Resisting Allegory: Interpretive Delirium in Spenser's Faerie Queene

By Harry Berger and Jessica O'Reilly

Spenser is a delirious poet. He can’t plough straight. What he builds is shiftier, twistier, than anything dreamed up or put down by M. C. Escher.

So begins Resisting Allegory, in which the leading Spenser critic of our time sums up a lifelong commitment to the theory and practice of textual interpretation. Spenser’s great poem provides the occasion for a searching and comprehensive interdisciplinary exploration of reading practices¾those the author advocates as well as those he adapts or criticizes in entertaining a wide range of critical arguments with his celebrated combination of intellectual generosity and rigorous questioning.

Berger is interested in how details of the poem's language—phrases, images, figures on which we haven’t put enough interpretive pressure—disconcert traditional interpretations and big discourses that the poem has often been thought to serve. Central to this volume is an attention to the deployment of gender in conjunction with the Berger’s notion of narrative complicity.

Resisting Allegory offers a model of theoretically sophisticated criticism that never wavers in its close attention to the text. Berger offers a sustained and brilliantly articulated resistance not only to allegory, as the title indicates, but also to prevalent modes of cultural and historical criticism. As in all of Berger’s books, a lucid reflection on questions of method—based on a profound and richly theoretically informed understanding of the workings of language and of the historical situations of the people involved in it—are interwoven with an interpretive practice that serves as an exemplary pedagogical model. Berger attends to historical and political context while deeply respecting the ways in which text can never be reduced to context.

This distinctive and original book makes clear the scope and coherence of the critical vision elaborated Berger has elaborated in a lifetime of seminal and still-challenging critical arguments.

• Language: English
• Publisher: Fordham University Press
• Release date: Jan 7, 2020
• ISBN: 9780823285648

Harry Berger

Harry Berger, Jr., was Professor Emeritus of Literature and Art History at the University of California, Santa Cruz. His most recent books are Resisting Allegory: Interpretive Delirium in Spenser's ‘Faerie Queene’; Harrying: Skills of Offense in Shakespeare's Henriad; and The Perils of Uglytown: Studies in Structural Misanthropology from Plato to Rembrandt.

Book preview

EDITOR’S INTRODUCTION

David Lee Miller

Sixty-two years ago Spenser criticism woke up to the Berger alarm. It was 1957. Benny Goodman was old hat, but the New Criticism was in full swing. The New Critics didn’t think much of Spenser, though, and while there was plenty of good work filling the pages of scholarly journals, most of it was rather old-fashioned: A. S. P. Woodhouse’s "Nature and Grace in The Faerie Queene was still recent (and controversial). Once conspicuous irrelevance" had fastened itself like a burr under the saddle of Spenser criticism, there would be no going back.

Over the next three decades, Harry Berger, Jr., would publish sixty strikingly original and enduringly influential essays on Chaucer, Spenser, Shakespeare, Marvell, Milton, Plato, Alberti, Virgil, Dante, Da Vinci, Pico della Mirandola, Vermeer, Beowulf, Erasmus, More, Theocritus, the concept of cultural change, the theory of periodization, and the poetry of Robert Frost. A number of these were gathered into a pair of volumes issued in 1988 by the University of California Press: Second World and Green World: Studies in Renaissance Fiction-Making, edited by John P. Lynch, and Revisionary Play: Studies in the Spenserian Dynamics, with an introduction by Louis A. Montrose. Side by side on the shelf those two big books stand like pillars for a monument to critical intelligence, but that is the wrong image. Berger is no less restless in revising previous arguments than he is prolific in spinning new ones. In fact, he was just getting started. A dozen books later, on topics as various as Rembrandt, structural misanthropology, caterpillars, courtesy books, and Christian nudes, readers may be forgiven for feeling as if they have encountered not a monument but a whirlwind.

Not long after the appearance of Revisionary Play, Harry embarked on what we can now recognize as the third stage in his career-long engagement with Spenser: a series of seven essays published between 1991 and 2005 that revisit Books 1–3 of The Faerie Queene. Taken together, these essays develop both a comprehensive interpretation of the 1590 text and a finely crafted theory explaining and justifying a method of reading against the allegorical grain. Three essays on Book 1 have been revised and incorporated into the first chapter of Resisting Allegory, which now offers a full-dress reading of the Legend of Holinesse. Four more essays, two each on Books 2 and 3 of the poem, are reprinted here. They include "Narrative as Rhetoric in The Faerie Queene," a landmark of Spenser criticism that first appeared in English Literary Renaissance in 1991 and set the table for the work to come.

Clearly these essays were always parts of a single project. Together they add force and momentum to the argument launched in their various debut appearances. That argument is still timely. Berger’s work has always aged well: Essays from the early ’60s appear in the 1988 collection with no loss of pertinence. Unlike much practical criticism from that era, they are still intellectually bracing, demanding the reader’s full concentrated attention. The essays in this volume are just as tightly wound, and perhaps this is one reason much Spenser criticism that has appeared since the journal articles were published has not reckoned with their careful readings and arguments—ironically, given the scrupulous attention Berger himself tenders to the criticism he reads. The publication of Resisting Allegory offers us another chance to take stock.

•

Three features of the work gathered here especially stand out. First, the work represents a remarkable rereading and synthesis of the modern critical tradition. Throughout these essays Berger is actively engaged with the work of critics across three generations. By this I don’t mean that his footnotes are mini-bibliographies, but that he quotes, criticizes, adjusts, and assimilates the work of major critics from C. S. Lewis and Northrop Frye through William Nelson, Paul Alpers, A. Bartlett Giamatti, John Webster, Michael Murrin, Donald Cheney, and Richard Lanham to A. Leigh DeNeef, Stephen Greenblatt, John N. King, Maureen Quilligan, Lauren Silberman, Jonathan Goldberg, Arthur Kinney, David Norbrook, Louis Montrose, Susanne Wofford, and Linda Gregerson. Jeff Dolven remarks on this aspect of Berger’s work in Harry Berger’s Intellectual Community: More than any other critic I know, he builds his arguments out of encounters with the thinking of other members of his guild, our guild.¹

Berger comments in Narrative as Rhetoric that the convergence of critical arguments he is orchestrating suggests the emergence of a consensus, and together these essays powerfully advance that claim. It is by no means a neutral or easy synthesis: The argument carries a sharp polemical edge that cuts against the practice of much historicist and cultural criticism. It’s also distinctively American, for although Berger does cite a few English critics, much English and Irish criticism has pursued the cultural studies agenda targeted by Berger’s main line of argument. But the American Spenser Berger has constructed by rereading both the poem and the critical tradition will demand that future interpreters, no matter their provenance, method, or allegiance, come to term with its arguments.

One reason this collection deserves attention is that its assumptions and procedures are so carefully articulated. This is the second feature that stands out clearly. Berger is as brilliant and assimilative a theorist as he is an interpreter. Beginning with the introduction, which lays out in a few bold, elegant strokes the theory and method that inform the argument, Berger offers a remarkable reflection on critical method. For all that theory seems to have been the rage in academic criticism during the late twentieth century, there are not many books on Spenser that interweave a sustained and inventive interpretive practice with a lucid and equally sustained reflection on questions of method. It is this combined emphasis that allows Berger to incorporate so much of the critical tradition into his own argument without simply amalgamating incompatible approaches: He is actively sorting, evaluating, and disentangling critical arguments even as he acknowledges their achievements with unfailing generosity.

The third remarkable feature of this collection is the brilliance of the critical readings Berger develops. He has been among the most perceptive and original commentators on The Faerie Queene, and indeed on almost all of Spenser’s poetry, since his first book appeared. Generally acknowledged as at once a forerunner of, an inspiration for, and a leading participant in the Renaissance of Spenser studies during the 1960s, Berger has never stopped reading and rereading both the poetry and its most persuasive critics. The essays in this volume, written and published after Revisionary Play appeared, represent a powerful new turn, a challenging reinvention of his own, and everyone else’s, critical practice by the leading Spenser critic of our time.

In responding to this work, we don’t have to start from scratch. In essays that appeared together in the 2009 volume A Touch More Rare, Judith Anderson, Katherine Eggert, and Catherine Gimelli Martin engaged directly and at length with Wring Out the Old and Narrative as Rhetoric.² I take up their arguments here to advance the conversation these chapters should occasion.

Anderson begins with Harry’s claim in Wring Out the Old that Acrasia is the objectification of male hysteria, remarking that although this pointed ideological critique is true so far as it goes, Where misogyny is brilliantly exposed and rejected, I’m worried that androcentrism might still be alive and well, and indeed, reinstated (78). Accordingly, she sets out to find the mark of something female (80) persisting within both the Bower and the traditional, misogynistic male discourses that dominate there. The argument reaches its high point in Anderson’s commentary on the two descriptions of Acrasia, and in particular on the second, at 2.12.77–78. In a moment that strikes me as unusual in her criticism, Anderson responds to the beauty of the verse simply by describing the sense of wondrous pleasure it evokes and the disorientation this pleasure causes:

If I affirm that the erotic appeal of both stanzas is subtle, complex, seductive, and enveloping, I’m not sure what this says about Acrasia’s gender or about my own implication in male discourse or even whether such discourse can remain pristinely male once it involves me. I frankly believe, however, that the appeal of these stanzas exceeds sharply defined barriers of gender and other rational determinants. More simply put, I experience some kind of exhilaration or wonder in and at these lines, whatever their origin. (88)

Anderson goes on to gloss this effect by way of Jean-Luc Nancy’s The Ground of the Image, with its modes of "participation and contagion" (89). This move from confessional candor to theoretical precision is, in my view, as good as criticism gets. It is a perfect example of the quality Ayesha Ramachandran ascribes to Anderson’s work in a recent review essay:

In [Anderson’s] approach, there lurks a powerful response to Rita Felski’s call to imagine a form of post-critical reading that does not look behind the text—for its hidden causes, determining conditions, and noxious motives—but in front of the text, reflecting on what it unfurls, calls forth, makes possible.³

In other words, Anderson’s response to Berger is powerful and exemplary because it finds a way not just beyond wall-to-wall male discourse, but also beyond wall-to-wall ideological critique.

Within Judith Anderson’s learned, rigorous, and insightful body of work, this essay stands out for me because it is so playful, affectionate, even self-revealing, and these qualities clearly come forward in response to Harry both personally and critically—they are part of what his own work unfurls, calls forth, makes possible in the critic who matches him joke for joke, insight for critical insight. Harry’s affective responses to the text are on display throughout his commentary in the exuberance and hilarity that mark its style; Anderson brings off an impressive balancing act in responding generously to these qualities even while catching the master off guard.

Katherine Eggert’s Harry Berger’s Genius: Porting Pleasure in the Bower of Bliss might seem at first to be diametrically opposed to Anderson’s emphasis on pleasure, and there are certainly differences: Where the pleasure Anderson describes exceeds sharply defined barriers of gender, Eggert is seeking in the poem a point of entry for a feminine poetics (103). She begins this readerly quest with a counterintuitive but intriguing reading of the Bower’s Genius as anhedonic (101):

What does it mean to port pleasure? Astonishingly, it turns out not to mean either to experience pleasure or to convey it to another person, because this first Genius of the poem does neither.

If to port is to carry, then this Genius is all about vehicularity; his porting merely carries pleasure, it does not transmit pleasure, and thus the vehicle of Genius drives us to no tenor, not even one preposterously far away. (95)

As with Anderson’s recourse to Nancy on the theory of the image, Eggert here is insisting on the poetic surface divorced from meaning, although to different effect. Genius’s porting in her account becomes Guyon’s passing: Passing forth allows him to pass by and pass up (99). But this anhedonic mode yields finally to a

very Acrasian subversive motion and subversive thought …: one where the eye wanders on the surface of things and never seeks an Idea to go with that perceived Image, and where that wandering eye is enough to hint—just hint—at an alternative poetics, one that Patricia Parker, drawing from Valéry, associates with a delaying of [metaphor’s] movement towards meaning or object.⁴

This recasting of deferral as sweet reluctant amorous delay offers a mode of pleasure that has traditionally been gendered feminine, as it is in the passage I am echoing from Paradise Lost. The similarity to Anderson’s essay is instructive, for it is her own quest to discover the mark of something female in the text that leads finally, in her account, to a form of pleasure that exceeds gender categories even as it contaminates the aesthetic with the erotic.

One wonders whether Eggert’s reading of the Bower might lead to a similar conclusion, and in an important sense, it does. As readers, we get there through a version of the reverse jump cut (101) Eggert notices in Guyon’s and the Palmer’s double-approach to the Bower (they arrive in stanza 42 and again in stanza 69). Turning back from this 2009 essay to "Spenser’s Ravishment: Rape and Rapture in The Faerie Queene" (2000) we find the same critic arguing that

The Faerie Queene … if only intermittently, hints at poetry as a vehicle for rapture, a suffusion of delight that suspends the quest and admits a multiplicity of both erotic and epistemological pleasures. This rapture is felt, I want to demonstrate, by female as well as male characters in the poem, and specifically in scenes that a number of Spenser critics have identified as allegories of the writing and/or reading of poetry.⁵

Rereading this provocative account of pleasure in the poem, I can’t help thinking that the next step should be to consider—with Anderson à la Nancy and Felski—the pleasures of the poem, most especially in the Bower.

Catherine Gimelli Martin offers a different kind of resistance in her reading of Narrative as Rhetoric. It is an ambitious critique, one that addresses both the theoretical argument of this important essay and its close reading of the Phedon episode in canto 4 of The Faerie Queene, Book 2. It is also, so far as I am aware, the only full-scale effort to address the method and theory central to this volume. The argument Martin puts forward is indeed shrewd:

Why must revisionary reading proceed only in one direction, toward irony, self-parody, and subversion; why might not the mixed modes of allegory and oral narrative sophisticate without canceling the moral lessons that Spenser’s Letter to Ralegh claims that The Faerie Queene teaches? (151)

I am not sure that the essays collected here offer an answer to this question. Surely, it is not coincidental that the values informing Harry’s readings of Spenser happen to be those of late-twentieth-century academic liberalism.

To suggest that a question has not been answered is not, however, to assume that it can’t be. After all, no critical practice can operate outside of ideology; a critic without a point of view is a critic not worth reading. Berger’s answer is, as ever, to point to the text: Here, he implicitly says, is where I see the values of feminism (for instance) coming into play. Here is how they operate in the text. And, to her credit, that is where Martin engages him, opposing his reading of the Phedon episode with her own.

She tackles Berger’s reading head on, maintaining that not even the closest of close readings suggests that Phedon’s tale is self-exculpatory (152). The evidence she offers in support of this counterclaim suggests that Berger may have overstated his case, but there is surely a strong element of self-exculpation in the tale. I point to the text:

With hart then throbbing, and with watry eyes,

Fayre Sir (qd. he) what man can shun the hap,

That hidden lyes unwares him to surpryse?

Misfortune waites aduantage to entrap

The man most wary in her whelming lap,

So me weake wretch, of many weakest wretch,

Vnweeting, and vnware of such mishap,

She brought to mischiefe through her guilful trech,

Where this same wicked villein did me wandring ketch. (4.17)

It was a faithlesse Squire, that was the sourse

Of all my sorrow, and of these sad teares. (4.18.1–2)

Phedon points first at the female personification Fortuna (Miss Fortune) and then at his false friend, but the text that displays these evasions points back at him.

Martin’s close readings display some of the special pleading she finds in Berger, for example when she paraphrases the Palmer’s belated advice to Phedon as root out ‘love’ entirely if it grows from ‘filth’ or lust (154). The stanza in question is well known as a rhetorical tour de force:

Wrath, gealosie, griefe, loue do thus expell:

Wrath is a fire, and gealosie a weede,

Griefe is a flood, and loue a monster fell;

The fire of sparkes, the weede of little seede,

The flood of drops, the Monster filth did breede:

But sparks, seed, drops, and filth do thus delay;

The sparks soone quench, the springing seed outweed,

The drops dry vp, and filth wipe cleane away:

So shall wrath, gealosy, griefe, loue die and decay. (4.35)

The stanza is difficult because it combines the tight patterning of the clauses with inversions of syntax: [D]o thus expell and do thus delay govern and direct the rhetoric as imperatives. There is no exculpatory if: The Palmer simply says that love is a monster bred of filth, and the passage rightly figures as a prime exhibit for the gynophobia Berger and others attribute to Temperance as it is represented in the narrative.

One final example. Near the end of her essay, Martin refers to Guyon’s combat with Furor, a monster who immediately overthrows himself in combat with a truly temperate knight, Sir Guyon (2.4.8). Again, I point to the text:

His rude assault and rugged handeling

Straunge seemed to the knight, that aye with foe

In fayre defence and goodly menaging

Of armes was wont to fight, yet nathemoe

Was he abashed now not fighting so,

But more enfierced through his currish play,

Him sternly grypt, and hailing to and fro,

To ouerthrow him strongly did assay,

But ouerthrew him selfe vnwares, and lower lay.

Spenser is famous for scenes in which combatants become indistinguishable, so the mistake is pardonable, but it is indeed simply a mistake. The subject of the verbs after line 2 is uniformly the knight, and it is Guyon who overthrows himself in combat with Furor. To read the stanza otherwise makes nonsense of the encounter, which ends with the Palmer’s warning to Guyon that Furor is not, ah, he is not such a foe, / As steele can wound, or strength can ouerthroe (10.4–5).

It would be going too far, I think, to read Guyon here as a figure for the critic, wrestling with Harry only to overthrow herself. Martin’s essay is too appreciative, too honorific, even, to be characterized as combative, and Harry’s response—if we had it—would be a model of courteous conversation. I cite all three essays discussed here as model responses to Berger in the hope that they may inspire further, equally serious and respectful, critical engagements with work that is too important to ignore. In an essay published in 2006, I ended a sampling of modern criticism on the 1590 Faerie Queene with an appraisal of the essays collected here:

Berger’s latest essays on The Faerie Queene I–III, gathered into a forthcoming book from Fordham University Press, bring the present account to its close.… I conclude with Berger not only because he was a harbinger of (and a major figure in) the Spenser renaissance of the 1960s, but also because he has never stopped intensely reading and rereading the poem and its critics, citing generously from and arguing shrewdly with peers and predecessors as well as younger scholars. The result is a half-century of work that in my judgment comprises the most sustained and impressive body of commentary in the history of Spenser criticism.⁶

Publication of Resisting Allegory may have been delayed, but it offers a fitting close to a distinguished and utterly distinctive body of work. In that close, may Spenser criticism find many new beginnings.

RESISTING ALLEGORY

INTRODUCTION

On Texts and Countertexts

Language speaks to us about the ways we speak the language.

—A. LEIGH DENEEF, Spenser and the Motives of Metaphor

Spenser is a delirious poet. He can’t plough straight. What he builds is shiftier, twistier, than anything dreamed up or put down by M. C. Escher. I began thinking and writing about The Faerie Queene in the early 1950s, and that poem has never let me go because it has never let me in, has kept me digging outside its crooked walls for five decades in a responsive delirium of interpretation, mucking about at the alexandrine foot of the castle, furrowed by its aura, looking for the unlikely wicket, the machicolated flicker of light, the unraveled end of a Phaedrian lifeline winding through secret passages toward the Transcendence and toward the moment when finally I might give myself to be stigmatized not (as I already well know) by the beams of that Transcendence but by the harrowing teeth of its Judgment, its Proscription, for it will ask me why I dared presume and it will proclaim where I went wrong and how I, unworthy to be there, must find my way back out or be forever lost in its sandy bowels. I am The Faerie Queene’s Joseph K. But what follows is not my testimony, not my confession, only the delirium momentarily straitened and reduced to its method before it releases itself into the obliquity of example.…

Interpretive practices are polarized by the tension between two orientations, two differently accented modes of sign use put into play by speakers and auditors, or by writers and readers, or by the bodies of signification they mutually produce. On the one hand, there is a set toward the message, the story, whatever is being referred to or talked and written about. This includes a set toward the communicative transaction between sender and receiver—includes, that is, not merely the informational context but also the rhetorical form of the content.

On the other hand, there is a set toward the production of meaning and toward the content of the form, and this includes a set toward the possibility that the text may generate a message about the message. Let’s provisionally distinguish these two sets by calling the first one innocent reading and the second one suspicious reading. Let’s imagine that between them they polarize a continuum, a differential field of reading. And let’s give this continuum the form of an interpretive shuttle.¹ To ride the shuttle is to weave a fabric and pattern of reading that responds differentially to the pressures from each pole.

The two poles—one oriented toward the message and the other toward messages about the message—may be correlated to the two-level structure of the sign: First, the sign is coupled to a referent; then, within the sign, the signifier or verbal expression is coupled to the signified, the expression’s content, or sense, or meaning. Imagine the standard form of the dictionary entry. The first word in the entry occupies the place of the term being defined and thereby becomes the signifier. The words that follow it define its meaning (or its grammatical use in the case of function words like prepositions) and thus become its signified, or range of signifieds. Together, the signifier and the signified constitute the sign. The sign is the whole dictionary entry; it is the defined word.

Notice that the referent of the sign is not the same as its signified, but exists at a different level. Normally the sign doesn’t refer to itself or its parts. Within the sign, the signified designates or marks out a particular range of referents. But there are no referents in the dictionary when it isn’t in use. I said designates or marks out, not indicates or points out. The referent has to be marked out, distinguished by language, before the sign can be used to point it out. Considered strictly as a signifying code, a system of signifiers and signifieds, language makes reference possible but it doesn’t refer. Only language users, sign users, refer. For the sign to refer to the thing for which it stands, it has to be picked up by a sign user and pointed at the thing.

Pointing the sign at the thing—which could of course be another sign—is an act of reference that converts the thing into a referent. I use the expressions point to and point at to dramatize the notion that referring is an act, something sign users do. Thus if a sign is "everything that, on the grounds of a previously established convention, can be taken as something standing for something else," the something else is its referent.² Signs refer only when language users point them at things outside the signs they use. I note in passing that although signs normally don’t signify their signifiers and signifieds, they can be made to do so, just as things normally positioned as referents can be made to signify their signs.

The explanation I just gave is based on lexical interactions. Even there, you can glimpse the potential for complexity in the fact that one signifier can have many signifieds and one signified can couple with many signifiers. But when we go beyond the lexical and sentential levels, problems and complications multiply. For example, since reference involves sign users, the study of reference necessarily implicates the study of self-reference. This inevitably leads to the study of transactional relations between and among sign users.³ And as Teresa de Lauretis insists, the study of sign use includes the study of the materiality and the historicity of the subject itself—the study of a subject touched by the practice of signs, … physically implicated or bodily engaged in the production of meaning, representation, and self representation, and therefore, as she reminds us, the study of a gendered subject.⁴

Another problem is posed by the tendency to syncopate the two pairs of terms, signifier/signified and sign/referent. The reasoning here is that since both the signifier and the sign can be said to signify, why can’t the signifier be coupled to the referent, and if it can, what’s the point of distinguishing between signified and referent? Many interpreters make do with only the terms signifier and signified. For example, the distinction between signified and referent is often elided out in cultural critique and deconstruction. But at the same time, something else also tends to get elided: the distinction between what language does and what language users do. Granted that we interpreters always look at examples of language use—never at language pure and simple, but at discourse or language in use (dictionaries are products of discourse)—we can still uncouple the language from the language user. That is, we can distinguish what the users do with their language, what they mean by it, what they see and hear in it, from what their language says, does, and means regardless of what users say, do, mean, and read or hear.

Parallel to the distinction between innocent and suspicious reading is a distinction between documentary and mischievous sign use. I call sign use documentary when the objects constructed by the desire or decision to focus on sign/referent relations tend to be scenes of instruction and persuasion, sources of evidence, purveyors of doctrine, instruments of knowledge and doxa, producers of docility—in a word, documents. In the documentary mode signs are used and represented in a way that directs attention beyond them to their referents, a way that either doesn’t invite or even actively resists any effort to explore the activity going on within the signs themselves. When deployed in normal communication, signs are expected to be as inconspicuous as good servants in showing you the way to their referents. But in mischievous sign use, the signs either fail to be good servants and disturb you by getting in your way or else they behave like seditious servants who act up and get in your face.

Both innocent reading and documentary sign use fix on sign/referent relations. Innocent reading is what we all do as ordinary readers in the street. Suspicious reading and mischievous sign use fix on signifier/signified relations. Readers and other sign users may shuttle back and forth between these two positions or activities, and it is to mark their movements that I introduce the terms textualization and detextualization. To textualize is to shift attention from the documentary mode focused on sign/referent relations to the suspicious mode focused on signifier/signified relations. Textualization names an activity in which attention is redirected from a particular set of sign/referent relations to relations and interactions of various kinds between the signifiers and signifieds in that set.

I use the word text to designate the product of this textualizing activity, and I use textuality to designate its appearance. Mieke Bal defines text in a similar way—as the product of a mode of reading that allows for a continuous shaping and reshaping of sign events. She opposes this mode to the realist mode of reading for a content … modeled on reality at the expense of awareness of the signifying system of which the work is constructed.⁵ Under this definition, text differs both from actual writing, the script that represents words, and from the messages transparently conveyed by the writing.

Whatever invites or is produced by interpretation, whatever looks like it wants to be or has been interpreted, whatever we decide we want to interpret, assumes the status of text. The category thus includes any kind of writing and, more generally, any kind of sign system. It includes the goings-on among the semiotic, phonological, prosodic, lexical, rhetorical, and grammatical registers of the signifying context. It includes the effects of different media or channels on those events.⁶ And it doesn’t exclude the way all these interactions affect and modify the act of reference. The text produced by suspicious reading or mischievous sign use will always be a comment on the message, on the reference, and on the apparent intentionality (vouloirdire) of the writing.

The problem is to try to determine where this comment comes from, the writing, the reader, or some combination of both. For if there is mischievously opaque writing that invites textualization, mustn’t there be mischievously transparent writing that actively and knowingly resists it? And shouldn’t we also allow for the oppositional writing that represents the resistance to textualization in an ironic or parodic mode of critique?

If to textualize is to redirect attention from sign/referent relations to the various kinds of interplay between signifiers and signifieds, then to reverse this operation is to detextualize. On the one hand, textualization provisionally dethrones the sign/referent relations and puts them in brackets in order to free up the seditious particles bouncing about and colliding crazily within the nucleus of the sign. On the other hand, detextualization places the product of textualization in brackets, restores the transparency of the sign, and reestablishes the hegemony of the referent. But it doesn’t restore documentary innocence. Rather, it produces an imitation or simulacrum of the document. Detextualizing enacts or stages a defense against textualization, and for this reason I call its product a countertext.

Countertexts come in many forms. Books can be countertexts; textbooks definitely are. The Bible is the site of continual documentary and countertextual struggle. The body and cosmos are major cultural countertexts. Within and between them is an array of supporting countertexts that includes genders, lineages, ethnicities, rituals, religions, governments, and the variety of institutional discourses investigated by Foucault and others. These compose into the detextualized frame of reference—into the worldview or framework or dominant ideology in terms of which acts of reference are shaped, encouraged, and validated or violated. Whenever textualization burrows into what Jean-François Lyotard calls the universe of the phrase,⁷ it does so within this detextualized framework. The texts it produces are always relative to some particular state or aspect of the framework. To be more precise, they’re relative to two states of the framework: One is the context of the current state of documentary culture, and the other is the context of the current state of interpretive discourse and practice.

It may be the pleasure or jouissance of the textualizing act that produces the familiar paranoia of interpretation. Anyone who spends time moving back and forth across the interpretive shuttle between innocence and suspicion, or between documentarity and mischief, is liable to suspect that since all documents may enact the defense against interpretation, all may be countertexts in disguise. Riding the shuttle produces doubt not only about the givenness and closure of the sign/referent linkage but also about the innocence of the documentary standpoint that privileges them.

Countertextuality has a long history. We all know that one picture is worth a thousand words. Or at least we’ve heard that; we’ve been told it since the early middle ages, when visual images were called the bible of the illiterate, who could look at them while they were hearing ritual and homiletic speech acts.

In his famous vindication of images, Gregory the Great (c. 540–604) writes that the picture is for simple men what writing is for those who can read, because those who cannot read see and learn from the picture the model they should follow. Thus pictures are above all for the instruction of the people. Some seven centuries later, around 1285, William Durand uses Gregory’s dictum to introduce a discussion of church pictures: What writing supplies to him who can read, a picture supplies to him who is unlearned and can only look. Because they who are uninstructed thus see what they ought to follow: and things are read, though letters be unknown.

A little later Durand again cites Gregory on the value of pictures:

Paintings appear to move the mind more than [written] descriptions: for deeds are placed before the eyes in paintings, and so appear to be actually going on. But in [written] descriptions, the deed is done as it were by hearsay: which affects the mind less when recalled to memory. Thus it is that in churches we pay less reverence to books than to images and pictures.

Notice that this is more tendentious than the preceding statement, which only vindicates images as a kind of stopgap, a second-best if not a last resort. This formulation implies on the contrary that even if you can read words you’ll get more out of reading pictures. The instruction will be more effective because more affecting. The indoctrination will be more thorough, will go deeper, because the medium of instruction is more vivid and immediate. Here we slide into other familiar places: the discourse of paragone, the comparison of visual, verbal, and auditory arts: the discourse that grew up around Horace’s ut pictura poesis; the Reformation project of getting the power of interpretation out of the clutches, and out of the spaces, of Catholic ritual and mythology.

My point about the tendentious statement is that it is a theory of interpretation. To be more precise, it is a theory of disinterpretation, for if one picture is worth a thousand words, then a thousand words are reducible to one picture. On this theory the best method of reading involves looking through the language at the referent, that is, reading as if visualizing. The theory brings to bear an injunction against suspicious reading and mischievous writing. It is a form of censorship that aims to discourage the textualizing activity.⁸

Another version of this theory seems to invite interpretation by the promise of hidden meaning that it withholds from all who do not have the key. In his important study of iconoclasm in the English Reformation, Ernest Gilman observes that the Renaissance’s odd fascination with Egyptian hieroglyphs instances a

conception of the visual image as concealing a language.… The Egyptian priests had preserved their arcane wisdom in ideograms.… Here was a mute and symbolic language of ideas (Valeriano) at once more universal and indelible than other tongues, and more directly expressive of divine wisdom than written language. In the hieroglyphic image, meaning presented itself purely and instantaneously in the lexicon of things—in rebus rather than in verbis—without the mediation of words.⁹

This involves a more explicitly elitist rationale than that of the Gregorian mode, but the goal is the same. As Plato had shown in the Phaedrus and the Timaeus, the pictographic writing of the Egyptians differed from the more democratic alphabetic writing that represented oral speech in that the interpretation of hieroglyphics was restricted to the priesthood.¹⁰

Gilman also discusses the example of George Sandys, whose frontispiece to his translation of Ovid’s Metamorphoses is followed by a brief poem called

The Minde of the Frontispeece, And Argument of this Worke, a poem that not only clarifies the allegory but further suggests by its title that the frontispiece is itself the image of a Minde, of the integral idea of the