jump to main content

PDF Version of this article
This Article as pdf

Contents of Volume 23
This Volume

View the whole debate
This Debate

View a list of articles with a similar Topic
Related Topics

More Articles on this Author
More Articles on this Author

You are here: All IssuesContents of Volume 23 → Nicholas Halmi
DebatesImage and Imagism → Nicholas Halmi
AuthorsT. E. Hulme → Nicholas Halmi
AuthorsEzra Pound → Nicholas Halmi

© Connotations 23.1 (2013/2014): 127-39
N.B. For purposes of citation, page numbers of the printed version are inserted in square brackets.

Symbolism, Imagism, and Hermeneutic Anxiety: A Response to Andrew Hay

Nicholas Halmi

I.

In an article published last year in Connotations, Andrew Hay proposed an intriguing analogy between the image as postulated in Ezra Pound's program for Imagist poetry and the symbol as conceived by various Romantic–era writers and defined normatively in my book The Genealogy of the Romantic Symbol (2007). The analogy is between relationships constitutive of the respective semiotic phenomena: the relationship of ontological content to meaning in the Romantic symbol, and of presentation to reception in the Imagist poem. While the symbol is supposed to represent the fusion of contrary states (Halmi, Genealogy 1–2), Pound's Imagist distich “In the Station of a Metro”—“The apparition of these faces in the crowd; / Petals on a wet, black bough” (Personæ 111)—is asserted to represent the fusion of an objective image with the reader's subjective response to it: “Just as the Romantic symbol necessitates a union—whether it is the contingent and the absolute, or the temporal and the transtemporal—Pound's poem works through a yoking together of different contingencies: the reader and the image” (Hay 314).

Proceeding from Pound's own statement that the poem embodies “the precise instant when a thing outward and objective transforms itself, or darts into a thing inward and subjective” (“Vorticism” 286), [→page 128] Hay interprets the poem's indistinct temporality (“timeless instant”; Hay 312) as enacting not only the poet's transformation of perceptual act (seeing the faces) into poetic image (the petals) but also the reader's interpretation of that transformation: “the image serves authorial intentionality, but this intention sits alongside the faculties of the poem's reader” (314). By anticipating, if in the most generalized way, the process of its reception, the poem in effect conjoins “the internality of the reader and the onticity of the image” (314).

Pleased as I am to see him making a literary critical use of my discussion of the symbol, I find Hay's analogy unconvincing for two reasons. The first is the heterogeneity of the terms of comparison. Hay understands the Image (a word I shall capitalize when referring to its application to Imagist poetry) to incorporate proleptically the reader's response to it. If that is correct, then the Image acknowledges, however qualifiedly or problematically, that it must be recognized as an image in order to function as such. In contrast, by identifying the meaning of the symbol with its ontological content—“Meaning here is simultaneously being itself,” F. W. J. Schelling taught in his lectures on art in 1802–03 (411)—Romantic theorists in effect denied the symbol an instituted character, in Gadamer's sense of the term (cf. 159–60).1 In other words, because its meaning was supposed to inhere in it, the symbol could be conceived as meaningful even when it was not empirically recognized or recognizable. So while the Image anxiously anticipates its recipient, the symbol remains sublimely indifferent to whether it has one or not.

Secondly, to the extent that it offers a critique of the Imagists' own theorization of their practice, Hay's account of the Image is what Niklas Luhmann would call a “second–order observation” (94–95), concerned with the understanding of representation rather than with the description of phenomena. But the concept of the Image is at least a second–order observation of something real and specifiable, namely poetic images, whereas the Romantics' concept of the symbol lacks a corresponding first–order observation. Indeed a central contention of my book is that the semiotic phenomenon theorized under the rubric [→page 129] of the symbol neither did nor could exist because the concept itself was incoherent, conflating the logically incommensurate relations of identity/difference and part/whole by founding a semiotics of identity (in which symbol is claimed to be the same as what it refers to) on a metaphysics of participation (in which the symbol is claimed to be a part of what it refers to). Hence the second sentence of the Genealogy: “This is not a study of poetic imagery” (1).

Admittedly, some critics have contested this insistence that the symbol was purely a theoretical construct; but since Hay accepts my description of the symbol, it is fair to criticize his use of that description in an analogy which presupposes exactly what I deny, that an image in a literary work could ever conform to the Romantic concept of the symbol. From my perspective, the symbol does not afford, as Pound's Imagist program does, the opportunity of being tested against poetic practice. Imagism is a theory of poetry, the Romantic symbol a theory of the meaningfulness supposedly inherent in the structure of the world itself. The latter is a contribution not to poetics but to aesthetics, for it is concerned with aisthêsis in its root sense, perception, and more particularly with that reformed mode of perception which Shelley, stealing a phrase from Coleridge, advocated as lifting the film of familiarity from the world of quotidian experience (see 533). If Imagism “represents a naturalizing of the poetic sign” (Levenson, Genealogy 150), then the Romantic theorization of the symbol constitutes a naturalizing of the symbol for the sake of making nature symbolic.

II.

The foregoing objections notwithstanding, Hay's juxtaposition of the theories of the Romantic symbol and the Poundian Image prompts reflection on whether there might not in fact be some underlying affinities between them. Or rather I should say renewed reflection, for since the 1960s critics of Modernism have tended to reject Frank Kermode's [→page 130] identification of the Poundian Image, under the awkwardly tailored philosophical clothing furnished by its stepfather T. E. Hulme, as a successor to the Romantic symbol in a line of descent through French symbolisme. In Romantic Image (1957) Kermode had argued that, notwithstanding Hulme's professed antipathy to Romanticism and Pound's explicit dissociation of Imagism from Symbolism, both shared the symbolist dream of liberation from discursiveness: “What it comes down to in the end is that Pound, like Hulme, like Mallarmé, and many others, wanted a theory of poetry based on the non–discursive concetto. In varying degrees they all obscurely wish that poetry could be written with something other than words, but since it can't, that words may be made to have the same sort of physical presence 'as a piece of string'” (136). In this account, the Imagist emphasis on the visuality of the Image, as in Hulme's description of poetry as “a visual concrete” language, a compromise for a language of intuition which would hand over sensations bodily” (“Romanticism” 70), amounted to an attempt to mitigate or repress the semiotic character of language.

Opposing Kermode's interpretation of Imagism as a rather naïvely conceived mode of verbal pictorialism, unable or unwilling to acknowledge its nineteenth–century intellectual ancestry, Herbert Schneidau (see 14–15, 29–31) and Donald Davie, followed with qualifications by Marjorie Perloff, accept Pound's own contention that Imagism was not a continuation of Symbolism by other means, but in fact “a radical alternative to it” (Davie 43; cf. Perloff 159). What Pound understood Symbolism to consist in may be gauged from two brief paragraphs in his essay “Vorticism.” Here he identifies Symbolism vaguely with “mushy technique” and more specifically with the semantically associative use of imagery: “a sort of allusion, almost of allegory […] a form of metonymy” (281).2

For examples of what he was taking about, Pound needed to look no further than the early poetry of W. B. Yeats, with its “golden apples of the sun” and “silver apples of the moon.” Surveying the contemporary poetic scene in January 1913, Pound criticized Yeats, whom he [→page 131] identified with the French Symbolists, for his belief in “the glamour and associations which hang near words”—in contrast to the nascent group of Imagists, who, following Ford Madox Hueffer (later Ford), favored “an exact rendering of things” (“Status Rerum” 125–26). Always an eager pedagogue, Pound undertook personally to wean Yeats off his rhetorical imprecision and sentimentalism during their winter together in Stone Cottage, in Sussex, in 1913, his efforts being rewarded by Yeats's volume Responsibilities of the following year. Reviewing that book, Pound affirmed that the poetical “fogs and mists since the nineties” were finally being dispelled by the “hard light” evident, for example, in the first five lines of “The Magi”—“a passage of imagisme” (“Later Yeats” 380; see also Levenson, Modernism 133).3 Yet Pound's reference in “Vorticism” to metonymy recalls a principle enunciated by Mallarmé in his “Réponses à des enquêtes sur l'évolution littéraire” (1891) and paraphrased by Arthur Symons in The Symbolist Movement in Literature (1899), the book that introduced the generation of Pound and Eliot to the French Symbolists. Trying to account for the abstruseness and unintelligibility, as he concedes it to be, of Mallarmé's late style, Symons exhorts the reader: “Remember his principle: that to name is to destroy, to suggest is to create” (71).4 Symons imagines Mallarmé's compositional process as the sequential substitution of images increasingly remote from the sensation in which the poem originated, perfection from the poet's perspective achieved when the path back to that originary sensation has been completely effaced and the reader, who sees only the final choice of images, is thoroughly bewildered.

III.

If Pound's assessment of Symbolism was polemically reductive, its purpose, Perloff proposes, was to throw into relief the distinctiveness of the Image in being neither polysemous nor mimetic, the latter term understood here to mean “pictorial.” Symbols were acceptable to [→page 132] Pound to the extent that they resisted, or at least did not demand, translation into ideas. Thus in the “Credo” section of his “Retrospect,” a statement cobbled together from various essays and notes and published in 1918, Pound declared that “the proper and perfect symbol is the natural object, that if a man use 'symbols' he must so use them that their symbolic function does not obtrude; so that a sense […] is not lost to those who do not understand the symbol as such, to whom, for instance, a hawk is a hawk” (259). Four years earlier, in the essay “Vorticism,” which Perloff characterizes as an “anti–Symbolist manifesto” (161), Pound had even more emphatically denied the Image, under its new rubric of Vorticism, semantic translatability: “The image is not an idea. It is a radiant node or cluster […] a VORTEX, from which and through which, and into which, ideas are constantly rushing” (289). Even if the Image itself cannot be paraphrased, however, the theory of it can be, and it amounts to two principles that were also central to Romantic symbolist theory: first, the self–identity of the image, a condition for which Coleridge coined the term tautegory—that is, “meaning the same thing as itself,” in contradistinction to allegory, in which the image means something different from itself (see Halmi, “Coleridge” 353–55)—and second, the irreducibility of the image to any particular meaning.

To be sure, in the poetical work generally considered in the 1920s and 1930s to be Pound's masterpiece, the sequence Hugh Selwyn Mauberley (1920), the poet manifestly violated his own dictum that “[t]o use a symbol with an ascribed or intended meaning is, usually, to produce very bad art” (“Vorticism” 284). Conceding this contradiction, Perloff observes that the beginning of the poem's third section, which rings the changes on the contrast between an idealized past and the debased present, offers up eight symbols in eight lines, each of which could be replaced by another without substantially changing the passage's poetic effect:

The tea–rose tea–gown etc.
Supplants the mousseline of Cos,
The pianola “replaces”
Sappho's barbitos.
[→page 133] Christ follows Dionysus,
Phallic and ambrosial
Made way for macerations;
Caliban casts out Ariel. (Personæ 186)

That Mauberley appears unambiguously to be a social commentary, deploring the degradation of consumer capitalism and the monstrosity of the recently concluded World War, accounts for much of its attraction to students of Modernism: “the age demanded,” among other things, a poem about the age. But the poem is uncharacteristic, Perloff argues, precisely because the substitutability of its images suggests that Pound proceeded from ideas rather than from images; and she accordingly displaces Mauberley from the center to the margins of the poet's achievement, as an anomalously transparent satire with a greater affinity to Eliot's roughly contemporaneous Waste Land (whose symbolism Pound's editing made more prominent) than to Pound's own subsequent Cantos, which are distinguished by linguistic indeterminacy (167–70, 181–83). Interpreting Mauberley in relation to Pound's evolving conception of his poetic project, Perloff is entirely convincing. One may still question, however, how qualitatively differentiated innovative linguistic indeterminacy and old–fashioned semantic obscurity are in readers' experience.

Like Perloff, Hugh Kenner and, more recently, Daniel Tiffany (see 21) define the essence of Imagism and Vorticism as the rejection of verbal pictorialism. For his part Kenner identifies such pictorialism with post–Symbolist lyric poets of the 1890s rather than with the Symbolists themselves (186), but he insists no less emphatically than the other two critics on the fundamentally non–visual, non–mimetic character of the Poundian Image. Referring to “The Return,” published in the 1912 volume Ripostes, Kenner observes that none of the poem's imagery is visual or sculptural: “It is wholly linguistic” (191). Here are the first three stanzas:

See, they return; ah, see the tentative
        Movements, and the slow feet,
        The trouble in the pace and the uncertain
        Wavering!
[→page 134] See, they return, one, and by one,
With fear, as half–awakened;
As if the snow should hesitate
And murmur in the wind,
        and half turn back;
These were the “Wing'd–with–Awe,”
        Inviolable.

Gods of the wingèd shoe!
With them the silver hounds,
        sniffing the trace of air! (Personæ 69–70)

Whether or not one agrees wholly with Kenner's contention that the poem's rhythm, as opposed to its imagery, defines its meaning, it is certainly the case that the patterns of repetition with slight changes and the absence of an overall syntax encompassing the individual syntactical units give the poem's purely linguistic dimension a greater prominence than would be expected if its significance were supposed to reside primarily in the meanings of its images. This is not to say that “The Return” is devoid of ideas, for indeed the first stanza enacts rhythmically an idea that can easily be paraphrased in prose: “The gods, returning now, do so in unstable meters” (Kenner 190). But the poem's subject is the writing of poetry itself, not the return of the classical gods. As in the poem “In a Station of the Metro,” our attention is directed less to the external referents of the verbal images employed by the poem than to their juxtaposition within the poem. “Words,” Kenner explains, “without loss of precision, have ceased to specify in the manner of words that deliver one by one those concepts we call 'meanings'” (187).

IV.

In the Imagist poem, therefore, outwardly directed referentiality, though not absent, is subordinated to an inwardly oriented sequentiality. Paratactic indeterminacy becomes the self–authenticating expression of a shaping consciousness that, recognizing its own contingency, has renounced the metaphysics of ontological participation [→page 135] on which the Romantic concept of the symbol had been founded. As Pound's biographer A. D. Moody observes, the hawk that is “always a hawk and not any other thing” in an Imagist poem is already “a subjective mental object” (225), no longer a bird in the sky but a bird in the mind. Thus when the poet presents the faces seen in the metro station as “petals on a wet, black bough,” both the faces and the petals are in the mind, and “under the pressure of attention they become identified the one with the other in a further apprehension charged with unexpected significance” (Moody 225). The ontological relation of the face in the mind to the face in the underground or of the petal in the mind to the petal on the bough is irrelevant.

In contrast, such connections between images and objects are the very subject of Baudelaire's “Correspondances” (1857), which I adduce because it is less a Symbolist poem itself than a versified statement of a Romantic natural philosophy that attributes to discrete phenomena resonances of a unifying order by which all are related to one another.5

In this sonnet the imagined interactions of things—smells, colors, and sounds—are compared to echoes heard from afar and merging in an obscure and profound unity: “Comme de long échos qui de loin se confondent / Dans une ténébreuse et profonde unité” (Œuvres 1: 11). While assuming one kind of correspondence, between word and object, the poem enunciates another, between physical phenomenon and metaphysical order. The two kinds are inversely related, the semiotic functionality of the language becoming invisible in proportion as the symbolic functionality of the natural objects evoked becomes credible. When, in the first quatrain, Baudelaire tells us that nature is a temple in which living pillars sometimes release mysterious words, and that we pass there through forests of symbols that watch us with knowing looks, his words are supposed to direct our attention beyond themselves to the phenomena they interpret:

La Nature est un temple où vivant piliers
Laissent parfois sortir de confuses paroles;
L'homme y passe à travers des forêts de symboles
Qui l'observent avec des regards familiers. (1: 11)

[→page 136] Still, a nagging question remains: how would we know that we were walking through forests of symbols if Baudelaire didn't warn us? How would we know otherwise that amber, musk, benjamin, and incense, having the expansion of infinite things (Ayant l'expansion des choses infinies), chant the ecstasies of the mind and senses (chantent les transports de l'esprit et des sens)? The poem's deictic purpose undermines its very claim for the symbolic action of nature.

To summarize the argument thus far: the Romantic symbol and the Poundian image differ profoundly from each other inasmuch as the one is conceived as the objective manifestation of a harmoniously organized totality encompassing the individual subject, and the other as the subjective appropriation of a mentally represented external object. “Images in verse,” T. E. Hulme averred, “are […] the very essence of an intuitive language” (70). And Pound concurred: “An image […] is real because we know it directly” (“Vorticism” 283). Intuitive, non–discursive apprehension of that kind is a fantasy, however, and moreover the very fantasy—as Kermode recognized—that links Pound, in spite of himself, to the Romantic theorists of the symbol. Behind that fantasy, I suggest, is the hermeneutic anxiety signalled in my title: a preoccupation with the tendency to error and imprecision inherent in any self–governing semiotic system. Hence the appeal to an authority, whether the order of nature or the individual mind, in which the communicative act could be claimed, although paradoxically, to be wholly contained, and the need, indeed the possibility, of interpretation thereby eliminated. The epistemological reliability of semiosis was to be assured no longer by the grounding of the process in an extrasemiotic reality—as, for Dante, the triune God is, enfolding all signs into himself like the leaves of a book being closed (cf. Paradiso 33.85–87)but rather by its self–referentiality, even if that self–referentiality was defined in metaphysical terms that are themselves open to question. No interpretive slippage would occur in the semiotic vacuum of self–referentiality, or more precisely, of self–identical referentiality.

[→page 137] The philosopher Hans Blumenberg once observed that a weakness of subjective idealism is that, although it can conceive the world as the self–objectification of the subject, it cannot guarantee that the world thus objectified will conform to the subject's wishes (see 298). A comparable problem attended both the Romantic symbolists and Pound in that the intuitive, non–discursive communicative acts they posited needed to be communicated discursively to others if they were to be known at all—if, in the case of the Image, a radiant node of rushing ideas was to be distinguished from a mere opacity of meaning. Just as the forest of symbols could not speak for itself, neither could the petals on the wet, black bough. The Imagist poem therefore required its own theoretical, discursive correlative, the Imagist manifesto, although that cannot have been what Pound had in mind when, sounding strangely like Wordsworth in the Preface to Lyrical Ballads (1800), he spoke of the need to “bring poetry up to the level of prose” (“Vorticism” 280).

University of Oxford

Notes

1. Translations throughout are mine.
Back to Main Text

2. Pound first published “Vorticism” in the Fortnightly Review in September 1914, then reprinted it as chapter 11 of Gaudier–Brezeska: A Memoir (1916). Whitworth offers a brief contextualization of the critical positions of Kermode and Perloff with long extracts from both critics.
Back to Main Text

3. This is Yeats's stanza as quoted by Pound:

Now as at all times I can see in the mind's eye,
In their stiff, painted clothes, the pale unsatisfied ones
Appear and disappear in the blue depth of the sky
With all their ancient faces like rain–beaten stones,
And all their helms of silver hovering side by side. (“The Later Yeats” 380)

Pound omits the final three lines: “And all their eyes still fixed, hoping to find once more, / Being by Calvary's turbulence unsatisfied, / The uncontrollable mystery on the bestial floor” (Variorum Edition 318). On Pound's misleading use of the term objective in his contrast between Yeats and Hueffer (“Status Rerum” 125–26), see Levenson, Genealogy 150.
Back to Main Text

[→page 138] 4. Mallarmé himself had written: “To name an object is to suppress three–quarters of the pleasure of a poem […] to suggest it, that's the dream. This is the perfect use of the mystery that constitutes the symbol: to evoke little by little an object in order to display a state of the soul” (869).
Back to Main Text

5. With regard to Baudelaire's English reception as a symboliste poet, it is worth noting that Symons first included a chapter on him, and a very brief one at that, only in the third edition of The Symbolist Movement in Literature (1919). (Baudelaire had died nineteen years before Jean Moréas published the manifesto of the Symbolist mouvement in Le Figaro.) To be sure, Hugo Friedrich argues that Baudelaire's relation to Romantic Naturphilosophie is itself problematic in that his metaphysics is an “empty ideality”: because we never learn from Baudelaire exactly what the “henceforth understood language of flowers and things” is (Friedrich refers to the poems “Elévation” and “Correspondances”), we can “fill his words with no other content than that of absolute mysteriousness [Geheimnishaftigkeit] itself” (48–49). With the substitution of the word meaningfulness for mysteriousness, however, Friedrich's conclusion about Baudelaire would apply perfectly to Romantic descriptions of the symbol.
Back to Main Text

Works Cited

Alighieri, Dante. Porgatory and Paradise. Trans. Henry Francis Cary. Chicago : Thompson & Thomas, 1901.

Baudelaire, Charles. Œuvres complètes. Ed. Claude Pichois. 2 vols. Paris: Gallimard, 1975.

Blumenberg, Hans. Arbeit am Mythos. 1979. Frankfurt a.M.: Suhrkamp, 1984.

Davie, Donald. Ezra Pound. London: Fontana, 1975.

Friedrich, Hugo. Die Struktur der modernen Lyrik: Von der Mitte des neunzehnten bis zur Mitte des zwanzigsten Jahrhunderts. 1956. Hamburg: Rowohlt, 1973.

Gadamer, Hans–Georg. Wahrheit und Methode: Grundzüge einer philosophischen Hermeneutik. 6th ed. Tübingen: Mohr, 1990.

Halmi, Nicholas. “Coleridge on Allegory and Symbol.” The Oxford Handbook of Samuel Taylor Coleridge. Ed. Frederick Burwick. Oxford: OUP, 2009. 345–58.

———. The Genealogy of the Romantic Symbol. Oxford: OUP, 2007.

Hulme, T. E. “Romanticism and Classicism.” The Collected Writings of T. E. Hulme. Ed. Karen Csengeri. Oxford: Clarendon P, 1994. 59–73.

Kenner, Hugh. The Pound Era: The Age of Ezra Pound, T. S. Eliot, James Joyce and Wyndham Lewis. 1972. London: Faber & Faber, 1975.

Kermode, Frank. Romantic Image. London: Routledge, 1957.

Levenson, Michael. A Genealogy of Modernism: A Study of English Literary Doctrine 1908–1922. Cambridge: CUP, 1984.

———. Modernism. New Haven: Yale UP, 2011.

Luhmann, Niklas. Die Kunst der Gesellschaft. Frankfurt a.M.: Suhrkamp, 1995.

[→page 139] Mallarmé, Stéphane. “Réponses à des enquêtes sur l'évolution littéraire.” Œuvres complètes. Ed. Henri Mondor and G. Jean–Aubry. Paris: Gallimard, 1945. 866–83.

Moody, A. David. Ezra Pound: Poet. Vol. 1. Oxford: OUP, 2007.

Perloff, Marjorie. The Poetics of Indeterminacy: Rimbaud to Cage. 1981. Evanston: Northwestern UP, 1999.

Pound, Ezra. Early Writings: Poems and Prose. Ed. Ira Nadel. London: Penguin, 2005.

———. “The Later Yeats.” The Literary Essays of Ezra Pound. Ed. T. S. Eliot. London: Faber & Faber, 1954. 378–81.

———. Personæ: Collected Shorter Poems. Ed. Lea Beachler and A. Walton Litz. 1990. London: Faber & Faber, 2001.

———. “A Retrospect.” Early Writings: Poems and Prose. Ed. Ira Nadel. London: Penguin, 2005. 252–65.

———. “Status Rerum.” Poetry 1 (1912/13): 123–27.

———. “Vorticism.” Early Writings: Poems and Prose. Ed. Ira Nadel. London: Penguin, 2005. 278–91.

Schelling, F. W. J. Philosophie der Kunst. Sämmtliche Werke. Ed. K. F. A. Schelling. 14 vols. Vol. 5. Stuttgart: Cotta, 1859. 355–736.

Schneidau, Herbert. Ezra Pound: The Image and the Real. Baton Rouge: Louisiana State UP, 1969.

Shelley, Percy Bysshe. “A Defence of Poetry.” Shelley's Poetry and Prose. Ed. Donald Reiman and Neil Fraistat. 2nd ed. New York: Norton, 2001. 509–38.

Symons, Arthur. The Symbolist Movement in Literature. Ed. Richard Ellmann. New York: Dutton, 1958.

Tiffany, Daniel. Radio Corpse: Imagism and the Cryptaesthetic of Ezra Pound. Cambridge, MA: Harvard UP, 1995.

Whitworth, Michael. “Modernism and Romanticism.” Modernism. Ed. Michael Whitworth. Oxford: Blackwell, 2007. 63–98.

Yeats, W. B. The Variorum Edition of the Poems of W. B. Yeats. Ed. Peter Allt and R. K. Alspach. New York: Macmillan, 1957.

Readers are invited to submit their responses to the editors at editors@connotations.de.

Back to Top