Temporal Thinking in Modern Lyric by Katherine M. Miller B.A., Wellesley College, Wellesley MA, 2006 M.A., Brown University, Providence, RI, 2010 A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Department of English at Brown University PROVIDENCE, RHODE ISLAND May 2014 c Copyright 2014 by Katherine M. Miller This dissertation by Katherine M. Miller is accepted in its present form by Department of English as satisfying the dissertation requirement for the degree of Doctor of Philosophy. Date Mutlu Konuk Blasing, Professor of English, Advisor Recommended to the Graduate Council Date Paul Armstrong, Professor of English, Reader Date Ravit Reichman, Associate Professor of English, Reader Approved by the Graduate Council Date Peter Weber, Dean of the Graduate School iii Acknowledgements For their guidance and support, I owe many thanks to my dissertation committee. Mutlu Blasing’s criticism has been a touchstone for me, and every chapter of this dissertation has been shaped by her sharp advice. For helping me hone my questions, and for guiding me toward challenging and complementary bodies of work, I am grateful. My conversations with Paul Armstrong helped me to broaden the perspective of my argument—to consider my claims about poetry in the light of narratology, hermeneutics, and phenomenology—and also to look beyond it to new approaches and critical contexts. Early in the dissertation, Ravit Reichman prompted me to consider the ethical implications of my reading of the temporality of modern lyric, and her generous readings have been important throughout the writing process. My time at Brown has been shaped by the support of many faculty members, fellow grad- uate students, and administrative staff. Lawrence Stanley, Copp´elia Kahn, and George Landow provided pedagogical mentorship. The New England Americanist Collective, advised by Deak Nabers, and the Harvard American Colloquium have read and commented on large portions of the dissertation. Lorraine Mazza and Ellen Viola have been helpful, generous, and kind. I am grateful to my graduate student peers for their contributions to my thinking and for their collegiality and friendship. Writing can be a lonely exercise, and it would be neither rewarding nor possible without my family and friends. Thank you to my parents and sister, to Jacqueline, and to Megan and Michelle. Thanks to Peach. Many thanks, and all my love, to Th´eo. I know we shall not cease from our exploring. iv Contents Acknowledgments iv 1 Introduction 1 2 “this will happen again—where? when?”: Typological Thinking in H.D.’s Trilogy 43 3 “Every poem an epitaph”: Rhythm, History, and Genre in Four Quartets 71 4 Repetition and Revision in the Poetry of W. B. Yeats 109 5 Hart Crane: The Bridge and the Limits of Genre 152 v C HAPTER O NE Introduction 2 At the point in which sound is about to be ruined in the abyss of sense, the poem looks for shelter in suspending its own end in a declaration, so to speak, of the state of poetic emergency. —Giorgio Agamben, “The End of the Poem” Each of these compositional generators—line and design—is murderously disposed towards the other. . . . The poem can succeed only if neither principle is perfectly satisfied. In other words, it can succeed only upon condition that it fails. —Simon Jarvis, “The Melodics of Long Poems” I. The Temporality of Modern Lyric My work examines the ways in which modernist lyric poems activate specific ways of thinking through their formal and thematic engagements with time. One particularly persistent line of literary criticism has theorized the temporal mode of lyric as a kind of perpetual present that resists the passage of time and strives for access to the eternal. This line of thought has led to a larger assumption about the closed nature of lyric time: that is, that the “now” that lyric insists upon makes it fundamentally less capable than other literary genres of incorporating other tenses, of positing relations between temporal actions or events, or of navigating between story and discourse. I argue that this assumption will not withstand a sustained examination of how time works in lyric and the thinking that it enables. Because the temporality of a poem is part of its non-paraphrasable content, it is difficult to use it to produce a “reading” of a text. I argue that what the temporal qualities of lyric may reveal instead is how a poem thinks, how we think when we read a poem, and what poetic modes of thinking can offer—ethically as well as aesthetically—that other cognitive modes cannot. I argue that lyric poetry operates not in a timeless present tense, but according to a typological structure that calls for a continual, mutually reinforcing process of anticipation and retrospect. This temporal process requires the reader to consider each phoneme in a lyric poem as a possible sonic or semantic “figure” of a later “fulfillment,” which in turn may be a figure in one 3 or more patterns. Typology, sometimes called figural interpretation, is most commonly identified with a method of Christian exegesis and a related mode of historical interpretation. However, as Erich Auerbach has shown, figura acquired associations with non-logical relations of appearance or similitude well before Tertullian: in the early Roman empire, “[a]s one may well imagine, it is the poets who were most interested in the shades of meaning between model and copy, in changing form and the deceptive likenesses that walk in dreams” (21). Auerbach is concerned with how, “on the basis of its semantic development a word [figura] may grow into a historical situation and give rise to structures that will be effective for many centuries” (76). What I call the typological mode in lyric is a non-theological mode of interpretation, a kind of typology without telos. Lyric may thematize human yearning for access to an eternal, atemporal realm, but the poetic structures that express these sentiments are necessarily temporal. Because we do not know the end in advance, we must anticipate fulfillments of patterns that are still forming, and must revise the patterns we look forward to and those we have already identified as we read. For example, take W. B. Yeats’s “A Coat”: I made my song a coat Covered with embroideries Out of old mythologies From heel to throat; But the fools caught it, Wore it in the world’s eye As though they’d wrought it. Song, let them take it For there’s more enterprise In walking naked. After the first four lines—which suggest an enclosed (ABBA) rhyme scheme and modified ballad meter, and constitute one complete syntactic phrase—we might expect that these patterns constitute a structural figure, that the next four lines will respond in kind. Instead, in line 5, the phrase is prematurely broken; in line 7, the enclosed rhyme scheme breaks down, replaced by what appears to be an ABAB scheme; in line 9, this rhyme scheme, too, is disrupted, and we find 4 that the last 6 lines of the poem shift from what might have been a quatrain and an ending couplet to two enclosed-rhyme tercets. And yet, for all of these disruptions, we feel, as Nicolas Abraham puts it in his discussion of poetic duration, that “[t]he successive semantemes emerge as the only ones possible, and yet each one of them surprises me by being just as it is” (59). At the level of the word, the title figures the “coat” at the end of line 1, the “throat” end rhyme in line 4, the internal slant rhyme of “caught” in line 5, and, by semantic opposition, the concluding foot of the poem, “naked.” The last lines feel appropriate because they surprise us and because this surprise is tempered by a sense of retrospective necessity. In the last word, the poem dis-mantles the semantic assumption we might make from the title and unexpectedly rhymes an adjectival foot with a verb-object phrase (“take it”); here, in sound as in sense, the figure finds its seemingly unavoidable fulfillment. In modernist poetry, the typological mode of lyric may complicate or directly contradict theories of time, history, and subjectivity advanced by the poets in their prose works and by critics who assume a direct relationship between a poet’s verse and his or her literary or social criticism. This mode offers readers a model for ethical reading that is distinct from literary genres such as narrative, drama, and the non-fiction essay. It requires readers to imaginatively inhabit different temporal positions successively and simultaneously, and—quite apart from any logical system—it keeps the possibilities of the past in play even after they seem to be fixed. The typological mode also calls into question some commonplaces of narrative theory, such as Frank Kermode’s characterization of poetry as an “art of the timeless prison” (174), or the distinction Paul Ricoeur draws between poetry and narrative, which follows from the observation that “time becomes human to the extent that it is articulated through a narrative mode” (Time and Narrative Vol. 1, 52). The processes of narrative may be analogous to typology in lyric poetry, but lyric typology is not a subset of narrative operation—rather, it is antecedent to the operations of plot. Moreover, the effects of lyric typology are more concentrated than narrative patterns because lyric figures participate in patterns generated by sound as well as sense, and construct patterns at the level of 5 the line as well as the semantic design (see Jarvis). In this study, I focus on lyric poems written in the period of literary modernism, roughly 1890–1945, when poets’ attention to temporal themes and their experimentation with formal patterns contributed to a heightened sense of the relation of lyric poetry to time. Modernist lyric poems call attention to the time of reading as well as to the time of composition, and to the aesthetic, ethical, and cognitive effects of the revisions that may be made to a given poem or set of poems across time. The long poems that I classify as lyrics, such as H.D.’s Trilogy, Eliot’s Four Quartets, and Crane’s The Bridge, test the assumed material length of a lyric poem and the temporal duration it takes to read one. Finally, because many of the poems in this study were composed near the end of the poet’s career, they may be interpreted as commenting on earlier works and participating in typological patterns that exceed the level of the poem—typology at the level of the book, the collected volume, or the poet’s oeuvre to that date. Although none of the poems I consider are the last poem in that oeuvre, much less the “last words” of the poet, late-career poems in particular suggest that “if all a poet’s words aspire to be last words, then every poem is potentially a deathbed poem, a written guarantee that no poet will ever die without a lasting legacy” (Fuss 899). These lyrics thus retrospectively modify the legacy of a career and provide a provisional fulfillment. My claims about the typological mode in modernist lyric engage with the questions posed by several distinct debates in poetics, genre theory, and modernist studies. To wit: What texts qualify as modernist poems? What poems qualify as lyric? How should we consider “lyric time” in relation to the time of other literary genres? What kind of thinking does lyric enable that other literary genres do not? And, finally, what are the ethical implications of applying the typological mode to poetry generally, and to modernist lyric poetry in particular? In the sections that follow, I address the aspects of these debates that are most relevant to an understanding of my argument and to the way that I position the poems of H.D., Eliot, Yeats, and Crane. 6 II. New Modernisms and Historical Limits The term “modernity” . . . . designates more generally the problematical possibility of all literature’s existing in the present, of being considered, or read, from a point of view that claims to share with it its own sense of a temporal present. In theory, the question of modernity could therefore be asked of any literature at any time, contemporaneous or not. —Paul de Man, “Lyric and Modernity” The definition of literary modernism, and the relation of literary modernism to modernity, have long been contentious subjects. Arriving at a definition of modernism is “all the more difficult,” Lawrence Rainey writes, “because the term has been extended to cover disparate yet cognate fields, acquiring different chronologies and shades of meaning as it moves from art history to architecture to music and intellectual history” (xx). If literary modernism bears a direct relation to the conditions of modernity—if it tends to precede, develop alongside, or react to modern life—then, as de Man observed, in theory it is a label without historical precision. Modernity itself may describe both the movement that turns away from literature and the condition that makes it possible: The continuous appeal of modernity, the desire to break out of literature toward the reality of the moment, prevails and, in its turn, folding back upon itself, engenders the repetition and continuation of literature. Thus modernity, which is fundamentally a falling away from literature and a rejection of history, also acts as the principle that gives literature duration and historical existence. (de Man, “Literary History and Literary Modernity” 162) And yet, for literary critics, modernism remains a necessary term. It is, as Michael Levenson observes, “at once vague and unavoidable”: “Anything more precise would exclude too much too soon; anything more general would be folly” (vii). 7 I define modernism as the historical period from 1890 to 1945 when Anglophone lyric poets directly and repeatedly took up questions of time and history in their poetry, essays, and manifestoes1 and, in lyric poetry, introduced disruptions or innovations of the formal patterns that are observable only through time. In prose, their theories of human and literary history share a tendency to undercut evolutionary, progressive, or strictly linear models of time;2 in poetry, these concerns may be reflected both thematically and formally, although the non-semantic patterns of logic in the poem may or may not run parallel to the poem’s paraphrasable “argument.” This historical period is broader than some definitions of modernism—e.g., Levenson’s focus on 1908–1922 (when the doctrines of “Impressionism, Imagism, Vorticism, and Classicism” developed)—and narrower than others—e.g., Donald Wesling’s, which expands the field of modernity (in which “modernism” is a “subparadigm shif[t]”) from the origins of Romanticism in 1795 to the present (7, 99). It also rejects some common methods of narrowing modernity3 in favor of historical and linguistic delimitation. This is because, in addition to the periodizing concerns noted above, lyric poems are most usefully read together when they are read in their original language. This definition of modernism pushes back against the expansionist movement in recent literary criticism. Take, for example, the new modernist studies, which promises an enlargement of the modernist field’s historical period and geographic boundaries as well as its objects of study. In their introduction to Bad Modernisms and, more recently, in “The New Modernist Studies,” Douglas Mao and Rebecca Walkowitz describe the turn in modernist scholarship toward 1 Take, e.g., Pound’s famous formulation of Imagist poetry: “An ‘Image’ is that which presents an intellectual and emotional complex in an instant of time” (“A Retrospect”). Lawrence, in free verse, seeks to inaugurate the “plasm[ic]” poetry of the present, the “seething poetry of the incarnate Now” (“The Poetry of the Present”). Stevens presents the poetic moment as always in development, in flux, on the cusp: past and future are both impermanent (21), and the imagination “is always at the end of an era. . . it is always attaching itself to a new reality, and adhering to it” (22). 2 For Yeats, history takes the form of the gyre, a structure that builds with increasing intensity until it collapses upon itself and reverses course, only to eventually repeat the process again; Hulme suggests a continual alternation between what he terms romantic and classical periods; Lewis describes a struggle between what he terms time-minds and space-minds; and Eliot, in addition to suggesting that all poets return to and incorporate the existing literary tradition in their work, later comes to propose a model of literary cycles (“What is a Classic?”, “The Music of Poetry”). 3 For example, de Man’s focus on “geographical, spatial terms” and “general and theoretical” themes (“Lyric and Modernity” 167). 8 “extend[ing] the designation ‘modernist’ beyond such familiar figures as Eliot, Pound, Joyce, and Woolf. . . and embrac[ing] less widely known women writers, authors of mass cultural fiction, makers of the Harlem Renaissance, artists from outside Great Britain and the United States, and other cultural producers hitherto seen as neglecting or resisting modernist innovation” (Bad Modernisms 1). Methodologically, they note, “the new modernist studies has moved toward a pluralism or fusion of theoretical commitments, as well as a heightened attention to continuities and intersections across the boundaries of artistic media, to collaborations and influences across national and linguistic borders, and (especially) to the relationship between individual works of art and the larger cultures in which they emerged” (2). Work in this vein4 has helped put pressure on the mostly male, mostly Eurocentric modernist canon; illuminated the political and social contexts for works long accepted as aesthetically modernist; and, following the work of Andreas Huyssen and others, interrogated the “high modernist dogma” (Huyssen ix) that insisted on a strict separation between high art and mass culture. Taken to the extreme, however, the “temporal,” “spatial,” and “vertical” transformations of modernist studies—which expand literary modernism beyond the rough historical period of 1890–1940, beyond European and North American literatures, and beyond forms of “high art” to popular art and other cultural productions, respectively (“The New Modernist Studies” 737)—threaten the stability of modernism as a meaningful critical concept. For example, Susan Stanford Friedman’s “definitional excursions” regarding modern/modernity/modernism, culminating in her theorization of “a planetary approach to modernism that breaks the Anglo-European hold on the field” (“Musing” 475), suggest that “[e]very modernity has its distinctive modernism” (475); that is, given that the “specific conditions of modernity” may arise at different moments and in different locations “for different genders, races, nations, and so forth” (474), we may apply the term modernism equally to “the 4 See, e.g., Laura Doyle and Laura Winkiel’s edited volume, Geomodernisms: Race, Modernism, Modernity (2005); Matthew Hart’s Nations of Nothing But Poetry (2010); and Jessica Berman’s Modernist Commitments: Ethics, Politics, and Transnational Modernism (2011), among others. 9 early modernities of India in the fourteenth-fifteenth centuries or the postcolonial modernities of Africa after 1960” (480), to the stream-of-consciousness novels of James Joyce or the realist novels of Monica Ali. In particular, we should be suspicious of “the West’s narrative of itself” as “the story of its own invention of modernity,” a story in which “‘we’ did it first in the West” and “‘they’ did it elsewhere: derivative and belated” (476-477). The objections to the new modernist studies and to Friedman’s “planetary” approach tend to focus on the utility of modernism as a term that refers to a specific historical period and/or a specific aesthetic movement. For example, David James and Urmila Seshagiri, who support the critical turn from the “mythos of Western metropolitan modernism” to a “more nuanced, historically balanced, and demographically diverse understanding of cultural production in the early twentieth century” (87), also argue for the necessity of certain historical and conceptual limits to the modernist label. They note that, despite “numerous” and “varied efforts to refute modernism’s early-twentieth-century genesis, the decades between 1890 and 1940 repeatedly emerge as the chrysalis for modern literature’s form-breaking work” (91). In their view, stricter periodization actually helps preserve the formal significance of modernism as a movement and, perhaps more surprisingly, might help promote the transnational turn in modernist scholarship: “Temporal contraction, or periodization, helpfully counterbalances the spatial expansiveness of the new modernist studies. Indeed, giving precision to modernism’s place in time facilitates a more historically robust understanding of how it has moved across cultures” (90). More broadly, James and Seshagiri insist that “modernism’s formal promises for narrative fiction will make political, ethical, and aesthetic sense in the future only if modernism denotes a historically precise activity instead of connoting radical artistic responses to every modernity’s upheavals” (97). The consideration of modernism as “a moment as well as a movement,” a “period and a paradigm” (James and Seshagiri 87) represents a refinement of the new modernist studies rather than a rejection of its broad goals (it does, however, reject Friedman’s “planetarity”). A different, more pointed line of criticism takes aim not only at the expansion of modernism as a category but 10 also at the hermeneutic position from which many critics associated with the new modernist studies approach canonical modernist works. Responding to the collection of essays in Bad Modernisms, Charles Altieri argues that “the idealization of subversion affords a prime instance of the pathos involved in being able to idealize from the past only its resistance to conventional idealization. . . . The more we become programmatically suspicious of authorial intentions, the more we seem left only the principle of subversion to justify the critical energies that we devote to our studies” (“How” 767). In his view, the line of discourse that “idealizes various forms of negation like subversion and critique” produces an essential misreading of modernism because “subversion and critique were very rarely ends in themselves [for modernist writers] but functioned as means to clear the air and the ground for positive visions” (765). By privileging authorial intention in this way, Altieri aligns himself with Best and Marcus’s model of surface (as opposed to symptomatic) reading, in which “a true openness to all the potentials made available by texts is also prerequisite to an attentiveness that does not reduce them to instrumental means to an end” (Best and Marcus 16). Although the poems in this study are aligned with an aesthetic conception of high modernism that, Friedman argues, is neither “desirable nor possible” (474) for contemporary critics, they nevertheless suggest some avenues for expansion. First, Temporal Thinking in Modern Lyric is a transatlantic project, comprising poets who were born in the United States (Crane, Eliot, H.D.) and in Ireland (Yeats), who adopted British citizenship by choice (Eliot) and through marriage (H.D.), and who had complicated and evolving relationships with their respective national literatures. H.D., the only woman poet in this group, has until recently been read primarily through the lens of feminist theory rather than through the aesthetic, political, and historical frameworks that have been applied to her male counterparts. Moreover, while my argument subscribes to the conception of modernism as a particular historical period, the works that I choose push the boundaries of this period beyond 1940. Each chapter pays particular attention to poems that were written after the poet had already composed a large body of work, 11 including poems often more strongly associated with the modernist moment (e.g., Eliot’s The Waste Land or H.D.’s Sea Garden). This selection is deliberate, allowing me to consider the poem’s relation not only to other literary-historical sources but to the poet’s own body of work. In this sense, these particular lyric poems simultaneously contribute to the modernist movement as we understand it today and look back on a modernism that has already passed. More broadly, modernist lyric presents opportunities for “travel” that are distinct from the transnationalism or transculturalism of other literary genres. Jahan Ramazani’s concept of “traveling poetry,” which I adopt and modify, holds that “[l]yric highlights how lines of thought, analogy, and cross-cultural reading—whether strong ligaments or tenuous filaments—connect disparate human experiences” (290). That is, lyric allows for “an imaginative enactment of geographic movement” (283) that proceeds from line to line or stanza to stanza, or even between or within words. Eliot’s “The Waste Land” operates this way, for instance, when the image of the “Falling towers” immediately invokes “Jerusalem Athens Alexandria / Vienna London / Unreal” (373–376). Sonic patterning or punning can also contribute to the effect of movement, as when, in Wallace Stevens’s “Bantams in Pine-Woods,” “Chieftan Iffucan of Azcan in caftan” suggests a conflation of Mayan, Aztec, Middle Eastern, and even Appalachian landscapes; here, Ramazani claims, “[r]iding the back of such caravans of sound, poetry traverses real and semireal landscapes—perhaps more nimbly than less sonically rich, more prosaically referential forms” (286). As opposed to travel in the novel—often represented by “the plot-driven excursus into a foreign land”—travel in poetry “may occur at the level of a substituted letter, a varied rhythm, or a pivoting line” (291). Thus, for instance, in H.D.’s Trilogy, “here am I, Amen-Ra whispers, / Amen, Aries, the Ram” (TWDNF 21). This kind of travel is not just geographic or spatial, as Ramazani emphasizes, but temporal as well. Structurally, it is typological. When Ramazani quotes Dionisio D. Mart´ınez’s “Hysteria,” he writes: “The lyric ‘I’ enables the poet to weave chiastically together both Sandburg’s windy Chicago and the prohibitions in Tiananmen Square, both China and North America” (302). However, typology describes this movement better than rhetorical chiasmus, 12 because what the poem does exceeds the operation of inverted parallelism. Tiananmen Square is both a fulfillment, in time and space, of Sandburg’s Chicago, and a figure that finds provisional fulfillment in “the Great Walls of North America.” Modernist lyric poetry, then, expands our geographic and temporal horizons by virtue of its formal properties and its typological mode. My focus on Anglophone modernism in particular is driven not by any allegiance to an idea of development in which these artists innovate and others follow, but simply by my conviction that the non-semantic properties of language are vital to the reading of a poem—that I cannot do the work I want to do by reading a translation in English—and by the fact that, given their shared language and the temporal overlap of their careers, these artists worked with similar literary canons and responded to many of the same historical events. All that said, I believe that the urge to link modernity and literary modernism to such an extent that one reads both through the epistemology of “planetarity” is a misguided one. On the one hand, it should go without saying that the conditions of modernity have arisen at different places in different times—sometimes even in the same places at different times, and at the same time in different places. However, it does not follow that all cultural productions that contribute to or respond to these events are modernist works. To put it more boldly, when was there ever a place or time in which poets were not confronted, to greater or lesser extent, by social and technological change, the threat of war or its aftermath, anxiety about or nostalgia for the past, and the hybridity that stems from contact between cultures? When Friedman speaks of modernism, she seems to be speaking of the conditions of creativity in general, even of the human condition. If “modernism” simply connotes these conditions, no critic would want to limit the term to Anglo-European artists of the twentieth (or any other) century.5 However arbitrary its limits must always be, periodization remains a necessary construct for critics who seek to describe aesthetic movements whose significance derives, at least in part, from their historical context. 5 Here, “modernism” faces a semantic challenge that other periodizing labels do not. Terms that refer to specific historical conditions (e.g., Antebellum American literature, Victorian literature) cannot be meaningfully expanded. And there is no comparable critical pressure, for example, to expand the term “romanticism” to include all conditions associated with the various connotations of “romance.” 13 III. The Transhistorical Lyric Genre In lyric studies, as in modernist studies, one of the most pressing critical debates has to do with the definition of the field. What kind of poem counts as a lyric poem? Is the lyric a genre that is—or ought to be—bound by history, culture, and language? Does lyric transcend these particular contexts? Or is it more properly thought of not as a historical genre, but as a universal mode? Unlike the new modernist studies, however, the new lyric studies seeks to narrow the range of texts that fall into its field—in this case, the range of poems that may be considered lyrics—by calling on critics to properly historicize the category of lyric and the reception of lyric poetry, and to apply more precise generic and modal labels to their objects of study. “The notion that poetry is or ever was one genre,” Virginia Jackson argues, “is the primary symptom of the lyricization of poetry: the songs, riddles, epigrams, sonnets, epitaphs, blasons, lieder, elegies, marches, dialogues, conceits, ballads, epistles, hymns, odes, eclogues, and monodramas considered lyric in the Western tradition before the early nineteenth century were not lyric in the same sense as the poetry that we think of as lyric” (“Who” 183, emphasis mine). Jackson, along with Yopie Prins and Meredith Martin, among others, advocates a way “forward” for lyric studies that “look[s] back” to the “history of lyricization” (“Who” 186). In the view of these critics, the historical and formal emphases of the new lyric studies would help to remedy the “focus on modern and postmodern poetry” that “tends to occlude a broader historical range of poets who disturb and disrupt lyric utterance through various kinds of metrical mediation” (Prins 230). Moreover, this orientation would seem to allow a way for scholars to reclaim the study of poetry and poetics from some of the negative ideological associations that have been attached to lyric poetry in particular. If lyric is assumed to be subjective to the point of solipsism, a form of high culture that has removed itself from the concerns of politics and history and even from the pleasure of reading, then perhaps it would be best to let the lyric “I” speak to itself in some corner of the discipline, while we lend our ears to the voices of the other poetic forms and genres that the lyric has heretofore subsumed. 14 This twenty-first-century narrative of the ways in which the lyric came to dominate our conception of poetry often explicitly associates the ideology of lyric that it critiques with modernism or with the midcentury New Criticism. For example, Jackson argues that, in the case of Emily Dickinson, many twentieth-century critics “aligned themselves with modernist canons of taste specifically against the taint of nineteenth-century sentimental excess” (Dickinson’s Misery 209). More broadly, she notes that, “for New Criticism generally,” the lyric became “the test case, the zero-sum game, of literary interpretation” (93); in fact, “lyric became a metaphor for the New Criticism, in the sense that both the genre and the critical perspective on that genre came to stand for one another—so much so that the ahistoricism attributed to New Critical close reading became confused and identified with an inherent ahistoricism of the lyric genre itself” (93). I agree with Jackson insofar as we both seek a broader definition of lyric than the dominant post-Romantic strain and resist the ideological association of lyric poetry with New Criticism and ahistoricism. However, I think it is equally important to historicize the ideological charges against the lyric, which precede the reaction to New Criticism, as well as the alternate versions of lyric which have persisted alongside the narrower post-Romantic version. In fact, some of the claims of the new lyric studies have antecedents in precisely the movements that they align with the lyricization of poetry. For example, Jonathan Culler has linked the new lyric studies to Ren´e Wellek’s “Genre Theory, the Lyric and Erlebnis” (1970), in which Wellek argues that “[o]ne must abandon attempts to define the general nature of the lyric or the lyrical” (quoted in Culler, “Lyric” 884) and “proposes that we focus instead on describing particular genres, such as the ode, elegy, and song, their conventions and traditions” (“Lyric” 884). Further, as I point out in Chapter 4, modernist poets themselves criticized the vagueness of “lyric” as a category as well as the ascendancy of lyric poetry at the expense of other genres. Howard Blake, an American poet and critic of the early twentieth century, warned readers of his “Thoughts on Modern Poetry” (1935) that the “call for a pervading and persistent lyricism is both destructive and paralyzing” (190), and lamented the fact that “[t]he problems a modern poet must 15 confront seem to paralyze any wide assimilative powers”—i.e., the “positive, determined, ruling conception of life” and “universality” necessary for epic—“and cause, as escape, concentration on technique” (194). In “The Three Voices of Poetry” (1953), T. S. Eliot argued that “lyric” was an “unsatisfactory” term. The Oxford English Dictionary, which he cited in this essay, defined the term thus: “Now the name for short poems, usually divided into stanzas or strophes, and directly expressing the poet’s own thoughts and sentiments” (quoted by Eliot, 105). However, as Eliot noted, “there is no necessary relation between brevity and the expression of the poet’s own thoughts and feelings” (105). Many short poems called lyrics lack such subjective expression; many long poems include it; and yet this personal expression, in itself, may not be the defining mark of lyric either. Since Eliot’s time, the OED definition of lyric has not been made more specific. In fact, it has not been substantively changed at all, except to broaden the potential reach of the term by qualifying “short poems” with the clause “(whether or not intended to be sung).”6 Given the general lack of agreement about the boundaries of lyric as a genre and the negative associations that have accrued to it in the wake of the New Criticism (more on this later), the repeated calls for critics to historicize lyric studies and to focus on narrower poetic genres or subgenres have a certain appeal. But in this project, my aim is to draw attention to what might be gained by treating lyric as a transhistorical and transnational genre. Like Siobhan Phillips, I would argue that Jackson’s “focus on poetry’s cultural role could recommend a practice of increasing particularization that neglects some transcultural possibilities of verse writing” (23). As Phillips notes, “[c]urrent readers may be able to renew the study of lyric not only by dividing this label into the subclassifications of Jackson’s list. . . but also by expanding this label into the larger classification of poetry and thereby considering the rhetorical possibilities of verse itself” (23). My argument stops short of allowing lyric to dissolve into “the larger classification of 6 This change echoes another observation of Eliot’s, which immediately follows his charge that the term “lyric” is “unsatisfactory”: “We think first of verse intended to be sung. . . . But we apply it also to poetry that was never intended for a musical setting, or which we dissociate from its music” (105). 16 poetry”—although the poems in the chapters that follow may stretch the qualities of lyric, that designation is important to my reading of them—but it shares Phillips’s concern with “considering the rhetorical possibilities of verse itself.” In my use of the term “lyric genre” to describe a category of poems that can include a long modernist lyric like Crane’s The Bridge alongside Classical odes, metaphysical Renaissance poems, or nineteenth-century ballads, I rely on Jonathan Culler’s conception of genre in “Lyric, History, and Genre” (2009).7 Culler argues that the theory of genre has been divided into two camps—the theoretical and the empirical—and that this separation “creates the sort of confusion that contributes to the tendency to dismiss genres” (880). In the theoretical approach, genre is “based on some claim about elementary possibilities of thought, representation, or discourse”; by contrast, empirical genres are “lists of whatever genres people believe exist, some based on forms, others on content—classifications that do not seem very logical” (880). However, “[i]f one avoids the temptation to separate generic categories into the theoretical and the empirical but insists that genres are always historical yet based on some sort of theoretical rationale, they are more defensible as critical categories, essential to the understanding both of literature as a social institution and of the individual works that take on meaning through their relations to generic categories” (881). Here Culler cites as a model Ralph Cohen’s “process theory of genre” (Culler, quoting Cohen 216) in which genres are “open-ended and in a constant process of historical transformation” (Culler 882). When applied to the lyric, this theory allows critics to attend to the historical, cultural, and material contexts of specific poems while still considering how they function as part of a larger critical category: Observing particular shifts in the lyric does not. . . prevent one from maintaining a broad conception of lyric as a genre and its historical tradition. Conceiving of a broad 7 Another point of reference is Wai Chee Dimock’s “Genres as Fields of Knowledge.” However, in this essay Dimock considers genres to be purely empirical, defined only by their dynamic membership, and thus “resting always on some fluid continuum, with tributaries flowing into every individual instance” (1380). My contentions about genre are better aligned with Culler’s synthesis of the empirical and theoretical definitions of genre. 17 range of possibilities for lyric in many periods and languages can help prevent a certain narrowing of the conception of lyric and a tendency, understandable given the realities of literary education today, to treat lyric on the model of narrative, so that the dramatic monologue becomes the model of lyric. A broadly historical and, I would add, transnational conception of lyric enhances critical understanding. (885-886) This capacious and somewhat flexible conception of lyric can help critics resist the post-Romantic “lyricization of poetry” while also retaining a focus on those “important traditional aspects and potentials of lyric—especially those features, from rhythm and sound patterning to performative address, by which it differentiates itself from narrative fiction and mimetic modes generally” (896). It also, in my view, provides a better model for the kinds of transcultural and transhistorical work that the new modernist studies wants to promote. Modernism may be transnational, but it is not transhistorical; the lyric genre, however, is both. At its best, the new lyric studies promises to historicize the lyric by attending more closely to distinct poetic forms, meters, and genres in their social, cultural, and literary-historical contexts. In this respect, it shares Mutlu Blasing’s criticism of the kind of historicization that accepts a definition of lyric “as essentially a late-eighteenth- and nineteenth-century European invention,” a historicism that effectively “universalizes a historically and geographically specific model of a subject” (4). However, whereas Blasing provides a theoretical account of lyric poetry as a non-mimetic “formal practice that keeps in view the linguistic code and the otherness of the material medium of language to all that humans do with it” (2), a practice that is a “foundational genre” (4) across languages and cultures, the new lyric studies seems less interested in what lyric poems across time might share as formal practices than with how they differ from a specific post-Romantic notion of lyric. My project presumes that reading modernist lyric poems as particular historical instances of this transhistorical formal practice can tell us more about their temporal and ethical appeals than, for instance, apportioning them to narrower poetic subgenres 18 (e.g., separating modernist dramatic monologues from modified ballads) or classifying them with other experimental cultural productions (e.g., reading modernist verse alongside early twentieth-century print advertisements or silent films). To test this assumption, I focus on poems that push some of the traditional boundaries of lyric. In particular, three of the four chapters that follow consider long, multi-part lyric poems that encompass non-lyric (e.g. narrative, epic, or dramatic) elements. IV. The Long Poem and the Limits of Lyric I hold that a long poem does not exist. . . . That degree of excitement which would entitle a poem to be so called at all, cannot be sustained throughout a composition of any great length. After the lapse of half an hour, at the very utmost, it flags—fails—a revulsion ensues—and then the poem is, in effect and in fact, no longer such. —Edgar Allan Poe, “The Poetic Principle” This is great poetry. . . . But such a passage, by itself, is not In Memoriam: In Memoriam is the whole poem. It is unique: it is a long poem made by putting together lyrics, which have only the unity and continuity of a diary, the concentrated diary of a man confessing himself. It is a diary of which we have to read every word. —T. S. Eliot, “In Memoriam” What value did modernist poets place on the lyric as a genre, and what poems would they have recognized as belonging to this category? There is no question that the modernists, like many of today’s critics, associated the lyric genre primarily with short, non-narrative poems which expressed the emotions of the poet or of an imagined speaker. But their attitudes toward the lyric label were not unified, and often varied for individual poets over time. For example, early in his career, in “The Return of Ulysses” (1896), Yeats predicted a renaissance of dramatic verse which would emerge from the “obscurity into which all poetry that is not lyrical poetry has fallen” (146); later, in “The Symbolism of Poetry” (1900) and in “Poetry and Tradition” (1906), he wrote of lyric as a less developed form of poetry than epic or drama;8 in “Art and Ideas” (1913), looking 8 In “The Symbolism of Poetry,” Yeats framed the lyric as the primary stage in a sort of organic aesthetic evolution: “A little lyric evokes an emotion, and this emotion gathers about it and melts into their being in the making of some 19 back at his earlier verse, he lamented that “in our poems an absorption in fragmentary sensuous beauty or detachable ideas, had deprived us of the power to mould vast material into a single image” (255); and in his radio broadcast “Modern Poetry” (1936), writing again of that earlier moment, he declared that “[t]he aim of my friends, my own aim, if it sometimes made us prefer the acorn to the oak, the small to the great, freed us from many things that we thought an impurity” (91). Eliot shared this ambivalence about the aesthetic priority of lyric verse over other genres, especially vis-`a-vis verse drama, and he pointedly avoided applying the term to his own work. For example, in “The Three Voices of Poetry,” he suggested “meditative verse” (106) rather than lyric as the term that best described the “first voice” of poetry, “the voice of the poet talking to himself—or to nobody” (96). In “Yeats,” a retrospective essay delivered in 1940, Eliot praised his peer’s dramatic verse by declaring that “[t]he lyric poet—and Yeats was always lyric, even when dramatic—can speak for every man, or for men very different from himself. . . and it is only his imaginative power of becoming this that deceives some readers into thinking that he is speaking for and of himself alone” (253–4). However, he also noted that Yeats was a lyric poet “in a sense in which I should not think of myself, for instance, as lyric; and by this I mean rather a certain kind of selection of emotion rather than particular metrical forms” (254). This equivocation about the lyric, especially among those whom we would name as its accomplished modernist practitioners, indicates that, at the micro level, there was no broad agreement about its aesthetic value, and certainly none about its generic superiority. At the macro level, the critical trends that had constrained the lyric to “short poems, usually divided into stanzas or strophes, and directly expressing the poet’s own thoughts and sentiments” predated modernist and even nineteenth-century theorizations of lyric verse. Jackson, as noted above, great epic; and at last, needing an always less delicate body, or symbol, as it grows more powerful, it flows out, with all it has gathered, among the blind instincts of daly life, where it moves a power within powers, as one sees ring within ring in the stem of an old tree” (273). In “Poetry and Tradition,” he again argued for a hierarchy of literary genres, although he specified that such a system is always subjective: “We must not ask is the world interested in this or that, for nothing is in question but our own interest, and we can understand no other. Our place in the Hierarchy is settled for us by our choice of subject matter, and all good criticism is hieratic, delighting in setting things above one another, Epic and Drama above Lyric and so on, and not merely side by side. But it is our instinct and not our intellect that chooses” (210). 20 traces our contemporary “idealization” of lyric to Romanticism and to that movement’s tendency toward inwardness and abstraction. Mark Jeffreys, pointing specifically to the emphasis on shorter poems, locates the shift in the conception of lyric still earlier: The most important transformations of the usage of the term lyric occurred during the Renaissance, long after Aristotle and well before Hegel, when short inscriptive forms such as the epigram, newly distributed by printing presses and emblem books, mingled with Petrarchan song forms and then reemerged in the confusion of miscellanies and comprehensively titled individual collections of shorter verse forms. . . . From a welter of shorter poetic genres, lyric gradually emerged as the most common catchall category, and only in the nineteenth and twentieth centuries was it mythologized as the purest and oldest of poetic genres and thus transformed into a nostalgic ideological marker. (197) In these accounts, the abbreviation of lyric did not originate with modernism but was intensified by it for a variety of reasons. For example, literary movements such as Imagism required shorter poetic forms, emphasizing both subjective perception and the ephemerality of a single moment in time. In addition, the modernist reaction to nineteenth-century verse included an aversion to the seemingly bloated length of poems that “might have been expressed in prose” (Yeats, “Modern Poetry” 91) as well as critical defenses of comparatively neglected verse forms such as the “bright, hard precision” of the short metaphysical poems of the seventeenth century (Eliot, “Marvell” 167) or the “precise interpretive metaphor” of Guido Cavalcanti (Pound, “Cavalcanti” 162). And, from a purely practical standpoint, short lyric poems were better suited than epics for publication in journals and little magazines, and short poems or short excerpts of longer poems were easier for editors to anthologize in books.9 9 Of course, the issue of publication did not prevent poets from writing longer poems, and in fact allowed some of them to use (and profit from) the same work in different contexts. Many modernist poets who wrote long lyric poems published them in shorter individual sections as they completed them, before the entire work was finished and collected. This was the case for H.D’s Trilogy, Crane’s The Bridge, and Eliot’s Four Quartets. On the other hand, some 21 Given that many of the restrictive shifts in the definition of lyric preceded literary modernism, what could be the value in reading modernist poetry according to a broader conception of lyric—a conception that the poets under consideration might not recognize? If we grant that a modernist lyric poem need not be short, how does such a poem retain the attention and interest of readers without resorting to narrative? At what point does it cease to qualify as lyric? Moreover, given the presumed ascendency of the short lyric form, what might have driven modernist poets to disavow the lyric label, or to write long, ostensibly non-lyric poems? The answers that I offer in this project are twofold. First, I argue that a more encompassing idea of lyric verse is not ahistorical or anachronistic; rather, it has persisted alongside the narrower post-Romantic notion of lyric. Second, I borrow Culler’s notion of genre as both theoretical and historical. “Given the historicizing inclinations of criticism these days,” he writes, “it is important to stress that conceptions of genres are not just accounts of what people of a particular period thought; it is crucial to the notion of genre as model that people might have been wrong about them, unaware of affinities or ignoring continuities in favor of more striking novelties, or recognizing only an attenuated version of a larger tradition” (883). When we consider lyric in the context of its “larger tradition,” we may find that it includes narrative and dramatic elements, and precedes them generically; that its voices are not always consistent with a lyric I; and that it engages with time in ways that are not purely ahistorical or fixed in a perpetual present. It is my contention that, even when the notion of lyric is expanded to this extent, it remains distinct from the broader categories of poetry or verse. However, the line between lyric and non-lyric poetry becomes more difficult to draw as the length of the poem increases. Once the province of lyric has been opened up to include long poems, questions about the boundaries of the genre arise in many historical periods. For the purposes of my argument, one of the most relevant points of comparison is the nineteenth-century English long poem. In Narrative poets objected to the practice of anthologizing short excerpts from longer lyric poems that were already complete. In The ABC of Reading, Pound writes: “It would take a bile specialist to discover why the Oxford Book of Verse includes the first five of the strophes [of John Donne’s “The Ecstasy”] and then truncates the poem with no indication that anything has been omitted” (140, Pound’s emphasis). 22 Means, Lyric Ends, Monique Morgan examines four such works, arguing that each one “uses narrative techniques to create lyrical effects, effects that manipulate readers’ experience of time and shape their intellectual, emotional, and ethical responses” (2). Quoting a portion of Poe’s disavowal of the very possibility of a long poem, she asks why the form was popular with so many major poets of the nineteenth century. “One possibility,” she writes, “is that the epic may have carried a lingering prestige . . . . A second possibility is that poets were responding to a new narrative form. . . [i.e.] the novel” (16). In addition, while the novel was gaining critical and commercial popularity, lyric poetry had benefited from a “rise in prestige during the Romantic period” (15). The tension in trying to write a long lyric poem stemmed from the fact that, “[t]raditionally, lyric poetry is viewed as brief and intense, while long poems are assumed to have narrative interest to sustain their length” (15). Thus: “The author of a long work, especially one that requires many sittings to read, written in the lyric mode may find it difficult to maintain the illusion of a suspended moment of time. To write a work of this length in this period virtually requires generic experimentation and innovation. . . . Lyric had become more highly valued [than epic], the individual had become a more central focus, and modern times had encroached on the mythic past” (18). Morgan points out that, “[a]lthough contemporary narrative theory has expanded the range of texts it considers, poetry remains underrepresented and the novel remains the dominant source of examples and case studies” (2). Her analyses of Wordsworth’s The Prelude, Byron’s Don Juan, Elizabeth Barrett Browning’s Aurora Leigh, and Robert Browning’s The Ring and the Book help to rectify this exclusion by “qualify[ing] some generalizations that often are based too exclusively on the realist novel” and by “connecting specific uses of local poetic features (especially simile, personification, and alliteration) to larger narrative strategies” (2).10 Like Morgan, in the second chapter of this study, I examine some of the commonplaces of narrative theory in the context of a 10 What she calls “local poetic features”—instances of sonic repetition and figurative language—are features of literary language and of language generally, and what she calls “narrative strategies” are structures that are often at work in lyric. For the purposes of this chapter, however, I focus not on these differences but on our different conceptions of lyric time. 23 long lyric poem. However, unlike Morgan, I do not accept the traditional sharp distinction between lyric time and narrative time; rather, I use the concept of typological thinking in lyric verse—a kind of thinking that works at both the semantic and the sonic levels, and that requires readers to constantly negotiate between retrospection and anticipation—to point up similarities in the way that narratologists have theorized the temporality of the novel. For example, Frank Kermode finds a point of origin for his concept of modern “concord fictions” (59)—narrative fictions that “relate events that appear to be discrete and humanly inexplicable to an acceptable human pattern” (59)—in theological and literary-historical typology. In relation to the process of reading, Ricoeur proposes a similar structure, one in which the act of recollection disrupts the linear model of time and allows the reader to locate “the end in the beginning and the beginning in the end” (180). Peter Brooks, focusing on emplotment, observes that repetition (and divergence from repetition) “confound[s] the movement forward to the end with a movement back to origins” (108), such that readers, while ultimately driven toward the end of the text, are constantly referring to earlier moments in order to achieve “retrospective illumination” (108). Narrative theories such Kermode’s, Riceour’s, or Brooks’s are often presumed to be inapplicable to lyric poetry because lyric is associated with a timeless present, what Culler calls the “time of enunciation” (“Lyric” 887) and Morgan an “essential simultaneity” (12), while narrative takes place over time and creates meaning through the temporal organization of its plot. I have briefly addressed my objections to the binary view of lyric and narrative time, and will return to them in the next section. Here, I want to focus on another difference between lyric poetry and narrative prose, which is lyric’s resistance to plot summary or semantic paraphrase. Compared to the lyric, narrative lends itself fairly easily to these levels of summative description: e.g., even when a narrative unfolds non-chronologically—when the time of story does not match the time of discourse—one can analyze how meaning is created by the organization of events and actions in a particular order. Conversely, the fewer narrative aspects a lyric poem incorporates, the more difficult it is to paraphrase, and the more banal or bizarre is the paraphrase that results. This 24 is because, unlike prose narrative, lyric poetry is structured in verse, and the sonic relations in and between verses contribute to the effect of the poem and provide a counterpoint to narrative elements such as emplotment or syntactic assertions of cause and effect. These sonic or melodic relations resist any attempt at paraphrase (or, for that matter, translation), but they may structure the temporality of the poem in much the same way that narrative theory suggests the organization of events, actions, or images does for the novel. In lyric poems—and especially in long lyric poems, which often contain narrative elements—the sonic/semantic tension complicates our temporal reading experience even while it drives the verse we read. Giorgio Agamben has described this tension in terms of “metrical segmentation” and “semantic segmentation,” and posits that enjambment—the moment at which these two aspects lie in the balance, each awaiting resolution—“constitutes the only criterion for distinguishing poetry from prose” (109). In his reading, the end of the poem thus presents a crisis because it “would imply a poetic impossibility: the exact coincidence of sound and sense” (113). Simon Jarvis offers a similar argument with regard to long poems, suggesting that they are driven by two “compositional generators”: “line” (the line of verse, aligned with melody and sound) and “design” (“schemata” such as “plot, allegory, argument, and so on”) (“Melodics” 610). These terms cannot be mapped directly onto Agamben’s, because line is not identical to sound, and design does not strictly refer to sense or to narrative: “Both melody and thinking are at work in both line and design, as the intonational cues essential to argument, narrative, rhetoric, and syntax collide or go along with those held out by metre and rhythm” (612). However, much like Agamben, who frames the end of the poem as a confrontation between the patterns of sound and sense and as a “state of poetic emergency” (113), Jarvis argues that “every long poem ends in failure, a failure which is at the same time the condition of the possibility of its life,” because line and design both drive the poem and yet each “is murderously disposed toward the other” (“Melodics” 610). Jarvis’s argument about long poems is distinguished by his insistence upon the thinking 25 that is produced by the unit of the line as well as by the “melody” of the poem. This is in contrast to most narrative theory, which disregards poetry precisely because the formal arrangement of verses and the sound patterns within and among them are considered as, at best, a means of heightening the semantic emphasis of the poem and, at worst, mere decoration for the thinking that happens syntactically.11 For Jarvis, melody is not “the opposite of thinking,” but another kind of thinking—perhaps even “a medium for thinking and thinking about historical experience” (“Melodics” 611, 612). Moreover, “[m]elodic repertoires live and die in history, and part of their death is that tunes turn into signs” (615). That is, particular forms start as “real vehicle[s] for melodic transformation” and, after repeated use, become “markers” of a certain kind of poetry (615). The example Jarvis uses, William Collins’s “The Passions. An Ode for Music,” shows how a poem may refer to a given form—in this case, the Pindaric Ode—and also irregularize it: There is a way in which, in fact, this poem represents a kind of giant mono-stanza, where the overall shape—a curve away from regularity to irregularity and back again—leaves a more important impression on the reader than the shape of any individual stanza. Rhymes tie quatrains to irregular paragraphs. . . . It is, in fact, almost impossible to produce a rhyme scheme of this poem, because if you take the whole poem as a single unit, you run out of letters in the alphabet, but if you break it down into stanzas, you have no way of expressing its hardest-thinking rhymes, the metastrophic ones. (615–16) These sorts of metastrophic patterns appear in each chapter of this project, and are particularly relevant to my reading of the “Atlantis” section of The Bridge. Jarvis writes that “[t]he living long poem tends to destroy the metacommunicative legibility of its melodics, precisely in order to 11 The dismissal of the cognitive aspect of the verse form is not limited to narratologists, of course. Pound’s “Treatise on Metre” is emblematic of this line of thought: “Symmetry or strophic forms naturally HAPPENED in lyric poetry when a man was singing a long poem to a short melody which he had to use over and over. There is no particular voodoo or sacrosanctity about symmetry. It is one of many devices, expedient sometimes, advantageous sometimes for certain effects” (199–200). 26 recover them as an array of devices for thinking” (619). His paper eschews the term “lyric” and takes its examples from Classical and eighteenth-century English poetry, but his argument applies neatly to the ways in which modernist lyric poets adopt, distort, and transform various historical “melodic repertoires.” V. Lyric Time The experience of a poem is the experience of both a moment and of a lifetime. It is very much like our intenser experiences of other human beings. There is a first, or an early moment which is unique, of shock and surprise, even of terror (Ego dominus tuus); a moment which can never be forgotten, but which is never repeated integrally; and yet which would become destitute of significance if it did not survive in a larger whole of experience. . .” —T. S. Eliot, “Dante” There are essentially two critical positions on the nature of lyric time. The first presumes that lyric time differs from the time of other literary forms (e.g., narrative, epic, drama) insofar as it represents a sort of timeless present; the second most often retains a special temporality for lyric while describing poems that, for various reasons, represent exceptions to this lyric rule. One often-cited example of the first view is George Wright’s association of simple present verbs with the “lyric present.” The use of the lyric present tense, he argues, introduces a kind of abstraction that is particular to poetry; it “appears to offer as actual, conditions we normally accept only as possible, special, figurative, provisional” (569), and it “helps to elevate, to make not merely permanent but monumental and mythical that virtual experience we find at the center of the poem” (572). Jonathan Culler’s identification of apostrophe as the governing figure of lyric, and his consequent description of lyric as “the triumph of the apostrophic” (“Apostrophe” 66), share this underlying assumption about the lyric tense. In “Apostrophe,” which appeared three years after Wright’s essay, he proposes that if “the time of the apostrophe—a special temporality which is the set of all moments at which writing can say ‘now’”—is the time of lyric, then lyric time is 27 “a time of discourse rather than story” (66).12 Finally, in Lyric Time, Sharon Cameron argued that “the lyric voice is solitary and generally speaks out of a single moment in time. . . . The heart of the lyric’s sense of time might be specified. . . by its propensity to interiorize as ambiguity or outright contradiction those conflicts that other mimetic forms conspicuously exteriorize and then allocate to discrete characters” (23). In this sense, she noted, it might “be more accurate to say that its meanings are spatial rather than temporal. . . The lyric’s resistance to temporal extension creates the illusion of temporal wholeness, as well as of the temporal simultaneity of its moments” (Cameron 196). The view of lyric time as a time of simultaneity or as a resistance to the passage of time—in either case, as a kind of perpetual present—has been dominant in the decades since the 1970s.13 For the most part, the critics that offer alternatives to this theory of lyric time retain at least some aspects of its logic. For example, both James Longenbach and Heather Dubrow describe exceptions to the lyric present that function as such precisely because they represent hybrids of lyric and drama (Longenbach) or lyric and narrative (Dubrow). Thus, in Matthew Arnold’s “Empedocles on Etna,” which Longenbach describes as a “dramatical lyric, a lyrical drama,” there is a tension between lyric timelessness and the sequential and exteriorized conflicts that are characteristic of drama (Longenbach 848); in John Donne’s “Apparition,” a variety of narrative—a “ghost of future events”—distinguishes that poem’s relationship to time from the 12 Of course, de Man also associates apostrophe with lyric, but to far different effect. “What we call lyric, the instance of represented voice,” he argues, is characterized by, among other things, “the tropological transformation of analogy into apostrophe or the equivalent, more general transformation which, with Nietzsche’s assistance, we took as our point of departure: the transformation of trope into anthropomorphism” (“Anthropomorphism” 261). In his view, anthropomorphism (and thus lyric) arrests rhetoric in that it “freezes the infinite chain of tropological transformations and propositions into one single assertion or essence” (241). And lyric “is not a genre, but one name among several to designate the defensive motion of understanding, the possibility of a future hermeneutics” (261). 13 See, for example, Adam Potkay’s “Wordsworth and the Ethics of Things” (2008), William Harmon’s “English Versification: Fifteen Hundred Years of Continuity and Change” (1997), or William A. Madden’s “Framing the Alices” (1986). Robert von Hallberg’s (2003) phrasing is particularly forceful: “From Sappho to Pound,” he writes, “lyric poetry has committed like no other genre to moments of immediacy—not to narratives or continuity of any kind. Poetry is an art of resistance against all discourses that reach into the past and the future” (12–13). The consolidation of this view is part of what Jackson refers to as the lyricization of poetry. “At the risk of making a long story short,” she writes, “it is fair to say that the progressive idealization of what was a much livelier, more explicitly mediated, historically contingent and public context for many varieties of poetry had culminated by the middle of the twentieth century. . . in an idea of the lyric as temporally self-present or unmediated” (9). 28 lyric present as traditionally conceived (Dubrow 266). In the same vein, Morgan’s analysis of nineteenth-century long poems as products of the tension between lyric and narrative presumes that, in its purest form, “lyric is a suspended moment that stops the time of narrative and focuses instead on the ‘now’ of composition and reception” (4).14 A notable recent exception to both of these lines of criticism is Siobhan Phillips’s The Poetics of the Everyday, which argues that the midcentury American poets Wallace Stevens, Robert Frost, Elizabeth Bishop, and James Merrill use “particular but related means” to “convert the necessity of quotidian experience into an aesthetic and experiential opportunity” (1). Phillips links their poetics of “ordinary” everyday experience (3) to the process of repetition, which is implicated both in psychological self-fashioning and in aesthetic development (15). Moreover, repetition is itself a temporal phenomenon that is constitutive of poetic form: Ordinary verse emphasizes a fact so basic as to be easily overlooked: poetry is defined and constituted by its use of repetition. . . . Their [Bishop’s and Merrill’s] formalism might therefore show why and how a specifically poetic literature can realize the possibility of a commonplace life. Just as important, perhaps, their formalism might also prompt a salutary revision of generic definitions—definitions that often equate poetry and timelessness. (21) However, Phillips distinguishes the quotidian repetition of the poets she analyzes—their “ordinary rhythm” (4)—from “the common temporal extremes of stasis and change” that she sees as particularly endemic to an earlier modernist period. That is, the poetics of the everyday “does not seek the eternity of T. S. Eliot’s ‘still point’. . . but it does not accept the fluidity of Bergson’s evolutionary stream, either” (4). Poets at the height of modernism, in her view, responded to their 14 Models of lyric time that do not rely on generic hybridity to escape the timeless present tend to focus on poems that are formally experimental in some way. For example, reading Cole Swensen’s Oh as a poem that calls for a mode of “fractal reading” by virtue of its use of different scales and its fragmented visual presentation, Lynn Keller argues that, “in Swensen’s series of fractal lyric pages, the expanded present is in tension with references to dissolution and death. . . and/or it is disrupted by connections—syntactic, thematic, or sonic—to other moments of the text” (152). 29 “culture of time” in a binary fashion, by seeking either the self-effacing “destruction of sequence” (e.g., William Carlos Williams) or “the atemporal instant, the all-encompassing idealism of an all-powerful self” (9, 10). By contrast, the poets in her study “complicate both the enduring link of modernism and timelessness and the related, equally enduring connection of timelessness and subjectivity. . . [they] aspire to neither the potential solipsism of a private time nor the potential self-effacement of an external order” (10). Like Phillips, I propose an alternative to the equation of lyric poetry and the timeless moment. However, my readings reject the binary form of resistance to time that she reads into the poetics of early modernism. When Phillips writes that the poets in her study “complicate both the enduring link of modernism and timelessness and the related, equally enduring connection of timelessness and subjectivity,” she suggests that this reputation might be appropriate for the poets more commonly associated with high modernist aesthetics (e.g., Eliot, Williams, Pound). My project shares her interest in divorcing modernist lyric poetry from the qualities of timelessness and solipsism, but it undoes these associations in the poets with whom, for her, they seem to originate; in Chapter 2, for example, I specifically question the notion that Eliot’s Four Quartets aspires to the eternal rather than the temporal and that it can be opposed to Bergson’s dur´ee. Moreover, while Phillips offers “quotidian poesis” (3) as one thematically oriented, historicized alternative to lyric time, my conception of typological thinking considers the poems of H.D., Eliot, Yeats, and Crane as an especially salient body of work for a concept that might be extended to lyric poetry more broadly. That is, typological thinking is a model that better describes the temporality of lyric as a genre. This model is present in the poems in my study not because they offer special cases of lyric-epic or lyric-narrative hybridity—as might be argued in the cases of The Bridge and Trilogy, respectively—but precisely because they are lyric poems. Typological thinking is an engagement with time that lyric poems produce through the interaction of their semantic and non-semantic registers. This process involves both anticipation, as each verse points forward to a cumulative series of semantic units and formal fulfillments, and 30 retrospection, as each verse points back to a series of patterns visible and/or audible only in hindsight. Put differently, each word and each unit of sound in a lyric poem has the potential to function as both figure and fulfillment, and may play different roles in the linguistic patterns that construct meaning and the patterns that are formed by meter, rhythm, and sonic repetition over time. When Agamben singles out enjambment as the fundamental condition of poetry, he points to the moment when the poetic line is both most unstable and ripest with possibility—when, in the case of an end rhyme, there is a “disjunction between a semiotic event (the repetition of a sound) and a semantic event” (110). In other words, one phoneme may be the fulfillment of more than one figure, and it may also be the figure of a future fulfillment. Consider the first three stanzas of “To Brooklyn Bridge,” the prologue to Crane’s long, multi-part poem: How many dawns, chill from his rippling rest The seagull’s wings shall dip and pivot him, Shedding white rings of tumult, building high Over the chained bay waters Liberty— Then, with inviolate curve, forsake our eyes As apparitional as sails that cross Some page of figures to be filed away; —Till elevators drop us from our day . . . I think of cinemas, panoramic sleights With multitudes bent toward some flashing scene Never disclosed, but hastened to again, Foretold to other eyes on the same screen; [...] Reading these lines through the lens of the lyric present, we might note that “[t]he actions described seem suspended, removed from the successiveness of our ordinary time levels, neither past, present, nor future, neither single nor repeated, but of a different dimension entirely” (Wright 565). We can, after all, identify a solitary voice speaking in the simple present (“I think of 31 cinemas”); in the fourth stanza, this voice even performs the apostrophe promised by the poem’s title, calling on “Thee, across the harbor, silver-paced.” Yet the present tense in this opening is anything but simple: the first stanza introduces the idea of a future sequence of uncertain duration (“How many dawns . . . / The seagull’s wings shall dip . . .”), while the second proposes a break in this sequence (“Then, with inviolate curve, forsake our eyes”), a break described via a complex simile whose vehicle also points to an unmarked future moment (“Some page of figures to be filed away”). The disappearance of the seagull lasts “[t]ill elevators drop us from our day,” an action which implies a sequential pattern of work that complements or counterpoints the bird’s pattern of flight. We happen upon the simple present and the first-person voice only in the third stanza, where both are immediately complicated by the suggestions of repetition (a “flashing scene” that is “hastened to again”), prophecy (“foretold to other eyes”), and negation (“never disclosed”), and by the shift in scale from the “I” to the “eyes” of the multitude. Even if we were to presume that these various temporal references are all subsumed by the simultaneity of the lyric present, we would be faced with several elements traditionally ascribed to narrative. The poem is not purely apostrophic or meditative; time passes, and a compressed drama unfolds. Over the course of 11 stanzas, a “bedlamite speeds to [the bridge’s] parapets,” and, after “[t]ilting there momently, shrill shirt ballooning,” falls to his death; work carries on, as “the cloud-flown derricks turn”; and “dawns” give way to noon, afternoon, and the darkness of night. Here, what might be “simply successive” is conferred with meaning, “becomes charged with past and future” (Kermode 46); in other words, these verses organize meaningless temporal succession into the significance of fictional plot. The achievement of plot may feel all the more satisfying because, like many of Crane’s poems, “To Brooklyn Bridge” begins with semantic and syntactic opacity—Where is the subject and verb? Whose rippling rest is referred to? And what kind of rest is “rippled”?—as well as seemingly random sonic patterns. The first three pentameter stanzas flirt with alliteration and assonance within and across lines, internal rhyme, rhyming couplets, and an ABCB rhyme scheme; none of these patterns seems to govern the poem, and thus we continue to 32 listen for all of them in the expectation that one may be the figure of a future fulfillment (the ABCB scheme does, in fact, come to dominate, but not until the poem’s last stanzas). The idea of the lyric present is useful insofar as it insists upon the lyric as a discrete genre that engages with time in a different way than other literary genres do. Yet it is ultimately unsatisfactory because, even when the lyric present is described as the time of discourse—as a present that “repeats itself for readers in a future not even imagined” (Culler, “Lyric” 895)—it denies the ways in which lyric poems weave together various tenses and incorporate narrative elements such as plot. Conversely, while narratology can help explain instances of emplotment within lyric poems, and may be useful in describing the semantic aspects of typological interpretation, it cannot fully account for the ways in which lyric poetry works in time. Lyric often contains narrative, but it also exceeds the temporal engagement of narrative by virtue of its voice, which can speak from the time of discourse (but not only from the time of discourse), and by its form, which requires readers to look continually forward and backward, to identify and revise patterns of semantic and semiotic events. As Jarvis puts it, the argument of verse may be “generated out of melody and not simply poured into it” (“Melodics” 616). A novel may may engage in typological thinking in the sense that one event in the story may prefigure, in retrospect, an event that takes place later on, but such an event would not be equally available to a melodic pattern in the way that any phoneme in a lyric poem always is to both the semantic and sonic registers. To draw a finer distinction, a narrative poem may be generated by line and (plot-related) design, but without the lyric voice it lacks the ability to speak from temporal points outside of narrative. By virtue of the lyric “I,” lyric poems “justify themselves as invitations to imaginative participation within what the text elicits from its ways of bringing the world and psyche into language” (Altieri, “Lyrical,” 287)—ways that include temporal engagement. And this brings us to the question of what the ethics of the typological mode in lyric might be. VI. The Ethics of Modernist Lyrics 33 “We distrust poetry, whose rhythms already mark out our gestures and bewitch our every movement [. . . ]; we distrust everything that, in our lucid lives, is played out against our will.” —Emmanuel Levinas, Difficile Libert´e; trans. S´ean Hand, modified and quoted by Leslie Hill, 988-989. “[T]he poet strips the word of others’ intentions. . . . . Rhythm, by creating an unmediated involvement between every aspect of the accentual system of the whole. . . destroys in embryo those social worlds of speech and of persons that are potentially embedded in the word” —M. M. Bakhtin, The Dialogic Imagination, 297, 298. The idea that lyric poetry operates in a kind of perpetual present has contributed to a vein of criticism that views the lyric as anti-historical and, by extension, less ethical than narrative. From this critical position, modernist lyric—which is often characterized by the alienation of the subject, the aestheticization of politics, and an obsession with technical experimentation—is perhaps the least ethical genre of all. Friedman alludes to the suspicious qualities of twentieth-century, Anglo-European modernism when she asks: “Could it be that the anxieties about the geohistorical and generic expansion of modernist studies represents an uncanny desire to re-establish a particular early twentieth-century Western aesthetic style as the sina qua non of modernism? What is the ethics of that interminably repeated comfort zone?” (474). The first question is rhetorical; the answer to the second, one assumes, is not one that contemporary scholars should wish to perpetuate.15 From this point of view, a narrowly periodized modernism will always refer to closed, already-determined texts and will always carry with it the taint of an art-for-art’s sake that is formally associated with autocracy and imperialism. The problem of ethics for modernist lyric poetry is thus a multi-part problem. It involves, first, the ideology of modernism; second, the ideology of lyric; third, the relations that have been posited between these categories and Anglo-American New Criticism, a movement with which 15 In this sense, Friedman’s argument complements or extends some well-known critiques of the politics of high modernism, including Andreas Huyssen’s After the Great Divide, Raymond Williams’s The Politics of Modernism, and Michael North’s The Political Aesthetic of Yeats, Eliot, and Pound. 34 they have often been conflated; and fourth, their relation to postmodernism, a movement to which they have often been opposed. These issues are beyond the scope of this introduction, but I address some aspects of them here to clarify what I mean by the model of ethical reading that the typological mode in modernist lyric might offer. In the 1980s and 1990s, the reaction against New Criticism often included a reaction against the lyric poems that were often the object of mid-twentieth-century close readings. Directly opposed to both New Criticism and lyric, Mark Jeffreys argues, were the values of subversion and openness ascribed to postmodernism: In the context of the recent struggle to clear away New Critical poetics and to make room for a postmodernist poetics emphasizing openness in place of closure, intertextuality in place of the individual text, a fiction of polyvocality in place of the fiction of the “speaker,” and the appearance of subversive play in place of the appearance of ironic control, lyric became a metonymy for New Critical ideology and, in the extreme case, for all of Western literary authority since Plato. This trope depended for its rhetorical force on the New Criticism’s late-Romantic assumptions about lyric monologism and historicity, the same assumptions that were supposedly under attack. (203) More recently, Jennifer Ashton has addressed the connections between modernism, postmodernism, poetry, and New Criticism in the light of critical attempts to rehabilitate modernist poetry by ascribing to it postmodern values. She argues that contemporary critics have gone too far in this effort and that, contrary to their claims, “the modern/postmodern divide remains intact, both historically and theoretically” (2). In her view, modernism remains committed to the idea of a closed, autonomous text, one that does not invite readerly participation; postmodernism, by contrast, equates “experience with interpretation” (10). Like 35 Jeffreys, then, Ashton argues that postmodernism continues in the New Critical tradition, in this case by unwittingly subscribing to the intentional fallacy. Thus, “[i]f there are any postmodernists avant la lettre in the history of this trajectory [of twentieth- and twenty-first-century poetry], they are not Stein or (Riding) Jackson; they are I. A. Richards, William K. Wimsatt and Monroe C. Beardsley. . . . the legacy of the New Criticism is not modernism’s ‘autonomous lyric’ but postmodernism’s ‘open text’” (27). This reading would preserve the integrity of modernism as a historical period and aesthetic movement and dissociate it from the New Critical ideology, but it would deny it the openness and indeterminacy that have been granted to (some) modernist poets by critics such as Marjorie Perloff, Ron Silliman, and Jerome McGann. In addition to the ideologies often associated with lyric by virtue of its connection to modernism and/or to New Criticism, an ethics of modernist lyric poetry must confront the assumption that underlies its opposition to postmodernism: the idea that lyric poetry is a rigidly monologic genre which, by definition, denies the diversity of speakers and of types of speech within a language. This view stems in large part from M. M. Bakhtin’s comparison of poetry and the novel in The Dialogic Imagination, in which he argues that “[t]he language of the poetic genre is a unitary and singular Ptolemaic world outside of which nothing else exists and nothing else is needed. The concept of many worlds of language, all equal in their ability to conceptualize and to be expressive, is organically denied to poetic style” (286). In order to write a text that qualifies as a poem, he claims, “[t]he poet must assume a complete single-personed hegemony over his own language, he must assume equal responsibility for each of its aspects and subordinate them to his own, and only his own, intentions” (297). By contrast, “[t]he novelist working in prose (and almost any prose writer). . . . welcomes heteroglossia and language diversity of the literary and extraliterary language into his own work” (298). These assertions track nicely with the claim of the lyric present tense: one voice speaks from a single moment in time, and thus the lyric poem is doubly alienated from sociality and from history. Attempts to defend modernist lyric poetry against charges of its generic solipsism have 36 focused mostly on relations between subjects.16 These arguments help to refute the view of lyric as antisocial, but most of them ignore the assumption of lyric’s temporal isolation,17 which—just as much as assumptions about its romantic subjectivity—has contributed to the view of lyric as a genre that is inherently and systematically less ethical than narrative. If, as I have been suggesting, lyric has never functioned within the confines of the “lyric present,” and in fact requires a temporal engagement that is just as complex as (but distinct from) the operations of narrative or drama, then many of the objections to lyric’s supposed temporal isolation and ahistoricity immediately fall away. If we assume that lyric time is not defined as a perpetual present or as a grasping for the eternal, how might we reconceive the ethics of lyric? In particular, what happens when we assume that lyric is governed by a typological orientation? At this point, I must address the dangers of a certain kind of typological thinking as well as what I see as the promise of the typological mode in lyric. In the theological sense, typological interpretation offers a mode of non-logical, non-causal association that comes at the expense of historical narrative; this mode may easily be, and has often been, self-serving and totalizing in its revisions of the concrete historical past and historical narrative. The context of the first epigraph to this section, in which Levinas describes his “distrust” of poetry, underlines the links between this theological typology, poetry, and the most unethical of all historical interpretations. In 16 For example, Mara Scanlon addresses the Bakhtinian position directly, arguing through a reading of Robert Hay- den’s “Night, Death, Mississippi” that “dialogic poetry is possible” and that “a second dialogue, that between the poem and the answerable reader who attends the text, is implored, demanded, and even enacted by the lyric’s mobilization of voices and forms, including its use of call-and-response traditions” (2). Rachel Cole revisits lyric closure in order to ask whether there is “ethical potential. . . to be found in lyric strategies of economy and delimitation” (383). In the case of Wallace Stevens, she suggests, “the sonic intensity of lyric speech seems to represent. . . a force that can conciliate without resolving differences—not between sound and sense but between persons (actors and audiences, speakers and readers) who stand before one another as willful agents” (394). Starting from a similar theoretical position, James Phe- lan’s “rhetorical literary ethics” considers “the literary text as a site of multilayered communication between author and audience, one that involves the engagement of the audience’s intellect, psyche, emotions, and values” (630). 17 Two notable exceptions here are Blasing’s analysis of the historical “I” in lyric (e.g., “The special emotional and historical texture of the mother tongue that makes for the history of an ‘individual’ can return only in the specific language that produces itself on that ‘foundation’” (51)) and Culler’s emphasis on the performativity of the lyric present (“What lyrics demand of the world is often something to be accomplished by the performativity of lyric itself” (“Lyric” 887)). My argument complements such claims about lyric’s sociality, but it remains separate from the former (in attending to typological pattern in lyric rather than the historical “I”) and contradicts the assumption of the lyric present in the latter. 37 “Persons or Figures (On Emma¨us, by Paul Claudel),” the philosopher describes Claudel’s Christian typology as an non-logical system, “a state of logic that surpasses logic or precedes it” (120). When Claudel’s typological interpretation is applied systematically to the Bible, as it is in Emma¨us, the effect is stultifying and reductive: “If every pure character in the Old Testament announces the Messiah, if every unworthy person is his torturer and every woman his Mother, does not the Book of Books lose all life with this obsessive theme and endless repetition of the same stereotyped gestures?” (121). Such an application mistakes the nature of its object of interpretation and, in consequence, restricts human freedom: Sacred History is not the interpretation of a pi`ece a th`ese [problem play], albeit transcendent, but the articulation through human freedom of a real life Are we on stage, or are we in the world? Does obeying God involve receiving a role from Him or receiving an order? We distrust theatre, the petrification of our faces, the figure that our person weds. We distrust poetry, which rhythmizes and bewitches our gestures; we distrust everything which, in our lucid lives, plays on in spite of us. It is for this reason that the Claudelian exegesis ultimately disconcerts us. Man as a person, as an agent of history, seems less real to him than man as a figure or statue. (121, translation modified) Taken to the extreme, this line of interpretation justifies the most unspeakable of acts as the preordained fulfillment of a figure. “Does not Cain prefigure the Jewish people and Abel the Sacrificed Lamb? This is a courtly explanation of all our woes subsequent to exile, Auschwitz included” (122). I would take pains to differentiate what I call typological thinking in lyric from this kind of totalizing theological-historical typology. I borrow the concept of typology from Auerbach’s “Figura” precisely because this essay shows the development of the concept prior to its institutionalization as a model of theological and historical interpretation for the Christian church 38 and subsequent to it, as a mode of narrative in Dante. From its original, concrete meaning of “plastic form” (11) to its abstraction as a rhetorical figure, figura came to signify “any unilateral or indirect form of expression” (26). Already, before its appropriation by early Christian typology, figura was associated with the non-logical, unparaphrasable ways of meaning that we associate with poetry. Afterward, when theologians such as Tertullian and Augustine applied these ways of meaning to history—for example, relating two events in the Hebrew Bible and in the New Testament by an “accord or similarity” (29)—they retained in figural phenomenal prophecy a “firmly teleological view of history and the providential order of the world” (56). With respect to lyric, I think it is useful to return to that notion of a pattern that may be revealed in retrospect by non-logical accord without attaching to it a particular teleology. Every phoneme in a lyric poem is in this sense a figure charged with both symbolic and semantic fulfillment. If and when they are reached, these fulfillments may not coincide; moreover, they are always provisional, and themselves point forward and backward to other potential patterns. This continual revisability resists teleology even when, in the purely semantic register, poets signal typological and seemingly teleological readings of historical narratives. For example, the reverse chronology that H.D. employs in the fragmented narrative of The Flowering of the Rod, along with sonic associations such as “Mary” with “Magdala,” “Mara, bitter,” and “Mary-Myrrh,” and with the “Mother of Mutilations, / To Attis-Adonis-Tammuz and his mother who was myrrh” (The Flowering of the Rod 16), suggests a revision of Christian history and a cross-cultural realignment of myth that is enabled by a poetics of figure and fulfillment. Hart Crane, linking American rivers and cities to their analogues and namesakes in the Biblical and Classical worlds, connecting Native American history to Greek mythology to Whitman’s vision of the United States, and all, always, to “that great Bridge, our Myth, whereof I sing!” (“Cape Hatteras”), mythologizes his subject in part by lending it teleological significance. But even in these cases, the typological mode that governs both sonic and semantic patterns does not allow one stable or determined vision to dominate. That is also the price of figura, in the sense in which 39 I am using it: typological thinking allows non-logical patterns to proliferate but never to anticipate a fixed design. Fulfillments are also figures, and figures are always revisable in retrospect. If the typological mode offers an ethics for modernist lyric, then, it is an ethics unhitched from philosophy and logic. Here I follow the arguments of critics such as Altieri, Blasing, and Jarvis on the ways in which lyric poetry exceeds philosophy. For example, Altieri, responding to the ethical criticism of Wayne Booth and Martha Nussbaum, argues that “many of the values basic to [literary] experiences emerge in the modes of challenge and provocation, and not simply in exemplary cognitive judgments” (291). More broadly, Blasing writes that, “at the discursive level, the threat of poetry is not a threat of anarchy. . . . Rather, it is the threat of a different system underwriting—and, therefore, in effect overruling—the order of reason” (1). Jarvis argues that “[i]f there is to be a real relationship between philosophy and poetry, it must be one in which poetry can sometimes think better than philosophy, or at least think things that philosophy cannot” (“Unfree” 292). Structurally, typology describes the non-logical process by which lyric “connect[s] disparate human experiences” (Ramazani 283) without appropriating or exoticizing them. Altieri, relying on what he calls an ethos-based rather than an ethical criticism, proposes that lyrics be treated not as “examples of something that philosophy can clarify and help assess” but as “manifestations of the qualities and powers that establish what is possible within certain ways of engaging the world” (293). One of these ways of engaging the world is through time. The typological mode of lyric poetry calls for a constant renegotiation between past, present, and future; it requires readers to imaginatively occupy different temporal positions, and to consider simultaneously multiple temporal patterns. The ethics that it offers involves the demand to suspend our certainty about the past—to grant it a certain vitality—while also imagining what is possible in the future. Because it works formally and thematically, symbolically and semantically, this mode may even underlie lyric poems that seem to promulgate a closed narrative or a totalizing version of history. Attending to it requires our consideration of linguistic as well as 40 extra-linguistic patterns, including versification, rhyme, images, puns, rhetorical figures, syntax, alliteration, rhythm, and meter. Melody, as Jarvis has shown, constitutes its own kind of thinking. Considered as an element of typological thinking, even rhythm—one of the aspects of poetry determined to be most suspect by both Bakhtin and Levinas—may be a liberating and disruptive mode of thought rather than a beguiling constraint on the logical argument of a poem. VII. Chapter Summaries Chapter 2, “this will happen again—where? when?”: Typological Thinking in H.D.’s Trilogy,” considers the question of temporality in a long lyric work. Like many modernist poems, Trilogy dramatizes its own placement in history, engages in revisionary myth-making, and adopts the associative logic of Freudian psychoanalysis. But, in my view, its governing mode of thinking is typological; typology motivates the poem’s structure both semantically and sonically, and the temporal vision that results is generatively destabilizing rather than prescriptively closed. My argument for this typological mode draws on Erich Auerbach’s history of figural interpretation and Kenneth Burke’s analysis of the rhetoric of trope. Broadly, I argue that H.D. takes advantage of the paradoxical nature of typology—its assumption of predetermined structures of meaning, and its perpetual openness to revision—in order to make history available for rewriting. Following Agamben’s definition of poetry in “The End of the Poem,” I also suggest that typology may be a constitutive mode of lyric generally. The third chapter, “‘Every poem an epitaph’: Rhythm, History, and Genre in Four Quartets,” focuses on a poem that has often been read as the atemporal work of a conservative convert to the Church of England. Against this line of criticism, I read the model of time in Eliot’s Four Quartets as one of both simultaneity and radical openness. In this poem, Eliot transforms the insights he borrows from contemporaneous theories of time, including Bergsonism, to introduce a non-teleological, non-progressive typology in which “History may be servitude, / History may be freedom” (“Little Gidding” III, 162-163). This is not, I argue, simply a means of allowing us to 41 evade history, or to revise it to suit our own purposes. Rather, Eliot grants to the past a kind of viability that liberalism, in his view, does not; he takes it seriously, not just as something that has happened, but as something that is happening and will continue to happen, something that may have implications that we have yet to foresee. The fourth chapter, “Repetition and Revision in the Poetry of W. B. Yeats,” examines the temporality of Yeats’s poems as they relate to his revisionary practices: in addition to revisions made fairly soon after a poem’s initial publication, Yeats also revised many poems decades after he first composed them and altered their placement in his collected volumes. I read the temporal thinking in selected early and late poems in light of this practice and against some of his more famous political, historical, and ethical arguments in prose. Reprising questions of genre that arise in Chapter 3, this section provides readings of “Song of the Happy Shepherd,” which was originally composed as an epilogue for the verse play The Island of Statues; “The Lamentation of the Old Pensioner,” first published in 1890 and revised in 1895 and again in 1925; and “The Circus Animals’ Desertion,” which, perhaps more than any of Yeats’s poems, may be read as a meditation on the revisions—ethical and political as well as textual—that spanned his long career. “Hart Crane: The Bridge and the Limits of Genre,” the last chapter, takes up the perennial question of Crane’s success or failure as a poet in the context of his long poem The Bridge. I ask what difference it makes whether we regard this poem as lyric or epic—in other words, whether it succeeds or fails at satisfying the epic intention that early critics assigned to Crane. Why not, as some scholars have done, compromise by calling it a modern long poem or a “modernist epic”? The genre of the poem determines Crane’s place in the modernist canon and his place in literary genealogies (e.g., whether he is a late Romantic or an early language poet). Moreover, for many critics, it determines the poem’s relationship to history and the other, and provides support for reading Crane as either an essentially public or private poet. I argue that The Bridge both fails as an epic and succeeds in fulfilling Crane’s creative ambition by producing a defense of the temporal properties of the lyric. By returning to a sustained examination of the lyric genre, this 42 chapter also brings into focus the larger claim of this study: namely, that modernist lyric, rather than seeking to escape time, stages temporality in the reading experience in a way that highlights lyric’s ethical capabilities. C HAPTER T WO “this will happen again—where? when?”: Typological Thinking in H.D.’s Trilogy 44 In the modern view, the provisional event is treated as a step in an unbroken horizontal process; in the figural system the interpretation is always sought from above; events are considered not in their unbroken relation to one another, but torn apart, individually, each in relation to something other that is promised and not yet present. – Erich Auerbach, “Figura” It was not that he conjured up the past and invoked the future. It was a present that was in the past or a past that was in the future. – H.D., Tribute to Freud They make a group, a constellation, they make a groove or a pattern into which or upon which other patterns fit, or are placed unfitted and are cut by circumstances to fit. – H.D., Tribute to Freud I. Introduction Modernist lyric poems like H.D.’s Trilogy are often presumed to be both subjectively and temporally solipsistic. What could be more self-involved, after all, than the lyric’s commitment to the interior experience of a single speaker, its prescribed, delimited forms? Defenses of lyric poetry tend to answer this charge in terms of the ways in which lyric poetry might function as a protest genre1 or, more recently, in the ways it invokes relations between subjects, especially the relation between speaker and reader.2 But an equally important aspect of the hostility to lyric stems not from the isolation of the subject’s voice but from the poem’s limits in time. Implicit in the idea of the lyric’s refusal of the other is the suggestion that lyric temporality is similarly restricted. A speaker who consistently privileges her own interiority must also privilege her own 1 For example, Theodor Adorno argued that lyric opposes the dominant culture; as “a form of reaction against the reification of the world,” it “implies a protest against a social condition which every individual experiences as hostile, distant, cold, and oppressive” (quoted in Jeffreys 199). However, as Mark Jeffreys notes, “[t]he pragmatic political uses of lyric remain sharply circumscribed by the same assumptions about its essentially monological, subjective, introspective, and ahistorical nature that inform the models of lyric as an inherently reactionary genre” (200). 2 Rachel Cole summarizes several of these in the course of proposing an “ethical intelligence” in “lyric strategies of economy and delimitation” (383). In her reading of Wallace Stevens, delimitation helps broker an accord “based on a devotion of attention to the reader’s pleasure as well as the speaker’s, the other’s pleasure as well as one’s own” (396). 45 temporality; if lyric utterance is closed to other voices, it cannot be open to other experiences of time. Even defenders of lyric seem to confirm the idea of lyric’s temporal closure when they emphasize the particularity of the lyric “now,” a temporality that distinguishes the lyric utterance in part by distancing it from the time of narrative.3 These arguments for lyric openness thus simultaneously reinforce the notion of a closed, “presentist” orientation in lyric poetry.4 To reconsider the supposed closure of lyric, then, we must not only attend to how the lyric engages with the other, but also to how it exists in and engages with time. Much modernist lyric poetry seems to invite this kind of examination by treating time as a central problem, and it can be useful to approach the question of poetic temporality by asking whether such poems enact their arguments formally as well as semantically—or whether, in fact, there are different arguments at work. To understand these readings in the context of lyric poetry generally, we will need to ask broader questions as well. For example: How are the formal strategies of repetition and anticipation different from those that structure narrative plot? What is the relationship between logical and temporal priority in a given poetic figure? Is the idea of the lyric “now” adequate to the temporality of modernist lyric poetry? And, finally, how does the lyric think in time? That is to say, what role might temporality play in the kind of cognition that happens exclusively in verse?5 In this chapter, I consider these questions in the context of a long lyric work, H.D.’s Trilogy. Like many modernist poems, Trilogy dramatizes its own place in history, engages freely in mythopoesis, and adopts the associative logic of Freudian psychoanalysis. But its governing mode of thinking is typological; typology motivates the poem’s structure both semantically and sonically, and the temporal vision that results is centrifugal and generatively destabilizing rather 3 See, e.g., Jonathan Culler’s argument in “Why Lyric?”: “If narrative is about what happens next, lyric is about what happens now—in the reader’s engagement with each line—and teachers and scholars should celebrate its singularity, its difference from narrative” (202). 4 By this I mean to suggest not only that we should rethink the bounds of the lyric present, but also that we should consider how theories of the present might influence our reading of lyric. For example, a conception of the present as the only tense in which reality can be grasped has very different implications from one in which the instant of the present is so brief as to be unreal itself. 5 To address this last question, I will be drawing on Simon Jarivs’s work on cognition and verse prosody. 46 than prescriptively closed. In this respect, H.D.’s version of typology differs from the more rigid, dogmatic, or linear typologies of other modernist poets or of Christian exegesis. My argument for this typological mode draws on Erich Auerbach’s history of figural interpretation and Kenneth Burke’s analysis of the rhetoric of trope. I also follow Simon Jarvis’s proposition concerning cognition in verse: “If there is to be a real relationship between philosophy and poetry,” he writes, “it must be one in which poetry can sometimes think better than philosophy, or at least think things that philosophy cannot” (“Unfree Verse”). In what follows, I explore the thinking that results from Trilogy’s temporal mode. II. Trilogy’s Critical Contexts Trilogy foregrounds its formal and thematic concern with time in ways that are both common to the lyric genre and distinctively modernist. First, as its name suggests, Trilogy is made up of three books, and each of these is further divided into a set of 43 numbered poems, most of which are ordered in couplets; as a result, the tense of the lyric “now” is fragmented into a series of present moments that extend over 129 sections. Its epigraph, “for Karnak 1923 / from London 1943,” juxtaposes two significant places and dates in H.D.’s biography and signals the poems’ recurring interest in histories personal, literary, and mythological. In addition to the lyric present and to received and perceived accounts of the past, Trilogy occasionally gestures toward prophecy and, by way of its mythological subjects, toward a spiritual realm that transcends time. In Trilogy, when it comes to time’s form, the only stability is one’s interpretive position; in the “sub-conscious ocean,” the “Fish / move two-ways” (WDNF 4). This self-conscious grappling with temporality places H.D.’s work squarely in the high-modernist context of poet-critics like T. S. Eliot and Ezra Pound. Like “The Waste Land,” H.D.’s Trilogy is highly allusive; like the Cantos, it draws from a range of mythologies and is as comfortable forming association by metonymy and (imagined) etymology as by the more 47 traditional resources of metaphor and historical proximity. As much as H.D.’s poetic practice seems of a piece with this aspect of the modernist canon, however, it is in other ways alien to it. Her long poem differs from other canonical modernist works both in its approach to history and in its formal techniques. For example, Rachel Blau DuPlessis describes Trilogy as a poem “built in lyric sections of argument,” one that is “thoroughly syntactic, not (as in Pound’s Cantos) fragmented . . . . [it is] a major serial poem of meditation and cultural assessment” (“Revisionary” 118). Susan Gubar claims that H.D.’s “palimpsestic or encoded revisions of male myths” serve as a response to the inheritance of “uncomfortable male-defined images of women and history” (298). These techniques gesture toward a “magical potent center” that the poet seeks “because it can serve as an escape from the repetitions of history and because it is an entrance to creative stasis. Such a center is neither temporally nor spatially bound” (313). In these well-known accounts, the poet of Trilogy either recuperates the past from a female perspective or transcends it in order to create as a female artist. These late-twentieth- and early-twenty-first-century interpretations of H.D.’s poetics point to the importance of the poet’s gender to her approach to history and to her use or rejection of historical material. This aspect of her craft is indeed significant, and is one reason that feminist scholars have both championed her poems and emphasized their resonance with her biography. Taken to the extreme, however, the emphasis on gender can detract from features of H.D.’s poetics that place it among the modernists for different reasons. Even as her project in Trilogy involves a kind of destabilization of history and narrative, its implications extend much further than the gendered subversion or revision of masculine myth. If her oeuvre is not completely consonant with the poetry and criticism of her male contemporaries, neither is it necessarily antagonistic to their work; in fact, it may be particularly productive to reexamine her poetry in relation to a shared inheritance of ideas about the poet’s confrontation with literary history. In this chapter, I propose that typological thinking—which manifests in H.D.’s revisionary myth making, the palimpsest, “recovered” etymology, and other forms of wordplay—is the mode that best 48 articulates this relation to history and the material world. Although I suggest that typology is the essential structuring principle in Trilogy, typological readings have not been dominant in the H.D. criticism. Instead, critics have tended to focus on individual strategies that, I believe, contribute to this overarching typological mode. One of these is H.D.’s mythopoesis, her gendered rewriting of traditional, masculine-centered myths. DuPlessis argues that, in general, H.D.’s “mythopoetic works made self-consciously critical interpretations of major cultural myths involving gender” (“Revisionary” 124). Moreover, Trilogy’s rehabilitation of the female figure in myth “shows the emergence, in [H.D.’s] post-Freud writing, of massive female symbols, resolutions and myths, of women-identified quests, and of female heroes who act in narrative making explicit critique of phallocentric culture” (“Gender Authority” 76). In a similar vein, Susan Acheson claims that, in Trilogy, H.D. attempts to “revise the eschatological plots of traditional Christianity,” incorporating “esoteric theories relating to astrology and to the myth of Atlantis” as well as “the suppressed Franciscan eschatological theory of the ‘eternal gospel’ and the ‘third age’” (188–189). The revisionary aspect of H.D.’s use of myth, rather than the typological relations that she establishes, has been at the center of this line of thought.6 Another strategy that has garnered much attention is H.D.’s “lifelong fascination with the palimpsest” (Gubar 298). Literally, the palimpsest refers to a surface on which markings have been repeatedly inscribed, erased, and/or written over, and which thus bears the trace of multiple texts or images. Conceptually, it allows for associations between words based on “traces” that 6 More recent work in this vein has focused on the affinities of H.D.’s revisionary myth with early-twentieth- century models of archaeology, anthropology, and politics. In “Thesmophoria: Suffragettes, Sympathetic Magic, and H.D.’s Ritual Poetics,” Edward Comentale reads H.D.’s poetry in the context of contemporary feminist movements that “turned to classicism as a viable form of aesthetic politics” (472), and suggests that her “ritualized” poetry “reveals the effective unity of often gendered binaries, such as subject and object, public and private, work and world” (484). In “Egypto-Modernism: James Henry Breasted, H.D., and the New Past,” Marsha Bryant and Mary Ann Eaverly connect H.D.’s attention to and interest in Egypt with the Egyptology of the archaeologist James Henry Breasted, whose “‘New Past’ . . . positioned the United STates as Egypt’s cultural inheritor” (435). They argue that “H.D. came to share with Breasted a belief in Egypt’s cultural primacy. And like Breasted, she would seek a monumental past that did not simply reinshrine the traditional Greco-Roman one” (435). I would agree that “Egypt’s ancient archaeological sites— especially Karnak—gave H.D. and Breasted “a generative space in which to create new pasts unencumbered by the weight of traditional Eurocentric frameworks” (450), but I would also suggest that in Trilogy, this “generative space” is opened up less by Breasted’s “New Past Theory” than by the mode of typological thinking that governs the poem. 49 need not be related by subject, geographical or temporal origin, or scholarly etymology. In H.D.’s work, part of the function of the palimpsest is to disrupt historical continuity by linking words from different epochs, but the concept is also responsible for revealing certain linear histories that at first appear to be hidden; it may also structure the reader’s experience, as “the immediate impression of a stable pattern that is then overlaid with the far more problematic effects of actually interpreting the text” creates “a kind of palimpsest. . . of readings that shape our response to the poem” (Graham 163). Typological thinking links these dual projects of rewriting and recovering the myths and narratives of the past while simultaneously pointing to a future fulfillment that is always yet to be realized. A third category that might be said to govern Trilogy’s structure is the dream logic of psychoanalysis. In the Interpretation of Dreams, Freud’s explanation of the dream-work includes direct comparisons to artistic composition in general and to language in particular. H.D., a student of Freud, was also his analysand; as she was composing parts of Trilogy, she was simultaneously writing Tribute to Freud, a remembered account of her sessions with Freud in Vienna in the 1930s.7 The formal associative connections that drive Trilogy, whether at the level of concept or sound, have direct parallels in the psychoanalytic concepts of displacement and condensation. Thematically, too, critics have read Trilogy as a personal, psychological response to the trauma that H.D. endured in London during World War II (which in turn recalled the trauma of her experience during Wrold War I).8 But again, in my view typological thinking subsumes the logic 7 In Brenda S. Helt’s reading, Tribute to Freud shows that H.D. interprets history primarily in a psychoanalytic vein. My aim here is not to contradict this analysis of H.D.’s prose theories of time; in fact, her lyric may sometimes accord with such theories. However, lyric temporality encompasses more than the content of a prose argument. 8 See, e.g., Julie Goodspeed-Chadwick, who argues that “H.D. writes corporeality and female identity into war narrative, performing female embodiment within a site of trauma” (59); in particular, Trilogy “investigates the com- ponents of trauma and exclusion that pertain to women at large during wartime, while imagining remedies through poetic witnessing and a community of woman-centered spirituality and healing” (68). Like Goodspeed-Chadwick, Sarah H. S. Graham views Trilogy primarily as a wartime poem and as a response to war trauma, but her analysis places less emphasis on a specifically gendered response. Moreover, in Graham’s reading, Trilogy is not a poem that “imagin[es] remedies through poetic witnessing”; rather, The Walls Do Not Fall functions as a “complete response to war” (180), Tribute to the Angels as “a performative peace poem” (193), and The Flowering of the Rod as “a poem of war-weariness and depression” (194). In my view, the wartime context of the poem contributes to the significance of its typological mode, which allows the poet to confront the force of history by showing it to be revisable. Graham recognizes this operation in Trilogy, 50 of psychoanalysis as well as the poem’s biographical basis in war trauma. For all of its dealings with the past, Trilogy is not limited to the experience of a single life; rather, it points beyond the present to the future and to a divine realm that operates beyond measured time. Finally, my reading of H.D.’s poetry as typological shares some basic assumptions with Adalaide Morris’s recent book, How to Live / What to Do: H.D.’s Cultural Poetics. Like Morris, I believe that H.D.’s poems are instructive not just for “what” but also for “how they think” (2), and that we ought to read them as “acts of cognition” (10). Also like Morris, I argue that reading the poems this way requires an explicit attention to sound.9 However, where Morris is concerned mainly with the cultural work that the poems perform, and often explains their thinking in terms of the visual field, I focus on the explicitly temporal dimensions of these acts of cognition.10 These dimensions may be considered independently of their links to specific cultural events and materials, but they always retain an ethical dimension insofar as they require the reader to imaginatively occupy different temporal positions and to anticipate future verses even as the relations within and among the poems are constantly revised. Broadly, I want to argue that the although she does not name it as such (“the practice of ‘reading back’—in which earlier passages are enriched or clarified by later ones—is tacitly encouraged” (164)), and she connects the way in which “H.D.’s poets can see meaning in random events” to the historical “interpretation” of wartime bombings: “it was almost convention during air raids to assign a specific purpose to the enemy bombing that it simply did not have. This meant that the actions of the pilots—who thought themselves lucky to hit anything at all—were considered part of a mysterious plan, and the bombing of one place rather than another seemed to observers to have a concealed significance” (178). The historical parallel is fascinating, yet it need not suggest that Trilogy’s mode has a “peculiarly war-related significance”; rather, the typological orientation of the poem—which suggests an interpretive strategy that precedes the technology of bombs and airplanes—encompasses wartime events and also points beyond them. 9 As she puts it: “To foreground sound in poetic language is to engage a proliferation of meaning that cannot be halted, an ongoingness of signification that exceeds binary oppositions and resists closure . . . . however, this openness is not a rejection of communication. . . but rather a generation of conceptual links through which poets and readers access, activate, refresh, and sustain the cultural materials through which crises are understood and addressed” (8). My emphasis in this chapter is on the ways that sound may generate conceptual as well as non-conceptual links that both “resis[t]” closure and seem to ordain a certain structure in retrospect. 10 For example, in Chapter 6 Morris links H.D.’s Tribute to the Angels and Nathaniel Mackey’s Djbot Baghostus’s Run by noting that both are “constructed stereoscopically”: “They work, that is, like an optical instrument that creates three-dimensional illusions by bringing into a single focus photographs of the same scene taken from slightly different angles” (185). In Chapter 3, she describes the “alchemical” aspect of Trilogy by noting that each “image” in the poem “is a cell that embodies the code of the whole”; “[t]hese images prepare our understanding of the poem’s larger sweep” (113). My reading is most consonant with Chapter 2, in which Morris points explicitly to the “aural glyphs or phonograms” (42) that “play out for the ear as well as the eye” (43), but her focus is less on the temporal orientation of sonic play than on the way in which it organizes concepts such that the poet can “use words to think culture away from catastrophe” (49). 51 sonic play and recurrence of key terms that Morris identifies in H.D.’s body of work in general can be read, in Trilogy, through typology: this is how H.D.’s wartime poem thinks. III. Typology and Modern Lyric Figural typology, whether applied to a historical or a literary text, necessarily involves a particular kind of meaning-making from a particular point of view. In his long essay “Figura,” Erich Auerbach provides a description of this mode of interpretation that will help situate the discussion of typology in Trilogy that follows. The essay is divided into four parts, the first of which is devoted to tracing the meanings that accrue to the term figura in classical rhetoric. The Latin word, which “originally meant ‘plastic form,’” shares “the same stem as fingere, figulus, fictor, and effigies” (11). Over the course of the last century B.C.E., it came to acquire more abstract meanings: “side by side with the original plastic signification and overshadowing it, there appeared a far more general concept of grammatical, rhetorical, logical, mathematical—and later even of musical and choreographical—form” (15). Finally, “by the end of the republican era, figura was firmly ingrained in the language of philosophy and cultivated discourse, and during the first century of the Empire its possibilities continued to develop. As one may well imagine, it is the poets who were most interested in the shades of meaning between model and copy” (21). Ovid, in his poetry of metamorphosis, was perhaps the most adept at exploiting these “mobile, changeable, multiform, and deceptive” qualities of figura (23). But at this point, the mutable potential of figura did not yet extend to events ordered in time. The ways in which poets could apply the concept of figura expanded significantly in the early centuries C.E., when the term came to describe a specific form of historical and textual interpretation. In the next three sections of his essay, Auerbach examines the typological sense that figura takes on in the early Christian church, the development of typological (or figural) interpretation in Christian thought, and the application of the figural view of history to art in the 52 Middle Ages. For my purposes, the most important sections of Auerbach’s essay are the two that center on the history of typology. He locates the origin of “the strangely new [typological] meaning of figura” in the writings of Tertullian, the early Christian writer. In his work, “the naming of Joshua-Jesus is a phenomenal prophecy or prefiguration of the future Saviour; figura is something real and historical which announces something else that is also real and historical. The relation between the two events is revealed by an accord or similarity” (29). Typological interpretation thus casts a first historical event, usually located in the Hebrew Bible, as a figure; the second, situated in the New Testament, is its fulfillment. The connection between the two is established by virtue of the mode of interpretation itself, which need not have any correspondence with temporal or causal logic. Rather, “the vague similarities in the structure of events or in their attendant circumstances suffice to make figura recognizable; to find it, one had to be determined to interpret in a certain way” (29). In Auerbach’s view, this method of reading history predominates in Europe from the age of the first evangelists, who were armed with the typological logic of the Pauline epistles, until the end of the Middle Ages. In Western religious and political thought, its influence extends well into the modern era. As a mode of poetic composition and interpretation, typology has important implications for modernist poetry in general and H.D.’s Trilogy in particular. Broadly, because typology encompasses both historical and textual analysis, it enables a way of thinking history and language together that upsets assumed orders of events and words. Despite its disruption of empirical models, traditional typology also insists on the materiality of its concepts. Thus, in Christian figural thought, “the fulfillment is often designated as veritas. . . and the figure correspondingly as umbra or imago; but both shadow and truth are abstract only in reference to the meaning first concealed, then revealed; they are concrete in reference to the things or persons which appear as vehicles of the meaning” (Auerbach 34). This interpretation, in turn, depends upon the double valence of the concept or event, which exists simultaneously in linear time and in the timeless space of the eternal.11 As a result of this mobility, events in the typological view of 11 This aspect of figural interpretation intersects in interesting ways with psychoanalytic criticism of H.D. In “H.D.’s 53 history are never fixed: history, “with all its concrete force, remains forever a figure, cloaked and needful of interpretations”; it “remains open and questionable, point[ing] to something still concealed” (58). In the modern, empirical view of history, only the interpretation of the event is subject to change; the event itself retains a static character and fixed temporal location. Typological thinking turns this relationship on its head. Wrenched from its empirical context, “the fact is subordinated to an interpretation which is fully secured to begin with: the event is enacted according to an ideal model which is a prototype situated in the future and thus far only promised” (59). Extended to the literary realm, the juxtapositions and rearrangements that typology allowed made it a particularly productive mode for modernist poets who sought to break with the immediate past while claiming connections to other, more distant traditions.12 The concept of the figure, with all of its attendant connotations, is especially vital to an understanding of H.D.’s work. Like many of her contemporaries, she was concerned with revising a certain version of the historical narrative and finding new historical markers—or a new order for old ones—by which to understand modern experience. When one considers the positions that Trilogy suggests in terms of human and literary history, it becomes apparent that neither of these categories can be regarded as sufficient; neither can contain the potential meanings that lie latent in the word. For H.D., words have a particular power because they refer both to things in the world and to sacred truths, and the relation between the two realms need not respond to logic or reason. Speech contains the shattered “secret” of the gods (WDNF 8), hand words retain this potency even when their original meanings are veiled. Moreover, the promise of the word is not Helen in Egypt: Aging and the Unconscious,” Brian Brodhead Glaser argues that the central question of Helen in Egypt could be formulated thus: “How does one integrate this inner life that has never been marked by time, that has always seemed to be at once ancient and undying, into a conscious life shaped by one’s sense of inevitable finitude and the consequent interest in retrospection, understanding, and historical synthesis?” (101). Of course, we cannot equate the unconscious with the eternal, or the conscious with the event in time. But the attempts to navigate between these realms, or to integrate them in some way, seem to call for a similar mode of poetic thinking. 12 For instance, typological thinking strongly influences the structure of Pound’s Cantos. In “Designs on History,” Mutlu Blasing argues that Pound extended Puritan typology to the literary realm as a way of laying claim to literary authority: “The Puritans and Pound alike figure radical breaks with immediate precedents as returns to ‘purer’ origins, which they think to access through certain texts” (4). In this sense, typology also offers Pound “a model of intratextual coherence that does not do violence to the Cantos’ open form” (8). 54 confined to the past. Rather, past traces can be prophetic, awaiting the poet who will reveal their origins and bring about their fulfillment. In other words, Trilogy depends upon a kind of typological thinking for both its structure and meaning in which “facts” are subordinated to an interpretation which is always in process and hence always revisable. This sense of perpetual revision, which emphasizes the necessity of adjusting one’s expectations as well as one’s sense of the past, sets H.D.’s typological mode apart from other modernist poets and from the sort of Biblical exegesis in which a stable telos determines the interpretation of a given event. In The Walls Do Not Fall, the first book in Trilogy, H.D. suggests such a typological relationship beginning with the epigraph: “for Karnak 1923 / from London 1943.” The poet had visited the Karnak temple complex in Egypt, in 1923; twenty years later, she began her long wartime poem in a London devastated by the Blitz. The first section in this book sets up these two locations as figure and fulfillment in a relation revealed by physical resemblance: there, as here, ruin opens the tomb, the temple; enter, there as here, there are no doors: the shrine lies open to the sky, the rain falls, here, there sand drifts; eternity endures: Although figural prophecies are almost by definition visible only in retrospect, the figures themselves are “prophecies of something that has always been” (Auerbach 59). Writing for the past, from a war-torn present, the poet acknowledges a relationship that has always existed but has only recently, temporally, been revealed. Immediately preceding these stanzas, the “Luxor bee, chick and hare . . . continue to prophesy / from the stone papyrus”; following them, the appearance of Samuel, the Biblical prophet, and “the Pythian,” the prophetess at the Temple of Apollo at Delphi, emphasize the prophetic relation that the poet posits between the sacred Egyptian temples and the English metropole. 55 Notably, this first section also closes with a meditation on the original sense of figura, as “plastic form” (Auerbach 11). These stanzas extend the poet’s meditation on history and set up the temporal explorations of the following poems: the bone-frame was made for no such shock knit within terror, yet the skeleton stood up to it: the flesh? it was melted away, the heart burnt out, dead ember, tendons, muscles shattered, outer husk dismembered, yet the frame held: we passed the flame: we wonder what saved us? what for? The images of “husk” and “frame” recall H.D.’s description in a letter to Norman Holmes Pearson, also in 1943, of the past as a “skin or husk” that one must periodically cast off (Pearson 28). What remains, in both instances, is the essential form, frame, or mold, which may express a series of different, perishable shapes, but which will endure unaltered; put another way, the essence is stable even as the aspect fluctuates. Auerbach’s etymological examination of figura is tellingly relevant here: “strictly speaking, forma meant ‘mold,’ French ‘moule,’ and was related to figura as the hollow form to the plastic shape that issues from it” (13). In Trilogy, the original sense of figura lends force to the figural (or typological) relationships between images, words, and sounds within and across the sections in each book. The two-way potential of figura in Trilogy is perhaps best represented by a series of synonymous images of containment which includes the seed, the box, and the chrysalis. In section 25 of The Walls Do Not Fall, the speaker describes how her “heart shell / breaks open” long after “your bennu bird / dropped a grain / as of scalding wax” (H.D.’s emphasis). As “scalding wax,” the object suggests the capacity to take the imprint of the past, to reproduce the image of what has 56 come before. As a grain, however, it points to a mutable, generative future. When such a grain has “taken its nourishment” (25), it will reveal its binding, predetermined form. Within the three-part form of Trilogy, the grain demonstrates this potential as it returns in the last book. (It is worth noting here that Trilogy was not originally conceived as a three-part poem. In this sense, H.D. “fulfills” the promises made in The Walls Do Not Fall only with the benefit of hindsight; when the first poem was composed, the subsequent books were not yet envisioned.) In The Flowering of the Rod, when the Lady comes before Kaspar, he sees “a grain, a flaw, or a speck of light” and “the speck, fleck, grain or seed / opened like a flower” (30). At that moment: . . . the flower, thus contained in the infinitely tiny grain or seed, opened petal by petal, a circle, and each petal was separate yet still held, as it were, by some force of attraction to its dynamic centre; and the circle went on widening and would go on opening he knew, to infinity. (31) In the context of Trilogy, this moment fulfills the promise latent in the seed in the first two long poems. More broadly, these lines speak to the “infinite” history and capacity contained within “infinitely tiny” images, objects, or words. Earlier, in The Walls Do Not Fall, the speaker asserts: “. . . I know, I feel / the meaning that words hide; / they are anagrams, cryptograms, / little boxes conditioned / to hatch butterflies. . . ” (39). The word, as seed, contains an infinitely expansive potential; it carries the inheritance of the past along with the capacity to represent present and future realities.13 13 The almost dizzying concentration in such moments is balanced, as Acheson suggests, by the poem’s relatively 57 Alongside this set of rhyming images, H.D. emphasizes the disruptive and constructive functions of typology through repeated allusions to characters in Biblical, Classical, and Egyptian myth. Partly, these references serve to revise the gendered qualities of western mythology. But the implications of these newly forged juxtapositions extend beyond the feminist perspective. Typological thought allows H.D. to weave mythologies from different historical periods, languages, and cultures into a universal narrative with broad creative potential. Section 41 of The Flowering of the Rod—in which Kaspar contemplates the significance of his past decision to separate a pair of jars—offers an explanation of how the figure works: . . . we are permitted to wonder if it had possibly something to do with the vow he had made— well, it wasn’t exactly a vow, an idea, a wish, a whim, a premonition perhaps, that premonition we all know, this has happened before somewhere else, or this will happen again—where? when? (H.D.’s emphasis) Elsewhere in Trilogy, the poet seems to explicitly address the status of the figure in time. In The Walls Do Not Fall, she asserts an equivalence between figure and fulfillment that denies any temporal effects: “In no wise is the pillar-of-fire / that went before / different from the pillar-of-fire / that comes after” (36). As in traditional typology, however, this equalizing move does not efface the materiality of the event. Rather, . . . every concrete object short lines and frequent spaces: while Trilogy sometimes seems to “obsessively enunciat[e] itself, spinning out words and adding more and more lyrics” (191), it “is always open to the opaque and destabilizing other, the unsettling silence, the space after the end” (193). 58 has abstract value, is timeless in the dream parallel whose relative sigil has not changed since Nineveh or Babel. (15) The “Dream,” which we can perhaps equate to spiritual understanding in Auerbach’s model, “acts as go-between, interpreter” and “merges the distant future / with most distant antiquity” (20). In Tribute to the Angels, section 24 explores the figural idea that events in time may be both chronological and eternal: Every hour, every moment has its specific attendant Spirit; the clock-hand, minute by minute, ticks round its prescribed orbit; but this curious mechanical perfection should not separate but relate rather, our life, this temporary eclipse to that other . . . Because typological figuration fixes meaning in the realm of the eternal, the order of events in temporal chronology need have no bearing on their order in poetic narrative. In broad strokes, The Flowering of the Rod recounts a series of events beginning with Jesus’ birth and ending with his death and resurrection (and, possibly, the second coming). But because the stronger relation between these events is figural, not temporal, the order in which the events are recounted moves back and forth in a direction that starts from the future and points to the past.14 The first poem evokes an apocalyptic scene in which the select “we” will “leave the smouldering 14 This overall figural scheme may accommodate more specific allusions or resonances. For example, Acheson proposes that H.D. may be alluding to the argument in a book she owned and annotated, H. S. Bellamy’s The Book of Revelation is History, in which “Bellamy argued that the events recounted in Revelation were memories of geological catastrophes such as had drowned Atlantis, and that the last book of the Bible ought to be placed first, followed by Genesis which was really an account of re-creation” (198). 59 cities below.” Meditations on resurrection lead back, in section 11, to the scene of crucifixion. In the next section, Mary Magdalene washes Jesus’ feet with her hair and the substance from the “alabaster jar”; in section 13, we move back yet again, to her original attempt to acquire this substance. Sections 21 to 26 take us slightly forward, again to the gathering and the washing of feet, while sections 27 to 34 circle back in an extended meditation on Kaspar’s mystical moment of recognition or revelation when Mary lets down her hair. Section 43, the last in the book, concludes with a memory of Kaspar before the other, Virgin Mary, at the scene of Jesus’ birth. The jar that he offers her here is both figure and fulfillment of the jar that the other Mary will bring to her son, who is himself a third manifestation in the series (“the fragrance came from the bundle of myrrh / She held in her arms”). In its view of temporal history, H.D.’s typological project resonates with the figural techniques of other modernist poets, notably with Pound and Eliot. In Auerbach’s discussion of typology, he claims that the mode has always had a “historical function”; in the case of the early Christians, “its integral, firmly teleological view of history and the providential order of the world gave it the power to capture the imagination and innermost feelings of the convert nations” (56). Mutlu Blasing, in her essay on Pound and Puritan typology, extends this idea of historical function to the Puritans, who “used biblical patterns to ‘justify their errand into the New World and to defend their claim to national election” (Rowe, qtd. in “Designs on History” 6), and ultimately to Pound, for whom “typology serves a historical function by representing a literary-historical rupture as a return to a truer tradition” (7).15 Like Pound, H.D. uses typology as a means of appropriating the authority of prior texts and revising their chronological relations. However, whereas Pound uses the figure in an attempt to “perceive the single pattern . . . informing myths of the death and rebirth of ancient fertility gods, Odysseus’s journey to the underworld, Christ’s death and resurrection, [and] Dante’s journey” (14), such that his typology may lead to “a totalitarian, closed system of meaning that enfolds historical development” (16), 15 For the full context of Rowe’s argument, see Karen E. Rowe, Saint and Singer: Edward Taylor’s Typology and the Poetics of Meditation. Cambridge: Cambridge University Press, 1986. 60 H.D. participates in a typological reordering that opens up texts and words to designs that may disrupt the orders that she first sets out to establish. In this sense, H.D.’s typological mode is closest to Eliot’s. This is not only because H.D. and Eliot’s poems lack the dogmatic political stance of much of Pound’s verse, which demands that all figures conform to a certain historical vision. As early as “Tradition and the Individual Talent,” in which Eliot claimed that the “ideal order” of the “existing [literary] monuments” was “modified by the new (the really new) art among them” (38), he was proposing an organization of literary history in which the significance of past literary events was always alterable (in contrast to Pound’s model, in which this significance was fixed). His most fully developed treatment of typology occurs not in his prose essays, but in the lyric poetry of Four Quartets, which I turn to in the next chapter. The model of time that Four Quartets offers is one of both simultaneity and radical openness: the past never loses its singular character and concentrated potential, but its significance may be realized only retrospectively and is always open to revision. This model is somewhat more prescriptive than H.D.’s—“In my beginning is my end” (“East Coker” I, 1) offers room for temporary ignorance while implying a predetermined order—but it also introduces a non-teleological, non-progressive typology in which “History may be servitude / History may be freedom” (“Little Gidding III, 162–163). Thus far, I have considered typology in H.D. mainly from the point of view of the content, or sense, of the words that act as figures. But typology also works in Trilogy purely in the register of sound. In the figural view of history, “events are considered not in their unbroken relation to one another, but torn apart, individually, in relation to something other that is promised and not yet present” (Auerbach 59); similarly, H.D.’s figural poetry wrenches words from their historical and linguistic contexts in order to open up new interpretations and new patterns. Her figures are “concrete” not necessarily in the historical sense, but in the linguistic sense, in the letters on the page and in the sounds they generate. The connections forged by her linguistic figures thus allow a move into the second sense of time in typology, the non-linear, eternal space of equivalence. In 61 founding these typological relations, the sounds are enough, absent any logic of etymology or even translation, to establish and generate associative patterns. For H.D., the figure is a way of foregrounding repetition of every kind, of anticipating “this has happened before” and “this will happen again” (FR 41, H.D.’s emphasis). Thus: For example: Osiris equates O-sir-is or O-Sire-is; Osiris, the star Sirius, relates resurrection myth and resurrection reality through the ages; (WDNF 40) Here, the poet constructs relations between the name of the Egyptian god of the underworld, Osiris, and the English, masculine titles “sir” and “Sire” it contains; a rearrangement of the letters also allows a link with Sirius, the dog star, and with its associated goddess, Isis; finally, the female goddess suggests a subject rhyme with the Virgin Mary, thus linking Egyptian and Christian “resurrection myth.” The punning deconstruction of the “Osiris” palimpsest is closely followed by a series of verbs that further illustrate its function: “cover,” “reveal,” “uncover,” “correlate,” “recover” (WDNF 40).16 The latter term, in particular, speaks to the double function of the palimpsest and the double temporal impact of typology. The trace and the figure both face two ways: H.D. “retell[s] and rewrit[es] and add[s] to a palimpsest even as she realizes that such an approach is a regression. . . she hides what she seeks to reveal” (Gubar 315). More broadly, the concept of the palimpsest enables a “working over and working through narratives, thus altering and linking them differently as they are retold and reinterpreted” (DuPlessis, “Revisionary” 16 Morris identifies this sort of wordplay with the phonogram, a set of words in which matched consonants establish a pattern that is both “mutable and unalterable”: “Although they are open to change because of their variant vowels, their aural architecture—their ‘walls’ or consonant structures—do not fall” (42). 62 120–121).17 The palimpsest permits the poet to rediscover the past and, in doing so, to read it as a figure. In Trilogy, the figure has transcendental, mystical implications that point outside the poem as well as formal implications that point within the text itself. If “Osiris” includes and expands to “O-Sir-is” and “O-Sire-is,” to “Sirius” and “Isis,” this is because the arrangement of the letters on the page allows each incarnation of the word to make sense, to seem predetermined, only in retrospect. This kind of formal repetition is much more complicated than, for instance, a rhyme scheme, but it plays upon a similar system of expectation and fulfillment. For the reader, the effect is dual. On the one hand, Trilogy creates and sustains a network of associations that are legitimized by the pattern of the past (a pattern similarly created and sustained by the poet). On the other hand, the poem effectively exposes these patterns as artificial—not false, but constructed and endlessly revisable, and entirely dependent on perspective. This is because . . . my mind (yours) has its peculiar ego-centric personal approach to the eternal realities, and differs from every other in minute particulars (WDNF 38) These lines are the poem’s response to those who say that the “search for historical patterns” has “been done to death before” and ask: “what new light can you possibly / throw upon them?” (WDNF 38). The answer, in effect, is that the search will be, and must be, repeated infinitely. The parallels are always shifting, and as history progresses in time, the typological pattern—if not its interpretive mode—is subject to infinite change. 17 DuPlessis here is referring to H.D.’s later poem Helen in Egypt, but her comment may be equally applicable to Trilogy. 63 IV. Figure and Figura Whether in the realm of sense or sound, typological thinking in lyric relies on the forging or finding of relations that seem in retrospect self-predictive. This essential structure raises a question of consequence to lyric poetry, and to poetry in general: namely, do alliteration, rhyme, and other formal arrangements of sound prefigure typological thinking, or vice versa? Put another way, does sense or sound determine the form that poetry takes? Trilogy does not necessarily answer this question, but the relations it posits demand its consideration. The poem’s figural wordplay exposes a typological structure of expectation and fulfillment that applies to the register of sense as well as sound. As one way of thinking through the relations of primacy in these patterns, it may be helpful to examine the way that poetic figures (as tropes) contribute to H.D.’s typological project. In general, tropes allow the poet to express the sense of something more precisely than the signifier of the thing itself, and some tropes create this impression, to greater or lesser degrees, through a typological approach to form. My point of departure here will be Kenneth Burke’s discussion of metaphor, metonymy, synecdoche, and irony in “Four Master Tropes,” which suggests that tropes themselves operate in specific temporal ways. Burke’s system distinguishes between the four tropes that he identifies even as he argues that each one, when applied to the fullest extent, implicates all of the others. Each one is also associated with a “‘literal’ or ‘realistic’ application”: metaphor is aligned with perspective, metonymy with reduction, synecdoche with representation, and irony with dialectic (503). The first trope, metaphor, Burke defines as “a device for seeing something in terms of something else. It brings out the thisness of a that, or the thatness of a this” (503, Burke’s emphasis). This trope moves in one direction only, from the physical to the spiritual: “Language develops by metaphorical extension, in borrowing words from the realm of the corporeal, visible, tangible and applying them by analogy to the realm of the incorporeal, invisible, intangible; then in the course of time, the original corporeal reference is forgotten, and only the incorporeal, metaphorical extension survives” (506). This conception of metaphor recalls Emerson’s description of language 64 itself as “fossil poetry”: “As the limestone of the continent consists of infinite masses of the shells of animalcules, so language is made up of images or tropes, which now, in their secondary use, have long ceased to remind us of their poetic origin” (217). Extending Burke’s description here, we might say that as metaphor enables a move from the concrete to the abstract, it similarly allows a move from the chronological perspective to one that is typological, or to one that is timeless or out of time. Metonymy, the second trope, brings the one-directional movement of metaphor full circle. Its “basic ‘strategy’” is to “convey some incorporeal or intangible state in terms of the corporeal or tangible” (506). The poet can “regain the original relation [now obscured by metaphor] in reverse, by a ‘metaphorical extension’ back from the intangible into a tangible equivalent. . . and this ‘archaizing’ device we call metonymy” (506). Hence, metonymy is associated with the (re)embodiment of the word, and in this sense it mimics the typological interpretive move, in which two “concrete, historical” manifestations are mediated by the “spiritual factor” of the “understanding, intellectus spiritualis” (Auerbach 32). Burke’s third trope, synecdoche, has a more explicitly typological character. This is because it involves substitutions—of “part for whole, whole for part. . . cause for effect, effect for cause”—and “all such conversions imply an integral relationship, a relationship of convertibility, between the two terms” (508). Moreover, Burke claims that in some sense “the well-formed work of art is internally synecdochic, as the beginning of a drama contains its close or the close sums up the beginning, the parts all thus being consubstantially related” (509); thus, for example, in The Ancient Mariner, the Albatross is from the beginning “something to be murdered, and it implicitly contains the future that is to become explicit” (509, Burke’s emphasis). One thinks also of Chekhov’s rule whereby, “If in Act I you have a pistol hanging on the wall, then it must fire in the last act” (Ilia Gurliand, qtd. in Rayfield 203)—although, of course, the pistol is not a figure of its later use, but an object of concentrated potential whose selection, as a dramatic or narrative detail, is significant. Put another way, the “internally synecdochic” work that Burke describes is 65 one that is full of figures; it “contains” the future that is revealed in its narrative (and, I would add, linguistic and extra-linguistic) fulfillments. Following this progression of figural typology, we might expect the final trope in Burke’s system to be the most typological of all. In some ways, this is true: irony is defined as the “perspective of perspectives” (512) which views all positions as equally valid. This perspective, like that of the typological interpreter, flattens all hierarchies of chronological order and empirical truth. In the ironic or dialectical view of history, there are “elements of all. . . positions. . . existing always, but attaining greater clarity of expression or imperiousness of proportion of one period than another” (513). Burke concludes his essay by suggesting that irony “moves us into the area of ‘law’ and ‘justice’ . . . that involves matters of form in art (as form affects anticipation and fulfillment) and matters of prophecy and prediction in history” (516). Although typology never appears in this essay by name, these lines on the subject of “irony” essentially posit the extension of typological interpretation from history to narrative or poetic form. My reading of Burke is meant to suggest that, if the most powerful tropes in poetry point towards and extend into each other, they also point towards a figural (typological) structure of meaning. In fact, we might say that there is something in the cognition that happens through figures that is inherently typological. Consider Proust’s definition of art in Past Recaptured: “Truth—and life too—can be attained by us only when, by comparing a quality common to two sensations, we succeed in extracting their common essence and in reuniting them to each other, liberated from the contingencies of time, within a metaphor” (qtd. in Kern 218). Like Proust, many critics explain literary figures in spatial terms precisely because what the figure does disrupts time; it establishes connections that have nothing to do with linear cause and effect. In this way, the operation of trope has a direct parallel in Auerbach’s description of figural thought, in which a concrete figure (shadow) and fulfillment (truth) become abstract in relation to the meaning revealed by their juxtaposition. However, unlike an individual trope, the typological mode enables a kind of thinking that works across time, both forward (in anticipation) and 66 backward (in retrospect)—not just in terms of narrative plot, but, for example, within and among the sonic patterns that structure the verses, poems, and books of H.D.’s lyric Trilogy. Finally, in addition to Burke’s “four master tropes,” there is one trope that enacts the concept of typology yet more exactly.18 Metalepsis, which “figures a ‘consequent’ as an ‘antecedent,’ or the other way around” (Lyric Poetry 31), has been described as a kind of “meta-trope, or figure of linkage between figures” (Hollander, qtd. in Lyric Poetry 32).19 Following Hollander, Blasing notes that metalepsis, unlike other tropes, is essentially a “diachronic figure” (32). Because of the connections that it alone enables, we could say that “historical and narrative meaning in general—all meaning that rests on the order of precursors and fulfillments, causes and effects, where each gains significance in a reciprocal motivation across time—rests on metalepsis” (32). Crucially, “if the intentionalizing trope of metalepsis also moves along the lines of like sounds and phonemic connections. . . it would seem to be involved in the specifically poetic motivation of sound as sense and of sense as sound” (32, Blasing’s emphasis). Here, again, we return to the question that Trilogy implicitly poses. If we deconstruct the hierarchy that prioritizes the content of a word over its form, viewing sound and sense as equally compelling components, can we ever know which is the figure that prophesies the poem? By virtue of its involvement in poetry’s sonic and semantic motivations, metalepsis provokes a self-conscious assessment of the generative logic of Trilogy. Consider, in the first book: Depth of the sub-conscious spews forth too many incongruent monsters and fixed indigestible matter such as shell, pearl; imagery 18 For this idea, and for the discussion of metalepsis that follows, I am indebted to Mutlu Blasing. 19 For John Hollander’s full definition, see The Figure of Echo: A Mode of Allusion in Milton and After. Berkeley: University of California Press, 1981. 67 done to death; perilous ascent, ridiculous descent; rhyme, jingle, overworked assonance, nonsense, juxtaposition of words for words’ sake, without meaning, undefined; imposition, deception, indecisive weather-vane; disagreeable, inconsequent syllables, too malleable, too brittle (WDNF 32) Here, the poet seems to entertain doubts about a poetry that is generated by the “juxtaposition of words for words’ sake.” At the end of this sentence, however, we find that all these criticisms have come from others, from outside the poem (“you”). In fact, the poem ultimately rejects an order that could regard syllables as “inconsequent.” Rather, syllables and sound units construct an order even in their seeming “nonsense”; “even the erratic burnt-out comet / has its peculiar orbit” (WDNF 32). We can see the interaction of sound and sense at work earlier in the first book. For instance, in section 21: “here am I, Amen-Ra, / Amen, Aries, the Ram” (H.D.’s emphasis). Although their common association with the ram loosely links these deities, equally and perhaps more important is the logic of sound, what H.D.’s imagined critic might call “overworked assonance, nonsense.” Moreover, the statement “here am I” seems to demand the title that follows it, “Amen-Ra,” whose three syllables mirror its meter and nearly repeat, in a different order, its sounds. Similarly, when the poet undertakes a rehabilitation of Mary’s name in Tribute to the Angels, the force of the argument depends largely on aural links. Thus, in section 8, “a word most bitter, marah, / a word bitterer still, mar” becomes “mer, mere, m`ere, mater, Maia, Mary, / Star of the Sea, / Mother” (H.D.’s emphasis). This progression depends primarily on the shared sounds that link each word in the series; without these, the transformation of the content (i.e. “bitter” to “Star of the Sea”) would be groundless. Reading these poems, one has the sense that the sounds 68 are the driving source of the narrative movement. The last section of The Walls Do Not Fall concludes “possibly we will reach haven, / heaven” (H.D.’s emphasis), its last line revising the penultimate word by a single letter. But it is this one letter, this one shift of a vowel, that fulfills the sense and sound of a haven that did not yet know it was incomplete. The wobbly status of truth in these lines is not unique to H.D.’s poetics. Rather, it is her explicit acknowledgement of the transformative and generative power of individual sounds within a typological system, and her complicity with this power, that sets her apart among her contemporaries. Although typology, by its nature, exceeds rational logic, within its own system it imposes a “universal vision of history” (Auerbach 52) that, taken to the extreme, can result in totalization. Put another way, “typological figuration comes with metaphysical authority and confirms a totalitarian, closed system of meaning that enfolds historical development” (Blasing, “Designs” 16). For a poet like Pound, the possibility of constructing a closed, telos-driven system is one of the attractions of typology, even if “the specialness of poetic language interferes with, modifies, and may even subvert the poet’s ostensible ideology” (17). For Eliot, a typological orientation does not fix the historical narrative but keeps the possibilities of the past in play. Finally, for H.D., typological thinking tends to open up texts, to make them available for rewriting: “the pages, I imagine, are the blank pages / of the unwritten volume of the new” (TA 38). The temporal mode of Trilogy is similarly open, pointing to time’s fluidity and to the revocability of a seemingly static system. V. Conclusion: Typological Thinking and Trilogy’s Lyric Openness Trilogy may be unique in the sort of typology it adopts as a governing mode, but typological thinking is not unique to Trilogy. Among H.D.’s contemporaries, Eliot’s approach is probably the closest parallel; Pound, as mentioned earlier, invokes a more totalizing mode of typology in works like the Cantos. More broadly, typology might be said to operate, on some 69 level, in all modernist lyric, and in lyric generally. The temporality of any lyric utterance is necessarily different from that of narrative because meaning must be generated without the temporal movement of plot. In lyric, words and sounds gain significance as, in the retrospect of reading, their conceptual or sonic accord becomes apparent. Tropes achieve their effects by bringing together words or concepts that may share no other logic than the meaning they mutually produce. If we define typology as a mode of interpretation in which figure and fulfillment relate primarily by the truth they point to, then these features of lyric are necessarily typological. More specifically, in this chapter, I have been suggesting that H.D.’s typological mode in Trilogy is distinct from other literary and Biblical typologies in that it is open and destabilizing as opposed to closed and prescribed. This might seem counterintuitive, especially as typology involves a kind of predestination; to believe that something is a fulfillment, one must also believe that its figure had an original and inalterable purpose before the two were juxtaposed. Here, it may help to consider Agamben’s definition of poetry as a discourse characterized by enjambment: in “The End of the Poem,” he writes that “verse is the being that dwells in [the] schism [between sound and sense] . . . . And the poem is an organism grounded in the perception of the limits and endings that define—without ever fully coinciding with, and almost in intermittent dispute with—sonorous (or graphic) units and semantic units” (110). Enjambment itself is defined by the possibilities that exist between the end of one line and the beginning of another; all is undecided until the next line begins, at which point a seemingly necessary fulfillment is reached only to be followed by another schism or gap. Similarly, H.D.’s typology allows for multiple configurations of meaning that nevertheless reach certain limits with each new phoneme and phrase. In this mode, every unit of sense or sound is a potential figure awaiting a fulfillment that is identifiable only in retrospect. Finally, the typological mode of Trilogy argues for the openness of a certain kind of modernist lyric that calls for a corresponding openness in the act of reading. This is because typology necessarily involves an other, an interpreter—in the case of lyric, a reader. In a 70 typological reading of history, the interpreter provides the only stability; the salience of events may be arranged and rearranged according to the interpreter’s position. In the typological reading of lyric that H.D.’s poetry seems to require, readers provide this stability by making connections between words and sounds that anchor the text to certain meanings while, at the same time, revising this chain of significance throughout the poem. In this way, the typological operations of Trilogy ensure that its lyric temporality remains open, extending beyond the perpetual “now,” and that the poem attends to—in fact, depends upon—the interpretive act. C HAPTER T HREE “Every poem an epitaph”: Rhythm, History, and Genre in Four Quartets 72 Ridiculous the waste sad time Stretching before and after. – “Burnt Norton,” 174-175 Just how many clocks we contain still isn’t clear. The most recent neuroscience papers make the brain sound like a Victorian attic, full of odd, vaguely labeled objects ticking away in every corner. – Burkhard Bilger, “The Possibilian,” The New Yorker, April 25, 2011. I. Introduction Was ever a poet more “acutely conscious of his place in time”1 —and especially of his place in relation to time past—than T. S. Eliot? In “Tradition and the Individual Talent” (1919), Eliot’s early, seminal essay, he pointed to the artistic necessity of the “historical sense,” that is, “a sense of the timeless as well as of the temporal and of the timeless and of the temporal together” that “involves a perception, not only of the pastness of the past, but of its presence” (38). This concern with the present reality of the past is incessant in Eliot’s work. If anything, with the passage of time, the urgency of his prescription only increases. Twenty-five years after “Tradition,” he warns of “a new kind of provincialism” gaining currency, one that is “not of space, but of time; one for which history is merely the chronicle of human devices which have served their turn and been scrapped, one for which the world is the property solely of the living, a property in which the dead hold no shares” (“What is a Classic?” 130). Here, what had been an aesthetic flaw becomes a “menace” (130); the living, figured as the sole proprietors of the world, have weirdly disinherited the dead. In one sense, Eliot’s orientation toward time in these literary essays makes possible his conservative social criticism. The heightened rhetoric of “What is a Classic?” in particular is typical of Eliot’s late prose, which more famously laments the fragmentation of the “spiritual organism of Europe” at midcentury (“Notes towards the Definition of Culture” 302) and proposes, 1 “Tradition and the Individual Talent,” p. 38. 73 as the only viable social and political alternative to totalitarianism, a return to the tradition of Christianity (The Idea of a Christian Society). The concept of the unified historical sense that Eliot applies to literary history, a sense that flattens time by taking an atemporal perspective, complements his vision of a unified Christian Europe that subscribes to its own transcendence by the spiritual realm. Moreover, to the extent that Eliot’s late poetry echoes the temporal themes of his prose, it would seem reasonable to assume that the verse and the essays are governed by the same conservative ideology. This is, in fact, an assumption that underlies a significant portion of the scholarship on Eliot’s Four Quartets.2 In another sense, however, Eliot’s engagement with time in his literary essays and especially in his poetry complicates, and even undercuts, some of his more dogmatic cultural assertions. This chapter challenges the reading of Four Quartets as a poem that valorizes the atemporal—and thus, as a poem that supports or depends upon a certain conservative ideology—by looking more closely at Eliot’s thematic engagement with the philosophy of time and at the poem’s formal patterns of repetition and change. In line with Eliot’s claim in “The Music of Poetry” that “the poem means more, not less, than ordinary speech can communicate,” I consider the way that rhythm operates in the poem as a means of both establishing and disrupting readerly expectations about temporal organization (111). This reading of Eliot’s poem raises questions about the traditional theoretical opposition between lyric time and narrative time, about whether and how poetry engages us in temporally specific ways or performs specific kinds of cognitive work. Put simply, I am proposing that the thinking Four Quartets produces depends upon the experience of time that is possible only through the poem’s typological mode. The specificity of this thinking is highlighted, I suggest, by the first lines of the poem, which have their origins in another genre entirely. I focus on the period of Eliot’s deepest engagement with philosophies of time, in the 2 See, in this vein, the claims by Michael North, in The Political Aesthetic of Yeats, Eliot, and Pound, and John Xiros Cooper, in T. S. Eliot and the Ideology of Four Quartets. Contemporary conservative philosophers make the same assumption; see, e.g., Roger Scruton, “T. S. Eliot as Conservative Mentor.” 74 decade immediately leading up to and following the Second World War. Eliot’s major poetic work of the period, Four Quartets, is, like H.D.’s Trilogy, almost entirely a wartime poem; also like Trilogy, despite its title, it was not originally conceived as a work of multiple parts. The first poem in the series, “Burnt Norton” (1935), borrows its opening lines from a draft of Murder in the Cathedral, Eliot’s dramatization of the death of the archbishop Thomas Becket. “East Coker” and “The Dry Salvages,” published in 1940 and 1941 respectively, immediately follow another play, The Family Reunion (1939), whose driving tension is its characters’ relation to a past that is “unredeemable” without the intercession of the spiritual realm; the criticism from the same period includes The Idea of a Christian Society. The gap between the last poem, “Little Gidding” (1942), and Eliot’s next play, The Cocktail Party (1949), is significantly longer, but both works continue to comment on and revise the temporal philosophies in the earlier poems and plays. The same approximate period includes the essays “The Music of Poetry” (1942), “What is a Classic?” (1944), and “Poetry and Drama” (1951), a critical retrospective of Eliot’s dramatic works and their employment of his verse mode. In a reading of Eliot’s poetic temporality, it might seem reasonable to focus solely on his poetry. But a consideration of the criticism and drama that is contemporaneous with Four Quartets should serve to clarify, not muddy, our view of his engagement with time in verse. In “The Music of Poetry,” which opens with a reassessment of his early critical career, Eliot notes that “the poet, at the back of his mind, if not as his ostensible purpose, is always trying to defend the kind of poetry he is writing, or to formulate the kind that he wants to write”; hence, “what [the poet-critic] writes about poetry. . . must be assessed in relation to the poetry he writes” (107). This claim clearly invites application to his own work. Yet even if Eliot did not explicitly encourage readers to consider his prose arguments in light of his verse, and vice versa, the thematic and verbal echoes among his late works would indicate that these poems, plays, and essays are in conversation with each other and would suggest that Eliot’s thinking about time develops in important ways across genres. 75 Of course, even within a single work of poetry, Eliot’s consideration of the temporal and the timeless takes place on multiple levels, of which the thematic is only sometimes the most pronounced. In Four Quartets, for instance, the overarching title of the work proposes a relation between poetic and musical temporality, while the geographically specific titles of the individual poems point to local moments in space at distinct moments, across time, and out of time. Divisions within the poems and differences in the rhythms that structure them affect the temporal experience of the reader, as do the repetitions and revisions of certain refrain-like phrases (e.g. “time past and time future,” the “still point” of the turning world) and images (e.g. the rose garden). The work’s literary allusions, whether in the form of direct quotations, translated phrases, archaic spelling, or borrowed meters, juxtapose textual origins of great temporal and spatial variety; its internal echoes and references to other works in Eliot’s oeuvre direct our attention to a sustained reflection that transcends any one poem or part of a poem. All of these aspects participate in the temporal argument of the poem, although they are not always in agreement (both because time is inherently disjunctive, and because semantic statements about time in verse may indicate different arguments than the work of form). I examine the development of Eliot’s philosophy of time across genres, with special attention to the way his ideas function in Four Quartets. Part of the way that poetry means “more” than “ordinary speech” (Eliot, “The Music of Poetry” 111) is through the rhythm that we perceive in the poem’s sound and the visual lineation that structures our experience of time as we read it. Reading the time of Four Quartets through its non-semantic properties allows us to see how a poem might think differently than an essay; in particular, the conservative anti-historicism on display in Eliot’s works of social criticism is not necessarily compatible with the line of thought that governs his poetry. I propose that the temporal thinking in Four Quartets is typological, although not on the model of H.D.’s open-ended, future-oriented typology. Eliot’s typology is backward-looking, oriented to an active past that, like Bergson’s duration, perpetually weighs on and eats away at the present. It is this sense of an unremitting and always-increasing past that 76 requires the typological moment—the intersection of the temporal and the timeless—to establish pattern and meaning. In other words, Eliot’s typological thinking is both a reaction to the “presence” of the past—which is variously a redemptive ideal, an empirical fact, and a kind of nightmare—and the manner in which the temporal and the timeless can be thought together. II. Poetic Rhythm in Modernist Verse In an early essay, “Reflections on Vers Libre” (1917), Eliot suggests that “the very life of verse” lies neither in meter nor in the (futile) attempt to shrug off meter—after all, “even the worst verse can be scanned” (36)—but in the “contrast between fixity and flux,” an “unperceived evasion of monotony” (33). In other words, poetry of all kinds depends upon the presence of a kind of patterning that we might very well call rhythm. But this claim, if we accept it, still leaves open some very basic questions. What distinguishes prosody from rhythm? How does its structure interact with the semantic content of a poem? How does rhythm function in non-metrical verse? More fundamentally, what are the defining qualities of rhythm in lyric poetry? No aspect of poetry is more closely aligned with time than its rhythm: we experience its patterns over time, through time, perhaps even as time.3 The history of the word itself bears witness to the extent of this conceptual relationship. The Greek root of rhythm, rhythmos, meant movement or flow; the Latin rhythmus added an association with calculation and measurement;4 in English, one entry in the OED incorporates both senses, defining rhythm as the “the measured flow of words or phrases in prose, speech, etc.” (emphasis mine). This etymological shift parallels a longstanding philosophical debate about the nature of time: whether we should think of it as a continuous flow, like a river, or as a series of discrete moments, like a string of beads. Rhythm, like time, invites this paradox because it depends upon the continuous experience of patterns that encompass—or seem to—isolatable components. And it shares with time two other paradoxes as 3 See Blasing, Lyric, p. 55. 4 Brogan, p. 1067. 77 well: like time, rhythm appears to be at once universal and highly subjective; also like time, it can be theorized as a product of mind or as productive of mind, a precondition for conscious experience. Faced with these shared, contradictory conditions, it can be tempting to define rhythm by revising Augustine: “What then is [rhythm]? Provided that no one asks me, I know” (230). In a chapter that deals with Eliot’s temporal poetics, however, the question must be asked. Recent examinations of rhythm have emphasized its ability to surpass or even work against the semantic properties of a text. Peter Dayan and David Evans note that, in artists as different as Val´ery and Woolf, “rhythm comes to represent something that is as elusive in its source and matter as it is essential to literature; it not only invites disruption, it actually is disruption, disruption of our common sense of how we ought to be able to understand and situate the value of what we read” (153). Rhythm, then, both depends upon the material of language and does violence to it; it can impose order on a literary text or, as easily, upset it. Moreover, what we mean by rhythm exceeds its instantiation in verse, and verse itself exceeds its representation in language. Simon Jarvis argues: “Whether or not we all know exactly what language is, where it starts and where it stops, verse is not in any event a wholly contained subset of it. Language is one of the materials of verse. Others are paralinguistic or extralinguistic—rhythms which do not merely derive from language, but which are imposed on it in a series of transformative mutilations” (“Unfree Verse” 292). We are closer to a definition here, albeit a negative one: rhythm is not meter; it is not reducible to language; and, despite its tendency toward pattern, it is not always a force of order. All of these characteristics are reflected in Derek Attridge’s conception, which may be the most useful for the discussion that follows: rhythm is “a patterning of energy simultaneously produced and perceived; a series of alternations of build-up and release, movement and counter-movement, tending toward regularity but complicated by constant variations and local inflections” (3). Eliot, like many poets, believed that the rhythmic “patterning of energy” was the necessary precondition for the process of poetic composition. In “The Music of Poetry,” Eliot’s 78 most direct theorization of the embodied experience of poetic composition, he observed that “a poem, or a passage of a poem, may tend to realize itself first as a particular rhythm before it reaches expression in words, and that this rhythm may bring to birth the idea and the image” (114).5 Mutlu Blasing, noting the frequent association between rhythm and poetic genesis, observes that the non-semantic, affective qualities of language become more available to us through rhythm: “the hold of the referential function relaxes with the entry into a rhythmic state, allowing for a greater sensitivity to the perceptual qualities of sounds” (Lyric 49). Studies of childhood development confirm that a sense of rhythm is the necessary ground not just for poetic composition but for the acquisition of language in general. Rhythm is essential to our ability to understand speech as meaningful because it prepares us, as infants, to intentionalize the sounds that we hear around us.6 And while rhythm itself is neither purely mimetic nor purely acoustic, it is nevertheless a phenomenon that persuades us, rhetorically, to interpret sound as a representation.7 Insofar as this quality of rhythm is learned, it is part of a social process; insofar as it allows each of us to speak as an intending “I,” it is also a mechanism for individuation. The association of rhythm with the conditions necessary for speech brings us to another point: the relationship between rhythm and the act of will on the part of the reader (or, put more broadly, between poetry and phenomenology). T. V. F. Brogan notes that the first law of rhythmicity is “that a pattern exists so long as it is perceived as such” (1067, emphasis mine). This rule speaks to the difficulty in isolating rhythm in poetry: although it participates in both content and form, rhythm cannot be reduced to a component of either; rather, it must be perceived temporally, by a living subject. Blasing, drawing on the work of psychoanalyst and philosopher Nicolas Abraham, points up this quality when she observes that “the rhythmic event happens only if we will it” (Lyric 56). The “will” of the reader in this sense allows for involuntary as well as 5 This idea is not original to Eliot. Val´ery makes much the same point in his essays on poetry, for instance, and Shelley’s notebooks suggest a similar process of composition. See Evans, pp. 166-167, and Fehrman, Poetic Creation, p. 93. 6 See, e.g., Kovelman et al. (2012), and Blasing, Lyric, pp. 53-54. 7 Blasing, Lyric, p. 55. 79 voluntary responses to rhythm, which are part of its power. Abraham’s essays on rhythm, which incorporate both the psychoanalytic and the phenomenological traditions, enable a theoretical account of the structure and function of rhythm in poetry that proves especially useful for our reading of the temporality of Eliot’s verse. In Outline of a Phenomenology of Poetic Expression, Abraham theorizes rhythm as an event that depends upon a “change of attitude” (21) in the perceiving subject, as when a passenger in a train suddenly begins to experience the movement as rhythmical. In his account, “instead of saying a ‘perception of rhythm,’ it would be better to speak of a ‘rhythmization of perception’” (21). In a later essay, Rhythmizing Consciousness, Abraham proposes that the titular concept is distinct from cognitive consciousness in that, rather than relying on observation and prediction, its future “results from its own decisive act of will” (72). Rhythmizing consciousness makes patterns out of sensory data not by setting expectations to which the data must conform, but by conforming its expectations to the data it receives: “integrating the accident into the whole, rhythmizing consciousness becomes the expectation of what has just been constituted. And the appearance of the now-expected recurrence becomes a brilliant confirmation of its initially endangered power” (72). In other words, rhythmizing consciousness operates according to a kind of non-teleological typology, a structure of figure and fulfillment that realizes its seemingly pre-determined pattern only in retrospect and anticipates patterns that are necessarily open to revision. To say that rhythm is typological in this sense—that it involves, in Abraham’s terms, a “retrospective synthesis” (77)—is simply to confirm that it is, as we noted earlier, “a patterning of energy simultaneously produced and perceived” (Attridge 3). The vocabulary that Abraham introduces to describe what he calls a “phenomenological rhythmics” helps explain how this works from the basic level of the phoneme: the poetic term, the phenomenological equivalent of the poetic foot, includes the prosteleut¯e, “the set of emergences preceding and leading up to the teleut¯e,” and the teleut¯e, which resolves the tension inherent in the term (80). As a rhythm develops, it may involve a series of increasingly complex prosteleut¯e-teleut¯e patterns. A first term 80 may be a component of a secondary term, which may then be a part of a tertiary term, and so on. This progressive layering of figure and fulfillment roughly parallels what Attridge calls the “hierarchical arrangement” of beats in metered verse (62), in which, for example, one beat may be subordinate to another beat within a line, which may in turn be subordinate to another beat within a stanza. For Abraham, each term contains a figure-fulfillment structure and also participates in a larger pattern of figure and fulfillment; the term sets up an expectation for its own return and, if it is unexpected, is retrospectively assimilated. He notes that this “progressive ascent” is “a characteristic process of unification, a process through which the rhythmizing consciousness hurls itself into the conquest of time” (83). This process may appear to be a totalizing one, but it is also clearly open to revision: rhythmizing consciousness assimilates, but it can never prescribe. The rhythm of poetry is one obvious field of application for Abraham’s ideas. In fact, he argues that the typological operation of rhythm may be the distinguishing feature of all “authentic poetry” (59). The act of reading a poem involves both retention and protention, and as we read we continually “fill out” the phonetic and semantic aspects of the protention’s continually moving horizon.8 Rhythm and rhyme make up the “mobile framework” of the protention, in which the semantic elements are “very indeterminately foreshadowed” (59). The best poems exhibit a mix of the “unexpected” and “necessity” in the filling out of the protention, such that “the successive semantemes emerge as the only ones possible, and yet each one of them surprises me by being just as it is” (59). In this account, the difficulty in composing an “authentic” poem lies not in generating a sense of surprise—which can be achieved by following an order or pattern with any kind of nonsense or randomness—but in choosing a sequence of words in which every level of language seems reciprocally determined. And it is on this count that, according to Abraham, the 8 These phenomenological terms are originally Edmund Husserl’s. See, e.g., On the Phenomenology of the Con- sciousness of Internal Time (1893-1917): “Immanent contents are what they are only as far as, during their ‘actual’ duration, they point ahead to the future and point back to the past. But in this pointing back and forth something else remains to be distinguished: In each primal phase that originally constitutes the immanent content we have retentions of the preceding phases and protentions of the coming phases of precisely this content, and these protentions are fulfilled just as long as the content endures. These ‘determinate’ retentions and protections have an obscure horizon; in flowing away, they turn into indeterminate retentions and protentions related to the past and future course of the stream” (89). 81 “various attempts at poetic revolution” have failed: without the sense that its particular sequence of phonemes and semantemes fulfills our expectations—expectations that may only be clear in retrospect—a text can never be truly poetic. This idea has important implications for our reading of rhythm in Eliot, and for modernist poetics generally. Can rhythmizing consciousness assimilate the deviations from meter characteristic of much modernist verse? Can poetic necessity be achieved in a poem that is rhythmically uneven? Does the rhythm of a modernist poem differ in any meaningful way from the rhythms found in other genres of literature, or in speech itself? To address these questions, I turn now to Eliot’s critical and poetic engagements with philosophies of time. III. “History may be servitude, / History may be freedom” (LG III, 162-163) Eliot’s thematic engagement with time in his late work is also, arguably, a form of resistance to time. After his conversion to Anglicanism in 1927, his work tends to treat the struggle against time as a spiritual struggle for access to the eternal realm that, in Christian theology, transcends history and the material world. Too often, however, critics take Eliot’s Christian ideology as the definitive guide to the engagement with time that his poetry produces. For example, John Xiros Cooper argues that Four Quartets primarily offers an escape from history, that it “authorizes the serviceable obliteration of personal history by a kind of divine act of reification” and that it “relieves [readers]...of social and political responsibility in the present” (170). Even A. David Moody, who argues that Four Quartets is no less “humane” than more thematically intersubjective works, accepts the idea that the poem’s spiritual theme may limit the function of its words. In his view, because the poem wants to point to the spiritual realm, “beyond any sense which words can make,” it must work by “alienation and negation”: “Words and things have their specific density refined out of them” (151). The idea that the rhetoric of Four Quartets serves mainly to administer “consolatory Christian ideas and practical quietism” (Cooper 165) seems especially satisfying from the point of 82 view of some cultural critics. For example, Cooper observes that “the conservative cultural history which Eliot championed in the 1930s and 1940s presumed a particular subjective position for his mandarin readers,” and his reading of Four Quartets rests on the notion that the poem makes these same rhetorical assumptions. According to this politically charged reading, Four Quartets is dominated by an anti-historical ideology that differs in subtlety, but not in kind, from the ideology of his social criticism. From a literary point of view, however, this stance is problematic. To assume that the late poetry valorizes the timeless over the temporal or historical, or interior over intersubjective experience, is to ignore a much more complicated engagement with time, one that takes place not just in the poem’s thematic or propositional content but in its sound, syntax, lineation, and rhythm. Take, for example, Cooper’s reading of the following lines in “The Dry Salvages”: Fare forward, travelers! not escaping from the past Into different lives, or into any future; You are not the same people who left that station Or who will arrive at any terminus. . . (III, 137-140) In Cooper’s argument, this passage puts forth a view of human experience in which the “actions, beliefs, and events which constitute the past” no longer “constitut[e] a matrix of determinations” and the future need not “be seen nihilistically as the kingdom of ceaseless self-fashioning” (174). In other words, in spite of the content of the first two lines, Cooper argues that the passage offers readers an escape from both past and future. Such a vision might have been particularly compelling for those 1940s intellectuals “who needed a way of distancing themselves from past commitments,” allowing them refuge in interior reality and in aesthetic experience (174). But it would also have been a fundamental misreading of the poem, and not only because, as Cooper claims, Eliot’s real aim was “to lead secularized consciousness through the aesthetic towards the religious awareness that lies beyond it” (154). In Four Quartets, the past is exactly that which does not disappear; it haunts, and the poetry is complicit in this haunting. 83 Rather than offering a way out of time, the address to the travelers in “The Dry Salvages” elaborates on a recurring theme in Four Quartets: the idea of temporal flux. In DS III, we see this primarily in the evolution of the subject: the travelers “are not the same people who left that station” (133); likewise, “time is no healer: the patient is no longer here” (132). Eliot reprises this theme in The Cocktail Party, in only slightly different terms, in the speech of Sir Henry Harcourt-Reilly: “we die to each other daily,” he declares, as “What we know of other people / Is only our memory of the moments / During which we knew them. And they have changed since then” (I, iii). In this sense, “at every meeting we are meeting a stranger” (I, iii). In both passages, the corollary of constant change is the condition of constant belatedness, a condition which applies to language as much as it does to the subject. In “East Coker,” for instance, “one has only learnt to get the better of words / For the thing one no longer has to say, or the way in which / One is no longer disposed to say it” (176-178). Here, the referential function of language is unsatisfactory not only because it is inherently belated, but also because it is imprecise. And yet, for lack of an alternative medium, we return to saying things in words. In the verses themselves, each enjambed line pushes us forward in search of semantic and syntactic completion and requires an ongoing revaluation of what has already been read. The meaning depends on our continuous movement through time, not our transcendence of it. The theory of time presented in these lines suggests a relation to the past that is anything but escapist. First, if time is a continuous flux, our present understanding of the external and internal worlds can only consist in what has happened and who we used to be; we change, but we remain charged with the past. In this model, we can never cast off the past because we are always already living it. Second, although past moments cannot be fixed in time, they do not therefore cease to exist, and their consequences are not limited to the lives of those who lived through them. To the contrary: “the past experience revived in the meaning / Is not the experience of one life only / But of many generations. . . ” (DS II, 97–99). The idea of time as a flux thus coexists with the seemingly contradictory notion of the simultaneity of all time: the past, present, future, and 84 even the counterfactual. This is the vision that greets us in the opening section of “Burnt Norton”: “Time past and time future / What might have been and what has been / Point to one end, which is always present” (44–46). Hugh Kenner notes that here, “we have very nearly stepped into some world where happenings are simultaneous, the past actual, what might have been really so, our first world still here. . . . But this would be reality, of which human kind cannot bear very much, so it is as a concession to our weakness that we are shut up in the present moment” (296). Similarly, the compound ghost of LG II embodies “that simultaneity of the literary past which has been Eliot’s theme since 1917” (321). For the critics who define Eliot’s late work by its rejection of the temporal world, the lines that I have been focusing on may simply be an aberration from the dominant, anti-historical strain in Four Quartets, a way of building into the poem the idea that “Only through time time is conquered” (BN II, 89). To be sure, there are passages in Four Quartets that express a longing for temporal transcendence. For example, in DS V, . . .right action is freedom From past and future also. For most of us, this is the aim Never here to be realized. . . (224–227) Similarly, in LG III, the “use of memory” is “For liberation—not less of love but expanding / Of love beyond desire, and so liberation / From the future as well as the past” (156–159). Before we accept the primacy of the resistance to time in Four Quartets, though, we should note the choices that Eliot made in the process of composing and revising the poem. The rough draft of DS III, for instance, already claims that “Time is no healer,” but the clause that follows and modifies this statement—“the patient is no longer here”—is added only in a subsequent version. The constantly changing train travelers are also a late addition, elaborating on the idea of temporal flux and the instability of the subject. Meanwhile, the most significant revisions to the first draft of “Little Gidding” are the deletions of lines that would have heightened the poem’s elevation of the 85 timeless over the temporal.9 Eliot’s marginal notes to “Little Gidding” indicate that one motivation for these changes is a concern with the poem’s didacticism.10 But the overall effect of the revisions is broader than a simple shift in tone. Along with the poem’s insistence on the effects of temporal flux and the necessity of retrospection, these changes should call into question the idea that, on a thematic level, Four Quartets simply rejects the temporal or offers a politically convenient escape from the past. In the sections that follow, I attempt to identify the temporal orientation that the poem produces instead. IV. Eliot’s Bergsonism The criticism that emphasizes the role of Eliot’s theology or politics in Four Quartets tends to do so at the expense of the poem’s engagement with philosophical theories of time, of which the most notable is Bergsonism. Eliot’s relationship to Henri Bergson, the theorist of dur´ee and the e´ lan vital—from his attendance of Bergson’s Paris lectures in 1910–1911 to his later disavowal of “the fallacy of progress, which is a Bergsonian fallacy”11 —is beyond the scope of this chapter. But the ways in which Eliot’s poetry thematically coincides with and diverges from Bergsonism are telling, and Bergson’s philosophy offers some useful ways for thinking about formal temporality in verse. Understanding the temporal thinking of Four Quartets depends in part on our ability to recognize the way it takes up and transforms several of Bergson’s concepts. These include the belatedness of present perception; the notion that the past is irreversible and unrepeatable, and the future unpredictable; and the idea that movement through the temporal flux is progressive. Finally, the middle way that Bergson finds between mechanism and finalism, which is a kind of retrospective synthesis, corresponds to the typological operation of what Abraham calls rhythmizing consciousness and that operates at multiple levels in Eliot’s poetry. 9 See, in the first draft, the last lines of section III.2 and the first stanza of section IV (reproduced in Gardner 230–231). 10 See Gardner 229. 11 Quoted in Schwartz 298. 86 Consider the passage that I have been referring to in “The Dry Salvages.” When Eliot insists that the travelers are “not the same people who left that station” (DS III, 139) and that “one has only learnt to get the better of words / For the thing one no longer has to say, or the way in which / One is no longer disposed to say it” (EC V, 176-178), he is adopting a position on the subject’s experience of the flux of time that is very similar to the description of temporality in Bergson’s early work. In Matter and Memory (1896), Bergson writes: “Your perception, however instantaneous, consists then in an incalculable multitude of remembered elements; in truth, every perception is already memory. Practically, we perceive only the past, the pure present being the invisible progress of the past gnawing into the future” (150). Eliot’s use of this concept in his poetry allows him to broach questions of phenomenology that are still relevant in both the humanities and the sciences. For instance, the belatedness of the subject and of language itself in Four Quartets finds its correlate in the neuroscience that shows that the visual brain must wait for “different stimulus features” that “move through different processing streams and are processed at different speeds”; this biological fact “push[es] perception into the past” and “means that awareness is postdictive” (Eagleman). In other words, as Bergson intuited philosophically and Eliot echoed poetically, what the brain perceives as the present is really an interpretation of past data. In Bergson’s thought, the past is copresent with us as we move irreversibly through time. This idea is perhaps most vividly expressed in Creative Evolution (1907), in which the past is figured as a kind of unwelcome doppelg¨anger: its invasion of our present is always happening and always immanent. “In its entirety, probably, [the past] follows us at every instant; all that we have felt, thought and willed from our earliest infancy is there, leaning over the present which is about to join it, pressing against the portals of consciousness that would fain leave it outside” (5). In Four Quartets, too, the past persists in the present, whether we will it or no. But the poem parts ways with Bergson on the question of whether duration, which allows the past to linger in this way, also implies temporal progress. In his late work, Bergson marries the idea of duration to the 87 idea of progressive evolution: “evolution implies a real persistence of the past in the present, a duration which is, as it were, a hyphen, a connecting link” (CE 22). For Eliot, however, the idea that the present is an evolved condition betrays the very past that brings it into being. In “The Dry Salvages”: It seems, as one becomes older, That the past has another pattern, and ceases to be a mere sequence— Or even development: the latter a partial fallacy Encouraged by superficial notions of evolution, Which becomes, in the popular mind, a means of disowning the past.12 (II, 85–89) The pattern of the past is not arbitrary, not “mere sequence” or pure repetition, but neither is it indicative of development toward a higher order. Eliot’s “other” pattern allows us to make meaning from the past and to coexist with it without assuming that the passage of time itself is socially, biologically, or politically teleological. The anti-progressive stance in Four Quartets has many correlates in Eliot’s prose criticism. George Orwell, in his review of “The Dry Salvages,” observed: “Obviously a skepticism about democracy and a disbelief in ‘progress’ are an integral part of [Eliot’s oeuvre]; without them he could not have written a line of his works” (455). These views are perhaps most explicit in Eliot’s social criticism, in which, for example, he warns against the notion of collective cultural, spiritual, or artistic progress.13 His literary criticism posits various versions of cycles, rather than linear development, in literary history: for example, “What is a Classic?” notes a kind of creative destruction in the ways that classic poets “exhaust” the language; “The Music of Poetry” describes the cyclic prominence of “developers” and “innovators” in any given language. It is important to note, however, that the alternative to progress in Eliot’s temporal model is 12 As noted above, Eliot restates this admonition against “disowning” the past in “What is a Classic?”. 13 For instance, in Notes Towards the Definition of Culture: “The Christian Faith also, psychologically considered— as systems of belief and attitudes in particular embodied minds—will have a history: though it would be a gross error to suppose that the sense in which it can be spoken of as developing and changing, implies the possibility of greater sanctity or divine illumination becoming available to human beings through collective progress. (We do not assume that there is, over a long period, progress even in art, or that ‘primitive’ art is, as art, necessarily inferior to the more sophisticated)” (296). 88 neither unmitigated decline nor pure repetition. The idiom of poetry must reflect the state of the spoken language, which is constantly changing. Forms may return, but only after a period of abandonment and then of transformation; they “have to be broken and remade” (“The Music of Poetry” 31). In his rejection of pure repetition, Eliot again incorporates Bergsonism into his theory of time. For Bergson, the unrepeatability of the past is predicated on the irreversible nature of the temporal flux: “consciousness cannot go through the same state twice. . . . We could not live over again a single moment, for we should have to begin by effacing the memory of all that had followed” (CE 5–6). The impossibility of absolute repetition in lived experience has several implications relevant to our reading of Eliot. First, it raises the question of the work that formal repetition does in poems or other literary texts. In one sense, repetition is impossible here too: if I read Four Quartets aloud, I will speak the word “time” 71 times, but each utterance will be different from all the others by virtue of the duration that precedes it, the context of the words that surround it, the inflection of my voice, even the biological fact that I will be older at each reading. If I read the poem silently, I am still faced with the fact that the “music” of the word at each utterance “arises from its relation first to the words immediately preceding and following it, and indefinitely to the rest of its context; and from another relation, that of its immediate meaning in that context to all the other meanings which it has had in other contexts, to its greater or less wealth of association” (“The Music of Poetry” 25). In another sense, of course, repetition is not only possible in poetry, but one of its underlying preconditions. All linguistic patterns—in rhyme, meter, syntax, sound, rhythm, the visual appearance of the line—rely on the alternation of repetition and variation. If pure repetition is impossible, formal repetition is necessary to our experience of verse as verse. Among other things, it allows us to make non-causal and non-semantic connections between words, to detect and participate in a poem’s rhythm, and to experience poetic closure; above all, perhaps, it makes us acutely conscious of the passage of time. It is in this second sense that repetition functions within and among the poems and poem 89 sections in Four Quartets; the way that these repetitions make meaning relies upon the fact that pure repetition cannot take place. Second, Bergson’s view of repetition has implications for the possibility of prediction. If something cannot be repeated, its future state cannot be foreseen, “for to foresee consists of projecting into the future what has been perceived in the past” (CE 6). This explanation of the unforeseeable nature of creation is part of Bergson’s larger refutation of mechanism and, to a lesser extent, of finalism. Mechanism makes time “useless,” “even unreal,” because it “regard[s] the future and the past as calculable functions of the present” (37). Finalism, which is end- rather than origin-driven, “substitutes the attraction of the future for the impulsion of the past” (39). However, for Bergson this method is more flexible than mechanism—its principle is “essentially psychological”—and as such, it is impossible to disprove definitively (40). In fact, Bergson’s philosophy shares with finalism the conception of a holistic universe, the representation of “the organized world as a harmonious whole” (50). Its divergence from teleology lies in the source of this harmony. For Bergson, the pattern in creation is always achieved in retrospect: Of course, when once the road has been traveled, we can glance over it, mark its direction, note this in psychological terms and speak as if there had been pursuit of an end. Thus shall we speak of ourselves. But, of the road which was going to be traveled, the human mind could have nothing to say, for the road has been created pari passu with the act of traveling over it, being nothing but the direction of this act itself. At every instant, then, evolution must admit of a psychological interpretation which is, from our point of view, the best interpretation; but this explanation has neither value nor even significance except retrospectively. (51) This approach helps explain the status of the future in Bergsonism. It is not “contained in the present in the form of a represented end”; however, “once realized, it will explain the present as 90 much as the present explains it, and even more; it must be viewed as an end as much as, and more than, a result” (52). In this sense, Bergson adopts a typological approach that is not also teleological. An end is not known in advance, but is recognized in retrospect; events in the present may be figures, but cannot be known as such until they are fulfilled. This is a model that—like Eliot’s concept of the music of poetry—proposes harmony but also “admits of much discord” (50). Eliot’s reception of this aspect of Bergsonism is essential to our understanding of the temporal orientation of his poem. Up to a point, his typological thinking tracks closely with the description of harmony in Creative Evolution. As early as “Tradition and the Individual Talent,” in which he claimed that the “ideal order” of the “existing [literary] monuments” was “modified by the introduction of the new (the really new) art among them” (38), Eliot was proposing an organization of literary history in which the significance of past literary events was always alterable. But the most fully developed treatment of typology occurs in Four Quartets. The first lines of “Burnt Norton” announce this theme—“Time present and time past / Are both perhaps present in time future, / And time future contained in time past” (1-3)—and the first and last lines of “East Coker”—“In my beginning is my end” (1), “In my end is my beginning” (209)—apply it to an individual life. Lines like these, which set the tone for the poems that follow, suggest a telos-driven typology, one that contradicts Bergson’s insistence on the retrospective nature of harmony and the duration of lived time. Yet other moments in these poems question the notion of a pattern given ex-ante, or at least the possibility that such a pattern can be foreseen. In “East Coker”: . . . There is, it seems to us, At best, only a limited value In the knowledge derived from experience. The knowledge imposes a pattern, and falsifies, For the pattern is new in every moment And every moment is a new and shocking Valuation of all we have been. (81–87) 91 The future will indeed show us a pattern, but not the pattern that the past invited us to foresee. “Little Gidding” reiterates the futility of human prediction: . . . And what you thought you came for Is only a shell, a husk of meaning From which the purpose breaks only when it is fulfilled If at all. Either you had no purpose Or the purpose is beyond the end you figured And is altered in fulfillment. (30–35) These passages, which revise the declarations made earlier, might be said to both articulate and enact the process they describe. The passage from “Little Gidding,” in particular, repeatedly creates the illusion of end-stopped lines (31, 32, 34) whose meaning changes as the result of the modifications or additions in the lines that follow. For all of the formal patterns and verbal echoes that seem to govern Four Quartets, no reader would claim to predict the whole of it from an earlier part; to do so would be as absurd as a belief in the explanatory powers of the “usual / Pastimes and drugs” (“The Dry Salvages” 194–195) proffered by the fortuneteller and the psychoanalyst. Still, on reaching the end of the poem, one might sense that it fulfills an earlier promise.14 The model of time that Four Quartets offers is one of both simultaneity and radical openness. The past never ceases to be real, but its significance to those of us who exist in time may be realized only retrospectively. This model transforms the insights it borrows from Bergson to introduce a non-teleological, non-progressive typology in which “History may be servitude, / History may be freedom” (LG III, 162–163). This is not, I would argue, simply a means of allowing us to evade history, or to revise it to suit our own purposes. Rather, Eliot grants to the past a kind of viability that liberalism, in his view, does not;15 he takes it seriously, not just as 14 See Robinson, whose theory of the poem as a pretended illocutionary act suggests that it functions as a promise to readers. 15 See, for instance, the passage in his 1935 essay, “Religion and Literature”: “These liberals are convinced that only by what is called unrestrained individualism will truth ever emerge. Ideas, views of life, they think, issue distinct from independent heads, and in consequence of their knocking violently against each other, the fittest survive, and truth rises triumphant. Anyone who dissents from this view must be either a mediaevalist, wishful only to set back the clock, or else a fascist, and probably both” (104). 92 something that has happened, but as something that is happening and will continue to happen, something that may have implications that we have yet to foresee. In other words, the vision of Four Quartets is one in which the dead do hold shares in the world. Eliot—although no great interrogator of privilege as a social critic—here puts to the test our epistemological privilege as inhabitants of the present. Finally, the formal correspondences between Eliot’s typological thinking in Four Quartets and Abraham’s theory of rhythm suggest that the theory of time that emerges in this poem has consequences for both aesthetics and phenomenology. Although its form is not as regular as the forms that Abraham indicates would be characteristic of “authentic” poetry, Four Quartets nevertheless constructs patterns on every level of language that make each word in its sequence seem both unexpected and, in retrospect, uniquely necessary. In this way, the poem invites the rhythmization of the reader’s consciousness; its irregular patterns induce us to anticipate and, later, to revise our expectations in accordance with what we read. What I am calling here the poem’s typological mode has parallels in phenomenology and hermeneutics (for example, in Wolfgang Iser’s theory of reading, or Maurice Merleau-Ponty’s description of time as a “network of intentionalities” (484)). And this brings us, again, to the question of genre. What kind of thinking is Eliot’s poem capable of that could not be achieved in essays, narrative, or drama? V. Time and Lyric Theory One way that theorists of lyric poetry have attempted to distinguish this mode from other genres of literature is in its temporality. According to this line of thinking, if narrative depends upon the sequence of beginning, middle, and end, lyric resists the passage of time, remaining in a kind of perpetual “now” that is both in and out of time. Jonathan Culler, in his 1977 article “Apostrophe,” suggested that the time of this poetic figure, “which is the set of all moments at which writing can say ‘now’” (66), is the time toward which all lyric tends. The lyric “now” is thus a now “of discourse, of writing,” rather than the “moment in a temporal sequence” which is 93 the now of narrative or story (68). Sharon Cameron’s extended study of lyric time in the work of Emily Dickinson took this distinction as a given: “Unlike the drama, whose province is conflict, and unlike the novel or narrative, which connects isolated moments of time to create a story multiply peopled and framed by a social context, the lyric voice is solitary and generally speaks out of a single moment in time” (22-23). More recently, the status of lyric as a transhistorical genre about which such broad claims can be made has been called into question. Virginia Jackson underscores the relatively recent conflation of the lyric poem with poetry in general—a condition she calls the lyricization of poetry—and puts the normative definition of the lyric as “a short, nonnarrative poem depicting the subjective experience of a speaker” (183) into its literary-historical context. This line of study suggests that the distinct poetic genres that have been subsumed under the abstract notion of lyric may have different temporal orientations. As a corollary, it opens up a line of inquiry into how poems that differ from the normative definition of lyric differ temporally from other poetic genres or from the temporality of drama or prose narrative. Still, it is a commonplace to define the time of lyric—which sometimes stands in for all poetry—as a kind of timeless now. This temporality distinguishes the lyric poem from other literary genres, but the distinction comes at a cost. If we accept that lyric poetry is defined by its tendency toward a hermetic, interior perpetual present—that it is, as Frank Kermode would have it, an “art of the timeless prison” rather than an art of time and change (174)—then we will be hard-pressed to deny the claim that it is anti-historical and, by extension, less ethical than narrative.16 One response to this problem would be to follow Jackson’s literary historical approach. In the case of Four Quartets, we would find a poem that is, by her definition, nearly anti-lyrical: not short, not entirely non-narrative, and peopled by multiple voices. If Eliot’s poem is not a lyric in the normative sense, then we need not attribute to it the limitations of the lyric 16 The assumption that the relationship of narrative to history is morally superior to that of poetry is well entrenched in literary criticism, and not just in narrative theory. For example, Cooper argues that Four Quartets replaces the narrative model of history with “history as lyric repetition” (169), a model that ultimately relieves readers of their “social and political responsibility in the present” (170). 94 now. And yet this conclusion is not as satisfying as it ought to be, because we have yet to establish what, if anything, distinguishes the temporality of this poem from the engagement with time that takes place in other literary genres. If the time of Four Quartets is not the lyric now, is it the same as—or subsumed by, or a subset of—the time of narrative? The generic borders of Four Quartets may be porous, but it still seems that this poem, like all poems, retains an important formal and temporal specificity that engages our thinking in distinct ways. To understand the particular qualities of Eliot’s temporal thinking in Four Quartets, it is helpful to recognize some of the points of contact between the temporal operations of lyric poetry and narrative. The mode of typological thinking discussed in the previous section, which allows for revised interpretations of the past in light of the unpredictable future, has much in common with some forms of narrative theory. Frank Kermode, who describes plot as the interaction of the successive march of time (chronos) and the significant event (kairos), traces the rise of modern “concord fictions” through a lineage that includes theological typology and Eliot’s literary-historical typology. He argues that typological structure and interpretation is a transhistorical phenomenon, a now-secularized principle that “recurs from the New Testament through Alexandrian allegory and Renaissance Neo-Platonism to our own time” (59). In fact, when Nils Bohr uses the principle of complementarity to explain the relevance of both classical physics and quantum mechanics, “he is really doing what the Stoic allegorists did to close the gap between their world and Homer’s, or what St. Augustine did when he explained, against the evidence, the concord of the canonical scriptures”; the “motive is the same, however the methods may differ” (62). All of these systems are attempts to meet the basic human need for concord—for the imposition of meaning and form on the world’s chaos—but literary fictions may be the most satisfying because they are “consciously false” (64). In the modern era, Kermode suggests, poetry provides a certain level of satisfaction—“the poet finds his brief, fortuitous concords”—but “the novelist accepts need, the difficulty of relating one’s fictions to what one knows about the nature of reality, as his donn´ee” (133). 95 Kermode’s theory of concord fictions elevates narrative, and especially the novel, primarily in the literary-historical sense, as a medium particularly suited to the sensibility of the twentieth century. For Paul Ricoeur, narrative is a category of the poetic, but “the problem posed by narrativity is. . . both more simple and more complicated than the one posed by lyric work” (TN I, 81). It is more simple “because the world. . . is apprehended from the angle of human praxis rather than from that of cosmic pathos” (81); more complicated, because the category of narrative encompasses both fictional and historical discourses and their inevitable “interweaving” (82). But the most distinguishing aspect of narrative is its engagement with time. In “Narrative Time,” Ricoeur proposes a reciprocal relationship between narrativity and temporality, one in which the reader’s memory of the narrative plot enables a reversible structure of expectation and fulfillment such that she can read “the end in the beginning and the beginning in the end” (180). Time and Narrative spells out the implications of such a model: “time becomes human to the extent that it is articulated through a narrative mode, and narrative attains its full meaning when it becomes a condition of temporal existence” (TN 1, 52). What narrative offers is not simply concord, but “discordant concordance,” a concept derived from Aristotle’s analysis of tragedy (42). The modernist novel, by loosening the illusion of linear time, explores “uncharted modes of discordant concordance,” the “hierarchical levels that form the depth of temporal experience” (TN 2, 101). What is at stake in experimental narrative form, then, is more than the novelty or surprise that is necessary for aesthetic experience. If the operation of narrative is understood to be essential to the human experience of time, then more temporally complex narratives allow us to experience more completely what it is to be human. The central component in the correspondence between human and narrative time is the concept of emplotment. As many critics have pointed out, plot is as essential to our everyday experience of time as it is to the structure of the novel. When we perceive the uniform sounds a clock makes as a series of ticks and tocks, for instance, “we use fictions to enable the end to confer organization and form on the temporal structure” (Kermode 45). This is “a model of what 96 we call a plot, an organization that humanizes time by giving it form” (45). Peter Brooks’s psychoanalytically inflected Reading for the Plot observes that this operation works in two directions: emplotment imposes meaning on seemingly meaningless repetitions, and repetition is necessary to the concords created by plot, allowing readers to impose a meaning on time and even to reverse its course (108). Narrative, as opposed to poetry, “must ever present itself as a repetition of events that have already happened” (99). Thus, any event in a narrative plot is “both the recall of an earlier moment and a variation of it: the concept of repetition hovers ambiguously between the idea of reproduction and that of change, forward and backward movement” (99-100). In a similar vein, Ricoeur, again working from a reading of Aristotle’s Poetics, argues that a complex tragic plot depends upon a series of reversals, each of which takes place in “the work’s time, not the time of events in the world” (39). In the literary work, the fact of objective temporal succession is made subordinate to another kind of logic. We perceive this work as a whole because “the ideas of beginning, middle, and end are not taken from experience,” but are “the effects of the ordering of the poem” (39), an ordering which makes the narrated events seem necessary rather than accidental. Broadly, we might say that narrative emplotment is understood as always temporal in nature, and always dependent upon the pattern composed of 1) pure chronology and 2) the events that take their significance from the figures that, in retrospect, they fulfill.17 What should we make of the correspondences between the observations of narrative theory and the temporal operation of Four Quartets? First, it is worth noting that, although I selected these passages precisely for their relevance here, their treatment of poetry is fairly representative of narrative theory as a genre. Brian McHale notes that “even the indispensable poems, the ones that narrative theory seems unable to do without [e.g. the Homeric poems for Plato, Genette, and Sternberg, or Pushkin’s Onegin for Bakhtin], tend to be treated as de facto prose fictions; the poetry drops out of the equation” (11–12). That is to say, poetry is contemporary narrative theory’s “blind spot,” one that is all the more glaring when we note the 17 Another useful point of comparison in narratology here is Gerard Genette’s Narrative Discourse. See especially the discussion of analepsis (retrospection) and prolepsis (anticipation) (pp. 33–85). 97 many ways in which narrative and poetry have historically intersected (12). Kermode closes The Sense of an Ending with a reading not of Flaubert or Dickens, but of Wallace Stevens; Ricoeur’s argument throughout Time and Narrative depends upon his reading, in the first volume, of Aristotle’s Poetics, which offers a theory of lyric and epic poetry as well as drama. If we do not hesitate to classify their arguments as theories of prose narrative, perhaps we should. At the very least, we should not accept the idea that when poetic texts seem to confirm or conform to these arguments, they indicate the subordinate status of poetry to narrative. This would be true only if we read poetry solely along its semantic and syntactical registers, as if it were prose. If we are attentive to the elements of poetry qua poetry that narrative theory tends to ignore—e.g. its sound, meter, rhythm, and lineation—we find that these are exactly the qualities by which poetry exceeds the narrative engagement with time. The patterns in each of these registers are sensitive to changes in the others (for example, the choice of end-stopped or enjambed lines affects the rhythm of a poem, and an established rhythm can train us to expect the coincidence of syntactic and line units), and they may work with or against the semantic thrust or narrative structure of a poem. Attridge observes that, “although the revision of interpretation as the reader progresses is something that happens in prose as well, poetry possesses a special power to present meaning as a constantly changing process” (18). Like many critics, he suggests that the transition between poetic lines is the place in the poem where this effect is most strongly felt. In a similar vein, McHale proposes that the defining feature of poetry is the way in which it negotiates gaps (in lineation) and silences (in oral poetry); his argument draws from Rachel Blau DuPlessis’s concept of “segmentivity” and John Shoptaw’s “countermeasurement” (14–18). Giorgio Agamben locates the necessary quality of poetry more specifically in the enjambed poetic line, which propels the poem forward in time even as it necessitates the backward look of revision. All of these definitions suggest that the formal, non-paraphrasable aspects of poetry participate in the effect of a poem at least as much as, if not more than, the semantic content of its words, and that they do so in part through our experience of time. But these aspects of poetry do 98 more than help us produce a “reading” of the poem in question. As Simon Jarvis puts it, “the assumption that poetics subserves hermeneutics presses connoisseurship of verse style so rapidly and so forcibly into interpretive service that it leads to falsified and unconvincing claims about verse style and its supposed effects” (“For a Poetics of Verse” 932). With this caveat in mind, I want to close this section by offering a reading of the decidedly non-narrative verse in “Burnt Norton” IV: Time and the bell have buried the day, The black cloud carries the sun away. Will the sunflower turn to us, will the clematis Stray down, bend to us; tendril and spray Clutch and cling? Chill Fingers of yew be curled Down on us? After the kingfisher’s wing Has answered light to light, and is silent, the light is still At the still point of the turning world. (127–136) This section departs from the sections that surround it in its relatively short length, stanza organization, and incorporation of regular meter and end rhymes. The nine-syllable, four-beat first line sets up what could be iambic tetrameter, with an extra unstressed syllable and an initial inversion; however, the effect is more like triple meter, which lends a light tone to a semantically somber, end-stopped line. We move more slowly through the next line, both because the uncertainty of the meter may tempt us to add a beat to “cloud” in addition to “black” and the first syllable of “carries,” and because the end of the line coincides with the end of the sentence. The next line extends the nine-syllable line to 13 syllables, and rather than resolving the matter of duple or triple meter, blends both; the enjambed ending follows a punctuated pause in the middle of the line, and finds its rhyme not at the end of the next line, but in the phrase before the caesura (“bend to us”). Meanwhile, the first and last words of the fourth line (“stray” and “spray”) answer the end rhymes of the first two lines, and the enjambed ending leads into a final two-beat line that echoes the consonance in lines 2 and 3 but refuses to rhyme. Throughout the stanza, we move 99 from an observation about the past, in the present perfect, to the present tense, to a question posed about the future. We are propelled toward the next stanza both because we seek an answer to this question and because we seek an overarching formal structure in a stanza that is increasingly patterned and increasingly irregular. The second stanza at first refuses resolution even more radically than the first. The opening line is just one enjambed word, “Chill,” which might be either an imperative verb or, as it turns out, an adjective; it forms part of a question that curls through the first three lines. Until the final line, every line in this stanza leads with an enjambment into a longer line, so that we get progressively more material with which to make pattern and meaning but no regular meter and no answer to the questions posed. The penultimate line is the longest in the section, and seems still longer because of the long vowels and the threefold repetition of “light”; it leads, again by enjambment, into a final line which at last seems to offer the coincidence of formal and semantic units. Here, we return to a four-beat line in which the final, stressed word (“world”) echoes the sounds that precede it in the line and rhymes with the ending of the second line. The gnomic quality of this closing sentence seems to position it as a universal truth, perhaps along the lines of the poem’s Heraclitean epigraphs. In this sense, we have passed through the temporal tenses to the space of the eternal. Yet there is something uneasy about the closure on offer here. The two stanzas almost seem to stand on their own, offering a formal movement from order to disorder and back again and a statement that seems to answer to the questions provoked by the first two lines. But there are some curious semantic gaps: with the sun gone, why would the sunflower turn to “us”? Why would the climbing clematis and the yew both curl “down” to us? We might guess that this is because “we” are a source of light, but the poisonous yew, with its “Chill / Fingers” and homophonic suggestion of a “you,” frustrates our association of the first-person plural with light and life. What we have here is not so much a narrative or a lyric as a series of rhetorical fragments: a sing-song couplet, two riddle-like questions, and a statement that is about an answer 100 (of light to light) but not an answer to the questions posed. The gaps at the ends of the increasingly enjambed lines encourage us to feel this disjunction, while the sound patterns offer a different mode of meaning-making. The repetition of sounds, rhythms, words, parts of words, and lines call the past into the present and also—because they necessarily occur in a non-identical context—remind us that what they repeat is incommensurable, not retrievable in repetition. The irregular meter emphasizes that change is our only reliable expectation: this is not chaos, in which “change” itself would be meaningless, but a network of potentialities that encourages the rhythmization of consciousness. The temporality of this section is not the normative timeless now of lyric nor the chronologically mobile now that narrative manipulates in the service of plot; it is a “now” moving through time that does not privilege what we think of as the present. To be in time here is to return to a past that we know will be revised, and to generate expectations about a future that resists prediction. VI. Poetry, Rhythm, Genre To understand whether Four Quartets produces a different kind of thinking about and through time because of its poetic qualities or because of something else—Eliot’s religious or political views, say, or his overarching personal literary “style”—we can look to his contemporaneous publications in other literary genres. Eliot’s serious investment in the genres of poetry, drama, and literary criticism makes his work an ideal test case for the question of the specific aesthetic and temporal effects of particular literary forms. The case is made all the more interesting for the interrelatedness of these genres and the fluidity of their boundaries, especially between poetry and drama. In her 1936 review of Eliot’s Collected Poems 1909-1935, Marianne Moore characterized his art as one “which from the beginning has tended toward drama” (359–360); in a review the same year, Edwin Muir noted that his “dramatic approach [had] influenced the form of poetry away from the purely lyrical” (354). Robin Grove, commenting on Eliot’s drama, argues that we might even “consider the printed verse as itself a series of gestures, speeches, quasi-theatrical occasions—beginning with the titles (‘The Love Song of J. Alfred 101 Prufrock’), and going on through the monologues, ensembles, scenes in the pub, the rose-garden, and the rest” (160). Four Quartets, although arguably more lyrical than some of Eliot’s early verse, is itself a product of this genre-bending tendency and an elaboration of some of his dramatic themes: in fact, the first fourteen lines of “Burnt Norton” were originally intended as a speech for the Second Priest in the verse drama Murder in the Cathedral.18 Eliot’s 1951 essay “Poetry and Drama” takes up the question of the relation between these two genres, and especially of the value of writing drama in verse, by evaluating his own theatrical projects of the 1930s and 1940s. He starts with the assumption that poetry written for the stage “must justify itself dramatically, and not merely be fine poetry shaped into a dramatic form” (132). The converse of this is true as well: not all drama written in verse qualifies as poetry. Verse drama “will only be ‘poetry’ when the dramatic situation has reached such a point of intensity that poetry becomes the natural utterance, because then it is the only language in which the emotions can be expressed at all” (134). When this effect is achieved, it is through “a kind of musical design. . . which reinforces and is one with the dramatic movement” (136) to such an extent that the audience remains unconscious of the poetic quality of the lines. And yet, in the ideal poetic drama, this poetic design is entirely necessary, because the drama deals with something beyond “the nameable, classifiable emotions and motives of our conscious life” (146). This “peculiar range of sensibility”—a “fringe...of feeling which we can only detect, so to speak, out of the corner of the eye”—“can be expressed by dramatic poetry, at its moments of greatest intensity” (146). The ideal verse drama thus “touch[es] the border of those feelings which only music can express,” but only the border; “we can never emulate music, because to arrive at the condition of music would be the annihilation of poetry, and especially of dramatic poetry” (146). In “Poetry and Drama,” Eliot is essentially writing from the point of view of the poet; the emphasis is on the advantages that poetic effects may lend to drama, with little attention given to the way that drama might enhance what we think of as non-dramatic lyric poetry. How far can we 18 See Gardner 79. 102 extend this limited argument about the benefits of mingling genres? What Eliot had in mind was not simply a conflation of form; “no play,” he remarked, “should be written in verse for which prose is dramatically adequate” (“Poetry and Drama” 132). Nor was it the translatability of content; after all, there was presumably some reason why “Burnt Norton” had to become a poem. In 1953, in an essay called “The Three Voices of Poetry,” Eliot comes closer to defining the possibilities that are opened up or foreclosed when poetic and dramatic voices interact. The first voice, he theorized, was “the voice of the poet talking to himself—or to nobody”; the second “the voice of the poet addressing an audience”; and the third, “the voice of the poet when he attempts to create a dramatic character speaking in verse” (96). The point of distinguishing these voices is not to draw hard and fast lines between the kinds of verses in which each is dominant, but to show that almost every “real poem” contains more than one voice. At one end of the spectrum, “if the author never spoke to himself, the result would not be poetry, though it might be magnificent rhetoric”; at the other, “if the poem were exclusively for the author, it would be a poem in a private and unknown language,” and this “would not be a poem at all” (109). In order to refute the idea of a truly private poetry, Eliot has to distinguish what he means by poetry in the “first voice” from “what we call loosely ‘lyric poetry’” (105). His critique of lyric here is surprisingly consonant with Jackson’s recent comments on lyricization. By the 1950s, the normative definition of lyric had already expanded beyond the kind of poetry that was sung or set to music to “poetry that was never intended for a musical setting,” such as the verse of the metaphysical poets (105). The unsatisfactory OED definition of “lyric,” which Eliot quotes here, is “now the name for short poems, usually divided into stanzas or strophes, and directly expressing the poet’s own thoughts and sentiments” (105). But, as he notes, “there is no necessary relation between brevity and the expression of the poet’s own thoughts and feelings”; if all such poems qualify as “lyric,” then it is so broad a category as to be meaningless. When Eliot speaks of lyric in association with the first voice of poetry, then, he is referring to the kind of lyric that expresses personal feeling; length has nothing to do with it (in fact, because of this semantic confusion, he 103 prefers the term “meditative verse” (106)). Even when we work with this more limited definition of lyric in the first voice, the poem cannot be reduced to a solipsistic utterance. The moment the author turns his work over to a set of unknown readers, Eliot argues, is “the consummation of the process begun in solitude and without thought of the audience, the long process of gestation of the poem, because it marks the separation of the poem from the author” (109). A verse that never completes this process of separation is not what we could properly call a poem at all. And yet: despite Eliot’s praise of the effect of poetry in drama, his claim for the necessary mixture of the three poetic voices, and his refusal to recognize as poetry a verse without an audience, his criticism maintains a firm distinction between poetry and drama as genres. “In writing verse for the stage,” he says, “both the process and the outcome are very different from what they are in writing verse to be read or recited” (“Three Voices” 102). There is not only a “difference,” in fact, but an “abyss” between the two (102). The nature of this abyss has everything to do with form; this includes what we have been discussing as rhythm, and, of course, what both form and rhythm do to our experience of time. As early as 1921, Eliot suggested that a poet’s serious philosophical interests had a place in poetry so long as the poet “turn[s] them into poetry, and not merely meditate[s] on them poetically” (65). Thirty years later, the abyss between poetry and any other verse text still turned on the difference between poetry qua poetry and poetically treated material. In non-dramatic poetry, the impetus to the poem lies in that “fringe” of unclassifiable feeling that we can access only on the border between sense and sound. This kind of a poem simply “replace[s]” the “‘thing’ for which the words had to be found” (106), a ‘thing’ which cannot even be identified “until it has been transformed into an arrangement of the right words in the right order” (106). In this sense, “the ‘psychic material’ tends to create its own form” (110), but in a manner that is, in retrospect, mutually constitutive; there is always “a simultaneous development of form and material” (110). By contrast, dramatic or quasi-dramatic poetry can access the “fringe” of feeling only at its height, and it remains motivated by action; moreover, its form “is to some extent given” (110). 104 How wide can the abyss really be between what Eliot calls “poetry” and material that is already treated poetically—that is already, in fact, written in verse? Consider the following passage from Murder in the Cathedral, one of Eliot’s earliest theatrical projects and the original context for the lines with which “Burnt Norton” begins. Here, the First Tempter, a friend from the past, reminds Thomas Becket of the pleasure he formerly took in his friendship with Henry II, and encourages him to accept the king’s terms for peace: Thomas: You talk of seasons that are past. I remember Not worth forgetting. Tempter: And of the new season. Spring has come in winter. Snow in the branches Shall float as sweet as blossoms. Ice along the ditches Mirror the sunlight. Love in the orchard Send the sap shooting. Mirth matches melancholy. Thomas: We do not know very much of the future Except that from generation to generation The same things happen again and again. Men learn little from others’ experience. But in the life of one man, never The same time returns. Sever The cord, shed the scale. Only The fool, fixed in his folly, may think He can turn the wheel on which he turns. (184) These lines are typical of the versification of the dialogue in Murder in the Cathedral, which Eliot later described as “the versification of Everyman,” in that they are characterized by “an avoidance of too much iambic, some use of alliteration, and occasional unexpected rhyme” (“Poetry and Drama” 139–140). They are typical, too, in their content: in this play, whose events are somewhat limited—Eliot later summarized the plot as, “A man comes home, foreseeing that he will be killed and he is killed” (“Poetry and Drama” 140)—the dialogue tends more toward persuasion and introspection than toward the kind of communication that motivates action. Nor are the themes or the images specific to drama: the rejection of progress and prediction is echoed thematically in Four Quartets, as are the images of the spring that comes in winter (the “transitory blossom / Of snow” in “Midwinter Spring” in LG I, 15-16) and the wheel of life (the “bedded 105 axle-tree” around which living patterns revolve in BN II, 48). What distinguishes the dialogue of Murder in the Cathedral as drama, then, has less to do with the language itself than with its generic presentation. Lines that could be excised and presented as poems or parts of a poem are here put in conversation with each other and made part of a dramatic narrative. On this reading, Murder in the Cathedral would seem to call into question the divide that Eliot insists exists between dramatic verse and the lyric verse that we call non-dramatic or non-narrative. But there are at least two reasons why such a conclusion is problematic. First, by the 1950s Eliot had determined that Murder in the Cathedral was, as a verse drama, a “dead end” (“Poetry and Drama” 139), not least because the historical remoteness and religiosity of its subject matter made the audience more likely to “put up with poetry” (139), when they should really “be made to hear it from people dressed like ourselves, living in houses and apartments like ours, and using telephones and motor cars and radio sets” (141). Second, and more important, the experience of hearing these lines in the theater would be substantially different from the experience of reading them on the page, either as part of the script or set off independently as a poem. That these experiences are different is obvious—watching a play will always be different from reading it—–but what most distinguishes the dialogue in Murder in the Cathedral when it is read as poetry has to do with what happens at the enjambed ends of lines. The space that follows “I remember” refuses to be filled—what is not worth forgetting?—and nags at us in a way that the grammar of the sentence spoken aloud would not. The Tempter’s enjambment, which keeps propelling us to the next line in search of the action, the verb, contributes visually to his argument for Thomas’s future with the king. When Thomas’s resolved, end-stopped lines slide into lines of increasingly strong enjambment, rhyme, and rhythm, the logic of his argument also shifts at a non-semantic level. The fact that these lines were written as theatrical dialogue does not mean that they cannot be read as poetry, but their poetic effect will not be complete unless we present them visually, as a poem. The problems that Eliot identified in Murder in the Cathedral as a drama—its historical 106 setting, its lack of action, its tendency toward noticeably poetic idiom—are consciously stripped from his later dramatic works. In The Cocktail Party, a social drama very loosely based on Euripedes’s Alcestis, Eliot resolved “to avoid poetry which could not stand the test of strict dramatic utility: with such success, indeed, that it is perhaps an open question whether there is any poetry in the play at all” (“Poetry and Drama” 144). He also, he says, “tried to keep in mind that in a play, from time to time, something should happen; that the audience should be kept in the constant expectation that something should happen; and that, when it does happen, it should be different, but not too different, from what the audience had been led to expect” (144). The passage below comes at the end of the play; Edward and Lavinia Chamberlayne, reunited, and having received the advice of the enigmatic psychiatrist Reilly and their friend Julia Shuttlethwaite, are about to host a second cocktail party: Edward: I never shall learn how to pay a compliment. Lavinia: What you should have done was to admire my dress. Edward: But I’ve already told you how much I like it. Lavinia: But so much has happened since then. And besides, One sometimes likes to hear the same compliment twice. Edward: And now for the party. Lavinia: Now for the party. Edward: It will soon be over. Lavinia: I wish it would begin. Edward: There’s the doorbell. Lavinia: Oh, I’m glad. It’s begun. The poetry in the Chamberlaynes’ dialogue, if it is poetry, is less formally poetic than the dialogue between Thomas and his Tempter: the rhythm is very close to the rhythm of prose speech, the language is almost entirely free of alliteration or rhyme, and most of the lines are end-stopped. There are no imagistic or verbal echoes here of Eliot’s lyric work. However, the theme would be quite at home in Four Quartets: in the temporal world, repetition is necessary even as pure return is impossible (“so much has happened since then”). We can anticipate the temporal event (“Now for the party”), anticipate its end (“It will soon be over”), and recognize it once it has passed (“Oh, I’m glad. It’s begun”), but the event itself, the pure presence of it, eludes 107 our grasp; the beginning of the second cocktail party, if we can mark it at all, takes place in the non-verbal tone of the doorbell during the play’s performance. The coincidence of theme does not mean that Eliot has turned this concept into poetry rather than “meditate on [it] poetically” (“The Metaphysical Poets” 65). But the semantic content does turn our attention to the repetitions of the language itself: and again, to be attentive to the full force of these repetitions, we need to read the lines as they appear on the page. In particular, the last, short exchanges between Edward and Lavinia divide the lines between them, so that the transformation from anticipation to retrospect, mediated by phrases that we hear as repeating with a difference what has already been said, takes place in the visual gaps. VII. Conclusion: “. . . yet the words sufficed / To compel the recognition they preceded” (BN II, 48–49) What does it mean, finally, to say that the typological mode of Four Quartets depends upon the way it thinks as a poem? Eliot wrote that “a philosophical theory which has entered into poetry is established, for its truth or falsity in one sense ceases to matter, and its truth in another sense is proved” (“The Metaphysical Poets” 65). In this chapter, I have been concerned with this “other” sense, the way in which the linguistic and extra-linguistic properties of poetry participate in a kind of thinking that is not entirely translatable to other genres. We can only detect the argument that this thinking produces if we accept that it is not reducible to the poem’s paraphrasable content. In the case of Four Quartets, this means that we must be open to the possibility that Eliot’s engagement with time might be something other than anti-historical. At the non-semantic level, the typological orientation of Four Quartets makes our experience of the poem one in which each sound and unit of sound seems reciprocally determined. Each new word surprises us but also seems, in retrospect, uniquely necessary. This operation, which transforms some of the insights of Bergsonism, corresponds to what Abraham calls “rhythmizing consciousness.” The kind of thinking that it produces animates the past that we 108 carry with us so that, at any moment, its significance in the network of the temporal realm might shift. The instability of this copresent past lends it a plenitude of possibility that we normally associate only with the future: with it, we shed our temporal “provincialism” (“What is a Classic?” 130) and formally re-enfranchise the dead. The thinking about and through time that takes place in the lyric verse of Four Quartets cannot be faithfully replicated by either narrative or drama. This would be true even if the poem were paraphrased, in a novel, as a lecture, or if it were presented as a series of monologues or spliced into the dialogue in a play. However, to insist that the status of the text as a long lyric poem makes a difference, we do not need to define or enforce strict boundaries between genres. Eliot’s work, from his dramatic monologues to his verse plays, his didactic lyrics to his literary social criticism, gives the illusion of generic isolation the lie in any case. When we read his verse, or any verse, as a lyric poem—whether or not this verse has narrative or dramatic qualities—we enter into a temporal engagement with the text that enables specific modes of thinking: “Every phrase and every sentence is an end and a beginning, / Every poem an epitaph” (LG V, 224-225). The non-semantic aspects of the language of a lyric poem are activated in a way that encourages us to consider the simultaneous interactions of different patterns; the endings of lines give us the space to weigh these networks and, through the rhythm that we both perceive and project, to anticipate their direction. C HAPTER F OUR Repetition and Revision in the Poetry of W. B. Yeats 110 A writer must die every day he lives, be reborn, as it said in the Burial Service, an incorruptible self, the self opposite of all that he has named “himself.” – Dramatis Personae, 1896-1902 (1935) Talk to me of originality and I will turn on you with rage. I am a crowd, I am a lonely man, I am nothing. Ancient salt is best packing. – Introduction for the Scribner Edition of Yeats’s Collected Works (1937) I. Introduction Time in the poetry of W. B. Yeats is a slippery, unstable, many-versioned thing. There is the verse of the young Yeats, precociously nostalgic, who, at the age of 24, wrote of an amorous couple who had reached “the hour of the waning of love”1 and a lover who, now “full of tears,” regrets the time when he “was young and foolish.”2 A few years later, Yeats, now a champion of the Irish Literary Revival, declared folk art, an art of memory, “the soil where all great art is rooted.”3 In addition to recording Irish folktales, he set out to retell the stories of Irish heroes in verse and verse drama. Beside both of these is the Yeats who was fascinated with the occult and convinced of the presence of past spirits; this was the poet who saw history coming to the end of an era, who prophesied a brilliant, violent turn in time to come, and who theorized strange and complex patterns linking events and people across time. Near the end of his life, there was also the Yeats who looked back on these various selves with disillusionment, sorrow, and some compassion. It would be difficult enough to track the thematic, philosophical, and political ruptures in Yeats’s thought, and the way that these complicate any unified idea of time we might try to draw from his poetry, if the composer of the poems were not always also their reviser. However, the poet being what he was, and the poems being what they are—verses revised, 1 “The Falling of the Leaves,” 5. 2 “Down by the Salley Gardens,” 8. The tone of lamentation that Yeats favors in this period reflects his selection of source materials and themes, if not his personal feelings. In a note to the 1889 version of this poem, Yeats writes: “This [poem] is an attempt to reconstruct an old song from three lines imperfectly remembered by an old peasant woman in the village of Ballysodare, Sligo, who often sings them to herself” (quoted in the Norton edition, p. 10). 3 From “By the Roadside” (1901), in The Celtic Twilight, p. 191. 111 rewritten, and recontextualized, often in the service of Yeats’s evolving idea of his life’s work—it is impossible to trace the conceptualization of time in Yeats’s lyric poems without considering his practices of revision. Over the course of his career, Yeats revised his poetic work in two qualitatively different ways. First, he edited, altered, and sometimes entirely rewrote the verses of individual poems. This kind of revision took place both during and after the initial period of composition. In 1932, when he was 67, Yeats wrote to Olivia Shakespear, “I spend my days correcting proof . . . . I keep saying what man is this who in the course of two or three weeks . . . says the same thing in so many different ways.”4 The manuscript record confirms this account of his compositional process, which, if not tortured, was certainly exacting. Yeats routinely went through several manuscript versions—as many as 185 —before publishing a given poem for the first time. Sometimes this process of revision also involved a shift in genre: the ideas for many poems first appeared in Yeats’s notebooks as prose themes,6 while other poems began, like Eliot’s “Burnt Norton,” as part of a longer dramatic verse. For the vast majority of these poems, the event of publication did not mark the end of the revisionary process, but rather the beginning of a new cycle of alterations. Yeats’s post-publication changes were sometimes limited to a poem’s title, to a few words, or to a few key lines. But many poems were heavily revised or rewritten for subsequent publications, and continued to attract Yeats’s editorial hand decades after their original appearance in print. The other major form of revision in Yeats’s poetry comes into view at the level of the book. In addition to drafts of the poems, Yeats’s notebooks contain provisional versions of the 4 Letter to Olivia Shakespear, June 30, 1932. Collected Letters, Unpublished Letters (1905-1939), 5692. 5 In his introduction to The Early Poetry, George Bornstein calls attention to Yeats’s persistent habit of revision, and observes: “The previously unpublished poem ‘Pan’ offers perhaps the fullest surviving example of the evolution of a lyric: nine manuscript versions survive, each with revisions, to yield a total of eighteen still extant stages” (2). 6 Curtis Bradford notes: “Yeats almost always began work on a poem by composing what he called a ‘sketch’ or ‘subject’ in prose. These subjects state the content of the poems and note the principal images to be developed in them” (Work 4). Unfortunately, not all of these sketches have survived. Wayne K. Chapman observes that Yeats “sometimes recorded his themes in a diary or journal before setting himself to the hard work of transforming them into verse. . . . In most cases, however, if there ever was a prose ‘theme’ to match the poem that grew out of it, those themes have been lost, being the most perishable of pre-texts” (26). 112 order in which these poems would appear in various volumes. As with the individual poems, each new, individual book of poetry marked both an end and a beginning—a fulfillment and a figure—in Yeats’s artistic process. Beginning with the 1895 Poems, Yeats periodically issued collected volumes of his lyric work which both altered the lists proposed in previous books and omitted some of the poems on those lists. The significance of these recontextualizations is such that Hugh Kenner suggests, in “The Sacred Book of the Arts,” that “the unit in which to inspect and discuss [Yeats’s] development is not the poem or sequence of poems but the volume, at least from Responsibilities (1914) to A Full Moon in March (1935)” (578). George Bornstein extends this argument to Yeats’s earliest books of poetry, noting that “Yeats carefully arranged and rearranged their contents into meaningful wholes, with the next context exposing new meanings (or new fault lines) for any given poem as well as for the entire sequence” (Vol. II, 2). This self-conscious patterning took on more urgency, and a more self-consciously narrative shape, as Yeats grew older. James Pethica argues that Yeats’s “last sequence of poems served him as a personal meditation on mortality, but also as a carefully managed dramatic narrative which might condition critical responses after his death” (xxiii). In fact, the last thing that Yeats wrote before his death, Pethica suggests, may have been a draft “List of Contents” for his final book of poems (xxvi). In this chapter, I examine the ways in which Yeats’s practices of revision both participate in the temporal arguments of specific lyric poems and suggest a typological structure for his work at the level of the career. In particular, I focus on three lyrics that represent different modes of revision and distinct moments in the development of Yeats’s style. First, “The Song of the Happy Shepherd” (1885) allows us to consider what happens when Yeats changes the genre of a text; when he alters the verse in the process of composing it, in the first printed version, and in later printed versions; and when he chooses to reproduce the poem in different contexts and in different orders relative to other poems. “The Lamentation of the Old Pensioner” introduces a more radical kind of revision. Yeats first published this poem, then titled “The Old Pensioner,” in The Scots 113 Observer (November, 1890); editions published in later years included only minor changes until, for Early Poems and Stories (1925), Yeats rewrote the entire poem and added a third stanza. Finally, the manuscript record of “The Circus Animals’ Desertion” (1939) offers us a view of Yeats’s revisionary process at the end of his career in verses that examine the difficulty of maintaining a perpetually creative mode and presents a retrospective judgment of the poet’s most famous “old themes.” In these poems, revision is related to practices of formal repetition in individual poems, and also to form as repetition. Yeats’s choices in rewriting, repeating, and recontextualizing his work point to a scale of reading in which every revision and every new poem contributes to a larger figural pattern. II. Critical Contexts The textual criticism in the decades since Yeats’s death, in 1939, has made it clear just how problematic it is to attempt to identify an authoritative version of any individual poem or collection of poems in Yeats’s oeuvre. When faced with multiple versions of a poem, each approved and published by the poet at a different point in time, how is an editor to choose the one version that should be read today? One popular strategy is to defer to Yeats’s last known intention, whether in the form of the last published version of a poem or the last authorized changes made to a manuscript draft. Richard J. Finneran, who oversaw the 1983 and 1989 editions of Yeats’s Collected Poems, explains that in these volumes he “tried to present Yeats’s poems as they stood at the time of his death, in terms of both text and arrangement” (“Text and Interpretation” 18). He notes that his edition “was not offered—and could not have been offered—as a new ‘Definitive Edition”: by the 1980s, “the very concept of such an artifact had been largely discredited,” a problem compounded, for any critic hoping to circumvent the difficulties in establishing intention and authority, by the poet’s “notorious handwriting” as well as “his lifelong habit of revising and rearranging his verse” (18). Still, he maintains that his best option as an editor was to try to “reproduce what Yeats in fact did or clearly intended to do, rather than what others assumed he 114 would have done—or should have done” (18–19). This means, for instance, that he follows Yeats’s idiosyncratic punctuation rather than the changes various editors made to it (26–27). Finneran’s editorial philosophy, although dominant in much Yeats criticism, is not the only recent approach to his work. In his introduction to the Early Poetry, Bornstein notes that “more traditional textual scholars like Donald Reiman or Stephen Parrish still argue against overvalorizing the final state of a text,” with Reiman preferring “versioning” to “editing” (Vol. II, 3). Along these lines, early textual studies of Yeats, including Thomas Parkinson’s W. B. Yeats, Self-Critic (1951) and W. B. Yeats: the Later Poetry (1964), Jon Stallworthy’s Between the Lines (1963) and Vision and Revision in Yeats’s Last Poems (1969), and Curtis Bradford’s Yeats at Work (1965), shed light on Yeats’s creative process by reconstructing the manuscript development of his poetic works. More recently, Chapman has updated this project in his essay collection, Yeats’s Poetry in the Making (2010), which argues that an organic principle governs the development of Yeats’s work; “from a rudimentary ‘thought’ worked out in prose, Yeats’s poem grows and becomes a text in a canon” (13). The work of these textual scholars has informed—and overlaps with, to some degree—the principles of French genetic criticism. In their introduction to this school of thought, Daniel Ferrer and Michael Groden write that, “for geneticists, instead of a fixed, finished object in relation to which all previous states are considered, a text becomes—or texts become—the contingent manifestations of a diachronous play of signifiers” (5). The difference between their work and traditional textual scholarship is that “a textual critic will tend to see a difference between two states of a work in terms of accuracy and error or corruption, whereas a genetic critic will see meaningful variation” (11). Finally, in their discussions of textual criticism in the Cornell editions of Yeats’s manuscripts, both Bornstein and Pethica point to another line of thinking, championed by James McGann, which views the text and its authority as social phenomena, not products of an individual intention. Pethica remarks: “In the case of Yeats—an inveterate and almost obsessive reviser, for whom the self and the self represented in the text were always in complex dialogue—the traditional notion of intentionality is one that 115 should always be applied warily” (xxix). My approach, keeping this latter caution in mind, draws from elements of several of the strategies described here. I treat the revisions to and reconfigurations of Yeats’s poems as operations in a career-long project that is itself subject to change over time. In this schema, the later version of any given work is not necessarily to be privileged in itself; rather, it both records and enacts a change in the typological structure of Yeats’s oeuvre and of the aspects we come to view as “necessary” in his life work. To understand these changes, I rely on several scholarly accounts of Yeats’s creative process, especially George Bornstein’s edition of The Early Poetry (Vols. I and II) and James Pethica’s Last Poems, both part of the Cornell series of Yeats’s manuscript materials, and Warwick Gould’s “Writing the Life of the Text: The Case of W. B. Yeats,” which applies some of the principles of genetic criticism to the textual history of “The Lamentation of the Old Pensioner.” My goal in this project is not to suggest that the final versions of the poems are necessarily inevitable, authoritative, or aesthetically superior to any other possible version of the poem. However, every provisional end is important insofar as it reconstitutes the figural relations among the versions that precede it. Unlike genetic critics, I do not aim to level the hierarchy of virtual and received texts, although I am sympathetic to the ambition of genetic criticism to “discover a plurality of virtual texts behind the surface of the constituted text” (Hay 22). Rather, I am interested in what it means to treat revision as a form of repetition that operates in a field with an ever-shifting horizon (the repetition is, in fact, what shifts that horizon). Put differently, I want to trace Yeats’s poetic thinking as a process that takes place over the textual life of the poem. By examining Yeats’s practices of revision in his poems, I seek to engage not just with the textual criticism on Yeats but with other critical accounts of his understanding of time. One prominent line of scholarship considers Yeats’s treatment of time in the service of a larger argument about his political orientation. The arguments in this vein, which often come to opposing conclusions—Jonathan Allison notes that Yeats “has been described variously as fascist, 116 patriotic Tory, colonial, and ‘establishment’ nationalist” (186)—reflect the fact that Yeats adopted very different and sometimes conflicting political ideas at different points in his career. At one end of the spectrum is Conor Cruise O’Brien, who argued in 1965 that Yeats’s late poems “had often a political theme, sometimes a political intent,” and that his politics “were not fundamentally inconsistent, vague or irrelevant to his ‘real self’ and were, in his maturity and old age, generally pro-Fascist in tendency and Fascist in practice on the single occasion when the opportunity arose” (48). Elizabeth Cullingford, responding in large part to readings like O’Brien’s in Yeats, Ireland, and Fascism (1981), argued that there is a “sharp distinction between [Yeats’s] willingness to take political risks and try out extreme positions, and the foundation of consistent and reasonable belief to which he always returned” (234). In The Political Aesthetic of Yeats, Eliot, and Pound (1991), Michael North placed these and other critical accounts of Yeats’s politics in the context of a larger argument about the relation of aesthetic modernism to the Enlightenment project of modernity. His chapter on Yeats proposes that the poet’s philosophy is ultimately incompatible with Fascism, but not because of any liberal tendencies; rather, “despite the hints in the table7 that fascism attracts Yeats because of its greater respect for both sides of the social and political dichotomies, he finally rejects or supersedes fascism because of a conviction that these cannot be equally served” (73). In other words, Yeats’s deep-seated belief in the “eternal condition” of human conflict and the “unchangeable” results of the modern project—the belief that there is “nothing for politics to do in the future it foresees”—precludes any possibility of the reconciliation that fascism might promise (73). Not surprisingly, the critics that reach different political conclusions about Yeats also find different temporal emphases in his work: his support of Irish national literature can be seen variously as indicative of a romanticization of the past or of a progressive hope for the future; his fascination with the occult, as a sign of his wish to commune with the past or his desire to access a means of prophecy; his theory of the gyres, as a totalizing view of history or a revolutionary 7 North refers to Yeats’s “Genealogical Tree of Revolution,” “a table reflecting notes he made on politics during 1933” (71). 117 one. In particular, much of the critical disagreement hinges on the interpretation of Yeats’s theorization of repetition. From very early on in his career, Yeats posited a relationship between repetition and the eternal. See, for instance, the last stanza of “Ephemera” (1889), which suggests that eternity is achieved not by what endures but by what repeats: The little waves that walked in evening whiteness, Glimmering in her drooped eyes, saw her lips move And whisper, “The innumerable reeds I know the word they cry, ‘Eternity!’ And sing from shore to shore, and every year They pine away and yellow and wear out, And ah, they know not, as they pine and cease, Not they are the eternal—’tis the cry.” In his prose, Yeats variously suggested that repetition acted as the return of the past and as a prophetic device; his fascination with the possibility of lived repetition prompted both delight in the perfection of an afterlife constituted by living life backwards8 and the belief that it was necessary to live as if Ireland’s war “were to go on endlessly.”9 In the last decade of his career, Yeats began to group these ideas explicitly under the rubric of Nietzsche’s eternal return.10 For example, in “A Dialogue of Self and Soul” (1933), the “Self” declares: “I am content to live it all again / And yet again” (57–58); “I am content to follow to its source / Every event in action or in thought; / Measure the lot; forgive myself the lot!” (65–67). In “Purgatory,” a short play first produced in 1938, the titular mode of afterlife is one in which the dead “Re-live / Their transgressions, and that not once / But many times, they know at last / The consequence of those transgressions / Whether upon others, or upon themselves” (32–35). The political implications of 8 See, e.g., in The Trembling of the Veil (1922): “When we are dead according to my belief, we live our lives backward for a certain number of years, treading the paths we have trodden, growing younger again, even childish again, till some attain an innocence that is no longer a mere accident of nature, but the human intellect’s crowning achievement” (245). 9 Unpublished manuscript, quoted in Richard Ellmann, Yeats the Man and the Masks, p. 117. Ellmann presents this work as a continuation of the ideas Yeats expressed in a 1904 speech. 10 Although references to something like the eternal return are especially prominent in Yeats’s work in the 1930s, he was introduced to the idea as early as 1902, when New York lawyer and patron John Quinn sent him a copy of Thus Spake Zarathustra. See the chronology in the Collected Letters, Vol. III (xix). 118 this idea depend upon whether, for Yeats, the operation of eternal return was essentially deterministic or part of a process that could eventually lead to transformation.11 In the context of literary modernism, Yeats stands out both for the longevity of his poetic career, spanning the 1880s to the 1930s, and for the metamorphoses which marked his style and subject matter over the course of these decades. Despite his claim that in his revisions he seems to say “the same things in so many different ways,” and his speculation that his works may have but a “sole theme,”12 it seems clear that Yeats’s repetitions of the “same things” are always also transformative. It is this tension in his work—the insistence on the eternal and unchanging that is nevertheless subject to perpetual revision—that requires the range of critical contexts that I have reviewed here.13 The thematic addresses to time and eternity throughout Yeats’s work draw attention to the temporal operations of his poetic form. Any formal argument, however, would be incomplete without a consideration of Yeats’s practices of revision, rewriting, and recontextualization, all of which participate in the poem’s material history as forms of repetition. Moreover, the interaction between internal form and content and the revisions that take place across time affect the inferences we would draw about the relationship between Yeats’s theorization of time and history in his poems and his political orientation. In what follows, I want to suggest that reading Yeats’s 11 Ellmann argues that the problem posed by the eternal return for Yeats—that “the soul. . . could not be expected to try to attain beatitude through escaping permanently from the wheel”—led him to turn to a new theory, the “Unity of Being,” which could better unite art and life (268). Without discounting the importance of this new model in Yeats’s later work, I would suggest that the eternal return retains its importance for Yeats throughout the 1930s. See, for example, “Purgatory” (quoted above), “The Gyres” (“Those that the Rocky Face holds dear, / [. . . ] shall [. . . ] / . . . disinter / The workman, noble and saint, and all things run / On that unfashionable gyre again” (18, 19, 22–24)), and “Lapis Lazuli” (“All things fall and are built again / And those that build them again are gay” (35–36)), all works dating from the last year of his life. 12 Letter to Olivia Shakespear, June 30, 1932. Collected Letters, Unpublished Letters (1905-1939), 5692. 13 Yeats’s engagement with the concept of the eternal return suggests several other critical lenses. In addition to Nietzsche’s introduction of this idea in The Gay Science (1882) and in Thus Spake Zarathustra (1884), Jacques Der- rida’s work considers the repeatability of the signifier (e.g., in Of Grammatology (1967) and introduces the concept of hauntology (in Specters of Marx (1993)). Gilles Deleuze, in Difference and Repetition (1968), proposes a theory of repetition that engages explicitly with Nietzsche, and J. Hillis Miller, in Fiction and Repetition (1982), uses Deleuze’s notion of Nietzschean repetition to identify non-mimetic repetition in various novels. However, Auerbach’s concept of figural interpretation remains the most relevant to the connections within and among Yeats’s lyric poems that I am interested in describing here. 119 repetition as a kind of revision, and vice versa, can help us understand his approach to words in time. III. From “Epilogue” to “Song” Yeats’s practices of revision and repetition naturally seem more prominent later in his career, when he had developed a broader body of work from which to draw. But the manuscript and print history of even his earliest work shows that revision was always already a part of Yeats’s creative process. Take, for instance, “The Song of the Happy Shepherd,” originally published in 1885.14 “Song” is the opening poem in “the expanded version of The Collected Poems (1933), which as of June 22, 1937, Yeats had planned to publish ‘in about two years’ time’” (Finneran, “Preface,” xxiii). The most recent edition of Yeats’s Collected Poems, which is based on this edition, reflects this order, making “Song” the first poem in the volume. The story of how “Song” came to occupy this position in Yeats’s collected works is a complex one, involving a series of transformative revisions in the manuscript and print versions of the verses themselves and in their juxtapositions with the poems that Yeats chose at various times as their context. However, to understand how the revision and recontextualization of this poem participates in Yeats’s poetic thinking, the print history is worth laying out at length. The most obvious difference between the lyric poem that opens the Collected Poems and its early manuscript draft is a generic one. The draft began with stage directions: “enter a Fawn / Holding a shell.” In the October, 1885 issue of the Dublin University Review, the poem was positioned as “An Epilogue” to two pieces of dramatic verse that had been published in the same journal earlier that year, “The Island of Statues” and “The Seeker” (published in the April–July and September 1885 issues, respectively). In its next print appearance, in The Wanderings of Oisin and Other Poems (1889), Yeats reframed “An Epilogue” as a lyric poem and changed its 14 For the manuscript history of this poem in what follows, I rely largely on George Bornstein’s edition of The Early Poetry: Manuscript Materials. 120 title to “Song of the Last Arcadian.” The new title preserved the Arcadian setting of the verse drama that it originally followed and announced, like “Epilogue,” the sense of an ending. However, the finality suggested by both titles is complicated by the order of the poems in WO. “Island of Statues,” Yeats’s first published work, is the last of 32 poems in this book; “Song of the Last Arcadian,” composed as its epilogue, precedes it as the tenth poem in the volume. This order marked an explicit change from the order of publication and also from two lists of poems—presumably an early plan of the order of the works in WO—that Yeats kept in a notebook along with a draft of “The Wanderings of Oisin.”15 In the first of those lists, “Epilogue” was the penultimate poem, appearing after (although not immediately next to) “Island of Statues.” When the poem was next published, in Poems (1895), it appeared under its final title, “The Song of the Happy Shepherd,” and it opened the last section in that book, “Crossways.” The division of the poems in this collected volume into “Crossways,” which comprised Yeats’s earliest poems, and “The Rose,” which contained poems mostly from The Countess Kathleen and Various Legends and Lyrics (1892), is preserved in later collections, with one important difference. In the 1895 version, the two sections are in reverse chronological order, with “The Rose” preceding “Crossways”; in the 1933 collected volume, the chronological order is restored. Thus, “The Song of the Happy Shepherd,” originally composed as the final monologue in Yeats’s first published verse drama, becomes the opening poem in the last collected volume of his work. In his introduction to the manuscript version of “The Island of Statues,” Bornstein notes that Yeats’s choices with respect to the positioning of “Song” effectively make the play it followed “the suppressed origin of his poetry” (9). At the same time, “Island” itself was both an end and a beginning, as it “mark[ed] the final work in the series of mostly unpublished productions which Yeats recalled in 1914 when preparing Reveries over Childhood and Youth”: “I had begun to write poetry in imitation of Shelley and of Edmund Spenser, play after play—for my father exalted dramatic poetry above all other kinds—and I invented fantastic and incoherent plots” (9). 15 See “Appendix Two: Revisions in the Tables of Contents through 1895,” in The Early Poetry, Vol. II. 121 Whether we regard “Song” as a sort of stand-in for “Island,” representing the origin of Yeats’s poetry, or as the lyric conclusion of this early series of dramatic verse works, the poem is always pointing both forwards and backwards. Read as an individual lyric poem, “The Song of the Happy Shepherd” both thematizes the repetitions and transformations of language and engages with the revision of sonic and rhythmic patterns in its form. Facsimiles of the manuscript version of the poem show that Yeats revised it significantly during the initial period of composition; the publication record of the poem indicates that he continued to make changes to it for years afterward. In its first print appearance, in 1885, the poem consisted of 62 continuous lines. By 1895, when Yeats published his first collected volume, Poems, he had cut the poem to 57 lines, which were now divided into three uneven sections (of 21, 23, and 13 lines, respectively).16 In both the 1885 and 1895 versions of the first section (lines 1–21), the speaker despairs of a world that has exchanged the spiritual nourishment of dreams for the fascination of “Gray Truth,” laments the fleeting nature of the temporal world, and suggests that words, not actions, survive the ravages of time thanks to their repeatability. The 1895 version of this section is reproduced below: The woods of Arcady are dead, And over is their antique joy; Of old the world on dreaming fed; Gray Truth is now her painted toy; Yet still she turns her restless head: But O, sick children of the world, Of all the many changing things In dreary dancing past us whirled, To the cracked tune that Chronos sings, Words alone are certain good. Where are now the warring kings, Word be-mockers?—By the Rood Where are now the warring kings? An idle word is now their glory, By the stammering schoolboy said, 16 Between these two versions, the poem was also reprinted in The Wanderings of Oisin and Other Poems (1889). 122 Reading some entangled story: The kings of the old time are fled. The wandering earth herself may be Only a sudden flaming word, In clanging space a moment heard, Troubling the endless reverie. In this opening section, a series of oppositions—what Yeats called “antinomies”—leads us dialectically to the question of the temporal status of repetition. First, the speaker sets up an opposition between the temporal and the timeless. In the temporal world, the present moment is an inferior successor to a lost idyll, and no human action, however heroic, can stop the force of time that will turn this moment, too, into the past. The one human product that can resist this force—“words alone”—opens up two further oppositions, the first between words and actions and the second between the two modes of preservation that allow words to persist. Words may last because they endure in their static material form, available to be read by people across time, or because they can be recollected in memory and reproduced by different people at different temporal moments. The latter mode requires—is, in fact, constituted by—cognitive recall and oral repetition. The “idle word” of the “stammering schoolboy” may be a pale representation of a singular heroic action, but it is the only form in which this “glory” is preserved. Still, the end of the section, in characteristic Yeatsian fashion, problematizes any seeming resolution of the opposition between mortal action, unique and unrepeatable, and the immortal word, eternal because it recurs across time. Is the earth ageless because she cycles endlessly in her orbit or because, despite the illusion of this repetitive movement, she is a singular, “sudden flaming word”? The sonic patterns of repetition in this section participate in the oppositional debates it sets up in several ways. First, the thematic concern with repetition as transmission is emphasized by the rhymes at the ends of mostly end-stopped lines. Second, consider the rhythm: the first eight lines follow an almost unwavering iambic tetrameter, with every line but the fourth beginning 123 with an unstressed, monosyllabic function word and every line but the seventh ending with a syntactically complete unit. (This pattern was even stronger in the manuscript draft, in which Yeats prefaced “truth” with “And” or “But” rather than “Gray.”) The first internal pause, signaling the apostrophe to the “sick children” in the sixth line, lends extra momentum to the first enjambment in the seventh line; we are “whirled” to the end of the eighth line, and then, in the ninth, the iambs unequivocally break down with Chronos’s “cracked tune.” The tenth line, “Words alone are certain good,” is the end of the poem’s first sentence, but the beginning of a new rhythmical pattern. Like the next three lines, it begins and ends with a stressed syllable; because these framing syllables are also complete words, and because the line lacks an unstressed syllable, it wobbles between iambic and trochaic forms. Without this line, lines 11–13 might be set off as a kind of internal tercet, distinct from the rhythms that precede and follow them; with it, however, they seem to follow rhythmically and sonically (with line 12 echoing the beginning and end of line 10 in “Words” / “Word” and “good” / “Rood”) from the end of the poem’s first sentence. The metrical variation that this internal quatrain/tercet introduces infects the rest of the section, with the lines that follow alternating between iambic and trochaic forms. As the discussion of lines 10–14 above suggests, the repetition of sounds and of entire words introduces connections among lines that are rhythmically separate or disconnected by sense units. With very slight revisions, the section could be divided into four quatrains of alternating rhymes followed by a final quatrain of embraced rhymes. What prevents this pattern from prevailing is the presence of an “extra” line.17 Arguably, this is line 5, which adds a final “a” to an “abab” unit, and also draws attention to the return of this end rhyme in lines 15 and 17. “Fed,” “head,” “said,” and “fled” thus echo the first abrupt loss of the “dead” in line 1 and, in a way, belie it; what is “dead” returns, again and again, albeit in altered form. The importance of this transformation is underscored by Yeats’s manuscript version of the section, where he toyed with but ultimately abandoned lines such as “And all unworded things are dead,” “Happy 17 I borrow the terminology of “extra” and “missing” lines from Helen Vendler. See, in particular, pp. 94–100 in Our Secret Discipline. 124 utterance of the dead,” “telling of Chimeras dead,” and “all the heroes [sic] courts are dead.”18 With these lines omitted or replaced, “dead” lives on only in the repetition-with-change that is the operation of rhyme.19 Many of the differences between the 1895 version of the poem and the manuscript draft are, like this one, differences of subtlety. For instance, the original version of line 18, “The very world its self may be,”20 would have echoed on “world” (3, 6), “whirled” (8), and the multiple repetitions of “word” and “words”; its revision as the “wandering earth” preserves the sound and sense at a level that is less incantatory and more muted. The second section, which originally proceeded from the first without a break, extends the poem’s formal and thematic engagement with repetition: Then no wise worship dusty deeds, Nor seek—for this is also sooth— To hunger fiercely after truth, Lest all thy toiling only breeds New dreams, new dreams; there is no truth, Saving in thine own heart. Seek, then, No learning from the starry men, Who follow with the optic glass The whirling ways of stars that pass— Seek, then, for this is also sooth, No word of theirs—the cold star-bane Has cloven and rent their hearts in twain, And dead is all their human truth. Go gather by the humming sea Some twisted, echo-harboring shell, And to its lips thy story tell, And they thy comforters will be, Rewording in melodious guile Thy fretful words a little while, Till they shall singing fade in ruth, And die a pearly brotherhood; For words alone are certain good: 18 NLI 3726, 69v , The Early Poetry, Vol. I, 204-205., and NLI 3726, 70r , The Early Poetry, Vol. I, 206-207. 19 Curiously, in the manuscript drafts “dead” even seems to lead to its own replacement rhyme. In one version, Yeats appears to write “Then worship not an Idle dead” (NLI 3726, 70r , The Early Poetry, Vol. I, 206-207). This may have been a misspelling of the word Yeats eventually chose, “deed,” or “dead” may have suggested the change to him. 20 NLI 3726, 70r , The Early Poetry, Vol. I, 206-207. 125 Sing then, for this is also sooth. Notice, once more, the presence of an “extra” line in this 23-line section. The disruption again seems to stem from the fifth line of the section (line 26), which this time adds a superfluous “b” to the “abba” of the first four lines. As with the “dead” rhyme in the first section, this rhyme (“sooth” / “truth” / “truth”) reappears (and is the only end rhyme to do so) later in the section, as the enclosing end rhyme of lines 31 and 34 (“sooth” / “truth”) and 41 and 44 (“ruth” / “sooth”). In this section, Yeats does not shy away from exact repetition in the end rhymes. In fact, he employs a double kind of rime riche, as “truth” is made to rhyme with itself and also with “sooth,” its archaic synonym. The thematic concern with “truth” in lines 22–44, underlined by the sonic repetition of the word, proceeds through several stages, each of which takes the form of a directive. First, the speaker warns against looking to the “dusty deeds” of humans for permanence or authority, reminding us of the ephemeral actions of the kings in section 1. In fact, looking for any kind of permanent, empirical truth is unwise: “there is no truth, / Saving in thine own heart.” The “starry men,” who seek to abstract universal truth from the repetition of particular events, the “whirling ways of stars that pass,” have no grasp of “human truth.” Rather than follow their false path, we are instructed to take up “Some twisted, echo-harboring shell” that we might tell it our “story.” This shell will, presumably, preserve the truth of our “own heart” in its echoes, so long as they last. Again, “words alone are certain good,” and singing is “also sooth.” One line of thinking in this section suggests that truth, which is internal and subjective, can be made eternal through its preservation and repetition in language. This is the repetition of incantation and ritual, a power that calls into being the thing it names. But the section also hints at another effect of repetition, which is to destroy the thing that it repeats by evacuating it of meaning. We are prepared for this slide into nonsense by the accumulation of rime riche, starting with the “extra” rhyme in line 26, and by the description of the shell as not just “echo-harboring,” 126 but “twisted.” In the manuscript version of the poem, there is a strong parallelism between lines 37 and 38: “And to its lips thy story tell / And it they cormferterte [sic] shall be.”21 The published version (1885) introduces some pronominal confusion in line 38 with “they”—but of course, it is the multiplying echo, and not the shell itself, that offers comfort. In the 1895 version, the echoes “reword” thy “words” in “melodious guile” until they “shall singing fade in ruth, / And die a pearly brotherhood.” The destructive effect of the echo here is more explicit than in the drafts, where the shell rewords our words only “Till they have sung their way from ruth,” “For ruth and joy have brotherhood / And words alone have certain good,”22 or in the 1885 published version, where “...they shall singing fade in ruth; / For ruth and joy have brotherhood, / And words alone are certain good.” Insofar as the last sentence in this second section offers a solution to the problem of ephemerality addressed in the first, we have arrived at a potential end point for the poem. However, the third section disrupts this sense of closure, marking a shift that is both tonal and rhetorical: I must be gone—there is a grave Where daffodil and lily wave, And I would please the hapless faun, Buried under the sleepy ground, With mirthful songs before the dawn. His shouting days with mirth were crowned; And still I dream he treads the lawn, Walking ghostly in the dew, Pierced by my glad singing through, My songs of old earth’s dreamy youth: But ah: she dreams not now—dream thou! For fair are poppies on the brow: Dream, dream, for this is also sooth. As the speaker takes his leave here, he turns from the general observations of the first 21 NLI 3726, 70v , The Early Poetry, Vol. I, 208-209. 22 NLI 3726, 70v , and NLI 3726, 71r , The Early Poetry, Vol. I, 208-211. 127 section and the directives of the second to a specific memory and a present location. The “grave / Where daffodil and lily wave” was followed by a more extended pastoral description in the early versions of the poem,23 but the 1895 version shifts our attention with more economy to the grave’s occupant and to the speaker’s primary activities in this section, “singing” and “dreaming.” Variations of these words repeat via internal rather than end rhyme, but the effect, like the “truth” rhyme, is to push sonic repetition to the edge of sense. Singing, we were told in the previous section, is a path to truth, allowing us to give the past a kind of permanence in language. In this section, the singing is for one who already belongs to the past, and of “old earth’s dreamy youth.” In fact, the shepherd’s song is not only about the earth’s dreamy past, recalling the old world that fed on “dreaming” in section 1, but takes place in dream. This we discover in line 51 (“And still I dream. . . ”), the “extra” line as well as the exact midpoint of the section, which links the first sentence with the second and connects present desire (to sing to the deceased “before the dawn”) with dreams of the past (“still I dream he treads the lawn”). If we count the poem itself as an instance of song, it also ends with the instruction to dream, because dream, like song, is “also sooth.” Here, the return of the “truth” rhyme from section 2 lends a sense of closure to the final lines. More important, perhaps, is the fact that the semantic and sonic parallelism between lines 44 and 57 cements the link between song and dream. We can sing of the past, giving present voice to words in memory, or dream of the past, reanimating its sounds and images in the imagination; we can also sing to ensure the preservation of words in the future, and, if we are gifted with prophecy, our dreams will point to things to come. So we are told, in epilogue or happy song, by this sooth-sayer.24 23 I.e., in the manuscript, the 1885 version published in the DUR version, and the 1889 version published in WO. 24 I would argue that this variation on “sooth”—the sooth-sayer, or truth-teller, as clairvoyant—is an echo we are meant to generate. The entry for “sooth” in the Oxford English Dictionary online notes that the word is “In common use down to the first half of the 17th cent.; after this app. obsolete (except perh. in sense 4c [by my, your, etc. (good) sooth.]) until revived as a literary archaism, chiefly by Scott and contemporary writers.” Of the entries for “soothsayer,” the first definition, “One who speaks the truth; a truthful or veracious person,” is obsolete; the second, “One who claims or pretends to the power of foretelling future events; a predictor, prognosticator,” is not. Yeats’s fascination with prophecy suggests that his use of “sooth” invokes both the past (the archaic synonym for “truth”) and the future (the enduring sense of sooth-saying as fortune-telling). 128 My argument about the operations of form in this poem, and especially about the operation of repetition, builds on Helen Vendler’s observations about Yeats’s early lyric work—not including “The Song of the Happy Shepherd”—in Our Secret Discipline. Like “The Moods” (1893) and “The Old Men Admiring Themselves in the Water” (1902),25 “The Song of the Happy Shepherd” possesses several “extra” lines which lend both the rhyme schemes and the stanzas themselves a kind of uncertainty. Like “Who Goes with Fergus?”, the poem is marked by “ambiguous relations among its sense-units,” or a lack of coincidence between syntactic patterns and end rhymes.26 Finally, it also displays what Vendler calls “magical” techniques, or “non-rational, non-etymological relations between words” which are “as important to Yeats as logical, semantic, or etymological connections” (107). In Vendler’s account, these repetitive techniques—techniques like “anaphora” and “the repetition of phrases, lines, and sounds”—produce a sense of “reduplicative spell-casting” (109). The spell-like quality of the poem’s sounds imbues the poet with a magus-like mastery, and from this position he can turn the repetition of sounds from a soporific device to a revolutionary one. “Repetition, by spell-casting, is the guarantee of revolution; transformative or revolutionary spells must be repetitive first, and then [. . .] revolutionary” (109). The argument that Vendler builds about the “magical” techniques on display in Yeats’s early work contributes to a larger point about cognition and poetry. “Poetic thinking,” she writes, “proceeds not only by propositions and images, as we sometimes think, but also by such lexical and phonemic slippages and linkages [as we find in a poem such as Yeats’s ‘The Arrow’]” (110). The most common vehicle for these slippages and linkages are the sonic and rhythmic patterns that are constructed by repetition and disrupted by variation. In the case of “Song of the Happy Shepherd,” this kind of thinking is evident, for instance, in the rhyme scheme of the last section: a rhyming couplet followed by an alternating quatrain, an “extra” line tacked to this quatrain, another couplet, and an enclosed quatrain in which the first and last end rhymes bring back the 25 See pp. 99–105. 26 See p. 103. 129 rhyme from the last enclosed quatrain in section 2. So far as “The Song of the Happy Shepherd” conforms to the patterns Vendler recognizes in poems like “The Moods” (1893) and “He Wishes for the Cloths of Heaven” (1899), we can read it as evidence that Yeats was experimenting with “magical” techniques well before the 1890s. But, as I have already hinted, this poem also invites us to consider a kind of repetition that Vendler rarely engages with in her book: the repetition that occurs when a poet revises his poems. This kind of repetition also seems to participate in poetic thinking, especially in the thinking of a revisionary poet like Yeats. To understand how Yeats’s revision contributes to his poetic thinking, it is important to distinguish between three kinds of repetition: the formal repetition within a given poem, the repetition-with-change constituted by the poet’s act of revision, and the material repetition involved in reprinting a poem at different points in time. For Yeats, as for most poets who habitually revise their work, the latter two kinds of repetition are unquestionably related. From almost the beginning of his career, Yeats revised past poems with the intention to republish them in new, profitable contexts. Warwick Gould notes that, in the early years of his career, “Yeats had too many publishers because none who believed in him made money from him. Their deferrals led him to insist on short term contracts and new textual occasions, to a finality sought yet deferred, to provisionality” (11). When Yeats proposed his first collected volume, in October 1894, to the English publisher T. Fisher Unwin, he noted that the book could include several poems on which the copyright had recently expired or was set to expire, and mentioned that he had already received an offer for the project that would “satisfy [his] whim to the utmost.”27 As this letter suggests, the terms that Yeats reached with his publishers for his work necessarily affected his plans for subsequent projects. Still, the fact that Yeats’s revisionary repetitions have material as well as aesthetic motivations does not mean that we should discount their cognitive work. The labor that Yeats spent revising the poems for his 1895 volume, for instance, is out of all proportion to a purely 27 Collected Letters, Vol. 1 (1865-1895), p. 402. 130 economic motive. By December of 1894, having started work on the book, he wrote to his sister: “I have been doing next to no articles for all my time has been taken up with the new poem & the revision of the old ones. I am half through the revision of ‘The Wanderings of Oisin’, now ‘The Wanderings of Ussheen’. I do the new poems always before dinner & work at the old one after, & am getting on very slowly.”28 For several months after this, Yeats continued to revise lyrics for the book that would become Poems. Although he wrote to Unwin in January 1895 that he was “very anxious for my book to come out this spring,”29 in February, he informed the publisher that he was “getting on gradually with the revision—like most things it takes a little longer than I expected.”30 As this correspondence shows, while Yeats’s periodic re-issuing of already-published poems was in part a financial decision, the revisions he made to the poems were substantial and took time away from other creative projects. The consensus among textual scholars of Yeats is that Yeats viewed his revisionary work as part of a long-term aesthetic project; as Kenner put it, Yeats “was an architect, not a decorator; he didn’t accumulate poems, he wrote books” (578). In this sense, we might logically add the repetitions involved in revision to Vendler’s category of “magical” techniques; doing so allows us to view Yeats’s use of transformative repetitions at work on a different scale, one that uses the book or even the bibliography as the unit in which poetic thinking takes place. Because it is one of Yeats’s earliest published poems, and one that he continues to reprint in different contexts over the course of his career, “The Song of the Happy Shepherd” allows us to consider how the revision and recontextualization of a poem affect our reading of it. To understand the effects of Yeats’s revisionary repetition on the thinking of poems written at different points in time, I turn again to “Song”’s publication history. For Yeats, a poet so clearly invested in shaping his reputation through the selection and order of the poems that would represent his life’s work, the introductory lyric to his Collected Poems could not have been a 28 Collected Letters, Vol. 1 (1865-1895), pp. 419-420. 29 Collected Letters, Vol. 1 (1865-1895), p. 429. 30 Collected Letters, Vol. 1 (1865-1895), p. 439. 131 matter of indifference. One reason that he chose “The Song of the Happy Shepherd” to open the volume may be that it was, in fact, the earliest published work represented there; “Island” was never reprinted in book form. Another may be the new context that the poem acquired. Divorced from its dramatic context, “Song” was eventually paired with another lyric poem, originally called “Miserrimus” (in The Dublin University Review, October 1886, and in WO, 1889), and retitled “The Sad Shepherd” (in Poems, 1895). Yeats first placed these poems next to each other in a draft list for the order of Poems; there, “The Sad Shepherd” and “Song of the Last Arcadian” were the eighth and ninth of ten poems in “Crossways.” In the published version, of course, the poems were moved to the beginning of this section, their order was reversed, and “Song of the Last Arcadian” became “Song of the Happy Shepherd.” The decision to juxtapose these poems and to retitle them in a way that suggests their logical pairing necessarily influences our reading of them. Structurally, the poems constitute the two sides of an opposition in which a possible means to truth is offered (in “Song of the Happy Shepherd”) and undercut (in “Song of the Sad Shepherd”).31 It is only as a part of this structure, with its expanded interpretive horizon, that “Epilogue”’s new title makes sense. On its own, the shepherd’s song seems more equivocal than his description as a “happy shepherd” would suggest; the “mirthful” songs proposed in the third section glorify a past era and are offered to the dead. Weighed against “Song of the Sad Shepherd,” however, the first poem is undeniably hopeful. In the first poem, the speaker instructs his audience to seek comfort in the repetition of the “echo-harbouring shell” and “truth” in the acts of singing and dreaming. In the second poem, the speaking “I” disappears; instead of a lyric address, we are presented with a narrative poem in which the principal character seems to follow the advice given in the first. The “man whom Sorrow named his friend” looks in vain to inorganic natural phenomena—the stars, the sea, the dewdrops in a valley—for spiritual solace, finally turning, as the first shepherd told us we should, 31 Bornstein has remarked that “it is more the pattern of opposites than any particular truth in the early work that anticipates [Yeats’s] entire career” (Vol. II, 1). In fact, the oppositional pairing represented here by “The Song of the Happy Shepherd” and “The Sad Shepherd” recurs even in Yeats’s Last Poems, which closes with “The Circus Animals’ Desertion” and “Politics.” On this juxtaposition, see Pethica xxxv. 132 to a seashell. His reasoning echoes the advice in the first poem: [...] I will my heavy story tell Till my own words, re-echoing, shall send Their sadness through a hollow, pearly heart; And my own tale again for me shall sing, And my own whispering words be comforting And lo! My ancient burden may depart. With the poem’s closing lines, however, we find that this advice is a dead end. The transmission of story or song via repetition leads not to the comfort of permanence, but to nonsense and oblivion: Then he sang softly nigh the pearly rim; But the sad dweller by the sea-ways lone Changed all he sang to inarticulate moan Among her wildering whirls, forgetting him. What had been the last word, then—an epilogue, or the song of the “last Arcadian”—becomes the naive first half of a discourse on truth, language, preservation, and loss. If “words alone are certain good,” they are nevertheless subject to corruption and confusion. The pairing of the songs of the two shepherds, in which formal closure coincides with thematic death, introduces an elegiac tone that is common to much of Yeats’s poetry. This effect can be produced in smaller units, within individual poems, or in larger units, at the level of the book or collected volume. James Pethica, remarking on Hugh Kenner’s observation that Yeats constructed a kind of textual death in both Collected Works (1908) and The Tower (1928), argues that “deathly endings. . . are sufficiently to the fore throughout Yeats’s work to suggest that the rhetorical management of both physical and poetic closure was always crucial to him” (xxiv). The apparent finality of each textual closure was, however, no impediment to revising formerly “closed” works or to generating new material; quite the contrary. “For the revising author, the text 133 is a pre-text (in both senses—an avant-texte and an excuse or occasion),” notes Warwick Gould. “Revising authors proceed by a series of closures” (28). Thus, if “Yeats seems from an early age to have intuited that the poet seeks most often to ‘record what’s gone’ . . . and that writing is consequently always in some sense concerned with death and loss” (Pethica xxv), he also mines each textual loss for more poetic material. IV. “I spit into the face of Time / That has transfigured me” When Yeats first drafted “The Lamentation of the Old Pensioner,” he had already turned from from the pastoral, exotic settings of his early lyrics, including “The Song of the Happy Shepherd,” to the speech rhythms of Irish peasants and the subjects of their stories. In 1888, he published Fairy and Folk Tales of the Irish Peasantry; the next year, he published The Wanderings of Oisin and Other Poems and began work on the play The Countess Kathleen. In November 1890, Yeats wrote to the editor W. E. Henley, “I enclose two short poems for Scots Observer if possible. ‘The Old Pensioner’ is an almost verbatim record of words used by an old Irish man” (Letters 232). In fact, the “almost verbatim” account was already second-hand, recounted to Yeats by his friend George Russell (AE). Yeats described the meeting between Russell and the old man in “An Irish Visionary”;32 Russell wrote his own account of the event in his unpublished autobiography.33 Both versions emphasize the pensioner’s feeling of abandonment—in Russell’s version, his countrymen ask, “who’s that ould fella”; in Yeats’s, “Who is that old fellow there?”—and his idiomatic cry—according to Russell, “The fret is on me. The fret is on me!”, and, per Yeats, “ ‘The fret’ [Irish for doom] ‘is over me’ .” From the start, then, “The Lamentation of the Old Pensioner” is Yeats’s revision of the 32 Published in the National Observer, October 3, 1891, and republished as “A Visionary” in The Celtic Twilight, 1893. 33 Peter Kuch quotes the relevant section in “A Few Twigs from the Wild Bird’s Nest,” and notes that “most of this section of the autobiography was published in The Dublin Magazine, XIII, pp. 6-11, January 1938” (301). The quotations in the following sentence are from Kuch, 301, and “A Visionary” (1894), 24. 134 retelling of a conversation. It is also one of the poems that Yeats chose not just to revise but to rewrite completely, at a remove of several decades:34 [1895 version]35 [1925 version] I had a chair at every hearth, Although I shelter from the rain When no one turned to see, Under a broken tree, With ‘Look at that old fellow there, My chair was nearest to the fire ‘And who may he be?’ In every company, And therefore do I wander now, That talked of love or politics And the fret lies on me. E’er time transfigured me. The road-side trees keep murmuring. Though lads are making pikes again Ah, wherefore murmur ye, For some conspiracy, As in the old days long gone by, And crazy rascals rage their fill Green oak and poplar tree? At human tyranny; The well-known faces are all gone: My contemplations are of time And the fret lies on me. That has transfigured me. There’s not a woman turns her face Upon a broken tree, And yet the beauties that I loved Are in my memory; I spit into the face of Time That has transfigured me. Juxtaposed in this way, we can see that the two versions of the poem are radically different: no line from the first version is preserved in the second; the later version adds a third stanza; and the idiomatic refrain from the first version, “And the fret lies on me,” is dropped in favor of more sophisticated diction in lines that name “time” as the force that has “transfigured” the pensioner.36 The nature of the old man’s speech in the earlier poem, which incorporates the 34 My approach in this section is similar in some ways to David Ben-Merre’s in “The Brawling of a Sparrow in the Eaves: Vision and Revision in W. B. Yeats,” in which he argues “that Yeats’s ‘The Scholars’ can only be understood as the product of revision, and is, hence, a lyric poem that proceeds not on the page but in time” (71). 35 Both versions here are taken from the Norton edition of Yeats’s Poetry, Drama, and Prose. 36 There is a particular tension in the new refrain, as “transfiguration”—in addition to bearing within it the root of 135 remembered words of the men who no longer know him, as well as an apostrophe to the “road-side trees,” suggests that he is distractible, confused, and speaking primarily to himself. The later version, by contrast, is much more focused: this is an old man who speaks in the elevated terms of “conspiracy,” “contemplation,” and transfiguration, who can name the object of his contempt, and who resists it—“I spit into the face of Time / That has transfigured me”—rather than simply lament its effects. The rhyme scheme reflects this shift, as the simple monosyllabic rhymes of the earlier poem (see / be / me, ye / tree / me) are replaced by more complicated, asymmetric rhymes (tree / company / me, conspiracy / tyranny / me, tree / memory / me). Still, even without the title, which remains unchanged from 1895 to 1925, the later version is recognizable as a revision of the poem of the 1890s. Both speakers, old men, reflect upon the passage of time using similar images and phrases with similar semantic content: “I had a chair at every hearth,” “My chair was nearest to the fire”; “The road-side trees keep murmuring,” “. . . I shelter from the rain / Under a broken tree”; “The well-known faces are all gone,” “. . . the beauties that I loved / Are in my memory.” The connection is even stronger at a purely sonic level. Both poems use a standard four-line ballad stanza in alternating 4-3 ballad meter, rhyming abcb, with an added couplet; the resulting sixain rhymes abcbdb, with the final line serving as a refrain. Moreover, the basic b rhyme that holds these lines together is the same in both poems, despite the turn from perfect to asymmetric rhyme. At the level of form, then, the 1925 ballad is not so much a revision as a repetition. The remarkable sonic continuity from the first version of the poem to the last draws attention to subtle differences in rhythm. On one hand, the near-perfect iambic meter in the first version lends it a chant-like quality that is enhanced by its end-stopped lines and the simplicity of its words (in the first stanza, 33 of 37 words are monosyllabic; in the second, 27 of 34 are). Both stanzas reward sonic anticipation, reproducing the meter our ear is trained to expect, and end with figure and figuration—alludes to the Biblical account of Jesus’s transfiguration on the mountain, a moment in which the realms of the temporal and the eternal meet. Transfiguration in this sense points away from the temporal realm, while the speaker in Yeats’s poem complains of transfiguration as a temporal phenomenon. 136 a refrain that offers a measure of closure: “And the fret lies on me” interrupts the regular stress pattern, encouraging us to pause for two beats at the middle of the line before returning to a familiar iamb in the final foot. By contrast, the second version breaks the iambic pattern with the reversed first foot in the second line (“Under a broken tree”) and continues to frustrate any expectation we might have for a return to sing-song meter in the lines that follow. Although most lines can be scanned to match the rhythm of the first version, this requires more work on the part of the reader; trisyllabic words such as “company,” “politics,” and “memory” must be scanned as cretics, for instance, rather than dactyls, in order to conform to the iambic ballad meter. On the other hand, the second version of the poem propels the ear and eye forward in a way that the regular iambs of the first version do not. The alternation of enjambed and end-stopped lines, as well as the proliferation of two- and three-syllable words, leave the reader fewer natural pauses, and the refrain pushes the iambic meter forward rather than instituting a central pause. In fact, the last lines of each stanza reproduce the transfiguring effects of time as well as its merciless repetition: the speaker arrests the transfiguration that takes place between line 6 (“E’er time transfigured me”) and line 12 (“That has transfigured me”) by reproducing the latter exactly in line 18, but the effect is more suggestive of the presence of an underlying and uncompromising repetitive force—Chronos’s “cracked tune”—than of the vanquishing of time. The political aspect of the 1925 version of “Lamentation” seems to invite a comparison between the poem’s disillusioned speaker and the poet himself, now a 60-year-old former proponent of Irish nationalism writing in the aftermath of his country’s civil war. Before we draw too many conclusions about Yeats’s personal politics from this poem, however, we should note the connection between the poem’s treatment of politics and its temporal anxiety. In fact, the disengagement of the speaker who was once in the midst of those who “talked of love or politics” is not so much political per se as systematic; one has the sense that, whatever flaws the old man’s former political program had, their correction would not have changed his dissatisfaction with the present moment. Events in time, revolutionary or not, have come to seem to him the passive 137 components of an inevitable temporal cycle: the identity of the cause no longer matters, as young men will always be agitating about “some conspiracy,” and the evils they resist are part of an eternal condition of “human tyranny.” Turning his anger from the political arena to the temporal condition of being, the old man acknowledges the transfiguring effects of time even as he resists them: whereas, in the first version of the poem, the speaker simply declared that “the well-known faces” were “all gone,” in this version he maintains that “the beauties that I love / Are in my memory.” The temporal concerns of this version of the poem are of a piece with other works that Yeats was producing in the mid-1920s. A Vision, which details the system of gyres and lunar phases that he and his wife Georgie had developed, and which Yeats called a “last act of defense against the chaos of the world,”37 was finished in 1925 and first published in 1926. In a letter written that year to the artist Thomas Sturge Moore,38 Yeats laid out some of his evolving thoughts on time, being, and the “world of sense data”: Forty years ago the S.P.R.39 succeeded in transferring mental images — numbers, geometrical forms simple drawings — between two people (I) in the same room (2) in different rooms (3) in different towns. From that moment all philosophy based upon the isolation of the individual mind became obsolete. The ecclesiastics of the mechanical philosophy met this & all evidence of psychical research by the demand that no fact of this kind could be accepted until — as you put it – “the conditions are known in which it will certainly recurr.” This was an evasion for psychical facts belong to mind which never does the same thing in exactly the same way twice because all moments of being are unique. They in fact demanded that mind should become mechanism before they would consider its action. No too people see exactly the same world of sense data ever — whether these sense data are mental or phisical images nor 37 Collected Letters, Unpublished Letters 4525, letter to Edmund Dulac, April 23, 1924. 38 Collected Letters, Unpublished Letters 4855. Errors are in the original. 39 The Society for Psychical Research. 138 does one person see the same world twice. The material may be the same but the pattern is different. I read a book of Catholic apologetics a few years ago which contended with much proof that the Church had not condemned Galileo’s doctrine but merely Galileo for teaching it before he could fully prove it. It is a subconscious trick which must tempt every creed, which is incompatible with some fact, to perpetually demand to have that fact proved over over again. There need be no end to this process especially if the ecclesiastics — scientific or theological — do not look at the proofs. The belief that all is experience does not mean that there is no truth unknown to us for there are unknown minds, but it does mean that there is no truth where there is no mind to know it [...] This letter explains why occult theories and experiments need not be subject to empirical tests, which call for the exact repetition of original conditions and final results: “psychical facts belong to the mind which never does the same thing in exactly the same way twice.” This Bergsonian observation is followed by the claim that “no too [sic] people see exactly the same world of sense data ever... nor does one person see the same world twice.” Here is a way out of the potentially totalizing cycles suggested in “Lamentation” and in A Vision: it may be that lads will always be “making pikes again,” but they will never be doing so in exactly the same world, under exactly the same conditions. The other side of the speaker’s rage against transfiguring time, which allows nothing to stay the same from moment to moment, is the relief that “all moments of being are unique.” At this point, we might turn to Yeats’s revision of “Lamentation” from a different angle: what was different about the world of the poem when Yeats repeated it, by revising it, in 1925? Warwick Gould has detailed the process and context of this revision, noting that “the poem survived with fairly minor changes [after its initial publication in 1890] until Yeats began to work upon Early Poems and Stories (1925). . . . There he tinkered with it on the galleys, alongside 139 another vigorously altered poem, ‘The Dedication to a Book of Stories selected from the Irish Novelists’ ” (16). On its own, the new poem shows a shift in thematic concern, a “preoccup[ation] with civil unrest (topical in post-Civil War Ireland) and with the transfigurations of time” (18). However, Gould argues that the full import of the 1925 rewriting is apparent only when one considers the rest of the book in which it was to appear. Early Poems and Stories, he notes, was the first book to include the poem as well as “A Visionary,” Yeats’s prose account of Russell’s encounter with the old man who was the inspiration for the poem. Moreover, “if the rewriting of the poem for its new context involved a new reading of ‘A Visionary,’ tinkering with the poem necessitated the editing of the prose account” (17). For the 1925 publication, Yeats cut the original ending of “A Visionary,” a paragraph in which he speculates that the old man and Russell both “have within them the vast and vague extravagance that lies at the bottom of the Celtic heart” and reflects on the “great Celtic phantasmagoria.” The new version ends simply with Russell’s story of the old man ranting and weeping on the mountain. This significant change in tone, Gould argues, must be understood in the context of the book: “its economy. . . demanded that ‘A Visionary’. . . retain something of the unfinished misery of the old pensioner in the early version of the poem (‘The well-known faces are all gone / And the fret lies on me.’)” (18). Another important context for the poem’s revision is Yeats’s evolving use of ballad form. “No other modern poet,” Helen Vendler writes, “was so determined as Yeats to make the ballad a substantial and lifelong part of his oeuvre” (144). From early ballads like “Moll Magee” (1889), in which the content “remains within one branch of the Wordsworthian tradition, in that his protagonists are, for the most part, poor, overworked, or old” (113), Yeats went on to incorporate occult topics, political material, and even metaphysical investigations; by the 1930s, “a Yeatsian ballad sequence could embrace almost any lyric subgenre and any subjective attitude” (126). “The Lamentation of the Old Pensioner,” in its early and late versions, seems to anticipate several of the major steps that Vendler identifies in Yeats’s experimentation with the ballad genre. Like Moll Magee, the speaker fits the convention of the downtrodden ballad protagonist. However, the 1890 140 version of the poem has already departed from the traditional four-line ballad stanza. Vendler notes that, “by 1916.... The abcb ballad quatrain comes to seem less a self-contained unit than a base upon which interesting things can be built” (118); for instance, “Yeats’s two ballads on the Easter Rising, “Sixteen Dead Men” and “The Rose Tree” both “extend the stanza to six lines, abcbdb (120). In this, they follow the model of “Lamentation,” composed more than 20 years earlier. As Vendler’s description of the evolution of Yeats’s use of ballad would suggest, the 1925 version of “Lamentation” shifts the ballad toward the province of lyric: the speaker not only seems to reflect on past and present from a single moment in time, but also thematizes the urge to stop the propulsive, chronological force that traditionally drives both narrative and ballad. This revision anticipates later ballads like “What lively lad,”40 in which, Vendler argues, “rather than a narrative’s being unfolded, a special set of frustrated emotions is being defined” (128), as well as Yeats’s final ballad, “The Black Tower,” which “poises itself at a single crucial moment” (143) and in which Yeats “wrests the innate narrativity of the ballad as far as it can go in the direction of instantaneity” (144). Vendler locates Yeats’s most significant lyricization of the ballad in four ballad sequences composed after 1925: “It is in. . . ‘A Man Young and Old’ (1926–1927), ‘A Woman Young and Old’ (1927–1928), ‘Words for Music Perhaps’ (1929–1932), and ‘The Three Bushes’ (1936) that Yeats, using ballad conventions, most fully dedicates himself to creating lyrics rather than narratives” (123). From Vendler’s description of these poems alone, we could infer that they have grown directly or indirectly out of Yeats’s experimentation with the “Lamentation” poems. For instance, the first sequence, “A Man Young and Old,” contains eight ballad poems written in the familiar abcbdb sixain; these poems, whose governing title alludes to the passage of time, “become ballads about psychic alteration” and thematize metamorphosis (124). But the connection is stronger than that, as Warwick Gould has shown: Th[e] fresh perspective [provided by revising “The Lamentation of the Old 40 Part of the ballad sequence “A Woman Young and Old” (1927–1928). See Vendler 126–128. 141 Pensioner” and “A Visionary” for Early Poems and Stories] opened up a whole new vein of poems for Yeats, beginning with four he first published in the London Mercury in April 1926 as “More Songs of an Old Countryman”. Why “More”? The title must have seemed obscure to most readers. On the (undated) manuscript Yeats tries out, “Songs of a mad old man”, then “More Songs of an Old Pensioner”. The first poem, later “His Memories”, associates Yeats himself (nearing 60) with the pensioner, whose “broken” thorn-tree becomes the locus (and focus) for Yeats’s seventeen-year-old memories of the short-lived consummation of his love for Maud Gonne. (18) “Revision and new writing,” Gould notes, “were not merely interdependent activities [for Yeats], revision of an old poem frequently made the next new poem possible” (20).The drafts that Gould mentions, which stem directly from the revision of “Lamentation” in 1925, are in fact the early materials of “A Man Young and Old,” which “led in turn to “Words for Music Perhaps” ... and to “A Woman Young and Old” (20). If these sequences represent, as Vendler suggests, “Yeats’s utmost effort to bring the ballad under the vault of lyric tradition” (138), then one point of origin for this effort must lie in the 1925 revision of “Lamentation”—and perhaps even earlier, in the original version of this ballad. Much remains the same, or repeats, between the two poems, including the non-traditional ballad form as well as some of the content. What has changed from 1895 to 1925 is not simply the poet’s political stance or generational perspective, but the extent to which he will push the boundaries of the ballad genre: the sonic and semantic engagement with the question of time in the later version recalibrates our expectations about what a ballad can do and prepares us for further revolutions in Yeats’s ballad sequences. V. “Why brood upon old triumph? / Prepare to die” By the late 1930s, Yeats’s obsession with mortality and regret no longer seemed so proleptic. He had passed his seventieth birthday, and, as James Pethica puts it, “his political 142 marginalization in the Irish Free State, the renown of his own earlier work, the death of some of his closest friends, and other factors had made him feel increasingly a figure from the past” (xxxi). His letters from the period refer repeatedly to the fluctuating state of his health, which he sometimes blamed for delays in writing (“I have had much illness & lassitude & great accumulation of letters”41 ) and sometimes credited with spurring him on to new projects (“I have been in a very excitable state — the result I suppose of illness — strangely miserable & this has given my poems a new poignancy”42 ). It was against this backdrop that Yeats began a poem, “alternately titled ‘Despair’ and ‘On the lack of a theme’ in one late draft,” which “emerged out of a sudden and debilitating sense of creative bankruptcy which followed his completion of New Poems in the Autumn of 1937” (Pethica xxx). This poem is, of course, “The Circus Animals’ Desertion,” the 1939 version of which is reproduced below: I I sought a theme and sought for it in vain, I sought it daily for six weeks or so. Maybe at last being but a broken man I must be satisfied with my heart, although Winter and summer till old age began My circus animals were all on show, Those stilted boys, that burnished chariot, Lion and woman and the Lord knows what. II What can I but enumerate old themes, First that sea-rider Usheen led by the nose Through three enchanted islands, allegorical dreams, Vain gaiety, vain battle, vain repose, Themes of the embittered heart, or so it seems, That might adorn old songs or courtly shows; But what cared I that set him on to ride, I, starved for the bosom of his fairy bride. And then a counter-truth filled out its play, 41 To Miss Lidell, March 13, 1937. Collected Letters, Unpublished Letters, 6859. 42 To Dorothy Wellesley, March 24, 1937. Collected Letters, Unpublished Letters, 6883. 143 “The Countess Cathleen” was the name I gave it, She, pity-crazed, had given her soul away But masterful Heaven had intervened to save it. I thought my dear must her own soul destroy So did fanaticism and hate enslave it, And this brought forth a dream and soon enough This dream itself had all my thought and love. And when the Fool and Blind Man stole the bread Cuchullain fought the ungovernable sea; Heart mysteries there, and yet when all is said It was the dream itself enchanted me: Character isolated by a deed To engross the present and dominate memory. Players and painted stage took all my love And not those things that they were emblems of. III Those masterful images because complete Grew in pure mind but out of what began? A mound of refuse or the sweepings of a street, Old kettles, old bottles, and a broken can, Old iron, old bones, old rags, that raving slut Who keeps the till. Now that my ladder’s gone I must lie down where all the ladders start In the foul rag and bone shop of the heart. No other poem in Yeats’s corpus so explicitly confronts the conflicting complex of emotions bound up in the artist’s retrospective judgment of his own work: despair, defiance, shame and inspiration all radiate from this backward look. But it is not the work only that is judged in retrospect. As the poem suggests, the most abstract or fantastic art grows out of the artist’s life, from his “heart mysteries.” One must be judged with the other; if there is a pattern of order, an aesthetic unity, it must be found in both life and work. The kind of order that Yeats might have been seeking is suggested in his narrative account of his life and in his correspondence. In his autobiographical writings, Yeats repeatedly expressed 144 a tendency—or a wish—to view the events in his life as if they were governed by a kind of typology. In Autobiography, recounting the year when he met Maud Gonne, he wrote: “I was twenty-three years old when the troubling of my life began. . . . In after years I persuaded myself that I felt premonitory excitement at the first reading of her name” (225). In The Trembling of the Veil, in an episode describing his first acquaintance with John Synge, he observed: “I am often astonished when I think that we can meet unmoved some person, or pass some house, that in later years is to bear a chief part in our lives. Should there not be some flutter of the nerve or stopping of the heart like that of [sic] Macgregor experienced at the first meeting with a phantom?” (244). If Yeats desired access to a secret order in life, he seemed certain that such an order was present in his work. In 1922, he wrote to Olivia Shakespear of the collections Later Poems and Plays: “As I turn the pages I find much unity of meaning & some curious prophesies. ‘The Unicorn from the Stars’ is now being fulfilled in Munster where great houses are being burned. . . . Then too I wrote ‘Kathleen-ni-Houlihan’ when Ireland was entirely at peace & ‘The Kings Threshold’ before anybody had hunger-struck.”43 A letter to Ethel Mannin in 1936 expressed a similar conviction: “If you have my poems by you look up a poem called ‘The Second Coming.’ It was written some sixteen or seventeen years ago & foretold what is happening.”44 At the time Yeats wrote “The Circus Animals’ Desertion,” the task of imposing an order on his artistic output as well as his biography must have seemed increasingly urgent. Of Last Poems, the volume in which this poem was published, Pethica notes: “It has long been understood that one of Yeats’s key concerns in his final volume of lyrics was to interpret his impending death, and to inflect others’ interpretations of it. His last sequence of poems served him as a personal meditation on mortality, but also as a carefully managed dramatic narrative which might condition critical responses after his death” (xxiii). “The Circus Animals’ Desertion” is in many ways the poem that determines this narrative, whatever we may make of it. In order of composition, the poem, first completed in November 1937,45 preceded all the others that he grouped with it. In a 43 Collected Letters, Unpublished Letters, 4204. 44 Collected Letters, Unpublished Letters, 6530. 45 Although one version of the poem was completed at this time, Yeats did not make his final revisions to the poem 145 letter to Dorothy Wellesley that month, Yeats wrote: “I have finished my book of lyric poems, & a new poem which is the start of a new book. I will not send it you until I have done the poem that follows & explains. For five weeks I could not write a line, & then the poem came & a new mass of themes.”46 In Yeats’s draft list of contents for the posthumously published Last Poems, the poem that “started” this book is in the penultimate position, followed by “Politics.” Pethica argues that this placement “effectively completes a thematic turn away from the bitterness and disillusion of the opening sequence, thereby allowing the volume as a whole to replicate in large the move made within that poem alone” (xxxv). However, the ordering of the poem also invites some doubt as to Yeats’s intended tone: “ ‘Politics’ offers a closure which immediately complicates this resolve, letting things end, in Bradford’s view, on a ‘mocking note’” (Pethica xxxv).47 How we read “The Circus Animals’ Desertion,” both as an individual poem and as a component of this volume, thus regulates our interpretation of Yeats’s largest, last “dramatic narrative.” It is one of the ironies of “The Circus Animals’ Desertion” that its formal achievement—a key to its internal order and to its order within Yeats’s body of work—is so hard-won. To be sure, the poem thematizes, and opens with, the difficulty of poetic composition: “I sought a theme and sought for it in vain.” However, Yeats’s mastery of the poem’s form in its final version elides the struggle of the manuscript stages; we forget Yeats’s description, in “Adam’s Curse” (1903), of the labored attempt to make a line of verse seem “a moment’s thought,” and we forget that this poem, like any poem, might have been other than what it was. The published version of “The Circus Animals’ Desertion” is so well wrought, in fact, that Helen Vendler holds it up as an exemplar of its stanza form: “No poem of Yeats better shows the cumulative possibilities of the ottava rima stanza: the form is self-contained enough to make a strong freestanding unit, but short enough so that an allusion to a previous stanza, by word or reference, is instantly grasped by the reader and added to the ongoing effect of the whole. It entertains a narrative, but it also entertains for nearly a year. See Pethica xxxii and Chapman 27. 46 Collected Letters, Unpublished Letters, 7113. 47 Pethica quotes Curtis Bradford, “Yeats’s Last Poems Again,” Yeats Centenary Papers, MCMLXV, ed. Liam Miller (Dublin, 1968), p. 275. 146 philosophical reflection” (276). Vendler emphasizes the spatial aspect of this form, arguing that ottava rima seemed to be for Yeats “not so much a stanza as a station, a place, a location” (264). However, her description of the form in “The Circus Animals’ Desertion” also highlights its complex temporal qualities: the stanzas are both discrete and continuous, and their mutual echoes ask us to hold them together in our mind, so far as this is cognitively and temporally possible. Writing of another ottava rima poem, “Among School Children,” Vendler remarks that “no poem can obliterate or disavow the reflections in which it has engaged en route to its arrival at a ‘solution’ ” (289). These observations, although made in reference to the reflections that take place in the published poems, seems especially relevant to the excisions, alterations, and false starts that went into the creation of the “ongoing effect of the whole” in “The Circus Animals’ Desertion.” Yeats’s creative struggle in this poem is most in evidence in the record of its last stanza. Below are two versions of this stanza,48 the first dated November 1937 (the month he wrote to Dorothy Wellesley that the poem was “finished”): [NLI 13,593 (47), 5r ] [NLI 13,593 (47), 9r ] Why brood upon old triumph? Prepare to die O hour of triumph come and make me gay! For all those burnished chariots are in flight. If burnished chariots are put to flight O hour of triumph come & make me gay; Why brood upon old triumph, prepare to die; For even For on the edges Even at the approach of the unimaged night Even at the approach of the un-imaged night Man has the refuge of his gaety; Man has the refuge of his gaiety; A dab of black enhances every white A dab of black enhances every white, vigour Tension is but the vigor or the mind, Tension is but the vigour of the mind, Cannon the god & father of mankind. Cannon the god and father of mankind. 48 Here I follow Pethica’s transcriptions in Last Poems. I match the stanzas line to line, so that there are blank spaces in the second version where in the first there are inter-line changes or corrections. In the typescript, the stanza is continuous. 147 Yeats drafted multiple versions of this stanza before finally rejecting it. In the earliest version, the first line (“Why brood upon old triumphs, Prepare to die”) is followed by tentative variations on this theme: “Renounce immortality,” “learn to die.”49 In later iterations, although Yeats settles on a single version of this self-commandment, buries it after an apostrophe to the “hour of triumph,” and makes the “flight” of the “burnished chariots” a future possibility rather than a present reality, the specter of death continues to determine the stanza and prevent its satisfying closure. The speaker protests altogether too much; his bravado is as forced as the rhyme scheme that marries “die” to “gay” and “gaiety.” As Pethica notes, the “swerve” toward death in this stanza, “which initially seems to gesture towards a stark new realism. . . nonetheless leads only to a new round of self-persuasion” (xxxii). If the original closing stanza is such a failure, what makes its replacement a success? In describing Yeats’s mindset in writing the poem, Pethica argues that it “embodies a stark confrontation with the possibility that he would be unable to avoid trailing off into infirmity, silence, or, worse still, repetition” (xxxi). However, I want to suggest that repetition is actually the crux of this poem, and especially of its closing stanza. The opening lines of the poem frame the problem of the artist’s lack of inspiration as a problem of repetition: “I sought. . . and sought. . . / I sought. . .” In the first line of the second section, in which Yeats sits in retrospective judgment of his past work, the speaker despairs, “What can I but enumerate old themes,” proceeding then to do exactly that. By the end of this section, it seems natural to suggest—as Yeats originally tried to do—that what the poem requires is an abrupt break with the repetitions of the past. And what could halt the stale repetitions of a life better than death? What could be more alien, more unknowable, more original, than the “un-imaged night”? Here is the new project, the way out of living habit: prepare to die. And yet, as we have seen, the ending that attempts to rejoice in this new imperative is unsatisfactory. The ending that Yeats finally arrived at takes an entirely different tack. It would be 49 NLI 13,593 (47), 3r , Last Poems 373. 148 simplistic to say that the stanza achieves closure via formal repetition alone, although it does embrace repetitive diction (“Old kettles, old bottles. . . / “Old iron, old bones, old rags. . .,” “. . . ladder’s gone / . . . . ladders start”) and bookends the description of the roots of the “masterful images” with an opening two-line question and a closing (slightly extended) couplet. Nor is it enough that the stanza thematizes repetition, although it does this too, following the train of thought of a speaker who seeks the source of his abstract creations. The directive to “lie down where all the ladders start,” prescribing a return to origins, could not be more different from the earlier instruction to “prepare to die.” Rather, what magnifies our satisfaction in both of these kinds of repetition is the patterning that ties the repetition in this stanza to the rest of the poem. The opening question reminds us of the speaker’s lament in the poem’s opening lines and implicitly asks us whether, over the course of the poem, anything has changed; its closing line, locating creative inspiration “In the foul rag and bone shop of the heart,” answers the speaker’s resistance in the first section (“Maybe at last being but a broken man / I must be satisfied with my heart”) and echoes the “themes of the embittered heart” and “Heart mysteries” in stanzas 2 and 4.50 In fact, we might say that this final stanza is simply a radical revision of the first, refracting its anxieties through the retrospective stanzas of the second section. The poet, ready now to acknowledge the origins of the “circus animals,” can look forward only because he has first looked backward. The kind of repetition that Yeats theorizes in this revised stanza is distinct from the pure repetition (in which meaning is impossible) or degenerative repetition (in which meaning is diluted or corrupted) that the poem’s speaker initially fears. This is a repetition that leads not to stagnation or clich´e, but to creative possibility, much in the way that literary genre is itself a structure of repetition—in the way, for instance, that, Yeats found generative frames in received forms such as ottava rima and ballad. One way to understand this mode of repetition is to characterize it as typological: a figure and its fulfillment can be thought of as a kind of repetition 50 Vendler argues that, in the final stanza, the poem in fact “comes to rest on the most important of its repeated words—heart—its ‘mysteries’ now grossly demystified” (278). 149 with change, a repetition that is recognized as such in retrospect and that invites us to anticipate future iterations in the “cumulative possibilities” (Vendler 276) of a text.51 If we move our horizon of interpretation from “The Circus Animals’ Desertion” to Last Poems and to Yeats’s body of work as a whole, the poet’s revisionary repetition turns out to be a key to his aesthetic project. Chapman, seizing on the image of the ladder in the final stanza, argues that it summons a chain of tree-images in Yeats’s source material and his own work, including “the biblical tree of Jesse, the tree of life depicted on the cover of The Secret Rose. . . the “Rose Tree” in one of his Easter rebellion poems, rose-trees grown to one root in “Three Bushes”. . . trees of generation flocked with birds. . . in ‘Sailing to Byzantium’”; in sum, the image calls to mind “a host of trees in a canon nearly finished with the completion of this poem” (28). Pethica, commenting on the theme of the poem more generally, notes that “‘The Circus Animals’ Desertion’ seems to reject the youthful Yeats as ‘embittered’ in his unrequited love, and as ‘vain’ . . . in his dreaming as his hero Oisin, yet ‘Politics’ potentially takes a reader back to The Wanderings of Oisin, the opening poem of his canon, echoing Oisin’s own wishing-back to his youth, and thereby giving a reincarnate structure to Yeats’s whole oeuvre” (Pethica xxxvi). We can trace chains like this back to Yeats’s earliest works; think, for instance, of “The Song of the Happy Shepherd,” which declares “there is no truth, / Saving in thine own heart,” long before Yeats turned back to the “heart” in “The Circus Animals’ Desertion,” and which anticipates the regrets of old age even as it serves, for the older poet, as a document of youth and an introductory poem for his life’s work. It would be a mistake, however, to try to trace any of these repetitive chains back to an original term. 51 Another way is to compare it to the Deleuzian theory of repetition that emerges from his reading of the eternal return, which in some respects is analogous to typological thinking. Deleuze introduces his concept of repetition by emphasizing its difference from those terms that we might associate with it: it concerns not “cycles and equalities” but “non-exchangeable and non-substitutable singularities” (1). Further, he associates repetition specifically with poetry; citing the Romanian linguist Pius Servien, he opposes “the language of science, dominated by the symbol of equality, in which each term may be replaced by others” and “lyrical language, in which every term is irreplaceable and can only be repeated” (2). Finally, consider Deleuze’s citation of the French poet P´eguy: if “it is not Federation Day which commemorates or represents the fall of the Bastille, but the fall of the Bastille which celebrates and repeats in advance all the Federation Days” (1), might we not say that it is not the last line of Yeats’s poem that represents its preceding stanzas, but the first line that “repeats in advance” the lines to come? 150 V. Conclusion The intersection between repetition and revision in Yeats’s poems is in some ways particular to the work that he left behind. Yeats’s inveterate practice of revision was remarkable, if not unique, and the manuscript record that remains of his drafts and typescripts is unusually detailed. Not all poets revised their work to anywhere near the same extent, and not all poets’ manuscripts have been preserved and edited as Yeats’s have been. Moreover, Yeats brought a specific set of ideas about time, history, politics, and the afterlife to his individual poems and to his editorial work. For Yeats, art and life were ideally to be united, and he employed the same mode of revisionary typology to both: Maud Gonne’s name alone ought to have provided him with a premonition of the importance she was to have in his life, just as the prophecy in “The Second Coming” ought to have been legible to its European readers in the aftermath of the Great War. Revision was, for Yeats, not just a matter of improving upon or updating an individual poem, but the way he made emerging patterns in his work readable; recontextualized poems commented on each other in new ways, and revisions to any individual work reverberated throughout his oeuvre. Despite the idiosyncrasies of Yeats’s work, however, there are some aspects of his revisionary repetition that may be generalizable to other poets. First, revision calls attention to the poem’s place in time: it marks the differences in a poem at different points in time, tracing the evolution of a text whose form we sometimes seem to read as preordained, and also points out of time, to variations of the poem that were not but might have been. Second, revision is an essential part of poetic thinking. This much is apparent to any reader who has ever browsed the archives; the thinking that takes place in any given “finished” poem is the result of a longer cognitive process that takes place, at least in part, in other, earlier versions of the text. But revision participates in poetic thinking in another way, as well: it extends the horizon of what Vendler calls Yeats’s “magical” techniques, the sonic repetitions that invoke the transformative, “incantatory power of reduplicative language” (109) and that seem to lead to a “self-generating” poetry (108). Poetic thinking proceeds according to Vendler’s “lexical and phonemic slippages and linkages” 151 (110), the non-logical connections which act as mutually constitutive figures and fulfillments in a typological mode. In her examples, these slippages and linkages are often repetitions with difference: identical words that repeat in different contexts, or words whose sounds train us anticipate similar sound patterns. Revision, it seems to me, is this kind of repetition writ large: the act of revision repeats a poem with internal alterations or in different external contexts, and in so doing participates in—and to some extent generates—the typological thinking in the poem and in the poet’s body of work. C HAPTER F IVE Hart Crane: The Bridge and the Limits of Genre 153 Very roughly, it concerns a mystical synthesis of “America.” . . . . it may be too impossible an ambition. But if I do succeed, such a waving of banners, such ascent of towers, such dancing etc, will never before have been put down on paper! – Hart Crane to Gorham Munson, February 18, 1923 . . . however bad this work may be, it ought to be hugely and unforgivably, distinguishedly bad. – Hart Crane to Gorham Munson, March 5, 1926 The labor in locating the interrelations between sources, facts and appearances in all this is, believe me, difficult. One may be doomed to the kind of half-success which is worse than failure. – Hart Crane to Yvor Winters, November 15, 1926 I. Introduction Hart Crane’s assessment of his most ambitious poetic project vacillated, as the epigraphs above suggest, from the optimistic to the despairing: The Bridge would cement his place in the American literary scene, he thought, or it would be his critical undoing. The first reviewers of The Bridge followed Crane’s example in framing their evaluations of the poem in terms of its success or failure, and later critics, often aiming to refute these early readings, have also responded to the success/failure binary.1 These analyses may take for granted different assumptions—for example, that Crane was a late romantic, a modernist, or a herald of Language poetry; or, that his poems are marked by language that is hopelessly idiosyncratic, by “substantive or ontological privacy,”2 or by verse that is “difficult only if and when it is understood” by readers3 —but all are predicated, as the early reviews were, on the question of genre. To wit: is The Bridge an epic or a lyric poem, a 1 See, for example, Joseph Riddell, “Hart Crane’s Poetics of Failure,” or Susan M. Schultz, “The Success of Failure: Hart Crane’s Revisions of Whitman and Eliot in The Bridge.” 2 Dean 91. 3 Grossman 159. 154 kind of “half-success” of one of these genres, a superior synthesis of both of them, or another kind of poem entirely? The question of the genre of Crane’s long poem has important implications for critical debates about Crane’s legacy, American modernism, and, more broadly, lyric poetry or the “lyricization” of poetry. Many recent analyses of Crane, for instance, defend the literary value of The Bridge by making the case for its status as an epic or a special variety of epic that incorporates history and the other in a way that lyric poetry cannot. In this chapter, I want to see what happens when we take the poem as Crane described it following its publication—as a “a long lyric poem, with interrelated sections”4 —and, in the process, to put pressure on certain ideas about the limitations of lyric language. For example: is the temporality of a lyric poem always restricted to a kind of perpetual present? Is it fundamentally—that is to say, generically—incapable of addressing such “ambitious” ideas as time and history or ethical engagement with an other? Is the orientation of a lyric poem always inward, exclusive—essentially solipsistic? My aim in examining the genre of The Bridge is not to supply a new definition of either epic or lyric poetry. However, since these generic definitions have been, and continue to be, contested territory, I lay out some of my underlying assumptions here. Epic poetry tends toward the narrative and the mimetic, focuses thematically on action or event, and often aspires to treat subjects of cultural or national significance; it instrumentalizes history in the service of nation and empire. In this, I loosely share J. K. Newman’s historical definition of epic as “a long narrative poem. . . that treats a single heroic figure or a group of such figures and concerns an historical event. . . or an heroic quest or some other significant mythic or legendary achievement that is central to the traditions and belief of its culture” (361) or, more concisely, Ezra Pound’s: “epic in the real sense is the speech of a nation thru the mouth of one man” (175).5 Lyric poetry I take to 4 HC to YW, June 4, 1930. Complete Poems and Selected Letters 642. 5 Since I am dealing with the question of Crane’s poem, I focus on the definition of epic poetry. However, as Newman notes, poetry is not the only form of modern epic. “If [would-be writers of epic] choose verse for their mode, they accept all the risks of a genre not now organically related to their culture. . . . More traditional poets either find themselves compelled to treat of lofty themes in a way that is bound to be lifeless and unconvincing, because it 155 be verse that is primarily non-narrative and non-mimetic, verse that foregrounds the material and sonic properties of language.6 As such, it may be opposed to epic poetry. However, it does not follow that lyric is limited to one voice or to one moment in time, that it is always “about what happens now” as opposed to “what happens next” (Culler, “Why Lyric?” 202). Nor, as Paul De Man argued, is lyric fundamentally either more primitive or more modern than other genres.7 In contrast to Virginia Jackson, whose work I will address in more depth later on, I am interested in the transcultural or transhistorical properties of lyric generally, and the ways that Crane’s lyric voice engages these properties as it speaks to and for others in time.8 II. Critical Reception of The Bridge Many early reviewers, Crane’s friends and peers among them, panned his long poem when it appeared in 1930. These attacks, along with their personal and aesthetic motivations, have been mistakes the polyphonic and potentially even comic nature of the genre, or to ignore the decorum of [Classical epic] in an extravagant effort to impress. If they turn away from mythical loftiness. . . they run the opposite risk of seeming to have forgotten their commitment to heroic sublimity” (374). 6 Even this basic definition is not universal. For instance, James William Johnson writes that “[i]n its modern meaning, a lyric is a type of poetry which is mechanically representational of a musical architecture and which is thematically representational of the poet’s sensibility as evidenced in a fusion of conception and image” (715). My emphasis on the non-mimetic function of lyric poetry is more aligned with Mutlu Blasing’s description of lyric poetry as “a formal practice that keeps in view the linguistic code and the otherness of the material medium of language to all that humans do with it—refer, represent, express, narrate, imitate, communicate, think, reason, theorize, philosophize. It offers an experience of another kind of order, a system that operates independently of the production of the mean- ingful discourse that it enables. This is a mechanical system with its own rules, procedures, and history. It works with a kind of logic that is oblivious to discursive logic” (Lyric Poetry 2). 7 In “Lyric and Modernity,” De Man describes this paradox of lyric: “On the one hand, lyric poetry is often seen not as an evolved but as an early and spontaneous form of language” (168); on the other, “the social projection of modernity known as the avant-garde consisted predominantly of poets rather than of prose writers,” and lyric is sometimes “considered as the best means of access to a discussion of literary modernity in general” (169). This confusion stems from misguided attempts to make poetry conform to a “genetic historicism” (185): “To claim that the loss of representation is modern is to make us again aware of allegorical element in the lyric that had never ceased to be present, but that is itself necessarily dependent on the existence of an earlier allegory and so is the negation of modernity . . . . [t]he worst mystification is to believe that one can move from representation to allegory, or vice versa, as one moves from the old to the new, from father to son, from history to modernity” (186). 8 My position here is consonant with Siobhan Phillips, who argues, that, “[w]hile Jackson’s insights are invaluable in their exposure of assumptions, her focus on poetry’s cultural role could recommend a practice of increasing partic- ularization that neglects some transcultural possibilities of verse writing. . . . Current readers may be able to renew the study of lyric not only by dividing this label into the subclassifications of Jackson’s list, and thereby considering the historical applications of many different works in verse, but also by expanding this label into the larger classification of poetry and thereby considering the rhetorical possibilities of verse itself (23). 156 well documented elsewhere.9 Until recently, however, comparatively little attention has been paid to the generic component of the reviews, despite the fact that many of them are preoccupied with questions of genre, including the intended genre of the poem, the proper generic classification of the published version, and the implied aesthetic and ethical consequences of this classification. For example, Yvor Winters, who, in his admiring review of White Buildings in 1927, had called Crane “among the five or six greatest poets writing in English” (49), in 1930 reduced his assessment of Crane’s contribution to “several charming minor lyrics and many magnificent fragments” (165). As the premise for this drastic critical reversal, the opening paragraph of Winters’s essay on The Bridge is worth quoting in its entirety: It is necessary, before attempting to criticize Mr. Crane’s new book, to place it in the proper genre and to give as accurate an account as one is able of its theme. The book cannot be called an epic, in spite of its endeavor to create and embody a national myth, because it has no narrative framework and so lacks the formal unity of an epic. It is not didactic, because there is no logical exposition of ideas; neither Homer nor Dante will supply a standard of comparison. The structure we shall find is lyrical; but the poem is not a single lyric, it is rather a collection of lyrics on themes more or less related and loosely following out of each other. The model, in so far as there is one, is obviously Whitman, whom the author proclaims in this book as his master. (153, emphasis added) In part, Winters suggests, Crane’s failure to compose a proper epic is a result of his choice of model: whereas “destiny for Vergil was a clear and comprehensible thing and had a clear relationship to a complete scale of human values,” “Whitman found all human values about equal and could envisage good only as an enthusiastic acceptance of everything at hand” (156). With no hierarchical system of values to hold them together, Whitman’s poems are thus “little save 9 See, for instance, Mariani 351–355, Irwin xi-xii, Edelman 2–3, and Bloom 268. 157 boundless catalogues” (156). Crane not only neglected to follow an epic predecessor with an appropriate value system, Winters claims, but he also lacks such a system himself, and is (or ought to be) aware of this failing: “there is a great deal of evidence that Mr. Crane suspects continually the inadequacy of his [own] belief” (160). In Winters’s estimation, The Bridge is plagued by “mawkish and sentimental” attempts to depict human relations (157); its lines “hypnotize” but fail to convince (161), and its details have no meaning because, for Crane, “meaning is not available” (162). Following Crane’s death, Allan Tate added to the critical narrative of Crane’s progressive decline as a poet by delivering a similar pronouncement on the place of The Bridge in his career. Crane “never again had such perfect mastery of subject” as he did when he wrote “Praise for an Urn,” in 1922, Tate declared; moreover, “he never afterwards knew precisely what his subject was. That is why The Bridge is such a magnificent failure: a great talent is engaged upon the problem of stating a position that is fundamentally incapable of definition” (212). Although Crane’s long poem is “presumably an epic” (211), it suffers from a “vagueness of purpose” that cannot be remedied by the “apparently concrete character” of its governing symbol, “the Brooklyn Bridge” (213). In sum, while Crane might have intended to spur his artistic progress by writing an epic, this shift in genre could not disguise his creative limitations: the poet’s “more ambitious later project permitted him no greater degree of self-definition than the more literary symbolism of his youth” (213). Many of these early reviews imply that Crane’s failure is not just generic, but ethical; in fact, the generic and the ethical seem to be bound up together insofar as the ability to write an epic is associated with a heightened range of human understanding and degree of moral judgment. Winters partially explains Crane’s failure to produce an epic by referring to “one’s suspicion that Mr. Crane is temperamentally unable to understand a very wide range of experience” (157–158) and claims that Crane’s “attempt to emotionalize a theme to the point where both he and the reader will forget to question its justification” is ultimately “a form of hysteria” (162). In a similar 158 vein, Tate observes that Crane’s “world had no center, and the compensatory action that he took is responsible for the fragmentary quality of his most ambitious work” (214). Howard Blake, writing in the Sewanee Review in 1935, suggests that the common failing in both Crane and Eliot’s attempts at epic is their tendency toward a subjectively limited lyric. “Hart Crane, with some success, aimed at the epic goal in The Bridge,” he acknowledges, and “Mr. Eliot’s Waste Land remains. . . the one poem our time has fostered with unmistakeable epic adumbrations” (189). Eliot’s poem even achieves a kind of diversity of expression—we are “aware of his having expressed many men”—but “[t]he weakness remains: he articulates only the Mr. Eliots” (189). Blake concludes that “this limitation seems to be an integral part of lyric poetry, but is lamentable when we have need for a more universal portraiture” (189).10 Crane seized on the generic component of these contemporary attacks in his defense of The Bridge. In the last letter he would ever write to Winters, his once-close friend and frequent correspondent,11 he denied that he had set out to write an epic in the first place: “Your primary presumption that The Bridge was proffered as an epic has no substantial foundation. You know quite well that I doubt that our present stage of cultural development is so ordered yet as to provide the means or method for such an organic manifestation as that.”12 Instead, Crane claimed, The Bridge was a kind of expanded lyric, and “[t]he fact that [it] contains folk lore and other material suitable to the epic form need not therefore prove its failure as a long lyric poem, with interrelated sections” (emphasis added).13 Crane also recognized that Winters’s criticism of the poem’s genre was infected by personal conflict. “At any rate,” he noted, “you seem to have appropriated a good many of my ethical and privately expressed aesthetic convictions, and applied them pretty much to the advantage of your initial persuasions against certain features and 10 Blake reiterates this judgment later in his essay: “Mr. Eliot’s nice writing and Hart Crane’s broad splashes have this in common: both are subjective; intent (consciously in Mr. Eliot’s case, I believe, unconsciously in Hart Crane’s) on words, not sections; lacking any positive, determined, ruling conception of life (a belief which automatically excludes exceptions); lacking in universality” (194). 11 Mariani 353. 12 HC to YW, June 4, 1930. Complete Poems and Selected Letters 642. 13 HC to YW, June 4, 1930. Complete Poems and Selected Letters 642. 159 directions repugnant to you in The Bridge.”14 It is true that, before Winters’s review appeared, Crane had often suggested that “The Bridge” aspired to something very like epic. In 1923, writing to Gorham Munson, he declared that his “new enterprise” involved “a mystical synthesis of ‘America.’ History and fact, location, etc. all have to be transfigured into abstract form that would almost function independently of its subject matter.”15 Two years later, when he wrote to Otto Kahn, the banker who would become his patron, Crane described the project in very similar terms: his long poem “aim[ed] . . . to enunciate a new cultural synthesis of values in terms of our America.”16 Writing again to Kahn in 1927, Crane offered a more detailed plan of the poem. What his “Myth of America” was after, he explained, was not a “purely chronological historic angle” but rather “an assimilation of this experience, a more organic panorama, showing the continuous and living evidence of the past in the inmost vital substance of the present.”17 Crane’s letter also reiterated a theme that was common in his correspondence to Kahn, his father, and other friends who had loaned or might loan him money: to achieve his ambitious project, he would need freedom from interruption and economic concerns, unbroken stretches of time to think and write. “The Aeneid was not written in two years—nor in four, and in more than one sense I feel justified in comparing the historic and cultural scope of The Bridge to that great work,” he concluded. “It is at least a symphony with an epic theme.”18 Crane’s statement to his patron may well have been inflected by anxiety about justifying the cost of his support and the cultural and aesthetic value of his project. It should not be surprising, then, that his statements to his peers on the subject of the poem’s genre were more equivocal. In an early letter to Winters, while their literary friendship was still flourishing, Crane placed The Bridge’s relation to epic in the context of a broader argument about modernity and 14 HC to YW, June 4, 1930. Complete Poems and Selected Letters 642. 15 HC to GM, February 18, 1923. Complete Poems and Selected Letters 321. 16 HC to OK, December 3, 1925. Complete Poems and Selected Letters 424. 17 HC to OK, September 12, 1927. Complete Poems and Selected Letters 554. 18 HC to OK, September 12, 1927. Complete Poems and Selected Letters 558–59. 160 literary form. A modern writer would, he suggests, have to “overcom[e] the limitations” of epic, and “[t]his will have to be done by a new form,—and, of course, new forms are never desirable until they are simply forced into being by new materials. Perhaps any modern equivalent of the old epic form should be called by some other name, for certainly, as I see it, the old definition cannot cover the kind of poem I am trying to write except on certain fundamental points.”19 Both “old epic form” and The Bridge, however, were “concerned with material which can be called mythical.”20 The poem he wanted to write would achieve “nothing less ambitious than the annihilation of time and space, the prime myth of the modern world.”21 What can we make, then, of Crane’s generic ambition for The Bridge? Should we privilege the parallel he draws between his poem and the Aeneid; the distinction he suggests between old epic and a new form that might carry that genre into modernity or, conversely, mark its end; or his denial, in his final letter to Winters, that The Bridge was ever intended as an epic in the first place? Recent critical analyses of The Bridge have taken up these questions, re-examining the early critics’ claims about its generic status and Crane’s statements on the topic in light of the poem’s structure and themes and the aesthetic effects that are generated within and among its various sections. For instance, John Irwin, in his preface to Hart Crane’s Poetry: “Appollinaire Lived in Paris, I Live in Cleveland, Ohio,” suggests that Winters’s and Tate’s early appraisals of The Bridge may have been the result of a combination of “professional ressentiment” (xii), personal bias—“[b]oth Tate and Winters . . . remembered personal quarrels with Crane . . . [which] may have made the two academic poet-critics feel that the high-school-educated, ‘roaring boy’ of modern American poetry could not possibly have written an epic” (xii)—and the mixed features of the poem itself. Irwin notes that, “in its effort to be a synthesis of American myth and history,” The Bridge “was strategically an epic,” but “it was tactically a lyric, seeking to achieve that strategic epic goal through a disjunctive series of lyric epiphanies in its sections” (xii). 19 HC to YW, November 15, 1926. Complete Poems and Selected Letters 495. 20 HC to YW, November 15, 1926. Complete Poems and Selected Letters 495. 21 HC to YW, November 15, 1926. Complete Poems and Selected Letters 496. 161 The synthesis between epic and lyric that Irwin proposes is characteristic of recent defenses of Crane’s project, many of which seek to justify the poem’s epic vision by emphasizing the ways in which it expands or modifies the traditional notion of the epic genre. In his annotated edition of The Bridge, for instance, Lawrence Kramer writes that the poem “is commonly thought of as a compact epic, although, unlike most of its precursors in the genre, it has no single hero and records no single narrative” (xv). “Nonetheless,” he argues, “the poem does concentrate on a trio of traditional epic themes, which it seeks to reinterpret in relation to modern life”: there is “the voyage or journey, which forms the fundamental action in virtually all of the poem’s sections”; “the relationship between past and present, the basic armature of history”; and “the possibility of understanding history as a purposive, even redemptory process, in particular, one destined to find its consummation in the New World” (xv–xvi). In this view, the poem’s deviation from traditional epic is explained by its experimental modernism: “Like many modernist texts, it is a kind of collage” (xv). Similarly, Daniel Gabriel argues that “despite The Bridge’s resistance to conventional epic and narrative continuity—its really quasi-epic status—it is unprofitable to read it merely as a lyrical sequence, or as a long lyrical poem” (2). More broadly, Gabriel claims that American modernism as a whole was a movement “preoccupied with genre” and that this “preoccupation gave us Crane’s The Bridge and a number of related works” (1). Specifically, “The Bridge emerges out of a conflict between Crane’s native [lyric] abilities and his epic [modernist] ambition” (1). Gabriel’s book positions The Bridge as a “modernist epic” (4) and makes a powerful argument for the formal and literary-historical significance of this genre. The Bridge, Gabriel posits, is “a hybrid of lyric and epic modes, a mixture of two distinct but related discourses” that “assume a dialectical relationship, and yet form an unconventional synthesis (a disunified unit) through a complex and elusive structure” (2). This choice of genre is neither idiosyncratic nor a mark of Crane’s failure but rather signifies his ambition to participate in a larger modernist project. Crane, like Eliot, Pound, and Williams, felt “compelled to include history in [his] 162 rebellious respect for tradition” and, “[a]t the same time,” found that “history as chronology or as knowledge became disrupted by another precept: radical disjuncture” (8). Formally, “[a] crucial feature of The Bridge and the modernist epic at large is fragmentation, both in style and point of view. Excising narrative elements, the modernist epic offered a metanarrative of the modern” (4). Ethically, the modernist epic also challenges the “I/Other” binary: “[r]ather than a recession of the lyric self,” Gabriel “perceive[s] an interaction of lyric and epic” in The Bridge such that “[t]he self becomes reconstituted as an agent in a field of possible others” (6–7). The recent appraisals of Crane seem to offer a critical compromise: they position the poet as a generic innovator who surmounted the limits of lyric and epic by synthesizing both, a late romantic who nevertheless earned his place as a modernist par excellence. And in many ways, these critics point out, The Bridge supports this reading. So why should we not close the question of genre by designating the poem a “modernist epic” (Gabriel), a “compact epic” (Kramer), or a “strategi[c] . . . epic” (Irwin)? First, such definitions seem to defer to Crane’s earliest detractors insofar as they assume that rehabilitating The Bridge requires reading it as an epic poem; they address the early reviews’ criticism of the poem as a failed epic by offering a new model of epic to which the poem perfectly conforms and which it even helps bring into being.22 Second, and more seriously, this privileging of epic implicitly authorizes the idea that there is something formally or ethically inferior about lyric poetry as a category. For the early critics, the slide from epic into lyric was linked to the danger of the poet becoming a “stylistic automaton,” a producer of “inexact poetic verbiage” (Winters 164) or a purveyor of “the current solipsistic verse” (Blake 195). The ethical consequences of this generic failing included, Tate argued, Crane’s inability to 22 My argument here is analogous in some ways to Lee Edelman’s. In Transmemberment of Song: Hart Crane’s Anatomies of Rhetoric and Desire (1987), he noted that the critics who defended Crane in the wake of criticism by Winters, Munson, Tate, et al. “did not question in any fundamental way the theoretical basis for the earlier negative evaluations of Crane’s poetry; rather, they attempted to show that the verse, properly read, did in fact fulfill the re- quirements articulated by the critics of the 1920’s and 1930’s. They sought to show, therefore, that the poetry was not incoherent, was not confused, was not seriously obscure. In consequence, the task of explication that they undertook bordered frequently on apologetics. . .” (2). These critics focused on defending Crane at the level of thematics, whereas Edelman shines a late on the rhetorical complexity of Crane’s work. My claim focuses on the generic accusations of the early critics and the attempts of more recent scholarship to show that Crane either fulfills or exceeds the requirements of the epic genre. 163 “define the limits of his personality and to discover its moral properties in an appropriate symbolism”; instead, he could “only assert a quality of will against the world, and at each successive failure of the will he [could] only turn upon himself” (214). By reading The Bridge as a poem in the lyric tradition, I am effectively conceding to the classification of the early critics while questioning the formal, temporal, and ethical limits that they associate with lyric poetry. III. The Bridge and Genre Theory Just as Crane scholars have begun to reexamine the hermeneutic significance of the genre of Crane’s work, studies of poetry and genre more broadly have raised questions about the critical tendency to “lyricize” wide varieties of writing in verse. Virginia Jackson’s work in this area both historicizes the formation of what we now think of as the lyric genre and probes our assumptions about its relationship to the public and private spheres. As she makes clear, the story of the idea of lyric after romanticism is, in large part, a story of how critics came to conceive of its temporality: . . . that shift in genre definition [of lyric before and after the eighteenth century] is primarily a shift in temporality; as variously mimetic poetic subgenres collapsed into the expressive romantic lyric of the nineteenth century, the various modes of poetic circulation . . . tended to disappear behind an idealized scene of reading progressively identified with an idealized moment of expression. While other modes—dramatic genres, the essay, the novel—may have been seen to be historically contingent, the lyric emerged as the one genre indisputably literary and independent of social contingency, perhaps not intended for public reading at all. By the early nineteenth century, poetry had never before been so dependent on the mediating hands of the editors and reviewers who managed the print public sphere, yet in this period an idea of the lyric as ideally unmediated by those hands or those readers began to emerge and is still very much with us. (7, emphasis added) 164 In this chapter, I build on Jackson’s critique of the lyricization of poetry to the extent that her work questions the reduction of the temporality of the poems we read as lyric to “an idealized moment of expression.” However, my concern is not so much to overturn the critical dogmatism that has subsumed different modes and sub-genres of poetry under lyric as to ask whether we have mistaken the capabilities of even the most generically typical “romantic lyric.” In particular, I want to question what we think of as the temporal limits of lyric and, by extension, the ethical limits that are often presumed to be attendant on lyric time. Crane’s The Bridge lends itself to this kind of examination precisely because so many evaluations of its success or failure have been dependent on an argument about its genre. I take seriously Crane’s insistence to Winters that the poem is not an epic but a “long lyric poem, with interrelated sections.” Certainly, by placing his poem in the category of lyric, Crane is shielding it from its supposed epic faults. But he is also, I think, defending the kind of work that lyric alone can do. By emphasizing those 1930s critics who bemoan Crane’s lyricism—who imply that a shift in genre from epic to lyric constitutes a kind of aesthetic fall; that the scale of epic is wide-ranging and masculine, while lyric is solipsistic and feminine; and that the epic genre enables a kind of historical and ethical thinking of which the lyric is fundamentally incapable—I am also asking us to reevaluate the supposed critical dominance of lyric in the early twentieth century.23 Finally, I am pointing up the implied connection between the supposedly limited subjectivity and temporality of lyric poetry and its ethical deficiencies. Before offering an analysis of The Bridge itself, I want to turn to some of the work that has shaped recent critical assumptions about genre and about the temporality of lyric. Sharon Cameron, in her two major books on the poetry of Emily Dickinson, presents some of the most 23 That is, by the height of the modernist period, critics were already questioning something very like the “lyriciza- tion” of poetry and demanding greater theoretical and terminological specificity. For example, Blake’s 1935 “Thoughts on Modern Poetry” rails against “the dogma that all poetry must be lyric poetry to be good,” a dogma associated with “a pervading and persistent lyricism” that is “both destructive and paralyzing” (190). T.S. Eliot, in “The Three Voices of Poetry” (1953), also criticized the unsatisfactorily broad usage of the term “lyric” in the mid-twentieth century. 165 important claims in this field.24 In Lyric Time: Dickinson and the Limits of Genre (1979), she suggests that Dickinson’s oeuvre is emblematic of the tension between the temporal and eternal orders at work in lyric generally. Moreover, she takes as axiomatic the idea that lyric as a genre is distinguished by its temporal position: All literature attempts to regulate chronology, much literature to defeat it, but the strategic ingenuities of the novel, drama, and lyric regale us variously with the dreams they have wrought. Unlike the drama, whose province is conflict, and unlike the novel or narrative, which connects isolated moments of time to create a story multiply peopled and framed by a social context, the lyric voice is solitary and generally speaks out of a single moment in time. (22–23) As Jackson notes, “[m]ost readers in 1979 would have shared Cameron’s assurance that the solitude of the lyric voice was its transhistorical feature” (42). Just as important, however, was the restriction of the lyric utterance to the present tense—a kind of temporal solitude—which was likewise assumed to be a defining quality of the lyric genre.25 In Choosing Not Choosing: Dickinson’s Fascicles (1992), Cameron opens up the contextual frame that she assumed in her earlier book by attending to the various versions of Dickinson’s poems and their placement in the original manuscript notebooks. Considered in this light, she argues, “Dickinson’s fascicles [manuscript packets] trouble the idea of limit or frame on 24 As both Jackson and Cameron use Dickinson’s verse to make larger arguments about lyric poetry and lyric time, their claims may apply to a wide range of poetry. Their analyses may translate particularly well to Crane’s poetry since, as many critics have pointed out, Dickinson and Crane share important rhetorical and stylistic affinities. (See, for example, Cameron 183, Bloom 266, and Blasing’s “Hart Crane: Inscribing the Sublime.”) However, insofar as their evaluations of Dickinson focus on the question of the manuscript context and the editorial choices made in posthumous publications of her work, they are less applicable to Crane. Although Crane revised many sections of The Bridge over a period of several years, and sometimes included drafts of these in letters to friends, he always intended to publish the poem as a structurally unified work, and we have no doubt about the final version of any given section or of the intended order. 25 Only two years earlier, for instance, Jonathan Culler had argued that “the time of the apostrophe—a special temporality which is the set of all moments at which writing can say ‘now’” (“Apostrophe” 66)—was the time “towards which the lyric strives” (68). 166 which. . . our suppositions of lyric fundamentally depend” (5), and variant words in these poems “raise the question of what counts as the identity of the text in question” (5). The new context of this reading sometimes causes Cameron to revise her earlier interpretations. For instance, she argues that when the poem “I saw no Way—The Heavens were stitched—” is read as “an isolated lyric,” it “seems like an exercise in solecism, as well as solipsism, having not only no referent but also no context: barely comprehensible” (31). However, “to read the poem differently in the context of the poem that precedes it on the same bifolium in Fascicle 31 is to see that there could be a referent for that experience” (31). More broadly, while Cameron still holds that “Dickinson’s poetry exaggerates the temporal features of the lyric as a genre,” she “reformulate[s]” this claim with the observation that “once read in the fascicle context the poems no longer look representative because they no longer look like lyrics” (184). One of the larger theoretical payoffs of Cameron’s argument in her later book is her claim that “Dickinson’s poetry dramatizes the impossibility of wholeness understood as boundedness” (182). Partially, “the poetry dramatizes the impossibility of wholeness in its preference for synecdoche,” but “[m]ost stunningly, Dickinson dramatizes the impossibility of wholeness by privileging variants as the essence of her poetry” (183). Moreover, this emphasis on variants leads to a view of poetry that is “open” and “multiply determined” (188). In Cameron’s revised assessment of Dickinson’s poetry, “what is repeatedly insisted upon is radical connection—as poems are not divided, as the dead and the living are not divided, as the “I” and the “you” are not divided, as speech from the center and speech from the margin are not divided, as the presumptive speech of the speaker and the presumptive speech of the reader are—can be said—not to be divided” (188–189, emphasis added). As should be evident by now, one of the major consequences for Cameron of formally contextualizing Dickinson’s poems is the opening up of their temporality, which in turn points to a different ethical position for the lyric voice. No longer divided from past and future, no longer over-determined, this voice is no longer necessarily so solipsistic. In Jackson’s view, however, 167 Cameron’s mistake is to assume that poems like Dickinson’s can be recontextualized in such a radical manner and still remain lyrics (even if they “no longer look like lyrics”). “It should follow,” Jackson argues, “that if all lyrics depend on limits or frames, and Dickinson’s writing turns out not to have such limits or frames, then Dickinson did not write lyrics—or did not write poems that conform to ‘our suppositions of lyric’” (43). Instead of setting Dickinson’s poems apart form lyric conventions, however, Cameron “holds tenaciously to the lyric as the genre governing Dickinson’s composition and, more suggestively, governing the interpretation of that composition” (Jackson 43). In addition to Cameron’s work, another point of departure for Jackson’s theory of the “lyricization” of poetry—and for my claims about The Bridge—is the body of work on genre theory. Throughout the nineteenth and early twentieth centuries, lyric, epic, and drama were considered the three major transhistorical literary genres, a division that was presumed to originate in Classical philosophy. In The Architext, G´erard Genette showed this tripartite division to be a “retrospective illusion” (2) that actually emerged much later, with romanticism: he explains that, contrary to the claims of eighteenth-century criticism, the dithyramb that Aristotle discusses in the Poetics is not equivalent to modern lyric, while Plato, in the Republic, “deliberately leaves out all nonrepresentational poetry—and thus, above all, what we call lyric poetry—and a fortiori all other forms of literature” (9). Moreover, treatments of Aristotle in the post-Romantic period tended to conflate genre with mode, i.e., with “situations of enunciating” (12); in Genette’s view, “the phenomenon of genre inextricably merges the phenomena—among others—of nature and of culture” (69), while modes may be transhistorical, preliterary, “a priori forms” (71n77). One of the consequences of this critical misreading of genre, Genette argues, is that “poetry” in the twentieth century came to be understood as “what the preromantics meant by lyricism” (59); “[i]n other words: for more than a century, we have considered as ‘more eminently and peculiarly poetry’ . . . exactly the type of poetry that Aristotle excluded from his Poetics” (60). 168 The historicization of genre theory raises several questions that are pertinent to our consideration of Crane and the genre of The Bridge. How far can we extend lyric, epic, or drama before these labels becomes meaningless? Can a genre really be transhistorical? Does a “hybrid” genre always imply progress—and if so, what kind of progress? De Man warns us that “[t]he worst mystification is to believe that one can move from representation to allegory, or vice versa, as one moves from the old to the new, from father to son, from history to modernity” (“Lyric and Modernity” 186). Part of what Genette’s history of genre systems shows us is how this mystification gained a critical foothold: following Schlegel and H¨olderlin, the dramatic was assumed to be the synthesis of the epical and the lyrical, and thus their aesthetic superior (41); following Schelling, lyric was assumed to be the more basic, and thus aesthetically inferior, category, as art “begins with lyric subjectivity, then rises to epic objectivity, and finally attains dramatic synthesis” (42). That Schelling’s sequence was the one that endured is evident in the comments of post-Romantic writers like Hugo and Joyce (43) and, I would add, in the assumptions that govern much of the early and contemporary criticism on The Bridge. That is, the early critics accused Crane of failing to “rise” to epic objectivity from lyric subjectivity, while later critics defend Crane by arguing that The Bridge both meets and exceeds the standards of epic by achieving a lyric-epic synthesis. That Gabriel, for instance, does not call this synthesis drama, but rather “modernist epic,” does not detract from the hierarchy implied in his argument. As to the question of the precise border between lyric and epic, or between lyric and other varieties of twentieth-century verse poetry, any definitive attempt at an answer invites both formal and historical exceptions. As Derrida argued, “[a]s soon as the word ‘genre’ is sounded, as soon as it is heard, as soon as one attempts to conceive it, a limit is drawn. And when a limit is established, norms and interdictions are not far behind” (“The Law of Genre” 56). At the same time, what he called “the law of the law of genre” is “precisely a principle of contamination, a law of impurity, a parasitical economy” (59). To belong to a genre, a text must invoke its predecessors through repetition, but it must also deviate from them. In this sense, Derrida’s remarks on the 169 archive also apply to genre (which is, in itself, a special kind of archive): “The [genre]: if we want to know what that will have meant, we will only know in times to come” (Archive Fever 36). Perhaps we can ask, then, what lyric poetry would have meant in 1930, following the publication of The Bridge, along with what it has come and will come to mean since then. IV. Crane’s Long Lyric As a test case for determining the generic status of Crane’s long poem, I turn to “Atlantis,” the section that bookends The Bridge, coming first in terms of composition—Crane began drafting this section in 1923—and last in terms of the shape of the poem as a whole, since he ultimately placed it at the end of the published work. Bloom calls it Crane’s “most representative” (275) and “most ambitious” poem, as well as the only section that can stand as “a synecdoche for The Bridge” as a whole (278). My reading also assumes that the poem has a synecdochic relationship to The Bridge, taking its formal and conceptual patterns to be both representative and constitutive of the larger work. At the same time, I want to suggest that, like Dickinson’s oeuvre, Crane’s long lyric poem might “dramatize the impossibility of wholeness understood as boundedness” (Cameron 182) through its use of synecdoche. “Atlantis” stands for The Bridge, limiting it to the length of a traditional lyric, while also, like a variant, disrupting any notion of resolution or narrative wholeness. Similarly, its shifting lyric voice attends to a larger collectivity, alternately claiming a place among this group or standing in for it, and resisting the claim to speak for anyone other than “me.” Like the sections of The Bridge that precede it, “Atlantis” emphasizes its thematic connections to the narrative myths of classical antiquity: the titular island, the mythic and historical cities of Tyre and Troy, the characters named (Jason, Aeolus) and implied (Penelope, and thus Odysseus, through descriptions of the “weav[ing]” of song, “shuttling moonlight,” and the “towering looms” of the bridge), as well as the choice of modifiers ([s]ybilline, labyrinthine, 170 palladium), leave no doubt as to the epic tradition that Crane means to invoke. These references, along with other semantic and sonic echoes, link “Atlantis” to other sections in the poem. In its allusions to the Aeneid, for example, Irwin argues that “Atlantis” contributes to an overarching epic structure in The Bridge: in this reading, Crane’s “cipher-script of time” refers to “the recurrences of certain hidden (Crane might say ‘encrypted’) structures throughout history: the way, for example, that the mating of Maquokeeta and Pocahontas. . . repeats that of Aeneas and Dido. . . and, further, the way that these fatal conjunctions foreshadow the doom of a people” (205). However, the “symphony” that Crane produces here may have an “epic theme” even while operating as a “long lyric poem, with interrelated sections.” In his reading of “Atlantis,” Bloom compares Crane to Shelley, noting that both poets “tend to transfigure every genre into the condition of lyric” (274) even though “Adonais ostensibly is pastoral elegy while ‘Atlantis’ concludes an American brief epic” (274). Indeed, “Atlantis” appeals to “another logic,” an “apparent illogic” that Crane called “the logic of metaphor” (“Letter to Harriet Monroe” 166). This logic is the property not of epic poetry but of lyric, which operates according to “a kind of logic that is oblivious to discursive logic” (Blasing, Lyric 2), a logic that follows non-narrative patterns of connotation and sound. Further, the logic of Crane’s lyric poem is not restrictive but expansive; as Allan Grossman comments on another Crane poem, “The Broken Tower,” “[t]he poem conducts us (permits us to think our way through) from tragic logic (Aristotle) to the other comic logic of inclusion, the logic of ecstasy” (156). Take the first three stanzas of the poem, which, rather than foreground the traditional epic features of character, action, or plot, show a more characteristically lyric attention to “voices” and a “cry,” as well as an emphasis on movement, sonic pattern, and symbol: Through the bound cable strands, the arching path Upward, veering with light, the flight of strings,— 171 Taut miles of shuttling moonlight syncopate The whispered rush, telepathy of wires. Up the index of night, granite and steel— Transparent meshes—fleckless the gleaming staves— Sibylline voices flicker, waveringly stream As though a god were issue of the strings. . . . And through that cordage, threading with its call One arc synoptic of all tides below— Their labyrinthine mouths of history Pouring reply as though all ships at sea Complighted in one vibrant breath made cry,— “Make thy love sure—to weave whose song we ply!” —From black embankments, moveless soundings hailed, So seven oceans answer from their dream. And on, obliquely up bright carrier bars New octaves trestle the twin monoliths Beyond whose frosted capes the moon bequeaths Two worlds of sleep (O arching strands of song!)— Onward and up the crystal-flooded aisle White tempest nets file upward, upward ring With silver terraces the humming spars, The loft of vision, palladium helm of stars. From the opening prepositional phrase, these octaves establish and re-establish a spatial orientation with each first line (“Through the bound cable strands,” “And through,” “And on”), while internally, the octaves emphasize vertical movement (“Upward,” “Up the index of night,” “up bright carrier bars,” “onward and up the crystal-flooded aisle,” “White tempest nets file upward, upward ring”) in a kind of counterpoint (or syncopation) to the phrases we might expect to be associated with the horizontal structure of the bridge (“across,” perhaps, or “over”). The first octave also sets up the syntactic (but non-narrative) suspension that governs the poem, along with a set of sounds and images that will be repeated and transformed in the stanzas that follow. Of these, the most important is the alliteration at the ends of the lines: beginning with line 2, “strings”—which thematically telescopes the strings of a harp, the warp and weft of a loom, and the metal wires of the bridge—finds a basic alliteration in “syncopate” (3), an even closer match 172 in “steel” and “staves” (5, 6), and a perfect consonant match in “stream” (7), which finally leads back to the rime riche that closes the stanza. The last two lines produce the first reference in the poem to song, the origin of which is both multiple (“[s]ybilline voices flicker”) and, thanks to a turn to the subjunctive mood, singular and divine (“As though a god were issue of the strings. . . .”). The instability of reference and the frustration of semantic and syntactic fulfillment here is typical of the operation of Crane’s lyric. In the second octave, a single sentence stretches the syntactic delay of the first stanza—where we waited until the third and seventh lines, respectively, for the subjects of its two sentences—until the very last line, where the “seven oceans” finally “answer from their dream.” The seven preceding lines, echoing the number of oceans, are thus varieties of “moveless soundings” before the primary acoustic action (the “answer”). They reinforce the notion of a voice that is both compound and unified in the “cry” of the ships; they also introduce the concept of history, which is emphasized by the acoustic end rhyme with “sea” and the visual echoes of “cry” and “ply.” Moreover, because these lines force us to anticipate the subject of the sentence, to hold each prepositional phrase and modifying clause or sub-clause in mind through line breaks, phrase endings, and shifting grammatical moods, they simultaneously extend the moment of cognition required to read the poem and propel us forward in the hope of syntactic completion. They are, like Dickinson’s verse, “open” (Choosing Not Choosing 188); like the singular “vibrant breath,” they may be synecdochic, but they require us to suspend knowledge of what, exactly, they are a synecdoche for. The third stanza brings the first two “arching strands of song” to a preliminary thematic and sonic climax. Like the second octave, its lines are irregularly punctuated by end rhymes: the “bars” rhyme opens and closes the stanza (“spars,” “stars”), and within this frame the end slant rhymes of lines 2 and 3 are complemented by the interlaced rhyme of “aisle” and “file.” Like the first octave, it emphasizes the alliteration of ending words (“song,” “spars,” “stars”) and closes with the s-t-r consonance. It multiplies the references to vertical movement with “on,” “onward,” 173 twice “up,” and twice “upward.” And, in addition to representing the sonic convergence of the first two stanzas, it suggests several varieties of conceptual convergence: like “strings,” which called to mind the harp, loom, and bridge, “octaves” refers to the framework of the bridge and also to the musical interval and to the stanza structure of the poem itself; “bars,” similarly, can refer to architectural or musical structure, while the “humming spars” assigns a musical function to the ships’ masts. The repeated association of musical and physical structures also implies a convergence of the temporal and the spatial realms, of movement and stasis, and of acoustic and visual stimuli, a synaesthetic effect first encountered in the opening stanza (“miles of . . . moonlight syncopate”; “voices flicker”) that will recur in later stanzas (“swift peal of secular light” (65), “one tolling star” (90)). Finally, the middle line of this stanza gives us the first instance of a direct address to an other with its parenthetical apostrophe: “(O arching strands of song!).” The three octaves that follow further develop the poem’s emerging themes of time, myth, and history alongside abstract apostrophic addresses to the static bridge: Sheerly the eyes, like seagulls stung with rime— Slit and propelled by glistening fins of light— Pick biting way up towering looms that press Sidelong with flight of blade on tendon blade —Tomorrows into yesteryear—and link What cipher-script of time no traveller reads But who, through smoking pyres of love and death, Searches the timeless laugh of mythic spears. Like hails, farewells—up planet-sequined heights Some trillion whispering hammers glimmer Tyre: Serenely, sharply up the long anvil cry Of inchling aeons silence rivets Troy. And you, aloft there—Jason! hesting Shout! Still wrapping harness to the swarming air! Silvery the rushing wake, surpassing call, Beams yelling Aeolus! splintered in the straits! From gulfs unfolding, terrible of drums, Tall Vision-of-the-Voyage, tensely spare— 174 Bridge, lifting night to cycloramic crest Of deepest day—O Choir, translating time Into what multitudinous Verb the suns And synergy of waters ever fuse, recast In myriad syllables,—Psalm of Cathay! O Love, thy white, pervasive Paradigm . . . ! In the fourth stanza, the formal patterns that had marked the ends of lines in the first three stanzas move almost entirely inward: “rime” (25) finds its rhyme in “time,” at the third beat of line 30, which is transformed to “timeless” in the last line; “flight” (28) echoes “light” (26), and “yesteryear” sets up the rhyme for the concluding “mythic spears” (32). At the same time, different axes of spatial movement (the eyes that “[p]ick biting way up,” the “towering looms that press / Sidelong”) intersect with temporal movement from future to past (“[t]omorrows into yesteryear”) and “the timeless laugh of mythic spears.” The next stanza echoes the compression of future and past in the equivalence of “hails” and “farewells” as well as the transcendence of time by myth: “silence rivets Troy,” even as Jason is “[s]till wrapping harness to the swarming air”—“aloft there,” suspended in space and time—and beams are still, then and now, perhaps always, “yelling Aeolus.” Although the mythologies Crane refers to here “are deprived of enough fac¸ade to even launch good raillery against,” “much of their traditions are operative still—in millions of chance combinations of related and unrelated detail, psychological reference, figures of speech, precepts, etc.” (Crane, “General Aims and Theories” 161). Throughout “Atlantis,” as in The Bridge as a whole, thematic and sonic compressions and convergences attest to this persistence—and revision—of various pasts in the present moment. It is also in stanza 5 that, instead of the oblique references to disembodied “eyes” and the hypothetical “traveller,” the poem first refers to a “you.” The significance of this address is underscored by the references to calls or cries that frame it (the “hails, farewells,” the “whispering hammers,” “the long anvil cry,” the “hesting Shout[,]” the “yelling” beams) and by the “silence” that precedes it. Moreover, the outcome of this exclamation—“splintered in the straits!”—is the 175 culmination of the s alliteration that had again begun to gather force in this stanza. “Straits” provides a slant rhyme with the opening line, but conceptually reduces the “planet-sequinned heights” to a narrow passage; similarly, it transforms the diffuse “air” of line 38 to an anagram straitened by the “st–ts” chiastic frame. This stanza brings us perhaps the closest of any in the poem to epic—we are presented with character, location, speech, and action, all in the context of Classic myth—and yet there is no stable narrative; the emphasis, instead, is on the act of calling and its temporal persistence. Here, Crane establishes connection even while dramatizing its difficulty; he “reminds us that the death-blocked conversation between the dead and the living is identical to the conversation between the living and the living, which is enabled and blocked in life by our unmingling bodies” (Grossman 148). The sixth stanza, like the third, synthesizes the major thematic and sonic patterns of the two stanzas that precede it. Instances of calling proliferate in the apostrophes that crowd these lines and the various epithets designated to the object of that call (“Tall Vision-of-the-Voyage,” “Bridge,” “O Choir,” “Psalm of Cathay[,]” “O Love”). The theme of time also resurfaces, “translat[ed]” into the “multitudinous Verb” that is both “fuse[d]” into a unity and “recast / In myriad syllables”: logos here is at once word (verbum), action, and the result of the compound speaker’s translation. What it would mean for time to be translated to language is not entirely clear, but the next octave gives us three possible conceptions: We left the haven hanging in the night— Sheened harbor lanterns backward fled the keel. Pacific here at time’s end, bearing corn,— Eyes stammer through the pangs of dust and steel. And still the circular, indubitable frieze Of heaven’s meditation, yoking wave To kneeling wave, one song devoutly binds— The vernal strophe chimes from deathless strings! 176 These lines, which mark the beginning of the second half of the poem, describe a voyage that implies linear temporal progress—“[w]e left the haven”—even as the confusion between moving and static objects produces spatial disorientation (rather than the passengers on the ship moving forward, the harbor lanterns seem to fly backwards from them). We then find ourselves “here” at “time’s end,” peaceful, but also in some way opposed to the subject of the poem, as “Pacific” counters “Atlantic” and “Atlantis.” Finally, there is “the circular, indubitable frieze / Of heaven’s meditation, yoking wave / To kneeling wave,” a spatial conception of time (“frieze”) that calls for circular repetition so that seasonality may be invoked (with “vernal strophe”) in the context of eternal return (“deathless strings”). The speaking subject that describes these alternate conceptions is, significantly, the first-person plural “[w]e.” Although this voice is not identified until the second half of the poem, the “we” that includes a singular I and other(s), that speaks and stands for a larger whole, has been prepared for by the quick shuttling back and forth between “sibylline voices” and the singular issue of “a god,” the “labyrinthine mouths of history” and “one vibrant breath,” the “multitudinous Verb” and the “myriad syllables.” It also heralds a series of vacillations between we, “Thou,” and “me,” an instability that is dramatized in the following stanza by “the many twain” who sing within the “single chrysalis” constituted by the “Thou” this voice addresses: O Thou steeled Cognizance whose leap commits The agile precincts of the lark’s return; Within whose lariat sweep encinctured sing In single chrysalis the many twain,— Of stars Thou art the stitch and stallion glow And like an organ, Thou, with sound of doom— Sight, sound and flesh Thou leadest from time’s realm As love strikes clear direction for the helm. Who speaks, or sounds, for whom here? The properties of synecdoche are so stretched that, in the next stanza, images and actions are represented not by their parts, but by their 177 negatives: the “unshadow” that is “unborne,” the “white escarpments” that cancel the “black embankments” of the second stanza: Swift peal of secular light, intrinsic Myth Whose fell unshadow is death’s utter wound,— O river-throated—iridescently unborne Through the bright drench and fabric of our veins; With white escarpments swinging into light, Sustained in tears the cities are endowed And justified conclamant with ripe fields Revolving through their harvests in sweet torment. We have not been able to exit “from time’s realm,” but instead find ourselves “revolving” through the “sweet torment” of the temporal realm, from part to whole and back again, from assertion to (and through) contradiction. Wherever the eternal is posited, the eruption of the temporal is not far behind: Forever Deity’s glittering Pledge, O Thou Whose canticle fresh chemistry assigns To wrapt inception and beatitude,— Always through blinding cables, to our joy, Of thy white seizure springs the prophecy: Always through spiring cordage, pyramids Of silver sequel, Deity’s young name Kinetic of white choiring wings . . . ascends. Here, the “Thou” addressed is “Forever” a “Pledge,” a promise presupposing a future, much like the ideas of “sequel” and “prophecy” (which here provides a rhyme with “history,” linking back to the second stanza and refreshing this rhyme for the stanza to follow). “Always” is “always through,” always kinetic, always moving spatially and through time; its repetition in verses 76 and 79 enacts the “sequel” that it precedes. 178 Migrations that must needs void memory, Inventions that cobblestone the heart,— Unspeakable Thou Bridge to Thee, O Love. Thy pardon for this history, whitest Flower, O Answerer of all,—Anemone,— Now while thy petals spend the suns about us, hold— (O Thou whose radiance doth inherit me) Atlantis,—hold thy floating singer late! So to thine Everpresence, beyond time, Like spears ensanguined of one tolling star That bleeds infinity—the orphic strings, Sidereal phalanxes, leap and converge: One Song, one Bridge of Fire! Is it Cathay, Now pity steeps the grass and rainbows ring The serpent with the eagle in the leaves . . . ? Whispers antiphonal in azure swing. With the penultimate stanza, formal repetition brings to a head several patterns. “Memory” (81) echoes “prophecy” (77) and “history” (11) in its rhyme and in its temporal reference; in the lines that follow, “history” returns at the center of line 84, and the rhyme repeats again in the address to “Anemone” (85) and in the parenthetical “me” which is the first reference in the poem to a single speaking subject. Temporally, the stanza points forward and backward, to history and prophecy, as the speaking voice narrows from “us” (86) to “me,” “thy floating singer.” This narrowing, which is accompanied by other cancellations and erasures—memory “void[ed],” the heart “cobblestone[d],” the speaker asking a “pardon” for history and referring to that which is “[u]nspeakable”—might seem, if we accept the poem as lyric, to point to lyric’s inevitable hermetic tendency. However, Crane’s own criticism describes the paradox by which the creative principle, “when seeming most to narrow and confine the subject (to itself). . . proceeds to open and extend it by this purification” (Crane, “A Pure Approach . . . ” 177). Like Dickinson, he insists upon “radical connection” (Cameron, Choosing Not Choosing 188), which is revealed in this stanza precisely at the point where the speaker emerges from the connective patterns that 179 precede and produce “me.”26 Gabriel’s analysis of the last stanza of “Atlantis” makes the case for the way in which this poem, despite its lyrical tendencies, engages with history in a way that is traditionally assumed to be beyond the bounds of lyric: Hart Crane’s dialectic against necessity . . . can be read to some extent as politically progressive—as a non-escapist intervention with history—if we apply Benjamin’s powerful notion of the ‘now.’ This now is revolutionary to the extent that as it ‘leaps’ back in time it can also ‘revise’ the future. One can read the last line of ‘Atlantis,’ ‘Whispers antiphonal in azure swing,’ as a quiet but dialectical voice opposing inevitability: a sublime that though ceasing to roar is just as potent as a whisper. . . . Through the sublime, Crane might be telling us, we come not to the end of history but to the point where, potentially at least, history will not end us. (183) Gabriel says that in this passage he “can speak only for Crane” (183), and his claim subtends his larger argument about the status of The Bridge as a modernist epic. But what he describes here is not, I think, specific to Crane—although “Atlantis” provides a striking example of this temporal operation—or to the genre of epic poetry. Rather, it is characteristic of lyric: the lyric poem is constituted by language that, because of its emphasis on the non-rational logic of formal and 26 My sense here of the way in which Crane’s lyric voice is attentive to the other shares some affinities with Rachel Cole’s essay on Wallace Stevens, in which she argues that “[f]or Stevens, the sonic intensity of lyric speech seems to represent . . . a force that can conciliate without resolving differences—not between sound and sense but between persons (actors and audiences, speakers and readers) who stand before one another as willful agents. . . . Lyric poetry, and more specifically lyric repetition, with its ‘final,’ ‘perfect,’ ‘unalterable vibration,’ offers a technology for pro- ducing peace” (394). While Cole focuses on the way lyric may produce “accord” between persons by way of formal closure, I aim to point up the ways in which the temporal operation of lyric allows the lyric voice attend to others. As Levinas has argued, “Knowing conceals re-presentation and reduces the other to presence and co-presence. Time, on the contrary, in its dia-chrony, would signify a relationship that does not compromise the other’s alterity, while still assuring its non-indifference to ‘thought’” (Time and the Other 31). In the realm of poetry, when the lyric is assumed to be limited to the present moment, it follows that its subjective horizon is similarly narrowed. But lyric poetry is not only a poetry of the present, and the constant revision and anticipation of pattern required by lyric reading encourages a similar openness toward the engagement of the lyric voice with other voices (speaking to, as, for, with, etc.). 180 sonic patterns, continually calls on us to leap back, revising our expectations in retrospect, and to project forward, reforming the fulfillments we anticipate in the act of reading. It is, as Cameron argued of Dickinson’s work, a genre that is “open” and “multiply determined” (Choosing Not Choosing 188) rather than solipsistic and closed, and its capabilities are categorically different from those of epic. Crane’s fear, referenced in the epigraph to this chapter, that he would meet “the kind of half-success which is worse than failure” in “locating the interrelations between sources, facts and appearances” was unfounded in the sense that it could not apply to the genre in which he was writing: lyric poetry does not require stable referents and interrelations or a narrative plot, but instead, by asking us to read the non-discursive patterns that propel the poem forward and simultaneously alter our sense of its past, it insists upon—while dramatizing the difficulty of—“radical connection” between past and present, the dead and the living, the voice of the self and the other. 181 Works Cited Abraham, Nicolas. Rhythms: On the Work, Translation, and Psychoanalysis. Trans. Benjamin Thigpen and Nicholas T. Rand. Stanford, CA: Stanford University Press, 1995. —–. Outline of a Phenomenology of Poetic Expression. Rhythms: On the Work, Translation, and Psychoanalysis. 1–64. —–. Rhytmizing Consciousness. Rhythms: On the Work, Translation, and Psychoanalysis. 65– 106. Acheson, Susan. “‘Conceived at the Grave’s Edge’: The Esoteric Eschatology of H.D.’s Trilogy.” Literature Theology 12.2 (1998): 187–204. Agamben, Giorgio. The End of the Poem: Studies in Poetics. Trans. Daniel Heller-Roazen. Stanford: Stanford University Press, 1999 (1996). Allen, Nicholas. “Yeats, Spengler, and A Vision After Empire,” in Modernism and Colonialism, eds. Richard Begam and Michael Valdez Moses. Durham, NC: Duke UP, 2007. 209–225. Allison, Jonathan. “Yeats and Politics.” The Cambridge Companion to W. B. Yeats. 185–205. Altieri, Charles. “How the ‘New Modernist Studies’ fails the Old Modernism.” Textual Practice 26(4) (2012): 763–782. —–. “Lyrical Ethics and Literary Experience.” Mapping the Ethical Turn: A Reader in Ethics, Culture, and Literary Theory. Eds. Todd F. Davis and Kenneth Womack. Charlottesville, VA: University Press of Virginia, 2001. 30–58. Ashton, Jennifer. From Modernism to Postmodernism: American Poetry in the Twentieth Century. Cambridge, UK: Cambridge University Press, 2005. Attridge, Derek. Poetic Rhythm: An Introduction. Cambridge, UK: Cambridge University Press, 2008. Auerbach, Erich. “Figura.” Scenes from the Drama of European Literature. Trans. Ralph Man- heim. Minneapolis: University of Minnesota Press, 1984 (1959). 11–78. Augustine. Confessions. Trans. Henry Chadwick. Oxford, UK: Oxford University Press, 2008. 182 Bakhtin, M. M. The Dialogic Imagination. Ed. Michael Holquist. Trans. Caryl Emerson and Michael Holquist. Austin, TX: University of Texas Press, 1981. Ben-Merre, David. “The Brawling of a Sparrow in the Eaves: Vision and Revision in W. B. Yeats.” Journal of Modern Literature 31:4 (Summer 2008): 71–85. Berman, Jessica. Modernist Commitments: Ethics, Politics, and Transnational Modernism. New York: Columbia University Press, 2011. Bergson, Henri. Creative Evolution. Trans. Arthur Mitchell. Mineola, NY: Dover Publications, Inc., 1998. ——. Matter and Memory. Trans. N. M. Paul and W. S. Palmer. Brooklyn, NY: Zone Books, 1991. Best, Stephen, and Sharon Marcus. “Surface Reading: An Introduction.” Representations 108(1) (Fall 2009): 1–21. Blake, Howard. “Thoughts on Modern Poetry.” The Sewanee Review 43(2) (April–June 1935): 187–196. Blasing, Mutlu Konuk. “Designs on History: Ezra Pound, the Puritans, and Self-Fulfilling Prophe- cies.” The Calvinist Roots of the Modern Era. Eds. Aliki Barnstone, Michael Tomasek Manson, and Carol J. Singley. Hanover, NH: University Press of New England, 1997. 3–19. —–. “Hart Crane: Inscribing the Sublime.” American Poetry: The Rhetoric of Its Forms. New Haven, CT: Yale University Press, 1987. —–. Lyric Poetry: The Pain and the Pleasure of Words. Princeton, NJ: Princeton University Press, 2007. Bloom, Harold. The Anatomy of Influence: Literature as a Way of Life. New Haven, CT: Yale University Press, 2011. Bornstein, George. “Introduction.” The Early Poetry, Vol. I. 3–16. Bornstein, George. “Introduction.” The Early Poetry, Vol. II. 1–22. Bradford, Curtis. Yeats at Work.Carbondale, IL: Southern Illinois University Press, 1965. 183 —–. “Yeats’s Last Poems Again,” The Dolmen Press Yeats Centenary Papers MCMLXV, ed. Liam Miller. Dublin, 1968. 259–88 Brogan, T. V. F. “Rhythm.” The New Princeton Encyclopedia of Poetry and Poetics. Eds. Alex Preminger and T. V. F. Brogan. Princeton, NJ: Princeton University Press, 1993. Brooker, Jewel Spears, ed. T. S. Eliot: The Contemporary Reviews. Cambridge, UK: Cambridge University Press, 2004. Brooks, Peter. Reading for the Plot: Design and Intention in Narrative. Cambridge, MA: Harvard University Press, 1984. Bryant, Marsha, and Mary Ann Eaverly. “Egypto-Modernism: James Henry Breasted, H.D., and the New Past.” Modernism/Modernity 14(3) (2007): 435–453. Burke, Kenneth. “Four Master Tropes.” A Grammar of Motives, and a Rhetoric of Motives. Cleve- land: World Publishing Co., 1962. 503–517. Burwick, Frederick and Paul Douglass, eds. The Crisis in Modernism: Bergson and the Vitalist Controversy. New York: Cambridge University Press, 1992. The Cambridge Companion to W. B. Yeats. Eds. Marjorie Howes and John Kelly. Cambridge, UK: Cambridge UP, 2006. Cameron, Sharon. Choosing Not Choosing: Dickinson’s Fascicles. Chicago: University of Chicago Press, 1992. —–. Lyric Time: Dickinson and the Limits of Genre. Baltimore: Johns Hopkins University Press, 1979. Chapman, Wayne K. Yeats’s Poetry in the Making. New York, NY: Palgrave Macmillan, 2010. Cohen, Ralph. “History and Genre,” NLH 17(2) (1986): 203–18. Cole, Rachel. “Rethinking the Value of Lyric Closure: Giorgio Agamben, Wallace Stevens, and the Ethics of Satisfaction.” PMLA 126(2) (March 2011): 383–397. Comentale, Edward P. “Thesmophoria: Suffragettes, Sympathetic Magic, and H.D.’s Ritual Poet- ics.” Modernism/Modernity 8(3) (2001): 471–492. Cooper, John Xiros. T. S. Eliot and the Ideology of Four Quartets. Cambridge, UK: Cambridge 184 University Press, 1995. Crane, Hart. Complete Poems Selected Letters. New York, NY: The Library of America, 2006. Culler, Jonathan. “Apostrophe.” Diacritics 7(4) (Winter 1977): 59–69. —–. “Lyric, History, and Genre.” New Literary History 40(4) (Autumn 2009): 879–899. —–. “Why Lyric?” PMLA 123.1 (January 2008): 201–206. Cullingford, Elizabeth. Yeats, Ireland, and Fascism. London: Macmillan, 1981. Dayan, Peter, and David Evans. “Rhythm in Literature after the Crisis in Verse.” Paragraph 33:2 (2010): 147–157. De Man, Paul. “Anthropomorphism and Trope in the Lyric.” The Rhetoric of Romanticism. New York: Columbia University Press, 1984. 239–262. —–. “Literary History and Literary Modernity.” Blindness and Insight. Minneapolis: University of Minnesota Press, 1983 (1971). 142–165. —–. “Lyric and Modernity.” Blindness and Insight. 166–186. —–. “The Rhetoric of Temporality.” Blindness and Insight. 187–228. Dean, Tim. “Hart Crane’s Poetics of Privacy.” American Literary History. 8(1) (Spring 1996): 83–109. Deleuze, Gilles. Difference and Repetition. Trans. Paul Patton. New York, NY: Columbia UP, 1968 (1994). Derrida, Jacques. Archive Fever. Trans. Eric Prenowitz. Chicago: The University of Chicago Press, 1996. —–. Of Grammatology. Trans. Gayatri Chakravorty Spivak. Baltimore, MD: Johns Hopkins UP, 1967, 1974, 1976, (1997). —–. Specters of Marx: The State of the Debt, the Work of Mourning, and the New International. Trans. Peggy Kamuf. New York: Routledge, 1994 (1992). —–. “The Law of Genre.” Trans. Avital Ronell. Critical Inquiry 7(1) (Autumn, 1980): 55–81. 185 Dimock, Wai Chee. Genres as Fields of Knowledge. PMLA 122(5) (October 2007): 1377–1388. Douglass, Paul. Bergson, Eliot, and American Literature. Lexington, KY: The University Press of Kentucky, 1986. Doyle, Laura, and Laura Winkiel, eds. Geomodernisms: Race, Modernism, Modernity. Bloom- ington: Indiana University Press, 2005. Dubrow, Heather. “The Interplay of Narrative and Lyric: Competition, Cooperation, and the Case of the Anticipatory Amalgam.” Narrative 14(3) (October 2006): 254–271. DuPlessis, Rachel Blau. “H.D. and Revisionary Myth-Making.” The Cambridge Companion to Modernist Poetry. Eds. Alex Davis and Lee M. Jenkins. New York: Cambridge University Press, 2007. 114–125. —–. “Gender Authority: ‘Another Region of Cause and Effect, Another Region of Question and Answer.’” H.D.: The Career of that Struggle. London: Harvester and Bloomington: Indi- ana University Press, 1986. 70–100. Eagleman, David. “Brain Time.” What’s Next? Dispatches on the Future of Science. Ed. Max Brockman. New York, New York: Vintage Books, 2009. Edelman, Lee. Transmemberment of Song: Hart Crane’s Anatomies of Rhetoric and Desire. Stan- ford, CA: Stanford University Press, 1987. Eliot, T. S. “Dante.” The Selected Prose of T. S. Eliot. Ed. Frank Kermode. New York: Farrar, Straus and Giroux, 1975. 205–230. —–. Four Quartets. New York, NY: Harcourt Inc., 1971. —–. “In Memoriam.” The Selected Prose of T. S. Eliot. 239–247. ——. On Poetry and Poets. New York, NY: Farrar, Straus and Giroux, 2009. —–. “Religion and Literature.” The Selected Prose of T. S. Eliot. 97–106. —–. The Complete Poems and Plays: 1909–1950. New York, NY: Harcourt Brace Company, 1980. —–. “The Music of Poetry.” On Poetry and Poets. New York: Farrar, Straus and Giroux, 1957. 17–33. 186 —–. “The Three Voices of Poetry” On Poetry and Poets. 96–112. —–. “Tradition and the Individual Talent.” Selected Prose of T. S. Eliot. 37–44. —–. “What is a Classic?” The Selected Prose of T. S. Eliot.115–131. Ellmann, Richard. Yeats: The Man and the Masks. (1948) New York, NY: W. W. Norton & Company, 1999. Emerson, Ralph Waldo. “The Poet.” Ralph Waldo Emerson: Selected Essays, Lectures, and Poems. New York: Random House, 2007. 207–228. Evans, David. “Paul Val´ery and the Search for Poetic Rhythm.” Paragraph 33:2 (2010): 158–175. Fehrman, Carl. Poetic Creation: Inspiration or Craft. Trans. Karin Petherick. Minneapolis: University of Minnesota Press, 1980 (1974). Ferrer, Daniel and Michael Groden. “Introduction: A Genesis of French Genetic Criticism.” Ge- netic Criticism. 1–16. Finneran, Richard J. “Preface.” The Collected Poems of W. B. Yeats. xxiii–xxv. —–. “Text and Interpretation in the Poems of W. B. Yeats.” Representing Modernist Texts: Editing as Interpretation. Ed. George Bornstein. Ann Arbor, MI: University of Michigan Press, 1991 (1994). Forster, E. M. Aspects of the Novel. Orlando, FL: Harcourt, Inc., 1955. Friedman, Susan Stanford. “Planetarity: Musing Modernist Studies.” Modernism/modernity 17(3) (September 2010): 471–499. Fuss, Diana. “Last Words.” ELH 76(4) (Winter 2009): 877–910. Gabriel, Daniel. Hart Crane and the Modernist Epic: Canon and Genre Formation in Crane, Pound, Eliot, and Williams. New York, NY: Palgrave Macmillan, 2007. Gardner, Helen. The Composition of Four Quartets. New York: Oxford University Press, 1978. Genetic Criticism: Texts and Avant-textes. Eds. Jed Deppman, Daniel Ferrer, and Michael Groden. Philadelphia: University of Pennsylvania Press, 2004. 187 Genette, G´erard. Narrative Discourse: An Essay in Method. Trans. Jane E. Lewin. Ithaca, NY: Cornell University Press, 1980 (1972). —–. The Architexte: An Introduction. Trans. Jane E. Lewin. Berkeley, CA: University of Califor- nia Press, 1992 (1979). Glaser, Brian Brodhead. “H.D.’s Helen in Egypt: Aging and the Unconscious.” Journal of Modern Literature 28.4 (Summer 2005): 91–109. Goodspeed-Chadwick, Julie. “Validating War Experience through Literary Witnessing: The Poet- ics of the Poet-Prophet and the Politics of Trauma and Healing in H.D.’s Trilogy.” Mod- ernist Women Writers and War: Trauma and the Female Body in Djuna Barnes, H.D., and Gertrude Stein. Baton Rouge: Louisiana State University Press, 2011. 59–91. Gould, Warwick. “Writing the Life of the Text: The Case of W. B. Yeats.” Miscel´anea 30 (2004): 9–34. Graham, Sarah H. S. “The Echoing Spell of H.D.’s Trilogy.” Signets: Reading H.D. Eds. Susan Stanford Friedman and Rachel Blau DuPlessis. Madison, WI: University of Wisconsin Press, 1990. 297–317. Grossman, Allen. “On Communicative Difficulty in General and ‘Difficult’ Poetry in Particular.” True-Love: Essays on Poetry and Valuing. Chicago, IL: University of Chicago Press, 2009. Grove, Robin. “Pereira and after: the cures of Eliot’s theater.” The Cambridge Companion to T. S. Eliot”. Ed. A. David Moody. Cambridge, UK: Cambridge University Press, 1994. 158–175. H.D. Tribute to Freud. New York: New Directions, 1974 (1956). —–. Trilogy. New York: New Directions Books, 1998. Hart, Matthew. Nations of Nothing But Poetry. Oxford, UK: Oxford University Press, 2010. Hay, Louis. “Genetic Criticism: Origins and Perspectives.” Genetic Criticism. 17–35. Helt, Brenda S. “Reading History in The Gift and Tribute to Freud.” The Cambridge Companion to H.D. Eds. Nephie J. Christodoulides and Polina Mackay. Cambridge: Cambridge University Press, 2012. 188 Hill, Leslie. “‘Distrust of Poetry: Levinas, Blanchot, Celan,” MLN 120 (2005): 986–1008. Hollenberg, Donna Krolik, ed. Between History and Poetry: The Letters of H.D. and Norman Holmes Pearson. Iowa City: University of Iowa Press, 1997. Husserl, Edmund. On the Phenomenology of the Consciousness of Internal Time (1893–1917). Trans. John Barnett Brough. Dordrecht, The Netherlands: Kluwer Academic Publishers, 1991 (1980). Huyssen, Andreas. After the Great Divide: Modernism, Mass Culture, Postmodernism. Bloom- ington: Indiana University Press, 1987. Irwin, John. Hart Crane’s Poetry: “Appollinaire Lived in Paris, I Live in Cleveland, Ohio”. Baltimore, MD: The Johns Hopkins University Press, 2011. Iser, Wolfgang. The Act of Reading: A Theory of Aesthetic Response. Baltimore: Johns Hopkins University Press, 1980. Jackson, Virginia. “Who Reads Poetry?” PMLA 123(1) (2008): 181–187. —–. Dickinson’s Misery: A Theory of Lyric Reading. Princeton, NJ: Princeton University Press, 2005. James, David, and Urmila Seshagiri. “Metamodernism: Narratives of Continuity and Revolution.” PMLA 129(1) (2014): 87–100. Jarvis, Simon. “For a Poetics of Verse.” PMLA 125.4 (2010): 931–935. —–. “Musical Thinking: Hegel and the Phenomenology of Prosody.” Paragraph 28(2) (2005): 57–71. —–. “The Melodics of Long Poems.” Textual Practice 24(4) (2010): 607–621. —–. “Unfree Verse: John Wilkinson’s The Speaking Twins.” Paragraph 33(2) (2010): 280–295. Jeffreys, Mark. “Ideologies of Lyric: A Problem of Genre in Contemporary Anglophone Poetics.” PMLA 110(2) (March, 1995): 196–205. Johnson, James William. “Lyric.” The Princeton Encyclopedia of Poetry and Poetics. Eds. Alex Preminger and T. V. F. Brogan. Princeton: Princeton University Press, 1993. 713–727. 189 Keller, Lynn. “Singing Spaces: Fractal Geometries in Cole Swensen’s Oh.” Journal of Modern Literature 31(1) (Fall 2007): 136–160. Kenner, Hugh. The Invisible Poet: T. S. Eliot. New York, NY: McDowell, Obolensky, 1959. —–. “The Sacred Book of the Arts.” The Sewanee Review 64(4): 574–590. Kermode, Frank. The Sense of an Ending: Studies in the Theory of Fiction. Oxford, UK: Oxford University Press, 2000 (1966). Kern, Stephen. The Culture of Time and Space, 1880–1918. Cambridge, MA: Harvard University Press, 2003 (1983). Kovelman, Ioulia, Kira Mascho, Louisa Millcott, Alyssa Mastic, Bianca Moiseff, and Mark H. Shalinsky. “At the rhythm of language: Brain bases of language-related frequency percep- tion in children.” NeuroImage 60 (2012): 673–682. Kramer, Lawrence. Hart Crane’s “The Bridge”: An Annotated Edition. New York, NY: Fordham University Press, 2011. Kuch, Peter. “A Few Twigs from the Wild Bird’s Nest: Yeats the European.” In Jeffares, A. Nor- man. (ed.). Yeats the European. Gerrards Cross: Colin Smythe Ltd; Savage, Maryland: Barnes and Noble Books, 1989. 101–118. Lawrence, D. H. “The Poetry of the Present.” The Norton Anthology of Modern and Contemporary Poetry. Vol. 1. Eds. Jahan Ramzani, Richard Ellman, and Robert O’Clair. New York, NY: W. W. Norton Co., 2003. 960–963. Levenson, Michael H. A Genealogy of Modernism: A Study of English Literary Doctrine 1908– 1922. Cambridge: Cambridge University Press, 1984. Le Brun, Philip. “T. S. Eliot and Henri Bergson.” The Review of English Studies, New Series, 18:70 (May, 1967): 149–161. ——. “T. S. Eliot and Henri Bergson.” The Review of English Studies, New Series, 18:71 (August, 1967): 274–286. ´ Levinas, Emmanuel. Difficile Libert´e. Third Edition. Paris: Editions Albin Michel, 1976 (1963). —–. “Persons or Figures (On Emma¨us, by Paul Claudel).” Difficult Freedom. Trans. S´ean Hand. Baltimore: The Johns Hopkins University Press, 1990. 119–122. 190 —–. Time and the Other. Trans. Richard A. Cohen. Pittsburgh, PA: Duquesne University Press, 1987. Longenbach, James. “Matthew Arnold and the Modern Apocalypse.” PMLA 104(5) (October 1989): 844–855. Madden, William A. “Framing the Alices.” PMLA 101(3) (May 1986): 362–373. Mao, Douglas, and Rebecca Walkowitz. “The New Modernist Studies.” PMLA 123(3): 737–748. —–. “Introduction: Modernisms Bad and New.” Bad Modernisms. Durham, NC: Duke University Press, 2006. 1–18. Martin, Meredith. The Rise and Fall of Meter: Poetry and English National Culture, 1860–1930. McHale, Brian. “Beginning to Think about Narrative in Poetry.” Narrative 17:1 (2009): 11–27. Merleau-Ponty, Maurice. Phenomenology of Perception. Trans. Colin Smith. London and New York: Routledge, 2002 (1945). Miller, J. Hillis. Fiction and Repetition: Seven English Novels. Cambridge, MA: Harvard Univer- sity Press, 1982. Moody, A David. “Four Quartets: music, word, meaning and value.” The Cambridge Companion to T. S. Eliot”. Ed. A. David Moody. Cambridge, UK: Cambridge University Press, 1994. 142–157. Morgan, Monique R. Narrative Means, Lyric Ends: Temporality in the Nineteenth-Century British Long Poem. Columbus, OH: The Ohio State University Press, 2009. Morris, Adalaide. How to Live / What to Do: H.D.’s Cultural Poetics. Urbana and Chicago: University of Illinois Press, 2003. Newman, J. K. “Epic.” The Princeton Encyclopedia of Poetry and Poetics. 361–375. Nietzsche, Friedrich. The Gay Science (1882). Excerpted in The Nietzsche Reader, Ed. Keith Ansell Pearson and Duncan Large. Malden, MA: Blackwell Publishing, 2006. 207–237. —–. Thus Spake Zarathustra. Trans. R. J. Hollingdale. London: Penguin Books, 1961, 1969 (2003). 191 North, Michael. The Political Aesthetic of Yeats, Eliot, and Pound. Cambridge, UK: Cambridge University Press, 1991. O’Brien, Conor Cruise. “Passion and Cunning: An Essay on the Politics of W. B. Yeats.” First published 1965. Excerpted in Yeats’s Political Identities: Selected Essays. Ed. Jonathan Allison. Ann Arbor, MI: University of Michigan Press, 1996 (1999). 29–56. Orwell, George. “Points of View. T. S. Eilot.” T. S. Eliot: The Contemporary Reviews. 452–58. Oxford English Dictionary Online. “Lyric.” Accessed February 15, 2014. —–. “Sooth.” Accessed January 12, 2013. Parkinson, Thomas. W. B. Yeats, Self-Critic: A Study of His Early Verse and The Later Poetry. Berkeley and Los Angeles: University of California Press, 1951, 1964 (1971). Pearson, Norman Holmes. Foreword. Trilogy. By H.D. New York: New Directions Books, 1973. Pethica, James. “Introduction.” Last Poems, xxiii–liii. Phelan, James. “Rhetorical Literary Ethics and Lyric Narrative: Robert Frost’s ‘Home Burial.’” Poetics Today 25(4) (Winter 2004): 627–651. Phillips, Siobhan. The Poetics of the Everyday. New York, NY: Columbia University Press, 2010. Poe, Edgar Allan. “The Poetic Principle.” The Complete Works of Edgar Allan Poe. Kindle Edition, 2014. Potkay, Adam. “Wordsworth and the Ethics of Things.” PMLA 123(2): 390–404. Pound, Ezra. “A Retrospect.” The Norton Anthology of Modern and Contemporary Poetry. 929– 938. —–. “Cavalcanti.” Literary Essays of Ezra Pound. New York: New Directions, 1935. —–. Ezra Pound to His Parents: Letters 1895–1929. Ed. Mary de Rachewiltz, A. David Moody. Prins, Yopie. “Historical Poetics, Dysprosody, and The Science of English Verse.” PMLA 123(1) (2008): 229–234. Rainey, Lawrence. “Introduction.” Modernism: An Anthology. Malden, MA: Blackwell Publish- 192 ing, 2005. ix–xxix. Ramazani, Jahan. “Traveling Poetry.” Modern Language Quarterly 68(2) (June 2007): 281–303. Rayfield, Donald. Anton Chekhov: A Life. New York: Henry Holt Company, 1998. Ricoeur, Paul. “Narrative Time.” Critical Inquiry 7(1) (Autumn 1980): 169–190. —–. Time and Narrative. Vol. 1. Trans. Kathleen McLaughlin and David Pellauer. Chicago: The University of Chicago Press, 1984. Riddel, Joseph. “Hart Crane’s Poetics of Failure.” ELH 33.4 (Dec. 1966): 473–496. Robinson, Peter. Poetry, Poets, Readers: Making Things Happen. Oxford, UK: Clarendon Press, 2002. Scanlon, Mara. “Ethics and the Lyric: Form, Dialogue, Answerability.” College Literature 34(1) (2007): 1–22. Schultz, Susan M. A Poetics of Impasse in Modern and Contemporary American Poetry. Tuscaloosa, AL: University of Alabama Press, 2005. —–. “The Success of Failure: Hart Crane’s Revisions of Whitman and Eliot in The Bridge.” South Atlantic Review 54(1) (Jan., 1989): 55–70. Schwartz, Sanford. “Bergson and the Politics of Vitalism.” The Crisis in Modernism. 277–305. Scruton, Roger. “T. S. Eliot as Conservative Mentor.” The Intercollegiate Review. 39(1/2) (Fall 2003/Spring 2004): 44–54. Stallworthy, Jon. Between the Lines: Yeats’s Poetry in the Making. London, UK: Oxford UP, 1963. —–. Vision and Revision in Yeats’s Last Poems. London, UK: Oxford UP, 1969. Stevens, Wallace. The Necessary Angel. New York: Vintage Books, 1951. Tate, Allan. “Hart Crane and the American Mind.” Poetry 40.4 (Jul. 1932): 210–216. Vendler, Helen. Our Secret Discipline: Yeats and Lyric Form. Cambridge, MA: Harvard UP, 2007. 193 Von Hallberg, Robert. “Literary History and the Evaluation of Poetry.” American Literary History 15(1) (Spring 2003): 7–13. Wesling, Donald. The Chances of Rhyme. Berkeley, CA: University of California Press, 1980. Williams, Raymond. The Politics of Modernism. New York: Verso, 1989. Winters, Yvor. “Hart Crane’s Poems: White Buildings, by Hart Crane.” Poetry 30.1 (Apr. 1927): 47–51. Winters, Yvor. “The Progress of Hart Crane.” Poetry 36(3) (Jun. 1930): 153–165. Wright, George T. “The Lyric Present: Simple Present Verbs in English Poems.” PMLA 89(3) (May 1974): 563–579. Yeats, W. B. “A Visionary.” in The Celtic Twilight, 17–28. —–. “Art and Ideas.” The Collected Works of W. B. Yeats. Vol. IV: Early Essays. Eds. George Bornstein and Richard J. Finneran. New York: Srcribner, 2007. 250–256. —–. Autobiography. Excerpted in Yeats’s Poetry, Drama, and Prose. Ed. James Pethica. New York, NY: W. W. Norton & Company, 2000. 225–240. —–. from Dramatis Personae. Yeats’s Poetry, Drama, and Prose. 247–250. —–. from “Introduction” to the Scribner edition of Yeats’s Collected Works. Yeats’s Poetry, Drama, and Prose. 312–313. —–. “Modern Poetry: A Radio Broadcast.” The Collected Works of W. B. Yeats, Vol. V: Later Essays. New York, New York: Charles Scribner’s Sons, 1994. 89–102. —–. Last Poems. Ed. James Pethica. Ithaca, NY: Cornell University Press, 1997. —–. “Poetry and Tradition.” The Collected Works of W. B. Yeats. 180–190. —–. The Celtic Twilight. London: Macmillan and Co., 1894. —–.The Collected Letters of W. B. Yeats. Vols. 1–3 (1865–1904) and the Unpublished Letters (1905-1939). InteLex Past Masters Electronic Edition. Charlottesville, Virginia: InteLex Corporation, 2002. 194 —–.The Collected Poems of W. B. Yeats. Ed. Richard J. Finneran. Second Edition. New York, NY: Simon & Schuster Inc., 1996. —–. The Early Poetry. Vol. I: Mosada and The Island of Statues. Ed. George Bornstein. Ithaca, NY: Cornell University Press, 1987. —–. The Early Poetry. Vol. II: “The Wanderings of Oisin” and Other Early Poems to 1895. Ed. George Bornstein. Ithaca, NY: Cornell University Press, 1994. —–. “The Return of Ulysses.” The Collected Works of W. B. Yeats. 146–149. —–. “The Song of the Happy Shepherd.” Dublin University Review 1 (7): 1885, 230–231. Viewed via the Hathi Trust Digital Library. —–. “The Symbolism of Poetry.” 113–120. —–. The Trembling of the Veil (1922). Excerpted in Yeats’s Poetry, Drama, and Prose. 240–247.