Introduction: modernism's new literalism
“As we move into the twenty-first century,” observes Marjorie Perloff in a recent book, “the modern/postmodern divide has emerged as more apparent than real.”1 Coming not only from a distinguished critic, but also the foremost academic champion of an avant-garde that – whatever disagreements its individual members have about their place in postmodernism – has defined itself against modernism, this observation is a striking one. After all, the divide once seemed crucial to many literary historians, including Perloff herself. Why now does it seem irrelevant, or perhaps more to the point, why did it use to seem so fundamental? What was the crucial difference between modernism and postmodernism? That is, what is the difference between, say, T. S. Eliot or Wallace Stevens and the poets most often identified with postmodernism, particularly those affiliated with the language movement in American poetry (Charles Bernstein, Lyn Hejinian, Ron Silliman, Steve McCaffery, Barrett Watten, Bob Perelman, Susan Howe, Michael Palmer, to name a few)?2 Certainly by customary definitions, the difference would seem incontrovertible. Where, for example, the modernism of Eliot has been identified with the autonomy of the text (or what postmodernism calls the “closed” text) and the determinacy of its meaning, the postmodern text is “open” and its meaning is indeterminate. And where the participation of the reader was thought to be irrelevant to the text in modernism, it has become not just relevant but crucial to the text in postmodernism. But if the divide appears obvious in the context of these stark oppositions, Perloff has strong reasons for denying it.
For even when postmodern poetry was most committed to describing itself as a repudiation of modernism, it was also insisting on a continuity between its values and those of a certain subset of modernist writers. Laura (Riding) Jackson, Louis Zukofsky, and above all Gertrude Stein are invoked with almost ritual frequency as modernist practitioners of a thoroughly postmodern aesthetics.3 But this subset of postmodern modernists has proliferated to the point that now (as we will see in Perloff's own analysis) even Eliot has come to seem increasingly connected to the values of postmodernism – i.e., to the open text, to the solicitation of the reader's participation, and to the indeterminacy of meaning. As the modernist poets to whom postmodernism was once most opposed turn out today to be its most sympathetic precursors, the differences between them do indeed become “more apparent than real,” and what was announced as a break with the modernist tradition looks instead like its perpetuation.
The argument of this book, however, is that those differences, far from being merely apparent, are real, and that the modern/postmodern divide remains intact, both historically and theoretically.4 I am arguing first that the literary history that eliminated the divide is mistaken, which is to say that Stein and (Riding) Jackson (if not Zukofsky) are not committed to the open text and the values of indeterminacy; and second, that the theoretical difference between a literature committed to the text's dependence on readerly participation, and a literature not so committed – a literature committed instead to the irrelevance of the reader and to the absolute autonomy of what Stein calls the work that “exists in and for itself” – is fundamental.This project is thus at once both a literary-historical and a theoretical argument: it is an attempt to alter the currently received history of twentieth-century American poetry by showing that Stein and (Riding) Jackson have been and continue to be misunderstood as postmodernists avant la lettre. And it is meant to show that this historical misunderstanding is itself a function of a more pervasive theoretical effort – beginning for my purposes with the early New Criticism of the 1920s and continuing through the work of critics like Perloff herself – to displace what, in its broadest terms, we might call the “meaning” of a text by the reader's experience of it, a displacement Perloff calls “literalism.”5
Perloff announces the growing inconsequentiality of the modern/postmodern divide in a book called 21st-Century Modernism: The “New” Poetics. In putting scare quotes around “new,” she means to suggest that the poetics in question, far from being new, can be traced at least as far back as the earliest works of Eliot:
InThe Poetics of Indeterminacy (1981) I drew a sharp distinction between Eliot's symbolist mode and the more “literalist” indeterminacy of John Ashbery. Twenty years later, in the context of recent poetic developments, I would qualify my earlier reading by noting that the comparison was to the later Eliot, not the poet, then largely unknown, made familiar by Christopher Ricks's superb edition of the hitherto unpublished poems written between 1909 and 1917. (7–8)
Perloff goes on to explain how the Eliot whom she formerly saw as the antithesis of Ashbery's (and for that matter, the language poets') “poetics of indeterminacy” has earned this limited admission to the “New” Poetics. She cites J. C. Mays's reading of “The Love Song of J. Alfred Prufrock,” where “images [tend] to balloon away from their referents and assume an uncontrollable life of their own” (25–26), a reading that, Perloff discovers, resembles her own reading of Stein as “stress[ing] composition rather than representation, the play of signifiers rather than the pointing relation of signifier to signified” (54). But if terms like “uncontrollable” and “play” give us a “poetics of indeterminacy” in Stein and now Eliot, what makes their indeterminacy “literalist”? Perloff never explicitly defines the term, but what she means by it is perfectly clear when she says that Prufrock approximates “Constructivist notions of ‘laying bare the device,’ of using material form – in this case, language – as an active compositional agent, impelling the reader to participate in the process of construction.”6 Certainly any focus on the “material form” of the text – which in this context refers to its physical appearance on the page, the sounds of its syllables – cannot help but “impel the reader to participate” (25–26) if only because our eyes, ears, etc. are required to read or listen to it. But what is distinctive about literalism in this context is that the materiality of the text is also understood to produce its indeterminacy. Every text is material, but the literalist text understands its materiality as an invitation to its reader, and hence as the condition that makes every reading both different from and equal to every other in constituting the text.
Let us take a comparatively early example of a text from the language movement that does everything Perloff understands as “literalism”: Lyn Hejinian's Writing Is an Aid to Memory (1978), whose most visible formal feature is the ragged positioning of its lines in relation to the left margin. This raggedness is not random, however; it works according to strict principle, for each line is placed where the first letter of its first word would occur in an alphabet typed across the page in order from ‘a’ to ‘z’:
abcdefghijklmnopqrstuvwxyzhow ness posites
autobiography sees the world
by early beginning with the top narrate
much intention is retrospective and much
extension is prospective
for illusion of men can be so many scandals shifting quite but admirable victed
sequences. . .7
We can easily see what might count as “literalist” about this text, in the sense that here and throughout the poem its most literal constituents (the letters forming the words, rather than the meaning of the words) are its organizing principle. And the arrangement of those letters confronts us as well with the most literal mechanical conditions of their production (the typewriter keys striking, the carriage advancing). In a gesture that echoes Perloff's definition of literalism, Hejinian remarks in her 1983 essay “The Rejection of Closure,” that such formal devices as these not only “foreground process,” but also “serve to ‘open’ a poetic text” by “invit[ing] participation” in that process.8
Certainly the amputated suffixes and roots of words (e.g., “ness,” “posites,” “victed”) that are the linguistic hallmark of Writing Is an Aid to Memory invite the reader to entertain multiple possibilities: is it “evicted” or “convicted”? “Apposites,” “opposites,” or an alteration of “posits”? And what about the (necessarily “prospective”) “extension” of options for “ness,” which are as many as the adjectives we can bring to mind? But Hejinian imagines her readers as more than participants in the composition of the text; she imagines them as agents of its composition. Thus, the “open text,” she says, foregrounds not just “the process of the original composition” but also that of “subsequent compositions by readers,” becoming, in other words, not one composition but many (The Language of Inquiry, 43). Where the “closed text” is imagined to have a meaning that exists independent of the interpretations of its readers and therefore remains unaffected by them, the open text is reconstituted every time it is read. And because it is reconstituted every time it is read, there is no prior meaning to be discovered through interpretation. Rather, insofar as every encounter between the reader and the poem becomes a new composition, every new reader becomes a writer of the poem, so that the relation of the reader to the “open text” is no longer to understand what it means, but to become, again quite literally, who its author is. (And as we shall see later in this discussion, authorship under these conditions is no longer understood as producing a meaning or meaning to produce, but as literally causing an effect. Indeed, strictly speaking, we might say that the open text never really has any meaning and is thus never interpreted at all.)
If the hallmark of literalism is a text's ability to compel our attention to its physical features, and more generally, to make us think of language in terms of its material constituents – letters and phonemes overwhelming words and sentences – Perloff's chapter on the Russian Constructivist poet Velimir Khlebnikov seems to present us with an uncontroversial example of literalism. But her account of the literalist Khlebnikov also makes her account of the alleged literalist Eliot all the more controversial. For in its effort to foreground the material constituents of language, what Khlebnikov's “zaum” poetry of the teens and 1920s supposedly shares with the language poetry of the 1980s and 1990s, Perloff argues, is the desire to dissociate language from understanding: “But, from the perspective of contemporary poets like Susan Howe and Bruce Andrews, what is more interesting than phonemic repetition as such is Khlebnikov's own sense of how phonemic and morphemic play can produce a poetic language beyond (za) mind or reason (um) – what Khlebnikov and his fellow-poet Kruchonykh called zaum” (21st-Century Modernism, 123). We have already noted, in the example of Hejinian, how the participation invited by her “phonemic and morphemic play” is supposed to make the reader the producer of the text rather than the discoverer of its meaning. But if we can see the uncoupling of language from reason in that example, it's hard to see what makes this literalization of language count as going “beyond” reason. According to Perloff, “Khlebnikov's stress on the materiality of the signifier, the graphic and phonic characteristics of language” embodies the cause of “resistance to an Establishment ‘poetry’.” Once Khlebnikov's cause also becomes the “cause of Eliot or of Stein” as well as of “Concrete Poetry” (which was even more uncontroversially literal in its commitment to “material form” than Khlebnikov's zaum poetry), literalism seems not to involve pushing language beyond reason, but never to let it get there in the first place (128). In identifying zaum poetry with Concrete Poetry, Perloff suggests that both make the text “concrete” by making the reading of it consist of experiencing its form (registering the shape of the letters, words and lines) rather than interpreting its meaning.9 Or to turn this around, the concrete or zaum poem seems to make the reader into someone who experiences the poem rather than understands it by making the poem become an object rather than a text. In this respect poetic literalism – the transformation of readers into experiencing subjects and of texts into concrete objects – has an important analog in the history of art in the last century, where painting and sculpture undergo a similar transformation.
Indeed, the term “literalism,” used as a way of talking about how art becomes an object (or rather, never ceases being one), finds its first currency not in Perloff or in any of her twenty-first-century modernist texts (including the twentieth-century ones), but in “Art and Objecthood,” Michael Fried's 1967 essay on the emerging movement in painting and sculpture that is most often identified as “minimalism.”10 While Perloff does not cite Fried as a source for the term, her understanding of literalist poetry corresponds quite precisely to Fried's understanding of literalist “art.” In works like Tony Smith's six-foot cube entitled Die (1962) or Robert Morris's Untitled (Ring with Light) (1965–66), every material aspect of the work, including not just the visual and tactile (and even aural in some other examples) form of the object itself, but also of the environment in which it is beheld, is relevant to its status as object:
There is nothing within [the beholder's] field of vision – nothing that he takes note of in any way – that declares its irrelevance to the situation, and therefore to the experience, in question. On the contrary, for something to be perceived at all is for it to be perceived as part of that situation. Everything counts – not as part of the object – but as part of the situation in which objecthood is established and on which that objecthood at least partly depends. (Fried, “Art and Objecthood,” 155)
If, in the example of Hejinian's open text, the relevance of everything about the reader's experience makes her the author of the text in question, at the moment when everything about the object and its situation becomes relevant to the beholder's experience, the beholder's experience itself comes to constitute the object. Indeed, the beholder's experience is, as Fried explains, the “everything” on which the object's very “objecthood” depends. And so, inasmuch as “the experience of literalist art is of an object in a situation” it is “one that, virtually by definition, includes the beholder” (153).
When Fried refers to this relation in terms of the “special complicity that the work extorts from the beholder” (155) – i.e., her participation – the literalist object begins to look exactly like language poetry's “open text.” Moreover, this solicitation of the beholder translates into an aesthetics of indeterminacy in Fried's account of literalism, just as it yields a poetics of indeterminacy in Perloff's. For despite (or rather, entirely because of) the obdurate materiality of the object, the possibilities it affords are infinitely expansive, as many and varied as the beholders who might approach it: “The beholder knows himself to stand in an indeterminate, open-ended – and unexacting – relation as subject to the impassive object on the wall or floor” (155). But while Fried demonstrates the same set of concerns that Perloff does around the treatment of the work of art as an object of experience, and thereby, as something that exists necessarily in relation to a subject, “Art and Objecthood” is an argument against literalism for its repudiation of modernism, whereas 21st-Century Modernism celebrates literalism for its successful embrace of modernism. “Ours may well be the moment,” Perloff writes in the last line of the book, “when the lessons of early modernism are finally being learned” (200). Learning “the lessons of early modernism,” however, by which Perloff means learning the lessons of the writers who count as modernism's true avant-garde, also means unlearning the lessons of what she takes to be the critical legacy of mainstream modernism: “Of course, ‘Prufrock’ was. . .to become a celebrated modern poem, but the New Critical classic. . .is not ours” (27). If, in other words, the lessons attributed to Stein, Khlebnikov, and now the avant-garde Eliot are those of the “open text,” the ones that need to be unlearned are those of the New Criticism, with its notorious commitment to the autonomous (in Hejinian's terms, “closed”) text.
In this respect, literalism in poetry does seem to follow the same course of resistance as literalism in art, for according to Fried, what literalism rejects in modernist painting is precisely its autonomy, the idea that “what is to be had from the work is located strictly within it” (“Art and Objecthood,” 153). Whereas the modernist work of art not only makes no claims on the beholder, but “finds intolerable” the very idea of any relation to an audience, “literalist art,” writes Fried,
addresses itself to the beholder alone. Someone has merely to enter the room in which a literalist work has been placed to become that beholder, that audience of one – almost as though the work in question has been waiting for him. And inasmuch as literalist work depends on the beholder, is incomplete without him, it has been waiting for him. And once he is in the room the work refuses, obstinately, to let him alone.(163)
Like the literalist object that awaits the beholder, the “open text” awaits its reader, and both are “incomplete” alone. But resisting the autonomy of the text by making the text dependent on the reader's experience of it becomes problematic when language proponents like Perloff or Hejinian turn to Stein as their mascot. For if literalism refuses the autonomy of the work of art by calling upon the beholder (or reader) to participate in its situation – indeed, to create its situation – Stein, by contrast, insists on the autonomy of the work of art precisely by refusing any relation whatsoever between the work and anyone who might experience it, including the author herself. In short, Stein refuses literalism.
Stein begins a 1936 lecture called “What Are Master-pieces and Why Are There so Few of Them” by commenting on the fact that she finds herself before an audience, a situation, she argues, that is antithetical to the creation of masterpieces:
One of the things that I discovered in lecturing was that gradually one ceased to hear what one said one heard what the audience hears one say, that is the reason that oratory is practically never a master-piece . . .It is very interesting that letter writing has the same difficulty, the letter writes what the other person is to hear and so entity does not exist there are two present instead of one and so once again creation breaks down. I once wrote in writing The Making of Americans I write for myself and strangers but that was merely a literary formalism for if I did write for myself and strangers if I did I would not really be writing because already then identity would take the place of entity.11
Stein explains what she means by identity with the example of her relation to her dog: “I am I because my little dog knows me but, creatively speaking the little dog knowing that you are you and your recognising that he knows, that is what destroys creation” (Writings, 355). What matters here is not so much the latter recognition (although it too is crucial to her logic), but the dog's, since it will turn out that all kinds of things (especially things that are not masterpieces), none of which possesses its own faculties of recognition, can be functions of identity just as persons can. Insofar as the recognition in which identity consists arises out of a relation between an object and a subject who may as well be a dog, the relation is one of pure memory: all that is required to produce the dog's recognition, and in turn your identity, is its having been in your presence. Moreover, the object, whether it be the dog's mistress, the literalist work of art, or the “open text,” only achieves its identity – which produces what Fried would call its objecthood – out of the situation in which it is experienced. And even though such an object, as Fried explains, “must remain the center or focus of the situation,” nevertheless “the situation itself belongs to the beholder.” The dog's experience of having seen you before can only belong to the dog. Thus the object of identity – always the object of a subject's experience – can never be an entity because it can never, as Stein puts it, “exist in and for itself”; it can only exist for someone (Writings, 357). Indeed, the whole point for Stein of insisting that the masterpiece is an entity is to insist that it cannot be an object. And the whole point of insisting that it cannot be an object is to insist that what it is can never be a function of anyone's experience of it – or, to put this slightly differently, that what it is can never be a function of what it is for someone.
For Stein, then, a masterpiece can never be an “open text” because it can never “invite participation.” This is not to say, of course, that readers do not or cannot have responses to or experiences of a work of art (in a trivial sense we can't help but do so); only that their responses and experiences have nothing to do with what makes it art. This is nothing if not a commitment to the autonomy of the work of art, and in fact, it's a commitment to one of the most important, if only intermittently influential New Critical arguments for that autonomy, William W. Wimsatt and Monroe C. Beardsley's “The Affective Fallacy,” which contends that specific readerly responses – particularly emotional ones like happiness or sadness – are not only not required to grasp the meaning of a poem but are in fact altogether irrelevant to that meaning. But the reason these New Critical doctrines – the autonomy of the work of art and the affective fallacy – have been only intermittently influential is that they have always been at odds with two other, equally foundational, ones: namely, that the poem must not mean but be, and that paraphrase is heresy.12
The logic whereby the heresy of paraphrase entails the requirement that the poem must not mean but be, and further, the logic whereby both render impossible the kind of autonomy that Stein (and Fried) imagine for art, find concise expression in the aptly titled chapter called “The Poetic Experience” in I. A. Richards's Science and Poetry (1926). The famous phrase “heresy of paraphrase” occurs much later in the work of Cleanth Brooks, but the critical principle behind those words – that the best interpretation of a poem is the poem itself – is already in place when Richards urges that the best way to grasp the reasons for “thinking [poetry] valuable” is to “begin by reading slowly, preferably aloud, giving every syllable time to make its full effect upon us.”13 Because our focus in such an exercise is not, according to Richards, the sentences, or even exactly the words, of the poem, but the separate syllables of the words; and further, because what we are after is “the sound of the words ‘in the mind's ear’ and the feel of the words imaginarily spoken,” this “reading” of the poem (which is above all a repetition of the poem) produces not an account of the meaning of the poem, but an experience of what Fried would call its objecthood (Richards, Poetries and Sciences, 23). As we will see in chapter 3, Richards occupies a somewhat anomalous position in the New Criticism because in elaborating this claim for the sensory effects of the poem he explicitly embraces the readerly “affect” that Wimsatt and Beardsley (and for that matter, William Empson, John Crowe Ransom, Cleanth Brooks, and a host of others) reject as “fallacy.”14 By the latter critics' reasoning, Richards has already given up the grounds on which the poem can count as autonomous. But by Fried's (and, I am arguing, Stein's) reasoning, the others also give up the autonomy of the text when they commit themselves to locating the “meaning” of the text in what Ransom calls its “objective features” and what Brooks calls its “formal” features.15 Indeed, by treating the objecthood of the text as if it were equivalent to the meaning of the text, the New Critical commitment to the heresy of paraphrase cannot help but entail a commitment to the affective fallacy – if the meaning of a text is reducible to the text's objecthood, it can only consist of the reader's affect.16
Thus, when Richards says that “it is never what a poem says that matters but what it is” the moment when the poem has to “be” rather than “say” is also the moment when it becomes an object rather than, in Stein's terms, an “entity” or in Fried's terms, “art” (Richards, Poetries and Sciences, 33). For the moment when the text becomes an object is precisely the moment when it can no longer be autonomous, since everything that constitutes the text's objecthood – the “sound” and “feel” of its constitutive syllables – belongs entirely to the experience of someone – just what Stein insists it cannot do and still be a masterpiece. The New Critical poem becomes, in other words, the very kind of literalist text that Perloff says “foregrounds the material form of language” and “impels our participation in its construction” (21st-Century Modernism, 25–26).
In this context, it should hardly be surprising that someone like Perloff is fond of quoting Charles Bernstein's statement that “the poem said any other way is not the poem,” itself a paraphrase of Brooks's “heresy of paraphrase” doctrine (cited in 21st-Century Modernism, 12).17 Yet neither she nor Bernstein nor anyone else currently subscribing to that claim recognizes its patent repetition of the theoretical commitments of the New Criticism, and the recognition never takes place because the proponents of language poetry rightly understand themselves as committed not to the autonomy of the poem but to its objecthood. While the literalism celebrated by Perloff and by language poetry more generally appears to have corrected one New Critical mistake – that of equating objecthood with autonomy – it has simply reinstated the more foundational one – that of equating experience with interpretation. Perloff's concluding statement in 21st-Century Modernism, – that “the lessons of modernism are finally being learned” (200) is, therefore, in some sense right, if we take the modernism whose lessons are being learned to