On Materialities, Meanings, and The Shape of Things

On Materialities, Meanings, and The Shape of Things

The Shape of the Signifier: 1967 to the End of History
Walter Benn Michaels
Princeton and Oxford: Princeton University Press, 224 pp., $24.95

Lori Emerson reviews The Shape of the Signifier by Walter Benn Michaels.

Ben Underwood:

A recent essay by Michaels on the films of Christopher Nolan has appeared in an electropoetics collection assembled by Lori Emerson.


Walter Benn Michaels’ latest book stands in relation to an ever-increasing number of post-theory or post-critique books. At their most ambitious, these works seek to rehabilitate the left in the wake of the impossibility of political action under postmodernism; failing that, such post-theory books simply try to find a way to make judgments without drawing on norms, universal principles, or, as Jean-François Lyotard famously puts it, meta-narratives. In fact, the most recent addition - Terry Eagleton’s After Theory (2003), a book which supposedly deals a death-blow to the field of cultural studies - is by the very same author who brought us Literary Theory. Although deliciously shallow in its nasty assessment of the new breed of student who, he claims, is obsessively and uncritically interested in sex, the body, and Friends, Eagleton does touch on one of Michaels’ primary preoccupations: that if, according to high theorists such as Foucault and Derrida, we live in a world of pure difference, and if we therefore must make subjectivity, identity, and culture primary, then we are left without a way to account for, on the one hand, class and poverty and, on the other hand, meaning and interpretation. While Michaels’ account is significantly more sophisticated, the main difference between him and Eagleton, of course, is that Michaels has been against theory and so has been writing post-theory since the height of theory in the 1980s.

The Shape of the Signifier: 1967 to the End of History, then, combines his defense of intentionality in Against Theory (1985) with his critique of identity in Our America (1997); in other words, Michaels delineates a general cultural movement from questions about the ontology of the text to an insistence on the primacy of the subject (10). The argument, briefly stated, is that the simultaneous commitment to materiality and identitarianism is both inconsistent and problematic. As he states in a 2002 interview that appeared in The Minnesota Review, such a commitment is real and is bad - strikingly real and bad in that, following Michaels’ way of thinking, if you hold, say, Susan Howe’s views on the importance of the physicality of the text, you must hold that the subject, the identity of the subject, is crucial for registering the physical aspects of the text; therefore, by this logic, you must also hold George Bush’s views on terrorism as a war not of (political) beliefs but of identities, and since terrorists follow what’s being called a perverted form of Islam, theirs is an identity we do not even have to acknowledge. It’s precisely Michaels’ ability to put forth such well-oiled, even seductive, inferential arguments that makes his book fascinating, intellectually and personally challenging, yet also slyly problematic. But before I can speak to the difficulties I have with The Shape of the Signifier, some recounting of its key assertions is necessary.

Recalling the 1967 publication of Michael Fried’s Art and Objecthood which marked for many the end of modernism, Michaels frames this year as an end-point after which it became essential to pose the question of what kinds of subjects are entailed by what kinds of objects (13). And, using Howe’s work on Thomas Shepard and Emily Dickinson followed by Paul de Man’s thinking in Aesthetic Ideology as a template, Michaels outlines what in particular is at stake for subjects and texts which are based on a concern with the materiality of the signifier. One of Michaels’ critical innovations is to demonstrate that the prevailing concern with materiality, while commonly associated with visual, concrete and language poetry, is also exemplified in postmodernist writing ranging from science fiction by Octavia Butler and Kim Stanley Robinson, fiction by Richard Powers and Kathy Acker, to Conceptual art by Robert Smithson.

But Michaels’ reading of Howe en route to his reading of de Man is, I think, particularly revealing of the shape of his argument in The Shape of the Signifier. He begins his book by turning to Howe’s The Birth-Mark (1993) and her concern with editorial control and authorial intention vis-à-vis the often overlooked physical aspects of the text: the eighty-six blank pages in the manuscript of Shepard’s Autobiography and the smallest physical details of the page (142) in Dickinson’s fascicles. For Howe the problem is, first and foremost, the imposition of editorial control that limits both authorial intention and meaning; but the problem is also that of discerning the accidental from the intentional as well as - even more radically - how to preserve the text when an editor cannot possibly discern the accidental, the purely random [c]ancelations, variants, insertions, erasures, marginal notes, stray marks and blanks, (9) from the intentional. In other words, a lurking skepticism (or the fact that we may never know whether Shepard ‘meant’ for there to be eighty-six blank pages separating the two texts) as well as a dedication to the belief that poetry is a physical act (Interview 157) keeps Howe astutely between a defense of Shepard and Dickinson’s intentions and a defense of what may be purely random, purely meaningless stray marks; it also results in her assertion that the only acceptable edited version of a writer’s work is a facsimile.

Although Michaels is clearly a fan of Howe’s work, the same dedication to text and author that drives Howe also drives him to point out that once a text such as Shepard’s or Dickinson’s becomes a material object that must be preserved and once all of its physical features are equally important, the work not only ceases to be a text that can be edited but it also ceases to be a text. Moreover, once the text ceases to be a text, the author’s intentions become beside-the-point:

Indeed, despite the fact that our interest in the text’s materiality was provoked first by an interest in Dickinson’s intention, we can no longer have any principled interest in Dickinson at all … Thus the most radical form of Howe’s commitment to Dickinson produces a certain indifference to Dickinson - for the things that Dickinson didn’t care about (say, the kind of ink) must matter just as much as the things she did care about (say, the shapes of the letters). (Michaels The Shape)

Although Michaels never directly says as much, clearly, taken to its logical conclusion, a strict adherence to materiality simply does not reflect the way we actually think about texts and readers (and he doesn’t directly state his argument since, on the one hand, he assumes that readers are familiar with his earlier defense of intentionality and argument against identitarianism and since, on the other, the force of his argument comes from unpacking the deeper implications of not appealing to authorial intention in favor of the reader’s subjective experience [call it ‘reading’] of the text). For what careful readers of Dickinson, for example, would say they are not reading a text? What readers would claim that they do not think there is such a thing as a text? Further, given such a strict materialist stance, there is the issue of meaning - for if a defense of the purely random mark for the sake of treating the text as an object is an effacement of intentionality, it is also a turn away from meaning (at least as Michaels understands it) for the sake of what Paul de Man calls the text’s sensory appearance. But again, we need to ask: what reader would say that there is no meaning in the lines My Life had stood - a Loaded Gun - outside of the appearance of the lines in the poem-as-object? Another way of putting what concerns Michaels here might be that if every single feature of a text matters, then no features matter because to matter (or mean) there has to be defining criteria for what matters and what does not matter; without such criteria, we are left with a potentially endless collection of readers’ reports on their experience of the so-called text and on what meanings they derived from this experience. And, to complicate matters (and meanings) even further, Michaels also points out that the text as object is constituted not only by all of its physical features (a kind of objective, scientific enumeration of its attributes) but also by everything that can be seen by the reader (6). So we’re left with a text as an object that generates effects rather than meanings (a text that ‘does’) and a reader who does not interpret so much as experience the text.

Before I elaborate on what kinds of subjects are entailed by such an account of textual objects, I must first point out that de Man’s much more single-minded, much less elastic than Howe’s, allegiance to materiality is the catalyst for Michaels’ polemic. As he puts it:

De Man’s insistence on what he calls a material vision thus produces - inevitably, which is to say, necessarily - a replacement of the idea of the text’s meaning (and of the project of interpreting that meaning) with the idea of the reader’s experience and with a certain indifference to or, more radically, repudiation of meaning and interpretation both. (6)

However, Michaels may also be guilty of using Howe as a straw man. That is, no matter how satisfying Michaels’ logic or the tightness of his argument (taken solely on its own terms), and whatever the stakes for literary theory in the attack on de Man, crucial nuances are overlooked in Howe’s work. The problem here is not just that he ends up inaccurately representing Howe; rather, the overlooked nuances may be just what are needed to call into question the larger argument that Michaels wants to put forward.

For example, when he argues that the preservation of Shepard’s or Dickinson’s texts as a way to protect authorial intention actually obviates intention for the sake of the text as object, he neglects the fact that Howe’s attempt (along with attempts by any number of critics such as Johanna Drucker and Jerome McGann) to preserve the object is also an attempt to keep alive the author’s intentions; that is, such critical accounts draw attention to the impossibility of ever closing a text off to a single interpretation which also means that the text is what the author intended as well as what the reader reads. Krzysztof Ziarek succinctly puts it as follows:

Howe’s relentless problematization of the effects of the standardization and uniformization of a Melville or a Dickinson text draws attention to the practices of effacing the intrinsic unreadability and open-endedness of these texts, of their plural textualities, for the sake of producing an underlying version of experience as closed and readable. These practices transform experience from an open field of possibilities into a uniform, regulated pattern, into a univocal text which effaces contradictions and conflicts, or singularities, to paraphrase Howe. (275)

What’s at stake here is a fundamental difference in Michaels’ and Howe’s attitude toward what constitutes a text, a reader, the act of reading and interpreting texts, not to mention meaning. Howe understands a text as that which must be read as an artifact of the author’s act of writing and as that which bears meaning on countless different levels (from the paper it was written on, the way in which it was written in addition to the many vectors of meaning carried by each word, each combination of words); further, meaning here is also inevitably created as much by the author as by the text itself and the reader. Moreover, given this multi-layered complexity to texts, for Howe there’s no reason to believe that a text can’t be an object; Charles Peirce’s logical graphs, for instance, are certainly object-like but there’s no doubt that they can be read and interpreted. Michaels, by contrast, seems to want a text, a reader, and a model of reading or interpreting texts that’s considerably more conservative: first of all, from his point of view a text simply is understood to consist in certain crucial features (e.g., [and minimally] certain words in a certain order), and any object that reproduces those features … will reproduce the text (3); likewise a reader is understood to read texts for meaning which in turn simply is understood to be identical with what the author intended - a position that, as far as I can tell, Michaels first outlined in Against Theory and which he continues to hold in The Shape of the Signifier.

Thus, Michaels’ model of texts, readers and meaning is not compatible with Howe’s, and so his model may also not be compatible with the wide range of writers he critiques throughout this latest book - among them Robert Smithson, whom Howe (not coincidentally) has had a long-standing interest in. However, aside from the neatness of his reasoning, what still makes Michaels’ critique of a de Manian material vision powerful is that it can be used to account for certain trends in literature and the arts that are not touched on in The Shape of the Signifier. For instance, given what I outline above, where Michaels might critique language poets such as Charles Bernstein for his move away from a strict understanding of the function of texts, readers, and reading (see, for example, Michaels’ response to Marjorie Perloff in Modernism/Modernity 3:3), his argument can be used to critique standard accounts of language poetry that turn language poems into purely non-referential, purely meaningless works which can simply be accounted for by appealing to the reader’s perceptions and experiences. Again, not only does this strictly materialist account not reflect the way we actually think about texts, but it also ignores important early statements of poetics (or statements of intention) in The L=A=N=G=U=A=G=E Book by poets such as Bernstein in which he claims that his work comes not out the death of the referent or of referentiality, meaning, or intentionality in general, but out of the recharged use of the multivalent referential vectors that any word has … of releasing the energy inherent in the referential dimension of language (115).

Further, another problematic version of such a strictly material vision is perhaps even more epitomized by the current abundance of digital poems such as Brian Kim Stefans’ The Dreamlife of Letters or Maria Mencia’s Birds Singing Other Birds’ Songs that are, for all their aesthetic appeal, cinematic projections of words and letters doing or merely producing effects. The point is not that these works ignore intention or that they’re engaged with the letter, the word, as a material, manipulable object - rather the point is that they are in fact rife with intention, turning the reader into a passive viewer who has nothing to interpret and can only register the evolving shapes of the letters, thereby closing off the possibility of a text/reader along the lines of either Howe or Michaels. As such, Michaels’ critique of Judith Butler’s theory of resignification (in which to communicate means to risk having the other assign a different meaning to one’s utterance and which Michaels also identifies with the materialist stance) could just as well be a critique of certain accounts of language poetry or of this strain of what I’m calling cinematic digital poems; as he writes, … the ‘force of the performative,’ precisely because it goes ‘beyond all question of truth or meaning,’ will be to replace the understanding appropriate to the sign with the effect appropriate to the mark, to imagine a world in which what the text means will be entirely subsumed by what it does (66).


To clarify Michaels’ argument thus far, the turn toward ontologizing texts is a concomitant turn toward the subject since, given that a text’s meaning is now determined by what it does, the reader’s responsibility is to register the effects of the text; and, crucially, since each reader will see or experience the text differently, we are not only left without meaning (or, again, ‘meaning’ as Michaels understands it to mean) but we’re also left with a limitless multiplicity of experiential accounts which cannot be argued with and instead must simply be accepted as differing. The problems with such a way of thinking, he argues, are many: first, we are left with texts that are not texts, that cannot be interpreted, and that have no meaning (when, in fact, few would claim that there is no such thing as a misinterpretation); secondly, while it cannot be denied that texts do have effects, there’s no reason to assume that the text’s effects entirely replace meaning and neither is there any reason to assume that intentional effect entirely replaces intention (127); and finally, once the interpretation of texts is no longer about determining the correct interpretation and instead becomes one of registering the text’s effects, we no longer have any reason to disagree - how can anyone disagree with an experiential account? As Michaels puts it, … the difference between what you see and what I see is just the difference between where you’re standing and where I’m standing - literally, a difference in subject positions (10). In short, the commitment to the materiality of the signifier necessarily means a commitment to the primacy of the subject, the primacy of identity. And so we’ve arrived at the second end to accompany the end of modernism: the end of history.

Working from Francis Fukuyama’s The End of History and the Last Man (1992), Michaels argues that the end of history - marked, according to Fukuyama, by the end of the Cold War - not only means the end of ideological conflict between liberal capitalism and socialism, but it means the end of ideological dispute as such; in other words, the supposed triumph of the West (which has become a global triumph) ended dispute over how society should be organized and so also ended dispute over beliefs (the only driving force behind differing ideologies). So while the Cold War made beliefs essential, thereby also making disagreement possible, the end of the Cold War instead made identity essential, thereby making disagreement impossible. The issue is complex and multi-layered in a way that cannot adequately be expanded on in this review; but the impossibility of disagreement can partly be explained by the following from Michaels:

The differences between (and within) bodies may here be understood as underwriting the insistence on all the nonideological differences. It is, in other words, those differences that have nothing to do with differences in belief - racial difference, sexual difference, linguistic difference, even (and, in a certain sense, especially) cultural difference - that emerge as foundational … people with different bodies don’t thereby have different beliefs. And even people with different beliefs can be understood as not disagreeing with each other as long as their beliefs are understood to constitute a culture rather than an ideology …. Culture, in other words, has become a primary technology for disarticulating difference from disagreement. (16)

One’s culture, like one’s identity, is now seen as something one has - like the language that one speaks, it’s neither right nor wrong but rather just what one is or just the unique subject position one holds. Given the overwhelming dominance of culture as a defining feature of present-day America, it becomes clear that what disturbs Michaels about this post-Cold War transformation is not that people no longer disagree, but rather that this dominant axis of thought makes it impossible for one to say that one’s actions (either as an individual, a culture, or a country) are either right or wrong, defensible or indefensible - they simply are part of one’s identity. And so, to begin the slow return to the beginning of this essay, since economic inequality is tied to class and since class-struggle must be over if communism is over, then poverty becomes an identity that one has, an identity that ought to be respected and that need not be overcome.

Michael Hardt and Antonio Negri’s Empire is particularly the object of Michaels’ analysis as he argues they treat poverty as if it were an identity, or, more precisely, as if it were identity itself (180) - thereby ontologizing poverty (in the same way that Howe or de Man supposedly ontologize texts) as if it is an identity that ought to be acknowledged and respected rather than fought against or eradicated. However, for Michaels class cannot be replaced with identity - the former can be overcome whereas the latter cannot. (I should also point out that Michaels does demonstrates the complexity of the matter by acknowledging that the transformation of the poor into an identity is a way of reclaim[ing] the poor for the left such that once we turn poverty into a structure of identification … we can begin to think that our problem is that we’re all insufficiently ‘poor’ [181]). But while I sympathize with his assertion that poverty is hardly something that ought to be respected as if it were a mere difference between people, Michaels may again be guilty of overlooking certain nuances to the issue in favor of an all too neat argument that ties together the different strains of thought comprising his career as a thinker. For instance, on a mundane level, it certainly seems to be the case that Buffalo, where I’m presently writing this essay, is the home of a significant working-class identity that is not necessarily defined by income; the moment any one of my Buffalo-bred undergraduate students speaks up in class I’m immediately aware of how much more than simply money informs their thinking. On a less mundane level, Hardt and Negri are hardly simply proposing to ontologize poverty in the way that de Man seeks to ontologize texts - they are proposing, at the very least, that there is a new proletariat (distinctly different from the industrial working class who were defined by their waged labor and who therefore were defined simply by how much money they had) which participates in processes of ontological constitution that unfold through the collective movements of cooperation (402). In other words, a processual unfolding of ontological constitution is not identical to ontologizing - the former does not imply that culture is something one has in the way that one has a language, but rather that culture is something that one is continually, actively in the process of creating. Moreover, it also seems that Hardt and Negri are not speaking of identity at all since the scope of the new proletariat, or those whose labor produces and reproduces social life and so are not contained by the working day, is so far-reaching as to cease to be an identifiable identity at all.

The Shape of the Signifier, then, pulls together the major strains of Michaels’ thinking over the last twenty years or so to argue that the aesthetic commitment to the physicality of the text and to the subject position, as at least partly exemplified by Susan Howe’s concern with Shepard’s eighty-six blank manuscript pages, runs parallel to the political commitment to embrace cultural identity. While, to my mind, the project problematically comes to resemble a theory of everything, there’s no doubt Michaels has written a deeply provocative and deeply troubling book for artists and thinkers alike.


Many thanks to Benjamin Robertson, Joseph Tabbi, and the members of the Buffalo poetics listserv for their helpful suggestions.

Works Cited

Bernstein, Charles. Semblance. The L=A=N=G=U=A=G=E Book. Eds. Bruce Andrews and Charles Bernstein. Carbondale and Edwardsville: Southern Illinois UP, 1984.

Eagleton, Terry. After Theory. New York: Basic Books, 2003.

Fukuyama, Francis. The End of History and the Last Man. New York: Free Press, 1992.

Hardt, Michael and Antonio Negri. Empire. Cambridge and London: Harvard UP, 2000.

Howe, Susan. The Birth-Mark: Unsettling the Wilderness in American Literary History. Hanover and London: Wesleyan UP, 1993.

—. Talisman Interview, with Edward Foster. The Birth-Mark: Unsettling the Wilderness in American Literary History. Hanover and London: Wesleyan UP, 1993. 155-181.

Mencia, Maria. Birds Singing Other Birds Songs. http://www.m.mencia.freeuk.com/

Michaels, Walter Benn. Our America: Nativism, Modernism, and Pluralism. Durham: Duke UP, 1995.

—. Response. Modernism/Modernity 3:3 (September 1996): 121-126.

—. The Shape of the Signifier: 1967 to the End of History. Princeton and Oxford: Princeton UP, 2004.

—.  with Knapp, Stephen. Against Theory. Critical Inquiry 8:4 (Summer 1982): 723-743.

—.  with Williams, Jeffrey J. Against Identity: An Interview with Walter Benn Michaels. The Minnesota Review 55-57 (2002).

Stefans, Brian Kim. The Dreamlife of Letters. http://www.ubu.com/contemp/stefans/dream/index.html

Ziarek, Krzysztof. The Historicity of Experience: Modernity, the Avant-Garde, and the Event. Evanston, IL: Northwestern University Press, 2001.