhyme, the norm accepted in English poetry for most of its history. Rhyme, scorned by classicists as new-fangled ornamentation—Thomas Campion called it “childish titillation”—was defended on the grounds of its fidelity to the structure of the English language, and to its “most auncient English” vocabulary.9 George Gascoigne, in the earliest treatise on the making of
-07&3):.&48*5)0'
English verse in rhyme, instructed poets that “the more monasyllables that you vse the truer Englishman you shall seeme.”10 Even though the argument for the naturalness of rhyme was made for its expedience, it was predicated on ways of thinking so widely and unquestioningly taken for granted, it seems, that there was no need to bring them to the surface in these earliest defenses of rhyme. When Elizabethan writers recommended specifically monosyllables handed down from “naturall Saxon English,” they were essentially referring, without having to say so, to nouns. Nouns made up a vast proportion of the prized monosyllabic end rhymes, verbs a lesser share, adjectives (still considered, as in Latin, to be a species of noun) and adverbs together making up a very much smaller number than verbs.11 Since the position of end rhymes gives them a special prominence, their place would belong naturally to nouns and verbs, which are the parts of speech that chiefly determine the structure and carry the burden of meaning in a sentence, where for those reasons they are normally stressed. Nouns that name what we can see and touch are the first words we are taught as infants. They are the grammatical guides that help us most to get along in the world, as anyone knows who has traveled in a foreign country with only the most rudimentary knowledge of its language. Nouns are the first category to be explained in any treatise on English grammar, reflecting a hierarchical ordering of parts of speech that goes back to antiquity. In the earliest English rhyming dictionary of the only classes included are, in this order, nouns, adjectives, verbs, adverbs.12 Nouns and verbs of Saxon origin in particular tend to be the names for common things and events that make up ordinary life. (Many foreign nouns were imported in the sixteenth century to supply English with a vocabulary of terms for abstract concepts it had no words for.) Because these things and events are part of everyday experience, the nouns and verbs for expressing them were themselves recognizable realities. Their solid familiarity gave them precedence over less essential words belonging to other parts of speech, so that the hierarchical structure of English grammar supported the use in poetry of rhymes that contribute most to the making of “order and forme,” which Samuel Daniel in A Defence of Ryme claimed to be “pleasing to Nature.”13 The monosyllables paired in rhymes usually belonged to the same part of speech, so they had the same grammatical meaning. That agreement meant they filled the same function in structuring the mind’s understanding of order in the world. In sixteenth-century terms, pairing like grammatical
-07&3):.&48*5)0'
parts in rhyme induced the “certaintie” and “clozes” that Daniel believed “the diuine power of the spirit” imposed on the poet’s imagination to bring about harmony.14 Grammatical agreement was an important instrument in the preferred aesthetic of rhyming, in concert with other agreements in sound, in syllabic structure, in metrical accent, and in closeness of lexical meaning, or what Jakobson called semantic propinquity (using dove–love as an example).15 Sequences that bind their member rhymes together by their likeness in multiple features have threaded their recognizable way through English poetry at least since the sixteenth century: God–rod, breath–death, womb–tomb, lust– dust, tears–fears, sleep–keep, name–fame, prime–time, rhyme–chime, trees– breeze, heart– (in the sixteenth century often spelled hart) part–smart–dart. The series love–prove–move, pronounced alike at least into the eighteenth century (in poems by Pope, Gay, and Prior among others) has continued to be a useful formula, perhaps because historically love has been pointed out as a famously difficult word to find rhymes for.16 As recently as , G. S. Fraser used it as his example in a discussion of the poverty of rhymes in English by comparison with languages rich in words with like endings: “Amore in Italian rhymes with cuore, but love in English perfectly only with the undignified word shove or the trivial word glove” (a conservative statement, we shall see, in the light of practice in twentieth-century poetry).17 The first stanza of “Gascoignes good morrow” shows the focused interest and expertise of sixteenth- and earlier seventeenth-century poets in bonding like-sounding words by harmony among their various features, to create verses that might be perceived as reflecting the divinely given orders of nature and language. This was the prevailing ideal of poetry that the laws for rhymed verse first explained by Gascoigne were dedicated to fostering: You that have spent the silent night, In sleepe and quiet rest, And ioye to see the cheerefull lyght That ryseth in the East: Now cleare your voyce, now chere your hart, Come helpe me nowe to sing: Each willing wight come beare a part, To prayse the heavenly King.18
The stanza is composed of forty-four monosyllables, five words of two syllables, and one of either two or three: the pattern of regular iambic trimeter
-07&3):.&48*5)0'
expected of the last line makes it most likely that heavenly is contracted to two syllables, though since it is the most significantly charged word and is positioned in the last line, it might be allowed three. The metrical arrangement of syllables is reinforced by the end rhymes, themselves all monosyllabic nouns except sing, which as an infinitive works more like a noun than a verb. Beyond those syllabic and grammatical symmetries, the paired words are close to one another in lexical meaning, and not only those end words that agree in sound but also those that are acoustically unlike. That is, night has the same semantic orientation as lyght but also rest; lyght in turn is located by familiar biblical imagery in the same semantic field as East. The pairing of hart and part is so common in sixteenth-century poetry that their likeness in sound could persuade, as Hugh Kenner said of proverbial rhymes, “that they inhere in the workings of the normal mind”;19 part also has a semantic association with sing in the language of religious and other ceremonies because in choral music, which the poet in this stanza invites readers “to sing,” each voice is assigned its “part.” Likeness of syllabic structure, and of lexical and grammatical meanings, together with like sound, creates a symphony that draws into its concert even the rhyme words that are unlike in their vowel sounds: night, rest, lyght, hart, part all close on the sound of t. Only the “heavenly King” demands a rhyme of his own, but one that agrees with the uninterrupted accentual pattern of masculine rhyme. These likenesses between words linked as end rhymes were conceived as given: not only in the senses that the same pairs were always at hand, expected, and the more valued for having been so often repeated as to have become proverbial truths. George Herbert, in the last two stanzas of “A true Hymne,” stated or—to use his own preferred verb—copied out another understanding of rhyme as given: He who craves all the minde, And all the soul, and strength, and time, If the words onely rhyme, Justly complains, that somewhat is behinde To make his verse, or write a hymne in kinde. Whereas if th’ heart be moved, Although the verse be somewhat scant, God doth supplie the want.
-07&3):.&48*5)0'
As when th’ heart sayes (sighing to be approved) O, could I love! and stops: God writeth, Loved.20
Because agreements in many features of the rhymes together make the harmony in these stanzas, as they do in Gascoigne’s, the form of the verses agrees closely with their sense so that form and sense themselves can be said to rhyme. Herbert’s title, which could serve equally for Gascoigne’s morning hymn, implies as much: to call a hymn true was to describe it as being faithful to eternal verities and exact in form. Sixteenth- and earlier seventeenth-century prescriptions for end rhymes were based on the hierarchy of grammatical parts in English, and reflected the hierarchy of poetic kinds that the makers of the new vernacular verse promoted (probably encouraged by the examples of modern Italian and French poetry). Theorists like Puttenham condescendingly allowed scattered departures from the accepted norm of monosyllabic masculine rhyming in poems of lesser stature in the hierarchy of kinds: And the accented sillable with all the rest vnder him make the cadence . . . but alwayes the cadence which falleth vpon the last sillable of a verse is sweetest and most commendable that vpon the penultima more light, and not so pleasant: but falling vpon the antepenultima is most vnpleasant of all, because they make your meeter too light and triuiall, and are fitter for the Epigrammatist or Comicall Poet than for the Lyrick and Elegiack, which are accompted the sweeter Musickes.21
This view gave poets license to vary details of their rhyming style to suit verses of different kinds. The more elevated in rank demanded rhyme pairs closest in sound and in other features that would mirror concord in the creation. The humorous kinds that treat of disorderly conduct and upset values, accidents and tricks might be served by words mismatched in form and in lexical and grammatical meanings, which would represent conflict in the human world and untrustworthiness in its language: Sidney’s love–shove; Spenser’s handsomly–I; Donne’s kinde–blind-/[nesse], away–India, or–abhorre; Jonson’s love–Glove, embrace–Pythagoras, at a–taffeta; Herrick’s Go on–Predestination, Tiffanie–the–Sea. The revolution in English poetry in rapid progress by was a reordering of the hierarchy of kinds, again measured by new standards for the proper practices of rhyming. These were laid out in by Edward Bysshe in his respected treatise The Art of English Poetry. The kind of verse Bysshe took
-07&3):.&48*5)0'
for granted as norm and ideal was rhymed “Heroick Poetry,” defined to be lines coupled by end rhymes, each line made of exactly ten syllables arranged in iambs so that the final syllable would be strongly accented. In his treatise Bysshe explained the more minute formal regulations of the new verse, and included a dictionary of “such Words only, as both for their Sence and Sound are judg’d most proper for the Rhymes of Heroick Poetry.”22 By definition the strict criteria for heroic rhyme excluded schemes of alternating end rhymes and divisions into stanzas, structures preferred by earlier poets writing in the highly ranked “Lyrick and Elegiack” kinds; in Bysshe’s ranking “the Lyrick” was demoted to the same inconsiderable status as “Burlesque.”23 The stricter rules for ensuring the accent on the final syllable of the line and the closure at the end of the couplet promoted close agreement rather than difference in accent between the paired words, and since the rhymes were adjacent, their accents were sharply audible. At the same time, Bysshe’s catalogue of exclusions listed rhymes that are phonetically “too perfect,” which, though “allowable in the Days of Spencer and the other old Poets,” are “not so now; nor can there be any Musick in one single Note.”24 His example—Light–Delight—makes clear that the objection was specifically to pairs where both the vowel sound and the consonant before it are acoustically alike. Still, there is a more general suggestion implicit here that though there were more and stricter criteria for proper rhyming, “too perfect” agreement, overshadowing all the subtler differences between rhyme pairs, was deliberately to be avoided in heroic couplets. The same suggestion is implied by the silence in Bysshe’s treatise and in other discussions of heroic verse about the desirability of monosyllabic rhymes, which typically come close to perfect agreement and were in large part for that reason treasured and preserved by repeated use. In An Essay on Criticism Pope used Breeze–Trees (an instance of the “too perfect” type of Light–Delight) to make fun of the stock of “unvary’d Chimes, / With sure Returns of still expected Rhymes.”25 Simple boredom must account to some degree for the lack of enthusiasm for familiar, monosyllabic nouns as end rhymes; a sense of historical distance from earlier and more impoverished stages of English must also have played a part. Beyond those causes, there seems to have been at issue conceptions of the origins and nature of rhyme that differ from the sense of it as given that informed the earlier enthusiastic practice of monosyllabic end rhyming.
-07&3):.&48*5)0'
Four couplets into Pope’s Windsor-Forest is a set passage that, like Gascoigne’s and Herbert’s stanzas, calls attention to its matching of form to sense. These lines present themselves as a deliberate model for poetry, and specifically for heroic rhyme ordered “harmoniously,” “as the World” in its divinely intended form: The Groves of Eden, vanish’d now so long, Live in Description, and look green in Song: These, were my Breast inspir’d with equal Flame, Like them in Beauty, should be like in Fame. Here Hills and Vales, the Woodland and the Plain, Here Earth and Water seem to strive again, Not Chaos-like together crush’d and bruis’d, But as the World, harmoniously confus’d: Where Order in Variety we see, And where, tho’ all things differ, all agree.26
Like the earlier sort of poetry that Gascoigne’s morning hymn represents, Pope’s “Song” celebrates a natural order where parts “all agree.” Here that emphasis is conveyed by adjacent masculine rhymes, closely matched in sound, anchoring the ends of lines identical in number of syllables and in positioning of accent. Agreement in these features always creates the most immediate impression of harmony. Even so, the balance toward closeness over distance in rhymes does not have the same effects in Pope’s lines as in Gascoigne’s. Only four of Pope’s ten end words here are monosyllabic nouns, and of them only Flame and Fame are paired together. Song is linked in imperfect grammatical agreement with long; Plain is mismatched both grammatically and in syllabic structure with again. The distance is not disturbing, as it deliberately is in the examples quoted earlier of unruly comic rhymes, but it does allow space for “Variety” within the measured “Order” of rhymes. Pope’s end words do not belong to the meta-language of conventional rhyme that Gascoigne used when he joined night with lyght or hart with part. These were commonplaces passed on from poem to poem so that readers could participate easily by predicting the completion of the chime, singing their part in the song. Pope selected his words from the much larger and more various vocabulary. The echo of the first paired word by the second could not be predicted in these couplets (except perhaps the perfect match of Flame and
-07&3):.&48*5)0'
Fame, used more than once in Sir John Denham’s “Cooper’s Hill,” the model of descriptive verse imitated in Windsor-Forest). The introduction of variety and unexpectedness within the strict order of structure and sound in the rhymes makes readers aware of the poet’s acts of choosing and bringing the end words together in these couplets. Authorial presence is more strongly felt than in Gascoigne’s hymn, where the poet is a voice summoning us to join him in a choir of other voices all singing their given parts. Of course this awareness of the poet fetching rhymes is stronger in Pope’s satires, where the “uncommon,” “Low,” and “Miscompounded Words” allowed by Bysshe only in humorous verse widened the distances between rhymed words in lexical and grammatical meanings, as well as in syllabic structure. Still, Pope even in satires took care to avoid the most extreme form of disagreement between rhyme pairs that would result from drawing on those parts of speech Bysshe altogether outlawed as words that “ought not to end a Verse; as the Particles An, And, As, Of, The, &c.”27 In the heroic couplets of Pope’s pastoral poem, the role of the poet in shaping the “Description” is called to attention by that word itself; by “seem,” which shows him engaged in interpretation; and most strongly by the line “Where Order in Variety we see” (its internal rhyme adding to its variety and unexpectedness). We readers see the multitudinous classes of things in the phenomenal world unified in a single design because the poet’s rhymes are models of the way “tho’ all things differ, all agree.” In the closing pair of lines, the first proposes a definition of order in “the World”; the answering line defines the heroic couplet. Together they characterize the ideal of heroic poetry in rhyme, where, Pope wrote in An Essay on Criticism, “The Sound must seem an Eccho,” that is, a rhyme, “to the Sense.”28 The aesthetic of heroic rhyme agreed with the earlier ideal of symphonic rhyming in the sense that both assumed the existence of a grandly inclusive order in the phenomenal world and in language. Throughout this extended period, in spite of shifts in the hierarchy of kinds, English poetry accepted rhyme as (along with new highly developed and regularized standards of metrical correctness) the expressive instrument of some overarching, unifying authority. Even so, historical differences distinguished between the conceptions of rhyme so authorized. In the earlier period, proverbial rhymes were conceived as divinely given repositories of truth preserved and handed down in poems. In eighteenth-
-07&3):.&48*5)0'
century heroic rhyme, the poet was accepted as an intermediary authority gifted in the art of finding adequate form in rhymes to express universal truths, a process Pope summed up in his famous maxim: “What oft was Thought, but ne’er so well Exprest.”29 In revolutions of English poetry, rhyme has always been a charged issue, its defenders and attackers debating on one side or the other as to whether it fosters truth to nature in poetry or thwarts it. By the late s dissatisfaction with heroic rhyme became a force in the revolution against what the makers of new poetry considered the all-inclusive artificiality of eighteenth-century verse. This revolt surfaced, and has done intermittently ever since, in two forms. Young, looking back to Paradise Lost as the ideal model of what Milton called its “English Heroic Verse without Rime,” set an example along with Thomson’s and Cowper’s by eschewing rhyme altogether in long poems. 30 Wordsworth followed this course, while other poets of his time and later tried out rebellious experiments with various new ways of rhyming. To this point in the discussion it has been possible, with considerable simplification, to recognize and describe preferred forms of English rhymed poetry roughly from Gascoigne’s generation to Pope’s, and to extrapolate from them a common sense of language and its place in the world. To borrow from Clive Scott’s description of French verse of comparable periods, poetry in English before the later s was generally predicated on the assumptions that there is accessible to poetry “some underlying unity in the world,” that “the relationship between signifier and signified is direct, unmediated, and unproblematic,” that “semantic closure is perfectly achievable and truth utterable.”31 As these assumptions began to unravel—for the many complex reasons too often discussed to need summarizing here—the possibility of a new aesthetic with a dominant ideal of form in rhymed verse dwindled as well, and for the same reasons. Various poets experimented with new ways of rhyming that were informed by an assortment of literary and metaphysical attitudes and impulses, but showed a common interest in freeing poetry from the unnaturally strict regularity and the too predictable closure of the heroic couplet. That impulse encouraged, among several poets writing in the early s who were influential throughout the century, the preference for paired end words that differ more than they agree in formal features and in lexical and grammatical meanings. Loosening the likeness in sound between rhymed words was the most
-07&3):.&48*5)0'
widespread form of revolt, a practice that rhymed poetry in English has not turned away from since. Blake set an early example in such pairs as tomb– come and lawn–morn that later poets expanded by enlarging the gap. Some examples are Keats’s patiently–sigh; Elizabeth Barrett Browning’s fewer–poor; Emerson’s fell–oracle; Longfellow’s woods–solitudes; Dickinson’s indeed– instead; Swinburne’s love–enough. Coleridge launched a different attack on heroic rhyme in his experiments with classical meters. By their admission of trisyllabic feet, they invited end rhymes mismatched in syllabic structure and placing of accent: hospitality–jealousy; and especially paired feminine with masculine rhymes: unless–loveliness; suddenly–be. Keats tried out the curious effects of making the rhymed couplet straddle separate sentences: “Will trace the story of Endymion. / The very music of the name has gone.” Poets writing in traditional rhymes at least since the time of Shakespeare have allowed themselves prepositions as end words when they act as part of a verb in what linguists now call phrasal verbs or prepositional verbs.32 Instances are Shakespeare’s drinkes it vp–cup; Herbert’s creeps in–sinne; Milton’s bout– drawn out; Wordsworth’s passing by–jeopardy; Tennyson’s floated by–high. For a considerable number of nineteenth-century poets this kind of grammatical formula offered an effective escape from the artificiality of diction they objected to in eighteenth-century poetry. Browning heard in such phrases the authentic note of colloquial speech, particularly when the preposition stands out at the end of the line at some distance from the verb: “look backward for,” “Telling ought but honest truth to.” Hardy, above all, showed interest in such phrasal rhymes, using them to create a sense of a speaking voice coming through strictly ordered meters and rhyme schemes. He often ended lines with colloquial expressions like “worth living for,” “found herself in,” “know nothing of,” “thought nothing of,” of being specially useful for many of his poems by offering a rhyme with love that was neither trite nor trivial. Probably in large part because of Hardy’s effective use of these phrases, Marianne Moore praised him as the master of the “unaccented terminal rhyme.”33 In a very few, but striking, instances, Shelley violated traditional rhyming practices by different and more radically subversive use of end rhymes made from prepositions and other parts of speech on Bysshe’s list of words altogether excluded from serious rhymed verse: “the Particles An, And, As, Of, The, &c.” As late as , when the revolt against the heroic couplet could declare itself victorious, an encyclopedia article on grammar cited in the
-07&3):.&48*5)0'
Oxford English Dictionary (OED) called particles “inferior Parts of speech.” What made them attractive to Shelley in such instances, it seems, was their potential—when not built into a phrasal or prepositional verb—for enjambing lines of verse. A word like of when part of a phrase acts as a modifier or intensifier of the verb preceding it, as in to make of (where of gives the verb phrase a different meaning from to make). When not part of such a phrase, of has the function of connecting or relating, for instance, in the love of God; it points forward to the word that must complete its significance. Hence these lines from Shelley: The chasm in which the sun has sunk is shut By darkest barriers of cinerous cloud, Like mountain over mountain huddled—but Growing and moving upwards in a crowd, 34
or these: And still I felt the centre of The magic circle there Was one fair form that filled with love The lifeless atmosphere.35
Other rare instances of lines enjambed by means of articles, conjunctions, and prepositions are sprinkled about in later nineteenth-century verse, particularly in poems by Christina Rossetti, Swinburne, and Hardy. Gerard Manley Hopkins devised a sort of quasi-mystical system that allowed inclusion of those parts of speech in rhymes. His idiosyncratic sprung rhythm encouraged various departures from traditional practices of syllabic structure and placing of accent. His equally personal ways of thinking about rhyme—that “any two things however unlike are in something like,” that “all beauty may by a metaphor be called rhyme”—justified pairing words to any degree like in sound, however unlike in grammatical and lexical meaning: 36 England–and; the–thee; or–snowstorm; busy to–unvisited; on–Communion; of–Love. Before the nineteenth century, these end rhymes would have been allowed only in humorous poems (though not in Pope’s heroic satire). Once the nineteenth-century experiments with wider distances between rhymed words in sound, syllabic structure, and accent grew familiar in the work of greater numbers of poets not otherwise much alike, those new rhymes could be taken seriously. More or less extravagant disagreements in
-07&3):.&48*5)0'
these three features slowly began to be tolerated; they were expansions of traditional rhyme that did not destroy its conformity with our everyday patterns of thought and language. In the same period, rhymes like love–of were not so absorbed, as the reception of Hopkins, who died in , to some degree reflects. His poems first appeared as a collected volume in , but it was not until the mid-s that they were acclaimed, revealingly as the work of a modern poet. This valuation was due in considerable part to his radical innovations in rhyme. On the basis of Hopkins’s contributions to the movement toward allowing more extreme innovations in rhyming, T. S. Eliot in the mid-s in his introduction to a volume of poems by Marianne Moore, included Hopkins’s work among what he considered at that time to be “the best contemporary poetry.”37 It is not altogether an exaggeration to say that the possibilities of rhyme were reconceived in when Marianne Moore in a poem with the arresting title “The Just Man And” used an end rhyme of the with be.38 Her first readers had good cause to respond with amused bafflement not only to the grammatically odd title but to what they would have taken to be a radically unbalanced end rhyme pairing the most significantly charged verb in the English language with an article. We have evidence that even sophisticated readers of contemporary poetry, Moore’s fellow poets, had much the same sort of initial response that on a lesser scale we have to the sentence “Love rhymes with of.” William Carlos Williams allied himself with Moore in sharing the overarching aims of “all the moderns” to “separate the poetry from the subject,” to insist that “pure craftsmanship joins hard surfaces skillfully,” and to do without “ex machina props of all sorts, without rhyme, assonance, the feudal master beat, the excuse of ‘nature,’ of the spirit, mysticism, religiosity, ‘love,’ ‘humor,’ ‘death.’ ”39 Still, even Williams, who believed that “Marianne Moore is of all American writers most constantly a poet,” acknowledged the “incomprehensibility of her poems.”40 He complained that the “only help I ever got from Miss Moore toward the understanding of her verse was that she despised connectives,” another feature of the modernist aesthetic.41 Williams gave these accounts of his responses to Moore’s poems in the early s. In , Auden made what he later called his “silliest critical blunder” when he “could not make head or tail” of the work newly gathered in Moore’s Selected Poems.42 Wallace Stevens, commenting on that volume, described its aesthetic as “new,” its style as “unique,” and Moore as a “complete
-07&3):.&48*5)0'
disintegrator” and an “equally complete reintegrator.”43 Eliot, who edited the selection, introduced her as a poet of “original sensibility”—“Miss Moore has no immediate poetic derivations”—who “found her rhythm, her poetry, her appreciation of the individual word, for herself.”44 He praised her for “startling us into an unusual awareness” likely to cause “bewilderment” to the point where it would sometimes be “difficult to say what is the ‘subject-matter’ ” of a poem.45 Stevens registered the same response when he said of Moore’s poems, “Subject, with her, is often incidental.”46 Without admitting to the incomprehension that Williams and Auden openly confessed, Eliot’s introduction implies that readers of Moore’s poems must retrain their ways of thinking to understand the workings of her mind and language, as different from the readers’ as the operations of “a high-powered microscope.”47 The formal feature of Moore’s poems that Eliot thought readers would find most baffling, since he gave it most attention in his introduction, was her use of rhyme, “in itself an innovation in metric.”48 He began his lesson with a description of rhyme as it was practiced by poets writing in English before his present generation: In the conventional forms of rhyme the stress given by the rhyme tends to fall in the same place as the stress given by the sense. The extreme case, at its best, is the pentameter couplet of Pope. Poets before and after Pope have given variety, sometimes at the expense of smoothness, by deliberately separating the stresses, from time to time; but this separation—often effected simply by longer periods or more involved syntax—can hardly be considered as more than a deviation from the norm for the purpose of avoiding monotony.49
More recently, some poets contemporary with Moore and Eliot had made the modernist choice with Williams to dismiss rhyming altogether. Others had asserted their modernity with regard to rhyme by using it “here and there to make a pattern directly in contrast with the sense and rhythm pattern, to give a greater intricacy.”50 Moore, although she publicly claimed to despise rhyme on principle, was one of these (she owned a rhyming dictionary that she told Elizabeth Bishop was “indispensable”).51 This rhyming against the pattern of sense and stress Eliot separated into two kinds, “heavy or light,” saying that of “the light rhyme Miss Moore is the greatest living master; and indeed she is the first, so far as I know, who has investigated its possibilities.”52 What he meant he illustrated partly by quoting the opening end rhymes of two stanzas from “The Fish” (a poem with a
-07&3):.&48*5)0'
conventional-seeming title that turns out to be the grammatical subject of a sentence ending partway through the second stanza). The innovative effects of Moore’s kind of light rhyme, Eliot explained, were “sometimes obtained by the use of articles as rhyme words,” as in the second stanza from which he quoted these lines: an injured fan. The barnacles which encrust the side of the wave, cannot hide . . .
and from stanza : the turquoise sea of bodies. The water drives a wedge . . . 53
Both the words that initiate the rhymes in these two opening couplets would be called light because as articles they are regularly unstressed in a sentence and unaccented in verse. Words of this category (articles, conjunctions, and prepositions to which modern grammarians have added demonstratives) are fundamentally different from the parts of speech traditionally allowed in rhyming. They have a function in relation to other words but no inherent signifying capacity: by themselves they refer to nothing. The problem they present to rhyming is not the proverbial difficulty of pairing apples and oranges, or even apples and umbrellas, but something more like the challenge of comparing apples with D, which, like an or the, is a sign without what linguists call descriptive content. It is for these reasons as well as for their light stress that such words, described by Charles Tomlinson as “the humbler components of language,” were for centuries denied the prominent position of end rhymes in serious verse.54 Moore’s way of undermining the ranking of grammatical parts that dictated traditional rhyming practice was to seem innocently unaware of it. In “The Fish,” she slyly tucked her two stanzas beginning with the most subversive form of rhyme—an–fan and the–sea—among three other stanzas that are built in the same unconventional lineation and rhyme scheme but that play by traditional rules in being made out of acceptable parts of speech: wade–jade, sun–spun, pink–ink. Yet that grammatical distinction, which is the predication of our ordinary ways of thinking about the nature of language and how
-07&3):.&48*5)0'
it works, is blandly ignored. Or it is swept out of sight by insistence on the agreement among these two forms of rhyme by exact repetition of their other than grammatical features. The opening rhyme words in all five stanzas are notably similar in size. They are all constructed out of two, three, or four letters packaged in one syllable (letters and syllables are what Moore in “Poetry” partly meant by “the raw material of poetry”) and arranged in the identical, distinctive pattern of lines. This makes a strong impression of a single, clearly shaped design repeated all down one page as spaced in Selected Poems. The emphasis on exact structural repetition works against the hierarchical differentiation and demotion of an and the: these articles are among their equals. In some way that matters, this declaration defines the rhymes with satisfactory precision, clearing away the distortions imposed by convention, sifting the “genuine” from the “derivative” (to quote again from “Poetry”). The last three stanzas of “The Fish,” all down the facing page in this edition, bring this suggestion of conceptual equality a little closer to the surface; it may be as a way of testing our progress toward recognizing what has not been said in words but figured in their shape, structure, and arrangement (perhaps what Moore hinted by saying of her poems, or all poems, that “Form is synonymous with content—must be”): 55 All external marks of abuse are present on this defiant edifice— all the physical features of accident—lack of cornice, dynamite grooves, burns, and hatchet strokes, these things stand out on it; the chasm-side is dead. Repeated evidence has proved that it can live on what cannot revive its youth. The sea grows old in it.
-07&3):.&48*5)0'
The words that make up the opening rhymes in these stanzas (all departures from traditional rhyme following the directions of nineteenth-century experiments) do not closely mirror the look of those on the opposite page. They are made with anywhere from two-plus-hyphen to seven letters. Two are capitalized because they begin a sentence (not because lines of poetry used always to begin with capitals). One is followed by a period (marking the end of the sentence not coterminous with the stanza). Two are trisyllabic with the rhyme falling on the unaccented last syllable. One, ac- / cident, is broken off as if by accident, but of course by design, to make the components of the word fit the syllabic pattern of the lines. These stanzas offer a potpourri of irregular rhymes in various forms all different both from the radical combinations of an–fan or the–sea and from the traditional pairs like wade–jade. Even so, their identical arrangement of syllables—without regard to where the accent falls—is the measure and proof of their common nature and their equality: any word or even any syllable (Moore once pointed out that “Everyone writes in syllables!”) is raw material for rhyming.56 The opening end rhymes in the last three stanzas of “The Fish” also have in common that they use a different kind of word from those in the first five stanzas. Besides being more complexly built out of more than one syllable, these longer words contrast with the semantic simplicity of the earlier rhymes, almost like ones in a children’s book. We might be likely to describe external, ac-/cident, Repeated as words more conceptual, more explanatory, more weighted with meaning. These are all ways of describing them that make them seem more serious or dignified, deeper: certainly more so than an or the, but also than wade or fan or pink. And dead.—its finality visibly figured by the “.” that constitutes one of its parts—would be tempting to single out as the most important word in the poem. To give in to that temptation would be to disregard what Moore’s repetition of her stanza form based on exact syllabic count does to her craftily chosen end rhymes. It equalizes them as if they were digits, and mathematics is one of the preferred terms Moore used to describe her syllabic verse.57 Ranking words in a hierarchy of lexical meanings that makes dead. more important than -ed (or an or pink) is as distorting to an understanding of the nature and workings of rhyme as traditional grammatical snobbery. So is generic ranking. The trick of breaking off a syllable from the front of a word is pulled out of the bag of conventions associated with comic verse, but here the play
-07&3):.&48*5)0'
with ac-/cident occurs in a context that is not comic, where it makes a serious point: the synonymity of form and content. It seems that Moore’s rhymes in “The Fish,” matching words that belong to wholly unreconcilable parts of speech, are prized for their power to shock us. They can jolt us into new ways of thinking about rhyme and what it tells us about the nature of language and of poetry, making those concerns part of Moore’s characteristically opaque subject matter (along with such others as the unknowable mystery of the natural world, and its ways of overturning our sense of relations between time and space, outward and inward, surface and depth). One direction “The Fish” seems to be leading us is to see beyond, beneath, through the “external” conventions—their analogues may be wornoff “cornices”—that dictated centuries of rhyming practice. The particular lens “The Fish” implicitly offers as an instrument to focus our looking is the “Repeated / evidence” of its rhymes. Another poem of Moore’s brings end rhymes of the most iconoclastic kind into sharper focus by using them to open four of its five stanzas: those– echoes, and–hand, a–Persia, pear–square, to–su-. Like “The Fish,” the poem itself begins with what is called a run-on title: a phrase in the title space that starts a grammatical sequence in the text, in this instance a series of phrases and clauses with no predicate but stopped by a period thirty lines later in the second-to-last line of the second-to-last stanza: Those Various Scalpels, those various sounds consistently indistinct, like intermingled echoes struck from thin glasses successively at random—the inflection disguised . . .
The close, multiple likenesses between the three words of the title and the first three words of the opening line—grammatical units in identical order made of identical demonstratives and adjectives modifying monosyllabic plural nouns—are signals that the phrases are to be read in apposition. As grammarians define the term, it means that these parallel phrases have the identical reference; what they refer to, the first rhymed couplet points at. “Scalpels” and “sounds,” themselves naming two classes of phenomena seeming as nearly resistant to comparison as the grammatical classes of “those” and “echoes,” nevertheless both refer to, or are analogues for, rhyme. The rest of the long, incomplete sentence catalogues other entities: some “repeating . . . in reverse
-07&3):.&48*5)0'
order,” “all alike,” “of uniform / and at the same time diverse, appearances.” These phrases tell us the items listed have in common that they, along with “Scalpels” and “sounds,” are references to, or are analogues for, rhymes. In the last stanza, “these things” (the shift from “those” to “these” locating the listed items nearby where the reader can see them on the page) are collectively described as “rich / instruments with which to experiment; naturally.” Uncharacteristically, Moore’s poem comes close to explicitness here about what its subject matter might be: experimental rhyming. What its experiments discover about words, language, poetry, and their relation to “things” in the phenomenal world, the poem expresses only implicitly, very often obscurely. Collective terms given in the last stanza and in the catalogue preparing for it are unreliable in that they are not uniformly applicable to the items listed. For instance, “things” names a category that scalpels would usually be said to belong to (a more specific category would be “instruments”), but not sounds. We would ordinarily be more likely to call them by a term for intangible phenomena like effects. The inclusive description given in stanza three—“a collection of little objects”—would more emphatically exclude “sounds” because in our habits of speech object more strictly than thing refers to something we can see and touch, except in the grammatical sense of object. (That meaning, we shall see, comes close to the surface here.) The use of these ill-fitting collective terms seems pointed when contrasted with the way similar classifications in “The Jerboa”—“small things,” “little paired playthings”—are used to describe objects that all belong equally to those categories. Even when exclusively visible and tangible things are listed together as analogues for rhyme, they do not all have littleness in common. In the sequence of “pear / and three bunches of grapes, tied with silver: your dress, a magnificent square / cathedral tower,” the “square / . . . tower” is way out of scale in size with (though linked by sound to) “pear.” Beyond that, none of the items except “pear” and “grapes” seems to have anything in common with the others as we ordinarily experience them, so we find no grounds for comparing them. Other figures for rhyme in the catalogue present a different kind of obstacle to comparison, compounded by the problems that Moore’s characteristically idiosyncratic punctuation puts in the way of grammatical clarity. Some closely parallel phrases point to likenesses in form and meaning among various analogues for rhyme while they also intersect with other parallel phrases belonging to an altogether unrelated field and sometimes a different ontologi-
-07&3):.&48*5)0'
cal sphere. Here is a set of examples from stanza that shows this layering of parallels as if “successively at random”: . . . your raised hand, an ambiguous signature: your cheeks, those rosettes of blood on the stone floors of French chateaux, with regard to which the guides are so affirmative—those regrets of the retoucher on the contemporary stone: your other hand, . . .
The phrases “your raised hand,” “your cheeks,” “your other hand” (following “your hair,” “your ears,” “your eyes” in the first stanza) are closely drawn parallels between parts of the body “re- / peating . . . in reverse order” and rhyme. Then “your cheeks, those rosettes” seems to set up a different kind of parallel, not drawing on repetition of words, syntax, and semantic orientation, but on “intermingled echoes” of famous poems in the tradition epitomized by Robert Burns’s “My love is like a red, red rose.” The continuation in the next line of the clause begun by “those rosettes” pushes aside the conventional simile— “your cheeks are like roses”—and replaces it with an unexpected comparison joining things so remotely similar that the parallel is almost buried (as are the “lances all alike” that are “partly hid” by jeweled ornaments in stanza ). In the final phase of layering, those rosettes is joined, by the rhyme it initiates, with those regrets. The lexical incompatibility of these rhymed words is as difficult to resolve in its way as the grammatical disagreement between end words in the kind of rhyme represented by love–of (which Moore used once, in “The Paper Nautilus”). The untraversable lexical distance between rosettes and regrets (which also bear no discoverable resemblance to rhymes), points to the conclusion that the parallel depends solely on likeness of sound, absolutely independent of intelligibility. Here the rhyme is used as a lens to bring into focus a modern and modernist view of language and its relation to the phenomenal world that Malcolm Bowie described as fundamental to work by French poets of Valéry’s generation, who were Moore’s older contemporaries: Sound-system and sense-system are heterogeneous, and the mystical echoes which many nineteenth-century writers had heard travelling between them, and suggesting the existence between them of some unfathomable kinship, are heard no more. The relations between them are now a source of intellectual tension and excitement.58
-07&3):.&48*5)0'
In Moore’s poems, with special concentration in “Those Various Scalpels,” they are also an encouragement to obliqueness. The rhyme those rosettes–those regrets diagrams quite simply the great divide between sound and sense, while the parallel between “Those Various Scalpels” and “those various sounds” does something else that the poem wants us to find primary. Or so it seems to suggest in the question straddling its last lines: “But / why dissect destiny with instruments which / are more highly specialized than the tissues of destiny itself?” By describing the rhymes in the poem (including the one completed in the question itself, where which rhymes with rich / instruments with which to experiment) as “more highly specialized” than need be for discovering a new way of understanding rhyme, the closing question raises playful doubts about the necessity for the extravagant obliqueness of the poem’s procedures. Its collective categories that seem not wholly applicable to what they describe, and its layering of parallels, analogues, comparisons to rhyme based on distant or obscure or undiscoverable likenesses, are “too highly specialized.” The question hints at a simpler, primary understanding of the issues in the poem that the poem has not directly led us to, has disguised, hid, made indistinct or ambiguous: all words sprinkled inconspicuously among the items in the catalogue. This obscured but basic understanding is discoverable in the opening rhyme words that end adjacent lines in each stanza, even though here the exaggeratedly uneven lengths of the coupled lines put a wide distance between the rhymes so that their “echoes” are “indistinct.” As in “The Fish,” the first member of each pair is a single word by itself constituting the line, so that it is isolated for scrutiny as if in the visual field of a microscope. When we look at the one seemingly ill-fitting instance of pear, we can picture—cannot prevent ourselves from picturing—an object, much as it would be drawn on the page of a children’s book with “PEAR” written below. When we stare through the lens at those, and, a, or to, we see only letters making a word. We may hardly be aware of its sound because the echo that calls attention to it takes so long to be heard. Still, because pear is placed in a sequence of other words of similar size and identical position doing the same work of rhyming, they are all understood to belong to a common category. Although those, and, a, to cannot be included among “a collection of little objects” in the phenomenal sense, they can be objects in the grammatical sense—as Moore used the word in the title of her essay “Subject, Predicate,
-07&3):.&48*5)0'
Object”—because they all belong, equally with pear, to the category of word, and any word appreciated for itself, which Eliot praised Moore for doing, is a noun. In their own simple terms, those, and, a, to lead us to read the title sentence of this essay as stating the obvious: that “Love rhymes with of ” must be understood to mean “the word love rhymes with the word of.” What binds love and of together beyond their likeness of sound has nothing to do with their lexical and grammatical meanings, but everything to do with their common nature and status as words. Purely as words, stripped of all conventional misconceptions about them, they become nouns. Stevens was perhaps the first besides Moore among their contemporaries to recognize and use the ontological truth of this grammatical event in poems: he condensed it in his famous, playful, only seemingly obscure phrase The the.59 The primary understanding of language that Moore’s radical rhymes lead to is more widely based even than the recognition that sound and sense are heterogeneous. In “Those Various Scalpels,” her iconoclastic rhymes bring into sharp focus the unbridgeable ontological gap between words and the phenomena they point to “at random.” What Williams said of all true poems in his discussion of Moore’s can equally describe the words hers are made with: that they have “a separate existence uncompelled by nature or the supernatural.”60 Necessarily the rhymes they make do not echo some universal harmony but only their own phonetic effects, and the poet does not act in the poem as an intermediary authority who discovers the conjunction of language with phenomena of this world or beyond it. The poet of Moore’s “The Fish” is as rigorously excluded as grammar will allow, although the extravagantly strict arrangement and minute wordplay in the poem call attention to the suppressed poet’s verbal feats. “Those Various Scalpels” implies the presence of an agent shaping the raw material of the poem by another of Moore’s inconspicuous scatterings of words with like associations: struck, sculptured, sown, set, made, tied, Whetted. Beginning with struck in the opening lines, they hint at the work of an artisan skilled in using the tools of various trades. Simultaneously, the first stanza figures the poet as a performer at a street fair, striking sounds “from thin glasses successively at random,” publicly making private and precarious music. Moore’s small, simple opening rhyme words are ordinary, unspecialized verbal objects like glasses: everyday things put to use as musical instruments. The tune they play fulfills the antipoetic ideals of modernism. Though “indistinct,” creating what Sylvia Adamson described as the “illusion of a natural
-07&3):.&48*5)0'
speech which falls into rhyme almost accidentally,”61 the “sounds” made by the artisan–street performer–poet can be heard among the “highly specialized” experiments that are also a characteristic of modernist poems. Yet another of their defining features is their capacity to make fun of their own obscurity. Elizabeth Bishop remembered that Marianne Moore “made contemporary poets self-conscious about their crudities, afraid to rhyme ‘bone’ with ‘stone,’ ” and that once Moore congratulated her for the (Moore-ish) rhyme antennae with many.62 Auden admitted to having “stolen a great deal” from Moore, and by was himself rhyming love with of; Moore took amused notice that he was not “too fettered to use ‘who,’ ‘he,’ ‘the,’ or ‘which,’ as an end rhyme.”63 Richard Wilbur acknowledged having been “very charmed” by Moore’s poetry and “undoubtedly influenced” by it, and the repeated evidence of James Merrill’s rhymes makes it entirely likely that he learned from hers.64 The examples of these and other strong poets following Moore have made her most vehemently anticonventional rhymes accessible for poets to use for their own, quite different effects. Something like this assimilation or accommodation has followed every revolution in rhyming. According to Henry Lanz, when the “classical rules are violated . . . so frequently that the poets’ disregard for them can no longer be considered a mere license,” then for “many poets licenses become rules. Out of the chaos of individual experiments new principles begin to arise.”65 A further dimension of this process is that when subversive rhymes become accepted practice among the poets whose influence defines the norms for their generation, then those rhymes have a different role in the production of meaning. A sequence of couplets in Thom Gunn’s “Memoirs of the World” can show this transformation of anticlassical rhyming, even of Moore’s most radical, agrammatical type: of–love. The contrast in rhyming effects is sharpened because the passage from Gunn’s poem bears some resemblance to “The Fish” in that it leads the reader to look into the depths at aquatic creatures seen like “cells” through the lens of a microscope: I watch the cells swimming in concert like nebulae, calm, without effort, great clear globes, pink and white.—But look at the intruder with blurred outline that glides in among the shoals, colourless, with tendrils like an anemone’s
-07&3):.&48*5)0'
drifting all around it like long fur, gently, unintelligently. Where it touches it holds, in an act of enfolding, possessing, merging love. There is coupling where no such should be. Surely it is a devil, surely it is life’s parody I see, which enthralls a universe with its rich heavy passion, leaving behind it gorgeous mutations only, then night.66
The distances here between words paired in end rhymes, their mismatchings in sound, structure, accent, and lexical and grammatical meanings, are as exaggerated as in many of Moore’s experiments, but no particular attention is directed to the fact of their incompatibility. By contrast with Moore’s extravagantly imbalanced line lengths, the largely even lineation of Gunn’s couplets prevents the rhyme words from standing out to be specially noticed. They are meant to be taken for granted. Adding to this leveling effect, the rhyme pairs in this passage are all without exception mismatched in their various ways, as they are, for instance, in poems of Wilfred Owen, whereas Moore included instances of classical rhyming to call attention to the eccentricity of her experiments. Here even before the “cells swimming in concert” are disturbed in the second couplet— “But look at / the intruder”—the word concert is inharmoniously joined with effort. Irregular rhymes here are not mutations; they are the norm. Even of– love is not distinguished in its oddity from a near rhyme like concert–effort, the kind Eliot described as merely a deviation from the classical model and not an overthrow of it. These differences in verbal treatment offer concrete explanations of our differences in response to the same types of anticlassical rhyming. A more conceptual way of thinking about why we are not taken aback or baffled by Gunn’s rhyming, as we can be by Moore’s, is suggested in Clive Scott’s statement that the rhyming practices of a culture or period can be seen as “continually endorsing the intellectual and emotional stance of an ideology.”67 What Moore’s antitraditional rhymes express—that language and the phenomenal world exist without intersecting, as parallels that never meet—
-07&3):.&48*5)0'
has by now become so deeply embedded in the ideology of our culture and time that it is not held up to be questioned; Gunn’s rhymes no longer need to proselytize about these issues. Why, then, can we still be surprised and uneasily amused by Moore’s rhymes built on the model of love–of ? Besides the fact that they use every imaginable verbal means to startle, bemuse, and amuse their readers by their eccentric behavior, they are aided by circumstances of history and habit. Most readers of poetry even now are more familiar with, and therefore more at ease with, poems written over some years of traditional rhyming than with modernist and more recent poetry. Whereas, according to Wilbur, “most poets who choose to rhyme, nowadays, are troubled about using easy rhymes and are embarrassed about using their own rhymes more than once,”68 readers are more likely to be troubled by unfamiliar, anticlassical rhymes when they call attention to their departures from tradition. Or to describe differently the roles of habit and history in shaping our responses to Moore’s iconoclastic rhymes: our ideology is not easily compatible with what Marie Borroff described as our “naive or commonsensical account of the conceptual implications of our inherited parts-of-speech system.”69 The grammatical distinction between words in the separate and traditionally unequal classes that love and of belong to still has shaping power over our workaday understanding of language in relation to the things of this world. We still have a vague sense that the descriptive content of pear connects it with a piece of fruit, making it possible for us to conceive of them in relation to one another, which is what traditional rhyme teaches us to do. Whereas a is inherently empty of reference; in our hierarchical system of grammatical classes it could not legitimately be used in a rhyme with pear, which Marianne Moore vehemently insisted on doing, and Gunn might be quite likely to do without our noticing. It is tempting, but probably irresponsible, to think that this gap between our responses still shaped by inherited commonsense notions and our contemporary ideology is implicitly acknowledged in Gunn’s line “There is coupling where no such should be.”
$IBQUFS
5 ) & 4& / 4 & 0 ' 3 ): . & 4JEOFZBOE4IBLFTQFBSF
" / :0 /& 8 ) 0 4 & 5 4 0 6 5 to
write or read yet another discussion of Shakespeare’s sonnets is sure to feel something like the anxiety and dismay of the lover in Sonnet , who complains, “What’s in the braine that Inck may character, / What’s new to speake, what now to register?”1 No other poems in English suffer as much from an embarrassment of historical and critical commentary brought to bear on texts that have raised so many endlessly worried questions, so many of them unanswerable. Still, this chapter, which is an essay mainly in the sense it has in Sonnet , of a testing or trial, proposes another question about Shakespeare’s sonnets: this one rarely noticed more than in passing. The inquiry here is prompted by some of the known circumstances surrounding poetry when the sonnet was newly introduced into English, and by a peculiar feature of Shakespeare’s sequence. While his contemporaries working in this exciting new form experimented with a variety of translated and freshly invented arrangements of end rhymes, Shakespeare, whose sequence consisted of many more sonnets than most others, contented himself (with only a few variations) with a single existing rhyme scheme. His chosen form was not one preferred by most late Tudor sonnet writers, and not used for any sonnets in Sidney’s sequence, which otherwise greatly instructed Shakespeare in the making of his own. These anomalies, beyond their inconsistency with the usual practice of sixteenth-century sonnet writers, invite questioning for their seeming
5)&4&/4&0'3):.&
incompatibility with Shakespeare’s treatment of other formal dimensions of his sonnets. Unsurprisingly, he was a radical innovator in the multiplicity of metrical and syntactical patterns, as Stephen Booth has lavishly demonstrated, and in the kinds of end rhymes he deployed inside the framework that diagrammed the fixed number and disposition of the rhymes.2 The simplest inference to draw from this apparent inconsistency is that Shakespeare’s interests in the sonnet as a space for the exercise of rhyme were somehow, in some ways, different from those of poets occupied with the various structural possibilities presented by rhyme schemes, beginning with Sidney, who according to his chief editor tried out thirty-three different sonnet forms.3 This inference invites inquiry into the ways Shakespeare understood rhyme differently than other sonnet writers did in the sixteenth century, and into the ways he applied this understanding to the sonnet with different effects. It is one of the seemingly ineradicable clichés of English literary history attached peculiarly to sonnet writing in the late sixteenth century that it was a vogue or fad: one never reads of the vogue for the heroic couplet, or of free verse as a fad or fashion. In fact, the introduction of the sonnet—along with the simultaneous invention of unrhymed iambic pentameters (which Hamlet called blank verse)—is among the most consequential events in the history of English poetry. The surest sign of its enduring effect is the fascination the sonnet form had for a succession of later poets, such as Milton, Wordsworth, Keats, and Hopkins. Paradoxically, its undiminished importance was confirmed by the vigorous program of American poets in the early twentieth century to stamp out the sonnet form and to break the back of its prescribed metrical structure, what Ezra Pound called “the god damn iambic.”4 The presentation of the sonnet to English readers of poetry in their own language was an important contribution to a serious and considered project. Richard Tottel, a printer largely of law books but also the publisher of Surrey’s two books of the Aeneid in the earliest English blank verse, designed a miscellany of poems “in smalle parcelles” that could match those of “diuers Latines, Italians and other.”5 It was first published in and went through nine editions over thirty years. Tottel’s aim, besides the financial success that the miscellany spectacularly achieved, was to promote “the honor of the Englishe tong” with a display of “English eloquence.” There is no more reason to doubt the seriousness of this grand intention than to question the earnestness of Pound, Robert Frost, or
5)&4&/4&0'3):.&
William Carlos Williams in their challenge to the ascendancy of English over American poetry. To further this program, Tottel designed his Songes and Sonettes (the second term meaning both a particular kind of poem and any small poem) as a showcase for the sonnet, which was known of in England, but not usually known firsthand, as a courtly form associated with the name of the learned laureate Petrarch. Tottel displayed the sonnet prominently at the beginning of the book by his choices and placings of poems: opening with thirty-six by the Earl of Surrey of which fifteen are sonnets; then some ninety poems by Wyatt, about a third sonnets; three among Nicholas Grimald’s entries; and scattered among another ninety or so poems by unknown authors only nine sonnets, including a pair of which one is titled “That petrarke cannot be passed but notwithstanding that Lawra is far surpassed.” The entries by Surrey and Wyatt show them having experimented with various rhyme patterns in both their translated and original sonnets, but according to their different interests. Surrey worked mainly in the form of his own invention (later called the English or Shakespearean as distinct from the Italian or Petrarchan sonnet), whereas Wyatt preferred schemes he learned from Italian poets (though he regularly used the final couplet). George Gascoigne, a studious reader of Songes and Sonettes, was the most accomplished though not prolific maker of sonnets in the period between the first printing of Tottel’s miscellany and the earliest published version of Sidney’s sonnet sequence in . He was also the author of the first treatise in English on the art of English poetry, Certayne notes of Instruction concerning the making of verse or rime in English, printed in . In it he offered help in the making of end rhymes and a working definition of the sonnet, which he described as if the only poems worthy of that name were those built on Surrey’s model “of fouretene lynes, every line conteyning tenne syllables. The firste twelve do ryme in staves of foure lines by crosse meetre, and the last twoo rymiing togither do conclude the whole.”6 It seems likely that Gascoigne privileged the form invented by Surrey because his aristocratic title was the only author’s name on the title page of Songes and Sonettes, so that the book itself was specially and sometimes exclusively associated with him (as Tottel no doubt intended). When Sidney’s sonnet sequence—the first in English— appeared in print, his masterly demonstrations of the possibilities opened by variations on Italian rhyme schemes, along with his aristocratic authority, made him the model
5)&4&/4&0'3):.&
for other poets and won him laurels as England’s worthy challenger of Petrarch. Sidney himself alluded to this nationalist rivalry by making a mockmodest denial of it in Sonnet : “Nor so ambitious am I, as to frame / A nest for my yong praise in Lawrell tree.” Besides Sidney’s aspirations to win “a Poet’s name” for himself, he wanted to advance the honor of the English language by improving the eloquence of its poetry. To that end he carried forward Tottel’s national program in An Apologie for Poetrie, where he approved Songes and Sonettes by including it on a pitifully short list of the poems to date in English that he found to have “poeticall sinnewes in them.”7 On it he named only Chaucer’s “Troylus and Cresseid”; the “Mirrour of Magistrates”; his friend Spenser’s “Sheapheards Kalender”; and “the Earle of Surries Liricks” where Sidney admired “many things tasting of a noble birth, and worthy of a noble minde.” It was his intention in his Apologie, written sometime after when he had begun writing sonnets, not to take vpon me to teach Poets howe they should doe, but onely, finding my selfe sick among the rest, to shewe . . . that, acknowledging our selues somewhat awry, we may bend to the right use both of matter and manner; whereto our language gyueth vs great occasion, beeing indeed capable of any excellent exercising of it.8
Sidney hoped to show how English poetry might be improved in his generation and for the future by laying out the challenge posed to its poets by the historical transformation of the very structure of the language they had to work in. Without ruffling his elegantly easy manner, he issued a call to his fellow poets to meet this crisis with an extreme effort for the honor of English poetry: to make it new. What Sidney clear-sightedly recognized, what Gascoigne must have sensed when he offered instruction to a hypothetical young poet in the making of rhymed poetry, was that the loss to English of the inflectional grammar that had structured the line in the classical system of unrhymed verse necessitated a new kind of poetry where, Sidney said, “the chiefe life . . . standeth in that lyke sounding of the words, which wee call Ryme.”9 George Puttenham, in in The Arte of English Poesie, assessed the linguistic situation in the same dimensions, almost in the same words as Sidney: that mastering excellence in the new mode of line endings was essential to improve the work of English poets, rhyme being “the cheife grace of our vulgar
5)&4&/4&0'3):.&
Poesie.”10 It was the only means its apologists knew to defeat the accusation that English—lacking what Sidney called “those cumbersome differences of Cases, Genders, Moodes, and Tenses, which I thinke was a peece of the Tower of Babilons curse”—was inferior not only to ancient but to modern continental languages as an instrument for the making of poetry.11 Without the grammatical forms that still furnished Italian and French poets—England’s chief literary rivals among the moderns—with an abundance of like-sounding endings, poets writing in sixteenth-century English found their store of inflectional rhymes severely limited. In compensation, Sidney pointed out, English had the advantage of being a “mingled language,” which gave it more range and variety in the sounds, and in the syllabic structures and accentual patterns available for its rhymes. It still held on to its distinctive Saxon vocabulary, famous for sturdy monosyllabic words that were early recommended by Gascoigne for use as strong end rhymes, the accent necessarily falling on their only syllable. Added to this advantage, Sidney boasted, English poets could vary the effects of monosyllables at line end by infusions from French and other romance languages that might act as end rhymes of different multisyllabic structures and accentual designs.12 In this linguistic situation, where the challenge was to find adequate rhymes that would strongly define the structure of the line and the line group, the sonnet was the ideal space in which to discover and refine the possibilities of rhyming. The form’s absolute requirements—that it have fourteen lines with a fixed metrical count and an orderly arrangement of like-sounding words at line end—set limits to its shape. Within those constraints there are multiple possibilities for strengthening line endings by the choice and disposition of rhymes. No other of the fixed verse forms available in this period— or probably ever—offers such opportunities. Surrey and Wyatt recognized some of them, but it was Sidney who first exploited them to the utmost in his sonnet end rhymes. Shakespeare, we know, studied Astrophil and Stella in a copy, which must have encouraged, if it did not inspire, his interest in sonnet making.13 Elsewhere I have written at length about his detailed adaptations of particular sonnets by Sidney, and his assimilations of Sidney’s most profound discoveries of what could be made to happen in a sonnet by means of its structural complexities.14 To Sidney, John Thompson early and convincingly argued, the most important openings they gave him were for finding ways that spoken patterns of rhythm and syntax could be superimposed on metrical and rhyme
5)&4&/4&0'3):.&
schemes, creating expressive multiplicities new to lyric poetry in English.15 This argument, a premise throughout the coming discussion, needs no further support here. Shakespeare’s sonnets everywhere show how fully he understood and could put to work Sidney’s uses of end rhymes to accomplish these effects. Shakespeare’s sequence also shows that he turned his attention to rhyme in another direction, beginning with his virtually exclusive choice of the English sonnet form over Sidney’s adoption and adaptations of more complicated Italian models. Shakespeare seems, uniquely, to have recognized that new kinds of end rhymes could be made to act in different ways from Sidney’s if he were to explore fully the role of sense in rhyme. The English sonnet form expedited that enterprise. Although Sidney and later sixteenth-century theorists—Puttenham, Samuel Daniel, William Webbe in particular—were more expansive than Gascoigne in their discussions of rhyme, they recognized it just as he did, and as it is still commonly described: as a figure of sound consisting in the linkage of words that begin by making different sounds but have like-sounding endings. Even so, Gascoigne and later promoters of rhyme, in their instructions for finding acoustic matches between words by running through the alphabet (the technique of rhyming dictionaries beginning to appear in English in the late s), always insisted on the need to “searche the bottome of your braynes for apte wordes” to make rhymes.16 Apt rhymes meant pairings that “best fitte the sence of your matter in that place” rather than those that bend it solely to make the phonetic match, forcing links that the theorists liked to describe by the recently invented metaphor far-fetched.17 The concerns of these instructors in rhyme were for the compatibility of both linked words with the logical development of the poem and their suitability to its generic vocabulary: that is, with their decorum. In these cautions against far-fetched rhymes we can discern that the writers themselves, though they did not explicitly say so, sensed rhyme to be not quite simply a figure of sound. This unspoken awareness is nearer the surface in the preference of theorists and practitioners of late Tudor lyric poetry for pairings or sequences of end rhymes that are (the opposite of far-fetched) near to each other not only in sound but in their lexical meanings or associations. Such rhymes linguists now describe, using a less vivid spatial metaphor, as belonging to the same semantic field. The Italian sonnet form of octave—abba abba (by Sidney
5)&4&/4&0'3):.&
sometimes reconfigured abab abab)—is the strongest frame inside which to display this close-knit kind of rhyming because a structure of two quatrains phonetically bound together limits the eight lines to only two sets of rhymes at line end, so that each rhyme sound is repeated four times. This arrangement makes the acoustic likenesses insistent. When the words that furnish the rhyme sounds are monosyllabic, their phonetic and semantic agreements are particularly loud, because the accent must fall on the root of the word, which carries its lexical meaning. Sense and sound coexist in rhymes that make powerful markers at line end. To illustrate, one of the rhyme sequences in the octave of Sonnet in Astrophil and Stella is set off by bright answered aptly by light–sight–delight. The echoes can be heard to be so nearly exact, and the semantic links joined in sequence can so easily be traced, that sound and sense—what Gascoigne and other theorists also called rime and reason—seem to be bound together naturally. They seem to be again in Sonnet , where the identical words are linked but with light as the initiator of the rhyme chain. In these instances the words at line end have clearly been chosen to be together for their semantic as well as phonetic agreement, although conceptually the acoustic effects are the determinants since, but for their ight sound, the answering words could not be heard to echo bright in Sonnet or light in . Clearly vision, though lexically synonymous with sight, could not work in its place. Elsewhere the degree to which semantic agreement was taken into account in the making of rhyme is harder to measure: for instance in Sonnet , where one sequence of end rhymes is set off by the noun might unexpectedly answered by light–delight–night. To make the links in the rhyme chain hold together in sense, the reader would have to postulate some circumstance that would allow might a space in the same semantic field as its three answering words, a process that would overstep the limits of rhyming. The acceptance here of might purely for its acoustic agreement with the three semantically linked end words seems to make no expressive point. No special attention is called to its presence in the rhyme chain, leaving us to infer that so long as the meaning of a word did not violate the logical and syntactical development of line and quatrain, its lack of lexical connection was not a consideration. Its likeness of sound alone justified its inclusion because phonetic agreement is what makes rhyme possible. Though credible here, this conclusion does not stop us from judging the parenthesis in the opening line—“When sorrow (using mine owne fier’s might)”—to be a clumsy device altogether untypical of Sidney’s skill with end rhymes.
5)&4&/4&0'3):.&
Another perspective on the understood roles of sound and sense in this kind of rhyme linkage is suggested by Sonnet . Here one sequence of end rhymes in the first quatrain is initiated with night answered by light, but in the second the joined words are might–right. In each pair the originating and echoing words belong to the same semantic field, but the pairs themselves do not match. They mark a division that depends only on their different lexical content within the unifying phonetic design that forms the octave. Here sense plays the structural role, causing tension between inner semantic disagreement and outward agreement of sound. In Sonnet , the first line is “I might, unhappie word, o me, I might,” the end word then echoed by phonetically true rhymes semantically distant from might and from one another: night–right–delight. Instead of ignoring the lexical disagreement, this sonnet calls attention to it in the opening line by various exaggerated devices: by using the “unhappie word” to anchor both ends of the line; by a series of punctuated pauses; most of all by enjambment that surprisingly ends the line with the auxiliary might divided from the verb see by two incomplete negative clauses: I might, unhappie word, o me, I might, And then would not, or could not see my blisse:
Because might is an auxiliary to a verb, its semantic isolation is very much more prominent than the separation of might by its lexical content from the like-sounding nouns that answer it in : light–delight–night. “I might” makes a bold lexical/grammatical mismatch with its phonetic echoes, offered with a touch of bravado often heard in Astrophil’s voice. Here it seems to flaunt the power of rhyme sounds to make intelligible structures with words independent of or in contradiction to their sense. Occasionally Sidney’s pairings of end words for their exaggerated lexical disagreement with acoustic harmony are more insolent: the archaic verb form art coupled with the noun art in Sonnet and wit answered by Muscovite in Sonnet make comments on their own artful or witty disregard of sense in rhyme. The rhyme scheme of abba abba was Sidney’s preferred design for the octave, it would seem because its disposition of parts makes the strongest structure within which to arrange the end rhymes. Sonnet shows the outline of the template: Thought with good cause thou likest so well the night, Since kind or chance gives both one liverie,
5)&4&/4&0'3):.&
Both sadly black, both blackly darkned by, Night bard from Sun, thou from thy owne sunne’s light; Silence in both displaies his sullen might, Slow heavinesse in both holds one degree, That full of doubts, thou of perplexity; Thy teares expresse night’s native moisture right.
As the indenting of the lines (Sidney’s or the compositor’s) diagrams, they are framed by the a word of line and its match at the end of line , and bound together at midpoint by a pair of a rhyme words that themselves make the inner sides of the frame for two pairs of b rhymes. In the octave of this sonnet, the exact likeness of endings in the sound of ight is enhanced by the close joining together of light–might in adjacent lines, and by the monosyllabic structure of all four words, their accents on their roots. The b chain of rhymes is very different in design and its effects, because the end words are unlike among themselves in lexical content, structure, and to some degree even acoustic effect. Although they all echo the end sound of liverie, the variations in their counts of three, one, two, four syllables and in their accents with only one falling on the root, vary their acoustic effects (of pitch, extension, and other features). The two chains of end words, tightly bound by the abba abba rhyme scheme, are calculatedly weakened in their phonetic agreement—one sequence ending in the sound of t, the other in ie—and are irretrievably distant in lexical content. The result is that the two patterns of end rhymes seem to generate energy by pressing against the restrictions of the sonnet form that holds them together, rather than by the structures, we shall see, that Shakespeare liked to build out of connecting agreements and contradictions of sense. This catalogue of octaves illustrates various effects brought about by Sidney’s choices and dispositions of end rhymes. It supports the likelihood that he recognized overlapping and what seem to be mutually exclusive understandings of the role that sense plays in rhyme, and tested their applications to his designs for the sonnet form. At this point in the argument, it should be clarifying to review their differences in brief. The perfect agreement of sound and sense in rhyme chains like those quoted from Sonnets and encouraged belief that words alike in sound naturally agree in sense, or that sound originates sense much as a noise begets an echo. This preferred style of rhyming made it easy to think that light generates sight, or that might makes right. There is something almost magical
5)&4&/4&0'3):.&
in their visible and audible likeness, as in the charm still recited by children: “Star light, star bright, / first star I see tonight, / I wish I may, I wish I might, / have the wish I wish tonight.” Such agreements were received as true, meaning not only that they were exact, but also that they conformed to the natural order of things: its symphony or concord. Both terms were used by the theorists as synonyms for perfectly matched end rhymes and for the unfallen harmony of the created world. In Sonnet the end word might and in the pairing of night–light weave ties with other answering end words solely by the power of sound, a design encouraging to the view of phonetic likeness as so potent a unifying force that it could ignore or submerge irrelevant discrepancies in sense. This conception is pushed farther in Sonnet , where the calculated unexpectedness of “I might” at line end sets up a kind of rivalry between the lexical content of the verb might and the nouns that answer its sound: a contest in which phonetic identity overcomes semantic opposition. Another kind of rhyming that Sidney practiced in Astrophil and Stella (as did all poets, sometimes even Shakespeare) makes a further suggestion about how rhymes were understood to negotiate between sound and sense. This sort of rhyming can be illustrated graphically by the change of format shown below: the lists copy in order all the end words in two of Sidney’s sonnets: Sonnet on the left, on the right. They represent his only arrangements of rhymes for the octave, and his two preferred orders for the sestet, both rhyme schemes able to support this type of matching end words. Here Sidney’s particular choices are listed vertically as a pertinent reminder that words at the end of the horizontal line are deployed along a vertical axis, making them and their connectedness exaggeratedly audible and visible: face furniture pure place Grace sure endure enterlace guest such best
well sit dwell unfit tell it hell wit ease How please
5)&4&/4&0'3):.&
touch draw straw
no is kisse
Readers of poetry now are as aware as were Sidney’s contemporaries that sonnet sequences by late sixteenth-century English poets are virtually guaranteed to consist almost entirely of love poems. Surprisingly, then, in these two lists only the closing end word of Sonnet would give a clue to a hypothetical clueless reader that love is the matter of this sonnet, whereas there is no such giveaway sign in the end words of Sonnet . Even a reader familiar with the genre would not be able to locate its rhymes in a coherent space or to string them in a credible narrative. The same could almost be said of , but for the telltale kisse. To make the point about these end rhymes from a slightly different angle, a surprising number of them, though situated in the most prominent position at the junction of the horizontal and vertical axes of line and line group, are semantically empty or nearly so. In Sonnet sure and such, in it, How, is are by themselves almost without content; face, furniture, place, enterlace, draw, and well, sit, tell do not go beyond simple denotation. The words linked by sound do not form semantic connections, nor are there any between the alternating rhyme chains. Here the linked words spaced at regular intervals occupy a stabilizing position in an almost abstract pattern of sound that matches—as it were rhymes with—the quasi-mathematical pattern of the meter. Rhymes of this kind at line end belong to an intelligible system that yokes words by sound and metrical position so that, Astrophil says in Sonnet , they “flow / In verse,” which meant in measured lines. This system of neutral end rhyming was valuable for its structural effects, but also—one imagines at least for Sidney—as an alternative to predictable rhyme chains that insist on the perfect agreement of sound and sense. English monosyllables provided many, but they could seem somewhat worn by use. The offer of relief from repetition in the octave of the English sonnet, which does not demand that two rhyme sounds be repeated four times each, must have been one of its attractions for Shakespeare. By contrast, Sidney’s exclusive choice of the more intricate Italian design might be explained by his concentration on sound in end rhyme as the prime source of multiple structuring inside the sonnet frame. Shakespeare’s attention was tilted in another direction. Sonnets by Shakespeare that show his adaptations of sonnets by Sidney are a
5)&4&/4&0'3):.&
ready measure of the different emphases in their understandings of the role that sense plays in rhyme. Listed below on the left are the end words in Sonnet in Astrophil and Stella; on the right, those in Sonnet of Shakespeare’s sequence: I be me eye ly thee felicitie flie slave call have withall dwell hell
slaue pleasure craue leisure beck libertie check iniury strong time belong crime hell well
Of the two words—slave, hell—used at line end in both sonnets, slave points most tellingly to the connections between them. The situation of both poems is figured by the metaphor ubiquitous in sixteenth-century poetry of the lover as subject to a sovereign ruler, but the choice of the indecorously bitter slave in place of the conventional servant and its variants was distinctively Sidney’s. The word occurs seven times in his sequence, from which Shakespeare borrowed it to use in Sonnet and in several others variously imitative of Sidney, but never in the sonnet sequences of Daniel, Drayton, or Spenser. In both sonnets, the enslaved lover accepts the cruelty of his tormentor: Astrophil plea-bargaining “Or if I needs (sweet Judge) must torments have”; Shakespeare’s self-castigating “Oh let me suffer (being at your beck)”; and in both couplets, the lovers’ submissions are exposed as self-deceptions when they admit to being in “hell”: Alas, whence came this change of lookes? If I Have chang’d desert, let mine owne conscience be A still felt plague, to selfe condemning me: Let wo gripe on my heart, shame loade mine eye. But if all faith, like spotlesse Ermine ly Safe in my soule, which only doth to thee
5)&4&/4&0'3):.&
(As his sole object of felicitie) With wings of Love in aire of wonder flie, O ease your hand, treate not so hard your slave: In justice paines come not till faults do call; Or if I needs (sweet Judge) must torments have, Use something else to chast’n me withall, Then those blest eyes, where all my hopes do dwell, No doome should make one’s heav’n become his hell. That God forbid, that made me first your slaue, I should in thought controule your times of pleasure, Or at your hand th’account of houres to craue, Being your vassail bound to staie your leisure. Oh let me suffer (being at your beck) Th’imprison’d absence of your libertie, And patience tame, to sufferance bide each check, Without accusing you of iniury. Be where you list, your charter is so strong, That you your selfe may priuiledge your time To what you will, to you it doth belong, Yourselfe to pardon of selfe-doing crime. I am to waite, though waiting so be hell, Not blame your pleasure be it ill or well.
Shakespeare’s mirroring of Sidney’s sonnet in so many of its dimensions makes the differences in their choices of other end words than slave and hell significant pointers to their different emphases in rhyming. Most of the words at line end in Sidney’s sonnet are by themselves significantly neutral (felicitie is the clearest exception). They give little indication of semantic fields that would suggest the situation in the poem beyond that it involves “me” and “thee.” The end rhymes imply no story that would project the movement forward from line to line group, except that the closing couplet rhyme seems to arrive at some decisive recognition. Often in Sidney’s couplets (they occur in sonnets of ) both like-sounding words are semantically empty or neutral: this–is, all–shall; do–you; name–same are examples. All this is to say that the most active function of the end rhymes in Sonnet is not their signifying capacity but the usefulness of their phonetic effects for the definition of relationships among rhythm, meter, and syntax.
5)&4&/4&0'3):.&
By contrast, Shakespeare’s end words in Sonnet suggest a situation or tell a story: whether the rhymes are close in both sense and sound—pleasure– leisure, beck–check—or related in sense but unlike in sound—pleasure–craue, libertie–check. Because the words at line end have strong emotional connotations, by contrast with the neutrality typical of Sidney’s in Sonnet , they seem as it were to talk to or about each other. The situation of being a “slaue” implies deprivation of “pleasure” unless the situation is of being a slave to pleasure, or of perversely finding pleasure in being enslaved. These alternative suggestions are not interpretations of the sentence that weaves its way through the first quatrain. Instead they are associations acting like vibrations set off by the significantly charged end rhymes that turn out to have some subtextual aptness. The coming together and pulling apart of semantic contents perform something comparable to the structuring functions of Sidney’s purely phonetic effects. Shakespeare’s remaking of the episode in Sonnet in Astrophil and Stella as his own story in Sonnet borrows more end rhymes and makes more rearrangements of them. They show with stunning clarity the encouragement of the English sonnet form to his emphasis on the role of signifying in rhyme. Shakespeare took over the end rhymes of Sidney’s first, third, and eighth lines— ride, face, hide—and moved them into adjacent positions in his own lines , , . He also adapted Sidney’s end words grace and shone in lines and as disgrace and shine at the end of the second quatrain and beginning of the third: In highest way of heav’n the Sunne did ride, Progressing then from faire twinnes’ gold’n place: Having no scarfe of clowds before his face, But shining forth of heate in his chiefe pride; When some faire Ladies, by hard promise tied, On horsebacke met him in his furious race, Yet each prepar’d, with fanne’s wel-shading grace, From that foe’s wounds their tender skinnes to hide. Stella alone with face unarmed marcht, Either to do like him, which open shone, Or carelesse of the wealth because her owne: Yet were the hid and meaner beauties parcht, Her daintiest bare went free; the cause was this, The Sunne which others burn’d, did her but kisse. Fvll many a glorious morning haue I seene, Flatter the mountaine tops with soueraine eie,
5)&4&/4&0'3):.&
Kissing with golden face the meddowes greene; Guilding pale streames with heauenly alcumy: Anon permit the basest cloudes to ride, With ougly rack on his celestiall face, And from the for-lorne world his visage hide Stealing unseene to west with this disgrace: Euen so my Sunne one early morne did shine, With all triumphant splendor on my brow, But out alack, he was but one houre mine, The region cloude hath mask’d him from me now. Yet him for this, my loue no whit disdaineth, Suns of the world may staine, when heauens sun staineth.
Shakespeare’s sonnet follows the traditional English scheme, three sets each of two alternating rhyme sounds, but in lines to it clusters end words borrowed from Sidney into a sequence linked not by acoustic likeness but by shared semantic associations. Words that are relatively neutral when spaced at distances from each other are in their new proximity charged with significance. These lines make a distinct structural unit that cuts across the divide between the second and third quatrains, linking four different rhyme sounds into a chain of five words held together by sense. They hint at a sinister story already insinuated in the feminine rhymes at the beginnings of earlier lines, which make their own semantic chain: Flatter–Kissing: Guilding–Stealing. The linkage of face–hide–disgrace–shine–brow also carries forward the story line to its conclusion in the ugly feminine rhyme of the couplet. Shakespeare seems here to have tried out a multiplicity of possibilities opened by the English sonnet form for treating the end of the line so as to make signifying the dominant feature in the forward movement of the line group. In Sonnet in Astrophil and Stella, the relative freedom from connotations of most of Sidney’s words at line end, twelve of them monosyllabic, allows their agreement in sound to be heard with special distinctness. This effect is enhanced because the like sounds occur more often, and more often in adjacent lines, than in the English rhyme scheme. Although we cannot predict what the answering words will be, as we can when the first rhyme word belongs to a familiar chain of semantically related and like-sounding words (light–bright– sight), once Sidney’s cognitively neutral rhymes are in place, they seem inevitable. This happens because their various phonetic agreements are so strong; their appearance of inevitability, in turn, strengthens their structuring powers.
5)&4&/4&0'3):.&
In contrast, the sound effects of Shakespeare’s end words in Sonnet (except in the couplet) are not heard as sharply for multiple reasons, the most influential being that they demand more attention than Sidney’s to their semantic combinations. The second quatrain shows this process, which depends heavily on the alternation of like-sounding words demanded by the English sonnet: Oh let me suffer (being at your beck) Th’ imprison’d absence of your libertie, And patience tame, to sufferance bide each check, Without accusing you of iniury.
At the beginning we are distracted from the exact agreement of beck–check because their joining is delayed. The initiating word beck must first negotiate with the semantic content of libertie, which seems almost its contrary except that the paralleling of your beck, your libertie so close together points to their overlapping. The tormentor takes the abusive “libertie” of beckoning his enslaved lover; “libertie” is also the name for the slave master’s licentious behavior; it is a dangerous freedom that causes “iniury.” This kind of reading of multiple significations is demanded often by Shakespeare’s rhymes: Sonnet is another example. The first line—“From fairest creatures we desire increase”—has nearly oracular authority by virtue of the biblical associations of its end word. It is balanced by chiasmas with the sound of creature, another reminder of Genesis. Line —“That thereby beauties Rose might neuer die”—is a quasi paraphrase of the biblical command to increase and multiply, but the substitution of neuer die for the expected rhyme multiply tilts the sense of the first two lines: increase has no power over mortality. The second set of alternating end rhymes has further unsettling effects: But as the riper should by time decease, His tender heire might beare his memory.
The stability and amplitude of line might have been expected to evoke ease, please, or release as the echo of increase. Instead we meet decease, which echoes the sense but not the sound of die. Completing the rhymes of the octave, the trisyllabic word memory, ending on the weaker of its two accents, pits itself against its stronger partner die. Multiple, overlapping signification here, as in Sonnet , is a structuring principle and source of energy. It would hardly be an exaggeration to say that in this kind of rhyming, sense seems to generate sound.
5)&4&/4&0'3):.&
An added reason why the rhymes in Sonnet do not have the perfect pitch typical of Sidney’s is that the pairings do not as often match each other in structure and accent, especially in the octave (which requires four different rhyme sounds, whereas the sestet demands only three). In this sonnet, slaue–craue is a monosyllabic rhyme by its structure carrying the masculine accent, while pleasure–leisure, where the second syllable is unaccented, joins feminine endings that typically weaken or blur the sound at line end. The trisyllabic pair libertie–iniury is obedient to the rules of masculine rhyme, but the final accent is less distinctly heard because the stronger emphasis is given to the previous accent on the first syllable. Where rhyme pairs are not identically structured and accented, some attention gets diverted from the sound effects of words at line end to their sense. Another contribution to Shakespeare’s distinctive modes of rhyming is his surprising vocabulary. In Sonnet , where prime would have been the expected echo of time (making one of those prized proverbial rhymes that seemed to be sanctified truths), Shakespeare chose crime. He used that word again in three other sonnets, but it does not appear anywhere in the sonnet sequences of Sidney, Daniel, Drayton, or Spenser. An approximate count (made for this study without benefit of technology) of Shakespeare’s end words not to be found in these other sequences lists close to two hundred such entries, by contrast with a parallel count of about sixty end words found only in Astrophil and Stella.18 Shakespeare’s prolific vocabulary, making rich use of the advantages that Sidney attributed to English as a “mingled language,” is suited to the demand of the English sonnet for a greater number of varied rhyme sounds. Turned around, this statement is more to the particular point here: that the English form opened more spaces for unexpected pairings of words unfixed by convention and capable of multiple signification. Shakespeare owed his extraordinarily large and mingled vocabulary for rhyming partly to his acceptance of words that would make feminine rhymes, which, though not valued as highly as those with strong masculine accents, were allowed by sixteenth-century English theorists. Sidney made an unusual choice in exiling them from the sonnets in Astrophil and Stella (although they are plentiful in its songs), a move that must have been intended to ensure potent accents at line end, and perhaps to set an example in holding himself to the highest standards of sonnet writing. Shakespeare’s admission of feminine rhymes brought him not only a wider supply of words but certain kinds of words with capacities for special forms of signifying. He experimented with
5)&4&/4&0'3):.&
them as markers of new structures allowable within the frame of the English rhyme scheme. A number of sonnets show that Shakespeare used words having an unaccented final syllable for end rhymes because of their subversiveness. He treated them so that their disruption of the norms dictated for the structure and accent of words at line end contributed to their signifying content, tilting it toward suggestions of illicit or corrupt matter. We have already seen that happen in his uses of pleasure–leisure and disdaineth–staineth. He made a witty game of this subversive tendency in Sonnet by playing with the term feminine rhyme itself (borrowed by Sidney and other English theorists from the French). The sonnet is addressed to the effeminate young man the poet calls “the Master Mistris of my passion.” All the end rhymes, some descriptive of the young man and some of “womens” weaknesses from which the poet at least seems to be exonerating him, end on an unaccented syllable: painted, passion, acquainted, fashion, rowling, gazeth, controwling, amaseth, created, a doting, defeated, nothing, pleasure, treasure. This pattern of rhyming, unique in Shakespeare’s sequence, is the organizing principle that unites sense with sound. The formal subversiveness of the feminine rhymes adds not quite innocent insinuations to the semantic content of the end words, but those subversive hints are kept within bounds by the fidelity of the rhymes to their own perfectly disciplined (and prosodically accepted) rhyme scheme. Sonnet makes an unplayful game out of the signifying possibilities of metrical accent. The masculine rhymes, almost neutral in semantic content, in the first quatrain are replaced by two- and three-syllable end words with feminine accents in the second: Take all my loues, my loue, yea take them all, What hast thou then more then thou hadst before? No loue, my loue, that thou maist true loue call, All mine was thine, before thou hadst this more: Then if for my loue, thou my loue receiuest, I cannot blame thee, for my loue thou vsest, But yet be blam’d, if thou this selfe deceauest By wilfull taste of what thy selfe refusest.
The shift to feminine rhyming in lines through gives a different quality to the expression of the lover’s feelings than if Shakespeare had stopped at line
5)&4&/4&0'3):.&
end with forms of these verbs having rhyme sounds and final accents only on the root: receiue, vse, deceaue, refuse. These (hypothetical) masculine rhymes would bluntly expose the lover’s self-degradation, whereas the (actual) extra rhyming on weakly trailing suffixes is the poet’s softening, blurring, covering over of the sexual/monetary transactions of his lascivious friend. In the sestet another shift to end words that do directly say what they mean—theefe, pouerty, griefe, iniury, showes, foes—signifies the poet’s bitter recognition of his own self-deceptions. The structure and accent of the feminine end rhymes mark off the middle quatrain as a stage in the evolving drama of the lover’s feelings. Adding to the expressive capacity of the feminine metrical accent, Sonnet makes use of the grammatical expressiveness built into the present participle. This is another source of end rhymes avoided by Sidney in his sonnets, with the one exception of the masculine pair in where sing is echoed by triumphing (with much less emphasis on the final accented syllable than on the first). Probably he excluded such rhymes for the same reason that he forbade himself feminine end words: that their final accents did not have the sharpness and stability that made the phonetic effects of masculine rhymes structurally powerful at line end. In Sonnet , the participial rhymes are measured against the only masculine rhyme—estimate–determinate—which mimics the legalistic language of the faithless young man: Farewell thou art too deare for my possessing, And like enough thou knowst thy estimate, The Charter of thy worth giues thee releasing; My bonds in thee are all determinate. For how do I hold thee but by thy granting, And for that ritches where is my deseruing? The cause of this faire guift in me is wanting, And so my pattent back againe is sweruing. Thy selfe thou gau’st, thy owne worth then not knowing, Or mee to whom thou gau’st it, else mistaking, So thy great guift vpon misprision growing, Comes home again, on better iudgement making. Thus haue I had thee as a dreame doth flatter, In sleepe a King, but waking no such matter.
5)&4&/4&0'3):.&
Present participles are grammatically undetermined: they can act as verbs, adjectives, and verbal nouns. Here the poet exploits this indeterminacy as a mode of evasiveness and insinuation, turning the present participles into parodic weapons against the young man’s self-justifying judgments and against his own acquiescence to them. The noun possess or its verb form possessing are legal terms the youth would use to declare limits or boundaries of property or feelings; possessing turns this decisiveness into a fluid condition or process without ending. The ten present participles at line end make such parodic transformations of the young man’s arguments into the poet’s weak justifications and sly criticisms of them. These are then exposed in the couplet by the shift to atypically sharp feminine rhyme words in grammatically strong verb and noun forms: Flatter, which is what the parodic exonerations were meant to do; and Matter, a semantically neutral word used in the immediately preceding sonnet to refer to the “matter” of the poem. Grammatical forms contribute to sound effects in the end rhymes of this sonnet: grammatical meanings contribute richly to its sense. Their coexistence could be said to be its matter. Shakespeare’s few variations on the English rhyme scheme support the argument that sense in rhyme was for him a dominant organizing principle; that the comparative openness of the English rhyme scheme allowed him more space to implement this principle; that he was willing to modify the fixed form where the redistribution of rhymes would encourage the role of sense as a structuring force. His maverick rhyme schemes—my count found them in fifteen sonnets—cause disturbances only in the last six lines, leaving unchanged the accepted form of the octave, where the most important distinction between the Italian and English sonnet forms—which is also to say between Sidney’s rhyme patterns and Shakespeare’s—is made. Sonnet is an instance of Shakespeare having revised the expected disposition of rhymes to suit the particular situation or argument of the poem. Here he made this adjustment by echoing one of the end rhymes and end sounds of the second quatrain in the third: abab cdcd ecec ff. This peculiar arrangement stands out with particular emphasis, coming as it does in a poem belonging to the group known as the procreation sonnets, where the accepted English rhyme scheme—abab cdcd efef gg—is strongly delineated. The first four lines plead with the young man to store the treasure of his beauty (in a woman’s body) where it will beget that beauty in his child. The second and third quatrains are then bound together by the rhyme chain
5)&4&/4&0'3):.&
that repeats thee, just as the poet urges the youth to do, in “an other thee” or “refigur’d thee”: That vse is not forbidden vsery, Which happies those that pay the willing lone; That’s for thy selfe to breed an other thee, Or ten times happier be it ten for one, Ten times thy selfe were happier then thou art, If ten of thine ten times refigur’d thee, Then what could death doe if thou should’st depart, Leauing thee liuing in posterity?
This specially designed scheme of rhymes figures the sense of the lines. By contrast with Sidney’s nocturnal Sonnet , Shakespeare’s nocturnal Sonnet displays in its unconventional arrangement—abab cdcd aeae ff—their different emphases in rhyming, as it may be that Shakespeare meant it should: This night while sleepe begins with heavy wings To hatch mine eyes, and that unbitted thought Doth fall to stray, and my chiefe powres are brought To leave the scepter of all subject things, The first that straight my fancie’s error brings Unto my mind, is Stella’s image, wrought By Love’s owne selfe, but with so curious drought, That she, me thinks, not onely shines but sings. I start, looke, hearke, but what in closde up sence Was held, in opend sense it flies away, Leaving me nought but wailing eloquence: I, seeing better sights in sight’s decay, Cald it anew, and wooed sleepe againe: But him her host that unkind guest had slaine. If the dull substance of my flesh were thought, Iniurious distance should not stop my way, For then dispight of space I would be brought, From limits farre remote, where thou doost stay, No matter then although my foote did stand Vpon the farthest earth remoou’d from thee, For nimble thought can iumpe both sea and land,
5)&4&/4&0'3):.&
As soone as thinke the place where he would be. But ah, thought kills me that I am not thought To leape large lengths of miles when thou art gone, But that so much of earth and water wrought, I must attend, times leasure with my mone. Receiuing naughts by elements so sloe, But heauie teares, badges of eithers woe.
Astrophil’s waking “thought” is “brought” to fly away in the first quatrain, to be replaced by a dream vision lovingly “Wrought” with such fine “drought” (draftsmanship) that it can defy the laws of mind and body by making itself heard as well as seen. This ought rhyme chain, skillfully linked to wings– things–brings–sings by a series of enjambments, holds together the octave in a single sentence. Then the blank spaces at the end of line and at the indented beginning of make a sharp division—“I start”—between the octave and the sestet with its end rhymes pointedly differing in sound and sense from the two earlier rhyme sequences. The ought end sound of the dominant chain in the octave echoes only in the internal rhyme sound of nought in line . In the first quatrain of Shakespeare’s Sonnet , the sleepless lover wishes that his heavy body were immaterial as “thought,” which might carry him over the distance that separates him from his beloved. In the sentence that constitutes the second quatrain, “thought” is carried forward by the internal repetitions in thought and think in lines and . Then the end sounds of the earlier quatrain redouble in “But ah, thought kills me that I am not thought,” acting out by the reinforced linking of quatrains the futility and obsessiveness of the lover’s “thought.” Shakespeare may have learned much from Sidney’s uses of rhyme in Sonnet , and what it taught may have included recognition that he differed from Sidney in his conception of rhyme as a figure of sense as much as of sound. The impression of studied experiment in Sonnet is supported by the nocturnal sonnets just before and after it. In them Shakespeare made other rearrangements of the accepted English rhyme scheme by repeating end words and end sounds in unconventional designs—abab cdcd efef aa, abab cdcd dede ff—that suggest paralyses of or changes in feelings. In Sonnet , the first quatrain proposes the rhyme sounds that dominate the poet’s dreams and the unusual disposition of the end words: When most I winke then do mine eyes best see, For all the day they view things vnrespected,
5)&4&/4&0'3):.&
But when I sleepe, in dreames they looke on thee, And darkely bright, are bright in darke directed.
In the couplet, see and thee are joined by me in the situation that has been signified by the overlapping semantic fields shared by the end rhymes: All dayes are nights to see till I see thee, And nights bright daies when dreams do shew thee me.
The repetitive ordering of rhymes here, as in Sonnet , is instrumental in figuring sexual obsession. Sonnet mechanically rings changes on the same set of words and end sounds, and this is also true of and . Even so, their maverick rhyme schemes contribute to the buildup of burning concentration on the dramatic interactions of thee and me in Shakespeare’s sequence. In the years from the early s when Shakespeare is generally thought to have begun writing sonnets, to when his sequence was published in the only form we know it, he was mainly giving his time to writing dramatic blank verse (not the situation of any other authors of sonnet sequences). We might assume that practice in disposing end rhymes at fixed intervals within the strictly limited length of fourteen lines would not teach lessons in how to order unrestricted numbers of metrically regular lines without rhymes (that would substitute for lost inflections) to mark their endings. Yet Shakespeare’s inventive ways of treating rhyme inside the sonnet space lead to a different view of the possible connections between his sonnet rhyming and his unrhymed dramatic verse. Writing sonnets and writing blank verse posed essentially the same difficulty for late sixteenth-century English poets trying to control and direct the radically changed but still changing structure of their language. They had to devise ways to mark off the line horizontally as a metrical/syntactical construct, but the stabilizing marker at line end was also required to give a vertical direction to the line group. In any rhymed poem, the words that define the end of the line look forward because the initiating rhyme word predicts an echo from some later line end, but blank verse, having no such acoustic projection, needed end words with other features besides sound that would bind lines in a forward-moving sequence. Shakespeare’s sonnet rhymes put distinctive emphasis on the functions of their end position, which are to promote both stasis and sequence, because he tended to finish the line with a semantically packed word and to follow it with words related in sense that are in the parallel position in later lines, some like-sounding and some not. Together they tell a story or act out an
5)&4&/4&0'3):.&
unfolding situation, which is what is demanded also of the lineation in dramatic blank verse. To support—though necessarily in brief—the closing proposal of this chapter, that Shakespeare’s innovations in sonnet rhyming contributed to his mastery of this other challenging new verse form, I have chosen a sequence of lines exchanged among Hamlet, his stepfather king, and his queen mother in Act , scene , lines –: King. Ham. King. Ham. Queen.
Ham. Queen. Ham.
But now, my cousin Hamlet, and my son— [aside] A little more than kin, and less than kind! How is it that the clouds still hang on you? Not so, my lord. I am too much i’ the’ sun. Good Hamlet, cast thy nighted colour off, And let thine eye look like a friend on Denmark. Do not for ever with thy vailed lids Seek for thy noble father in the dust. Thou know’st ’tis common. All that lives must die, Passing through nature to eternity. Ay, madam, it is common. If it be, Why seems it so particular with thee? Seems, madam? Nay, it is. I know not ‘seems.’ ’Tis not alone my inky cloak, good mother, Nor customary suits of solemn black, Nor windy suspiration of forc’d breath, No, nor the fruitful river in the eye, Nor the dejected haviour of the visage, Together with all forms, moods, shapes of grief, That do denote me truly. These indeed seem, For they are actions that a man might play; But I have that within which passeth show— These but the trappings and the suits of woe.19
We have heard in Shakespeare’s sonnets various patterns of rhymes making line endings that are audible in and behind this dramatic exchange as ghostly echoes.
$IBQUFS
5 *5 -& 4 * / ( & 0 3 ( & )& 3 #& 3 5 ±4 ® -*55 - & # 0 0 , ¯
gave a great deal of thought to the titling of poems because we have evidence that he chose the title himself for each of his many poems, that in some instances he changed a title and not only for poems he had revised, that he formed his titles according to regular patterns, and that these formulations must have been carefully considered over a long period spanned by two authoritative manuscripts.1 This combined evidence in itself would be sufficient to have aroused interest in Herbert’s titles as unusual for a collection of shorter poems first published in , when careful attention to titling in such volumes was rarely given by either authors or editors.2 Yet the fact that Herbert’s titles have received fuller critical consideration than those of any other poet writing in English before the nineteenth century is due especially to what John Hollander has called their “amazingly radical” expressiveness. He discusses this feature in his suggestive essay, “ ‘Haddock’s Eyes’: A Note on the Theory of Titles,” as do other critics in studies of Herbert.3 Their common emphasis is on the individually “expressive character” of his titles, which identifies Herbert as the first English poet consistently to make the title “part of the poem’s fiction,” so much so that some of his seem to lead directly to uses of titles usually associated with later nineteenth- and twentieth-century poetry.4 We have more to learn from the titles of Herbert’s poems about what he thought a poem should be or do, about his notions of collecting his poems into a “little Book,” and about how it should be read, if we explore other 8 & , / 085 ) "5(&0 3 (&)&3 # &3 5
5*5-&4*/)&3#&35±4®-*55-�,¯
features of his titles. Our questions might include, What models existed for the titling of poems in collections of mixed or separate secular and sacred verse, and what did Herbert think about the relations between these two modes? What connections or distinctions did he make between his poems with their titles and these existing models? How did Herbert use titling as a way of defining the nature of his poems and the collection in which he arranged them? The casual and inconsistent habits of titling shorter poems still prevailing in Herbert’s lifetime are exemplified in the first edition of Donne’s verse, published in the same year as Herbert’s and also posthumously. These practices were common both to manuscript collections, such as the various surviving copies of Donne’s poems, and printed volumes of miscellaneous verse by one or a number of poets. What is representative about the titling in this edition, called on the title page simply Poems, By J. D., is the kinds of titles used and the sorts of irregularities in their assignment.5 Although there is no table of contents and all the poems except the satires are on pages with the running title “Poems,” the reader sees at once that in sections sometimes separately headed throughout the volume poems are grouped by genres, in this period usually called kinds. These divisions reflect choices made by the editor based partly on the various ways the poems were arranged in manuscripts available. Epigrams, epithalamions, satires each constitute one such group; almost all the epistles are together in another; poems called elegies are arranged in four sections; religious sonnets, in two groups both titled “Holy Sonnets,” are placed together. Within each of these sections, titling follows precedents associated with the genre and does so with considerable consistency within the group. Of sixteen poems headed “Epigrams,” fourteen have titles traditional for the genre; twenty-five poems constitute a group of epistles, each titled with the conventional formulation “To ——.” Love poems not fitting traditional definitions of the elegy and religious poems other than sonnets are not so clearly grouped by conventional precedent. The nonelegiac love poems are collected in two widely spaced groups, with one poem placed elsewhere without title. The first group starts abruptly without heading with an untitled poem (called “The Message” since the second edition).6 Poems with “Valediction” in the title are placed together in only one instance. The title of one, “A Valediction of my name, in the window,” is so phrased that it seems to have been given by the poet himself. Another, “Valediction to his booke,” uses the third person associated with titles
5*5-&4*/)&3#&35±4®-*55-�,¯
assigned by copyist or editor, and inaccurately describes the poem since it is addressed to the speaker’s “deare Love,” not “to his booke.” Some scattered poems are titled “Song,” and others that might be so called are without any title. The remaining love poems have the kinds of titles by which they are still known, which vary widely in formulation, as the titles in the generically separated groups do not. The religious poems not in sonnet form, also given an assortment of titles, are still less clearly grouped. “The Crosse” is set by itself between an epistle and an elegy, and “A Hymne to God the Father” comes alone after the satires. Between them and a section of elegies are two poems religious in subject but otherwise entirely dissimilar in form and title: “A Hymne to Christ, at the Authors last going into Germany,” and “The Lamentations of Ieremy, for the most part according to Tremelius.” Closest to a grouping like other sections in the volume are the poems on pages to . These begin, without heading, with three religious poems otherwise unlike—“Psalme ,” “Resurrection, imperfect,” “An hymne to the Saints, and to Marquesse Hamylton”—and end with another such trio—“The Annuntiation and Passion,” “Goodfriday, . Riding Westward,” “The Litanie.” But between these groups of three are the one poem in the volume called an epitaph and Donne’s only heroic epistle, “Sapho to Philaeris.” Instead of a section of religious verse comparable to the groupings of love poems, which the opening trio predicts, these pages read more like a miscellany of poems the editor did not know where else to place. Similar insecurities are reflected in the individual titles of the religious poems. Of the three widely spaced poems called hymns, two addressed to persons of the Trinity are in stanzas suitable to be set to music.7 In these respects they are like “Psalme ” rather than the third “Hymne to the Saints, and to Marquesse Hamylton,” which is in couplets resembling especially “Obesequies to the Lord Harringtons brother. To the Countesse of Bedford.” Since Hymne as well as Psalme meant “song,” its misuse here seems to reflect a desire for a title placing the poem in a genre where none was recognizable. The arrangement and titling of poems in the volume of Poems, By J. D. are attributable to some combination of Donne’s own habits, the practices of copyists, the accidents of transmission, the decisions of the editor, the vagaries of printing.8 In these features it is fairly representative of books—the term at this time was used indiscriminately for manuscripts as well as printed volumes—in which shorter poems of various kinds were collected, and reflects some commonly held assumptions about poems and their titles shared
5*5-&4*/)&3#&35±4®-*55-�,¯
by writers, copyists, printers, and readers in the earlier seventeenth century. The following are some of these assumptions. Wording of some sort in the title space, while more or less expected, was certainly not obligatory. Since titles were not essential, they seem not to have been considered a necessary part of the poem with a primarily expressive function (although we value the expressiveness we find in many of them). This must be a reason why titles apparently assigned by the editor (or in a manuscript by the copyist) seem to have been as acceptable as an authorial title might be, and perhaps more expected, especially in a posthumous book. It also helps to account for the inaccuracy or inappropriateness of many editorial titles. The functions of titles seem to have been conceived more often as practical than as individually expressive. A title could mark the beginning of a poem, dividing it more clearly than a space or simple line from the poem before it; it could separate the poem from another of its kind, which could be accomplished by mere numbering or by phrasing, as the title given by the editor of the edition to “A Valediction forbidding mourning” distinguishes it from the other valedictions (in surviving manuscripts it was called simply “Valediction” or had titles variously naming its occasion).9 Titles could help in orderly arrangement among miscellaneous poems, desirable perhaps, at least in part, so that the collection might more closely resemble prestigious classical and Renaissance volumes. Some other commonly held assumptions are reflected in the prominence given in the volume to titles that identify the poem as belonging to a traditional kind, either by naming the genre or by imitating the formulation of titles associated with it. These generic titles, besides aiding arrangement, were useful in establishing the credentials of the author for learning and skill in performance, while at the same time shaping the expectations of readers by placing the poem as an example of a traditional category. What Poems, By J. D. shows is that such categories were not as readily recognizable for many kinds of love poems or for religious poems. Gestures toward titling by genre for love poems that might be categorized as songs or for religious verses titled “A Hymne” are inconsistent and insecure, whereas other titles not referring to kinds for love poems and religious poems in the collection are far more various than titles given to poems within generic groupings. These patterns suggest more pressures brought to bear on the formation of titles for love poems and religious poems. This was the situation faced by
5*5-&4*/)&3#&35±4®-*55-�,¯
poets of Herbert’s generation when he made his own decisive choice to devote himself to writing religious verse. Herbert announced this dedication, we know from Isaak Walton’s Life, in a letter to his mother at New Year’s – accompanying a gift of two poems: —But I fear the heat of my late ague hath dryed up those springs, by which Scholars say, the Muses use to take up their habitations. However, I need not their help, to reprove the vanity of those many Love-poems, that are daily writ and consecrated to Venus; nor bewail that so few are writ, that look towards God and Heaven. For my own part, my meaning (dear Mother) is in these Sonnets, to declare my resolution to be, that my poor Abilities in Poetry, shall be all, and ever consecrated to Gods glory.10
We do not know if Herbert gave a title to either or both poems sent with the letter, but he refers to them there as “Sonnets,” and Walton calls them a single “Sonnet,” printing them with a stanza break between them. These attentions to their genre are appropriate for the nature of the protest to God of which the first sonnet consists and to which the second is a response: “Doth Poetry / Wear Venus Livery / only serve her turn? / Why are not Sonnets made of thee?” Herbert’s resolution was to reconsecrate the language of “Love-poems,” exemplified by sonnets, making “Poets turn it to another use,” so that once more its ornaments, “its Roses and Lillies speak” God’s praise. His choice of the sonnet form was itself part of this argument, since in English poetry sonnets had become the established genre most strongly associated with love poetry and with its most prestigious English practitioners of the sixteenth century. Herbert’s intensely personal decision to appropriate secular forms and language for sacred poetry, made around the age of seventeen, expressed attitudes and intentions not uniquely his own. Evidence that they were shared by other poets in this period is the publication of numerous collections of religious verse, especially sonnets, such as Barnabe Barnes’s A Divine Centurie of Spirituall Sonnets in ; Henry Lok’s Svndry Christian Passions, Contained in two hundred sonnets in ; Henry Drummond’s Flovvers of Sion; or Spirituall Poems in ; Donne’s two groups of “Holy Sonnets,” not published until but probably written close to the time of Herbert’s New Year’s sonnets.11 The sonnet was the single genre most often chosen by poets transforming secular poetry for sacred purposes, but others were appropriated to the same ends, especially the classical epigram in collections like Francis Quarles’s Divine
5*5-&4*/)&3#&35±4®-*55-�,¯
Fancies: Digested into Epigrammes, Meditations, and Observations in . Sonnets were usually untitled; epigram titles followed classical precedents. There were also collections of religious verse in miscellaneous forms, for instance by Robert Southwell.12 The Mvses Sacrifice; or, Diuine Meditations by John Davies, published in , contained a large assortment of poems on religious subjects individually titled, and a smaller group under the unclassical heading of “Essaies.” These collections show especially clearly the peculiar demands of titling for religious verse. Herbert almost certainly knew the religious poetry of these recent and contemporary English poets. He was clearly familiar with the secular classical and Renaissance models these religious poets as well as he often claimed to rededicate to religious subjects. An epistle prefacing a collection of Southwell’s poems by “The Author to his loving Cosen” repeats the commonplace criticism of poets for “abusing their talent, and making the follies and faynings of love, the customary subject of their base endeavours,” and defends religious poetry by reference to biblical urgings “to exercise our devotion in Himnes and Spirituall Sonnets.”13 Introducing his Partheneia Sacra . . . with Piovs Devises and Emblemes, Henry Hawkins defended his appropriations of secular verse for sacred purposes with an argument very like Herbert’s in the New Year’s sonnets: Wherin though the instruments I vse, may seeme prophane, so prophanely vsed now adayes, as Deuises consisting of Impreses, and Mottoes, Characters, Essayes, Emblemes, and Poesies; yet they may be like that Panthaeon, once sacred to the feigned Deities, and piously since sanctified, conuerted, and consecrated to the honour of the glorious Queene, and al the blessed Saints of Heauen.14
Herbert was therefore not remarkable in promising to rededicate poetry to God; he had available examples of religious verse in English in a variety of forms as he worked toward the fulfillment of that promise, ultimately embodied in the miscellaneous collection of religious poems published after his death in by the Cambridge University printers Thomas Buck and Roger Daniels with a title page reading The Temple. Sacred Poems and Private Ejaculations. By Mr. George Herbert. Evidence that Herbert worked on this body of poems over a long period and with carefully considered and consistent intentions about its character exists in two surviving manuscripts. The earlier, in which he collected poems
5*5-&4*/)&3#&35±4®-*55-�,¯
beginning probably shortly after the date of the New Year’s sonnets, contains revisions in his own handwriting that suggest work over a considerable period. The later manuscript, in which he made further revisions, some rearrangements, and omissions, and added around ninety poems, is in handwriting showing that it was copied at Little Gidding under the care of Nicholas Ferrar, who prepared a title page for the volume and made some corrections in the text in his own hand. It is thought to represent Herbert’s final intentions for the text and perhaps to be the copy from which the first edition of the poems was prepared.15 There is nothing unexpected in the fact that Herbert wrote a collection of religious poems, considering his announced dedication and in the context of parallel efforts by his recent predecessors and contemporaries, whose work he is likely to have read. Yet it is precisely in that context that we should find the nature of Herbert’s collection itself very surprising, in ways reflecting his special intentions as to its character and functions. These surprising qualities can be seen most immediately in the titles of the poems, which are astonishingly radical in more features than the individual expressiveness that makes many of them an integral part of the particular poems to which they attach. It is very unusual in this period for a poet to have a large number of poems transcribed in a manuscript under running titles at the top of each page and with individual titles for each poem. Even printed books in this period were not often so carefully arranged, manuscripts of miscellaneous poems even by a single poet much more rarely. Yet both the authoritative manuscripts of Herbert’s poems, like the first printed edition, have a particular title in the title space above each poem, and group them under running titles that are the same as those across the tops of pages in the volume: “The Churchporch,” “The Church,” “The Church Militant.” Herbert’s meticulous attention to these details is epitomized by the fact that in the earlier manuscript he inserted a running title on a page where it had been omitted in a previous stage of transcription.16 He preserved this carefully ordered arrangement in its outlines, despite some revisions and much expansion, from the earlier to the later manuscript, showing that it was an essential feature of his design for his poems. His arrangement of them in a meticulously ordered manuscript is in contrast with the treatment of shorter poems common among other poets, who seem most often to have written and circulated them on loose sheets, called “a Copie of Verses,” as Walton, for instance, described Donne’s “The Baite”
5*5-&4*/)&3#&35±4®-*55-�,¯
in in The Compleat Angler, even when they had long been in print.17 The prefatory letter to a posthumous volume of Edward Lord Herbert’s Occasional Verses refers to it as “This Collection of scattered Copies of Verse, composed in various and perplexed times.”18 The phrase “Copies of Verse,” while describing the way unpublished poems were typically transcribed and transmitted, may also have served to distance them from the poet so that he would not seem to have been personally responsible for making them public. George Herbert’s unusual treatment of his own poems in manuscript suggests that he must have begun quite early to think of them differently as a “little Book,” with implications for how he conceived of them individually and as a collection. The fact that he assigned titles to all the poems at a time when that format had not yet come to be considered essential, in a manuscript not even expected, itself shows the special importance Herbert gave to titling. Its value to him as a dimension of his poetry is further emphasized by the fact that he made his titles follow certain principles that he must early have set down for them and then obeyed with almost unvarying consistency over a long period and for a large body of poems. Such consistency in titling of miscellaneous shorter poems is remarkable in this period; Ben Jonson’s rare earliest printed poems stand out as the chief exception before Herbert’s. He saw to the publication in of a collection of secular poems in various forms—single and grouped couplets, epitaphs, an epistle, an anagram, a narrative fable—for which he chose the generic title Epigrammes, a term tracing back to Martial.19 He used for each of the poems (with only four exceptions) one or the other of two classical formulations for the genre of epigram: titles beginning with To mainly for addressing praise or with On more often for dispraise, as in “To Iohn Donne,” “On English Movnsier.” His care clearly shows in the volume both in fidelity to traditional generic titling and in consistency of formulation. His titles were individually chosen but follow generally recognized rules. Herbert consistently limited his choices for titles of individual poems—in both manuscripts—according to a few principles, but of very different kinds from those applied by Jonson. Herbert’s titles can be grouped for purposes of discussion in five types, which can include all the poems and which can be described along fairly clear lines, although of course a title (for example, a pun) can sometimes be classified according to more than one of these types. Herbert’s guiding principles for these chosen types of titles pertain first of
5*5-&4*/)&3#&35±4®-*55-�,¯
all to their substance—that is, how they purport to identify the poem, what dimension or feature of it they call to the reader’s attention most explicitly. He also established rules for how the various types of titles would be grammatically formulated, itself a device for directing how the poem should be read. Although none of the five types is absolutely without precedent in other collections of religious verse, the proportions and treatment of them in Herbert’s are unique and, in their uniqueness, revealing of his special intentions. Counting the poems in the later manuscript, which correspond to those in the first edition of The Temple, twenty loosely constitute a type in that they all purport in their various ways to identify the poem explicitly by an aspect of its form—its mode of expression or address, its style, shape, or rhetorical status: “Complaining,” “Sighs and Grones”; “The Call,” “The Answer,” “The Invitation”; “The Quidditie,” “A Wreath,” “The Quip.” Other titles refer to types of singing: “A true Hymne” and “Antiphon” (used twice, once in substitution for the earlier title, “Ode”). “A Dialogue-Antheme” and “A Parodie” are terms used for both musical and written kinds of composition, and “The Reprisall” is a punning title that includes the meaning of a musical response.20 Only four titles associate the poem with a distinctly literary genre: “Anagram,” “Dialogue,” “The Poesie,” “L’Envoy.” A small number of titles identifying the poem as belonging to a formal category of verse is not in itself unexpected in a miscellaneous collection of religious poetry at this time, when the paucity of recognized sacred genres created the need for other kinds of titles. Yet we shall see surprising features in Herbert’s choices and treatment of his few generic titles. Another type of title Herbert chose for a dozen poems uses names for church rites, like “The H. Communion” or feasts and seasons like “Easter” and “Lent”; he gave to sixteen poems titles taken from biblical texts or names or events: “The H. Scriptures,” “The Psalme”; “Marie Magdalene,” “Jordan”; “Dooms-day.” Both types purport to identify the poem by its subject or by a key image or text associated with it rather than by its form. In other collections of religious poems, titles like these are used, but in greater proportion, by poets such as Southwell, Davies, and Donne; they occur also in Donne’s sonnets of the “La Corona” sequence. Herbert titled some forty more of his poems with nouns referring to other than biblical persons, things, or actions. They purport to identify the poem by its subject or a key image associated with it. In these titles Herbert regularly used The before the noun (exceptions are some five isolated examples
5*5-&4*/)&3#&35±4®-*55-�,¯
without an article, such as “Artillerie,” and the group “Church-monuments,” “Church-musick,” “Church-lock and key,” “Church-rents and schismes”). Titles of this type also appear occasionally in collections of religious verse by other poets of the period, especially with nouns having distinctly religious meanings, for instance “The Crosse,” used as a title for poems by Donne and Herbert. This type of title but with nouns having secular as well as religious associations controlled by context, like Herbert’s “The Rose” or “The Pearl. Matth. ,” became common in sacred emblem poetry, of which most English volumes were published after The Temple.21 These also extended the range of nouns to include names for objects distinctly secular in their associations, like Herbert’s “The Pulley” and “The Bag.” It is his use of such emblematic titles for expressive purposes that has been given most attention in recent critical discussions, along with his titles of this type that combine secular and religious meanings in punning nouns like “The Foil” and “The Collar.” Herbert gave by far the largest number of poems in his collection a type of title consisting almost unvaryingly of a single noun referring to an abstraction purported to be the subject of the poem, unmodified by an article, a preposition, or an adjective. Not only did he prefer this type, which he used for close to seventy poems, but he repeated particular titles of this type more often for two or more poems: “Affliction,” “Employment,” “Justice,” “Love,” “Praise,” “Prayer,” “Sinne,” “Vanitie.” His remarkable preference points to his special intentions because this type of title was extremely rare in English religious poetry, scarcely less so in secular verse. There are only a few titles of this type in a very small number of collections of epigrams and emblems, usually for secular poems. In one such volume by Francis Thynne published in , the seventy-five poems headed “Epigrames” include seven, while the majority of titles use generic formulations for epigrams: “Of ——,” “To ——,” “Vpon ——,” “That ——.” Among the sixty-four “Emblemes” in the same volume, eighteen have titles of the type Herbert preferred; the other titles vary widely, from classical formulations for epigrams to proverbial sentences.22 There are also a very few titles of this type among epigrams written and translated by Robert Hayman in a volume published in .23 I have found no instances in collections of holy sonnets or in volumes of miscellaneous religious poems in various forms. In this context Herbert’s preference for titles of this type is remarkable; it is made specially pointed because he chose to use a high proportion of them in his collection, formulated in precisely the same way and sometimes repeated—“Affliction”
5*5-&4*/)&3#&35±4®-*55-�,¯
three times in the earlier manuscript, five in the later—for poems widely spaced in the collection. Herbert’s originality here is made very clear not only by the rarity of this type of title in collections by other poets but by the kinds of changes made in Herbert’s when they were transcribed into manuscripts by early readers or reprinted by editors of poetic miscellanies. They changed them in the direction of making titles of this unusual type conform to more familiar models: Walton described “Providence” as Herbert’s “Divine Contemplation on Gods Providence”; “Constancie” was referred to in wording that called attention to its generic status as a character poem: “The description of a CONSTANT MAN.” More often copyists altered Herbert’s titles of this type to fit a generic formulation by simple addition: “On Peace,” “On Sinne,” “Of Giddiness,” “Touching Prayer,” or more arbitrarily “Content” was copied under another formulation for epigrams, “Against Vaineglorie.” “Life” was renamed for its central image, “The Poesie”; and “Love” was expanded to “The Divine Love.” Titles of Herbert’s belonging to other types were also changed, although not so often; for example, “Charms and Knots” was renamed both “Epigramme” and “Riddle.”24 Although such changes in individual titles show the uneasiness of copyists and editors with Herbert’s unfamiliar choices (especially titles of the type he preferred), groups of his poems in early copies also show the transcribers’ indifference to his principles of consistency in titling. One such group of ten published in a miscellany of printed the first “Jordan” poem untitled and the other entries with titles all varying from the originals but in assorted ways. For example, “Charms and Knots” became “Epigramme,” “The Quidditie” was titled “What verse is and is not,” “The Sinner” was altered to “The sicke Sinner,” “Man” was expanded fully to “On what man is: being some Meditations on a Sermon preached that morning.”25 Such alterations of Herbert’s titles in early transcriptions point out what we can learn from his own careful treatment, both in the rules he proposed and quite strictly followed for them, and in the current practices and conventions he chose to avoid. Following these two lines of exploration, we can trace the design of Herbert’s titling. It is striking that he omitted titles naming a secular genre, even those commonly appropriated for sacred verse like the sonnet or the epigram, although his collection includes some fifteen sonnets as well as other poems that early readers renamed as epigrams. Perhaps it was to conform to this pattern that he omitted from both manuscripts the two early New Year’s sonnets, which
5*5-&4*/)&3#&35±4®-*55-�,¯
explicitly call attention to their generic status. In accordance with the same principle, he also avoided grammatical formulations for titles traditionally associated with secular genres like the epigram: “To all Angels and Saints” is his only title beginning with a preposition (by contrast with his Latin poems, the great majority of which use Ad ——, In ——, De ——).26 Still more surprising in the context of other collections of religious verse, Herbert even excluded titles that place the poem as an example of a distinctly sacred mode of utterance, such as “A Prayer to ——,” “A Thanksgiving for ——,” “A Confession of ——,” “A Meditation on ——.” Walton substituted this kind of title for Herbert’s preferred type when he described “Providence” as “his Divine Contemplation on Gods Providence,” a formulation that points to the performance of the poet whose presence is ignored in Herbert’s original title. Among his few titles that do identify the poem as belonging to a recognized art form, his preference was for sacred musical compositions usually sung by more than one voice: “A true Hymne,” “A Dialogue-Antheme,” “Antiphon.” He seems to have acted according to the same principle of shifting away from titles that focus on literary performance when he changed his early titles of “Ode” to “Antiphon,” “Poetry” to “The Quidditie,” “Invention” to “Jordan.” This calculated displacement of emphasis is still clearer in Herbert’s decision to exclude any reference in titles to the presence of the author, such as in the title of Donne’s “A Hymne to Christ, at the Authors last going into Germany,” or references by personal pronoun as in “Hymne to God my God, in my sicknesse” (from the second edition of Donne’s poems).27 He even excluded references to the poet by third-person pronouns (such as Jonson for some secular poems and Herrick later for religious verses used in titles of their own devising to imitate editorial distance).28 These exclusions are particularly pointed in titles for the many poems where Herbert speaks in the first person of “my verse,” “my lines,” “my rhyme,” or tells of how “I live and write . . . and relish versing.”29 Herbert also persisted with extraordinary consistency in leaving out not only titles referring to the poet but titles that name any other speaker. His choice was remarkable because these types of titles were especially common in religious as well as secular poetry of the period. Some, following biblical example, named a particular speaker, as in Donne’s “The Lamentations of Ieremy,” a formulation Herbert could have imitated for “The Sacrifice.” Much more common were titles identifying the poem as the utterance of a general-
5*5-&4*/)&3#&35±4®-*55-�,¯
ized or exemplary speaker, for instance, Davies’ “A Confession of a Sinner, acknowledging the miserie of humane frailtie,” “A Sinners acknowledgement of his Vilenesse and Mutabilitie.”30 In the expanded manuscript Herbert kept only one title, “The Sinner,” which names the speaker who describes himself in the first person, but since this title—unlike Davies’—does not identify the sinner’s mode of utterance, or subject, or condition, it functions almost as do titles that name an image associated with the poem rather than its speaker. One other title in the earlier manuscript, “The Publican,” named a figure described in the text in the third person to illustrate that “Man is a foolish thing.” Conforming this title to his preferred type, Herbert made a radical substitution in the later manuscript, retitling “The Publican” as “Miserie.”31 Herbert also excluded from all his titles other specifying references, for instance, to occasion or setting. His “Good Friday” is not grounded by the speaker’s presence in time and space as in the title of Donne’s “Goodfriday, . Riding Westward.” Herbert also left titles naming modes of expression like “Sighs and Grones” and “Complaining” wholly unlocalized, setting aside more conventional formulations such as the title of Davies’ “The sighes of a Pensiue Soule, groaning under the burden of sinne.”32 With similar effects, in titles that identify the poem as a distinct mode of address, like “The Invitation” or “The Answer,” Herbert did not specify to whom they are addressed or in what situation, by contrast with the titles given to Donne’s hymns “to Christ” or “to God” in “sicknesse” or on “last going into Germany.” Herbert’s are unlocalized so that they exist as categories. This status is supported by his consistent choice of the article the in these titles: “An Answer,” a title common in this period, would emphasize the poem as an example of a mode, before the category it exemplifies. He achieved a similar effect in titles of the type that identify the poem by an image associated with it, which he always formulated as “The Rose” rather than “A Rose,” “On a Rose,” “To a Rose,” titles used in collections by other poets. The tendencies among all types of titles in Herbert’s collection away from performance and toward categorization, even abstraction, are of course most fully embodied in the type of “Affliction,” “Love,” “Vanitie,” which can suggest a reason why he markedly preferred them. Another may be that titles of this type were so very rarely used in other collections of either sacred or secular poetry that they were virtually free of associations with the verse of other poets. In one respect they resemble conventional titles that name a genre of poetry or use a formulation associated with it, in that they constitute
5*5-&4*/)&3#&35±4®-*55-�,¯
a category under which the poem is classified; however, they do so with very different effects. The titles that early readers gave to Herbert’s poems, like “Epigramme” or the more surprising substitution of “Against Vaineglorie” for “Content,” introduce the poem as an example of a poetic genre that provides a form chosen by the poet for his individual performance. Avoiding them, Herbert used titles like “Content” to introduce the poem under a head—in this period also called a topic or place. It suppressed the source and form of the poem while associating it with entries in a recognized kind of collection of writings, the commonplace book. With its origins in ancient logic and rhetoric, the commonplace book was revived by continental humanists like Agricola, who advocated a method of arranging materials under general and familiar topics (for instance virtus, vitium, vita, mors), and like Erasmus, whose voluminous and influential compilations were ordered according to similar methods.33 One of the earliest and most popular English commonplace books was William Baldwin’s A Treatise of Morall Philosophy, published in eighteen editions between and . Materials in it were grouped under variously formulated headings including “Of Vertue,” “Of Repentance,” “Of Death not to be feared.”34 Sir Thomas Elyot’s prestigious collection of moral sayings, Banket of Sapience, was also published in numerous editions beginning in the earlier sixteenth century and continuously used.35 It contained nearly entries, more than half from Christian sources. They were arranged alphabetically in both the text and the table of contents under topics consistently formulated in the same way as Herbert’s favorite titles, and including some identical with his: “Auarice,” “Confession,” “Constancy,” “Deathe,” “Faithe,” “Grace,” “Justice,” “Judgement,” “Prayer,” “Prouidence,” “Vertue.” Herbert’s avoidance of formulations with prepositions in imitation of classical titles would have led him to prefer topics formed simply by a noun, like Elyot’s. These collections of quoted materials, often called storehouses, were stocked and arranged explicitly for the benefit of writers, to supply them with copy or copiousness on a variety of subjects. Whereas for early humanists the compositions to be supplied were typically orations and other classical genres (commonly assigned as school exercises), Protestant English writers emphasized the value of commonplace books for preaching. For this purpose Herbert’s ideal “Countrey Parson hath read the Fathers . . . and the Schoolmen, and the later Writers, or a good proportion of all, out of which he hath complied [sic] a book, and body of Divinity, which is the storehouse of his
5*5-&4*/)&3#&35±4®-*55-�,¯
Sermons.”36 Compilations were also prepared for more general use to aid understanding of the Bible for readers as well as writers. Herbert, describing his country parson as a model, recommended “a diligent Collation of Scripture with Scripture. For all Truth being consonant to it self, and all being penn’d by one and the self-same Spirit, it cannot be, but that an industrious, and judicious comparing of place with place must be a singular help for the right understanding of the Scriptures.”37 Under such commonplaces the collated passages were arranged. These were to guide readers among dispersed biblical verses related to the same topic, as Herbert describes them in the second of paired sonnets together titled “The H. Scriptures”: Oh that I knew how all thy lights combine, And the configurations of their glorie! Seeing not onely how each verse doth shine, But all the constellations of the storie. This verse marks that, and both do make a motion Unto a third, that ten leaves off doth lie: Then as dispersed herbs do watch a potion, These three make up some Christians destinie . . .
Herbert actually tells us here about the design of his own titling. His method, not used in any other collection of English poetry in this period, of repeating the same topic for the titles of widely “dispersed” poems could aid readers to trace “configurations” or “constellations” in his “verse” as they were trained to compare one biblical “verse” with another in commonplace books. What he describes is not a structure imposed as part of the maker’s performance but an exercise to be performed by readers seeking light on their own spiritual “destinie.”38 The possibility that early readers might have recognized and responded to affinities between Herbert’s titles and topics in commonplace books is strongly supported by the format of The Temple as it was printed in the first and all subsequent seventeenth-century editions. The volume followed the later of the two manuscripts by listing at the end all the poems included. In the manuscript their titles were set down beginning on a fresh page without heading in order of the poems’ appearances in the text. Titles repeated for poems widely scattered in the collection were also repeated where they came in the sequence in the list without cross-references. Poems were counted in the table (but not in the text) from to by numbers to
5*5-&4*/)&3#&35±4®-*55-�,¯
the left of their titles, and page numbers were given at the right. A list in this form would have been useful to a copyist or editor as a way of checking the number and order of poems reproduced, and was perhaps prepared for that purpose, but it would have been of much less help to readers. By contrast, the table (Figure ) printed at the end of The Temple under an ornamental design and headed “The titles of the severall poems contained in this book” is arranged alphabetically. Titles repeated for several poems are listed only once, in their alphabetical position, with the page numbers for each repetition given together. Even early editions of English commonplace books, like Elyot’s, included a table of contents. By Herbert’s time tables were virtually obligatory for such a volume, but scarcely even expected for collections of verse. Neither the first edition of Donne’s poems nor the much more carefully arranged second edition lists the contents, and even Jonson’s meticulously prepared volume of Epigrammes numbered them in the text but included no table. In rare instances when such lists were printed in collections of shorter poems, they were by title in order of appearance, as in Michael Drayton’s The Harmonie of the Church of , or for untitled poems still more rarely in sequence by first line, as in Thomas Campion’s A Booke of Ayres of . Barnes’s and Lok’s untitled sonnets were tabled alphabetically by opening phrases.39 Whoever decided on the inclusion and order of the table of titles in The Temple, presumably the printer, apparently saw that the titles in Herbert’s manuscript followed patterns that lent themselves to alphabetical arrangement (as those in the first edition of Donne’s poems, for example, would not). He also seems to have seen that Herbert’s titles would allow readers to find and compare poems by topic, as Herbert described in “The H. Scriptures,” which was their habit with entries in commonplace books. Accordingly the table at the back of The Temple, reproduced in Figure , was arranged in a format characteristic of such printed compilations. The example in Figure is from a commonplace collection of miscellaneous verse, Bel-vedere or The Garden of the Mvses. It was first published in , again with the same entries in , under the patronage of John Bodenham, a notable collector and compiler also associated with another immensely popular prose compilation printed with a very similar table, Politeuphuia. Wits Commonwealth. As the illustrations show, this table was not only printed in a form that could have been the model followed in The Temple but included entries resembling
;>< JG : Title index from George Herbert’s The Temple: Sacred Poems and Private Ejaculations (Cambridge: T. Buck and R. Daniel, ). Courtesy of Wellesley College, Special Collections.
An Alphabeticall Table, of the feuerall things handled iA thi~ Booke. .
I
. Botendttnce. vide Riches. Ab(ence.· vide LoueandFr~endjhip. Abjlinence. vide Gluttonie. vide Iriflice. Accufotion. I·,I7,I9,50,67,&c. A.Clions. · 4·o Loue. . vide Admiration. vide Cormfefl,Aduife,&c. Admonition. 11. vide Poreertie. Aduerjitie. 73 Aduife. vide Lufl. Adultcrie. I6I t..Affeelion. · A(flielion. 22Z Age. l:o8 Ambitim. Amitie.
A
.
; >< JG : Subject index from George Herbert’s table to Bel-vedere or The Garden of the Mvses (), reprinted from the original edition for the Spenser Society, issue no. , . Courtesy of the Houghton Library, Harvard College Library. *EC.HT..
5*5-&4*/)&3#&35±4®-*55-�,¯
and even identical with Herbert’s. The complete table of Bel-vedere contains, in addition to Herbert’s favorite title of “Affliction,” a number of other entries typical of commonplace books and also identical with titles in The Temple: “Auarice,” “Constancie,” “Death,” “Faith,” “Grace,” “Greefe,” “Heauen,” “Hope,” “Iudgement,” “Iustice,” “Life,” “Loue,” “Man,” “Misery,” “Nature,” “Obedience,” “Peace,” “Praise,” “Prayer,” “Prouidence,” “Repentance,” “Sinne,” “Time,” “Vanity,” “Vertue.” Such close resemblances would encourage the notion that The Temple was meant to be read, aided by a suitable table, as Herbert’s contemporaries were in the habit of reading commonplace books. Evidence that the poems may have been read that way comes from two telling sources. In Barnabas Oley’s preface to Herbert’s Remains, published in , the author describes the poems as if Herbert himself had compiled them as entries in a commonplace collection consisting of biblical texts and “passages” collated from theological writings and religious controversies (and under some topic like “Doctrine” Oley here made a brief collation of his own): And he that reads Mr. Herbert’s Poems attendingly, shall finde not onely the excellencies of Scripture Divinitie, and choice passages of the Fathers bound up in Meetre; but the Doctrine of Rome also finely and strongly confuted, as in the Poems, To Saints and Angels pag.. The British Church pg.. Church Militant, &c.40
In Walton’s Life of Mr. George Herbert, published in , he reported that “Mr. Farrer would say” of The Temple “that the whole Book, was such a harmony of holy passions, as would enrich the World with pleasure and piety.”41 Whether or not the quotation is precisely authentic, the word harmony was an appropriate attribution that would have had for Ferrar a more pointed meaning than the general or even the musical sense of an orderly arrangement. In this period harmony was also the term for a collation of passages from different writings on the same topics, usually religious, such as the famous concordances of biblical texts—called harmonies—prepared at Little Gidding.42 The quotation attributed to Ferrar shows that he or at least Walton thought The Temple could fittingly be called such a harmony. Herbert himself seems to have encouraged this recognition in deliberately giving prominence to titles not traditional for a book of poems but identical with topics in a commonplace book; by using many more of them than any other type of title; by spacing them throughout the collection; by repeating some for poems “ten leaves off” and more, a device expanded in the later
5*5-&4*/)&3#&35±4®-*55-�,¯
manuscript; by limiting his other titles in type and formulation so that they would share features of the preferred commonplace topics such as we have seen: their tendency toward categorization or abstraction, and away from localizing particulars and literary associations or performances. That is to say, Herbert seems to have used titling as a complex device for defining his “little Book” and how it should be read. The design of his titling suggests a different conception of sacred poetry from the earlier one, shared by other poets of the time, which Herbert announced in the letter to his mother and in the sonnets exemplifying it. They explicitly point to the poet’s appropriation for sacred ends of the form and language of a genre associated with secular love poetry, so that for them the title “Sonnets” would have been a fitting way of calling attention to the poet’s act of appropriation. This would be the effect if Herbert had, for example, chosen for the poem he called first “The Passion,” later retitled “Redemption,” a conventional title like “Sonnet” or “A Contemplation upon Redemption” or the abbreviated formulation “On Redemption” (a noun like Sonnet or Contemplation being presupposed). His carefully chosen title in its different formulation shifts attention away from the poet, and from the poem as an example of his performance in a literary genre. To strengthen this effect, Herbert for widely spaced poems with the same title never used the same form. In its association Herbert’s title allowed the reader to approach the text as if it were an entry in a commonplace book under the topic “Redemption.” In the fiction enacted, the volume is then a compilation of passages collated under commonplaces, that is, traditional topics drawn from an established corpus of writing (examples have shown that the same topics were included in volume after volume). The passages themselves are the “sayinges of the wise,” copied by the compiler to be consulted by himself or any reader seeking light on the topic.43 Both the title of the poem—its status as a common place assures—and the verses under it, imagined as having been copied by the compiler from another text, became as it were detached from the poet, whose existence is not acknowledged in the fictional transaction. The title therefore works as an important instrument for achieving the qualities of “transparency or self-effacement,” as Helen Vendler has named them, which were not part of Herbert’s original program for sacred verse, but which critics have recognized as ideals he later strove for in his poetry.44 The fiction of locating himself imaginatively in the situation of a compiler of sacred commonplaces could have been a means of escaping poetry associated with human invention
5*5-&4*/)&3#&35±4®-*55-�,¯
and entwined with self. It could have helped him to write as if performing the humble copyist’s task dictated to the poet-speaker in the second poem called “Jordan” (a substitution for its earlier title, “Invention”): There is in love a sweetnesse readie penn’d: Copie out onely that, and save expense.
Ultimately Herbert’s intention in privileging titles like entries copied into a commonplace book may have been to suggest a more particular association with the Psalter. This was the Old Testament book recited most often in the liturgical worship of the English Church. It was repeated in regular rotation at morning and evening prayer in a thirty-day cycle, and was recited twice through every twenty-four hours at Little Gidding.45 The Psalms were also specially recommended for private prayer. Donne urged the members of his congregation to use them because they ministered “Instruction, and satisfaction, to every man, in every emergency and occasion,” preferring them—“being Poems”—before all other books of the Old Testament; Herbert’s country parson recommended them to his parishioners for singing “at their work.”46 To aid private use of the Psalter as a “harmony of holy passions” for the encouragement of piety, it was often printed with some sort of “Table, shewing wherevnto euery Psalme is particularly to be applied,” where the reader could find psalms collated like passages in a commonplace book.47 The English translation of The Whole Boke of Psalmes, collected into Englishe Metre most commonly used for congregational singing was often printed with such a table to aid private devotion with instructions how to apply them: “ If thou wilt laud God with a psalme or Himine, sing the . psalmes”; “ If thou feelest Gods dreadfull threates, and seest thy selfe afrayd of them: thou mayst say the .. Psalmes.”48 In much the same way, seventeenth-century editions of The Temple, beginning with the seventh in , were printed with the table of titles moved to the front, and in back what the title page announces as “an Alphabeticall Table for ready finding out chief places” in the text.49 Reproduced in Figure is the first of thirty-five pages collating passages from the poems with suggestions for their use under topics some of which, like “Affliction,” are identical with the titles of the poems. The association of The Temple with the book of Psalms and the practice of using the two collections of poems in the same ways for private devotion must have been well established to have suggested the need for such a table, and was likely to have been perpetuated by it.
A
·T AB LE . ~
I
Aron's garme~tf jhould he ftiO wor n hy Mi.; nijlers P:1ge t68 Lin e t? Abrahat;J /;!(ought R.tlrgion wit h btm from the Ea/} 1 84-. 1 9 J/;flitJtnce, bow profitable . 79 . 7 Jbuj'e of things taf(eth not awa y their tlfo 7 9· 16 Ab11jivu1efs, the Jcum of wit 8, 29 9. I Jccount, fee Rul es Jllton, Tht glory of an aElion u, to do if for Got!s J.IDrJ I 7 8 · %. I ActiVe fPirits one !yli ve . ' IZ. · 13 71· 3 Jdorll tiQn of Sai nts , why unl awf uU 70. r A ft1fl1on jucceeaeth projperi ty ) 8. 15 , &c. H u Not ro be grieved for 164 . 1 I. or rather grie f for afjlvtiion u to he tu;·ned into gri ifjo r fin 164 . I 1• ho!JJ to carry our jelves thertiH 40. 7 it is advatlltt_(e to a ChrifltaH j). 90. 7· r 14. 2. 5. .njflt Z.1ons .;,/drotl helpeth to jup;le the lmJTt 12.% . 17 -2 .9 l 13: 1 n.fjl tflion to C~rifti. ms, J11(e rbe prtm ing •t}f to Trees 116. 2 ajjl tato tH compare d ~o Mol e; 1.19 . 1 aU ~ur afjl illio m, tJ()th tHg to Chr d l's fuj~ jtri•gs 53· 26 Chr ift ha1h his 11rt in otcr af· #•llsens 6 4· t. 7 6 s. 7 89. 2. S JLns, rl·t mrJ/} tbriving tr,Jde 89.3 mus ivu shereutJtl IJ .9, &c. jee Ru 'cs. •r , jtt Gvds.
A
-
I
; >< JG : Subject index from George Herbert’s The Temple: Sacred Poems and Private Ejaculations (London, ). Courtesy of Wellesley College, Special Collections.
5*5-&4*/)&3#&35±4®-*55-�,¯
Other evidence that The Temple was particularly associated with the Psalter as a devotional commonplace book is in Walton’s Life, in the phrasing of the message said to have been spoken by Herbert from his deathbed about the disposition of his “little Book.” Whether the quotation describing the collection is in Herbert’s own words or merely represents what he might have said, Walton must have thought it an appropriate characterization of the book and must have believed it would seem so to his and Herbert’s readers, who would also have been able to recognize the resemblance of the description to traditional characterizations of the Psalter: I pray deliver this little Book to my dear brother Farrer, and tell him, he shall find in it a picture of the many spiritual Conflicts that have past betwixt God and my Soul, before I could subject mine to the will of Jesus my Master, in whose service I have now found perfect freedom; desire him to read it, and then if he can think it may turn to the advantage of any dejected poor Soul, let it be made publick; if not let him burn it.50
This description seems to have been modeled on familiar ways of describing the Psalms as a collection of poems made up of materials of special value to individual readers above other books of the Bible. One such representative description, and of particular authority for English Protestants, was translated by Arthur Golding in from Calvin’s epistle to the reader prefacing his commentaries on the Psalms: The holy Ghost hath heere lyuely set out before our eyes, all the greefes, sorowes, feares, doutes, hopes, cares, anguisshes, and finally all the trubblesome motions wherewith mens mindes are woont to be turmoyled. The rest of the Scripture conteineth what commaundementes God hath enioyned to his servantes to be brought vnto us. But in this booke, the Prophets themselues talking with God, bycause they discouer all the inner thoughtes, do call or drawe euery one of vs to the peculiar examination of himself, so as no whit of all the infirmities to which we are subject, and of so many vyces wherwith we are fraughted, may abyde hidden.51
Using language close to that in the description Walton attributed to Herbert, George Wither stated his intention to illuminate the Psalms for private use by the reader of A Preparation to the Psalter, published in : How among them he may find true comfort in euery necessitie, would he well enure himselfe to these Poems. For, there are no temptations, affections, nor
5*5-&4*/)&3#&35±4®-*55-�,¯
afflictions, which a Christian may be subiect vnto, but in the Psalmes he shall finde both the formes of expressing them, and their means of remedie.52
Wither also published his own translation of The Psalmes of David in , with a prefatory letter to “Elizabeth Princess of Great Britaine” instructing her how to “make personall Application of these Hymnes vnto yourself”: For, they were Afflictions or Deliverances from Afflictions, which were occasions of euery Psalme: Therfore, none but they who have bene afflicted, cann relish the sweetnes, or understand, the depth of these Raptures . . . those many which are appliable . . . are the very same Afflictions, wherof (as a type of Christ) King David complained.53
The special dimension that the association with the Psalter added to the more general connection of Herbert’s collection with religious commonplace books can be located in what seems to have been the understanding of how “euery Psalme”—and by analogy any poem in The Temple—“is particularly to be applied” by the reader. Descriptions of the Psalter and The Temple as well as the tables collating the poems in them suggest that something more seems to have been meant than was expected of readers’ responses to other collections of religious verse. Their titles typically identified them as portraying a representative Christian’s generic inward experience, as in Lok’s Svndry Affectionate Sonets of a Feeling Conscience of . Titles of individual poems, for instance, Southwell’s “A Phansie turned to a sinners complaint” or Davies’ “The longing of the Soule to be with God,” claimed to be representative in the same way.54 Such titles fittingly identified the poems as the utterances of exemplary speakers in common situations like sickness, need, or peril; expressing generalizable feelings like fear, grief, comfort; performing ritual acts of confession, thanksgiving, praise. The reader was invited to recognize these portrayals as so general that they equally included everyone. Herbert, we have seen, entirely avoided such titles, preferring topics drawn from common wisdom. Paradoxically this decision then allowed him to enter poems that are generic portrayals of human “Affliction”—Louis Martz has pointed out that the word is used in noun and verb forms more often in the translation of the Psalms than in any other book of scripture—but also poems even under the same topic that draw on more particular and personal materials.55 The printer or whoever it was who added to the title page of The Temple the subtitle Sacred Poems and Private Ejaculations (not on the title page prepared by Ferrar) may have done so to point to this dimension of Herbert’s
5*5-&4*/)&3#&35±4®-*55-�,¯
poems. Walton in the Life also recognized it when he described “a Copy of Verses in his printed Poems; ’Tis one of those that bears the title of ‘Affliction’: And it appears to be a pious reflection on Gods providence, and some passages of his life.”56 Under impersonal topics like “Affliction,” “Employment,” “The Priesthood,” “Dulnesse,” “Deniall,” “Obedience,” “Home,” “Jordan,” Herbert could enter such “passages of his life” as the souring of his academic successes; his lack of occupation; the difficult choice of his vocation and its constraints; above all, the struggle to write poetry cleansed of self. Because they were not titled as performances of a poet or utterances of a speaker but as if they were entries copied under a commonplace, these poems could be at once detached from self-presentation and yet autobiographical. The inclusion of such poems, a considerable number of them added in the later manuscript, made his collection capable of being read as modeled on the Psalms in more particular ways than those of his contemporaries. For the Psalter was read by English Protestants in the sixteenth and seventeenth centuries simultaneously as a mirror of everyone’s inward experience and as a record of David’s (and in Calvin’s view also of other “Prophets”) personal struggles in particular situations. This understanding is reflected, for example, in Wyatt’s English versions of the penitential psalms, published in .57 The verses themselves are complaints, confessions, prayers, suited to congregational recitation and to the private devotions of every soul in affliction and desiring to repent. They are introduced by narrative prologues locating them in David’s biography so that they are grounded in particular experiences, as are poems in The Temple. The same ways of reading the Psalter as both generic and personal are suggested in the tables aiding readers’ particular application of the Psalms to themselves. Some entries guided them to recognize their participation in man’s fallen nature and condition: “ If for the imbecilitye of thy nature thou art wery with the continuall miseries and griefes of this lyfe, and wouldest comfort thy selfe, sing the . psalme.” Others directed readers to make “personall Application” of particulars in the Psalms to their individual situations, not just to see themselves in them as a member of a category—the soul, the sinner, the feeling conscience: “ If so agayne thou wilt sing in geuing thankes to God for the prosperous gathering of thy fruites, vse the . Psalme”; “ If thou art elect out of low degree, specially before other to some vocation to serue thy brethren, aduance not thy selfe to hye agaynst them in thine owne power, but geue God his glory who did chose thee, and sing thou: the [sic] Psalme.”58 The extensive and elaborately detailed
5*5-&4*/)&3#&35±4®-*55-�,¯
“Table for ready finding out chief places” printed with The Temple may have been designed on the model of harmonies of the Psalms to encourage more “personall Application” of particulars in the text than would have been served by the table of titles resembling topics in a commonplace book. It would have made possible the exercise of tracing how “dispersed” verses “make up some Christians destinie,” which Herbert instructed his readers how to perform in “The H. Scriptures.” The associations of Herbert’s poems with collections of religious commonplaces and with the Psalms as a “harmony of holy passions” were suggested by his carefully designed and consistently followed principles of titling. By submitting his poems to that self-imposed discipline, he achieved for them a freedom to be at once anonymous and intensely personal, detached from self and yet autobiographical, representative but private, gathering particulars into simplicity. These paradoxes are something understood in the poems. The likening of his collection to a “harmony” for use in private spiritual exercises may also have granted him another kind of freedom. It may have made it possible for him to imagine arranging his poems in a “little Book” that might one day be “made publick,” a plan that would otherwise most likely have been distasteful to a poet at this time who belonged to a great family and who was a priest.59 Above all to a poet who struggled for a language untangled from self and from human invention, such preparation of his own poems in a volume for publication would have been unacceptable unless, perhaps, he could think “it may turn to the advantage of any dejected poor Soul” who would use it as a devotional commonplace book modeled on the Psalter.
$IBQUFS
5 ) */( " / % 5 )* / ( 4 8PSETXPSUI 4UFWFOT "TICFSZ
and one of the most inescapably necessary, as anyone knows who tries to get through a sustained conversation or piece of writing without it or its plural form things. Formally, like the many other Anglo-Saxon monosyllabic nouns that gave us much of our everyday vocabulary, its uncompounded structure seems to give it the definite outline of a solid body, as is true of nouns that have clearly defined references like bird, earth, house, man, stone. According to Willard Quine: 0 $ %* # * 4 0 /& 0 ' 5 ) & 0 - % & 4 5 8 0 3 % 4 */ &/ ( - * 4 )
Linguistically, and hence conceptually, the things in sharpest focus are the things that are public enough to be talked of publicly, common and conspicuous enough to be talked of often, and near enough to sense to be quickly identified and learned by name; it is to these that words apply first and foremost.1
The simple structure of thing and its familiarity can make it seem to belong to the same class as nouns like those in the list above— until we think about it. The discussion here will begin that process by looking at the linguistic history of thing: at the stages over many centuries by which it eventually became a significantly potent instrument for poets with particular need for its unique, far from simple or single semantic possibilities. Because many of the earliest English words are structurally uncompounded nouns that name elemental or ordinary properties of daily life, we might expect—or at least I expected—that the earliest reference for thing would be a familiar material object. The OED gives for its first appearance,
5)*/("/%5)*/(4
late in the seventh century, the definition of a court or council, which soon elided into meaning a concern considered in such a place, but neither of these senses carried into Middle English. The earliest meaning to survive, surfacing in the ninth century, reconceptualized thing by immensely expanding its range of semantic reference to encompass “whatever one is concerned with in action, speech, or thought.” Unexpectedly, then, the earliest surviving sense of thing, instead of having the clearly defined primary signification typical of simple common nouns, was from its introduction into Middle English peculiarly open to a boundless accumulation of meanings that include the semantic references of every noun in English. What follows is a list culled from the OED of definitions for thing in the order of their earliest known appearances (not including slang uses that have abounded in every period: the Wife of Bath chose thynges as a euphemism for what the OED more euphemistically explains as “private parts”). Thing can refer to an entity of any kind, that which exists individually in fact or idea (); an affair, business, concern, matter (); a doing, act, deed, transaction, event, occurrence, incident, fact, circumstance, experience (); a material substance, material, stuff (); a living being, creature, plant (); a saying, utterance, expression, statement, prayer, jest (); an actual being or entity as opposed to a symbol of it (); that which cannot be described (); a being without life or consciousness; an inanimate object as distinguished from a person or living creature (). The absorption by thing of such superabundance of semantic references made it so significantly packed that it came to be indispensable to the writing and speaking of English. A sign of its inescapable presence is that its continual pairing with certain often-used adjectives led to the formation of compounds with their own significations: anything, everything, something, nothing. These compound nouns shift the emphasis to the adjectival prefix, emptying the root noun of definable meaning: any suppresses the particularity of thing; every generalizes it, expands it to be all encompassing; some dislocates it, making it indeterminate, unidentified; no, by denying it, points to absence, blankness, vacancy. These ubiquitous adjectival prefixes call attention to the mysterious, elusive character of thing. Its earliest surviving sense—whatever one is concerned with in action, thought, or speech—can, at least in some contexts, come close to being synonymous with anything or everything or something or nothing. For
5)*/("/%5)*/(4
certain poets—most markedly Wordsworth, Stevens, and Ashbery—this chameleon-like nature of thing has served their peculiar conceptual and imaginative needs. The noun thing and its plural form things are dominant in the personal idioms of these poets when they have tried to distance poems from the vocabulary of transcendence used most grandly by Milton.
Milton found a dozen places to use thing in Paradise Lost, a hundred for things, proportions that more or less agree with the proportional distribution of these grammatical forms in English translations of the Bible. There things is authorized to refer to “all things created, that are in heaven and that are on earth, visible and invisible.” This distinctively Pauline wording (invisible only appears five times in the whole of the Bible, all in Pauline Epistles) embraces a multitude of meanings that had by Milton’s time been absorbed into the noun thing. What is special about his uses of the plural form is that in some fifty appearances of it in Paradise Lost, things is modified by all, a pairing that Wordsworth adapted from Milton’s to his own intentions. As William Empson long ago pointed out, all is a “key word” or “motif ” (he counted appearances) in Milton’s epic and in his earlier poetry, where it is a prime expression of single-minded dedication to the pursuit of absolute truth.2 In Paradise Lost this purpose is held steady by the poet’s assurance that with supernatural guidance he would achieve “Things unattempted yet in Prose or Rhime”; that he would be granted transcendent vision “to see and tell / Of things invisible to mortal sight.” To Wordsworth, who used Milton’s epic in many of the ways that Milton used the Bible, these heroic acts of “Imagination to incite and support the eternal” qualified Milton’s writings along with the “prophetic and lyrical parts of the Holy Scriptures” as “grand store-houses of enthusiastic and meditative Imagination.”3 The absolute claims of the poet in Paradise Lost to transcendent knowledge of invisible things persuaded Wordsworth to believe that the Imagination is conscious of an indestructible dominion; the Soul may fall away from it, not being able to sustain its grandeur, but, if once felt and acknowledged, by no act of any other faculty of the mind can it be relaxed, impaired, or diminished.4
The phrase an indestructible dominion exudes the absolute assurance that Empson called attention to in Milton’s intonings of all, but the rest of the
5)*/("/%5)*/(4
sentence just quoted collapses Wordsworth’s confidence. The insecurity we hear in his admission that “the Soul may fall away” (from the imagination? consciousness? Eden? God?), and again in the grammatically blurred reference for the second it, exposes Wordsworth’s doubts about the possibility of visioning the invisible, unless perhaps in occasional veiled glimpses experienced by the imagination contemplating external nature. His uncertainty— in “There was a boy” Wordsworth immersed “the visible scene” in lake water reflecting “that uncertain heaven”—forced on him the imperative to find a language for expressing such intimations of transcendence as he believed or hoped to believe might come to him. Thing is a key word in Wordsworth’s visionary vocabulary. While borrowing countless phrases and their cadences, including all things, he put special reliance on his own peculiar uses of thing and especially things to distance his personal language from scriptural and Miltonic authority. The most conspicuous sign of his personal idiom is that he used the singular form in his poems some times, the plural at least (by contrast with the figures of and for thing and things in all of Milton’s poems combined). The essential fitness of this ordinary noun to Wordsworth’s urgent purpose consists, to begin with, in the simple, elemental immediacy it shares with bird and stone (by contrast with a Latinate philosophical term like entity, frequently used as a synonym for thing, but not by Wordsworth). This immediacy gives thing the suggestion of an embodied presence like bird or stone: it can stand for either. In its long linguistic history, thing has also been commonly understood to signify what we can only conceptualize, imagine, believe in; what is not confirmable by the senses; not utterable in the logic of grammar: invisible things. Wordsworth discovered in this noun a simultaneity of meanings suggesting mysterious power that might make it a space for the tenuous meeting if not the deep interfusion of external nature with the eternal. The many repetitions of thing and things in Wordsworth’s poetry prior to show his intense interest in their capacity to refer at once and equally to whatever is visible, embodied, and what is unseen, immaterial. Examples, necessarily a small sampling, of phrases, passages, and poems will illustrate how Wordsworth’s exploration of these linguistic possibilities is personal, idiosyncratic, strange by contrast with Milton’s biblically sanctioned ceremonial pronouncements. Whereas Milton’s intonations of all things are authoritatively determined, Wordsworth’s dwellings on thing, and especially things, are signs
5)*/("/%5)*/(4
of inner necessity that can make his repetitions seem at once uncertain and almost obsessive. Milton’s generic formula all things is grandly suited to scenes that celebrate “Omnipresence,” as in the blessing of the Sabbath by the “Author and end of all things”; it is appropriate to describe Edenic harmony when twilight “Had in her sober Liverie all things clad”; it conveys the indiscriminate carnage wrought in the world by Death, whose stench rises “from all things there that live.” Elsewhere Milton modified the formula by attaching a specifying adjective or descriptive phrase to suggest the innumerable things “in Heaven” and “on earth.” These expanded formulas, although more varied, have their own ritualistic cadences: “all things made,” “all things transitorie and vain,” “all external things,” “all things fair and good,” “all things visible in Heaven,” “all things under Heaven,” “all things living,” “all things new.” Although in these formulas all emphasizes encompassing magnitude, categories of things are differentiated by their place in the divinely instituted spatial, temporal, ontological, moral orders. Formulas in Paradise Lost that use words other than all to modify things (though it may be that traces of its ubiquitous sound can sometimes by heard in its absence) are built on the same hierarchical arrangements: “Ethereal, first of things,” “Things above Earthly thought,” “Things in their Causes,” “things of like kind,” “Things highest,” “things so high and strange.” Even “Things not reveal’d” can be placed in order because Milton’s language is capable of saying what they are not. In Wordsworth’s poems written before his conversion to orthodoxy, it was not his aim to escape from the authority of Milton’s biblical language, but to absorb into his own what he could that would support his uncertain hope of transcendent vision. To that end he imitated Milton’s reiterations of all things some times, and in occasional instances varied that ritual phrase by pairing things with other traditional biblical adjectives, as in “everlasting things,” “created things,” “invisible things.” These borrowings of Miltonic formulas do not clash with Wordsworth’s more personal invocations of things, but their presence calls attention to the distance between the two forms of expression. Wordsworth’s characteristic formulas do not clearly define things by their places in the “indestructible dominion,” visions of which Milton’s epic bard had been granted, because Wordsworth was skeptical of the power of his own imagination to believe. Instead, his invocations of things characteristically
5)*/("/%5)*/(4
treat the word as mysterious, unlocated, indescribable by adjectives. To encourage such an effect, he chose as a repeated form a prepositional construction joining noun to noun. The first attaches a quality or attribute to things but leaves its semantic references undetermined or even void: “the principles of things,” “the mystic cause of things,” “the ghostliness of things,” “the spirit of things,” “the depth of things,” “the heart of things,” “the soul of things,” “the light of things,” “The moral shapes of things,” “The spiritual presences of absent things.” Adding to the mysteriousness of these phrases, the first nouns, which might be expected to specify or otherwise delimit the undefined nature of things, are themselves opaque. They tend to veil or mute the significations of materiality packed into things that might otherwise be acknowledged as they are in the adjectives typical of Miltonic formulas quoted earlier: external, fair, visible, new, above, highest. This dimension to Wordsworth’s language of things is most luminously illustrated in “Tintern Abbey.” The poem is built on the quintessential Wordsworthian situation of the poet searching in a natural landscape, and in the memory of it, for intimations of transcendence. Because of its paradigmatic value, critics have often chosen this poem for discussion, among them Geoffrey Hartman: Wordsworth, at the beginning of “Tintern Abbey,” is striving toward the expression of a mystic feeling. . . . Whenever Wordsworth considers external nature of that familiar kind which alone enters into his experience, he is beset by an urge to consider it with the same enthusiasm that affects Milton contemplating a nature neither external nor familiar, that is, the very spirit of God seen through the mind’s eye; but this valley, these waters and cliffs, these pastoral farms never, even at the most intense, convey more than the suggestion of a possible sublimity.5
Hartman’s argument focuses on the problem for Wordsworth of connecting if not merging the visible character of the landscape with the immaterial, the unutterable, with what Milton confidently claimed to see: “things invisible to mortal sight.” The discussion here is interested in Wordsworth’s ways of using things to make possible—if possible—the connecting or merging. Hartman’s discussion begins by noticing that the description of the scene in the first verse paragraph is thick with demonstratives: “These waters,” “these steep and lofty cliffs,” “this dark sycamore,” “These plots of cottage-ground,
5)*/("/%5)*/(4
these orchard-tufts,” “These hedge-rows,” “these pastoral farms, / Green to the very door.”6 The repetition of demonstratives emphasizes the immediate visible presence of the natural objects that the poet contemplates. An added observation to the point of the argument here is that these objects are identified and specified by plural nouns with clearly defined references of the familiar kind that Wordsworth used to name what he classed elsewhere in his poems as “external things,” “visible things,” “common things that round us lie,” as distinct from “everlasting things.” Even more to the point, in this introductory passage of “Tintern Abbey” he withheld the chameleon-like word things. The second paragraph opens with a demonstrative that points to these same immediately present “waters,” “cliffs,” “farms,” but describes them as the poet remembered them during his long absence from the scene, using a noun that conceptualizes them as a collective abstraction: “These beauteous forms.” The poet senses a process taking place internally, “passing even into my purer mind,” that transforms his language into a mystical meditation: While with an eye made quiet by the power Of harmony, and the deep power of joy, We see into the life of things.
What were visible objects belonging to external nature, reconceived as “These beauteous forms,” are no longer demonstratively these, but simply things that are—among the multiplicity of semantic references packed into the noun— simultaneously embodied and ethereal, temporal and internal, visible and invisible. Their mysterious “life” can be seen by the outward eye, seen “into” by the visionary eye. Had Wordsworth been less uncertain of his own power to believe, he might have ended the poem with these quietly ecstatic lines. What immediately follows them is convincing as an expression of an intense personal struggle because, having admitted that this transcendent experience may be “but a vain belief,” Wordsworth tried to reformulate his claim to the vision at the end of the next passage, where the poet once more recalls his coming among “these hills”: And I have felt A presence that disturbs me with the joy Of elevated thoughts; a sense sublime Of something far more deeply interfused, Whose dwelling is the light of setting suns,
5)*/("/%5)*/(4
And the round ocean and the living air, And the blue sky, and in the mind of man: A motion and a spirit, that impels All thinking things, all objects of all thought, And rolls through all things.
The crescendo of something, dwelling, setting, living, thinking, things, things makes this vision of “all things” almost mesmerizing in its cadences. The poet, like the narrator in Paradise Lost, from a perspective outside the universe views it whole: its many suns, the curved surface of the ocean, “all things” and “the spirit” that rolls through them as the earth rolls round in its diurnal course. In these lines the poet senses that external nature, humanity, spirit are mysteriously merged by “something far more deeply interfused,” but he cannot see and tell what that “something” is. In that compound the accent—both semantically and metrically—falls on the indeterminate prefix some. The lines that follow this visionary passage quite explicitly return to the much loved earth: Therefore am I still A lover of the meadows and the woods, And mountains; and of all that we behold From this green earth . . .
And all that follows is a beautifully elegiac anticlimax that in the last lines returns the poem, like the poet after his years of absence, to what is visible to the outward eye: “these steep woods and lofty cliffs, / And this green pastoral landscape.” Wordsworth’s “To the Cuckoo” () is one among a number of contemporaneous poems that chooses the situation of hearing the sound made by an unseen bird to figure the poet’s longing for transcendent experience. The special interest of this discussion is in what Wordsworth’s unexpected introduction in the middle of the poem of the word thing does to the familiar Romantic situation, which has as its most famous antecedent the lines of Paradise Lost that invoke the “Bird” who “Sings darkling, and in shadiest Covert hid / Tunes her nocturnal Note.” In poems remembering Milton’s nightingale—along with Wordsworth’s to the cuckoo, Wordsworth’s and Shelley’s to the skylark, and Keats’s to a nightingale come first to mind—the possibility of visionary poetry depends on the power of the poet’s language to realize the bird as a symbol that unites
5)*/("/%5)*/(4
the visible and invisible. In none of these poems other than “To the Cuckoo” is the actual word thing used to stand for the bird. Here it makes its surprising appearance in stanza . “To the Cuckoo” opens, as do both Wordsworth’s poems to the skylark, with a conventional address to the bird:7 O blithe New-comer! I have heard, I hear thee and rejoice. O Cuckoo! shall I call thee Bird, Or but a wandering Voice?
The heightened exclamations in these lines are reinforced by the repeated trope of apostrophe in the concluding stanza —“O blessed Bird!”; but in stanza —“Thrice welcome, darling of the Spring!”—the sublimity of the apostrophic address is undercut by thing. What first makes this appearance of thing unexpected is its role as the answering end rhyme to “Spring!,” the season heralded in traditional lyric poems by “the darling buds of May” and the songs of birds: Thrice welcome, darling of the Spring! Even yet thou art to me No bird, but an invisible thing, A voice, a mystery; . . .
Wordsworth’s echo of Shakespeare’s most often quoted sonnet reinforces the expectation that sing or wing would be the rhymed response to “Spring!” The jolt of thing, almost bathetic in its ordinariness, is exaggerated by its strong position as the accented rhymed syllable at the end of a punctuated line. The more compelling surprise in the appearance of thing is its incompatibility with the apostrophe that invokes it, and ultimately with the apostrophic mode itself. This kind of poetry is, according to Jonathan Culler, “the pure embodiment . . . of the subject’s claim that in his verse he is not merely an empirical poet, but the embodiment of . . . the spirit of poetry.” It follows that apostrophe is particularly suited to poetry reaching for transcendence, because a poet “who successfully invokes nature is one to whom nature might, in its turn, speak. He makes himself poet, visionary,” or hopes to become such, perhaps by sheer intensity of address. Since apostrophe is predicated on the fiction that whatever the poet addresses can respond in a recognizable language, it is essentially a form of
5)*/("/%5)*/(4
personification, and “Thrice welcome, darling of the Spring!” is the most overt and detailed personification in the poem.8 It creates in miniature a human situation in which the poet acts as delighted host extending an ecstatic greeting on the occasion of a familiar and lovable visitor’s return. If Wordsworth had arranged the stanza so that it carried the personification into the following lines with a different epithet—“Even yet thou art to me / No bird, but an invisible [song]”—the renaming of the bird would not be shocking. Instead, Wordsworth chose “invisible thing,” a formula attached in the Bible to what is transcendent—distinct and unapproachably distant from what is ephemeral. It undermines the poet’s assumption—or deflates his hope—that he and the cuckoo might communicate as beings belonging to the same ontological sphere, because the epithet “invisible thing” points directly to the otherness of the bird. Located in the indestructible dominion, thing here has no defined reference; its existence is a sublime blank beyond the reach of human sight or language. This attempt to name the unknowable and therefore unnameable “thing” undermines the possibility that the poet’s apostrophic mode, to which he returns in the last stanza, can allow a transcendent vision of the natural and eternal merged in a living creature: O blessed Bird! the earth we pace Again appears to be An unsubstantial, faery place; That is fit home for Thee!
At most, the solid “earth we pace” on but do not dwell in “appears”—makes itself apparent or merely seems—to us as an ethereal “home” for the cuckoo, still longingly personified. The climactic apostrophe “O blessed Bird!” is a last and failed attempt to situate the cuckoo in both visible nature and the invisible dominion.
In Wallace Stevens’s Collected Poems, the word thing makes the first of its multitudinous appearances very near the beginning, in the opening lines from the second section of “Le Monocle de Mon Oncle”: A red bird flies across the golden floor. It is a red bird that seeks out his choir Among the choirs of wind and wet and wing.
5)*/("/%5)*/(4
A torrent will fall from him when he finds. Shall I uncrumple this much-crumpled thing? 9
Grammar dictates that the semantic reference of “this . . . thing” must be the “red bird,” a cardinal flying over fallen autumn leaves. Then “It” is personified, becoming the mythological bird of Romantic poetry that “pours forth” soaring harmonies, but this bird, battered by “wind” and drenched by “wet,” is “much-crumpled.” One association leads to another: “this much-crumpled thing” refers also to poetry of the Romantic past, its pages wrinkled because too often turned, and the asker of the question—the aging uncle seeing through only one eyeglass—is a stand-in for the modernist poet. His contemplated act of uncrumpling the “thing” would not make it new, but only smooth it for reuse. Still, if these lines themselves are an example of past poetry uncrumpled by a poet of the present, they show at least that it can be beautiful. Ten short poems later, in the opening section of “The Comedian as the Letter C,” the word thing appears in the narrative just as Crispin was starting out on his “introspective” voyage. He is another “distortion” representing the modernist poet who is a comic parody of the meditative Romantic poet on a quest for glimpses of the absolute, the essential, the unutterable: Here was the veritable ding an sich, at last, Crispin confronting it, a vocable thing, But with a speech belched out of hoary darks Noway resembling his, a visible thing, . . . Here was no help before reality. Crispin beheld and Crispin was made new.10
The narrator who hides behind Crispin tells us that his stand-in hero once had a transcendent experience and with some posturing confronted it, presumably by personifying it, invoking it with apostrophes in expectation of a response. He seems to have been sent an audible message, a sort of visionary tale, from “a vocable thing,” but as happens in Romantic poems, the answering “speech,” if there was any, was in a language not his own, or not intelligible to human ears. With characteristic sleight of hand Stevens ambiguously suspended “a speech belched out . . . / Noway resembling his,” so that the “speech” might be the mysterious, sourceless pronouncement of Kant’s famous formula untranslated in the first line, or it could equally be Crispin’s grossly mispronounced recitation of the German phrase in his own trivially
5)*/("/%5)*/(4
excited tone. Then it seems Crispin was granted a glimpse of some “visible thing,” but “Here was no help before reality” either. Finally, with miraculous suddenness, “Crispin beheld and Crispin was made new,” a momentary, falsely religious happy ending to a momentary Romantic illusion or disillusion. Much more flamboyantly than the lines from “Le Monocle de Mon Oncle,” this passage displays some of the characteristic fun and games of the modernist poet trying to escape his Romantic precursors. Typically, Stevens dodged questions from correspondents who wanted to know which poets had been his influences, for instance, in this response: “While, of course, I come down from the past, the past is my own, and not something marked Coleridge, Wordsworth, etc. I know of no one who has been particularly important to me.”11 By now readers are well prepared to be skeptical of such evasive or opaque statements by Stevens about his work, and its many-faceted relationship to Romantic poetry has been amply recognized. The intention here is to explore the as yet unexamined intimacy between Stevens and Wordsworth that we can discover embedded in these poets’ shared fascination with the expressive possibilities of thing and things. In its chameleon-like nature the noun thing unites the ordinary and extraordinary, visible and invisible dimensions of our experience, which is what these poets wanted to get at in their poems. In Stevens’s poetry he used thing slightly more than times, things some , numbers not far off the count of Wordsworth’s but staggeringly high when measured against the example of another modernist poet, William Carlos Williams. In his two-volume Collected Poems my count turned up fewer than instances of either thing or things (which, though scarcely used, rings out loudly in his battle cry for poetry with “No ideas but in things”). Among Stevens’s repetitions of things, are concentrated in “The Man with the Blue Guitar,” which makes a refrain out of variations on the formula things as they are, widely used in the nineteenth century as a religious, political, economic slogan (and as the title of William Godwin’s novel, Things as They Are or The Adventures of Caleb Williams). This often explicated poem has been commonly agreed on by writers about Stevens to have introduced his new style of “deceptive open plainness,” so described by Helen Vendler.12 Its vocabulary is markedly more spare, largely stripped of the gorgeous gaudiness in the first two books. The act of writing this pivotal poem also seems to have
5)*/("/%5)*/(4
deepened Stevens’s fascination with thing and things, which are repeated profusely in the later volumes in the texts of poems and in several of their titles: “Man Carrying Thing,” “Things of August,” “The Plain Sense of Things,” “Not Ideas About the Thing but the Thing Itself.” Obsessiveness is as obvious in Stevens’s repetitions of thing and things as it is in Wordsworth’s, because both poets not only use them often, but often repeat them in the same line or clusters of lines. Both also reuse the kinds of modifiers that describe or intend to attach semantic content to thing and things, and repeat syntactical arrangements that reveal or obscure their relations with the words around them. Though Stevens shared these forms of repetition with Wordsworth, he mainly avoided the precise biblical or Miltonic formulas that Wordsworth intermingled with his own distinctive visionary vocabulary. For instance, all things appears only half a dozen times in Stevens’s Collected Poems, other orthodox phrases even less often and then in distorted versions such as “things seen and unseen, created from nothingness.” He also stayed away from Wordsworthian mystifications like “the soul of things,” “the ghostliness of things,” “the mystic cause of things,” unless to subvert them. For instance, Wordsworth’s “the depth of things” rewritten in Stevens’s personal language turns bathetically into “the bottom of things.” Paradoxically, and paradox is a favorite modernist weapon against the Romantic sublime, Stevens’s peculiar personal schemes for treating thing and things are themselves mystifications, crafted so that they point out their departures from Wordsworth’s visionary language. Here are samples of Stevens’s typical replacements for his predecessor’s magical formulas: “a thing supposed,” “a thing not apprehended,” “a thing seen by the mind,” “A thing of shadows,” “A thing not planned for imagery or belief,” “the thing as idea,” “The idea as thing,” “Things unintelligible, yet understood,” “things submerged with their englutted sounds,” “things as the structure of ideas,” “things without a double,” “things beyond resemblance,” “Things dark on the horizons of perception.” Stevens’s phrases can trick us into the position of repeating about them what I said earlier in this discussion of Wordsworth’s parallel phrases: that “his invocations of thing and things characteristically treat them as mysterious, unlocated, indescribable.” Yet there is a conceptual world of difference in their distinctive phrasings that depends as much on their unlike grammar as on their dissimilar semantic content. That is, although the first noun in
5)*/("/%5)*/(4
Wordsworth’s ritual formulas—depth, soul, ghostliness—makes a vague suggestion only of some dimly perceived or unseen spiritual force, the preposition specifies that mystery to be a property “of” things. By contrast, in one of Stevens’s formulaic phrases the words that modify the noun tend to work more tentatively, questioning what might be “supposed” to be the nature of the thing or things. One of Stevens’s simplest devices is to surround thing or things with words like not, without, beyond, as in the phrases quoted above, but sometimes the negative is slyly implied: “a thing seen by the mind” is not visible to the outward eye; “A thing of shadows” has no substance. Often in passages that struggle with the mystery of thing or things, negatives—never, no, not, nor—combine with quizzically hopeful, conditional expressions—maybe, or, should be, As, like, Almost—as in these opening lines from section VIII of “The Auroras of Autumn”: There may be always a time of innocence. There is never a place. Or if there is no time, It is not a thing of time, nor of place. . . . There is or may be a time of innocence As pure principle. Its nature is its end, That it should be, and yet not be, a thing . . . It is like a thing of ether that exists, Almost as a predicate. But it exists, It exists, it is visible, it is, it is.13
Here negatives and words that express varying degrees of skepticism and affirmation combine with riddling paradoxes and dense philosophical puzzles phrased in grammatically intricate sentences, all schemes raising questions about the possibility that the mind might penetrate the mystery of thing itself. This skepticism shows in the avoidance of demonstratives—this, that, even the—or other specifying grammatical forms that would, like these in “Tintern Abbey,” grant thing an empirical as well as a transcendent existence. What we know of thing is given here only in two quasi comparisons: “as it is like . . . ether” (a broken-off revision of Milton’s and Wordsworth’s ethereal) “that exists / Almost as a predicate.” In the second figure, a “thing” is “as” a grammatical part of a sentence that says something about its subject but is not
5)*/("/%5)*/(4
its subject. This emptying of empirical content from thing seems to tilt it toward synonymity with nothing, recalling the quintessentially Stevensian line in “The Snow Man”: “Nothing that is not there and the nothing that is.” Then, inexplicably, the penultimate line bursts into passionate affirmation—“But it exists, / It exists, it is visible, it is, it is”—sounding more frantic than Wordsworth’s efforts toward convinced belief. This inexplicable reversal is paradoxically predictable of a poem by Stevens because the affirmation itself is “like a thing” that mysteriously is and is not something to be understood. In “Landscape with Boat” a “He” whom the poet recognizes as his “antimaster-man” is criticized for his inability to “suppose” contradictions, to entertain conditional propositions, to accept paradoxes that break into “parts” the notion of “A truth beyond all truths,” a monolithic “thing”: He never supposed That he might be the truth, himself, or part of it, That the things that he rejected might be part And the irregular turquoise, part, the perceptible blue Grown denser, part, the eye so touched, so played Upon by clouds, the ear so magnified By thunder, parts, and all these things together, Parts, and more things, parts. He never supposed divine Things might not look divine, nor that if nothing Was divine then all things were, the world itself, And that if nothing was the truth, then all Things were the truth, the world itself was the truth.14
These lines compose a catalogue of repeated words that turns a sequence of negatives—never, not, nor, nothing—into what sounds almost like a litany affirming the truth of “things together,” “more things,” “all things” that “look divine,” “all / Things” that make up “the truth.” The crescendo of words repeated in identical and similar phrases and cadences sounds like the most passionate visionary lines in “Tintern Abbey,” which are its climax but not its conclusion. What follows the heightened passages of both poems are anticlimactic descents to the beautiful, pleasurable but insufficient phenomenal world: in Wordsworth’s to “these steep woods and lofty cliffs” of an actual familiar landscape; in Stevens’s to a hypothetical view of a passing boat from a
5)*/("/%5)*/(4
“ balcony / Above the Mediterranean,” described as if the viewer were looking at an enticing travel poster rather than at the “Landscape” predicted in the title: Had he been better able to suppose: He might sit on a sofa on a balcony Above the Mediterranean, emerald Becoming emeralds. He might watch the palms Flap green ears in the heat. He might observe A yellow wine and follow a steamer’s track And say, “The thing I hum appears to be The rhythm of this celestial pantomime.”
The last sections of the two poems are similarly diminished by doubts of their own transcendental language. Wordsworth’s ends in muted sadness; Stevens’s in amusement, partly at himself. In its position as the title to the last entry in Collected Poems, “Not Ideas About the Thing but the Thing Itself” asks to be taken as a retrospective observation on the preceding body of poetry read as a perpetual search that has not reached its perhaps unattainable ideal. In the title’s position suspended above the poem, it states a mysterious proposition apposite to the poem but in no direct connection with it. Coming as if from the white space around it, the title does not name or place or characterize or criticize the poem. Even so, our habitual way of reading titles leads us to expect that in this one, however idiosyncratic, “Thing” has some semantic reference in the poem, which the first two stanzas define as “a scrawny cry,” then “A bird’s cry”: At the earliest ending of winter, In March, a scrawny cry from outside Seemed like a sound in his mind. He knew that he heard it, A bird’s cry, at daylight or before, In the early March wind. The sun was rising at six, No longer a battered panache above snow . . . It would have been outside.
5)*/("/%5)*/(4
It was not from the vast ventriloquism Of sleep’s faded papier-mache . . . The sun was coming from outside. That scrawny cry—it was A chorister whose c preceded the choir. It was part of the colossal sun, Surrounded by its choral rings Still far away. It was like A new knowledge of reality.15
Familiarity with Stevens’s positioning of poems in relation to Romantic tradition allows us to read this one in the context of earlier visionary poems and of Stevens’s “Le Monocle de Mon Oncle.” The “scrawny cry” is a revision of the “choir” the “red bird” sings in: the “much-crumpled” ecstatic poetry that pours from Romantic cuckoos, skylarks, and nightingales. This is an unromantic bird, no “darling of the Spring!” who comes in “the earliest ending of winter,” and “He” to whom its “cry from outside / Seemed like a sound in his mind” is not quite a Romantic poet. For he “knew” the bird and the sun came “from outside,” and “knew” also “That scrawny cry” (made particular and present by the demonstrative and descriptive adjective) could be heard even though “Still far away.” These unequivocating, affirmative statements express new certainties that “the Thing Itself” is both distantly surrounded by the mystery of “choral rings / Still far away” and yet immediately present “outside” the “mind” in external nature. The qualifying like in the last sentence does not negate the affirmatives. “It”—first referring to the “scrawny cry,” then to “The sun”—here seems to point to the writing of the poem: “It was like a new knowledge of reality” made known in metaphor.
Though famously elusive in his poems, John Ashbery in interviews has not played hide-and-seek with the inevitable, irritating question of influence that modernist poets typically dodged. He has more than once named Stevens as his early model though without giving more than a cloudy suggestion of their profound likenesses: Wallace Stevens was one of the first modern poets I discovered and who influenced me. But it’s hard for me to tell you now exactly in what ways. In some
5)*/("/%5)*/(4
of my early poems, such as “Le livre est sur la table”—I was half consciously imitative of Stevens, but I think that there has been a deeper influence which is, as I say, difficult for me to analyze now. I think that in this kind of meditative verse the things in a room and the events of everyday life can enter and become almost fossilized in the poems. “An ordinary evening in New Haven,” for instance, is that and also a meditation on life.16
Not deliberately, but—it will be argued here—not coincidentally, this acknowledgment of “deeper” affinity that Ashbery has sensed from the beginning between himself and Stevens makes important use of things. In the thirteen volumes of short poems by Ashbery where I could find this ordinary and extraordinary word, I counted (roughly) at least appearances of thing, of things. In both singular and plural forms, this word has played as dominant a part in Ashbery’s poems as in those of his continuing model, Stevens, and of his still active influence, Wordsworth. Ashbery seems to have recognized in their work the opportunity opened by the strangeness of the word thing: its contradictory, double capacity to stand for all other nouns (in Ashbery’s poems even for whole clauses like “the ‘really not the same thing at all’ ”) yet to have no intrinsic semantic content. From very early in his writing life, Ashbery apparently saw this paradox of a simultaneously overloaded and empty linguistic figure as a device for expressing the mysterious coexistence and disconnection of transcendent things and everyday things. Gestures that point to the presences of Wordsworth and Stevens show up under many guises in Ashbery’s poems. The most explicit signal their purpose when they call up a context in which thing or things appears in the work of the earlier poets. Here is an illustration from a poem in a volume published in , Hotel Lautreamont, showing that both Wordsworth and Stevens have been continuously present in Ashbery’s work. The title of the poem, “Cop and Sweater,” itself points to Stevens, who also liked to invent titles reflecting the randomness of things that we come across, or that come across us, in incomprehensible combinations daily. These are the opening lines: It’s about this undulation thing, how we were all beginners to get in on it when it began. Once that had happened, there was a different face on things: trees no longer came to the door; the seasons were always “forgetting” to include you in the list— that sort of thing.17
5)*/("/%5)*/(4
The voice in these lines makes us feel that “we” have come in late, because we are at first clueless about the semantic references of It’s, this, thing. Still, the lexical definition of undulation—a rise and fall of level—together with the associations of its overused ion suffix suggest that the “thing” we wanted to get in on early was some sort of gamble: like betting on the ups and downs of inflation, depreciation, recession, reorganization. Another, more grotesquely absurd pointer to the meaning of “this undulation thing” is the hilariously atrocious grammar of the phrase, an inevitable reminder of “the vision thing.” We bet on faith in visionary experience, only to discover that its deceiving promises were overlaid by another false invitation to belief: external nature—“trees,” “seasons”—“no longer came to the door” as it had—or so we were told—merged with “these pastoral farms, / Green to the very door” in “Tintern Abbey.” Although Wordsworth had hoped, though uncertainly, that physical nature might offer at least glimpses of the visible and invisible worlds united in things, “we” latter-day seekers who had hoped to win the gamble recognized that the natural world refused to “include” us. Refusing, in turn, to give in to Romantic disillusion, these lines treat casually the momentous historical divide between nature and the mind as a silly snub, “that sort of thing.” Of course the continued personifying of natural “things”—they make invitation lists—is a form of self-mockery, not Wordsworth’s “sort of thing” but sometimes practiced by Stevens and congenial to Ashbery. Obsession with things makes itself felt in Ashbery’s poetry, as in Wordsworth’s and Stevens’s, in countless repetitions of the same or similar patterned phrases, but his characteristic formulations are pointedly different both from Wordsworth’s biblical and visionary expressions and from Stevens’s quasiphilosophical sayings. Their precise formulas show up rarely in Ashbery’s poems (scarcity making their appearances powerfully effective). Much more typically he secularizes and often also parodies them, so that conceptual differences between him and his predecessors are vividly defined. To illustrate, Wordsworth’s “all things” in Ashbery’s language becomes “all those things,” “more than enough things,” “the overabundance of things,” “too many things to think about,” “extra things,” “too much of any one thing,” “How many / Other things can one want.” The magnitude and absolute certainty of the biblical-Miltonic formula are replaced by the sense of oppression and confusion of daily experience “full of more things” than—“What with one thing and another”—we can cope with.
5)*/("/%5)*/(4
Similarly, Stevens’s “a thing not apprehended” is reformed as “Things we mustn’t know,” “things that don’t coincide,” “the things that are / Truly impossible,” “so many things not to believe in,” “Some things are always left undecided.” Here Stevens’s belief that transcendence is not immediately knowable, not knowable in Wordsworthian glimpses, is exaggerated in Ashbery’s more convinced disbelief: a sense of disconnection that is alternately but sometimes simultaneously dismayed, irritated, passively accepting, amused. The idiosyncratic personal idiom that Ashbery uses about things is made in large part (with interruptions of highly rhetorical, philosophic, scientific, apocalyptic, and other vocabularies) out of idiomatic, often comically imprecise public language we hear and speak every day, which is not what Wordsworth intended when he aspired to write in “the language really spoken by men.” Stevens did use that casual spoken language, but sparingly. The simplest stock phrases that pervade Ashbery’s colloquial speech are the most familiar: “The whole thing,” “the main thing,” “the real thing,” “the big thing,” “the funny thing,” “the last thing,” “a good thing,” “a sure thing.” The casual style of these packaged phrases works sometimes to debunk the Romantic sublime while engaging in its concerns. Elsewhere these oftenrepeated idioms, pronounced with absolute certainty as if sanctioned by some central intelligence, are treated with amusement unheard in Wordsworth’s poems, combined with expressions not often heard in Stevens’s of fellow feeling with “we” who try to make ourselves understood in this blurry language. Its imprecision is not the opacity of Wordsworth or the obliquity of Stevens. The grammar of the phrases quoted just above works differently than Wordsworth’s or Stevens’s formulas for things because Ashbery’s minimally descriptive adjectives—whole, main, sure, and so on—leave the noun without specific properties or conceptual context. “I tried each thing,” the opening phrase of the opening poem in Self-Portrait in a Convex Mirror, seems to come out of nowhere. It is not inappropriate to borrow again the terms used earlier to describe Wordsworth’s and Stevens’s characterizations of things as “mysterious, unlocated, indescribable.” Even so, in Ashbery’s poems those effects mainly depend on the weird emptiness of his language about things, rather than on Wordsworthian veiled phrasing, or on the negatives and skeptical hypotheses of Stevens. Even these simplest idioms, which call most attention to the dominance of thing and things in our speech, are nevertheless strange when Ashbery puts them to work.
5)*/("/%5)*/(4
Very often Ashbery chooses more complex commonplaces for talking about things (not explicitly personifying them the way Wordsworth tends to do) as agents that set in motion our daily experiences: “The quirky things that happen to me,” “the way / Things have of just happening,” “the way things have of enfolding,” “when things go wrong,” “things can’t get any worse,” “things should pick up,” “In spite of things,” “This thing works both ways, you know.” In these clichés the absent semantic references of the contentless noun are imagined as acting in unaccountable ways, accidentally or coincidentally or magically making things happen. A contradictory pattern of stock phrases characterizes things as passive, inert obstacles that burden our ordinary lives: “I’ve a million things to do,” “I’ve already stopped to do things,” “Still another thing I need to do,” “to get out and do things,” “do as many things as possible,” “Some things we do take up a lot more time,” “things / Do get done . . . but never the things / We set out to accomplish.” Yet another strand of sayings, the most powerfully expressive, contributes to the sense in Ashbery’s poems that things escape from us into a mysterious ontological order where we can never quite fathom or get control of them: “Familiar things look a long way off,” “forgotten / Things that don’t seem familiar,” “things recently forgotten,” “all forgotten things,” “things you forgot,” “The thing we always forget to put in,” “forgetting / For a few moments some thing that should never be forgotten,” “Trying to remember things,” “I haven’t thought about these things in years,” “ It went away / Little by little, as most things do.” Whatever is forever slipping away from us is catalogued by lines in A Wave from the first two verse paragraphs of “Down by the Station, Early in the Morning,” the title itself the first line from an old, largely forgotten song for children: It all wears out. I keep telling myself this, but I can never believe me, though others do. Even things do. And the things they do. Like the rasp of silk, or a certain Glottal stop in your voice as you are telling me how you Didn’t have time to brush your teeth but gargled with Listerine Instead. Each is a base one might want to touch once more Before dying. There’s the moment years ago in the station in Venice, The dark rainy afternoon in fourth grade, and the shoes then, Made of a dull crinkled brown leather that no longer exists.
5)*/("/%5)*/(4
And nothing does, until you name it, remembering, and even then It may not have existed, or existed only as a result Of the perceptual dysfunction you’ve been carrying around for years.18
“It all wears out” is a vague enough sentence to encompass the motley “things” then listed—a texture, a sound, a place where nothing is stationary, a time of day, a pair of shoes—each a “moment” like a hoped-for Wordsworthian glimpse of the transcendental world. Here that hope is denied because the moment “no longer exists. / And nothing does”—Stevens’s “Nothing that is not there and the nothing that is”—unless, and this is Stevens’s supreme fiction, “you name it, remembering.” Existing though only as a word, it can be preserved, “fossilized” as Ashbery said of “the things in a room and the events of everyday life” in Stevens’s “An Ordinary Evening in New Haven” and in his own meditative poems. A feeling of loss or of getting “the whole thing wrong” is generated through Ashbery’s work by these repeated patterns woven around thing and things. They carry from one volume to another, from one poem to another, and in individual poems from one line to others, for instance, in the opening verse paragraph of “Grand Galop” from Self-Portrait in a Convex Mirror: All things seem mention of themselves And the names which stem from them branch out to other referents. Hugely, spring exists again. The weigela does its dusty thing In fire-hammered air. And garbage cans are heaved against The railing as the tulips yawn and crack open and fall apart. And today is Monday. Today’s lunch is: Spanish omelet, lettuce and tomato salad, Jello, milk and cookies. Tomorrow’s: sloppy joe on bun, Scalloped corn, stewed tomatoes, rice pudding, and milk. The names we stole don’t remove us: We have moved on a little ahead of them And now it is time to wait again. Only waiting, the waiting: what fills up the time between? It is another kind of wait, waiting for the wait to be ended. Nothing takes up its fair share of time, The wait is built into the things just coming into their own. Nothing is partially incomplete, but the wait Invests everything like a climate.
5)*/("/%5)*/(4
What time of day is it? Does anything matter? Yes, for you must wait to see what it is really like, This event rounding the corner Which will be unlike anything else and really Cause no surprise: it’s too ample.19
All things, which opens the two-line first sentence, is in the meditative mode that Ashbery shares with Stevens and Wordsworth. A disruptive set of lines follows, set off by vocabulary, syntax, and tone alien to meditative verse, after which the verse paragraph returns to its meditation. This longer set of lines picks up names and echoes All things, but narrows it to the things and makes near-rhymes of it with Nothing, nothing, everything, anything, anything surrounded by waiting, waiting, wait, wait, wait, time, time, really, really, it, It, its, it, it, its. This surface pattern is closely modeled on designs made by Stevens like the one in “Landscape with Boat”: part, things, part, part, parts, things, parts, nothing, all things, nothing, all / Things. Although the arrangements are very similar, Ashbery almost entirely avoids Stevens’s vocabulary of negatives and conditional affirmations that might draw lines between alternatives. In Ashbery’s exemplary design, “All things seem”—have the look of being—“names” that refer back to themselves and to other “things,” making a circular motion that appears on the page as a coming back of “things” to “things.” Outside it is the “Yes” that affirms the nameless “event”—death? Not a thing and “unlike any[ ]thing else,” the “it” is a transcendent experience outside the referentially empty, self-enclosed circle of language. Even less like what might happen in a poem by Stevens is the radical disconnection made by Ashbery in the disruptive set of lines that breaks in on the meditation, even though it recalls antipastoral catalogues by Stevens like this one in “The Man on the Dump”: So the sun, And so the moon, both come, and the janitor’s poems Of every day, the wrapper on the can of pears, The cat in the paper-bag, the corset, the box From Esthonia: the tiger chest, for tea.20
Typically, Stevens’s catalogue is woven grammatically into the lineation and argument of the paragraph, whereas Ashbery’s inserted sentences are so
5)*/("/%5)*/(4
foreign to the rest of the passage that they seem to belong to a different voice ignored by the meditative poet and forgotten by the later lines. The inset, coming out of nowhere, begins with three sentences that conjure up some urban street littered with objects that might end on the dump. The succeeding three sentences, joined by a casually illogical And, recite from what sounds like a calendar of school lunches. Every feature of the language in these lines is designed to shock us into feeling its disconnection from the abstract vocabulary suited to meditation on “All things” and their “referents.” The focus of incompatibility, coming midway in the line, is “The weigela does its dusty thing”: the only use in the entire verse paragraph of thing in the singular. The slang phrase takes the semantically empty noun out of the realm of abstraction, and links it to a noun with a clearly defined referent: a flower with a funny proper name. Many names of foods on the lunch menu are even more comically specific, evoking the sight and smell, texture and taste of each gooey item, which to the speaker is not merely a name but some thing you can put in your mouth, delighting in each swallow. Here the poet or poem seems to play Dr. Johnson’s game of refuting Berkeley by kicking a stone. If this outrageously disruptive set of lines came at the end after both meditative passages, it would subvert them, and there are such subversions: probably not so intended in poems by Wordsworth, deliberate in Stevens’s and Ashbery’s. Stuck part way in as an interruption, the misfitting lines have a somewhat different slant. They pose an irreconcilable contradiction reflecting in their own way the bewilderment and anxiety surrounding thing and things that has taken various forms in the poetry of Wordsworth, Stevens, and Ashbery. In Ashbery’s poetry, more than in Wordsworth’s and perhaps more even than in Stevens’s, the contradiction in our ways of thinking about thing and things is represented as a disconnection between linguistic abstractions and everyday experience. We know thing and things to be “really” signs referring only to each other. Even so, we experience them as what “really” happens to us, what we have to do, what we are forever losing or forgetting. This characteristic conflicted verse paragraph from “Grand Galop,” a kind of poem within a poem, comes to an inconclusion that invites a conventional critical conclusion that it does not work, which may be precisely the point. Ashbery could have been suggesting as much by what he said in an interview: Everything seems to be sort of transcendental, a kind of shadow of something else. I read a very beautiful passage in Borges recently about how certain things
5)*/("/%5)*/(4
in life—landscapes, remarks made by strangers, whatever—had a significance that we should have been aware of when it happened but weren’t. That seems to be basically what I feel about religion: that somehow we miss the boat at every possible moment. There is always the prospect of grace, but we shouldn’t count on it too much.21
$IBQUFS
015 * $ "- * - - 64 * 0 / 4 8PSETXPSUIBOE)JT*OIFSJUPST
archetypically titled “Poem,” the poet recognizes a remembered landscape of her childhood represented in a painting made by a great-uncle she never knew:
*/ &- *; " # & 5 ) # *4) 0 1 ± 4
Our visions coincided—“visions” is too serious a word—our looks, two looks: . . .1
a smilingly self-deprecating revision not to be taken unseriously. It pays tribute to the merging of a memory—“Heavens, I recognize the place”—and a visual illusion—“a wild iris . . . / fresh-squiggled from the tube”—in a scene made timeless by present participles: the munching cows, the iris, crisp and shivering, the water still standing from spring freshets, the yet-to-be-dismantled elms, the geese.
The transformed scene, pointedly not an apocalyptic vision, is a descriptive catalogue of familiar things in nature: “the little that we get for free, / the little of our earthly trust.” These lines, phrased in Bishop’s own characteristic low key, are tonally distant from Prufrock’s bathetic prophecy: And time yet for a hundred indecisions, And for a hundred visions and revisions,
015*$"-*--64*0/4
or Keats’s disillusioned questioning: “Was it a vision, or a waking dream?” Even so, Bishop’s revision is like these two famous earlier instances in alluding to a spectrum of questioning, distrustful, or disillusioned attitudes toward supernatural visions that we have come to identify with centrally important poems written since the beginning of the nineteenth century. There is surely no need to recite the much-discussed causes great and small that contributed to making poets uneasy with the word vision in its oldest sense of something seen with other than bodily sight: a prophetical or mystical revelation. The cultural situation that pertains directly to the comparatively microscopic focus of the discussion here can be summarized in this combination of circumstances: that some implicit uneasiness about supernatural visions as matter for poems of dominant kinds, particularly lyric and descriptive, coincided not coincidentally with the spread of scientific and philosophical theories about the nature of bodily vision; that these theories by the early eighteenth century began to interest a wide audience, including poets who made very specific use of them in their poems; that those poems, in turn, directly prepared the ground for the work of later poets where questioning, undermining, denying, or ignoring the possibility of supernatural visions is the explicit focus of the poem. Keats attributed the loss of nature’s power to yield adequate symbols of supernatural vision directly to seventeenth- and eighteenth-century studies of bodily seeing in a passage of Lamia written at almost the same time as “Ode to a Nightingale”: Do not all charms fly At the mere touch of cold philosophy? There was an awful rainbow once in heaven: We know her woof, her texture; she is given In the dull catalogue of common things. Philosophy will clip an Angel’s wings, Conquer all mysteries by rule and line, Empty the haunted air, and gnomed mine— Unweave a rainbow . . . (II, ll. –)2
Keats cannot have thought that the profound changes in attitudes toward nature and the supernatural reflected in poems by his contemporaries were caused solely by particular recent discoveries and theories. And he certainly
015*$"-*--64*0/4
knew that not all poets of his generation, not even those who were otherwise in sympathy with his writings, would agree with his disparagements of current scientific studies. Wordsworth, Coleridge, and Shelley we know did not. Still, the general accusation in these lines is convincingly sharpened when put beside the passage they particularly refer to in James Thomson’s long descriptive poem The Seasons, much imitated by his contemporaries.3 Wordsworth and Keats himself later borrowed from it, and Coleridge called its author “a genuine poet.” Thomson’s lines are prepared for by a lavish description of a sunset. Then “the grand ethereal bow / Shoots up immense”: Here, awful Newton, the dissolving clouds Form, fronting on the sun, thy showery prism; And to the sage-instructed eye unfold The various twine of light, by thee disclosed From the white mingling maze. Not so the swain; He wondering views the bright enchantment bend Delightful o’er the radiant fields, and runs To catch the falling glory; but amazed Beholds the amusive arch before him fly, Then vanish quite away. (“Spring,” ll. –)
The first two sentences of tribute to Newton’s “showery prism” in the sky borrow Miltonic favorites like ethereal, immense, awful, various to make the explanation of this much-celebrated scientific discovery sound as little as possible like “a dull catalogue of common things.” The elevation is most noticeable in the contrast between Thomson’s description of “dissolving clouds” that “unfold / The various twine of light . . . / From the white mingling maze” and Keats’s blunt accusation that “Philosophy” will “Unweave a rainbow.” Thomson’s lines nowhere convey a sense of coldness or dullness in the experience of seeing the sun’s rays strike the clouds so that the drops of rain form a prism, but the passage holds back no mysteries from “the sage-instructed eye.” A sense of the miraculous enters only with the uninstructed rustic who “wondering views the bright enchantment” that inhabits the natural world where he expects to find it. What he takes to be an authoritative vision of the invisible the poet calls an “amusive [deceitful, illusive] arch” that leaves the ignorant believer “amazed”: that is, caught in a “maze” from which Newton
015*$"-*--64*0/4
rescued illumined minds. Still, the story of the “swain”—a nostalgic archaism borrowed from Spenserian and Miltonic pastoral—is saved from smugness by the elegiac tone it takes toward the “falling glory” of nature that once promised glimpses of the supernatural. What Keats missed in Thomson’s lines was their sympathetic sense of some possibility now “vanished quite away” from the natural world that it was once believed to show forth in moments of vision. It is to the point here that throughout The Seasons—where the poet addresses God as “Light Himself” who “dwells awfully retired / From mortal eye” (“Summer,” ll. –)— Thomson used the word eye times according to Ralph Cohen, while by my count vision appears fewer than times, in phrases like “gay visions of unruly bliss,” “airy vision,” “narrow vision,” “weary vision,” “crude disjointed vision.” When Thomson used visionary, he meant it in the sense first given it in the eighteenth century, as an adjective to describe superstitious fantasies. This pattern of vocabulary supports the impression that Thomson’s description of the rainbow (and there are many like it by contemporaneous poets quoted in Marjorie Nicolson’s Newton Demands the Muse) hints at uneasy awareness that science and philosophy might undermine some part of poetry’s treasured language surrounding supernatural revelations.4 Visions divinely revealed through the medium of nature, since they are supernatural, can be embodied in any form—a bush, a ram, a loaf of bread, a woman—but above all they have been associated, in biblical literature and poetry steeped in it, with events in the heavens: unexpected effects of light from sun, moon, stars, and the reflections of them in earthly things. Keats chose the rainbow, an archetypal model recently demystified by Newton, as the focus of the unease that he and other poets of his generation came to feel more strongly than had Thomson’s generation. They expressed their dismay in poems built on two similar and sometimes overlapping designs so closely related that both models were practiced, and have been since, by different poets contemporaneously, and sometimes by the same poet. In poems following one design, the poet, when confronted by a mysterious event in nature traditionally celebrated as a conduit for supernatural revelation, finds that it can no longer promise to reveal a vision of “things invisible to mortal sight” yet formerly revealed to Milton’s blind bard. Poems on this model have invited a great deal of critical attention as quintessential Romantic texts: the poet’s experience of hearing the “immortal Bird” in “Ode to a Nightingale” is a famous instance.
015*$"-*--64*0/4
In other poems following the second model, which are the subject of this discussion, the poet reinterprets what are perceived to be more than ordinary natural events by substituting for divine revelation the mysterious perceptual effects of the phenomena we call optical illusions. The term seems to have been coined in the period when Newton’s Opticks was expected reading for the educated: Wordsworth was required to study it at Cambridge in the s. By that time the phrase had also become familiar to a popular audience, judging from its use in William Hooper’s Rational Recreations, already in its third edition in . Its appeal was that it explained with diagrams how to perform famous optical experiments (Hooper said “almost everyone has seen” those using the camera obscura) for private amusement.5 Optical illusions, particularly those caused by the very events in the sky long associated with supernatural appearances, have provided essential materials for the study of bodily vision, as Hooper informed his readers: “The science of optics explains the nature of vision, by investigating the causes of the various phenomena that arise from the refraction and reflection of light.”6 These scientifically useful phenomena have also been shaping motifs in a strand of poetry alive through the past two hundred years. Some deeply underlying reasons why poets have been fascinated with optical illusions are suggested by certain statements of R. L. Gregory, a grand master of visual science. One of his far-reaching propositions is that the “ability to read non-optical reality from optical images in the eyes is the miracle of visual perception.”7 From that premise derives a corollary pointedly applicable to this discussion of optical illusions as material for poems: that they “are more than mere errors: they may be experiences in their own right. They can illuminate reality.”8 Optical illusions have in common with other kinds of illusionary experiences that they are aroused by appearances seeming to contradict physical fact, so that their effects are mysterious, surprising, or bafflingly obscure. Illusionary experiences belonging to the category of optical illusions also have other, more definite dimensions peculiar to themselves, which have made them particularly interesting to poets since the beginning of the nineteenth century. It was then that they began to be used as the focus for a new kind of poem that invests its authority in some experience of an actual natural event not represented as a divine revelation, yet not wholly perceived as empirically accountable. The poems built according to this paradigm have their own
015*$"-*--64*0/4
features distinct from those poems—sometimes by the same author—that center on other kinds of illusions. Hallucination, according to the succinct definition of the term formulated by the nineteenth-century English psychologist James Sully, is distinguished for being a total rather than a partial “displacement from external fact.”9 In the nineteenth century particularly, certain poets became interested in hallucinatory experiences because they lent themselves to allegorical representations of states of consciousness that escape from actual conditions, typically those of urban, industrial society. Wordsworth’s “The Reverie of Poor Susan” shows this design in simple outline.10 The poet’s story tells of “a Thrush that sings loud” at the London street corner where Susan hears it as she passes each morning. Listening, she sees “A mountain ascending, a vision of trees,” a river flowing through Cheapside, “Green pastures.” Her “heart is in heaven,” but the scene fades, “And the colours have all passed away from her eyes!” Though the thrush, a common bird whose song excites Susan’s “vision,” is in some sense a forerunner of Keats’s uncommon nightingale, this poet does not ask, “Was it a vision, or a waking dream?” because he recognizes that what Susan sees is an allegorical representation of what is wholly internal to her imagination and memory. The question here, “What ails her?,” is the poet’s mocking imitation of society asking it, unaware of its own deeply ailing condition as the cause of poor Susan’s escape into delusion. “I went by the Druid stone,” the opening of Hardy’s “The Shadow on the Stone,” predicts a moment of supernatural vision that turns out instead to be what scientists now call a visual confusion.11 The poet in his garden, watching the “shifting shadows,” mistook one for the “shape” his dead wife used to cast there while she was gardening. He resisted the temptation to turn his head, refusing to “unvision / A shape which, somehow, there may be,” but his ways of naming the experience as “my imagining,” “my belief,” “my dream” do some of the work of unvisioning by denying a supernatural source. The empirical reality of a tree casting shade does the rest, not only because the shadows are deceptive—the root meaning of illusion is an act of mocking, deriding, deceiving the bodily eye or mind—but because the poet’s confusion is his own entirely personal error of perception deeply colored by his private grief. The poet’s visual confusion works as a simile: the “shadow” cast by the tree is like the “shade”—the ghostly apparition—of his dead wife. In contrast with both hallucinations and visual confusions, by Sully’s
015*$"-*--64*0/4
definition “optical illusions due to the reflection and refraction of light are not peculiar to the individual, but arise in all minds under precisely similar conditions.”12 In modern scientific terms, optical illusions are systematic, public, mass illusions. Though they induce errors of perception, those effects are excited by actual operations of nature, sharing the permanence of all endlessly repeated natural phenomena. Even given the knowledge of these defining dimensions of optical illusions, their effects still seem to be aroused by unpredictable, mysterious events beyond the ordinary, not accounted for in the phenomenal world. This paradox is smilingly explicated in an imitation of a traditional riddle poem by Gerard Manley Hopkins. Its pseudo-archaic simplicity is playfully combined with the pseudo naivety of a mid-nineteenth-century schoolboy reciting a lesson on the new theories of physiological optics, which, like the geometrical optics of the seventeenth and eighteenth centuries, used optical illusions as sites for experiments in bodily vision. These later discoveries expanded interest in perception as a creative process in which mind and senses—Hopkins’s “thought” and “eye”—interact with external nature: It was a hard thing to undo this knot. The rainbow shines, but only in the thought Of him that looks. Yet not in that alone, For who makes rainbows by invention? And many standing round a waterfall See one bow each, yet not the same to all, But each a hand’s breadth further than the next. The sun on falling waters writes the text Which yet is in the eye or in the thought. It was a hard thing to undo this knot.13
The childlike speaker seems to accept as matter-of-fact this series of propositions with all their twists or self-contradictions of but only, But each. Meanwhile the more prominent repetition and placing of Yet not, yet not, Which yet convey some sense of amazement or mystery: A Grammar of Contemporary English describes yet as a concessive conjunct that signals “the unexpected, surprising nature of what is being said in view of what was said before that.” Mystery challenges matter-of-fact. The final explanation of the riddle draws sun, waters, eye, thought together as if in cooperation that might seem to bring closure, to “undo this knot,”
015*$"-*--64*0/4
when what it does is hint at an answer to the lightly skipped-over riddle of “who makes rainbows by invention?” The misreading of the question as merely rhetorical—who on earth would go around thinking up rainbows?—is corrected by something beyond empirical phenomena: for who (not what) could invent a rainbow but God? This playfully serious riddle is a comment on what, by the s when it was written, was a recognizable kind of poem. Its distinguishing character is that it gives the space once filled by a supernatural vision to a natural event belonging in the category set apart by the late s as an optical illusion. And the distinguishing features of optical illusions, Hopkins’s poem reminds us, are that they are the effects of accurately observed phenomena, and that they are recognized by every observer under precisely the same conditions. The measurably short history of poems built on this plan, along with the special nature of their material, together bring them into unusually narrow focus. For these reasons, such poems allow us a rare chance to watch clusters of poetic conventions in the making. Their origins are in poems of Wordsworth’s early period. It was then that he made his explorations of the expressive possibilities in optical illusions as they are epitomized by the doubleness of the word vision, meaning both a divine revelation of invisible reality and the bodily seeing of empirical phenomena. His fascination with this doubleness shows in his many uses of the word in each of these senses, and by instances where it is charged with both meanings at once. “On the dark earth the baffl’d vision fails” from “An Evening Walk” (l. ) is an early example. He explored these possibilities intensively in a number of his greatest shorter poems, and in passages of The Prelude that have been much discussed, but without attention to their dependence on optical illusions as expressive instruments. The only notice I have found of their special potentialities is in a passage of Geoffrey Hartman’s The Unmediated Vision. It considers a remembered incident in The Prelude (I, ll. –), the boy’s trip in a stolen boat during which he saw a “huge peak” rear its head and grow until it “Towered up” between him and the stars. Hartman’s discussion identifies this as an instance of an optical illusion belonging to the subcategory of dilation.14 Another such is a similarly famous episode (VIII, ll. –) telling how the poet saw a shepherd “In size a giant, stalking through thick fog, / His sheep like Greenland bears,” borrowed from Thomson’s misty scene where “The shepherd stalks gigantic” (“Autumn,” l. ).15
015*$"-*--64*0/4
Hartman’s discussion of the boating incident makes the point, one crucial to the argument of this chapter, that optical illusions were or seemed to be uniquely capable of acting as the natural symbols Wordsworth needed: “Wordsworth is not writing fiction; this [dilation] is a correctly observed optical phenomenon which the sensitive eye may experience at twilight. . . . Even such purely sensory moments could become primary experiences and be interpreted as signs of Nature’s mediating presence.”16 Because this kind of experience is produced by fog, mist, vapor such as turned Thomson’s shepherd and Wordsworth’s into giants, instances of dilation were particularly congenial to the aesthetic of the sublime. Other types of optical illusion are the focus of poems by Wordsworth so closely interwoven that it is tempting, though probably mistaken, to think of them as having been designed together as a group. What they do demonstrate is that Wordsworth found optical illusions to be occasions that might satisfy his longing for adequate sources of material to fill the space traditionally given to supernatural visions. In “A Night Piece,” composed in , the word Vision (very often used by Wordsworth but very rarely capitalized) suddenly appears at the climactic point of the experience, making the poem the centerpiece among others most closely related to it. “A Night-piece” is a short narrative poem in which the traveler is witness to a sudden appearance of the moon racing across the sky. By urgently calling attention to the presence and placing of Vision, the poem forces us to think about the specially weighted expressiveness of the word itself. Its pointed absence from a description in The Seasons of the moon “breaking through the scattered clouds” and riding “up the pure cerulian” (“Autumn,” ll. , ) sharpens attention to the commanding position of Vision in “A Night-piece”: ———The sky is overcast With a continuous cloud of texture close, Heavy and wan, all whitened by the Moon, Which through that veil is indistinctly seen, A dull, contracted circle, yielding light So feebly spread that not a shadow falls, Chequering the ground—from rock, plant, tree, or tower. At length a pleasant instantaneous gleam Startles the pensive traveller while he treads His lonesome path, with unobserving eye Bent earthwards; he looks up—the clouds are split
015*$"-*--64*0/4
Asunder,—and above his head he sees The clear Moon, and the glory of the heavens. There in a black-blue vault she sails along, Followed by multitudes of stars, that, small And sharp, and bright, along the dark abyss Drive as she drives: how fast they wheel away, Yet vanish not!—the wind is in the tree, But they are silent;—still they roll along Immeasurably distant; and the vault, Built round by those white clouds, enormous clouds, Still deepens its unfathomable depth. At length the Vision closes; and the mind, Not undisturbed by the delight it feels, Which slowly settles into peaceful calm, Is left to muse upon the solemn scene.17
The narrative situation of “A Night-piece” imitates a typical occasion of late eighteenth-century religious verse abbreviated in titles of hymns such as William Cowper’s “The Shining Light,” “Looking Upward in a Storm,” “Peace After a Storm,” “Light Shining Out of Darkness.” On the basis of its grounding in fact, the event chosen for “A Night-piece” has also been used persistently as a classic example of optical illusion. According to Gregory’s explanation: When travelling, we see the moon and stars as apparently following our every movement. Intellectually, we know that they are still but are so distant that there is no appreciable parallax change. For objects at terrestrial distances, absence of parallax change with our motion occurs only for objects moving with us—so we see the heavens sharing our motions on earth.18
To his analysis Gregory appended the speculation that it may have been this optical illusion “which led man to believe that the stars do not merely look coldly down, but take an active interest in our individual affairs” by exerting what, in the vocabulary of occult sciences, was called influence. “A Night-piece” dispenses with “Jehovah—with his thunder, and the choir / Of shouting Angels, and the empyreal thrones—,” traditional announcements of supernatural visions that Wordsworth’s preface to the edition of The Excursion ridicules.19 Instead, the opening of “A Night-piece” describes a not unusual natural scene introduced in scrupulously flat, secular, quasiscientific language, only “veil” hinting at mystery. Inside the frame of “At
015*$"-*--64*0/4
length” (line ) and “At length” (line ), the static scene becomes an event in the “sky” visible to the bodily eye that seems to be, although the science of optics teaches that it is not, a mysterious contradiction of empirical reality. The pace changes as the narrator, who is first a detached observer of the traveler, becomes a passionate gazer at the “sky” renamed “the heavens.” The moon, now personified as a goddess in a chariot, “drives” the stars that “Wheel away” at great speed “Yet vanish not”; “still”—silently, motionlessly, continuously—“they roll along.” Then as the clouds draw together, “the Vision closes,” leaving the “mind” to “muse,” as poets do. The definite article in “the mind” and in “the pensive traveller” (in manuscript the adjective is musing), allows the subject to be both solitary and generic. The poet’s mind becomes the collective, perceiving human mind, which cooperates with the now illumined eye in its perception of this “Vision,” unordinary but not supernatural. In a note on the poem Wordsworth recorded that it was written just after what actually happened, as he actually experienced it, which is the impression made by the intense immediacy of the voice. This is an assurance that the appearance of the moon and stars in brilliant, silent motion is not an allegorical figure (like poor Susan’s “vision”), or a similitude (like the shadow on the stone that Hardy’s speaker wanted not to “unvision”), but a natural occurrence used symbolically. The stars are actually “still” in the firmament and “still they roll along” as they are actually perceived by the eye and mind, and as the perception is experienced by the human—any human—watcher of the skies. During and shortly after the marvelous year , Wordsworth explored the expressive possibilities of optical illusions in a number of poems that weave together a common vocabulary to describe a special kind of experience. A title he gave to one poem of – (later printed as a passage of The Prelude) broadly categorizes them: “Influence of Natural Objects in Calling Forth and Strengthening the Imagination in Boyhood and Early Youth.” Its rather pedantic title announces the conceptual—the philosophical and scientific—underpinning of this and the other related poems. Here are some lines where the poet remembers his youthful ice-skating exploits: how on winter evenings he “wheeled about,” and with boyish companions “hissed along the polished ice.” Sometimes he left them To cut across the reflex of a star; Image that, flying still before me, gleamed Upon the glassy plain: and oftentimes, When we had given our bodies to the wind,
015*$"-*--64*0/4
And all the shadowy banks on either side Came sweeping through the darkness spinning still The rapid line of motion, then at once Have I, reclining back upon my heels, Stopped short; yet still the solitary cliffs Wheeled by me—even as if the earth had rolled With visible motion her diurnal round! 20
The visual experiences of swiftly moving natural bodies in the heavens and on the earth here and in “A Night-piece” are closely parallel. They are even described in various grammatical forms of the same verbs: wheel, roll, gleam. The quietest yet most persistent echo between the two poems is the adjective/adverb/conjunction still always bound to verbs of motion: “flying still,” “spinning still,” “yet still . . . / Wheeled by me—.” Both poems exploit the rich possibilities of still for its paradoxically combined meanings of silent, motionless, uninterrupted. At the same time, these parallels show how differently Wordsworth could make the same optical illusion work as a natural symbol to replace the conventions surrounding supernatural revelations. The past tense here, whereas “A Night-piece” is wholly in the present, implies some separation between the poet’s selves as boy and man. The division is marked by the dash after the boy’s astonished notice: “yet still the solitary cliffs / Wheeled by me—.” What he witnessed then seemed to him both visibly real and magical, mysterious. The dash just at that point figures the intrusion of a reflecting mind that cannot be the child’s: “—even as if the earth had rolled / With visible motion her diurnal round!” What had been the boy’s experience of mysterious movements of natural objects the poet recognizes, on amazed reflection, to have been a vision that to the adult is only “as if” a sight of the earth’s invisible motion. The boy saw it rolling round what Wordsworth described in “A slumber did my spirit seal” () as its “diurnal course.” The grandeur in the sound of diurnal and in its combined etymological links to deity and journey invest the empirical event with mythic significance. Myth is the mode of the first part of “There was a Boy,” beginning with that ritual storytelling opening. Composed in , it is placed first in “Poems of the Imagination,” with only “To the Cuckoo,” itself a “visionary” tale, between it and “A Night-piece.” We know from a manuscript of “There was a Boy” in its early stages that it was originally conceived as a first-person memory of the poet’s childhood, like “Influence of Natural Objects.”
015*$"-*--64*0/4
The lines quoted here show more intimate internal connections with those poems and others that mutually depend on the effects produced by the same optical illusions: There was a Boy; ye knew him well, ye cliffs And islands of Winander!—many a time, At evening, when the earliest stars began To move along the edges of the hills, Rising or setting, would he stand alone, Beneath the trees, or by the glimmering lake . . . . . . or the visible scene Would enter unawares into his mind With all its solemn imagery, its rocks, Its woods, and that uncertain heaven received Into the bosom of the steady lake.21
In the first group of lines, earth’s diurnal course is acted out in present participles: rising, setting, glimmering, and—bringing a shock of mild surprise until we think freshly about the noun—evening. The repetition of or, carried into the later set of lines, contributes to the representation of the earth’s rolling motion if we understand or both as an indicator of alternatives—either rising or setting—and as an indicator of an appositive—rising, that is, setting. Though the wheeling motions of the stars are beautifully various, they are not substitutes for each other like ordinary alternatives. Their mysterious continuity is suggested in “I wandered lonely as a cloud” by a simile describing the unending line of golden daffodils to be “Continuous as the stars that shine; / And twinkle on the milky way” even in brilliant sunlight when they are invisible to the human eye. The second set of lines from “There was a Boy” intensifies the mythic atmosphere of the description by their mystifying grammar as much as by their complex interweaving of effects produced by two different optical illusions: the movement of the stars and the appearance of earth and sky in the lake. The same two are compressed in three lines of “Influence of Natural Objects” that describe the boy changing his own course “To cut across the reflex of a star [meaning both its reflection and its response]; / Image that, flying still before me, gleamed / Upon the glassy plain.” That “Image” and the “imagery” in “There was a Boy” are as solidly real and visible to the
015*$"-*--64*0/4
child as rocks, obliterating distinctions between internal experience and external phenomena, between subject and object, as the imagery enters the boy. There is a detailed but inscrutable description of the “visible scene” carrying into the boy’s mind its rocks and trees and “that uncertain heaven received / Into the bosom of the steady lake.” Boy, earth, sky seem to dissolve into one another. Without untangling the grammatical mysteries of this passage, it is still possible to say something about the shock of the more than mildly surprising adjective uncertain. The counterpoising of “uncertain heaven” with “steady lake” disallows the inference that the reflection is wavering in moving water. Though uncertain might suggest that the reflection, as often happens, is not as clearly defined as its original, that understanding is incompatible with the earlier clarity of the description. Something seems to be happening to the poet’s language. The tangled layering of phrases and clauses shows baffled effort to describe what is beyond the limits of grammar. Or does the poet feel that he has made a precarious choice in the word heaven because he has lost confidence in his mythic language? The demystified renaming of “glimmering lake” as “steady lake” allows that possibility. So do the closing lines of the poem, where the poet describes himself having stood “Mute—” at the grave of the boy: that is, at the loss of the certain visions he experienced in childhood, and perhaps at the sense that optical illusions may not work in poems in place of traditional visionary language. The sonnet “Composed by the Side of Grasmere Lake,” written in , is also concentrated on an optical illusion that is among the classic subjects of scientific experiments: the seemingly inexplicable mystery of mirror images. Here the effects of the illusion are treated differently by being described mainly in traditional mixing of classical mythology with biblical language: Clouds, lingering yet, extend in solid bars Through the grey west; and lo! these waters, steeled By breezeless air to smoothest polish, yield A vivid repetition of the stars; Jove, Venus, and the ruddy crest of Mars Amid his fellows beauteously revealed At happy distance from earth’s groaning field, Where ruthless mortals wage incessant wars. Is it a mirror?—or the nether Sphere
015*$"-*--64*0/4
Opening to view the abyss in which she feeds Her own calm fires?—But list! a voice is near; Great Pan himself low-whispering through the reeds, “Be thankful, thou; for, if unholy deeds Ravage the world, tranquillity is here!”22
When in the colorless, still landscape “A vivid repetition of the stars” is revealed in the water below, it is heralded by one of the biblical formulas that announces a supernatural appearance, “and lo!” The poet turns to myth to describe it, but to a borrowed nomenclature. In the octave he does not create a personal mythic language; there is no straining for words in the catalogue of divinities. Far from sounding uncertain, there is a tone of cheerful familiarity in its recitation: ruddy, fellows, happy. Beginning the sestet is a shift into the ordinary wording of the question “Is it a mirror?” followed by an alternative question in an invented mythical language, marked off by dashes, that reaches down into nature’s mysterious depths. Then it, too, is interrupted by a return to another of the formulas that traditionally announce a supernatural revelation, “But list!” The poet submits to the authority of a voice not his own, whispering through the “reeds”—traditional instrument of ancient poets—a resolution that ignores his troubled questioning and his tentative vision of nature’s hidden mysteries. The poem reveals that its author’s invented mythic language was unsustainable. This impression is later confirmed by a sonnet written in , “Composed During a Storm.” It is not a submission but a complete capitulation to another kind of traditional language: one associated with poems built on occasions of religious revelation such as Wordsworth earlier revised in “A Night-piece.” If that poem can be read as a revision of the conventional religious model, then “Composed During a Storm” is not a revision but a reversion to it: One who was suffering tumult in his soul Yet failed to seek the sure relief of prayer, Went forth—his course surrendering to the care Of the fierce wind, while mid-day lightnings prowl Insidiously, untimely thunders growl; While trees, dim-seen, in frenzied numbers, tear The lingering remnant of their yellow hair, And shivering wolves, surprised with darkness, howl As if the sun were not. He raised his eye Soul-smitten; for, that instant did appear
015*$"-*--64*0/4
Large space (’mid dreadful clouds) of purest sky, An azure disc—shield of Tranquillity; Invisible, unlooked-for, minister Of providential goodness ever nigh! 23
To call this poem a capitulation seems just, not because Wordsworth’s passionate effort to find adequate symbols for visionary experience in optical illusions is abandoned, but because the language of the poem is so forced, so stiffened by its rigid conventionality, that its certainties seem to lack conviction. Wordsworth’s search itself and its uncertainties are among his legacies to twentieth-century poets. His uses of optical illusions, even when they seem to fail him or he to fail them, have proved so potently expressive that poets after him have not only continued to explore them; they have made their own revisionary treatment of the conventions shaped and handed down by Wordsworth an active part of the subject matter of their own poems. In one by Wallace Stevens from The Rock, “Note on Moonlight,” there is, or at first seems to be, no comparable figure to the traveler or night-watcher of the skies. There is only the annotator who records “what one sees,” what “is to be observed” of a natural phenomenon: The one moonlight, in the simple-colored night, Like a plain poet revolving in his mind The sameness of his various universe, Shines on the mere objectiveness of things.24
Observations are recorded in the uninflected, impersonal passive voice—“the purpose to be seen”—of scientific writing designed to exclude the subjective responsiveness inevitable where there is human presence. Even so, that presence is evoked in the parenthetical simile of “a plain poet.” This is the poet of another poem in The Rock, “The Plain Sense of Things,” a poet impelled to imagine without imagining the world of physical nature: “The great pond, / The plain sense of it, without reflections.” The “plain poet” of “Note on Moonlight” is also Wordsworth’s, who—experiencing an optical illusion—saw without imagining something beyond sense: for instance, the effect of moonlight, which is to disclose the essential presence, say, Of a mountain, expanded and elevated almost Into a sense, an object the less . . .
015*$"-*--64*0/4
Or the moonlight’s effect is to “disclose” the traveler, “the figure waiting on the road / An object the more”: So, then, this warm, wide, weatherless quietude Is active with a power, an inherent life, In spite of the mere objectiveness of things, Like a cloud-cap in the corner of a looking-glass, A change of color in the plain poet’s mind, Night and silence disturbed by an interior sound, The one moonlight, the various universe, intended So much just to be seen—a purpose, empty Perhaps, absurd perhaps, but at least a purpose, Certain and ever more fresh. Ah! Certain, for sure . . .
To “the plain poet,” no longer a figure in a simile, these apparently mysterious activities of nature seem to close the divide between “the mere objectiveness of things” and “interior sound,” “a change of color in . . . the mind.” Or so Stevens seems to have interpreted what Wordsworth was looking for, what the nocturnal traveler was sometimes allowed to find without having known what he was looking for: a unifying “purpose” of “the various universe.” We can hear both assurance and needy reassurance in “Certain and ever more fresh. Ah! Certain for sure . . . ,” the trailing dots figuring Wordsworthian doubts about the adequacy of moonlight and reflections as symbols of nature’s mysterious being. At the same time the dots signify continuation and renewal, as in nature and perhaps also in the efforts of modern poets to give “fresh” life to poems by revising the conventions they have inherited. There is also sadness in the now personal voice, reflecting some sense of the widened distance between this poet and vital nature. The distance is imposed by scientific and philosophical thinking expressed in the largely abstract, impersonal language that dominates the poem as if “being was to be observed” rather than to be immediately experienced. Eliot, whose early poems did their ironic best to undo the persistent and pervasive influence of Wordsworth, in later poems acknowledged it by sympathetic revisions of his visionary poems: one is in the opening section of “Burnt Norton.” It begins by issuing an invitation from the poet directly to the reader: “My words echo / Thus, in your mind”; “Other echoes / Inhabit
015*$"-*--64*0/4
the garden”; “shall we follow / The deception of the thrush?”—Wordsworth’s and Hardy’s thrush, but also Keats’s “deceiving” nightingale. Once “we” have entered the garden, our guide begins to describe a symbolic setting that he already knows but we do not. It is a revision of a Wordsworthian secluded landscape: a “formal” garden strangely “empty” yet inhabited by the “dignified, invisible” ghostly echoes of dead poetic masters. Distanced behind this methodized pastoral setting is the particular ghostly presence of Wordsworth’s poet watching the sky reflected in Grasmere Lake: So we moved, and they, in a formal pattern, Along the empty alley, into the box circle, To look down into the drained pool. Dry the pool, dry concrete, brown edged, And the pool was filled with water out of sunlight, And the lotus rose, quietly, quietly, The surface glittered out of heart of light, And they were behind us, reflected in the pool. Then a cloud passed, and the pool was empty.25
In these revisionary lines, the waters of a lake rimmed with rustling reeds are displaced by a “drained pool,” not natural but manufactured out of “concrete,” its “dry” sides “brown edged” like old papers. This is a symbolic description of poems informed by the modernist aesthetic in the manner of Eliot’s “Preludes,” where “such a vision of the street / As the street hardly understands” is articulated in a dry, concrete vocabulary designed to strip vision of its traditional aura. In the lines just quoted, “we” who looked down into the “drained pool” no longer includes the reader. “We” are, rather, the narrator’s fellow poets, his contemporaries engaged in a version of Wordsworth’s search for adequate symbols in nature to do the work of language traditionally associated with vision. It follows that “they” who stood behind these modern poets, the dead masters, include Wordsworth in particular, who led them to the expressive possibilities of optical illusions. Following the description of the pool with no more warning than And, the language turns into magical incantation as it describes the mysterious effect of sunlight striking concrete. It filled the dry pool with water mythically begotten “out of sunlight,” and from the depths an exotic flower inexplicably appeared and rose “quietly, quietly” as if in secret. Most mysteriously, “The
015*$"-*--64*0/4
surface glittered out of heart of light,” a perception so strangely phrased that it seems to be translated from a foreign language with an unfamiliar system of grammar. The vision closed when a passing cloud blotted out the sunlight and the mythic language is displaced by an ordinary sentence dryly ending “and the pool was empty.” When Wordsworth’s poet in “Composed by the Side of Grasmere Lake” abandons his mythic vision, the cause does not seem to be that its language is false, but rather that it is not sustainable because nature’s revelations can only be glimpsed, and because the human mind is incapable of retaining them. When Eliot’s poet found the pool again empty, there seems to be a suggestion that the attempt at exotic mythic language to describe the optical illusion is itself empty. There is precedence for this device of calculated failure in Eliot’s early poems: for instance, in “Preludes,” where “vision” is degraded to “ fancies,” false or enhanced imaginings of the city as “some infinitely gentle / Infinitely suffering thing.” If Frost’s “For Once, Then, Something” is reduced to its simplest outline, the sort of reading of his poems he often invited and then thwarted, we can see that it, too, follows the paradigm of “Composed by the Side of Grasmere Lake.” It recounts an experience of a watchful observer who asks a question about something mysterious proposed by nature: the effects of the mirror illusion created by light striking water. Whereas this much can reasonably be said of the poem, its opening line blurs the outline by introducing a stubborn well-watcher who likes to do things the “wrong” way round; it follows that his experience as the poem describes it seems to turn the Wordsworthian model the wrong way round: Others taunt me with having knelt at well-curbs Always wrong to the light, so never seeing Deeper down in the well than where the water Gives me back in a shining surface picture Me myself in the summer heaven godlike Looking out of a wreath of fern and cloud puffs. Once, when trying with chin against the well-curb, I discerned, as I thought, beyond the picture, Through the picture, a something white, uncertain, Something more of the depths—and then I lost it. Water came to rebuke the too clear water. One drop fell from a fern, and lo, a ripple
015*$"-*--64*0/4
Shook whatever it was lay there at bottom, Blurred it, blotted out. What was that whiteness? Truth? A pebble of quartz? For once, then, something.26
Here the reflection of the sky in the water showed, instead of mythological deities, an engaging image of the poet as a wreathed god looking down from a classical heaven. It was not what the reminders of Wordsworth’s sonnet imply Frost’s poet had been looking for, but the sudden exclamation “Once” announces a uniquely intense moment when the well-watcher thought he had “discerned” what he sought. The verb implies detection of something hidden or “uncertain,” a Wordsworthian “Something more of the depths,” more mysterious than what the “Others” who taunted him could see by ordinary vision when they looked “Deeper down in the well” from a different angle. The dash following the “uncertain” glimpse sets apart the moment of more than ordinary vision: “and lo,” a drop blotted out “whatever” might have been revealed. Wordsworth used the traditional announcement “and lo!” in the beginning of his sonnet to herald the reflection of the classical planet-gods in the water below, and later another formulaic exclamation “But List!” to introduce a revelation by another classical deity. Because Frost shifted “and lo” to where it announces the collapse of the visionary glimpse, his poet is heard to mock his own portentous description of a tiny natural event: a drop of water falling into a well. The light mockery extends to Wordsworth’s conventional rhetoric as well, but not to his act of searching for a natural sign of secrets beyond the surface visible to bodily vision. The well-watcher at the end of the poem stubbornly continues his version of the search abandoned in the last lines of Wordsworth’s sonnet. Frost’s poet asks, “What was that whiteness? / Truth? A pebble of quartz?” Without recourse to the authority of a traditional poetic voice, he answers the question in his own skeptical, testing, tentative modern voice. The expressive potency of optical illusions has continued to inform some of the best work of poets of the generations more recent than the great one that included Stevens, Eliot, and Frost. Even when the Wordsworthian paradigms are filtered through the designs of these intervening poets, the history of the conventions built into this distinctive kind of poem still contributes to its special power. A glance at three examples can illustrate new freedoms in the treatment of optical illusions, and in the inferences drawn from them about what they make visible.
015*$"-*--64*0/4
The last section of James Merrill’s “Losing the Marbles” celebrates the transforming effect on things below of reflected “moon- or starlight,” an event that happens beside a “pool.” That is the Wordsworthian outline. It is blurred by the setting, a patio sociably filled with company, and by the choice of mirror-surface: not the water in the pool, but a scattering of marbles— “targets and strikers, / Aggies and rainbows”—given to the poet by a playful friend. More immediate models than any of Wordsworth’s would be the end of Stevens’s “Idea of Order at Key West” (probably the actual setting of Merrill’s imagined scene) where the reflected “glassy lights” of anchored fishing boats are perceived “Arranging” the sea and “enchanting night”; or the closing passage of Bishop’s “The End of March,” where the sun “for just a minute” made “the drab, damp, scattered stones” along the seashore look like “multi-colored” jewels set in a frame. The closing lines of Merrill’s poem begin in a casual, intimate first person that gradually, unobtrusively, ascends to a carefully restrained grandeur matching the effect of the night sky reflected in the marbles “embedded at random in the deck-slats / Around the pool”: By night their sparkle Repeats the garden lights, or moon- or starlight, Tinily underfoot, as though the very Here and now were becoming a kind of heaven To sit in, talking, largely mindless of The risen, cloudy brilliances above.27
There is an ease and amplitude in this description of what the mirror illusion reveals that does not preclude the questioning of it, which is a shaping convention of this kind of poem. There is a mild suggestion of a diminished scale in what is revealed—“as though,” “a kind of heaven,” “largely mindless”—but not grave uncertainty, disillusion, or defiance as in “Note on Moonlight,” “Burnt Norton,” “For Once, Then, Something.” The “very / Here and now,” what is “Tinily underfoot” has its own mysterious “brilliances,” which is what the optical illusion here beautifully reveals. Seamus Heaney’s “Seeing Things” ends with an admiring observer, perhaps a tourist, standing with companions “in bright sunlight” admiring the stone carvings on “the facade of a cathedral.” Its proximate model is Hardy’s “A Cathedral Facade at Midnight,” but there the observer is a Wordsworthian solitary who skeptically watched as the stiff sculptures of dead saints were
015*$"-*--64*0/4
mysteriously brought to life by the moonlight moving across them. In “Seeing Things” it is “Nothing else” but the sculptor’s art—like the great-uncle’s skill with brush and paint in Bishop’s “Poem”—that creates the visual illusion of motion and life in the stone: Lines Hard and thin and sinuous represent The flowing river. Down between the lines Little antic fish are all go. Nothing else. And yet in that utter visibility The stone’s alive with what’s invisible: Waterweed, stirred sand-grains hurrying off, The shadowy, unshadowed stream itself.28
Then in the very last lines nature reaches beyond art by creating a different optical illusion that is more overtly a modern revision of a Wordsworthian vision: All afternoon, heat wavered on the steps And the air we stood up to our eyes in wavered Like the zigzag hieroglyph for life itself.
The “hieroglyph” looks back at what is perhaps the most famous passage of The Prelude (VI, ll. –), where a vision is revealed to the traveler of the “Characters of the great Apocalypse, / The types and symbols of Eternity” written in nature. Here the “hieroglyph” is a sign of “life itself,” of what Merrill’s poem calls “the very / Here and now,” and is equally accepted and celebrated in a mysterious natural light. The closing group of lines in Bishop’s “The Fish” begins with an allusion to the moment in “I wandered lonely as a cloud” when Wordsworth’s traveler recalls his response to a visionary experience in nature with the wondering exclamation “I gazed and gazed”: I stared and stared and victory filled up the little rented boat, from the pool of bilge where oil had spread a rainbow around the rusted engine to the bailer rusted orange,
015*$"-*--64*0/4
the sun-cracked thwarts, the oarlocks on their strings, the gunnels—until everything was rainbow, rainbow, rainbow! And I let the fish go.29
By acknowledging Wordworth’s legacy, the allusion unobtrusively notices the radical revision of his poems of vision in the way Bishop treats the familiar optical illusion of a prism formed by sunlight striking a puddle of oil. This event has the effect on the observer of a sudden brilliant illumination, but the description of it never reaches for mythological or mystifying language to convey its unordinary power. The objects in the little fishing boat are called by their ordinary names, beginning with the unelevated “pool of bilge” where the rainbow formed and spread. The clauses and phrases of the eleven-line sentence are in their usual order, and without pronounced repetition until the very last line. Still, syntax and lineation by subtle cooperation weave together a seamless catalogue of objects in a setting without a trace of association with classical, biblical, or other traditional poetic vocabularies. Gradually the sentence takes on an unassertive incantatory rhythm, as if the poet were momentarily enchanted. The optical illusion in “The Fish” is a figure of the here and now, of ordinary natural and human life itself, of what Bishop in “Poem” calls “the little that we get for free,” which has its own moments of mysterious illumination. “The Fish,” in its watchful look at one such moment, is faithful to “the little of our earthly trust,” which is bestowed on us gratuitously—like rainbows. The “victory” of “The Fish,” in the context of the tradition we have traced, is that the optical illusion of the prism made by light on an oily puddle is effectively detached from what is in Western literature perhaps the most awful of apocalyptic visions: the “bow in the cloud” prophesied to Noah in Genesis, the “triple-color’d Bow” revealed in a vision to Milton’s Adam. Amid Wordsworth’s incalculable contributions to later poetry, he taught some of the twentieth century’s strongest poets a powerful way to end a poem, not with a supernatural vision but an optical illusion.
$IBQUFS
8*- -* " . $ " 3 - 0 4 8 * - -* " . 4±4 3& % & 4 * ( / * / (
8*--*". $ "3-04 8*--*".4,
along with other early modernist poets, marched under Pound’s standard for a revolution in poetry that would make it new. For almost a half a century, Williams experimented in poems with forms that would liberate poetry by exiling fixed stanza forms and meters, and by breaking the back of what—echoing Pound—he called “That cussed iambic pentameter!”1 To further the cause of invention, Williams struggled, much as did other poets in that time of daring exploration, to clarify his own experiments by expounding theories about them that were often both serious and in fun, that favored conviction over consistency. The wackiness of Spring and All is a joyful instance. Inevitably Williams gave the impression not always to have understood his own theorizing. At other times he seems ruefully to have recognized his incomprehension, even while offering to explain one of his inventions with “an example from my own work—not that I know anything about what I have myself written.”2 During his writing life and at least through the s, critics could not agree to disagree over the internal compatibilities of Williams’s ideas or their aptness to his poems. The critical differences, which were large, were not—and by now it seems could not be—resolved by efforts to untangle the web of theories. Although Williams never gave up his efforts to explicate them, he sometimes tried a simpler and more immediate way of making his formal intentions clear by showing his poems in alternative states. This kind of teaching was more in keeping with his ideals for the new poetry than abstract theorizing,
8*--*".4±43&%&4*(/*/(
since simplicity and spontaneity were among them—his famous call to battle being no ideas but in things—and since he had a “favorite image” of a poem as “a small (or large) mechanism or engine” made of parts that can be kept in working order only by “continual redesigning in each period of the world.”3 His chief instrument for “redesigning” was continual revision of alreadymade forms, including his own, a practice that he sometimes spoke of as alien to his aesthetic ideal of spontaneity, the agent of immediacy. He liked to tell inquirers that certain of his poems were made at a single sitting, and claimed with pride that “Very frequently the first draft was the final draft by the time I reached the third book, Al Que Quiere!” In A Voyage to Pagany Williams put in the mouth of his stand-in poet-doctor the opinion that it is of no “use to revise—for the presence is fleeting; revise and the thing escapes; only the footprints are left.”4 Though an apparent self-contradiction, and Williams was susceptible to them, the ways in which he understood revision coincided without canceling one another. In its most general senses it could mean designing “new forms (new associations of the elemental particles),” breaking down everything “to get at the essential,” escaping “traditional forms of the poem.”5 More particularly with respect to his own work, revision was an instrument for treating a poem as a process. The wider and narrower meanings, we will see, could easily overlap in his rewritings of poems. All poets must surely revise at least some of their poems at least to some degree, and must sometimes do so over and over, though few as drastically as, for instance, Marianne Moore. It is likely that periods of experiment in poetry encourage specially energetic exercises in revision. Particularly since the early twentieth century it has been common practice for a poet to publish in an altered form and in a new setting a poem that has already appeared elsewhere. The proliferation of magazines and anthologies that coincided—not coincidentally—with the beginnings of modernist poetry multiplied spaces for that kind of appearance, and poets dedicated to reform took advantage of the situation. What sets Williams somewhat apart from his fellow poets in this exercise is that he showed more interest than they in publicly exhibiting alternative versions of his poetry. He contrived various opportunities for a reader to measure exact changes in a poem. Sometimes he displayed side by side a revision with its original, on occasion accompanied by remarks about his intentions in making the changes. His guides to these practices were painters, from whom he borrowed the term exhibiting for his displays. Often he allowed an earlier
8*--*".4±43&%&4*(/*/(
poem to reappear in public—almost immediately or after a long interval—in a slightly or greatly changed version. Because this discussion focuses on his public displays of revision, only examples of that sort are talked about here. Because for reasons of space Williams used short poems in his demonstrations, only short poems—and for the same reason only some of them—will be explored. An easily penetrable example is his presentation of a poem twice titled “Young Woman at a Window.” The grammar of the title, stripped of the the, is an imitation of titles typically assigned to early twentieth-century paintings. They were often shown in pairs or series, often with the same label, to emphasize their multiple perspectives on their sometimes obscured subjects. What is most unusual if not unique about its publication is that it first appeared, simultaneously, in two forms in in the same autumn issue of The Westminster Magazine. The fact of their common display is an eye-catching signal of Williams’s own special interest in revision as an education in reading the kinds of poems that were being made new: Young Woman at a Window While she sits there with tears on her cheek her cheek on her hand this little child who robs her knows nothing of his theft but rubs his nose Young Woman at a Window she sits with tears on
8*--*".4±43&%&4*(/*/(
her cheek her cheek on her hand the child in her lap his nose pressed to the glass6
The two versions of “Young Woman at a Window” are placed in the magazine and on the page so that readers would naturally first read the poem beginning “While she sits / there.” It is an incomplete melodrama in six laxly measured couplets that leave us not very interested in the story we “know nothing of.” By contrast, the alternative version is a demonstration of how a word—in this instance tears—can proliferate patterns of sound and space that compress and release the energy of a poem. When shown the poems together, the reader is in a position to judge, or at least optimistically assume, that the more controlled version is the revision. Since the choice is not difficult to make, we can guess that Williams had already decided it when he planned to demonstrate the making of the poem as a kind of process. We can be sure at least of his ultimate preference between the two versions because in later appearances “Young Woman at a Window” was always printed only in the form originally placed second in the magazine. Williams played a different kind of instructive game when he juxtaposed two poems titled “The Locust Tree in Flower” in The Collected Earlier Poems. By placing them in reversed order, in time of their published appearances, the earlier one published in in Poetry here coming second, he forced readers to feel the full shock of the radically stripped down later version before turning to what is unmistakably a pastoral lyric: The Locust Tree in Flower Among of green stiff old bright
8*--*".4±43&%&4*(/*/(
broken branch come white sweet May again The Locust Tree in Flower Among the leaves bright green of wrist-thick tree and old stiff broken branch ferncool swaying loosely strung— come May again white blossom clusters hide to spill their sweets almost unnoticed down and quickly fall7
8*--*".4±43&%&4*(/*/(
The abstract design that dominates the abruptly presented revision reduces the eight three-line stanzas to what might be called four stanzas and a coda. Together they add up to thirteen lines consisting of thirteen words that are purged of a vocabulary he might have borrowed from Hopkins (as “wristthick,” “ferncool,” “come May”). And the thirteen-word version omits almost all the interpretive guidance of nouns and verbs: leaves, tree, blossom, cluster, sweets, and swaying, strung, hide, spill, fall are all cut out. The thirteen words neither state nor describe, “Yet the seeming numerically arranged words are not isolated” out of context, as David Walker and others have suggested, because the title—like many assigned to cubist paintings much admired and imitated here by Williams—answers in advance the riddle of which the poem consists: a tree in the early stage of its seasonal round.8 In effect, the title acts its traditional part as a prediction about the poem itself, which here diagrams the seasonal progress of the tree in the pared-down sequence of green, Old, branch, white, May, again. Williams publicly defended this minimalist experiment against its critics, but nevertheless included with it the less radical version, it would seem in order to give a demonstration of what a poem can work with and what, by a rigorous process of concentration, it can do without. The stripping of the tree’s surface is a demonstration of revision getting at “the essential.” “Love Song” first appeared in Poetry in in thirty-three uneven lines making six uneven divisions. It seems, judging from a letter Williams sent at the time to the magazine, that there were two versions of the poem made close in time or perhaps simultaneously, and that he was indifferent as to which of the versions would be published. The characteristically unsatisfactory explanation for his disinterest was that “technically I am only interested in the two main stanzas but of course, left by themselves, they make a bare looking whole.”9 Both versions disappeared from public view until much later, when they were included side by side in The Collected Earlier Poems.10 The poem from Poetry is placed first in the identical form (except for the omission of one comma) but with the startlingly odd title “First Version: .” On the following page the alternative version, reduced to seventeen lines, reverts to the title “Love Song.” The fact that he published the two versions together almost a quarter of a century after he made them, and even though he had been dissatisfied with them, tells more than his vague comment in his letter. It seems that he valued revision for its own sake, not merely as an instrument but as a principle, as a vitally necessary process keeping alive the power of invention, almost as a public duty.
8*--*".4±43&%&4*(/*/(
It is in this spirit that in the s Williams arranged for two very different printings of a poem explicitly about being an experimental poem. He titled the earlier one “A Possible Sort of Song,” and sent it—one suspects part jokingly—to a magazine called The Old Line. He inserted the second version in The Wedge with the less apologetic title “A Sort of a Song,” but still repeating in it his cries to battle: Compose. (No ideas but in things) Invent! Saxifrage is my flower that splits the rocks.11
In an accompanying comment he said of the poem: “It may be that my interests as expressed here are pre-art. If so I look for a development along these lines and will be satisfied with nothing else.” This is revision in the sense of making “new associations of the elemental particles.” The image of the flower as sharp as an ax that can split open rocks seems to derive from some ancient myth of a magic power that penetrates the surface of things to set free their essential being. The detailed readings that follow intend to inquire whether Williams’s exhibitions of poems in successive states suggest vocabularies to clarify the direction of experiments he was typically unable to explain with theories. Williams chose his essay on modern poetry in the May issue of Patroon as his platform or gallery for “demonstration” of a poem he thought worthy of “exhibiting.” He valued it because of its “useful quality for study in that it presents a simple image in the same sort of light that the Athenian placed the Venus—only not in the same context.” While defending the experimentation, Williams admitted in the essay his dissatisfaction with the poem’s form because “it lacks metrical emphasis.” Even so, he said, one would think “somebody with intelligence outside the art,” someone like a gallery owner perhaps, “would make themselves useful by exhibiting” the poem as he was doing by quoting it in his essay: The Girl with big breasts in a blue sweater was crossing the street bareheaded reading a paper
8*--*".4±43&%&4*(/*/(
held up close but stopped, turned and looked down as though she had seen a coin lying on the pavement—12
The opacity of the comment on lack of “metrical emphasis” is partly the result of Williams’s habit of using terms like metrics and stanza even while dedicating himself to breaking down the forms they have traditionally signified, partly due to his failure to fashion a consistent vocabulary for his own formal experiments. More enlightening than his explicit remarks are his successive versions of “The Girl.” In the same year as its appearance in Williams’s essay, he exhibited a revised version in a publication he shared with an artist friend: William Zorach / Two Drawings, William Carlos Williams / Two Poems. The only revision he made at this second stage was that, while retaining the lines of the poem unchanged, he separated its twelve-line unpunctuated sentence into six couplets. Evidently still dissatisfied with the poem’s lack of “metrical emphasis,” he revised the poem yet again, this time remeasuring all the lines to give each two stresses, cutting them down to five couplets, shifting them into the present tense, and making a few small but important changes in punctuation. Here is the final version as it is printed in The Collected Later Poems: The Girl with big breasts under a blue sweater bareheaded— crossing the street reading a newspaper stops, turns and looks down as though she had seen a dime on the pavement13
8*--*".4±43&%&4*(/*/(
The opening lines are a present-tense description something like the portrait “Young Woman at a Window.” Paradoxically, with stops, the poem starts to move, the poet’s eyes following the girl’s simple motions, their direction changed after the midline pause (the only one in the poem). Now description “turns” into action filled with suspense in each space between couplets. The lines and spaces have built to an intensity that makes it seem something very large in its smallness and simplicity has happened. The more than ordinary agent of the event, filling what is ordinary with wonder, is the fine precision of the poet’s seeing become imagined vision: “as though” he could see something that was not there so clearly as to recognize it as a “dime.” The playful last lines of Williams’s “Pastoral” of might be a silent coda to this playfully serious poem about the girl “with big breasts” crossing the street: No one will believe this of vast import to the nation.14
In Williams’s memoir I Wanted to Write a Poem, compiled thirty years into his career as a poet, he remembered thinking back over the revisions he had gathered together for The Complete Collected Poems of : I remembered writing several poems as quatrains at first, then in the normal process of concentrating the poem, getting rid of redundancies in the line— and in the attempt to make it go faster—the quatrain changed into a threeline stanza, or a five-line stanza became a quatrain, as in: The Nightingales Original Version
Revised Version
My shoes as I lean
My shoes as I lean
unlacing them
unlacing them
stand out upon
stand out upon
flat worsted flowers
flat worsted flowers.
under my feet. Nimbly the shadows Nimbly the shadows
of my fingers play
of my fingers play
unlacing
unlacing
over shoes and flowers.
over shoes and flowers. See how much better it conforms to the page, how much better it looks.15
8*--*".4±43&%&4*(/*/(
Critics, among them Marjorie Perloff and Henry Sayre, have used this demonstration of a revised poem annotated by Williams’s rather casual comment as support for their interpretations of his formal theories. For Perloff the later version of “The Nightingales” illustrates the coming into being of Williams’s “wholly distinctive visual”—distinct from aural—“prosody.”16 In Sayre’s view, the poet’s “awareness that the revised version ‘looks’ better has nothing to do either with the poem’s subject matter or with its aural dimension (though the revision does, of course, emphasize certain verbal repetitions).”17 Williams’s comment on the process of revising here is characteristically more enthusiastic than precise, and his memory is typically untrustworthy. He had forgotten, unless he had for illustrative reasons rewritten, the actual history of “The Nightingales.” What in his demonstration he called the “Original Version” was either a fiction suited to the purpose of his presentation, or an unpublished transitional stage between what was actually the first lineation of the poem (not mentioned by Perloff or Sayre) and the revised version in which it ultimately appeared. The earliest published form of the poem, presented in Sour Grapes in , is importantly different from the misnamed “Original Version” Williams chose to illustrate his revising process: The Nightingales My shoes as I lean unlacing them stand out upon flat worsted flowers under my feet. Nimbly the shadows of my fingers play unlacing over shoes and flowers.18
Whereas there is inevitably some difference in effect between a poem divided into stanzas of the same length and one with even a single extra line, there is always a decided distinction in kind between a stanzaic poem and one without internal spaces between lines. The first and most important revision of “The Nightingales” was not, as Williams claimed, the act of balancing the
8*--*".4±43&%&4*(/*/(
numbers of lines in the two sentences of the poem, but their division into stanzas marked by the white space between them. White space separating lines in a poem is always multiple in its generic effects, and when skillfully placed, can have special effects that contribute pointedly to the particular expressiveness of the poem. We can see and hear this happen in a paradigmatic example so familiar and so perfectly achieved that it needs no explication here, Wordsworth’s carefully untitled poem: A slumber did my spirit seal; I had no human fears: She seemed a thing that could not feel The touch of earthly years. No motion has she now, no force; She neither hears nor sees; Rolled round in earth’s diurnal course, With rocks, and stones, and trees.
A stanza space is like the pause at the end of a line in a poem, because it signals that something as yet unknown and different is to come, the expectation being intensified if the line is enjambed. Since the white space that comes after the last line of a stanza is wider than the pause between two immediately adjacent lines, it makes a more marked visual separation and a more audible one because more sustained silence is heard internally even when not voiced out loud. This being so, it predicts a greater change and makes room for that to happen between the lines. If Wordsworth had not separated the two little stanzas as he did, we would not experience the immensity of what they do not say, which is nevertheless expressed by the silence we hear and the emptiness we see in the white space. Inevitably length of line, placing of accent, arrangement of spaces must have their effects on the poem’s subject matter, and on its aural as well as visual dimension. As we will see, when Williams took out of “The Nightingales” the redundant line “under my feet” and replaced it with a stanza break, he made much more happen in the poem than the improved “look” of the newly matched quatrains. In the revised version of the poem, the equivalence of the two stanzas as they are balanced before and after the white space contributes with it to the sight, sound, and sense of the poem. In the two closely adjacent sentences as they originally appear in Sour Grapes, the pause at the end of the first five lines is marked by a period
8*--*".4±43&%&4*(/*/(
that calls for a slightly longer pause than do the ends of the preceding enjambed lines. The punctuation announces that something is finished and something unknown is to begin, but in the absence of a stanza break, the pause alone does not magnify expectations of what will happen (and the anticlimactic last line “under my feet” further dulls the sense of possibilities). Because the period makes an acoustic pause but no visible space between lines, there is less sense of importance attached to the new beginning, and the reader is less likely to look for sameness and difference between two parts of the poem that are unmatched in length as well as divided by syntax and punctuation. In the later version, the break between the balanced little stanzas makes space and time for tremendous change. After the white space, the poem seems to come to life as words and things begin to move about. The rather stolid sequence of discrete monosyllabic words separated by their different initial sounds, particularly in lines and , make the first set of lines seem as “flat,” as two-dimensional, as the woven woolen “flowers.” Then unexpectedly, “Nimbly,” words and things begin to “play / unlacing” freely among the lines: unlacing jumps from line of the first stanza to line of the second; shoes from line of the opening quatrain to line of the second, so that shoes pairs grammatically and, it seems, ontologically with flowers instead of “standing out upon” them incompatibly. Words begin to echo each other’s sounds more distinctly—Nimbly–fingers—and over off-rhymes with flowers. It is as if something magical happened in the white space, turning “shadows” into birds. That they are “Nightingales”—from Williams we might rather expect common American sparrows—suggests the agent of the magic may be experimentation, freeing the poet from traditional poems, allowing his imagination to arrange words and spaces between them so that new forms emerge from old. Although Williams’s “home-made” theories of poetry—to borrow Hugh Kenner’s often quoted epithet—seldom directly illuminate his practice, in a typically abstruse discussion of “Imagination” in Spring and All he made an analogy between birds and words (suggested perhaps by the rhyme) analogous to the image that emerges in his final version of “The Nightingales”: “As birds’ wings beat the solid air without which none could fly so words freed by the imagination affirm reality by their flight.”19 Williams’s poem called “Flowers by the Sea” was published in in the
8*--*".4±43&%&4*(/*/(
fall issue of Pagany. It was reprinted as quoted here in The Collected Earlier Poems: Flowers by the Sea Over the flowery, sharp pasture’s edge unseen, the salt ocean lifts its form flowers and sea bring, each to each, a change Chickory and daisies, tied, yet released seem no longer flowers alone but color and the movement—or the shapes of quietness, whereas the thought of the sea is circled and sways peacefully upon its plantlike stem20
The first three couplets separated by white space represent the motion of seawater as it “lifts its form”: “its” referring at once to the water, the edge of the shore, perhaps also the poem. The “form” is wavelike, repetitious, as are the alternating lines and spaces, yet it predicts a “change” that does not begin to happen until the fourth couplet with “but . . . / . . . whereas” (suggesting that Williams condensed lines in later revisions to avoid that delay). The enjambing of whereas heightens suspense, but then the poem collapses into overexplicitness in “the thought of the sea,” which must have been one cause of dissatisfaction that prompted Williams to make many and varied changes in the poem as it next appeared in Modern Things: Flowers by the Sea When over the blossomy, sharp pasture’s edge, unseen, the salt ocean lifts its form. Chicory and daisies, tied, released seem hardly flowers alone but color and a movement—or a shape perhaps. Whereas the sea is circled and sways harmlessly upon its plantlike stem—21
8*--*".4±43&%&4*(/*/(
Besides its disposal of the giveaway phrase “the thought of the sea,” this revision shows Williams trying to create a feeling of excited motion by breaking up sentences with enjambment over stanza breaks and by inserting periods midway through lines and syntactical units. The most extreme shift of movement happens in the line “perhaps. Whereas” enjambed over the white space. The last line, though an improvement over the earlier treatment, must still have seemed to need the kind of revision that Williams in fact made in the third and last version, printed in An Early Martyr and again in The Collected Earlier Poems as quoted here: Flowers by the Sea When over the flowery, sharp pasture’s edge, unseen, the salt ocean lifts its form—chicory and daisies tied, released, seem hardly flowers alone but color and the movement—or the shape perhaps—of restlessness, whereas the sea is circled and sways peacefully upon its plantlike stem 22
This third state incorporates many of the changes made in the previous version, the most prominent being the reduction to four couplets, but resists others. Instead of chopping up sentences, it threads a grammatically legitimate but mysteriously convoluted sentence from the opening through the closing line. The pivotal point of the poem, whereas, has more force because it marks a more significant change in the language than there is in the other versions. The lines that lead up to it make more use of a tentative vocabulary about what can be known of “form” in nature: seem hardly, but, or, perhaps. The search for nouns to see with—chicory, daisies, flowers, color, movement, shape—culminates (in this last version) in restlessness. The shape and sound of the word softly echoed by whereas figures the ebbing and continuing movement of the sea as the movement of the poem makes us able to experience it with our senses. Then the poem tilts in something like the manner of “The Nightingales.”
8*--*".4±43&%&4*(/*/(
A new and unforeseen event shaping itself in the white space is portended by whereas. Suddenly the poem shifts to a language that goes beyond the range of the senses (crudely stated in the first version by “the thought of the sea”) to where the ocean is “circled”—given its “form”—by some extraordinary agent, and “sways” on its invisible “stem.” The absence of punctuation after the last word allows the movement to go on silently and invisibly into the white space. In the successive stages of “Flowers by the Sea,” and in premonitions glimpsed in poems earlier in this discussion, it is possible to watch Williams reaching toward a kind of poem not recognized in his theories, indeed seemingly rejected by them, and in turn often ignored by critics looking in them for clarification. Among the most often cited is J. Hillis Miller, who accepts as Williams’s fundamental premise such of his familiar quotations as “The first thing you learn when you begin to learn anything about this earth is that you are eternally barred save for the report of your senses from knowing anything about it.”23 Miller’s definition of Williams’s epistemology is a rewriting of this and similar declarations: Objects for him exist within a shallow space, like that created on the canvases of the American abstract expressionists . . . and there is no lure of distances which stretch out beyond what can be immediately seen. Nothing exists but what stands just before the poet’s wide-awake senses.24
In spite of such strongly felt declarations, Williams’s and Miller’s, the opportunity to follow Williams’s revisioning of “Flowers by the Sea” allows us to read it as a counter to this generalization. The poem is an evocation of a force that reaches beyond the power of the poet’s imagination or the poem’s design to fill the ordinary with wonder. The movement of “Flowers by the Sea” is toward transcendence. Williams never stopped publishing poems in varying states of completion, or rather of revision, since his practice of leaving off punctuation at the ends of poems reflects his belief that at least hypothetically they are always susceptible to being redesigned, and that revision is the mother of invention. In his last volume, Pictures from Brueghel, he included a revision of a poem first published four years earlier, both versions with the title “The Loving Dexterity.” To map the direction and expressive effects of his changes, the two poems
8*--*".4±43&%&4*(/*/(
are shown here side by side as Williams displayed “The Nightingales,” with the earlier version on the left: The flower
The flower
fallen a pink petal
fallen she saw it
intact on the ground where Deftly
it lay she raised it
a pink petal
and placed it on its stem again
intact deftly placed it on its stem again25
The earlier poem is a variation of an arrangement exhibited in others of Williams’s experiments discussed earlier. The lines of the opening stanza constitute half the poem evenly balanced across the white space with the second set of lines. They make a design analogous to other beginnings of poems like “The Nightingales,” but more exaggeratedly motionless because without a main verb. That grammatical gap heightens suspense by straddling it over the only stanza break. Again characteristically, the second stanza turns still life into narrative. Strung out as a prose sentence, it would be merely an ordinary telling (in the subject-verb-object sequence required by Pound’s rules for good writing) of an everyday event like the girl crossing the street. Instead, because Deftly is capitalized and isolated by surrounding white space, it stands out as the pivotal word that sets the poem in motion, giving the suspended lines that follow it their more than ordinary importance. Because Deftly is synonymous in sound and sense with Dexterity in the title, it makes the active agent the design of the poem, of which “she” is a part. The more radically experimental form of the later version is of the sort that often incited Williams to offer theoretical explanations that critics tried often unsuccessfully to untangle. Its form is more complex, shifting spatial
8*--*".4±43&%&4*(/*/(
alignments and breaking up grammatical units so that from the beginning it makes the subject matter unordinary; it is not to be expressed or understood in the everyday order of subject-verb-object. There is inside the poem a hidden sentence that does not work like a sentence, strung between she and intact from lines to and twice straddling the white space. Read aloud or silently, the sentence demands that the poem be said more slowly than the earlier version. The greater number of stanza breaks dividing shorter lines gives the pauses an effect—like holding one’s breath—of mystery and suspense, to which the wide separation of subject she and main verb placed contributes. The pivot or turning place here is no longer Deftly, it seems, but intact, made to sound more forceful by internal alliteration—intact. It is also a more semantically charged word that describes the “flower” as an object mysteriously untouched, undefiled, left complete or entire. The agency in this version is “The Loving Dexterity,” which gives the poem a more than human power to reattach the “pink petal”—quintessential rose of poetic tradition—to its life-giving “stem.” As in “Flowers by the Sea,” and in another way in “A Sort of a Poem” and “The Locust Tree in Flower,” the movement of this poem reaches toward transcendence. Still faithful to the modernist ideal of making poems new, Williams’s last version of “The Loving Dexterity” is a redesigning of his own poem and also of what may be the trope of lyric poetry with the oldest uninterrupted history.
$IBQUFS
' 30 4 5 ± 4 ® 0 # 7 * 0 64 ¯ 5 * 5 -& 4
I was working on a book about the titles of poems, they almost always responded, “Then you must be writing about Stevens.” Sometimes they added “and Eliot,” “and Williams,” once in a while “and Moore,” but no one said, “Then you must be writing about Frost.” This pattern of response can tell a good deal about our habits of reading and some of the assumptions they are predicated on: what we think titles are and how they work—with the poem and for the poet or reader. The fact that no one thought to name Frost also says a good deal about his work, generally in relation to American poets who were his contemporaries, and particularly in his ways of titling, where he is no more “undesigning” than in his other formal decisions.1 We rightly assume of most titles what has been nearly always true—at least since the later seventeenth century—that they are chosen to fill the space above the text of the poem by the poet to tell something about it to the reader in abbreviated form, to answer some question the reader might ask, like what the poem is going to be about; what genre it belongs to; who is imagined saying or hearing it. Because historically titles have evolved in conventional forms designed to assert their presentational function, we usually take them for granted, calling them straightforward, by which we mean they can be trusted to do just what they claim. Habitually we ignore titles when reading poems, as if they were external to the whole design, demanding no more interpretive work than a sign on a door telling what we will find once we open it. 8 )&/ * 50 - % '3 *&/% 4
'3045±4®0#7*064¯5*5-&4
Challenges to these assumptions have come from those recent titles, and their rare predecessors before the late nineteenth century, that make some obvious bid for attention. If the language of the title itself seems to need interpretation, then we read it asking the same sorts of questions we habitually ask when we read poems. Stevens’s titling would come to mind as a subject to write about because he is perhaps the first poet to make such attentiongetting titles a specialty. Well-known examples are the elaborate wordplay of “Le Monocle de Mon Oncle,” the insistent metaphorical possibilities of “The Sense of the Sleight-of-Hand Man,” the wacky inconsequence of “The Woman Who Blamed Life on a Spaniard.” Similarly, the allusive incongruities in Eliot’s “Sweeney Agonistes,” Williams’s play with grammatical and lexical double meanings in “Spring Strains,” the suggestive incompleteness of Moore’s “Nevertheless” signal that we must ask questions about the title as a prerequisite for reading the text of the poem. Idiosyncratic, expressive titles like these earlier twentieth-century examples get the attention they ask for because critics know what to do with them. In contrast, most titles are overlooked (almost literally) because they seem to be simply declarative, redundant once we have read the poem, like the sign on a door when we have opened and gone through it. Even so, Frost can show us, they are not ignored by poets who must make the decision to use them, a choice more loaded for twentieth-century poets because it is made from the widened range of title forms available in the past hundred years. This practice of disregarding titles unless, by their egregious vividness or peculiarity, they prevent us from doing so is clearly visible in critical discussions of Frost because his insistent preference, especially in his first five volumes, which define him for most readers, was for simple, declarative titles designed to state what the poem is about: an unqualified noun or gerund, or a noun specified by The or A and modified by an adjective, the most common of traditional title forms. Titles in A Boy’s Will establish from the beginning Frost’s preference for these paradigms: “Stars,” “October,” “Reluctance,” “Waiting,” “Mowing,” “The Vantage Point,” “A Late Walk.” Such titles seem close to being unsusceptible to interpretation, as transparent as a windowpane. Their expressive neutrality is made stronger by Frost’s habitual choice of grammatical modifiers that are not evaluative. Typically they are common nouns doing the work of adjectives as neutral as the nouns they describe: from his first book examples are “Ghost House,” “Storm Fear,” “Wind and Window Flower,” “A Dream
'3045±4®0#7*064¯5*5-&4
Pang,” “A Line-Storm Song.” Their neutrality shows clearly in contrast with the more charged choices in Stevens’s “Of Bright & Blue Birds & the Gala Sun” or Williams’s “The Black Winds.” There seems little or nothing to raise questions about Frost’s apparently self-evident choices of transparently declarative titles, and in fact there has been no critical discussion of them. Virtually the only exceptions among the titles of poems in the whole of Frost’s work to have been given more than passing mention are those few untypically designed to invite attention to themselves, and therefore to their ways of working with the poem, because they do something more than or other than answer some question the reader might be expected to ask about it. These five or six critically noticed titles are worded so that they have multiple or even contradictory implications that make them less informative than ambiguous or enigmatic. Since what Frost called “doubleness, double entendre, and duplicity and double play” is the dimension he always said was the essence of a poem, the fact that such titles are extremely rare in his work, and when his American contemporaries were having fun with them, raises questions to be explored in this chapter.2 “Design” is the title of Frost’s most often included in discussions of the poem, probably because the interpretive questions it invites about itself are of a familiar kind, the same we ask about the language of the poem. Since we have been trained to think that double meanings are the stock-in-trade of twentieth-century poets, readers are likely to recognize the hints in “Design.” It signals that the poem will have to do with patterns (including its own, since it is also a commonplace of criticism that twentieth-century poems are likely to be self-referential), and that at the same time it will be about schemes (which might more obliquely include a reference to the poem, because if students of Frost know anything about him as an artist, they know he was—by his own admission—“not undesigning”). The signals of the title are answered in the octave by the poem’s interest in the intriguing pattern of small natural creatures all white (the key word in the unconventionally repetitive rhyme scheme of the sonnet). The pattern these little white things surprisingly make is likely to be “assorted” with the randomness of happenings along the road, but might be “assorted” like a carefully “mixed” variety of ingredients in a recipe (an alphabet soup of “characters”) or in a potion for a magic rite (“right”). Then the sestet, while carrying on the sense of “design” as pattern (just as the white rhyme is carried forward), brings still closer to the surface its suggestion of sinister plotting. By the couplet the
'3045±4®0#7*064¯5*5-&4
two meanings come frighteningly close to being synonymous in “design of darkness” (made out of, made by darkness), before the fear is questioned, pushed away, covered up by the partly skeptical, partly parenthetical, partly evasive last line: “If design govern in a thing so small.” “Design” is Frost’s only title constituted entirely of a double entendre, and scarcely any double meanings of single words occur in his other titles for poems, except in “The Need of Being Versed in Country Things” and “Nothing Gold Can Stay” (although they appear from the beginning in the titles of volumes: A Boy’s Will, Mountain Interval, A Further Range). The singular use of such a title for this poem might be accounted for by its appropriate significance. The insistent formal design the poem makes may be a figure for the designing act the poet performs to counter the darkness. Frost said of the poem both that it is “a set little sonnet . . . a kind of design,” and that “It’s a kind of poker-faced piece.” He also said, with more telling implications for its singular title, that the sonnet is “very undramatic in the speech entirely,” which may be a hint that its movement is—uncharacteristically—predicted if not predetermined by its title, which was among the many changes Frost made in recasting the earlier poem with the very different title “In White.”3 All the other revisions seem to follow from or be designed to support the double meaning of “Design.” The few other titles given more than a mention in discussions of Frost excite interest because they also offer multiple possibilities of interpretation, but by different means: “The Most of It,” “For Once, Then, Something,” and especially “Come In.” Here there are no words like Design, Versed, or Stay to signal their inherent doubleness; no unfamiliar words to be puzzled out like Gerontion; no words oddly combined, as in “The Emperor of Ice-Cream,” to suggest other than their declared meanings. There is, in fact, nothing unordinary about these phrases except their presence in the title space, where they refuse to act like titles because they seem not to predict anything of what the reader will find in the poem. Instead, these unusual title phrases signal the need to begin by asking questions about them before reading the poem, and then about their relation to it after we have read it, which is a curious reversal of conventional roles, since the title, coming first, has traditionally claimed to prepare the reader for what follows it. If we have some idea of the questions to ask about this kind of title, the reason is that Frost has taught us in his poems and by what he eloquently and often proclaimed were the prerequisites for reading them. We can recognize
'3045±4®0#7*064¯5*5-&4
these titles to be made out of what he called a “sentence-sound,” a verbal package familiar as an everyday idiom we have often heard or might hear in certain ordinary situations where it would be spoken in precisely definable but often multiple or even contradictory tones of voice. In lectures, essays, conversations, and letters Frost gave urgent instructions for reading this kind of phrase: to remember that it is “apprehended by the ear”; that it is “definitely and unmistakably indicated by the context”; and that it “often says more than the words. It may even as in irony convey a meaning opposite to the words.” That is to say, the sentence-sound must be read as dramatic speech, by contrast with the simple “declarative sentence used in making a plain statement” that “is one sound.”4 These lessons apply also to titles. Frost’s characteristically declarative titles make only one sound, as everyone knows who tries to say the title as well as the poem out loud. Even “Design,” despite its double signification, does not allow a variety of tones because it is formally abstracted from speech like a sign on a door saying “Entrance,” while a sentence-sound in the title space offers multiple possibilities to the “ear,” which is “the only true reader.”5 “Come In,” the title of Frost’s least often ignored (after “Design”), can serve as a paradigm of the way these sentence-sounds work differently in the title space from a sign saying “Entrance.” Fittingly, Reuben Brower, who gave to many students and their students what he learned directly from Frost about reading, has given us a demonstration of the questions we might begin to ask about “Come In.” Brower set an example still too rarely followed of reading the title as part of the whole design of the poem. His discussion begins by listening to the way the title “anticipates” in the sounds of its own voice the “doublings of tone in the poem.” “ ‘Come In’ combines the homely ‘come-in’ said to someone at the front door with the romantic invitation to ‘come into the garden,’ into a night retreat of Tennyson or Keats.”6 Brower’s demonstration then follows the sounds of this “delicate poise of tones” through the poem. For instance, “ ‘Hark’ belongs on one side to poetic and religious wonder, on another to . . . daily speech a generation back. Grandmothers used to say ‘Hark’ when they surely wanted to be heard.” With this kind of attentiveness to spoken tones, the discussion traces a pattern of details that suggests how strongly the voice in the poem expresses longing for and resistance to the seductions of the bird’s invitation, “Almost like a call to come in.” Following one of the lessons Frost preached and Brower passed on—that the questions invited by the rich possibilities of a poem’s language have no
'3045±4®0#7*064¯5*5-&4
set limits—we can explore in more detail the way the sentence-sound in the title space takes part in a dramatic interchange with the poem. Standing alone, “Come In” has to be an imperative, but its stresses are undetermined, so Brower could hear it as both a neighborly iambic invitation and a trochaic romantic seduction. It could also be an uninflected pyrrhic or an imperious spondee. The stress pattern would be determined by who says “Come In” and to whom, but the title itself has no settled dramatic situation unless the poem provides it. There the title phrase is said twice by the first-person speaker of the lines, but not as an imperative and not in the cadences possible in the title. In stanza it belongs to an anapest—“Almost like a call to come in”—but in the fifth it is ambiguously anapestic and iambic—“I would not come in” and “I would not come in.” These repetitions give no support for an easy identification of the speaker with whoever says the title, whereas the logic of the story actually disallows it unless the speaker is imitating what the bird’s song “Almost” sounded like, a seductive call. Or he might be warding off its temptations by domesticating it, making it sound like a neighbor’s invitation. Another possibility is that the imperative said before the poem is an echo of what the speaker first called out, which would predict the situation in the poem (though not its title) placed shortly after in the same volume, “The Most of It”: nature returns to the human being who calls out for a response from it “the mocking echo of his own” words. Or, since the poem “Come In” is itself a kind of echo of “Ode to a Nightingale” (with undertones of “The Darkling Thrush”), the voice heard in the title space may belong to poetic tradition, or to Keats’s bird luring the later poet “into the forest dim.” In this script, the question of tone is still quizzically and mischievously left unanswered. Does romantic poetry invite a twentieth-century poet seductively; as if to a friendly neighbor; indifferently; impatiently, because whoever is at the door stubbornly refuses to “come in”? All of these possibilities are allowed by the title “Come In,” making it a perfect fit for Frost’s description of how sentence-sounds—“Talking contraries”—work in his poems to open readings that “unsay everything I said, nearly.”7 That typically mischievous nearly works in precisely the way Almost (itself ambiguously iambic or trochaic) in the poem interferes with a settled hearing of the bird’s “call to come in.” No instructions for reading them come with Frost’s simply declarative titles, and no traditions of criticism have trained us to pay more attention to
'3045±4®0#7*064¯5*5-&4
them than as a way of identifying which poem is to be discussed. “Birches” is an exemplary instance: in its simplicity and directness it expects correspondingly simple acceptance as a self-evident or inevitable choice. There has been no speculation about the reasons for Frost’s decision to use that title, or how it works with the poem or for the poet and the reader, although the often mentioned fact that Frost first referred to the title in a letter of as “Swinging Birches” shows that characteristically he did not arrive at it simply or directly, that it was not self-evident to him.8 “Swinging Birches,” as well as “Birches,” would belong to Frost’s muchpreferred title forms. Both seem modest in claiming to do no more than answer in advance the simplest question the reader might ask, what the poem is going to be about. Grammatically abstracted from the sentence and free of ambiguous meanings or the associations built into spoken idioms, they are in themselves expressively neutral. They seem to give the reader no interpretive work to do with them or with their straightforward relation to the poem. Still, Frost’s deliberateness of choice here, as in all his artful simplicity and directness in other features of his poems, should alert us that we need to read at least some of his many titles in unobtrusive forms with as much attention as the few more obviously puzzling ones, though reading them must be a rather different process, requiring that we learn to ask new and different kinds of questions. “Swinging Birches” would be a congenial choice for Frost, who was the first poet to make present participles a preferred source for titles, so much favored by more recent poets that it no longer seems remarkable. Frost early discovered titling by present participles as a device for getting effects he seems specially to have prized in his first four volumes, where he chose them often and for poems he considered among his best, which are now best known. In A Boy’s Will there are four, one of them “Mowing” of which he wrote, “I come so near what I long to get that I almost despair of coming nearer.”9 North of Boston, with two such titles, gives “Mending Wall” the privileged opening place in the volume. Mountain Interval, where “Birches” appears, includes “Meeting and Passing,” “Putting in the Seed,” and “Range-Finding.” Of the three in New Hampshire, “Stopping by Woods on a Snowy Evening” is the poem Frost called “my best bid for remembrance.”10 The opportunities he exploits in these titles are made possible by unique features of the present participle. One is that it has the awareness of time that belongs to verbs, but in the special form of a continual present. This gives it
'3045±4®0#7*064¯5*5-&4
the kind of immediacy we also hear, for instance, in the sound of the speaking voice in the poem “Mowing,” while the grammatical suppression of the mower as the performer of the action allows the title “Mowing” to generalize it. This double effect would not have been achieved by a title in one of the other forms prominent in Frost’s early volumes, supposing the poem had been called “The Hay-Maker” or “Scythe Sounds.” Besides its power of generalizing without sacrifice of immediacy, the present participle may have attracted Frost as a source for titles by other effects of its unique capacity to act as a verbal noun and a pure nominative. In its gerundive sense it names an action that presupposes a performer who is unnamed; as a noun it names something done, an event that has grammatical status as a thing. Frost hinted at the figurative richness of this double possibility when he wrote: “Everything is an event now. Another metaphor. A thing, they say, is an event. Do you believe it is? Not quite. I believe it is almost an event. But I like the comparison of a thing with an event.”11 “Mowing” names an event, an ongoing action of someone making something happen, “my long scythe whispering to the ground,” and it names something whole and free-standing, like the “hay to make.” The present participle of the title says that the poem is indivisibly about a process and about some made thing, Frost’s characteristic definition of what a poem is: “Something you can do. Make a basket. Make a dress. Make an order of an hour in a class, you know; shape it. It’s a momentary stay against confusion.”12 The choice of present participle allows the title itself to make a figure for the poem “Mowing,” and for all poems. There is one more feature of the present participle that may help to account for its attractiveness to Frost, and even for its special fitness to title “Mowing,” and the poem originally to have been called “Swinging Birches.” The present participle has a marked stress built into it that in both these titles initiates a rhythm that can suggest the kind of movement the title names. In “Mowing” this is carried forward most prominently in the rhyming sounds and rhythm of whispering, Something, something, anything. In those lines of the later poem describing the actions of the “swinger of birches,” the rhythmic pattern that would have been set in motion by the original title builds in a crescendo of present participles: riding, launching, climbing, kicking. What these observations about Frost’s remarkable preference for titling by present participles are meant to show here is that the change from “Swinging Birches” to “Birches” for the title of so characteristic and important a poem was likely not to have been a casual decision, but one involving some sacrifice
'3045±4®0#7*064¯5*5-&4
of expressive possibilities that Frost had already made much of, and continued to develop, in the titles of other poems equally characteristic and important. This claim can be tested only by more observations and questions. Frost left on record one fact about the poem that may suggest a reason for the change of title: “‘Birches’ was two fragments soldered together so long ago I have forgotten where the joint is.”13 The revised title may have been designed to disguise the division, because it distributes the reader’s attention through the whole poem, whereas the original title would focus it on the act of “Swinging Birches” described in lines –. This emphasis would sharpen any sense of division between that passage and the earlier description, almost identical in length, of what ice-storms do to birches. It would also disturb the balance between the two descriptions: the “matter-of-fact” account of ice-storms and the “dream” preference for the game of swinging birches. By weighting the balance heavily toward the second passage, the unrevised title might have threatened the seriously fooling play in the use of matter-of-fact to describe the high poetic language about the reality of ice-storms, while the memory reimagined as a wishful “dream” is detailed in precisely measuring, spoken language. The poem would be ruined if it seemed to value “dream” at the expense of “matter” and “fact.” There is another piece of the poem’s history that may say something more about the change of title. Frost also revised the text of the poem by taking out a parenthetical line that came after line between the two long descriptive passages as “Birches” originally appeared in The Atlantic Monthly of August , and the following year in the first edition of Mountain Interval: But I was going to say when truth broke in With all her matter-of-fact about the ice-storm, (Now am I free to be poetical?) I should prefer to have some boy bend them
It seems plausible that this line was a clumsy attempt to solder the joint between the two fragments, but because the changed title did the work less crudely, the disastrously overexplicit parenthesis was expendable. Together these revisions point to larger issues at stake in Frost’s decision to retitle the poem. The title phrase “Swinging Birches” would make the second description the heart of the poem; the presupposition of an agent built into its present participle would bring to the foreground the performer of the continuously present action. Then the “swinger of birches,” delicately mythologized by that
'3045±4®0#7*064¯5*5-&4
epithet, becomes the hero of the poem whose “play,” one of Frost’s favorite metaphors for making poems, is a game “he found himself.” He is a “boy” like one Frost in a talk on “Attitudes Toward Poetry” described playing the game of hopscotch: You know how to play hopscotch. It’s laid out for you or you lay it out, and you come onto it, you step onto it, and step a figure in it . . . a boy began like this long ago . . . he was—fifteen years old, not much more than that. . . . And he sketches on the ground for himself that way, the court that he makes like hopscotch. Now he’s got to step into that again and again and make pretty figures out of his sentences.14
When Frost, to illustrate this story of “long ago,” recited the first stanza of “Ode on Solitude,” the boy hero turned out to be Pope. The figure of “some boy” in “Birches” is less like Alexander Pope than he is like Wordsworth’s mythologized Winander boy, but he is made to act as Pope does in Frost’s talk, as a figure for all poets, and in particular for Robert Frost: “So was I once myself a swinger of birches.” The detailed account of precisely how the boy played his game confirms the poet’s own boyhood practice of it long ago, and by analogy describes his later game of making poems. It must be played “carefully,” by keeping one’s “poise,” by taking “the same pains you use to fill a cup / Up to the brim, and even above the brim.” It requires learned skills, not “launching out too soon, / And so not carrying the tree away / Clear to the ground,” but still it demands taking risks “till the tree could bear no more,” just as Frost wrote in “Education by Poetry” about the danger of not mastering a metaphor: “You don’t know how far you may expect to ride it and when it may break down with you”: All metaphor breaks down somewhere. That is the beauty of it. It is touch and go with metaphor, and until you have lived with it long enough you don’t know when it is going. You don’t know how much you can get out of it and when it will cease to yield. It is a very living thing.15
Although Frost said “Birches” perhaps more often at public readings than any other poem, he never dropped characteristic hints—like those he gave about “Design,” for example—in talks or elsewhere that would alert readers to the dimension of the poem the unrevised title and unexcised line would have pointed to: that it has to do with making poems, and his poems. Perhaps again wanting to suppress this dimension, he later took out the revealing present
'3045±4®0#7*064¯5*5-&4
participle from the title originally to be “Making the Most of It.” That revision prevents the title from giving a clue to whether its truncated sentence-sound is spoken by the maker of the poem or “He” who acts in it. What, if anything, it signals is the poem’s refusal to answer that question or its corollary, whether “He,” another figure with ties to the Winander boy, is or is not a lesser or younger self of the poet. The only recorded instance when Frost, before saying “Birches,” offered the audience what he said was the “kind of explication I forbid,” shows him slyly supporting rather than undoing the title’s artfully straightforward claim to introduce a poem describing birch trees. His confidential explanation to his audience was only that “I never go down the shoreline [from Boston] to New York without watching the birches to see if they live up to what I say about them in the poem.”16 Still, even in this blandly noncommittal remark there may be a playful hint about the poem that might typify all poems: that the “matter” it works with—precisely how “birches bend”—is “fact.” This conviction would support Frost’s preference for simple title forms pointing directly to a thing or event in the poem that anyone after reading it might choose as its obvious focus, by contrast with more idiosyncratically expressive titles like those interesting to Stevens, Eliot, Williams, and Moore. These contemporaries could appropriately be included among poets who, in Richard Poirier’s words, “have proved far more accommodating than has Frost to critics who like their poems to be about poetry”: The reason, I think, is that while it is possible . . . to infer from Frost’s poems an interest in the drama of poetic “making,” he is some of the time even tiresomely determined not to surrender the human actuality of his poems to a rhetoric by which action is transformed immediately into ritual.17
“Directive” is treated with the same uncritical acceptance as Frost’s titles in simple, declarative forms like “Mowing” or “Birches,” as if he had chosen instead to call the poem “Giving Directions” or “The Road Back.” Listening to these alternatives, we can hear why “Directive” is not that kind of title, does not do the same work of pointing directly to some event or thing the poem is obviously about. The hypothetical alternative titles match the ordinary language of the poem itself, whereas the word Directive belongs to a very different, uncolloquial area of vocabulary. The voice convincingly dramatized in the lines by idiom and cadence could not conceivably say, “The road there, if you’ll let a guide give you a directive,” or “The road there is a directive way back.”
'3045±4®0#7*064¯5*5-&4
Uncharacteristically, Frost chose for the title of this poem a specialized term, scarcely if ever used in daily speech, associated exclusively with certain rare and limited but precisely definable situations. These are spelled out in dictionary definitions for directive as both noun (which is how we tend to read it in the title space) and as adjective (which is how the OED says it began, accounting for its suffix, from medieval Latin directivus): That which directs . . . spec. a general instruction how to proceed or act.
And: Having the quality or function of directing, authoritatively guiding, or ruling. Having the quality, function, or power of directing motion; causing something to take a particular direction in space. (Used especially of the force by which a magnet takes a north and south direction.)18
To take “Directive” as a pointer, like Frost’s earlier preferred titles, to some event or thing in the poem, we would have to put a common word such as Directions in its place by simply translating it, which Frost always argued was the way not to read his poems: I’m always hinting, intimating, and always on the verge of something I don’t quite like to say out of sheer delicacy. And the only thing I have against my friends, the teachers, is some of them are indelicate. They won’t leave it where I left it. . . . I don’t complain. I have complained of translating what I write. . . . And I’ve complained of people who’ve sat side of me when I’ve cracked a joke and said, “He means . . . .”19
Instead of translating the title “Directive,” we should raise questions about it, beginning with those we would ask by treating it as a kind of sentencesound. This is possible, though paradoxical—since it is a toneless term to be met more often in written documents than in speech—because it has built-in associations with certain definable situations. It pertains to the delivery and reception of general regulations from a distant authority that dictates correct behavior under its jurisdiction. It is a jarringly inappropriate word to refer to the situation of a traveler in an unfamiliar countryside being given directions by a native of the place to follow a road that leads to an abandoned town. Once we take the precise choice of the word Directive seriously, the fact that it belongs to an area of vocabulary unlike and even alien to the ordinary language Frost habitually uses, in both poems and titles, shows it to
'3045±4®0#7*064¯5*5-&4
be carefully designed, but not to give us guidance for reading the poem. It does not focus attention on an event or thing in it, as would “Giving Directions” or “The Road Back,” but instead raises the expectation of difficulties to come in our experience of it. Frost hinted at those pitfalls on relatively rare occasions when he included the poem in a public reading. For instance, “I’ll do it slow and you take it straight. But it’s full of dangers, sideways, off, and all that,” and again, “One of my diffidences is about poems like this to read aloud. Some of them, you know, are more open work, and this is a little more closed.” After saying the poem he added, “See, I don’t read it with the same certainty that I should. I feel a little afraid of it.”20 Uncertainties and fears begin to stir early in our experience of reading the poem as we submit, with some degree of resistance, to being directed by a selfstyled “guide” like Spenser’s Archimago, who “only has at heart” our getting lost. This common idiom, which makes the line sound convincingly like the guide’s own speech, is one that should make us uneasy. “I only have at heart” is among those packaged insincerities like “let me say frankly” or “for your own good” that familiarity has taught us to be wary of. Another is “You must not mind.” It is a repeated reassurance here that actually assures us we do mind, and have reason to as we hear the guide describe the ugly and threatening signs of nature reasserting itself over the abandoned town, its houses, orchards, and fields. What the guide wills us to see along the road “back” has none of the grandeur and freedom of the natural scene readers are led back to by the voice saying the opening lines of “Tintern Abbey.” In place of “lofty cliffs” there are giant boulders heaving up like the “great monolithic knees” of some primordial monster. All that remains are cellar holes the guide compares to “eye pairs” spying on us and a “few old pecker-fretted apple trees,” while Wordsworth’s revisitor found “plots of cottage ground” and “orchard-tufts.” Where he saw grass growing up to “pastoral farms / Green to the very door,” this guide describes a field shrunken to a raw and ugly “harness gall,” and nothing is left of the farmer’s house but an overgrown cellar hole “now slowly closing like a dent in dough” in silent parody of domestic living. The guide’s reassuring commands go on disturbing us. He tells us to soothe our fears with self-deceiving sentimentalities, for instance, with a comfortable picture of country life and work easily mocked by its differences from Frost’s farm poems: “Make yourself up a cheering song of how / Someone’s road home from work this once was.” To Make . . . up is not Frost’s verb
'3045±4®0#7*064¯5*5-&4
for making poetry, nor is the verb in “Then make yourself at home,” a still more fatuous self-deception in a place where all traces of home-making have been “lost,” presumably in failed struggling against the galling rural hardship and poverty Frost’s poems often describe. Most mockingly sentimental is the guide’s imperative to shed complacent tears over the meager “dishes” left “shattered” where the children once played “house”: “Weep for what little things could make them glad.” If we have been made uneasy enough by now, having listened with appropriate suspicion to the guide’s directions, we are “lost enough” to be prepared for his final imperative: Your destination and your destiny’s A brook that was the water of the house, Cold as a spring as yet so near its source, Too lofty and original to rage.
This is high poetry: certainly “lofty”—for instance, in using the noncolloquial sense of original as primal, the origin—and loaded with traditionally portentous words like destination, destiny, source (though the placing of the contraction destiny’s at the end of the line lightly emphasizes a slight dip in elevation). These lines are not the colloquial speech we are used to hearing from the voice in the poem, nor do they sound like descriptions of nature in Frost’s other poems, which we are reminded of by the next two casually parenthetical lines: (We know the valley streams that when aroused Will leave their tatters hung on barb and thorn.)
The personal pronoun We here is the first plural (since the seemingly sympathetic and communal us of line ) to join the speaker and the reader. In “We know” we are linked by familiarity with the same countryside, following some dozen uses of you that had, often it seemed with hostility, separated stranger from initiate. Having made this appeal to our shared experience, the guide lets us in on the secrets of what he has been up to: I have kept hidden in the instep arch Of an old cedar at the waterside A broken drinking goblet like the Grail Under a spell so the wrong ones can’t find it,
'3045±4®0#7*064¯5*5-&4
So can’t get saved, as Saint Mark says they mustn’t. (I stole the goblet from the children’s playhouse.)
This is not the way the guide has talked to us before. The voice here is confiding, and confident that “We” will catch the playfulness in the casual “like the Grail,” and in the rustic epithet “the wrong ones” for the nonelect. Now fully included among the initiated, we are ready to receive the final imperative: Here are your waters and your watering place. Drink and be whole again beyond confusion.
These lines return to the “lofty” solemnity and weighted symbolic language that introduced this closing passage, but its unfolding allows us to “get” them as the last move in the game the guide has told us he was playing. The fooling in the lines just before can only prepare us for this final, ringing directive if it is taken as a teasing shock, which Frost seems to have hinted in a talk “On Being Let In on Symbols” given in , six years after “Directive” appeared in Steeple Bush: A parable is a story that means what it says and something besides. And according to the New Testament the something besides is the more important of the two. It’s that the nonelect are supposed to miss and so not go to heaven. Saint Mark says so. . . . And these things are said so as to leave the wrong people out. I love that because it sounds so undemocratic. That’s not because I’m smart either, but I just love to be shocked, don’t you? I like to come right up against something like that.21
Since we have arrived where we were going—as travelers along the road to the abandoned town, as readers to the end of the poem—we have found, have been taught, what is “Back out of all this now too much for us.” There is no escape “beyond confusion,” but to go “back” to the “source,” living through our fear and uneasiness, is to know “confusion,” and in that knowing to be initiated into the game a poem plays to make a stay against it. Frost liked to tell audiences about an exchange he had with a man who complained to him of being confused: “All right, let’s play the game of Confusion. Are you ready? First I’ll ask you. Are you really confused?” “Yes, I am.” “Now you ask me.”
'3045±4®0#7*064¯5*5-&4
“Are you confused, Mr. Frost?” “No—I win!”22
The discussion of “Directive” to this point has proposed that the title plays its part in the unfolding of the poem by its jarring inappropriateness. Its unfitness involves it in the game of getting the reader lost, whether it encourages translation into a more helpful and undisturbing word like Directions, or if instead it promotes acceptance of a dictionary definition of Directive in good faith as a straightforward statement of what the poem is about. It would then require the reader to transform the actuality of the poem into a kind of allegory pointing beyond the traveler, the guide, the road leading back to the abandoned town, the brook, to a stratum of significance where the title would not jar. In this way “Directive,” because of its association with abstract or general commandments from a higher power, might be designed to tease the unwary into turning events and things in the poem into “symbol” and “meaning,” habits Frost often made fun of: “I’ve grown to hate the word symbol. . . . I wish we could change the subject a little and say typifies”; “I want something that hasn’t any meaning; I’m sick of meaning.”23 In such an allegorizing of the poem, the title would be an appropriate sign pointing directively to the high poetry, the allusively symbolic language of the closing passage. It would from the beginning prepare readers for the final imperative. The “wrong ones,” the type Frost also called “these ultimate people” for their tendency to “deal in” ultimate symbols and meanings rather than to “play with them,” would hear it as a solemn “Directive.”24 The guide, no longer a cranky countryman or even a Spenserian wizard, would be transformed utterly into a higher power commanding us to drink from the communion chalice (the broken goblet the guide had stolen and hidden for his own secret game) and be saved. Commenting on the critical reception of “Directive,” Frost hinted that those who elevated it above his other work might be “ultimate people” guilty of overlooking—looking “beyond”—the dramatic situation, the things and events of the poem, so that they have hold of the meaning but miss the experience: “This is the poem that converted the other group. The one these fellows have taken to build my reputation on. The boys [followers of T. S. Eliot] call it great. They have re-estimated me. This is great and most of the rest, trivia.”25 This identification of wrong readers of “Directive” with the “converted” recalls one of Frost’s most famous wordplays, which could serve as a gloss on the reading proposed here of the last line. Comparing himself to Eliot, he
'3045±4®0#7*064¯5*5-&4
wrote: “We are both poets and we both like to play. That’s the similarity. The difference is this: I like to play euchre; he likes to play Eucharist.” In “Directive,” the moves in “euchre” follow the steps in “the game of Confusion.”26 Paradoxically, “Directive” has more in common with titles by Eliot and other modernist contemporaries than is usual among Frost’s titles. Its resemblance is in being oddly angled to the poem by using a vocabulary alien to it, though far less insistently so than, for example, the liturgical title “The Burial of the Dead” for part one of “The Waste Land,” or than Stevens’s “No Possum, No Sop, No Taters” for a somber description of a frozen landscape. Again paradoxically, but not uncharacteristically, the far less obvious mismatch of Frost’s title makes it seem more straightforward while it is actually trickier, harder to pin down. It may be mainly for this reason that critical discussions of the poem have ignored the title. They assume it to be as transparently simple and direct as might be expected of Frost, even while recognizing that the poem itself quite obviously takes special risks in allowing traditional symbolic language and overt allusion into descriptions of the events and things out of which, typically for Frost, it is made. From the beginning Frost preferred to title his poems in declarative noun forms that seem clear to the point of transparency, almost invisible like a windowpane we look right through at what is beyond it. Such titles give up not only “the pleasure of ulteriority” in the doubleness of words and tones of voice so prized by Frost in poems, but also what he called their essential effects of “wildness” and “surprise.”27 Even so, because titles in these simple noun forms are clustered so thickly in Frost’s volumes, especially in the first five, in which he came to be best known, they accumulate a certain unassertive expressiveness. They convey a respectful sense of the familiar solidity and integrity, the inviolable thingness of things, and thing is a key word in the poems. In a letter to Louis Untermeyer letting him in on the fun to be had from ulteriority, Frost gives a clue to why the very simplicity and familiarity of his titles made them serve his designs better than the more exciting effects of expressively charged modernist titles: “I should like to be so subtle at this game as to seem to the casual person altogether obvious. The casual person would assume that I meant nothing or else I came near enough meaning something he was familiar with to mean it for all practical purposes. Well well well.”28 Certainly Frost’s characteristic titles, the many that are habitually ignored, seem “altogether obvious” and “familiar.” They invite the reader to “Come In” (and it is to the point here that Frost repeated the enigmatic title of that
'3045±4®0#7*064¯5*5-&4
evasive poem as the title for a volume of poems aimed, with the help of its illustrations and Untermeyer’s commentary, to appeal to young readers). They offer an easy way into the poem, predicting that it will not raise difficulties, or keep secrets, or intimidate the uninitiated. The seeming straightforwardness of his titles, then, is a form of ulteriority that contributes powerfully to the appearance his poems have of being accessible, perhaps what he hinted in a journal note: “A poem would be no good that hadn’t doors. I wouldn’t leave them open though.”29 It is this appearance of accessibility that most immediately divides them from a quintessentially modernist poem, which assaults the reader with a title that predicts the poem’s obscurity by its own. Frost’s supposedly obvious and self-evident titles are a finely crafted feature of his art, giving their unobvious support to Poirier’s claim that “Frost is a poet of genius because he could so often make his subtleties inextricable from an apparent availability” to a large and inclusive audience of readers.30
$IBQUFS
' 30 4 5 ± 4 %& 4 * ( /
*/+ " / 6" 3: , at the age of thirty-seven, Frost sent a letter to his longtime
friend and benefactor Susan Hayes Ward, literary editor of the prestigious magazine The Independent, with a postscript saying “Sonnet on the next page for my ‘Moth and Butterfly’ book.”1 Tacked on as if casually, it was a doubly designing strategy: a grateful allusion to her acceptance eighteen years before of his first poem ever commercially published, “My Butterfly: An Elegy,” and a playful hint that he was hoping soon to have put together a volume of his own poems that might, perhaps with her sponsorship, include the one she had not been shown before, “In White.” This we know was a serious ambition that he acted on some eight months later by moving his family and poems to England, but in the letter to Ward he made a joke about his imaginary book in the making, picturing it filled with fluttery, winged, vulnerable natural creatures, small and lightweight. His ambition became an actuality the next year when David Nutt published A Boy’s Will. The book included “My Butterfly,” prominently placed just before the closing poem, whereas another poem written in the same period and close in affinity, with the title “To a Moth Seen in Winter,” was left out, and only published in in A Witness Tree. Some postponements of presentation by Frost happened for casual reasons; more of them by calculation. The poem that Frost had joked of to Ward as belonging to his “ ‘Moth and Butterfly’ book” was also excluded from the careful arrangement he gave to A Boy’s Will, even though, joking aside, the postscript must have been in-
'3045±4%&4*(/
tended to suggest connections between “In White” and “My Butterfly” that would make them fit companions in the hoped-for volume. In fact, Frost prevented “In White” from ever appearing in print. Instead, sometime during the ten years between showing it to Ward and offering it to Louis Untermeyer for his anthology American Poetry , Frost radically revised “In White” and retitled it “Design.” There are understandable reasons for him to have rewritten the earlier poem, clearly recognizable when we compare it with “Design,” which eventually came to be one of Frost’s most widely discussed and admired poems. William Pritchard has described it as “exceptional in any terms.”2 Harder to account for are Frost’s successive treatments of and responses to his poem “Design,” some of which have been thoughtfully explored by Richard Poirier in Robert Frost: The Work of Knowing.3 The discussion here is indebted to his for pointing in the direction of further considerations. It is a complex story not only of a craftsman’s desire to make a better poem out of the material, but of what may have been genuine uneasiness, even fear, about the design in nature, or the absence of it, and also of a desire to withhold this hard poem for many years, fearing that its hardness might interfere with his self-presentation as an accessible poet, and to withhold it, furthermore, until, it could be presented, late in his career, in the context of other poems compatible with it. Lawrance Thompson recorded in some detail in his biography of Frost that he had been immersing himself in writings of William James and teaching selections from them to his students at Plymouth Normal School at the time he sent “In White” to Susan Ward, perhaps teasingly hoping the Reverend William Hayes Ward (her brother and the editor of The Independent) would see and be provoked to argument by its Jamesian unorthodoxy.4 An extensive passage capable of having that effect is quoted by Poirier from James’s Pragmatism on the question, debated since the eighteenth century by philosophers and theologians, of “design in nature.” James’s discussion is woven together by his dead-pan repetitions of the term design in various forms: Without nature’s stupendous laws and counterforces, man’s creation and perfection, we might suppose, would be too insipid achievements for God to have proposed them. This saves the form of the design-argument at the expense of its old easy human content. The designer is no longer the old man-like deity. His designs have grown so vast as to be incomprehensible to us humans. The
'3045±4%&4*(/
what of them so overwhelms us that to establish the mere that of a designer for them becomes of very little consequence in comparison. We can with difficulty comprehend the character of a cosmic mind whose purposes are fully revealed by the strange mixture of goods and evils that we find in this actual world’s particulars. Or rather we cannot by any possibility comprehend it. The mere word ‘design’ by itself has no consequences and explains nothing. It is the barrenest of principles. The old question whether there is design is idle. . . . Pragmatically, then, the abstract word ‘design’ is a blank cartridge.5
Frost’s adaptation is an attempt to confront the term design: In White A dented spider like a snowdrop white On a white Heal-all, holding up a moth Like a white piece of lifeless satin cloth— Saw ever curious eye so strange a sight? Portent in little, assorted death and blight Like the ingredients of a witches’ broth? The beady spider, the flower like a froth, And the moth carried like a paper kite. What had that flower to do with being white, The blue Brunella every child’s delight? What brought the kindred spider to that height? (Make we no thesis of the miller’s plight.) What but design of darkness and of night? Design, design! Do I use the word aright?6
The title of the poem is itself a verbal abstraction that suggests absence of meaning, “a blank cartridge,” by contrast with Frost’s characteristic titles made of solid nouns and active present participles. To match this title’s barrenness, the opening lines are unlocated and without context; they “have no consequences and explain nothing.” The first three identify three specimens of natural objects, all small and white, that are held up for inspection outside time and space in a non-sentence, again a contrast with the tangible, earthy sense of place and situation typical of Frost’s most familiar poems. This list of unmoving things is broken off by a dash followed by a rhetorical question coyly evoking a “curious eye,” perhaps a peeping and botanizing philosopher who is a stand-in for the questioner. He has no explanation of the static scene: the
'3045±4%&4*(/
“dented spider” suggests an accidental collision, but the incongruous harmony of misshapen or mutant creatures feels like the work of magic. Then comes a series of further questions in the first three lines of the sestet, at least the first and third suggesting an increasingly personal, uneasy urgency leading up to some climax. But it is ludicrously interrupted by the desperately obscure line , set apart in parentheses in a gauche attempt to excuse its disconnection. The redundant and emptily overexplanatory closing lines that must have been meant to be climactic do not convincingly build on the earlier traces of urgency. Instead they trivialize them. The delay of the governing word Design until the end of the poem turns it into a kind of unanswered riddle. Although the earlier lines show Frost to have been interested in James’s view that the abstract word design is “the barrenest of principles,” by the end “Design, design! do I use the word aright?” sounds anxiously self-questioning and also like a plea to some higher authority—perhaps what James slyly called “a cosmic mind”—for assurance that “design” has some meaning, though it be only “of darkness and of night.” The inadvertent redundancy of that phrase itself “has no consequences” and “explains nothing” in the poem. Besides the evidence in the postscript to Ward that Frost at one time thought his contemplated first book might include “In White,” there are other traces of this early plan in details compatible between “In White” and “My Butterfly”: Thou didst not know, who tottered, wandering on high, That fate had made thee for the pleasure of the wind, With those great careless wings, Nor yet did I. . . . And there were other things: It seemed God let thee flutter from his gentle clasp: Then fearful he had let thee win Too far beyond him to be gathered in, Snatched thee, o’ereager, with ungentle grasp. . . . I found that wing broken today! For thou art dead, I said, And strange birds say. I found it with the withered leaves Under the eaves.7
'3045±4%&4*(/
“Stars,” another poem included in A Boy’s Will, points to what James brusquely called “the design argument.” How countlessly they congregate O’er our tumultuous snow, . . . . . . As if with keenness of our fate, Our faltering a few steps on . . . And yet with neither love nor hate, Those stars, like some snow-white Minerva’s snow-white marble eyes Without the gift of sight.8
While arranging the final order of entries, Frost chose for “Stars” the gloss “There is no oversight over human affairs,” which might just as aptly have been applied to “In White.” Still, affinities like these would not have been strong enough to compensate for the ineptitude of “In White.” The lack of command of its argument, structure, diction, shifts in tone would not have escaped Frost for long. But as we learn from his masterly control over every dimension and detail of the revised poem, he continued to see possibilities for making “In White” into a poem of a powerfully different kind from those in A Boy’s Will or, as it turned out, in Frost’s next two books, published before “Design” first appeared in print in its finished form. “Design” is as unusual among Frost’s typically straightforward, informative titles as is “In White,” but otherwise radically unlike it. In place of the semantic blankness that meets the reader of the original title, “Design” is packed with simultaneous meanings that the title and its consequences in the poem refuse to separate for the reader. It seems that Frost chose the revised title to counter the coyly simplistic assumption at the close of “In White”—“Design, design! Do I use the word aright?”—that there could be a correct answer to the riddle of the spider, the flower, and the moth. All the other revisions seem to follow from or to be fashioned by the multiple meanings of the new title, beginning with the continuous presence throughout the poem of a speaking voice that allows at once coherence and forward movement that are missing from “In White.” This voice is the focus for revisions of line, phrase, word, and rhyme scheme.
'3045±4%&4*(/
The speaker variously evokes the pseudo-serious, playful, or sardonic tones of James’s Pragmatism, for instance, in his contemplation of Darwinian “chance-happenings” that can as easily argue “a diabolical designer” as “a man-loving deity.” Some of Frost’s contemporaries would have recognized a direct reference to the classic argument in Frost’s title (as do some readers now, for instance, this one, following Thompson and Poirier). Other readers of “Design,” hearing the speaker begin the story of his nature walk, would simply accept the ordinary, neutral sense of design as pattern: Design I found a dimpled spider, fat and white, On a white heal-all, holding up a moth Like a white piece of rigid satin cloth— Assorted characters of death and blight Mixed ready to begin the morning right, Like the ingredients of a witches’ broth— A snow-drop spider, a flower like a froth, And dead wings carried like a paper kite. What had that flower to do with being white, The wayside blue and innocent heal-all? What brought the kindred spider to that height, Then steered the white moth thither in the night? What but design of darkness to appall?— If design govern in a thing so small.9
Multiple signals given by the title are answered in the octave by the observer’s interest in the intriguing pattern of small creatures, all “white,” that he came on unexpectedly, like a nature lover catching a glimpse of a violet half hidden by a mossy stone. Or else he discovered what he was expecting to find: a visible demonstration of Darwinian nature at its work. The pattern these little white things surprisingly made was likely to have been “assorted” with the randomness of happenings along the road. Or perhaps it was “assorted” like a carefully “mixed” variety of ingredients in an alphabet soup of “characters” such as children like to spell with, or to stir into a brew for a make-believe magic rite (“right”). There is an absurd cheeriness in the speaker’s description of these grotesque activities, but the disturbing likeness of the moth to a “rigid” (revised from “lifeless”) inanimate object
'3045±4%&4*(/
becomes ghastly when in the last line of the octave we realize that what could be seen of the moth was only “dead wings” because the rest of the creature had by then disappeared inside the spider. The sestet, while carrying on the sense of design as pattern (just as the white rhyme is carried forward) brings closer to the surface its suggestion of sinister plotting. The observer begins to question the scheme of an unknown force that brought the spider to the “height” of its cruel power (one of the lines copied exactly from “In White” for its expressive intensity) and “steered” the moth to its fate (a substitute for the incomprehensible original line dismissing the moth with a parenthesis). Finally in the couplet the word design itself—unmentioned since the title—is not questioned to provoke the right answer for what it means, as at the end of “In White,” but to raise and explore its capacity for a multiplicity of meanings. The penultimate line—“What but design of darkness to appall?”—intensifies the speaker’s fear, even though the story he is telling is about an event that happened in the past. The question “What but” (which requires no answer) is dauntingly inevitable; the drawn-out sound of the second syllable in appall (replacing the redundancy of “darkness . . . night” in the original line) is ominous; and the suspension of the infinitive to appall without an object (to appall what?) hints that a pall is over all, like the “highway dust” in “The Oven Bird.” The two meanings of design come frighteningly close to being synonymous in “design of darkness”: made by darkness but also made out of darkness, as the poem is made. The ambiguous tone of voice in the last line—“If design govern in a thing so small”—belongs to the horrified observer of the nasty little scene, who is also the poet whose “If” is deeply skeptical but still inquiring of possibilities, radically unlike the frightened pleader for meaning at the end of “In White.” Frost’s unusual use of such a title for this poem is accounted for by its appropriate multiple significance. The insistent formal design the poem makes may be a figure for the designing act the poet performs to counter the darkness. Frost said of the poem that “it is a set little sonnet . . . a kind of design,” and another time that the “sonnet is the strictest form I have behaved in, and that mainly by pretending it wasn’t a sonnet.”10 “Design” is “set” in the sense of true to the tradition of its form and “set” in its adherence to rules of its own making, which the title calls to attention by its atypicality. The change in rhyme scheme in the sestet to acaacc in place of the undeviating a rhyme of “In White” in the corresponding lines reflects the force of design that can govern a poem, however “small.”
'3045±4%&4*(/
Frost also said, with more telling implications for its unusual title, that this sonnet is “very undramatic in the speech entirely,” which may be a hint that its movement is, uncharacteristically, predicted if not predetermined by its new title. It is surely one of the most important changes Frost made in recasting the earlier poem and its own peculiar title, “In White.” Delayed appearances of early poems, whether in their original form or as revised (often several times), are well known to have been habitual in Frost’s presentation of himself as poet. Even so, he considered it not in his interest to let this practice be publicly known for fear poems written earlier but newly presented might be thought of as “a falling off of power.” At stake were his reputation and his income: to support his large family he needed, he said, to “become my own salesman.”11 “To a Moth Seen in Winter” is a rare exception of an early poem not published until much later (), to which Frost attached a date of composition: “Circa ” (in private to a friend he dated it or ). Possibly this open admission of belatedness was meant to make acceptable the affinities of the long-postponed poem with “My Butterfly” and other entries in Frost’s first book. Much more characteristically he obscured the order of composition of his published poems to give himself freedom in the arrangement and timing of their presentation, and to preserve his privacy from attempts to track any “upward Darwinian line” in his thinking or to ask “Am I an atheist? Am I an agnostic? How deep do I go?”12 Writing privately to Untermeyer in , Frost admitted that “Design” was one of the calculated “resurrections from the past” that was a great deal older than the “lustrum of its book.”13 Yet in , when he sifted through poems not before printed in book form to offer Untermeyer for his anthology, he mentioned a handful of possibilities presumably of recent completion, and gave his own preference among them, that “You’ll want Design.”14 Of course Untermeyer accepted Frost’s recommendation, which was printed in the anthology along with four other suggestions given by the poet: “Fire and Ice,” “The Grindstone,” “A Brook in the City,” “The Witch of Coos.” Unaccountably he kept “Design” aside although he had so recently recommended it with unequivocal approval; although he must have known it to be stronger than the other three short poems in his selection; and although “The Witch of Coos” could have made a witchy companion to “Design.” Again leaving “Design” out of West-Running Brook in , Frost finally allowed it a space in A Further Range in , fourteen years after the poem had been printed in Untermeyer’s anthology, twenty-four years after Frost
'3045±4%&4*(/
had sent it unrevised to Susan Ward under the title “In White” as a possible entry for his “ ‘Moth and Butterfly’ book.” Once he had exposed “Design” to public opinion, he continued to include it in his collected and selected volumes, but it did not otherwise appear again in print elsewhere except in Come In and Other Poems, one of Untermeyer’s selections of Frost’s poetry carefully watched over by the poet himself. During the summer of , while Frost was struggling to find a sufficient number and suitable arrangement of poems for A Further Range, he confided to his friend and would-be biographer Robert Newdick an opaquely phrased reason why “Design” had not been previously collected in one of his volumes. His explanation was that he had been “fearful the experience was too special.”15 Newdick, apparently bewildered, wrote in his notes on Frost’s confession “What is,—what is not” too special an experience? Frost, it seems, had some sort of general answer in mind when, earlier in the same year, he set down in a letter to a friend his “formula” for a poem: “The subject should be common in experience and uncommon in books. . . . It should have occurred to no one as material.”16 Frost’s public explanation of his long delay in admitting “Design” into one of his own volumes has the ring of his characteristic dodging when responding to personal questions that probed the making of his poems. Speaking to an audience at Bread Loaf School of English in the late s, he told them he had “forgot I had—years ago” the poem “Design” (although he had deliberately put it aside very soon after having arranged a place for it in an anthology made by his most faithful editor). Then, according to his unlikely story, “someone turned it up and it began to get said about” (whereas it was generally ignored for something like fifteen years), “and I put it in the book.”17 All we can be sure of in Frost’s strange treatments of “Design” is that he meant us not to be sure, but his evasions—“I might easily be deceiving when most bent on telling the truth”—are themselves invitations to speculation.18
A Further Range deliberately opens up a range of interpretations, beginning with its title. Before publication Frost wrote to Untermeyer that “all I mean by a further range is the Green Mountains after the White,” a joking pretense of simplicity to match Untermeyer’s simple-mindedly serious suggestion that the title might be understood to mean Frost’s range of “interest in politics and
'3045±4%&4*(/
technique.” He saw to it in the dedication to the book that readers could not miss the duplicity he intended: To E. F. for what it may mean to her that beyond the White Mountains were the Green; yes, and beyond both these were the Rockies, the Sierras, and in thought, the Andes and the Himalayas; range beyond range even into the realm of politics and religion.19
Double meanings in the titles of Frost’s volumes and poems are not common among his typically forthright-seeming presentations, which may have been a component of his fear about “Design” that it might be too “special” to be understood. Something like the same uneasiness seems to be implied by the unusual and surely unnecessary explicitness of the dedication to A Further Range, and even more by the table of contents with its deliberately obvious first heading, “Taken Double,” and the comically redundant subtitles to the poems in this section. Their mock bluntness, many of them sounding like tabloid headlines, are a transparent pretense of clarification: “A Lone Striker / or, Without Prejudice to Industry” and “A Record Stride / or, The United States Stated” are typical examples that begin and end the section. “Taken Singly,” the heading for the next section, professes that the poems in it are not layered with multiple meanings, a claim that no steady reader of Frost would take at face value. Still, the poems here are mostly shorter like the pastoral lyrics in earlier volumes, and have the kinds of straightforward, declarative titles we associate with Frost’s poems of that kind. He placed “Design” in immediate proximity with such a poem, which by its different treatment of related matter might shift the reader’s attention, focusing it on what could seem less “special” to his readers. “On a Bird Singing in Its Sleep” is a sonnet in an unconventional rhyme scheme that is almost an image in reverse of Darwinian nature as it is reflected in “Design,” another sonnet in an unconventional rhyme scheme: On a Bird Singing in Its Sleep A bird half wakened in the lunar moon Sung halfway through its little inborn tune. Partly because it sang but once all night
'3045±4%&4*(/
And that from no especial bush’s height; Partly because it sang ventriloquist And had the inspiration to desist Almost before the prick of hostile ears, It ventured less in peril than appears. It could not have come down to us so far Through the interstices of things ajar On the long bead chain of repeated birth To be a bird while we are men on earth If singing out of sleep and dream that way Had made it much more easily a prey.20
The observer in this sonnet tells of an event in nature small enough to be told in the first two-line sentence, but unusual enough to be worthy of notice. What happened was a single half-sung tune he heard on a moonlit night, the song of a bird “half wakened” perhaps by some sound that caused it in selfprotection to break off its singing. Unlike the observer in “Design,” who asks questions of the miniature scene he cannot fathom, the bird watcher is affectionately familiar with the bird and its song. His meditation on the event is delivered to help “us” understand how in the design of nature it could be that the bird “singing out of sleep and dream that way” could have “ventured less in peril than appears.” Unlike the innocent moth and flower or the cruelly designing spider as they are personified in “Design,” the bird has “inborn” strengths “To be a bird.” It preserves itself: by caution perching “from no especial bush’s height” (mirroring in reverse the motions of the spider in “Design”), and by its cunning “ventriloquist” sound. Following the inward motions of its breathing (“inspiration”), it descends to safety through the enjambed lines that figure the Darwinian “chain of repeated birth” in the sestet. The sympathetic observer of this small happening, speaking steadily in rhymed couplets, is more even-toned than the intensely curious “I” in the other sonnet, whose voice registers a much more troubled and complex range of responses—sequentially but also simultaneously—to the perilous working of Darwinian nature. Only quietly frightening echoes of some such responses can be heard in the slowly developed lines of argument of the bird watcher’s meditation: “the prick of hostile ears,” “in peril,” “more easily a prey” (a chain of words linked by repeated sounds of p and r).
'3045±4%&4*(/
Each sonnet ends in an If clause, but taken singly they are differently proportioned. In “Design” the question asked in the penultimate line takes the form of a rhetorical question signifying that the answer is inevitable, not to be further questioned. Only the dash at the end of the line is a visual extension of it that marks resistance to “design of darkness,” a struggle to escape from a sense of inevitability into skepticism in the last line. If, with its strong metrical emphasis, is not a rhetorical but a pressing open question. It raises a doubtful possibility that is expressed by the strange, awkward, but in the end metrically correct, closing line: “If de / sign gov / ern in / a thing / so small.” In “On a Bird Singing in Its Sleep,” the If of the closing couplet finishes the uninterrupted sentence of the sestet, which is made of enjambed lines that figure the bird’s difficult descent among the “interstices of things ajar” in nature. The meditative speaker’s mastery of the grammatical convolutions in the sestet parallels the poet’s metrical triumph in the closing line of “Design.” In that sonnet the more outwardly evident struggle is fought by the poet alone, yet the “thing so small” that might free itself from the inevitable design of nature includes not only the poet’s lone voice but all other small things, among them the poem. In the sonnet placed to be read just after “Design,” the bird with its “little inborn song” is a companion figure for the poet, who understands and celebrates its triumph in his own. When two poems on adjacent pages of a book reflect each other but with pointed differences, we tend to read the second as a kind of answer or resolution to or denial of the first poem, which must have been Frost’s scheme in his arrangement of these two sonnets. If he had reversed their positions, “Design” could have seemed to mock with its skepticism the measured reflection on Darwinian nature in “On a Bird Singing in Its Sleep.” Instead, it is likely that he wanted the poem placed second to offer a less “special” experience: less probing, less predetermined, less seemingly designing. Having “On a Bird Singing in Its Sleep” on hand for A Further Range may have given him support for at last placing “Design” before the public in the sixth of his eight volumes. A selected list of anthologies that include Frost’s work shows that A Further Range, which widened his audience after it was chosen for both the Pulitzer Prize and the Book-of-the-Month Club, began to be represented in that sort of poetry collection in the early s. The poem most often chosen for them then was “Desert Places.” “Design,” by now a predictable anthology piece in any chosen selection of Frost’s work, only started to gain attention about a
'3045±4%&4*(/
dozen years later, after it was noticed by Randall Jarrell in the Kenyon Review in and by Lionel Trilling in a speech celebrating Frost’s eighty-fifth birthday; it was then printed in the Partisan Review in . The much publicized reevaluations of Frost’s work by these prestigious critics challenged the views of readers who liked to think of him as the “Poet Laureate of Vermont,” confronting them with a very different image convincing to very different readers. Trilling, mentioning only “Design” and “Neither Out Far nor In Deep” as examples, described their effects as “terrifying,” “dark,” “tragic.” Jarrell, in his much more expansively detailed review, pointed particularly to “Neither Out Far nor In Deep,” “Provide, Provide,” “The Most of It,” “Directive,” and above all “Design.” He declared it one of the lesser known of Frost’s “smaller poems” and “the most awful.”21 Jarrell wrote in summarizing the last two lines of the sonnet: “In large things, macroscopic phenomena of some real importance,” the poem says, “the classical mechanics of design probably does operate—though in reverse, so far as the old Argument from Design is concerned; but these little things, things of no importance, microscopic phenomena like a flower or moth or man or planet or solar system [we have so indissolubly identified ourselves with the moth and flower and spider that we cannot treat our own nature and importance, which theirs symbolize, as fundamentally different from theirs], are governed by the purely statistical laws of quantum mechanics, of random distribution, are they not?”22
He ended his discussion of “Design”: “This poem, I think most people will admit, makes Pascal’s ‘eternal silence of those infinite spaces’ seem the hush between the movements of a cantata.” As do most poets, Frost throughout his writing life carefully tracked the judgments of his work by critics, sometimes admitting to a devious version of the practice: “I’ve said it before, that I don’t read criticism about my work but I know all about it from my friends.”23 In his late years, although he had no further single volumes of his poems to invite new judgments, he used the argument ignited by Jarrell and Trilling, without mentioning them, to manipulate the experiences of his readers (as well as to insinuate his public status): “There’s a dispute as to how hard I am. Some think I’m too dark; others think I’m not dark enough.”24 The poems particularly fought over were virtually all from Frost’s last three volumes. At the very center of the controversy was “Design,” a maverick
'3045±4%&4*(/
from its conception as a possible candidate for inclusion in the “ ‘Moth and Butterfly’ book.” It came to be “wrongly regarded” by some, perhaps intentionally by Frost, as “a most celebrated poem of his sixties.”25
Frost’s treatment of “Design,” this tracking of its history has so far shown, can fairly be called mysterious. The hesitations and contradictions in his choices of whether or not, or when and where, to exhibit the poem display a kind of uneasiness that was oddly not connected with the quality of the poem or the judgments of it by critics (who in fact were scarcely aware of it until Jarrell woke them to it). So far as we know, Frost kept his troubled opinion of it to himself until, as he was on the point of publishing it, he confided to Newdick that he had been “fearful” (a stronger way to describe his feeling than, say, “doubtful” or “hesitant”) of its specialness. Whatever else he may have meant by that admission, we know from other statements beginning in his early years of publishing his poems that he did not want them to be accessible only to the “critical few” whose esteem “butters no parsnips,” but to the “general reader who buys books in their thousands.”26 That is to say, although Frost did not say it, what he had so long been afraid of in “Design” was its obscurity. He claimed (and then disclaimed to have said) that he disliked “obscurity in poetry,” and avoided that much-used term as a buzzword of fashionable criticism when scorning what he labeled “the new Movement” led by Pound and Eliot.27 He preferred to call their sort of poem by the workaday word hard—“There have been works lately to surpass all records for hardness”— and mocked them for wanting their poetry “to be mysterious. And if they want it to be—if they’ve got some secrets, let them keep them.”28 In his last years he continued to include in readings a considerable proportion of the earlier poems that were the favorites requested by many readers: “Birches” especially, and what he affectionately called his “little ones,” like “Mowing,” “The Road Not Taken,” “Stopping by Woods on a Snowy Evening.” Among them he slipped in some of the more recently published (although not all more recently written) poems that devotees of “Birches” would have been likely to have described as “too dark.” Frost’s remarks about these less well-known poems were sometimes overtly playful, as if to give assurance to the larger proportion of his listeners that the critical terms used admiringly by critics in the camp of Jarrell and Trilling
'3045±4%&4*(/
were exaggerated. About “Design” he teased: “Here’s one I don’t know so well. Do you want to hear it twice to see how bad it is?”29 Still, there is underlying the playfulness even here some unusual distancing of himself from the poem that he had hesitated so long before showing in a book. On more than one occasion he confessed to uneasiness not altogether pretended about this much-discussed poem. After again reading “Design” for a second time “the better to look at it,” he told the audience, “This is the kind of poem I am never sure of because I am too observing. . . . This one is a very special observation.”30 Or again, “This is a harder one. I may want to say this twice to you.” Following that favorite opening routine, Frost again set apart “Design” as if it were a different and—what was untrue—more recent kind of poem: “See, that hasn’t any tune at all” (by contrast with “Acquainted with the Night” and “Provide, Provide”). “That’s the new way to write” (perhaps a poke at “the new Movement” he associated with the obscurity of Pound and Eliot); “That’s getting all the resonance out of it that you can get out of it.”31 Frost also singled out certain other poems from his later volumes that not coincidentally were among those that critics had compared to “Design.” He surrounded these other atypical poems, especially “Directive,” with the same aura of not wholly disingenuous mystery. Speaking of “Directive”: “One of my diffidences is about a poem like this to read aloud. Some of them, you know, are more open work and this is a little more closed. . . . See, I don’t read it with the same certainty that I should. I feel a little afraid of it.”32 On another occasion: “This is called ‘Directive.’ And I’ll do it slow and you take it straight. But it’s all full of dangers, sideways, off, and all that, you know,” which is his description in everyday language of what is meant by obscurity in poetry.33 Frost’s words of warning about his special choices from among his later volumes sound as if intended to give readers who did not know the poems a pleasing shiver of anticipation. The joking seems also to have been aimed at critics who were afraid of the dark in these poems. Still, the playful cautions and mockery took different directions from his earlier evasions in the new vocabulary he used about the hardness of these special poems. About “Choose Something Like a Star” from “An Afterword” to his last volume, Steeple Bush, he said: “That’s preaching to myself, not you. You might not notice: I say in one place ‘Some mystery becomes the proud.’ . . . It took me a long time to accept obscure poetry, but I decided it was a lofty spirit made them obscure.”34
'3045±4%&4*(/
Taking language doubly is our only tool for responding to what Frost was up to in whatever he said about his poems, perhaps most of all his mysterious handling of “Design.” He once called it “a kind of poker-faced piece,” one skilled in not showing its hand. In the last year of his life, Frost told his audience about his poems: “So many of them have literary criticism in them—in them. And yet I wouldn’t admit it. I try to hide it.”35
$IBQUFS
" /0 / : . * 5 : 5IF-JUFSBSZ)JTUPSZPGB8PSE
5 )& " % +&$ 5 * 7 & " / 0 /: . 0 64 ,
which eventually gave us the noun anonymity, was brought into English from Greek in the late sixteenth century when it often showed off its learned source by its spelling anonymos. It was almost always used to describe a piece of writing or its author, and seems to have carried no generally held associations beyond the translation of its Greek root: “without name.” It stayed close to this lexical meaning until the first half of the twentieth century. Still, by the s the accumulated changes in the cultural assumptions about writers and writings that are without name expanded the implications of the adjective. Then in the early twentieth century, new meanings, some of them expressed by uses of anonymous to describe things unrelated to authorship, began to rub off on the adjective from the noun formed by the addition of a suffix. A noun made from an adjective belongs to a general class whose work is not mainly or usually to change the semantic role, that is, the root meaning, but to shift the grammatical class of the root (as in sad, sadness). In the first instances when anonymity appeared in English, which was not until the second quarter of the nineteenth century and then only rarely, and for more than fifty years after, it seems to have done just that work: to have been used only to describe authors or their writings in the state of being anonymous, of not being known by a name. The early history of the noun anonymity parallels closely the concurrent history of its root adjective. Adjectival nouns formed by adding the particular suffix ity (usually of clas-
"/0/:.*5:
sical or French origin) belong to a subset of the general class by being specially associated with what linguists call greater “lexicalization,” or “vocabulary construction”; when such nouns are formed, something new in the language “has come into existence which did not exist before,” so that their full meanings cannot be recovered from their formation. These nouns, then, can be powerfully compact signals of cultural changes because of the “concentration of presupposition that is packed into a lexicalization by the time it is successfully institutionalized.”1 The adjectival noun anonymity demonstrates this process of word making: it was institutionalized in the early twentieth century. The origin and history of this noun constitute a minuscule model of the sorts of energetic transactions that take place over time among words, poems, other writings, and the pressures in the culture that produces and is produced by them. This model also shapes the discussion here. Its line of argument will trace the literary history of a noun—together with its parent adjective—which has by now become so packed with presuppositions and preoccupations that we can often sense the structure of feelings associated with it to be present even in contexts where the word is not explicitly used. Poems are the spaces where this model can best be demonstrated, because in them the accumulated force of its concentration of meanings can be felt most powerfully.
"OPOZNPVT*UT3PPU.FBOJOH Like vast numbers of other words englished in the sixteenth century, anonymous was imported to serve a newly felt need. It became a conventional shorthand—soon abbreviated anon.—to sign writings whose authors were unknown, particularly poems that were for the first time offered in print to a public who had not had access to them when they were passed in manuscript among privileged circles of readers who might know their authorship without needing to be told. The term anonymous seems not to have been known to Richard Tottel in when he put into print for the first time a miscellany of poems from such a manuscript. In it he grouped unassigned entries under the heading “Vncertain auctors,” itself a borrowing from signatures already used in manuscripts: Incertus author, Jncerti Authoris, the autor unsertayn.2 In miscellanies of the s, many poems were printed with—in place of the author’s name— only a blank space beneath them, or initials, or a personal motto like “My lucke is losse”; or they were supposed in the title to be composed by someone
"/0/:.*5:
using a fictitious name, as in “A louing Epistle, written by Ruphilus a yonge Gentilman, to his best beloued Lady Elriza, as followeth.”3 By the end of the century, perhaps partly because the demand for such books grew as the supply of poems enlarged, the term anonymous was brought into the language. It showed up, under some poems where there is no author’s name or initials, in Englands Helicon of and A Poetical Rapsody of . Anonimus and Anomos are inscribed there, as well as Ignoto, or an occasional pseudonym such as “Sheepheard Tonie.” In Englands Helicon, a few poems are followed by comments from the editor, for instance: “These three ditties were taken out of Maister Iohn Dowlands book of tableture for the Lute, the Authours names not there set downe, & therefore left to their owners.”4 It is clear that in these contexts anonymous was a kind of name, or rather pseudonym, but since adjectives were still considered a category of noun in English as they were in Latin grammar, the free use of anonymous as belonging to either grammatical class says nothing special about the word. 5 In this period we know that it was considered altogether improper for gentlemen and persons of rank to appear in print as poets, so that others who wanted to display their wit as a way of advancing themselves in courtly circles were driven to publish verse unsigned but under fancy disguises that could be seen through: Spenser’s The Shepheardes Calender was the most successful of those projects. Probably because many of the poets in the early miscellanies were no longer living, their being named as the author of an entry did not necessarily signal loss of courtly dignity or lack of gentlemanly standing. Some of that latitude seems to have been granted to living poets in the same books, probably because the provenance of poems was so notoriously uncertain that accusations of unacceptable ambitions could not easily be brought against poets whose random verses showed up in miscellanies. Perhaps for the same reason, a poem or author said to be unknown was not assumed to have been buried in oblivion because of humble origin (an assumption first fully articulated in poetry in Gray’s “Elegy Written in a Country Church-yard” in ). In the first half of the mid-seventeenth century, the absence of an author’s name under a poem in a miscellany or on the title page of a book of verse by a single poet could be read as significantly charged in various, often simultaneous ways. Some social disdain still attached to publication by persons of rank (and by women), and political caution called for some anonymous appearances. Denham’s Cooper’s Hill was printed unsigned in , perhaps because
"/0/:.*5:
he was Sir John Denham, perhaps as much because he was well known as a royalist. There was also still some social pressure against acknowledging one’s printed juvenilia: Milton did not sign his name to A Mask Presented at Ludlow in , but announced it on the title page of Poems of Mr. John Milton in . Toward the end of the century, but more and more often in the eighteenth and much of the nineteenth centuries, so many poets published their work one time or another anonymously, and often not apparently for these traditional reasons, that it is tempting to attribute their unacknowledged appearances more to fashion, publishing tactics, or private caution than to the conventions then being largely emptied of their meanings by the pressures of many social changes. Prominent poets who published at least some of their work anonymously in this period include Dryden, Pope, Swift, Young, Johnson, Akenside, Gray, Beattie, Crabbe, Blake, Wordsworth, Coleridge, Shelley, Landor, Keble, Tennyson, Browning, Arnold, and James Thomson. Motives that impelled these and many lesser-known poets to nameless appearances are sometimes unknowable, or so layered that it would be only guesswork to explain them. Even so, we can see the widespread practice of anonymous publication in this period as a pattern that defines a now altogether unheard of literary situation that made the practice feasible then. This different set of circumstances and the assumptions they supported are almost unimaginable from our perspective. It must be that authors in England well into the nineteenth century could reasonably expect that a poem or book of verse would be bought, read, and reviewed even if it was published unsigned. It may even be that in some instances it would have a better start in the world if it appeared anonymously. That, for instance, was the argument Coleridge made in in a letter to the publisher Joseph Cottle, who apparently had his doubts about bringing out the Lyrical Ballads unsigned: As to anonymous Publications, depend on it, you are deceived.—Wordsworth’s name is nothing—to a large number of persons mine stinks—The Essay on Man, Darwin’s Botanic Garden, the Pleasures of memory, & many other most popular works were published anonymously.6
The reasonableness of this expectation about the fate of unsigned publications must have been built on the circumstances that many fewer volumes of poetry were printed than has been true since; that there was a much smaller and therefore more easily informed literary community of authors, publishers,
"/0/:.*5:
critics, and book buyers; and, more important, that readers had been brought up on different habits of response to poetry, which disappeared in the nineteenth century. To describe this transformation of readers’ expectations in the simplest way: before that time a poem was conceived of mainly as a skillfully made object fashioned according to formal conventions, rather than as a personal expression of its author. Sometime after, let us say, readers grew more in the habit of finding biographical connections between the poem and the poet, a practice naturally encouraged by their opportunities to trace among a number of works the shaping experiences and expressive tendencies of a poet known to be the author of all of them. This interest of readers in finding self-revelation in recent and new poems was pervasive enough by the s to make some poets anxious, early among them Tennyson and Browning at the crucial beginnings of their reputation making. Tennyson’s first published poems were included unsigned in in a volume inscrutably titled Poems by Two Brothers. In he allowed his name on the title page of Poems, Chiefly Lyrical, but gave to one of its entries the distancing title “Supposed Confessions of a Second Rate Mind Not in Unity with Itself.” Along with the scathing epithet, the anachronistic title formula “Supposed” seems to have been designed as self-protection, as a device to separate the author from the imagined (and unnamed) first-person speaker, but there were scarcely any details in the poem itself to support that distancing.7 It was actually discouraged in the text by loud echoes of poems by Shelley, especially “Ode to the West Wind,” where the invitation to conflate poet and speaker is irresistible. Apparently recognizing that his device of separating himself from his poem had failed, Tennyson tried to thwart what he later called the “absurd tendency to personalities” of “almost all modern criticism” by refusing to let the poem be republished for more than fifty years, even using a legal injunction to prevent a journal from printing it in .8 Browning chose the more direct tactic of withholding his name when he published his first poem in , Pauline: A Fragment of a Confession, but must have felt the protection too flimsy because, like Tennyson, he suppressed the poem immediately after interpretations appeared that identified the “author” with the maker of the confession. One such was printed in an unsigned review shortly after the poem was published: The author is in the confessional, and acknowledges to his mistress the strange thoughts and fancies with which his past life has been crowded. This is not
"/0/:.*5:
always accomplished with becoming dignity . . . in language worthy of one who evidently understood them so well; he sometimes runs slip-shod through his afflictions.9
Browning prevented reprintings of this supposed confession until , when, because his authorship of it had become widely known, he included it “with extreme repugnance” in an edition of his poems.10 The desire of poets to escape overpersonal interpretations of their poems added a new reason for anonymous publication when the old ones were fading or had virtually disappeared. Though this cultural change did not expand the meaning of anonymous beyond its lexical root, it did widen its implications. This stretching of anonymous was one of the pressures that created the need for the noun anonymity. It began to crop up in popular journals in the s and on into the early s, perhaps because by then an unsigned book was enough of a rarity to stir interest and debate about the reasons for anonymous publication and its legitimacy.11
"OPOZNJUZ"O"FTUIFUJ D*EFBM The anxiety Tennyson and Browning felt about the improper interest of readers in poetry as personal expression grew to be an almost obsessive concern among writers in this century. It informs the critical statements of poets in the earlier s, and their various efforts to make poems dramatic, meaning by that overworked word what Browning intended when he used it in in the advertisement to his Dramatic Lyrics to describe his poems as “always Dramatic in principle, and so many utterances of so many imaginary persons, not mine.” In their critical writings, synonyms for dramatic are words like impersonal, objective ; often they proposed the figure of the scientist as a representation of the poet, the poem as “a distillation,” “an amalgam,” a “metric figure.” The terms they used to describe their poems—Hardy’s dramatic impersonation, Pound’s persona, Yeats’s mask, Eliot’s objective correlative, Frost’s voice— are attacks on expression and confession. Still, the literary system in the early twentieth century no longer made anonymous publication a feasible choice for these poets, although they sometimes played games with pseudonyms: Pound’s “Alfred Venison,” Stevens’s “Peter Parasol,” Eliot’s “March Hare” and “Old Possum” are examples. By the s, anonymity was introduced into these critical discussions as a term for the concentration of qualities that make up this ideal literature,
"/0/:.*5:
particularly poetry. Sometimes the noun still referred to writings that were literally unsigned, for instance, in an article of in the Nation titled “The Cult of Anonymity.” It gives an account of a contemporary group of unidentified writers in Paris who were withholding their names from their published productions in an effort “to curb the exploitation of personalities, and to establish ‘the art as an ideal, not the ego.’ ”12 Whereas the noun here is attached to writings actually printed without name, its looser meaning—almost as an antonym for personality—smoothed the way for its use elsewhere for written pieces in fact signed by the author but exhibiting the impersonality associated with anonymity by removing the poet, so to speak, from the poem. One place we can watch that extension happening is in a pamphlet published in by E. M. Forster with his name on the title page under the calculatedly impersonal title Anonymity: An Enquiry, which makes the essay sound like a disinterested investigation of a public issue. It is not that, but an urgently felt argument for a return to the literary past when “writers and readers . . . did not make a cult of expression as we do to-day. Surely they were right, and modern critics go too far in their insistence on personality.”13 Forster, we know, had private reasons for wanting to deflect public interest in autobiographical expression; he distanced his argument that “all literature tends towards a condition of anonymity” from himself as novelist by choosing poems as his illustrations, both signed and not signed.14 By obliterating that distinction—between poems literally without name and those, though signed, that suppress the author’s personality—from his description of the literary ideal, he cut the noun anonymity from its root meaning. John Crowe Ransom went on with the argument in his essay “A Poem Nearly Anonymous,” first published in with the subtitle “The Poet and His Formal Tradition”: “Anonymity, of some real if not literal sort, is a condition of poetry. A good poem, even if it is signed with a full and well-known name, intends as a work of art to lose the identity of the author.”15 Ransom, arguing more tendentiously than Forster, took aim at “those moderns to whom ‘expression’ seems the essential quality of poetry,” and at young poets who begin misguidedly by trying to “write their autobiographies, following perhaps the example of Wordsworth.”16 He chose Lycidas (signed only J. M. as it was first printed) as his paradigmatic poem to make his case for the necessity of “training,” “technique,” skill in the making of a “fictitious personality,” to achieve the perfection that only “Formal Tradition” can give the poet: “Lycidas is a literary exercise, and so is
"/0/:.*5:
almost any other poem earlier than the eighteenth century.”17 This is a more pugnacious statement of what Forster meant when he chose as his models of literary anonymity the poets in the Greek anthology, “who would write and re-write the same poem in almost identical language, their notion being that the poem, not the poet, is the important thing, and that by continuous rehandling the perfect expression natural to the poem may be attained.”18 Ransom’s more decisive statement explicitly added spontaneity and sincerity to the antonyms for poetic anonymity in its aesthetic sense.
"OPOZNJUZ"$VMUVSBM.PUJG In an article by the literary journalist Henry Seidel Canby in titled “Anon Is Dead,” both anonymous and anonymity are key words in the explanation he proposed for the phenomenon discussed earlier: that in the twentieth century works of literature rarely appear in print unsigned, whereas before there were as many important books of all sorts first printed anonymously as with the writer’s name attached. The blame for this historical change is not to be laid to the “vanity” of authors, but to the longing for “escape from the deadly anonymity of modern life,” which generates the “passion for nonanonymity” in the “general man who feels his personality sinking lower and lower into the whirl of indistinguishable atoms to be lost in a mass civilization.” As a consequence, this “general man” longs for “vicarious experience” to be found in the intimate writings of “rich, glaring personalities” whose well-publicized names become legendary.19 Here anonymous still keeps strictly to its literal meaning, excluding the figurative sense that Forster and Ransom appropriated it for, as well as to its original attachment to writers and writings. Meanwhile anonymity has not only traveled a long way from its grammatical starting point but bears no other resemblance to the aesthetic ideal called anonymity than its link with impersonality, here denounced as a social evil. In the language of the article, anonymity has become, in effect, a new word with a new frame of reference. It is associated with “City living” which is “essentially impersonal,” but also with “standardized” feelings, with “science” and its “laws that . . . ignore individuality completely,” with all that to the writer constituted “mass civilization.” This disconnection allows the author of the article to paint himself into a lexical contradiction: that the disappearance of “anonymous”—in the sense of unsigned—in modern literature is due to the inescapable presence of “anonymity,” understood as a cultural phenomenon. The writer’s apparent unawareness
"/0/:.*5:
of the verbal oddity in using “anonymity” in this proposition, where it has nothing whatever to do with a piece of writing lacking the author’s name, shows that his generalizations, though heated, are not personal discoveries but packaged phrases that by the s expressed assumptions so deeply and commonly held they could be accepted without being questioned. They may have entered popular writing like Canby’s by the second quarter of the twentieth century from the vocabulary of sociology, where anonymity is a heavily charged term, but their origins, as the social historian William Cronin suggested in , are “in the lessons we learned from nineteenth-century Romantic writers like Wordsworth” and others who followed him.20 There, indeed, we can find a vocabulary of images repeated over and over with an expressive force that pressed toward the institution of anonymity in the cultural sense that Canby used it for automatically. In the last years of the eighteenth century, Blake’s poems, printed with his name on the cover, prophesied the condition of modern anonymity. “London,” beginning with its implacably impersonal title, forges a link between the suffering of modern city dwellers and society’s denial of their individual identity. The visionary who says, or chants, the poem reads “Marks of weakness, marks of woe” in “every face,” “every cry,” “every Man,” “every Infant,” “every voice,” while in cruel contrast “each charter’d street” where the solitary poet wanders among this undifferentiated humanity is at least granted “each” its singular identity.21 A more insidiously particular connection between the indifferent cruelty of urban society and namelessness is implied by the earliest to be printed of Blake’s two poems titled “The Chimney Sweeper.” The lines of the poem are spoken by the boy sweep, who was orphaned as an infant, perhaps before he had even a name. His speech is in part a kind of nursery rhyme full of affectionate nicknames for his child companions in labor, in contrast with the title of the poem, which names the boy only by the hellish work he is forced to perform. He accepts himself into that category when he announces his presence by the chimney sweeps’ cry, childishly lisping it “ ‘ ’weep! ’weep!’ ” The lesson he repeats in the last line in the language imposed on him by society teaches the children the same acquiescence by making them think of themselves in anonymous third-person plurals: “So if all do their duty they need not fear harm.”22 Wordsworth had completed a first version of his verse narrative about his residence in London by , only a few years after he must have read Blake’s poems, but it was not published until shortly after his death in (as Book
"/0/:.*5:
of The Prelude). Soon it became a locus for the concentration of attitudes and feelings since gathered around the noun anonymity. They are linked to its root meaning at the beginning of the narrative by lines remembering how the poet pondered what he had heard of London as a boy before he went to live there: Above all, one thought Baffled my understanding: how men lived Even next-door neighbours, as we say, yet still Strangers, not knowing each the other’s name. (ll. –)23
Here the poet’s voice sounds easy, smilingly imitating the familiar speech among “neighbours,” identifying with them in the comfortable plural “we.” By contrast, in passages opening and concluding his narrative, he denounces the city in the tone of a prophet viewing it panoramically from outside space and time: Rise up, thou monstrous ant-hill on the plain Of a too busy world! Before me flow, Thou endless stream of men and moving things! (ll. –) The comers and the goers face to face, Face after face . . . (ll. –)
And at the end: Oh, blank confusion! true epitome Of what the mighty City is herself To thousands upon thousands of her sons, Living amid the same perpetual whirl Of trivial objects, melted and reduced To one identity, by differences That have no law, no meaning, and no end— (ll. –)
The overpowering sense in these passages is of huge masses in senseless motion: “men” indistinguishable from one another, scarcely even from
"/0/:.*5:
“moving things”; reduced to synecdoches of “face after face”; unknown and unconscious living particles swept in a flood. The descriptive passages between these framing visions illustrate them in details that fill in the panoramic denunciations. Some picture lonely urban scenes of “empty streets” at late hours Of winter evenings, when unwholesome rains Are falling hard, with people yet astir, The feeble salutation from the voice Of some unhappy woman, now and then Heard as we pass, when no one looks about, Nothing is listened to. (ll. –)
Then there are glimpses among the “sundry forms,” “single forms and objects,” “commingled shapes,” “shapes which met me on the way” of some one suffering creature who emerges from the “huge fermenting mass of human kind” like a ghostly apparition: Behold, turned upwards, a face hard and strong In lineaments, and red with over-toil. ’Tis one encountered here and everywhere . . . (ll. –)
Or “See, among less distinguishable shapes / The begging scavenger with hat in hand.” For the young poet the epitome, the “type, / Or emblem” of humanity represented in London’s “overflowing streets” was the “blind Beggar, who, with upright face, / Stood, propped against a wall” like an object, with a “label” hung on his chest telling “His history, whence he came, and who he was,” without, perhaps, giving his name: And, on the shape of that unmoving man, His steadfast face and sightless eyes, I gazed, As if admonished from another world. (ll. –)
The city envisioned first in Blake’s poems and later as Wordsworth described it more fully in the version of The Prelude became an epitome of the modern condition for poets, essayists, novelists, and their readers in the
"/0/:.*5:
second half of the nineteenth century and on into the twentieth. Similar uses of urban materials, the same clusters of images, crop up everywhere, cumulatively, in phrases and paragraphs and as the structural design of whole works, making it irrelevant and often impossible to trace the formulations, beyond likenesses, to some particular source. A brief catalogue of examples shows how, in very different writings, what Raymond Williams has called “a common structure of feeling was being formed,” which in the twentieth century came to be gathered in the noun anonymity.24 Benjamin Disraeli made an allegory out of the same material that Blake had used in “The Chimney Sweeper” in his novel, Sybil, or The Two Nations.25 It tells an inserted tale of a child the narrator refers to as “the nameless one,” because when his mother went to work in “her factory” soon after his birth, she abandoned him without name, “baptismal or patrimonial.” After surviving most of his infancy on the street, one lucky day when he was five years old he joined a “crowd of men, women, and children” entering a factory, where he was given work because “A child was wanting in the Wadding Hole” (presumably a space only a small body could fit into). For that piece of good fortune “the nameless one” was given “even a salary, more than that, a name; for as he had none, he was christened on the spot—Devilsdust.” Like Blake’s chimney sweep, he was named by society for the filthy work that marked his only identity. The root meaning of anonymity, “without name,” becomes in this tale an allegorical emblem of the cultural horrors Blake denounced more than half a century earlier. In , Carlyle characterized what the title of his essay called “The Present Time” by images like Wordsworth’s of the streets of London where the human beings who made up its vast population were reduced to a grotesque mass of synecdoches: the tramp of its million feet is on all streets and all thoroughfares, the sound of its bewildered thousandfold voice is in all writings and speakings, in all thinkings and modes and activities of men.26
In , Arnold Toynbee, drawing on the structure of urban images already in place in literature, epitomized modern experience by another familiar synecdoche of unknown faces crowded in proximity: “In the new cities . . . the old warm attachments, born of ancient, local contiguity and personal intercourse, vanished in the fierce contest for wealth among thousands who had never seen each other’s faces before.”27 The choice of detail here conveys the
"/0/:.*5:
same experience as in Wordsworth’s many glimpses of strange faces, the sense of seeing only a sea of them above the shoulders of a crowd. Later novelists, both English and American, borrowed increasingly from this tradition. In in Demos, George Gissing described the “uniformity” and “squalor” of what he sarcastically called a “neighbourhood” in eerie depersonifications like Wordsworth’s “dead walls”: “each of these dead-faced houses,” “each separate blind window.”28 The streets of Chicago that Upton Sinclair wrote about in The Jungle of , where “the people who swarmed in them were as busy as ants—all hurrying breathlessly, never stopping to look at anything, nor at each other” could be the “ant-hill . . . / Of a too busy world” that Wordsworth wandered in a hundred years before, where “no one looks about, nothing is listened to.”29 In Our America of by Waldo Frank, the New York streets, “choked sluices” that “the brackish human flow pours through,” are squalid versions of Wordsworth’s “endless stream” pouring through London’s “overflowing streets” like a human “tide.”30 Lewis Mumford, in his widely read book of , The Culture of Cities, treated historically what Blake prophesied as a spreading plague that (in “London”) “blights” human lives. Mumford located it as a “blight that had its origin in England’s dark Satanic mills, as William Blake called them,” and called his own chapter on the spread of the metropolis in the century following Blake’s prophecy “A Brief Outline of Hell.”31 Whole poems are structured by this cumulative vocabulary of urban images in the work of two late Victorian poets, James Thomson and John Davidson. In their different choices among ways of using examples offered by Blake and Wordsworth to reflect their own senses of late nineteenth-century experience, they show what a variety of expressive possibilities this vocabulary offered. Thomson imitated the simple brevity of Blake’s lyric poems in one of his own, dated . By titling it “William Blake,” he made the nameless “He” in the first word of the poem a poet who “came to London town,” wandering lonely in a lonely crowd. Thomson accomplished this by using images that, because of their familiarity, had and still have generalizing power to make both the poet and the city into archetypes: There were thousands and thousands of human kind In this desert of brick and stone: But some were deaf and some were blind, And he was there alone.32
"/0/:.*5:
Thomson’s nocturnal vision titled The City of Dreadful Night was first published serially in in four numbers of the National Reformer, signed “B. V.” (initials of his nom de plume “Bysshe Vanolis,” a coded tribute to Shelley and Novalis). In it Thomson expanded on Wordsworth’s glimpses of London inhabited by ghostly figures that signaled from another world: “shapes” and “spectres,” “phantoms,” ranging among “mansions dark and still as tombs”— like Wordsworth’s “Courts / Gloomy as coffins”—or beings with “worn faces that look deaf and blind”; “Each man wrapt in his own doom”; whose “lonely sounding feet” and “bodiless voices” echoed in that “city of so lonely streets” and “weary roads”; where “one may count up every face he meets”; strangers who “speak to one another seldom”; who exist merely as “isolated units,” only the last a phrase belonging to Thomson’s generation that would not be found in The Prelude.33 John Davidson wrote of his London experience in poems written in the s and first decade of the s, for instance, in one with the title “London Bridge,” which actually refers to the railway station of that name but inevitably, and not altogether inappropriately, makes us think of the nursery rhyme with its lighthearted chant predicting doom to the city. The vision of London in this poem expresses a sense of isolation and especially of loss of identity in a rushing torrent of humanity, of inundation, death by drowning: in “London’s sea, immense / And turbulent, a brimming human flood”; “a brimming tide of mind / As well as blood”; “this human tide, / As callous as the glaciers that glide / A foot a day, but as a torrent swift.” Here immersion in the inexorable flood of humanity means the loss of everything in life that is “Distinctly personal, innate or earned,” all swept away “In the dull, rapid passage” flowing “from the station to the street.” Not even the myriad faces of strangers but only “the common face” can be seen above the “urban crowds” that represent the modern experience of “London Bridge.”34 This richly stocked collection of details, variously used in writings like these of many kinds, was from them absorbed into the forming of assumptions that became common by the early twentieth century, so that their pattern grew to be more than a description. Almost as firmly as the city itself, the constellation of imagery associated with it became an existential embodiment of a modern condition and state of consciousness. This linguistic event heralded the institution of anonymity as the code word for a cultural motif. The clearest evidences of the pervasiveness and depth of its assimilation are poems that count on readers to bring the assumptions concentrated in
"/0/:.*5:
this motif into play in their responses even to the smallest coded signals. A highly condensed example is a line from “The Lake Isle of Innisfree” of , the poem of Yeats’s best known in the early twentieth century: “While I stand on the roadway, or on the pavements grey.”35 Even out of context, the conventionally loaded words in combination make their points. Pavements are found in cities, and their material is by its nature “grey,” but in a poem of this period these details go far beyond what the words refer to, to suggest that the solitary speaker who says, or chants, the lines is lonely and estranged from his surroundings, which are isolating and alienating because colorless and standardized, dulling to personal feelings, indifferent to his individual self. Or think of Pound’s most famous early poem, of : In a Station of the Metro The apparition of these faces in the crowd; Petals on a wet black bough.36
The title line tells us that this is another poem with an urban setting, preparing us for certain quite specific responses to what follows, which can be intensely concentrated because its few signals are highly charged. A vision of the modern city that catches a glimpse of beauty usually found in natural things follows a familiar pattern—compare Wordsworth’s vision of London in “Compos’d upon Westminster Bridge, September , ”—and the perception that though this loveliness makes a few human beings rarely precious, it is as fragile and brief as flower petals has been a cherished truth in English poetry since at least the sixteenth century. Encoded in the phrase “faces in the crowd” is a more distinctively modern thought: that the actuality of the urban condition dulls even the owners of “these faces” into living their lives as anonymously—impersonally—and unconsciously as nonhuman creatures do. The poem itself, a sentence fragment that excludes the first person, is eerily displaced from the scene it envisions. To show one more poet exploiting the familiarity of the cultural motif expressed in this constellation of urban images, Frost—quite uncharacteristically—made a poem out of that material in , “Acquainted with the Night.” A catalogue of phrases—“the farthest city light,” “the saddest city lane,” “the sound of feet,” “an interrupted cry / . . . from another street”—associates the feelings of other urban wanderers—typically nineteenth-century poets—with the solitary “I” who says, almost chants, the lines.37 Those feelings are power-
"/0/:.*5:
fully expressed at the same time that Frost—quite characteristically—plays with their perhaps too ready familiarity and resonance, their formulaic quality, by a number of devices, including the repetitions in the lines of the poem itself (in particular the five in a row beginning with I ). The feelings of loneliness and estrangement this city wanderer tells about might be at once an inescapable modern condition and a self-induced mood. If the associations of the images were not familiar—and their familiarity is what the poem is partly about—their doubleness would be lost on the reader. A hypothesis of this discussion is that the specially wide appeal of these particular poems, their status as anthology pieces, has a great deal to do with their being expressions of the cultural motif of anonymity.
"OPOZNJUZ".PEFSOJTU1BSBEPY T. S. Eliot, icon and public voice of modernism, acknowledged debts to both Thomson’s “The City of Dreadful Night” and Davidson’s London verses, poems he read “between the ages of sixteen and twenty,” pairing them obviously because they shared what in writing about Davidson he called “a great theme.”38 In Davidson’s work in particular, Eliot remembered, he “found inspiration in the content of the poem, and in the complete fitness of content and idiom; for I also had a good many dingy urban images to reveal.” Beyond that, he credited Davidson with preparing him for the work of the French symbolists, who became his next models. Earlier, Eliot had given a slightly different account of discovering the material he needed: I think that from Baudelaire I learned first, a precedent for the poetical possibilities, never developed by any poet writing in my own language, of the more sordid aspects of the modern metropolis. . . . From him, as from Laforgue, I learned that the sort of material that I had, the sort of experience that an adolescent had had, in an industrial city in America, could be the material for poetry; and that the source of new poetry might be found in what had been regarded hitherto as the impossible, the sterile, the intractably unpoetic.39
The substitution of “an adolescent” for “I” is a thin layering over the confession that Eliot had personally experienced the sense of anonymity associated with the modern city.
"/0/:.*5:
To epitomize Baudelaire’s “significance” for him, he then quoted two lines (from a poem in Les fleurs du mal titled “Les sept vieillards”) that he thought summed it up: Fourmillante cité, cité pleine de rêves, Où le spectre en plein jour raccroche le passant! (City teeming like an ant-hill, city full of dreams, / where the spectre in broad daylight accosts the passerby.)
“I knew what that meant, because I had lived it before I knew that I wanted to turn it into verse on my own account.” Again “I had lived it” here is an uncharacteristic lapse into a personal mode. In turning the metropolis into “new poetry,” Eliot adapted this deeply embedded tradition of images to his own representation of modern anonymity, just as he said Baudelaire had done: “Inevitably the offspring of romanticism . . . he could, like anyone else, only work with the materials which were there . . . he must express with individual differences the general state of mind—not as a duty, but simply because he cannot help participating in it.”40 Eliot began to experiment with the possibilities of this ready material in his early work, beginning with the poems in the manuscript originally titled “Inventions of the March Hare,” where we find familiar images like these in the first lines of “Silence”: Along the city streets It is still high tide, Yet the garrulous waves of life Shrink and divide . . .41
Although he left poems in this manuscript unpublished, he did not soon leave their preoccupations behind. In “The Love Song of J. Alfred Prufrock,” which comes first in Prufrock and Other Observations (), the opening lines are an invitation to walk “through certain half-deserted streets”—no more deprived of vitality than Wordsworth’s “half-frequented” London streets— into a state of consciousness figured by the city with all its echoes. These lines strike the dominant note not only of this poem but of this first volume set mainly in the city, and of the rest of Eliot’s poems of his modernist period. Their culmination is in the much-discussed passage of The Waste Land at the end of “The Burial of the Dead” (to be quoted in full later), which begins with a prophetic invocation—“Unreal City”—and ends with a frag-
"/0/:.*5:
ment quoted from a poem by Baudelaire: “You! hypocrite lecteur!—mon semblable,—mon frère!” As this discussion has already stated and, it is hoped, demonstrated, the accumulated language about the modern condition of anonymity has so permeated our literature that it is pointless to trace to particular sources lines like Eliot’s in this passage, for instance, “A crowd flowed over London Bridge, so many,” or “And each man fixed his eyes before his feet.” Even so, Eliot himself gave multiple directions about his sources, as in the French quotation just above, in the later-canceled notes to The Waste Land, and in many passages like those already cited from his prose about the models who inspired his urban poems. What seems pointed in the context reconstructed in this discussion is Eliot’s persistent silence about Wordsworth’s extremely influential role in making the city a cultural epitome. It is inconceivable that Eliot did not know The Prelude: his own choice of the musical term “Preludes” for the title of a group of cityscapes is quite likely intended to set up a contrast between the impersonal purity of a poem made in the image of a musical composition and Wordsworth’s autobiographical narrative, about which the author himself said it was “unprecedented in literary history that a man should talk so much about himself.”42 It seems curious, too, that in paying so much attention to the lines of Baudelaire’s beginning “Fourmillante cité” that Eliot says were the most important inspiration for his own city poems, and that he quoted in the note to the line “Unreal City,” he did not hear their sympathetic likeness to Wordsworth’s invocation of his teeming city as an “ant-hill,” and to his many figurings of its inhabitants as ghostly. Of course there is no way to know to what degree a poet is deliberately suppressing, or is blocked from recognizing, or is merely forgetting the presence of another poet. Even so, Eliot has shown us both in his critical prose and in his early poems reasons why he never acknowledged Wordsworth even as a distant ancestor in the fashioning of urban images. In Eliot, in his since most famous essay “Tradition and the Individual Talent,” announced his “Impersonal theory of poetry”: in sum, that the “progress of an artist is . . . a continual extinction of personality,” achieved only through the immense labor of acquiring a true sense of “tradition.”43 He epitomized the enemy position by quoting Wordsworth’s most famous statement, that poetry is “emotion recollected in tranquility,” without ever mentioning who wrote it. Of course Eliot’s readers would be expected to
"/0/:.*5:
know the source, but the suppression of Wordsworth’s name seems in keeping with Eliot’s well-known program of dethroning Wordsworth, in which one of his preferred tactics was if possible to ignore him, to leave him out of consideration as a way of suggesting his unimportance to Eliot’s generation of poets. For the purposes of the argument here, Eliot’s refusal to name Wordsworth as the author of the famous “formula” (Eliot’s deliberately impersonal term) for poetry of “ ‘personality’ ” (Eliot put that word in quotation marks) can serve as an analogy for his silence about Wordsworth’s role in forming the tradition of urban poetry. Eliot was caught in what I am calling a distinctively modernist paradox. The materials he found best suited for the “new poetry” were “dingy urban images,” which could express the horror of modern cultural anonymity. The way of writing best suited for the “new poetry” promoted an escape for the poet into aesthetic anonymity. It would be an untenable position to recognize Wordsworth as an early discoverer and explorer of the territory staked out as uniquely theirs by poets for whom Wordsworth’s influential presence was the chief threat to their impersonal poetry. So it was necessary for Eliot to locate the origins of the urban imagery he needed no earlier than Baudelaire, whom Eliot called “the greatest exemplar in modern poetry in any language.” 44 This pronouncement stakes out another paradoxical position, since Les fleurs du mal was published in , only seven years later than The Prelude. Among the poems in Eliot’s first book, printed the same year as “Tradition and the Individual Talent,” the one that makes fullest use of traditional urban imagery is “Preludes.” It is the most useful choice of example here also because it is the one that performs the most visible experiments with the techniques to promote impersonality that Eliot is best known for in The Waste Land. “Preludes” suspends “grimy scraps” of city living in four separately numbered parts, each drawing material from the “thousand sordid images” and familiar synecdoches of other poems similarly set in “the sawdust-trampled street / With all its muddy feet” where the poet wanders. Although the imagery holds the parts together, the grammar dislocates them in this poem by all but eliminating the first person, who in the tradition of urban poetry is the lonely and estranged poet. Here it is not the poet who has a “vision of the street” but an initially unidentified You mentioned in the first word of part III, who turns out to be a woman with curlers in her hair looking through
"/0/:.*5:
the “shutters” of her window, a detail tied to the “broken blinds” and “dingy shades” of parts I and II. In those sections, and again in the last lines of IV, the only pronouns are generalized your or one. The phrase His soul, which opens part IV, is never located, unless it refers to whoever called the woman You: the poet displacing himself with a second-person pronoun?45 The first-person nominative intrudes only in four isolated lines just before the end of the poem, sounding very nearly as if spoken by the familiar compound ghost of a nineteenth-century poet: I am moved by fancies that are curled Around these images, and cling; The notion of some infinitely gentle Infinitely suffering thing.
These lines are close in manner to some parodic language Eliot used in an essay about Wordsworth: “he went on droning on the still sad music of infirmity to the verge of the grave.”46 In “Preludes” the droning of this I is erased by the next and coarsest line of the poem: “Wipe your hand across your mouth, and laugh.” (Said by an understudy for Eliot exorcising that ghost?) These and other techniques of dislocation and displacement are instruments of aesthetic anonymity finely tuned to express the great theme of cultural anonymity in Eliot’s early writings, above all in The Waste Land, the archetypal modernist vision of the city. They operate through the separated parts of the poem, and in sections within parts, as in the closing passage of “The Burial of the Dead”: Unreal City, Under the brown fog of a winter dawn, A crowd flowed over London Bridge, so many, I had not thought death had undone so many. Sighs, short and infrequent, were exhaled, And each man fixed his eyes before his feet. Flowed up the hill and down King William Street, To where Saint Mary Woolnoth kept the hours With a dead sound on the final stroke of nine. There I saw one I knew, and stopped him, crying: “Stetson! “You who were with me in the ships of Mylae! “That corpse you planted last year in your garden, “Has it begun to sprout? Will it bloom this year?
"/0/:.*5:
“Or has the sudden frost disturbed its bed? “O keep the Dog far hence, that’s friend to men, “Or with his nails he’ll dig it up again! “You! hypocrite lecteur!—mon semblable—mon frère!”47
Eliot revised the first line of this passage, causing “Unreal City” to float unlocated, by cutting out what originally followed it: “I have sometimes seen and see.”48 If this had been left in place, it would have worked like Wordsworth’s use of the first person in his invocation of the “monstrous ant-hill”— “Before me flow, / Thou endless stream of men”—to ground the vision in the personal experience of the poet wandering lonely in a modern city. Eliot allowed the first person to enter in line but, as he told his first readers in a note in case they missed the allusion, the line is not spoken in Eliot’s own voice telling of his personal experience, but by or through Dante describing another vision of hell. The eight lines around it are pronounless, sometimes incomplete sentences, in the passive voice. They happen grammatically, unattached like the ghostly synecdoche of the Sighs, as if without any located person saying or doing what happens in them. Then the passage tilts into a shockingly different mode, an absurdly dramatized speech by the stereotype of a suburban commuter who turns in the end into a poet—without knowing it?—by bursting into French—borrowed, the note says, from Baudelaire—in the last line. It addresses the reader as if from the grave, in the manner of an epitaph poem, much as in Eliot’s other favorite lines from Baudelaire the spectre accosts the passerby. In the tradition of poetry about the modern metropolis, the author’s voice expresses personal feelings of dislocation and displacement, as well as ascribes them to the culture the city embodies. In Eliot’s modernist poems, where structure and grammar exclude a continuous first-person speaker, it is the poem itself that is dislocated, its center of consciousness displaced. These techniques act out the theory of impersonal poetry proposed in “Tradition and the Individual Talent”: that “the poet has, not a ‘personality’ to express, but a particular medium . . . in which impressions and experiences combine in peculiar and unexpected ways.”49 Eliot’s formal techniques are the medium suited to working out the impersonal aesthetic of anonymity, in which Wordsworth was cast as the enemy. And they are the medium best fitted to represent the great theme of cultural anonymity, which to some unmeasurable but large degree was invented by Wordsworth, who in Eliot’s scheme invented the poetry of personality. It is the penultimate paradox of modernism that the
"/0/:.*5:
distinctively “new poetry” of its most famous practitioner belonged centrally, however inadvertently, to this Wordsworthian tradition. The ultimate paradox is the amalgam in Eliot’s modernist poems of two distinct twentieth-century meanings of anonymity, both packed with historically accumulated assumptions that give those poems their great and exemplary power: anonymity as an aesthetic ideal and as a cultural condition.
/05&4
*/530%6$5*0/
. Quoted in Lawrance Thompson, Robert Frost: The Early Years – (New York: Holt, Rinehart and Winston, ), . $ ) " 1 5 & 3
. W. H. Auden, “Don Juan,” in The Dyer’s Hand (New York: Random House, ), . . T. V. F Brogan, English Versification, – (Baltimore: Johns Hopkins University Press, ), . . Jurij Lotman, The Structure of the Artistic Text, trans. Ronald Vroom (Ann Arbor: University of Michigan Press, ), . . Sir Philip Sidney, “An Apologie for Poetrie,” in Elizabethan Critical Essays, ed. G. G. Smith (London: Oxford University Press, ), I:. . W. K. Wimsatt, “One Relation of Rhyme to Reason,” in The Verbal Icon (Lexington: University Press of Kentucky, ), . In this classic article Wimsatt distinguished Pope’s combinations of rhyme words from those of Chaucer, which belonged to a much earlier phase of English not yet fully emerged from inflectional grammar. Wimsatt’s example of Chaucerian rhyming is a sequence from the Knight’s portrait: he, worthynesse, ded, wildernesse, kyndenesse, where the esse endings are close to what Aristotle called homeoteleuton (identical or similar inflectional case endings). In contrast, Wimsatt characterized Pope’s rhyme pairs as more widely differing in parts of speech or in function of the same parts of speech, in lexical meaning, and in syllabic structure, differences heightened by the reliance of Pope’s couplets on parallel structure and epigrammatic point. . George Puttenham, The Arte of English Poesie, ed. Gladys Willcock and Alice Walker (Cambridge: Cambridge University Press, ), . . Ibid., –. . Ibid. My discussion here copies the model of Susanne Woods, Natural Emphasis: English Versification from Chaucer to Dryden (San Merino, Calif.: Huntington Library, ), , in following “the modern terminology” of stress to mean an inherent feature of the language and accent to mean the syllable emphasized metrically. . Thomas Campion, “Observations in the Art of English Poesie” and George Gascoigne, “Certaine Notes of Instruction,” in Elizabethan Critical Essays, II:, I:.
/05&450$)"15&3
. Gascoigne, “Certaine Notes of Instruction,” . . T. Walter Herbert, “The Grammar of Rhymes,” Sewanee Review (): . For discussion of the sixteenth-century conceptions of the noun-substantive and the nounadjective, see Anne Ferry, The Art of Naming (Chicago: University of Chicago Press, ), –. . Peter Levens, Manipulus Vocabulorum (Chicago: University of Chicago Press, ). . Samuel Daniel, “A Defence of Ryme,” in Elizabethan Critical Essays, II:. . Ibid. . Roman Jakobson, “Linguistics and Poetics,” in Language in Literature, ed. Krystyna Pomorska and Stephen Rudy (Cambridge, Mass.: Harvard University Press, ). . Percy Adams, Graces of Harmony: Alliteration, Assonance, and Consonance in Eighteenth-Century British Poetry (Athens: University of Georgia Press, ), . . G. S. Fraser, Metre, Rhyme and Free Verse (London: Methuen, ), . . George Gascoigne, The Poesies, ed. John Cunliffe (Cambridge: Cambridge University Press, ), . . Hugh Kenner, “Pope’s Reasonable Rhymes,” ELH (): . . George Herbert, “A true Hymne,” in The Works of George Herbert, ed. F. E. Hutchinson (Oxford: Clarendon Press, ), . In “Jordan II” Herbert uses the verb to copy out instead of to write to describe the act of making a poem. . Puttenham, Arte of English Poesie, . . Edward Bysshe, “A Dictionary of Rhymes, Preface,” in The Art of English Poetry (Menston, U.K.: Scolar Press, ), . . Ibid., . . Ibid., . . Alexander Pope, “An Essay on Criticism,” in Pastoral Poetry and an Essay on Criticism, ed. E. Audra and Aubrey Williams (London: Methuen, ), –. Wallace Stevens confessed that once “I made ‘breeze’ rhyme with ‘trees,’ and have never forgiven myself. It is a correct rhyme, of course,—but unpardonably ‘expected.’ ” . Alexander Pope, Windsor-Forest, –. . Bysshe, “Preface,” in Art of English Poetry. . Pope, “Essay on Criticism,” . . Ibid., . . See Edward Young, Conjectures on Original Composition (Leeds, U.K.: Scolar Press, ), : “Blank is a term of distinction; what we mean by blank verse is verse unfallen, uncurst; verse reclaim’d, reinthroned in the true language of the gods.” . Clive Scott, The Riches of Rhyme (Oxford: Clarendon Press, ), . . Randolph Quirk, Sidney Greenbaum, Geoffrey Leech, and Jan Svartvik, A Grammar of Contemporary English, rd ed. (London: Longman, ), . All mentions of grammatical terms and definitions refer to this invaluable book. . Marianne Moore, “Memory’s Immortal Year,” in The Complete Prose of Marianne Moore, ed. Patricia Willis (New York: Viking, ), .
/05&450$)"15&3
. Percy Bysshe Shelley, “Evening: Ponte al Mare, Pisa,” in The Complete Works of Percy Bysshe Shelley, ed. Thomas Hutchinson (London: Oxford University Press, ), . . Shelley, “To Jane: The Recollection,” in Complete Works, –. . Gerard Manley Hopkins, The Sermons and Devotional Writings of Gerard Manley Hopkins, ed. Christopher Devlin (London: Oxford University Press, ), ; The Journals and Papers of Gerard Manley Hopkins, ed. Humphrey House and Graham Storey (London: Oxford University Press, ), . . T. S. Eliot, “Introduction,” in Selected Poems (New York: Macmillan, ), xii. I have chosen to quote Moore’s poems exclusively from this edition because in it works of her modernist period were introduced to a larger public, and because she had not yet begun her radical program of revising them. The two poems quoted and discussed extensively in this essay, “The Fish” and “Those Various Scalpels,” are on pages –, –. . Margaret Holley, The Poetry of Marianne Moore (Cambridge: Cambridge University Press, ), . . William Carlos Williams, “Marianne Moore,” in Imaginations (New York: New Directions, ), , . . Williams, “Spring and All,” in Imaginations, , . . Williams, “Marianne Moore,” in Imaginations, . . W. H. Auden, “Nevertheless,” New York Times Book Review, Oct. , , ; “Marianne Moore,” in The Dyer’s Hand, . . Wallace Stevens, Letters of Wallace Stevens (Berkeley: University of California Press, ), , . . Eliot, in Selected Poems, xiv, ix, x. . Ibid., x, xi. . Wallace Stevens, “A Poet That Matters,” in Opus Posthumous, ed. Samuel French Morse (New York: Knopf, ), . . Eliot, in Selected Poems, x. . Ibid., xii. . Ibid. . Ibid. . Bishop, “Efforts of Affection: A Memoir of Marianne Moore,” in Elizabeth Bishop: The Collected Prose (New York: Farrar, Straus and Giroux, ), –. . Eliot, in Selected Poems, xiii. . Ibid. . Charles Tomlinson, “Introduction,” in Marianne Moore: A Collection of Critical Essays (Englewood Cliffs, N.J.: Prentice Hall, ), . . Moore, “Subject, Verb, Predicate,” in Complete Prose of Marianne Moore, . . Holley, Poetry of Marianne Moore, . . Linda Leavell, Marianne Moore and the Visual Arts (Baton Rouge: Louisiana State University Press, ), . . Malcolm Bowie, Mallarmé and the Art of Being Difficult (Cambridge: Cambridge University Press, ), .
/05&450$)"15&3
. Wallace Stevens, “The Man on the Dump,” in The Collected Poems of Wallace Stevens (New York: Knopf, ), . . Williams, “Marianne Moore,” in Imaginations, . . Sylvia Adamson, “Literary Language,” in The Cambridge History of the English Language, ed. Suzanne Romaine (Cambridge: Cambridge University Press, ), IV:. . Bishop, “Efforts of Affection,” , . . Auden, “Nevertheless,” in The Dyer’s Hand, ; Moore, “W. H. Auden,” in Complete Prose of Marianne Moore, . . Richard Wilbur, Conversations with Richard Wilbur, ed. William Butts (Jackson: University of Mississippi Press, ), . . Henry Lanz, The Physical Basis of Rime (Stanford: Stanford University Press, ), . . Thom Gunn, “Memoirs of the World,” in Collected Poems (London: Faber and Faber, ), . . Scott, Riches of Rhyme, . . Richard Wilbur, The Catbird’s Song: Prose Pieces – (New York: Harcourt Brace, ), . . Marie Borroff, Language and the Poet: Verbal Artistry in Frost, Stevens, and Moore (Chicago: University of Chicago Press, ), . $ ) " 1 5 & 3
. All quotations from Shakespeare’s sonnets are taken from Shakespeare’s Sonnets, ed. Stephen Booth (New Haven, Conn.: Yale University Press, ). They are located by sonnet number in the text. . Stephen Booth, An Essay on Shakespeare’s Sonnets (New Haven, Conn.: Yale University Press, ). . William Ringler, ed., The Poems of Sir Philip Sidney (Oxford: Oxford University Press, ), lviii. All quotations from Astrophil and Stella are taken from this edition and are located by sonnet number in the text. . Ezra Pound, The Letters of Ezra Pound –, ed. D. D. Paige (New York: Harcourt Brace, ), . . Tottel’s Miscellany, ed. Hyder Rollins, rev. ed. (Cambridge, Mass.: Harvard University Press, ), I:. All quotations from the Miscellany are taken from this edition. . George Gascoigne, Certayne notes of Instruction, in The Poesies, ed. John Cunliffe (Cambridge: Cambridge University Press, ), –. . Sir Philip Sidney, An Apologie for Poetrie, in Elizabethan Critical Essays, ed. G. G. Smith (Oxford: Oxford University Press, ), I:. . Ibid., –. . Ibid., . . George Puttenham, The Arte of English Poesie, ed. Gladys Willcock and Alice Walker (Cambridge: Cambridge University Press, ), . . Sidney, An Apologie for Poetrie, .
/05&450$)"15&3
. Ibid. . The publication of Sidney’s sequence was so important a literary event that Shakespeare would not have been likely to miss it. My more particular argument is that in the version of Sonnet “scarfe of clowds” replaces “maske of clowds,” which Shakespeare adapted from the version as “cloude hath mask’d” in his Sonnet . See Anne Ferry, The “Inward” Language (Chicago: University of Chicago Press, ), . . Ibid., –, –. . John Thompson, The Founding of English Metre (London: Routledge and Kegan Paul, ), , –. . Gascoigne, Certayne notes of Instruction, . . William Webbe, A Discourse of English Poetrie, in Elizabethan Critical Essays, I:. . Invaluable assistance came from Herbert Donow, A Concordance of the Sonnet Sequences of Daniel, Drayton, Shakespeare, Sidney, and Spenser (Carbondale: Southern Illinois University Press, ). . The Complete Works of William Shakespeare, ed. George Lyman Kittredge (Boston: Ginn, ), . $ ) " 1 5 & 3
. Accompanying facsimiles of these two manuscripts are introductory descriptions of them by the editors: Amy Charles, The Williams Manuscript of George Herbert’s Poems (Delmar, N.Y.: Scholars Facsimiles and Reprints, ); Amy Charles and Mario Di Cesare, The Bodleian Manuscript of George Herbert’s Poems (Delmar, N.Y.: Scholars Facsimiles and Reprints, ). See also F. E. Hutchinson, “Introduction,” in The Works of George Herbert (Oxford: Oxford University Press, ), l–lvi. . Quotations from The Temple are all from the facsimile of the first edition: The Temple (Menston, U.K.: Scolar Press, ). References are given by the title of the poem in the text only. . John Hollander, “ ‘Haddocks Eyes’: A Note on the Theory of Titles,” in Vision and Resonance (New Haven, Conn.: Yale University Press, ), . For the most extended discussion of Herbert’s titles, see Mary Ellen Rickey, Utmost Art (Lexington: University Press of Kentucky, ), –, –. Examples of briefer comments are Amy Charles, “The Williams Manuscript and The Temple,” in Essential Articles for the Study of George Herbert’s Poetry, ed. John Roberts (Hamden, Conn.: Archon Books, ), –; John Mulder, The Temple of the Mind (New York: Pegasus, ), –. . Hollander, “ ‘Haddock’s Eyes,’” –. . Quotations from the first edition are taken from the facsimile: John Donne, Poems (Menston, U.K.: Scolar Press, ). For relevant discussions of surviving manuscripts of Donne’s poems, see the textual introductions to Helen Gardner’s The Divine Poems and The Elegies and the Songs and Sonnets (Oxford: Oxford University Press, ), lvii–lxxxii, lxiii–xcv; Herbert Grierson, “The Text and Canon of Donne’s Poems,” in The Poems of John Donne (London: Oxford University Press, ), II:xxv– cxxiv.
/05&450$)"15&3
. All the love poems in this edition under the allusive but generically misleading heading of “Songs and Sonets” were given individual titles, including “The Message,” Poems, By J. D. (London, ), . . According to Walton, “A Hymne to God the Father” was “often sung to the Organ by Choristers of St Paul’s Church,” quoted by A. J. Smith, “Introduction,” in John Donne: The Critical Heritage (London: Routledge and Kegan Paul, ), . . For discussions of the transcription and circulation of Donne’s manuscripts, see Arthur Marotti, John Donne: Coterie Poet (Madison: University of Wisconsin Press, ); Alan MacCall, “The Circulation of Donne’s Poems in Manuscript,” in John Donne Essays in Celebration, ed. A. J. Smith (London: Methuen, ), –; Ted-Larry Pebworth, “John Donne, Coterie Poetry, and the Text as Performance,” Studies in English Literature (): –. . “Upon the parting from his Mistresse” and “To his Love upon his departure from her” are among manuscript titles cited by Gardner, The Elegies and the Songs and Sonnets, . . Isaak Walton, The Life of Mr. George Herbert (London, ), . . On English religious sonnets of the period, see William Stull, “Sacred Sonnets in Three Styles,” Studies in Philology (): –; “ ‘Why Are Not Sonnets Made of Thee?’ A New Context for the ‘Holy Sonnets’ of Donne, Herbert, and Milton,” Modern Philology (): –. . For a description of published collections by Southwell, see James McDonald and Nancy Brown, “Textual Introduction,” in The Poems of Robert Southwell, S. J. (Oxford: Clarendon Press, ), lv–lxxvii. . Ibid., . . Henry Hawkins, “The Preface to the Reader,” in Partheneia Sacra (Brookfield, Vt.: Scolar Press, ). . Charles and Cesare, The Bodleian Manuscript, xii. . Charles, The Williams Manuscript, xi. . Smith, John Donne: The Critical Heritage, . . “Epistle to Edward Lord Herbert from his Uncle, Henry Herbert,” in Occasional Verses of Edward Lord Herbert . . . Deceased in August, (London, ). . Bruce Smith, “Ben Jonson’s Epigrammes: Portrait Gallery, Theater, Commonwealth,” Studies in English Literature (): . . Rickey, Utmost Art, . . For a bibliography of such volumes, see Rosemary Freeman, English Emblem Books (London: Chatto and Windus, ). Herbert’s relation to emblem poetry is discussed by Charles Huttar, “Herbert and Emblematic Tradition,” in Like Season’d Timber, ed. Edmund Miller and Robert Di Yanni (New York: P. Lang, ), –. . Francis Thynne, Emblemes and Epigrames, ed. F. J. Furnival (London: Early English Text Society, ). . Robert Hayman, Qvodlibets (London, ), bound with his translation of Certaine Epigrams Ovt of the First Fovre Bookes of the Excellent Epigrammatist, Master Iohn Ovven (London, ).
/05&450$)"15&3
. These examples are from the collection of references to Herbert compiled by Robert Ray, “The Herbert Allusion Book,” Studies in Philology (): , , , , , , . . Ibid., . . Herbert’s Latin poems, which spanned much of his life, are discussed by Robert Wickenheiser, “Herbert and the Epigrammatic Tradition,” George Herbert Journal (): –. . Donne, Poems, . . An example in Jonson’s Epigrammes is “To his Lady, Then Mrs. Cary,” Ben Jonson, ed. C. Hereford, Percy Simpson, and Evelyn Simpson (Oxford: Clarendon Press, ), VIII:. Herrick used this style for titles of many individual poems like “His Confession,” “His Prayer for Absolution” on the opening page of his collection printed with the title page His Noble Numbers: Or, His Pious Pieces, Wherein (amongst other things) he sings the Birth of his Christ: and sighes for his Saviours suffering on the Crosse (London, ). . Some examples are “Deniall,” “Home,” the second “Jordan,” “The Collar,” “Dulnesse,” “The Flower.” . John Davies, The Mvses Sacrifice (London, ), , . . The most detailed study of Herbert’s revisions is by Janis Lull, The Poem in Time (Newark: University of Delaware Press, ). Revisions of “The Publican” are discussed on page . . Davies, The Mvses Sacrifice, . . A succinct account of this tradition is given by R. R. Bolgar, The Classical Heritage and Its Beneficiaries (Cambridge: Cambridge University Press, ), –. . The topics I have quoted are from A Treatise of Morall Philosophy (London, ?). Baldwin’s collection is described by Elizabeth Pomeroy, The Elizabethan Miscellanies (Berkeley: University of California Press, ), , . . The edition quoted is Banket of Sapience (London, ). In The Image of Gouernance (London, ), Elyot wrote that his commonplace book “in litle rome shewith out of holy scripture many wise sentences,” quoted by Stanford Lehmberg, Sir Thomas Elyot (Austin: University of Texas Press, ), . . George Herbert, “A Priest to the Temple: Or, The Country Parson” in Herbert’s Remains. Or, Sundry Pieces Of that sweet Singer of the Temple Mr. George Herbert (London, ), . . Ibid., . . These passages have been quoted as descriptions of structures in The Temple, for instance, the structure of individual poems by Chana Bloch, Spelling the Word (Berkeley: University of California Press, ), ; and the structuring of the volume, for example, by Sybil Severance, “Numerological Structures in The Temple: ‘Too Rich to Clothe The Sunne,’” ed. Claude Summers and Ted-Larry Pebworth (Pittsburgh: University of Pittsburgh Press, ), . . Michael Drayton, “The Spirituall Songes and Holy Hymns Contained in This Book,” in The Harmonie of the Church (London: Percy Society, ); Thomas Campion,
/05&450$)"15&3
“A Table of halfe the Songs contayned in this Booke,” A Booke of Ayres, in The Works of Thomas Campion, ed. Walter Davis (Garden City, N.Y.: Doubleday, ); “A Table to Find Out Any Sonnet Herein Alphabetically,” The Poems of Barnabe Barnes, ed. Alexander Grosart (London, ); Henry Lok, “A Table Of Feeling Affections, being the third Centenarie of Sonets,” in Svndry Affectionate Sonets Of A Feeling Conscience (London, ). . Barnabas Oley, “A Prefatory View of the Life and Vertues of the Author and Excellencies of This Book,” in Herbert’s Remains. . Walton, Life, . . The concordances or harmonies are described by John Ferrar, “A Life of Nicholas Ferrar,” in The Ferrar Papers, ed. B. Blackstone (Cambridge: Cambridge University Press, ), –; the fullest discussion is by Stanley Stewart, “Herbert and the ‘Harmonies’ of Little Gidding,” Cithara (): –. . The full title of the edition of William Baldwin’s commonplace book is A treatise of Morall phylosophye, contayning the sayinges of the wise. “Copying” as a metaphor for “human action uncontaminated by pride” in Herbert’s poems is discussed by Lull, The Poem in Time, . . Helen Vendler, The Poetry of George Herbert (Cambridge, Mass.: Harvard University Press, ), . . For a summary of liturgical uses of the Psalms, see John Wall, Transformations of the Word (Athens: University of Georgia Press, ), –. Recitations of the Psalter at Little Gidding are described in John Ferrar, “A Life of Nicholas Ferrar,” . . The Sermons of John Donne, ed. George Potter and Evelyn Simpson (Berkeley: University of California Press, , ), VII:, II:; George Herbert, “A Priest to the Temple,” . For discussion of traditional ways of thinking about the Psalms as poetry, see Anne Prescott, “King David as ‘Right Poet’: Sidney and the Psalmist,” English Literary Renaissance (): –. . The Psalmes Of Dauid, Trvely Opened . . . To the VVich is Added A briefe Table, shewing wherevnto euery Psalme is particularly to be applied, trans. Anthonie Gilbie (London, ). . The Whole Boke of Psalmes (London, ). . “A Table,” The Temple (London, ). . Walton, Life, . Walton’s ways of shaping the deathbed speeches of Herbert and Donne (both reported to have recited Herbert’s poesie, also quoted in Ferrar’s preface to The Temple—“Lesse then the least of Gods mercies”) are discussed by David Novarr, The Making of Walton’s Lives (Ithaca, N.Y.: Cornell University Press, ), ; Robert Ray, “Herbert’s Words in Donne’s Mouth: Walton’s Account of Donne’s Death,” Modern Philology (): . . “Iohn Calvin to the Reader,” in The Psalmes of Dauid and others. With M. Iohn Caluins Commentaries (London, ). . George Wither, A Preparation to the Psalter (London, ), . . George Wither, The Psalmes of David (Neatherlands, ). . McDonald and Brown, Poems of Robert Southwell, ; Davies, The Mvses Sacrifice, .
/05&450$)"15&3
. Louis Martz, The Poetry of Meditation (New Haven, Conn.: Yale University Press, ), . For other discussions of The Temple in various relations to the Psalms, see Bloch and also Coburn Freer, Music for a King (Baltimore: Johns Hopkins University Press, ). Barbara Lewalski states the generally held view that “Herbert seems to have conceived his book of lyrics as a book of Christian psalms” especially because it records “the full range of spiritual emotions . . . representing thereby to man the anatomy of his own soul,” Protestant Poetics and the Seventeenth-Century Religious Lyric (Princeton: Princeton University Press, ), . This essay argues that Herbert had also a more particular conception of the Psalter as a model for his book of poems, distinguishing it from other collections of religious verse of the period. . George Herbert: The Critical Heritage, ed. C. A. Patrides (London: Routledge and Kegan Paul, ), . . “Table of Dates,” in Sir Thomas Wyatt: The Complete Poems, ed. R. A. Rebholz (New Haven, Conn.: Yale University Press, ), . . The Whole Boke of Psalmes. . It may have been in part a desire to forestall criticism of such a plan that prompted Walton to follow Herbert’s deathbed directions regarding his “little Book” with the comment: “Thus humbly did this humble man think of this excellent Book,” Life, . $ ) " 1 5 & 3
. Willard Van Orman Quine, Word and Object (Cambridge, Mass.: Technology Press of MIT, ), . . William Empson, The Structure of Complex Words (Norfolk, Conn.: New Directions, n.d.), . . William Wordsworth, The Prose Works of William Wordsworth, ed. W. J. B. Owen and Jane Smyser (Oxford: Clarendon Press, ), III:, . . Ibid., . . Geoffrey Hartman, The Unmediated Vision (New York: Harcourt, Brace and World, ), . . William Wordsworth, Wordsworth: Poetical Works, ed. Thomas Hutchinson, rev. Ernest de Selincourt (Oxford: Oxford University Press, ), –. . Ibid., –. . Jonathan Culler, The Pursuit of Signs (Ithaca, N.Y.: Cornell University Press, ), –. . Wallace Stevens, The Collected Poems of Wallace Stevens (New York: Knopf, ), . . Ibid., –. . Wallace Stevens, Letters of Wallace Stevens, ed. Holly Stevens (New York: Knopf, ), . . Helen Vendler, On Extended Wings (Cambridge, Mass.: Harvard University Press, ), . . Stevens, Collected Poems, .
/05&450$)"15&3
. Ibid., –. . Ibid., . . Louis Osti, “The Craft of John Ashbery,” Confrontation (): . . John Ashbery, Hotel Lautreamont (New York: Knopf, ), . . John Ashbery, A Wave (New York: Penguin, ), . . John Ashbery, Self-Portrait in a Convex Mirror (New York: Viking, ), . . Stevens, Collected Poems, . . John Koethe, “An Interview with John Ashbery,” Sub-Stance (): . $ ) " 1 5 & 3
. Elizabeth Bishop, The Complete Poems – (New York: Farrar, Straus and Giroux, ), . . John Keats, The Poetical Works of John Keats, ed. H. Buxton Forman (London: Oxford University Press, ), II:. . James Thomson, The Complete Works of James Thomson, ed. J. L. Robertson (London: Oxford University Press, ). . Marjorie Nicolson, Newton Demands a Muse (Hamden, Conn.: Archon, ). . William Hooper, Rational Recreations, rd ed. (London, ), II:. . Ibid., . . R. L. Gregory, The Intelligent Eye (New York: McGraw-Hill, ), . . R. L. Gregory and E. H. Gombrich, eds., Illusion in Nature and Art (London: Duckworth, ), . . James Sully, Illusions, th ed. (New York: Da Capo Press, ), . . William Wordsworth, Wordsworth: Poetical Works, ed. Thomas Hutchinson, rev. Ernest de Selincourt (Oxford: Oxford University Press, ), . . Thomas Hardy, The Complete Poems of Thomas Hardy, ed. James Gibson (New York: Macmillan, ), . . Sully, Illusions, . . Gerard Manley Hopkins, Gerard Manley Hopkins, ed. Catherine Phillips (Oxford: Oxford University Press, ), . . Geoffrey Hartman, The Unmediated Vision (New York: Harcourt, Brace and World, ), . . Wordsworth, Poetical Works, ; Thomson, Complete Works (“Autumn”), . . Hartman, Unmediated Vision, –. . Wordsworth, Poetical Works, . . Gregory, Intelligent Eye, . . Wordsworth, Poetical Works, . . Ibid., . . Ibid., . . Ibid., . . Ibid., . . Wallace Stevens, The Collected Poems of Wallace Stevens (New York: Knopf, ), .
/05&450$)"15&3
. T. S. Eliot, Collected Poems – (New York: Harcourt, Brace and World, ), . . Robert Frost, Collected Poems, Prose and Plays, ed. Richard Poirier and Mark Richardson (New York: Library of America, ), . . James Merrill, The Inner Room (New York: Knopf, ), . . Seamus Heaney, Seeing Things (New York: Farrar, Straus and Giroux, ), . . Bishop, Complete Poems, –. $ ) " 1 5 & 3
. Linda Welshimer Wagner, ed., Interviews with William Carlos Williams (New York: New Directions, ), . . William Carlos Williams, The Selected Letters of William Carlos Williams, ed. John Thirlwall (New York: McDowell, Obolensky, ), . . William Carlos Williams, “An Approach to the Poem,” in English Institute Essays (New York: Columbia University Press, ), . . Stephen Cushman, “Williams Revising: The Worksheets of ‘Threnody,’ ” William Carlos Williams Review (): . . Williams, “An Approach to the Poem,” , , . . William Carlos Williams, The Collected Poems of William Carlos Williams, ed. A. Walton Litz and Christopher McGowan (New York: New Directions, ), I:. . William Carlos Williams, The Collected Earlier Poems of William Carlos Williams (Norfolk, Conn.: New Directions, ), –. . David Walker, The Transparent Lyric (Princeton: Princeton University Press, ), . . Williams, Collected Poems, I:, . . Williams, Collected Earlier Poems, –. . Williams, Collected Poems, II:. . The essay from Patroon, together with the first version of the poem, is given in Collected Poems, I:–. . William Carlos Williams, The Collected Later Poems of William Carlos Williams, rev. ed. (Norfolk, Conn.: New Directions, ), . . Williams, Collected Earlier Poems, . . William Carlos Williams, I Wanted to Write a Poem (Boston: Beacon Press, ), . . Marjorie Perloff, “To Give a Design: Williams and the Visualization of Poetry,” in William Carlos Williams: Man and Poet, ed. Carroll Terrell (Orono: University of Maine Press, ), . . Henry Sayre, The Visual Text of William Carlos Williams (Urbana: University of Illinois Press, ), . . Williams, Collected Poems, I:. . William Carlos Williams, Spring and All (New York: Frontier Press, ), . . Williams, Collected Poems, I:. . Ibid.
/05&450$)"15&3
. Williams, Collected Earlier Poems, . . Williams, Selected Letters, . . J. Hillis Miller, ed., William Carlos Williams: A Collection of Critical Essays (Englewood Cliffs, N.J.: Prentice Hall, ), . . Williams, Collected Poems, II:, . $ ) " 1 5 & 3
. Robert Frost, Selected Letters of Robert Frost, ed. Lawrance Thompson (New York: Holt, Rinehart and Winston, ), . . Reginald Cook, Robert Frost: A Living Voice (Amherst: University of Massachusetts Press, ), . . Ibid., , . . Frost, Selected Letters, , , , . . Ibid., . . Reuben Brower, The Poetry of Robert Frost (New York: Oxford University Press, ), . . Richard Poirier, “Robert Frost,” Paris Review (): . . Frost, Selected Letters, . . Ibid., . . Robert Frost, The Letters of Robert Frost to Louis Untermeyer (New York: Holt, Rinehart and Winston, ), . . Robert Frost, “Education by Poetry,” in Selected Prose of Robert Frost, ed. Hyde Cox and Edward Connery Lathem (New York: Holt, Rinehart and Winston, ), . . Cook, Robert Frost, . . Stanley Burnshaw, Robert Frost Himself (New York: George Braziller, ), . . Cook, Robert Frost, –. . Frost, “Education by Poetry,” , . . Cook, Robert Frost, . . Richard Poirier, Robert Frost: The Work of Knowing (New York: Oxford University Press, ), . . Oxford English Dictionary, nd ed. (Oxford: Oxford University Press, ). . Cook, Robert Frost, . . Ibid., , . . Ibid., . . Lawrance Thompson, Robert Frost: The Early Years – (New York: Holt, Rinehart and Winston, ), n. . Cook, Robert Frost, –, . . Ibid., . . Elizabeth Shepley Sergeant, Robert Frost: The Trial by Existence (New York: Holt, Rinehart and Winston, ), . Bracketed phrase is Sergeant’s. . Louis Untermeyer, Bygones: The Recollections of Louis Untermeyer (New York: Harcourt, Brace and World, ), .
/05&450$)"15&3
. –. . . .
Frost, “The Constant Symbol,” “The Figure a Poem Makes,” in Selected Prose, , Frost, Letters to Untermeyer, . Thompson, The Early Years, . Poirier, Robert Frost, x.
$ ) " 1 5 & 3
. Robert Frost, Selected Letters of Robert Frost, ed. Lawrance Thompson (New York: Holt, Rinehart and Winston, ), . . Frost: A Literary Life Considered (New York: Oxford University Press, ), . . Richard Poirier, Robert Frost: The Work of Knowing (New York: Oxford University Press, ), –. . Lawrance Thompson, Robert Frost: The Early Years – (New York: Holt, Rinehart and Winston, ), . . Poirier, Robert Frost, –. . Thompson, Frost: The Early Years, . . Robert Frost, Collected Poems, Prose, and Plays, ed. Richard Poirier and Mark Richardson (New York: Library of America, ), –. . Ibid., . . Ibid., . . Reginald Cook, Robert Frost: A Living Voice (Amherst: University of Massachusetts Press, ), ; Louis Untermeyer, The Letters of Robert Frost to Louis Untermeyer (New York: Holt, Rinehart and Winston, ), . . Untermeyer, Letters of Robert Frost, . . Stanley Burnshaw, Robert Frost Himself (New York: George Braziller, ), ; Cook, Robert Frost, . . Untermeyer, Letters of Robert Frost, . . Ibid., . . William Sutton, Newdick’s Season of Frost (Albany: State University of New York Press, ), . . Frost, Selected Letters, . . Cook, Robert Frost, –. . Burnshaw, Robert Frost Himself, . . Untermeyer, Letters of Robert Frost, –. . Frost, Collected Poems, . . Randall Jarrell, Poetry and the Age (New York: Ecco Press, ), . . Ibid., . The brackets in this passage are Jarrell’s. . Burnshaw, Robert Frost Himself, . . Cook, Robert Frost, . . Elizabeth Sergeant, Robert Frost: The Trial by Existence (New York: Holt, Rinehart and Winston, ), . . Frost, Selected Letters, . . Ibid., –; Frost, Collected Poems, .
/05&450$)"15&3
. Frost, Collected Poems, ; Cook, Robert Frost, . . Cook, Robert Frost, . . Ibid., –. . Ibid., . . Ibid., . . Ibid., . . Ibid., . . Burnshaw, Robert Frost Himself, . $ ) " 1 5 & 3
. Randolph Quirk, A Comprehensive Grammar of the English Language (New York: Longman, ), –. See also Otto Jespersen, The Philosophy of Grammar (London: George Allen and Unwin, ), –. . Hyder Rollins, ed., Tottel’s Miscellany, rev. ed. (Cambridge, Mass.: Harvard University Press, ), II:n. . . Hyder Rollins, ed., The Paradise of Dainty Devices (Cambridge, Mass.: Harvard University Press, ), ; Hyder Rollins, ed., A Gorgeous Gallery of Gallant Inventions (Cambridge, Mass.: Harvard University Press, ), . . Hyder Rollins, ed., England’s Helicon (Cambridge, Mass.: Harvard University Press, ), I:. . Anne Ferry, The Art of Naming (Chicago: University of Chicago Press, ), –. . Samuel Coleridge, The Collected Letters of Samuel Coleridge, ed. Earl Griggs (Oxford: Clarendon Press, ), I:. . Anne Ferry, The Title to the Poem (Stanford: Stanford University Press, ), –. . Hallam Tennyson, Alfred Lord Tennyson: A Memoir (London: Macmillan, ), ; Alfred Lord Tennyson, The Poetic and Dramatic Works of Alfred Lord Tennyson, ed. W. J. Rolfe (Boston: Houghton Mifflin, ), n. . Boyd Litzinger and Donald Smalley, eds., Browning: The Critical Heritage (London: Routledge and Kegan Paul, ), –. . Ibid., n. . Some examples are James Lane Allen, “Anonymity: ‘The Breadwinners,’ ” Critic (): –; Harry Smith, “The Science of Anonymity,” Good Words (): –; Coulson Kernahan, “The Question of Anonymity and Pseudonymity: Should Authors Deny the Authorship of Their Own Work?” Chambers’s Journal , th ser. (): –. . Henry Hazlitt, “The Cult of Anonymity,” Nation, Oct. , , . . E. M. Forster, Anonymity: An Enquiry (London: Hogarth Press, ), . . Ibid., . . John Crowe Ransom, “A Poem Nearly Anonymous,” in The World’s Body (New York: Charles Scribner’s, ), ; reprinted from American Review (May ): –. . Ibid., , . . Ibid., . . Forster, Anonymity, .
/05&450$)"15&3
. Henry Seidel Canby, “Anon Is Dead,” American Mercury (May ): –. . William Cronin, Nature’s Metropolis (New York: W. W. Norton, ), . The earliest use I have been able to find of anonymity in a description by a sociologist of the modern city is, unsurprisingly, in an essay of c. , influential in America, by Georg Simmel, “The Metropolis and Mental Life,” trans. Kurt Wolff, in Cities and Society, ed. Paul Hatt and Albert Reiss (Glencoe, Ill.: Free Press, ), . . William Blake, The Complete Writings of William Blake, ed. Geoffrey Keynes (London: Nonesuch Press, ), . . Ibid., –. . William Wordsworth, The Prelude, or Growth of a Poet’s Mind, ed. Ernest de Selincourt (London: Oxford University Press, ), . All quotations from The Prelude are from this edition; hereafter passages set as block quotations are cited in the text only, by line numbers. . Raymond Williams, The Country and the City (New York: Oxford University Press, ), . For more recent discussions of this material, see, for example, Robert Crawford, The Savage and the City (Oxford: Clarendon Press, ); William Sharpe, Unreal Cities (Baltimore: Johns Hopkins University Press, ); Richard Shusterman, “The Urban Aesthetics of Absence: Pragmatist Reflections in Berlin,” New Literary History (Autumn ): –. . Benjamin Disraeli, Sybil, or The Two Nations (London: Oxford University Press, ), –. . Thomas Carlyle, “The Present Time,” in Latter-Day Pamphlets (London: Chapman and Hall, ), . . Arnold Toynbee, “Industry and Democracy,” in Toynbee’s Industrial Revolution (London: David and Charles Reprints, ), . . George Gissing, Demos (New York: E. P. Dutton, n.d.), . . Upton Sinclair, The Jungle (New York: Doubleday, Page, ), . . Waldo Frank, Our America (New York: Boni and Liveright, ), . . Lewis Mumford, The Culture of Cities (New York: Harcourt Brace, ), , . . James Thomson, The City of Dreadful Night and Other Poems (London: Bertram Dobell, ), . . Ibid., –. . John Davidson, Fleet Street and Other Poems (London: Grant Richards, ), –. . W. B. Yeats, The Collected Poems of W. B. Yeats (New York: Macmillan, ), . . Ezra Pound, Personae (New York: New Directions, ), . . Robert Frost, Complete Poems of Robert Frost (New York: Holt, Rinehart and Winston, ), . . T. S. Eliot, “Preface,” in John Davidson: A Selection of His Poems, ed. Hugh McDiarmid (London: Hutchinson, ), xi. . T. S. Eliot, “What Dante Means to Me,” in To Criticize the Critic (London: Faber and Faber, ), .
/05&450$)"15&3
. T. S. Eliot, “Baudelaire,” in Selected Essays – (New York: Harcourt Brace, ), . . T. S. Eliot, Inventions of the March Hare, ed. Christopher Ricks (London: Faber and Faber, ), . . William Wordsworth, The Prelude , , , ed. Jonathan Wordsworth et al. (New York: W. W. Norton, ), ix. Herbert Tucker has pointed out to me that Eliot used the same tactic of neglecting urban materials in works of the Brownings and Tennyson. It is also the case that he neglects Whitman in this regard. But Whitman’s relation to his own urban material is rather different in its attitude toward anonymity. This could well be the subject of another study. . Eliot, “Tradition and the Individual Talent,” in Selected Essays, . . Eliot, “Baudelaire,” in Selected Essays, . Surprisingly, I have not found any scholarly discussion of the possibility that Baudelaire read The Prelude, although we know that he was familiar with some contemporary English poetry. . T. S. Eliot, Collected Poems – (New York: Harcourt, Brace and World, ), –. For relevant theoretical analyses by linguists of the workings of pronouns discussed here in Eliot’s poems, see Emile Benveniste, Problems in General Linguistics, trans. Mary Elizabeth Meek (Miami: University of Miami Press, ), –; Roman Jakobson, “Shifters, Verbal Categories, and the Russian Verb,” in Russian Language Project (Cambridge, Mass.: Harvard University Press, ), –. . T. S. Eliot, “Wordsworth and Coleridge,” in The Use of Poetry and the Use of Criticism (London: Faber and Faber, ), . John Paul Riquelme points out echoes of Wordsworth’s “Elegiac Stanzas,” ll. – (“I could have fancied that the mighty Deep / Was even the gentlest of all gentle Things”) in Harmony of Dissonances (Baltimore: Johns Hopkins University Press, ), . . Eliot, Collected Poems, . . T. S. Eliot, The Waste Land, facsimile ed., ed. Valerie Eliot (New York: Harcourt Brace Jovanovich, ), . . Eliot, “Tradition and the Individual Talent,” in Selected Essays, .
*/%&9
“Acquainted with the Night” (Frost), – Adamson, Sylvia, – Agricola, “Anon Is Dead” (Canby), Anonymity: An Enquiry (Forster), Anonymity, , , , , n; vs. anonymous, –; of authorship, –; as cultural motif, –, ; history of, –, –; modernism and, –; of urban society, – Anonymous, ; vs. anonymity, –; authorship, –; root meaning of, – Apologie for Poetrie, An (Sidney), Apostrophe: Wordsworth’s use of, – Arte of English Poesie, The (Puttenham), , – Articles, , Art of English Poetry, The (Bysshe), – Ashbery, John, , , ; “Cop and Sweater,” –; “Down by the Station, Early in the Morning,” –; “Grand Galop,” –, –; Self-Portrait in a Convex Mirror, ; on Wallace Stevens, –; A Wave, Astrophil and Stella (Sidney), , , ; rhyming structure in, –, “Attitudes Toward Poetry” (Frost), Auden, W. H., , , “Auroras of Autumn, The” (Stevens), Authorship: anonymous, – Baldwin, William: A Treatise of Morall Philosophy, , n
Banket of Sapience (Elyot), , n Barnes, Barnabe, ; A Divine Centurie of Spirituall Sonnets, Baudelaire, Charles-Pierre, , , n; Les fleurs du mal, ; “Les sept vieillards,” “Being Let In on Symbols, On” (Frost), Bel-vedere or The Garden of the Mvses (Herbert), , (fig.), “Birches” (Frost), , –, , Birds: as symbols, –, –, “Bird Singing in Its Sleep, On a” (Frost), –, Bishop, Elizabeth, , , , ; “The End of March,” ; “The Fish,” –; “Poem,” , “Black Winds, The” (Williams), Blake, William, , ; “A Brief Outline of Hell,” ; “The Chimney Sweeper,” ; “London,” Blank verse, , , , , n Bodenham, John, Booke of Ayres, A (Campion), Booth, Stephen, Borroff, Marie, Bowie, Malcolm, Boy’s Will, A (Frost), –, , – “Brief Outline of Hell, A” (Blake), “Bright & Blue Birds & the Gala Sun, Of ” (Stevens), Brogan, T. V. F., “Brook in the City, A” (Frost), Brower, Reuben, – Browning, Elizabeth Barrett, , n
*/%&9
Browning, Robert, n; Dramatic Lyrics, ; Pauline: A Fragment of a Confession, – Buck, Thomas, “Burial of the Dead, The” (Eliot), –, – Burns, Robert, “Burnt Norton” (Eliot), – Bysshe, Edward, ; The Art of English Poetry, – Campion, Thomas, ; A Booke of Ayres, Canby, Henry Seidel: “Anon Is Dead,” Carlyle, Thomas: “The Present Time,” “Cathedral Facade at Midnight, A” (Heaney), – Chaucer, , n; Troylus and Cresseid, Chicago, “Chimney Sweeper, The” (Blake), “Choose Something Like a Star” (Frost), City of Dreadful Night, The (Thomson), , Coleridge, Samuel Taylor, , , Collected Earlier Poems, The (Williams), , , , Collected Later Poems, The (Williams), Collected Poems (Stevens), , ; use of thing in, – Colloquial speech, “Come In” (Frost), , –, – Come In and Other Poems (Frost), Commonplace book, , n; tables of content, –; The Temple as, , ; titles and topics in, –, n Compleat Angler, The (Walton), Complete Collected Poems, The (Williams), “Composed by the Side of Grasmere Lake” (Wordsworth), –, “Composed During a Storm” (Wordsworth), – “Composed upon Westminster Bridge, September , ” (Wordsworth), “A Confession of a Sinner, acknowledging the miserie of humane frailtie” (Davies),
Conjunctions, , Cooper’s Hill (Denham), , – “Cop and Sweater” (Ashbery), – Copies of Verse, – “Corona, La” (Donne), Cottle, Joseph, Couplets: heroic, , –; Pope’s use of, –, n Cowper, William, , Critics: Frosts’ response to, , – Cronin, William, “Cuckoo, To the” (Wordsworth), , – “Cult of Anonymity, The,” Culture of Cities, The (Mumford), Daniel, Samuel: A Defence of Ryme, Daniels, Roger, Davidson, John, , ; “London Bridge,” Davies, John, ; “A Confession of a Sinner, acknowledging the miserie of humane frailtie,” ; The Mvses Sacrifice; or, Diuine Meditations, Defence of Ryme, A (Daniel), Demos (Gissing), Denham, John: Cooper’s Hill, , – “Desert Places” (Frost), Design, –; Frost’s focus on, –, –, –; in nature, – “Design” (Frost), , –, , , –, –, , ; publication of, –, ; as sonnet, – Dickinson, Emily, “Directive” (Frost), –, , ; analysis of, – Disraeli, Benjamin: Sybil, or The Two Nations, Divine Centurie of Spirituall Sonnets, A (Barnes), Divine Fancies: Digested into Epigrammes, Meditations, and Observations (Quarles), – Donne, John, , , , , n; “La Corona,” ; “Goodfriday, . Riding Westward,” ; “Holy Sonnets,” ; “A Hymne to Christ, at the Authors
*/%&9
last going into Germany,” ; “The Lamentations of Jeremy,” ; Poems, By J. D., –; titles used by, , n; “A Valediction forbidding mourning,” “Down by the Station, Early in the Morning” (Ashbery), – Dramatic Lyrics (Browning), Drayton, Michael: The Harmonie of the Church, Drummond, Henry: Flovvers of Sion; or Spirituall Poems, Early Martyr, An (Williams), “Education by Poetry” (Frost), “Elegy Written in a Country Church-yard” (Gray), Eliot, T. S., , , , , –, , , , n; “The Burial of the Dead,” –, –; “Burnt Norton,” –; “Inventions of the March Hare,” ; “The Love Song of J. Alfred Prufrock,” ; modernist poetry, –; on Moore’s use of rhyme, –, ; “Preludes,” –; “Silence,” ; “Tradition and the Individual Talent,” – Elyot, Thomas, ; Banket of Sapience, , n Emerson, Ralph Waldo, “Emperor of Ice-Cream, The” (Stevens), Empson, William, “End of March, The” (Bishop), End rhymes, , –, , ; in Moore’s poems, –; in Shakespeare’s sonnets, , –; in Sidney’s sonnets, –, – End sounds: in Shakespeare’s sonnets, – End words, –, , ; prepositions and particles as, – Englands Helicon, Epigrammes (Jonson), , , n Epigrams, – Essay on Criticism, An (Pope), , “Evening Walk, An” (Wordsworth), Experimentation, ; Williams’s, – Feminine rhyme: Shakespeare’s use of, –
Ferrar, Nicholas, , “Fire and Ice” (Frost), “First Version: ” (Williams), “Fish, The” (Bishop), – “Fish, The” (Moore): rhyming and structure of, –, – Fleurs du mal, Les (Baudelaire), Flovvers of Sion; or Spirituall Poems (Drummond), “Flowers by the Sea” (Williams), ; structure and revisions of, – Forster, E. M., ; Anonymity: An Enquiry, Frank, Waldo: Our America, Fraser, G. S., French: influence from, Frost, Robert, , , , ; “Acquainted with the Night,” –; “On Being Let In on Symbols,” ; “Birches,” , –; “On a Bird Singing in Its Sleep,” –, ; A Boy’s Will, –, , –; “Come In,” –; “Design,” , –, –, –, –, ; “Directive,” –; A Further Range, –, ; “For Once, Then, Something,” –; “My Butterfly,” –; use of titles, , , –; “In White,” , –; A Witness Tree, Further Range, A (Frost), , , –, Gascoigne, George, , , , , ; on rhyming, –, –, , , “Gascoigne’s good morrow,” –, “Girl, The” (Williams), – Gissing, George: Demos, Godwin, William: Things As They Are or The Adventures of Caleb Williams, “Good Friday” (Herbert), “Goodfriday, . Riding Westward” (Donne), Golding, Arthur, Grammar: particles in, –; of rhyming, , , , –; of titles, “Grand Galop” (Ashbery), –, – Gray, Thomas: “Elegy Written in a Country Church-yard,”
*/%&9
Gregory, R. L., , Grimald, Nicholas, “Grindstone, The” (Frost), Gunn, Thom, ; “Memoirs of the World,” – “’Haddock’s Eyes’: A Note on the Theory of Titles” (Herbert), Hallucination, Hamlet (Shakespeare): blank verse in, , , Hardy, Thomas, , ; “A Cathedral Facade at Midnight,” –; “The Shadow on the Stone,” Harmonie of the Church, The (Drayton), Harmony, , ; in book organization, – Hartman, Geoffrey, –; The Unmediated Vision, – Hawkins, Henry: Parthenia Sacra . . . with Piovs Devises and Emblemes, Hayman, Robert, Heaney, Seamus, ; “Seeing Things,” – Herbert, Edward Lord: Occasional Verses, Herbert, George, , n; Bel-vedere or The Garden of the Mvses, (fig.); “The H. Scriptures,” ; New Year’s sonnets, –; poem titles, , –, –, –; religious verse, , –; “A true Hymne,” –; The Temple, Sacred Poems and Private Ejaculations, –, –, (fig.), , , –, , n Herbert’s Remains, Heroic rhyme, ; Pope’s use of, – Herrick, Robert, , , n Hollander, John, “H. Scriptures, The” (Herbert), “Holy Sonnets” (Donne), Hooper, William: Rational Recreations, Hopkins, Gerard Manley, , , ; riddle poem, – Hotel Lautreamont (Ashbery), – Humor, , , “Hymne to Christ, at the Authors last going into Germany, A” (Donne), Hymns, , n
“Idea of Order at Key West, The” (Stevens), Illusionary experiences, . See also Optical illusions Indexes: use of, –, (fig.) “Influence of Natural Objects in Calling Forth and Strengthening the Imagination in Boyhood and Early Youth” (Wordsworth), – “Inventions of the March Hare” (Eliot), Irony: Eliot’s use of, – Italian: sonnets, , –, I Wanted to Write a Poem (Williams), James, William: Pragmatism, –, Jarrell, Randall, , “Jerboa, The” (Moore), Jonson, Ben, , ; Epigrammes, , , n Jungle, The (Sinclair), “Just Man And, The” (Moore), Keats, John, , , , , ; on science and supernatural, –, “Lake Isle of Innisfree, The” (Yeats), “Lamentations of Jeremy, The” (Donne), Lamia (Keats), Landscape, ; Wordsworth’s use of, – “Landscape with Boat” (Stevens), – Life of Mr. George Herbert (Walton), , , “Locust Tree in Flower, The” (Williams, – Lok, Henry, ; Svndry Christian Passions, Contained in two hundred sonnets, , “London” (Blake), London: as theme, –, “London Bridge” (Davidson), Longfellow, , “Losing the Marbles” (Merrill), Love poems, ; Donne’s, –, n “Love Song” (Williams), “Love Song of J. Alfred Prufrock, The” (Eliot), , “Loving Dexterity, The” (Williams), –,
*/%&9
Lycidas (Milton), – Lyrical Ballads (Coleridge), “Man on the Dump, The” (Stevens), “Man with the Blue Guitar, The” (Stevens), Mask Presented at Ludlow, A (Milton), Mathematics: in Moore’s syllabic verse, – “Meeting and Passing” (Frost), “Memoirs of the World” (Gunn), – “Mending Wall” (Frost), Merrill, James, , ; “Losing the Marbles,” Miller, J. Hillis, Milton, John, , , , , ; Lycidas, –; A Mask Presented at Ludlow, ; Paradise Lost, , ; Poems of Mr. John Milton, ; use of thing, , – “Miserie” (Herbert), Modernist poetry, , –, , , ; anonymity and, –; Marianne Moore’s, – “Monocle de Mon Oncle, Le” (Stevens), Monosyllables, ; rhyming and, , –, Moore, Marianne, , , –, , n; “The Fish,” –, –; “The Jerboa,” ; “Nevertheless,” ; “Those Various Scalpels,” –, , , ; Selected Poems, –; “Subject, Predicate, Object,” – “Most of It, The” (Frost), , , , “Moth Seen in Winter, To a” (Frost), , Mountain Interval (Frost), , “Mowing” (Frost), , , Mumford, Lewis: The Culture of Cities, Mvses Sacrifice; or, Diuine Meditations, The (Davies), “My Butterfly: An Elegy” (Frost), , , –, Myth: as theme, – Nature, ; design in, –, –; as theme, –, –, –; Wordsworth on, –
“Need of Being Versed in Country Things” (Frost), “Neither Out Far nor In Deep” (Frost), “Nevertheless” (Moore), Newdick, Robert, New Hampshire (Frost), Newton, Isaac: Opticks, Newton Demands the Muse (Nicolson), New Year’s sonnets (Herbert), – New York, Nicolson, Marjorie: Newton Demands the Muse, “Nightingales, The” (Williams): revisions, , – “Night-piece, A” (Wordsworth), –, North of Boston (Frost), “Note on Moonlight” (Stevens), – “Nothing Gold Can Stay” (Frost), “Not Ideas about the Thing but the Thing Itself ” (Stevens), – Nouns, , , , ; adjectival, –; in titles, – Obscurity, , Occasional Verses (Edward Herbert), “Ode on Solitude” (Frost), “Ode to a Nightingale,” , “Ode to the West Wind” (Shelley), Old Line, The (magazine), Old Testament, Oley, Barnabas, Omnipresence: Milton’s concept of, “Once, Then, Something, For” (Frost), –, Optical illusions, , , –, –; Frost’s use of, –; Wordsworth’s use of, –, , – Opticks (Newton), “Ordinary Evening in New Haven, An” (Stevens), Our America (Frank), “Oven Bird, The” (Frost), Owen, Wilfred, “Paper Nautilus, The” (Moore),
*/%&9
Paradise Lost (Milton), , –; use of thing in, , Partheneia Sacra . . . with Piovs Devises and Emblemes (Hawkins), Participles: present, , –, – Particles: as end words, – “Pastoral” (Williams), Pauline: A Fragment of a Confession (Browning), – Perloff, Marjorie, Personification, Petrarch, Pictures from Brueghel (Williams), Plymouth Normal School, “Poem” (Bishop), , “Poem Nearly Anonymous, A” (Ransom), – Poems, By J. D. (Donne), n; grouping and titling in, – Poems by Two Brothers (Tennyson), Poems, Chiefly Lyrical (Tennyson), Poems of Mr. John Milton, Poetical Rapsody, A, Poirier, Richard, ; Robert Frost: The Work of Knowing, Politeuphuia. Wits Commonwealth, Pope, Alexander, , , , n; An Essay on Criticism, , ; Windsor-Forest, – “Possible Sort of Song, A” (Williams), Pound, Ezra, , , ; “In a Station of the Metro,” Pragmatism (James), –, Prelude, The (Wordsworth), , ; on urban society, – “Preludes” (Eliot), – Preparation to the Psalter, A (Wither), – Prepositions, , , “Present Time, The” (Carlyle), Proverbial rhymes, , – “Provide, Provide” (Frost), Prufrock and Other Observations (Eliot), Psalmes of David, The, Psalms: Herbert’s works as, , –, – Psalter, , , , Pseudonyms, “Publican, The” (Herbert),
Punctuation: Wordsworth’s use of, – Puttenham, George, ; The Arte of English Poesie, , – “Putting in the Seed” (Frost), Quarles, Francis: Divine Fancies: Digested into Epigrammes, Meditations, and Observations, – “Range-Finding” (Frost), Rank: and rhyming styles, Ransom, John Crowe: “A Poem Nearly Anonymous,” – Rational Recreations (Hooper), Redesign: Williams’s, – Religion: commonplace books in, – Religious verse, , n; Donne’s, –, n; Herbert’s, –, –, –, –; titles of, –, –, –, –; Wordsworth’s use of, – “Reverie of Poor Susan, The” (Wordsworth), Revision: critics of, –; Frost’s, –; Williams’s, – Rhyme, rhyming, –-, , ; anticlassical, –; feminine, –; grammar of, , –; history of, –; line endings in, –; Moore’s use of, –, –; nineteenth-century, –; Pope’s use of, –; proverbial, –; sonnets, , , , –, –; standards for, – Rhyme chains, – Rhyme pairs, , –, –, , , ; anticlassical, – Riddle poems, – “Road Not Taken, The” (Frost), Robert Frost: The Work of Knowing (Poirier), Rock, The (Stevens), Sacred verse. See Religious verse Saxon language, , Sayre, Henry, Science, ; as theme, – Seasons, The (Thomson), – “Seeing Things” (Heaney), –
*/%&9
Selected Poems (Moore): critiques of, –; use of rhyme in, – Self-Portrait in a Convex Mirror (Ashbery), , – Semantic fields: in rhyming, – Semantic propinquity, “Sense of the Sleight-of-Hand Man, The” (Stevens), Sense-system, Sentence-sound, , “Sept viellards, Les” (Baudelaire), “Shadow on the Stone, The” (Hardy), Shakespeare, William, n; rhyme structures used by, , , –, –; sonnets, , –, , , –, –, , , –, – Sheapeardes Calender, The (Spenser), Shelley, Percy Bysshe, , , , ; “Ode to the West Wind,” Sidney, Philip, n; An Apologie for Poetrie, ; Astrophil and Stella, –, , –, –; rhyming by, –, , , ; sonnets by, , –, –, “Silence” (Eliot), Sinclair, Upton: The Jungle, “Sinner, The” (Herbert), “Snow Man, The” (Stevens), Songes and Sonnettes (Tottel), Sonnet (Shakespeare), Sonnet (Shakespeare), – Sonnet (Sidney), , Sonnet (Sidney), – Sonnet (Shakespeare), Sonnet (Shakespeare), Sonnet (Sidney), , n Sonnet (Shakespeare), –, n Sonnet (Sidney), , Sonnet (Sidney), – Sonnet (Shakespeare), , – Sonnet (Shakespeare), , , Sonnet (Sidney), – Sonnet (Sidney), , – Sonnet (Shakespeare), – Sonnet (Sidney), – Sonnet (Sidney), , Sonnet (Shakespeare),
Sonnets: Frost’s, , –, –; New Year’s, –; nocturnal, –; religious, –; rhyme structures in, –, –; Shakespeare’s, , –, , –, –, –, –, n; Sidney’s, –, , –; Wordsworth’s, – “Sort of Song, A” (Williams), Sound-system, Sour Grapes (Williams), , Southwell, Robert, , , Spenser, Edmund, ; The Shepeardes Calender, Spontaneity: Williams’s, Spring and All (Williams), “Spring Strains” (Williams), “Station of the Metro, In a” (Pound), Steeple Bush (Frost), Stevens, Wallace, , , , , n; “The Auroras of Autumn,” ; “Of Bright & Blue Birds & the Gala Sun,” ; Collected Poems, –; “The Idea of Order at Key West,” ; influence of, –; “Landscape with Boat,” –; “The Man on the Dump,” ; “Note on Moonlight,” –; “Not Ideas About the Thing but the Thing Itself,” –; “An Ordinary Evening in New Haven,” ; on Selected Poems, –; “The Snow Man,” ; use of thing, , –, “Stopping by Woods on a Snowy Evening” (Frost), , Storehouses, Structure: of poems, – “Subject, Predicate, Object” (Moore), – Subversion, Sully, James, , – Supernatural, , , , “Supposed Confessions of a Second Rate Mind Not in Unity with Itself ” (Tennyson), Surrey, Earl of, , , Svndry Christian Passions, Contained in two hundred sonnets (Lok), , “Sweeney Agonistes” (Eliot), Swinburne, Algernon Charles, Sybil, or the Two Nations (Disraeli),
*/%&9
Syllables: in Moore’s poems, – Tables of contents: use of, –, Temple, Sacred Poems and Private Ejaculations, By Mr. George Herbert, The, –, (fig.), , n; and Psalter, , ; format of, –; titles used in, , (fig.), – Tennyson, Alfred Lord, , , n “There Was a Boy” (Wordsworth), – Thing, , –; Ashbery’s use of, –; definitions for, –; poetic use of, , –; Stevens’ use of, –; Wordsworth’s use of, –, , – Things as They Are or The Adventures of Caleb Williams (Godwin), Thompson, John, – Thompson, Lawrance, Thomson, James, ; The City of Dreadful Night, , ; The Seasons, –; “William Blake,” “Those Various Scalpels” (Moore), –, , , Thynne, Francis, “Tintern Abbey” (Wordsworth), –, , , Titles, , , , n, n; Donne’s poems, –, n; Frost’s use of, –, , , –; Herbert’s use of, –, –, –, –, –, –, –; indexes to, , (fig.), ; and poem structure, –; of religious verse, –, –, –, –, –, Tomlinson, Charles, Tottel, Richard, –, ; Songes and Sonettes, Toynbee, Arnold, “Tradition and the Individual Talent” (Eliot), –, Treatise of Morall Philosophy, A (Baldwin), , n Trilling, Lionel, , Troylus and Cresseid (Chaucer), “True Hyme, A” (Herbert), – Ulteriority, ,
Unmediated Vision, The (Hartman), – Untermeyer, Louis, , , , Urban society: anonymity in, –, n; images of, –, n “Valediction forbidding mourning, A” (Donne), Vision(s), ; Eliot’s use of, –; supernatural, , , , , ; Wordsworth’s use of, – Voyage to Pagany, A (Williams), Walker, David, Walton, Isaak, ; The Compleat Angler, ; on Herbert, ; Life of Mr. George Herbert, , , Ward, Susan Hayes, , Ward, William Hayes, Waste Land, The (Eliot), –, , Wave, A (Ashbery), Wedge, The (magazine), West-Running Brook (Frost), “White, In” (Frost), , , –, , , Whole Boke of Psalmes, collected into Englishe Metre, The, Wilbur, Richard, “William Blake” (Thomson), Williams, Raymond, Williams, William Carlos, , –, , , ; “The Black Winds,” ; The Collected Earlier Poems, , ; The Collected Later Poems, ; “Flowers by the Sea,” –, ; “The Girl,” –; I Wanted to Write a Poem, ; “The Locust Tree in Flower,” –; “Love Song,” ; “The Loving Dexterity,” –, ; “The Nightingales,” –; “Pastoral,” ; “A Possible Sort of Song,” ; “A Sort of Song,” ; Spring and All, ; A Voyage to Pagany, ; “Young Woman at a Window,” – William Zorach / Two Drawings, William Carlos Williams / Two Poems, Windsor-Forest (Pope), – “Witch of Coos, The” (Frost),
*/%&9
Wither, George: A Preparation to the Psalter, – Witness Tree, A (Frost), “Woman Who Blamed Life on a Spaniard, The” (Stevens), Wordsworth, William, , , , , , , , ; on anonymity, –; “Composed by the Side of Grasmere Lake,” –, ; “Composed During a Storm,” –; “Composed upon Westminster Bridge, September , ,” ; “To the Cuckoo,” , –; Eliot and, –, , ; “An Evening Walk,” ; influence of, , –, –; “Influence of Natural Objects
in Calling Forth and Strengthening the Imagination in Boyhood and Early Youth,” –; “A Night-piece,” –; The Prelude, , , –, ; “The Reverie of Poor Susan,” ; “There Was a Boy,” –; “Tintern Abbey,” –, , , ; use of thing, , –, , –, Wyatt, Thomas, , , Yeats, William Butler, ; “The Lake Isle of Innisfree,” Young, Edward, , “Young Woman at a Window” (Williams), –