This action might not be possible to undo. Are you sure you want to continue?
2 (Feb., 1987), pp. 137-157 Published by: National Council of Teachers of English Stable URL: http://www.jstor.org/stable/377870 . Accessed: 30/04/2011 16:56
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at . http://www.jstor.org/action/showPublisher?publisherCode=ncte. . Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact firstname.lastname@example.org.
National Council of Teachers of English is collaborating with JSTOR to digitize, preserve and extend access to College English.
MarshallW. Alcorn, Jr.
One consequence of post-structuralist theory seems to be that it is no longer possible to assume the presence of an "objective" text. The text cannot be objective because, if it is considered as the encoding of an author's message, the text cannot remain tied to the integrityof its originatingor controllingpurpose. It loses its message as it is violated by readingstrategies. If we dismiss the importance of the authorand think of the text as an autonomouslinguisticartifact, the text still cannot exist as an objective entity, for it cannot control or deliver univocal meaning. As a verbally "autonomous"entity, the text cannot be "autonomous"; it is always on the point of dissolving. It is unable to enclose the boundariesof its subject, unable to limit the pluralitiesof its signification.These recent discoveries of linguistic unfaithfulness have changed the attitudes of many prominent critics. Stanley Fish, for example, observes that in his early years "the integrityof the text was as basic to my position as it was to the position of the New Critics" (9). Later in his career, however, Fish came to recognize a disquieting arbitrariness in what had been called with reverence, the "text." The text does not exist as a stable objective entity. The text becomes differentiallypresent and "objective" throughany one of the many interpretive methods of respondingcritics. The text "in itself" possesses "objectivity" only as black ink stains upon paper. Post-structural theorists emphaticallyrefute any claims for an objective text. Fish, for example, insists that "there isn't a text . . . if one means . . . [what Hirsch means] 'an entity which always remains the same from one moment to the next' " (vii). But Fish and other critics who dismiss the objective text nonetheless insist upon an interpretive vigilance in any act of interpretation. Paul de Man asserts that "[l]iterature as well as criticism-the difference between them being delusive-is condemned (or privileged) to be forever the most rigorous and, consequently, the most unreliablelanguagein terms of which man names and transformshimself" (19). BarbaraJohnson, who quotes de Man on this issue, explains that deconstruction"involves a reversal of values, a revaluation of the signifyingfunction of everythingthat, in a sign-basedtheory of meaning, would constitute noise" (279). As "noise" becomes seriously examined for
MarshallAlcorn is a visiting assistantprofessorat Tulane University. He has publishedarticles on Lawrence, Conrad,Coleridge,and on reader-response theory. He is currentlycompletinga book:
Narcissism in the Text: Reflections on the Semiotic Transfer of ldeals.
College English, Volume 49, Number2, February1987 137
its signifyingfunction, it calls into question the stabilityof the text's more apparent "intention" or "meaning." The perception of a textual content ("noise") that is "at war" with the text's more obvious rhetoricalintent is importantfor any theory of rhetoric. But this descriptionof deconstructivepractice also has important implications for understanding textual presence. The presence of "noise" in text implies two things. First, the notion of a dependablyencoded authorial intention disappears. But second, this unstable text offers, through the medium of "noise," clear, convincing, and verifiable-one might say "objective"-evidence of its inner incoherence. Post-structuralmethodology is, on the one hand, more relativistic or, from a traditionalperspective, more "subjective" than traditionalcriticism. Yet on the other hand, it is also more rigorous, far-sighted,and attentive to signifiersthan traditionalcriticism. This paradoxicalavowal of interpretiverigor and denial of textual presence is worth careful examination.If post-structural methodology is shunnedby many critics, it is shunnednot because it lacks clarity, method, and rigor (properties in fact usually allied with objectivity); it is shunned because somehow it seems more irresponsible.It seems, in a word, "narcissistic," and narcissisticmodes of readingseem to deny objective texts. Post-structuralist theory seems narcissisticbecause it makes increasinglyemclaims about the importanceof the reader's role in interpretingor conphatic stituting the text. Signifiersin the text do not disappear,but the thing signified by the signifiers becomes a product of reading strategies. Michel Foucault, for example, asserts that those "aspects of an individualwhich we designate as an author . . . are projections . . . of our way of handling texts" (127). Roland Barthes argues that "the birth of the reader must be at the cost of the death of the author" (148). Traditionalcritics recoil from these claims, arguingthat they over-emphasize subjective aspects of interpretation.George Steiner compares Derrida's Glas to the mirrors that "make up the autistic sovereignty of Narcissus," and he argues that Derrida'sreadingof Rousseau "is 'irresponsibility' in the exact, concrete sense of an inhibition of vital response, of a ruin in the arts of reciprocityand felt dialoguebetween text and reader" (8). Hayden White in a similarfashion complainsthat much currentcritical theory serves as mystification, and "[m]ystificationof the text results in the fetishism of writing and the narcissismof the reader" (265). Critical reactions to contemporarytheory frequently, either descriptively or pejoratively, circle about a set of concepts deriving their rhetorical force from the root adjective, "narcissistic." The concept of "narcissism"may be used explicitly in seemingly "descriptive" accounts of deconstructive or readerresponse methodology (see Brenkman); the term may be used with full pejorative intention to describe other critics and their methodologies; and finally, the term may be "bootlegged" into critical commentary,hidden beneath other terms associated with errant subjectivity. Yet however the term appears, it works determinedly for its author to distinguish one mode of aberrant subjectivity from anothermode of enlightenedsubjectivity. Writers who employ the notion of "narcissism," however, normally ignore the complex psychoanalytictheories (and paradoxicalmodes of subjectivity)implied by their concept. The term "narcissism" thus functions rhetorically and
Rhetoric, Projection, and the Authority of the Signifier
not analytically in contemporarycritical practice. Whether used to pronounce judgment on other critical practices or simply to describe or "explain" them, "narcissism" implies a great deal but means nothing very clear or concrete. Narcissism distinguishes "good" subjectivity from "bad" subjectivity, but it does not in fact reveal how one mode of subjectivitydiffersfrom another. As a rhetoricalterm, narcissismimplies a kind of self-indulgent,irresponsible, and insidiously imaginarydomination over any "other" posited in discourse. But critical theory has no clear taxonomy or hierarchythat names the variety and plenitudeof all the "others" of discourse. Since every instance of discourse is selective, every example of discourse always excludes some other. Certainly it is true that post-structuralist theory dominates and subjugates one kind of "other"-the author's apparentintention. But another "other," the signifieritself, is not dominatedor subvertedby the new methodology.Thus it would seem that, while post-structuralist theory is narcissistic in one sense, it is in another sense not narcissistic. Yet again, since the post-structuralistsignifierfunctions as an empty term, a term whose presence is only the consequence of a particular reading context and a particularreader's will to meaning, post-structuralpractice might well seem more narcissisticor solipsisticallyself-absorbedthan traditional practices. Such reasoning only reveals the circularity of our confusion over what narcissismmeans, for the traditionalcritic's assumptionof a meaning exterior to the signifier(in an author'sintentionperhaps)that he himself has securely discovered is certainlyas narcissisticand self-absorbedas post-structural assumptions. These observations, then, suggest that we have as yet no clear model for rigor in criticalresponse, and this confusion is the consequence of our more covert confusion over what we really mean or want to mean by "narcissistic" or errantforces of subjectivity. As long as we are unable to understand the function of narcissistic subjectivity, we will be unable to formulate criteria for accountabilityin reading. Narcissistic activity is an especially intricate,paradoxical,and complicatedphenomenon; critical theory has used the term genericallyand has failed to understand an importantsubset of narcissistic activities essential to the readingprocess. A more thoroughunderstanding these activities may help to disentangle of some apparentlyincompatibleassumptionsembodiedin various criticalperspectives. A fuller understandingof particularnarcissistic processes-narcissistic projection, idealization, and the role of language in narcissistic functions-can offer theoretical tools for defining a post-structuralist notion of textual "objectivity." In the discussion that follows I will clarify some common misconceptions about the nature of narcissismand projectionand employ recent developments in post-Freudian psychoanalytic theory to explain how projective activities are filteredand alteredby a certainnotion of textual objectivity:objectivity as definedby the text's materialsignifiers. The Narcissus myth is always a useful starting point for reflecting upon narcissistic phenomena. For the western world, the history of the concept of "narcissism" is tied to the image of Narcissus looking at himself in the river. Freud's 1914 essay on narcissism implicitlyalludes to the myth when it comments upon the doting self-admiration the narcissist. More recent analysts have used the of
myth to explore subjects of particularinterest to complex psychoanalytic theories of narcissism-subjects such as the narcissist's fear of self-loss in the beloved object and the narcissist's libidinalor sexual "cathexis" of himself. The myth of Narcissus thus continues to be a useful narrativefor reflectingupon the complexity and confusion of narcissistic behavior. But most readers overit simplifythe myth by remembering as a generalexplanationof narcissism. Most of us, I suspect, have a narcissisticunderstanding narcissism. We unof derstandthe concept by means of the myth, but the intelligibilitywe achieve is vitiated by assumptions we project into the most "apparent" of narcissistic mechanisms, the narcissistic gaze. We oversimplify the Narcissus myth when we use it to abstractwhat I shall call a "visual overlay" model for understanding the narcissistic gaze. The visual overlay model of the narcissistic gaze suggests that narcissismis a state of perceptionin which an image in the "real" world is covered up by an image projectedupon it from the self's imaginaryinner world. Accordingto this model, the narcissisticgaze functions to substitute, in a highly imperialisticmanner,one image for another. A "real" image available for all to see in the "external world" is unconsciously hidden and "covered up" by an unreal image, an image manufacturedby a wayward and undisciplined (narof cissistic) subjectivity.With this understanding the narcissisticgaze, an understandingonly too effectively dramatizedby the Narcissus myth, highly distorting modes of perception have become synonymous with projection and have come to characterizethe quintessentialact of a narcissisticsubject. Freud's work lends analytic rigor to the relationshipI have outlined and provides an almost unquestioned authority for establishing an emphatic link between projection and narcissism. In projection, an image is "projected" from the mindjust as an image is "projected"by a mechanicalilluminatingmachine. In projectionone image is covered up by anotherin a purely arbitraryway. The properties of the visual screen receiving the projectiondo not in any significant way determine or influence the characterof the projected image. This analogy was highly suitable for the point Freud wanted to emphasize about the subjective force of distortion.A "real" image-one that is "real," public, and available for all to see-is covered up, and the image that is superimposedis a private one, idiosyncratically manufacturedby the mind of the gazer. Projection involves, as Freud points out, a loss of reality. It occurs when "the original reality-ego, which distinguishedouter and inner by means of a sound objective criterion, changes into a purifiedpleasure-ego,which prizes above all the quality of pleasure" (79). Projectionthus reflects narcissistic self-indulgence;the "narcissistic" self does not want to see a realitythat is potentiallypainful. The Freudiantheory of projectionis more complicatedthan I have outlined, of and it diverges sharplyat many points from the understanding projectionone finds in the work of critics like Norman Holland. Nonetheless, the term projection has become popularand influential.Among reader-responsetheorists in particular, projectionhas come to define the essential mechanismof reading. Reading, thus characterized,seems to be a process whereby a reader's fantasies are substitutedfor the words in the book. Many advocates of reader-responsecriticism assert that readers project their personal stories upon literaturejust as a
Rhetoric, Projection, and the Authority of the Signifier
projection machineprojects movies upon a blank screen. The text, thus characterized, doesn't really exist. Fish has argued that there are no real differences between differenttexts (170), for the words of the text, despite our painstaking attention to them, do not guide or direct our experience. Words, instead, are empty and passive: blank space victimized by the reader's imperialisticwill to meaning. The "pictures" the reader "sees" in the text are determinednot by words, but solely by the film inside the projectionmachine, the fantasies animating the reader. Reading, thus, appears a highly narcissistic, or even solipsistic, activity. Generalizations such as these oversimplify the more recent work of critics like Fish, Bleich, and Holland, but it is clearly the case that as criticaltheory becomes increasinglypreoccupiedwith the reader, the role of the reader seems increasingly identified with essentially projective activities. Even Derrida's concern for the supplement, the slippage of signifiers, and the infective nature of reading contexts is frequentlyappropriated Americancritics as descriptions by of projection. Thus, Fish's claim that "the formal patterns" of literature "are themselves constitutedby an interpretiveact," Bleich's refutationof the "objective paradigm,"and Holland's rejection of a "biactive theory of text-readerinteraction" seem fully consonant with deconstructivetheory and largely support the view of the text as a blank screen animatedby the fantasies of readers (Fish 13; Bleich 98; Holland, "A TransactiveAccount" 198). Like Stanley Fish, Norman Holland is another prominentcritic whose best known work disavows any notion of textual objectivity. In a relatively recent essay on the "transactiveprocess" of reading,Hollanddisputes "New Critical" assumptions about the sovereignty of the text, assumptions that words in the text somehow limit or direct the response of the reader. Words do not direct meaning, Holland argues, because words have no real meaningin themselves. Words mean only insofar as they are consumed by projective processes. Holland's early articles on the readingprocess explicitly characterizereading as an essentially projective activity. In his highly influential PMLA essay, "Unity, Identity, Text, Self," Holland argues that the "overarchingprinciple" of interpretation is that "identity re-creates itself. . . . [A]ll of us, as we read, use the literary work to symbolize and finally to replicate ourselves" (124). The form and unity of the text, Holland suggests, is a function of the reader's highly arbitrary subjectivity. Holland originallybelieved in the "myth" of the objectivityof the text, but he was saved from his folly by empiricalresearch. Throughcarefulempiricalinvestigation of how readers actually read, Holland discovered that the variety of response to a text was enormous. The words in the text did not limit meaning;instead they seemed to proliferatemeaning. In his empiricalsamplingof student responses, Holland discovered the variety of response to literaryworks was so varied that meaningcould not be explainedby the "objective" natureof the literary work. "If you actually collect people's free responses to texts," Holland observes," they simply do not show a uniformcore (fromthe text) and individual variation (from the people). The responses have practicallynothing in common" ("Recoveringthe PurloinedLetter" 366).
Holland has recently modifiedhis theory of reader response to explore throughthe conceptof "feedback"a notionof textualcontrol(see "Psychoearliertheoretical analysisandFilm"),but Holland's positionwas highlyinfluled encedby the workof DavidBleich.Bleich'sarguments Holland conclude to that "storiesdo not have defensemechanisms playsdo not sublimate-but and his people do" ("A TransactiveAccount" 181). Bleich's work, particularly in Assumptions the Studyof Response"in Subjective chapter"Epistemological becauseit seemsto confront is Criticism, especiallyinteresting epistemological For issues in perception carefully. manyreaders,the worksucceedsin making with aestheticperception. Bleich, interpretation For is projection synonymous Bleich argues,do as understood "motivated (97). resymbolization" Signifiers, that acts. Instead,signifiers take on not announce meanings exist beforereading meaningthroughthe motives of readers.Since the motives for this "resymserves to express bolization"are characteristically unconscious, interpretation the unconsciousmotives of the reader.In effect, readingreadsprojection;it in he distinctive texts. "Theobjectof attention," argues, does not readanything of "is not the itemitselfbutis the response thosewho observeit" (98).Bleich's whatI havetermeda "visual thus of perception resembles description aesthetic "Motivated serves as Bleich's resymbolization" overlay"modelof projection. termfor projection. substitute issuescrucial an understanding textual for of Bleich'sengaging theoryignores claims.Bleich correctlyassertsthat all objectivityand conflatestwo different But imagesare subjective. he is led too hastilyfromthis assertionto another, This latterclaimrequires all different assertion: imagesare equallysubjective. denominator" bearswitness It carefulconsideration. suggeststhatno "common different of to a commonsourceinforming readings the same text. It suggests of exists amongdifferent thatno "familyresemblance" readings the sametext. to cannotbe "improved" referring the It suggeststhataestheticperception by cannot(in that the sourceof the perception, text. It suggests perception original Lacan'sterms)asymptotically "reality."Responsesto texts instead approach the proliferateendlessly and arbitrarily; differencesbetween readingsof the texts. betweendifferent sametext areequalto the differences II theorieslies in theirpropensity failurein manyreader-response The conceptual in as to see all subjective"distortions" perception equalin degreeand effect. disof Thisfailureis the consequence another largerfailureto graspimportant the protinctionsin subjective processes.Textsare not blankscreensreflecting
jections of a reader;textual signifiersdo things to projections. Readers who fail to encounter the signifiersof a text fail to encounter the text. Reader-response theory frequently assumes that because there can be no "correct" reading of signifiers, there can be no sense of accountabilityin reading. Such an assumption is naive. The "material" signifier does not guarantee a stable referential
Rhetoric, Projection, and the Authority of the Signifier
event, but does guaranteea stable perceptualevent that plays an importantrole in conscious, unconscious, and narcissistic readingprocesses. The materialsignifier offers itself as a criterionfor accountabilityin textual response; readings that fail to encounterthe actual signifiersof a text are failuresin reading. Bleich and others feel that since words can mean anything, the text does not offer anythingsignificantfor acts of attention. Such a view conflates the issue of objectivity with the issue of acceptability. Certainlyit is true that since words can mean many things, it is impossible to define a "correct" textual interpretation. But concerns for objectivity are not equivalentto concerns for "correct" interpretations.Acceptable or "correct" readingsof texts are measurementsof the satisfactions and rewards of interpretations,and acceptable readings need not be exactingly responsible to objective texts. Editors, operating within the ideological context of their historicalmoment, will publishcriticismthey find rewarding and reject criticism they find unpersuasive, unoriginal,or uninspiring. Since acceptable interpretations posit certain meanings as rewarding, acceptability is always a function of ideologicaland rhetoricalrelevance. Objectivityis a differentmatter. A judgment about objectivity in reading reflects a judgment made about a reader's encounterwith the materialsignifiers,or sequence of signifiersof a text. It does not reflect any judgment whatever of the complex and compelling systems of meanings the reader discovers through those signifiers. No critic disputes the fact that signifiers are objectively present in the text. Holland grants that there is text insofar as there are "words-on-the-page" (Five Readers Reading 286), and even Fish grantsthat signifiersare materialobjects ("markson the page") and that they are materiallypresent, thus objectively present for reading (173). Nonetheless, post-structuralist theory insists, like Fish, that "there isn't a text that remains the same from one moment to the next" (vii). Yet poststructural theory requires something in the text-the signifier-to remain the same. Thus, paradoxically,post-structuralpractice denies the iterabilityof the text, but requiresthe iterabilityof signifiers.The text disappears,but the letters and words of the text stand out ruggedly(in the form of noise) as unflinchingwitnesses to the betrayal of meaning.The objectivity of the text disappearsin one sense, but it reappearsagain with unmistakableinsistence in another sense. It reappearsin the materialityof the signifiersthat constitutethe text. The signifieris clearly materiallypresent, and "reading"is distinct from hallucination insofar as it describes a sequentialencounter with signifiers. But the objectivity of text, as it is understoodin these terms, is generally dismissed as theoreticallyinconsequential.Bleich, for example, denies the importanceof the material signifier and the "nominal meanings of words" because these things cannot "mean" anythingexcept themselves, and this "meaning" can never be transferredfrom one person to another except by repetition of mere dumb signifiers (111). This dismissal of dumb signifiers,however, is ill-considered. Most teachers would agree that there is a style of reading that fails to see the text, fails to register the material presence of the signifier and the sequence of signifiers that comprise a text. Many readings of texts fail to recognize major events in narrativeand fail to register the particularity vocabularydescribing of
these events. It is clear that these readingsare failuresin perceptionand failures to respond to the "objective" text. The repetitive encounter with dumb signifiers is not, then, as Bleich would have it, theoreticallyinconsequential. It is time to give the signifierits due. We must recognize that an encounter with the dumb signifieris essential to reading. To appreciatethe importanceof the dumb signifier(that can mean nothing because it can mean anything),we must explore some recent psychoanalyticobservations about language.There is an increasingrecognitionamongpsychoanalytic theorists that languageis somehow central to what is called the psychoanalytic process. MarshallEdelson, for example, suggests that "[p]erhapsonly a theory of language can begin to acount adequately for the complex phenomena embraced by psychoanalysis" (28). Edelson's own book carefully probes relationships between forms of literary language and forms of psychoanalytic exchange. Meredith Skura, a literary critic trained in psychoanalysis, explores similaritiesbetween literary experience and psychoanalytic experience. She argues that the "complex ways in which literarytexts elaborate and call attention to the play of consciousness . . . have a parallelin the way that these phenomena are handled in analysis-in the moments of integrationand insight" (11-12). The most widely influentialtheorist of the relations between language and psychoanalysis, however, is Jacques Lacan. Lacan argues quite directly that the crucialfunction of the psychoanalyticcure is the linguisticfunction. Lacan's theories postulate a self "anchored" by certain signifiers but also "split" and divided as it is drawntoward other signifiers.Lacan's theory of linguistic anchoringpoints definingthe self-system explains "the dominance of the letter in the dramatic transformation that dialogue can effect in the subject" of (Ecrits 154). Transformations the self, in effect, occur because of transformations in ways of speaking. What is importantfor critical theory here is that the materialityand the particularityof the signifierare crucial in the self's anchoring Lacan's examples of the role of lanand its splitting-and in its transformations. guage in the unconscious repeatedly call attentionto punning,homophony, and lexical allusion. In a more general way, Lacan argues that "the slightest alteration between man and the signifierchanges the whole course of history by modifying the moorings that anchor his being" (Ecrits 174). For Lacan, rhetorical pressure is a consequence of the self's being tethered to the particularity of words. Lacanian theory would support the generalizationthat close reading of of particularsignifiersmatters because the particularity the signifierexerts an inexorable rhetoricalpressureupon the matter of the self. Lacan's own writings perhaps fail to communicatethe portentousness of his discovery of the signifier's materiality. But it is clear from his comments that words and networks of words in their particularity operate for consciousness something like the bones of a skeletal system. They enable certain possibilities, and they disable others. Words are in a certain sense evasive and immaterial. Yet in another sense they are quite material, and they constitute nothing less than the true body of the self. "Languageis not immaterial,"he says. "It is a subtle body, but body it is. Words are trappedin all the corporeal images that
Rhetoric, Projection, and the Authority of the Signifier
captivate the subject; they may make the hysteric 'pregnant'" (Ecrits 87). Lacan's point here is that the symptoms manifestin psychological disorders, symptoms that bind, restrict, or paralyzeparts of the body and mind, thus preventing certain actions and thoughts, have their groundin the airy substance of words. Words thus exert a biological control over the body. Words, in all their seeming innocence and neutrality, can lead to paralysis, hysteria, hallucination, false pregnancy. Second, it is not simply that words hold sway over biological funcin tions, but that this control is maintainedby an unnervingparticularity words. The unconscious seems to admit to no paraphrase and no synonyms. Words matter in their particularmaterial signifying substance-both as marks and as sounds. Such acute consciousness of verbal particularityis importantfor Lacan because words function in psychoanalyticdiscourse by exploiting a distinctly literary function. "Poetry," Lacan writes, "is the creation of a subject assuming a new order of symbolic relation to the world" (Les psychoses 91). Since the linguistic creativity of poetry reflects the creative remakingof the self, Lacan argues that "psychoanalysis manipulates the poetic function" (Les ecrits techniques de Freud 106). Lacan's own writing and speech bristles with literary character.His languagecalls attentionto itself in a variety of ways: homophony, idiosyncraticsyntax, phonemicallusions, and a wide variety of tropes. But Lacan is not alone in emphasizingthe theoreticalimportanceof this "material"aspect of language. Edelson, for example, observes, "Words are the object through which one seeks for a way to handle the unconscious. Not even the meaningof the words, but words in their flesh, in their materialaspect" (11). If it is clear that the material presence of words matters enormously to the functions of the self, and especially to the unconsciousfunctions of the self, then it should also be clear that criticaltheory needs to examine how projective activities are animatedby the signifier'smateriality.If texts are not blank screens for projections, if instead projections are somehow "filtered" and networked by a text's signifiers,then we must find effective terms to describe this process. The transference theory of reader response that Mark Bracher and I proposed recently in "Literature,Psychoanalysis, and the Re-Formationof the Self: A New Direction for Reader-Response Theory" suggests that the "narcissistic" and projective processes of readingare modifiedby textual encounters. Bracherand I argue that mourningand idealizationare two psychologicalprocesses that alter a reader as he or she assimilates a text. To furtherappreciatethe complexity of narcissistic forces operatingin reading, however, we must analyze the relation between projection and idealization. Unfortunately,the term "projection" can refer to two distinctly different kinds of activity. On the one hand, projection can refer to a subjective replacingor deleting of an objectively present signifier (an avoidance of the perceptiblefeatures of the object). On the other hand, projection can refer to the subjective interpretingand contextualizingof signifiers actually encountered. For clarity, I will term the projective covering up of the text "projective occlusion," and the reworkingof signification"projective idealization."
The "projective occlusion" of the text designates the ability of projective forces to "cover up" the "objective" presence of texts. But not all projectionin reading is projective occlusion. Projective occlusion is quite different from "projective idealization," and much readingreflects processes of projective idealization ratherthan processes of projectiveocclusion. From a psychoanalytic perspective, the question of objectivity is equivalent to the question of whether projectionplaces somethinginto the text in an arbitrary and imperialisticmanner or uses some thing in the text as a stimulus for deriving (usually throughtropologicalprocesses of assimilation)personal significance. Both modes of response can occur-in both readingand perception. But these two things that can occur (and are often confused with each other) are not the same thing. Projective occlusion is not a personal reworkingof signifiersalready present in the text but an avoidance of signifiers,especially an avoidance of signifiersthat challenge one's values and sense of self. Projective idealization encounterspossibilities for meaningthat it organizesand reworks;projective occlusion denies the presence of some potential meaning that needs to be taken into account. Projectiveocclusion, further,is not equal among all readers and in it all interpretations; is a true misreadingof the text. It is inattentivenessdue to a variety of possible causes: fatigue, haste, distraction, and psychological defensiveness. As a defense, projective occlusion is characterizedby avoidance, denial, memory loss, and wholesale, arbitrary,and unconscious replacements of one signifieror set of signifiersfor another. We should acknowledgethat the unconscious replacementof signifiersis a component in everyone's reading. But it is a component we test by dialogue and attempt to subdue by re-reading.While re-reading cannot recover a "correct" interpretive reading, it can encounter more fully the particularityof signifiersand bringthem into the purview of consciousness, reworking them and creatively organizingtheir effect. Re-reading hence makes readinga discipline, not an opportunityfor hallucinationor visual slips of perception. In contrast to projective occlusion, projective idealization acknowledges the presence of the sequence of materialsignifiersconstitutingthe text. Rather than ignoringthe signifierin deference to its own fantasies, it attempts to use the signifiers to embody, articulate, elaborate, and even recursively reflect upon its values. Literaturecan act like a mirrorto reflect the arbitraryfantasies of the reader's mind, but most of us do not see only our own fantasies in literature.We use literatureto embody in particularways our "idealizations" (even if our idealization requires the deconstruction of idealizations). A "correct" reading of signifiersformulates a "projective idealization." When we respond fully to the material signifiersof the text, we use the text to embody in signifiersour partly conscious and partlyunconscious values. Further,we gain satisfactionfrom this process because the idealizationis made more concrete (and more representable) by passing throughwhat Lacan calls the "defiles of the signifier."Projective idealizationof the text, thus, refers to the mannerin which the ideals of the self are defiled by the signifiersof the text. Projective idealizationrefers to the process by which materialfrom the internalworld encounters materialin the external world and becomes modifiedby it. It is, of course, difficultto predict how
Rhetoric, Projection, and the Authority of the Signifier
any one signifierwill influencea reader. But for the present purposes of theory we don't need to be able to predict the influenceof a signifier.There may come a time when reader-responsetheory can examine basic styles of tropologicalresponses to signifyingchains, but for now we can content ourselves with a few general observations. Projective idealization, as an interpretiveresponse to the signifier,bears witness to attitudinalrecognitionand formulation:perhapsadmiration, denial, or assertion of value. Projective idealizationpasses throughnarcissistic mechanisms of thought to orient the self toward general themes or values that the signifierssuggest. Indeed, the particularity that materialsignifiers lend to this orientationmay well be the essence of the literaryeffect. But this effect, this projective idealization, however creative it may be, depends upon an encounter with the materialpresence of language, the materialpresence of the signifier,resisting, defiling,and exerting significantforce upon both the personal content of projection and the general linguistic field that orients any reader toward "reality." An appreciationof the importanceof the materialsignifierin "defiling"projective mechanismscan benefitcriticaltheory in a variety of ways. First, the material presence of the signifierserves as a necessary theoreticalpostulate for explaining what readers negotiate when they work together to seek meaning. Second, the materialpresence of the signifier(in whatevermeaningit may seem to have) serves as a resistance to projectiveocclusion. Regardlessof what words mean, their function in readingat this level is important.Words, by their material presence, color and influence the content of projection. The materialityof words and, more importantly,the literary effects made possible by this materiality (sounds, rhythm, puns, phonemic allusions, indeterminate meanings, onomatopoeia,visual patterns, etc.) make readingmore like social dialogue, and less like private hallucination. Because the reader must encounter and meaningfully "process" the signifier, critical theory must claim responsibility to an "objective text." Texts are not purely the productof a reader'sprojection.Texts have particular propertiesof their own. These particular properties,however, do not exist as categories of referentialmeaning;they exist as somethingwe might call rhetoric. Our discussion of the linguistic structure of the self encourages us to hypothesize a relation between the projective forces broughtto bear upon the text by the reader and the rhetoricalforces broughtto bear upon the readerby the text. In reading, projection has a boomerangeffect: projective moments become introjective momentsas the self invests itself narcissisticallyin the particularityof the signifiersit encounters. Words in their materialpresence assume certain energies from the projectiveforces workingupon them, but in doing so they defile those energies by entrapping them in a particularity of words. Words in this mannerabsorb projectiveforces and deflect them, thereby exerting "rhetorical" pressureupon self-functions. Rhetoricis a force empoweredby the linguisticsubstructureof the self. If the self is essentially disunifiedor if it is, as Cullerputs it, "dissolved as its various functions are ascribed to impersonal systems that operate through it" (33), it does not then follow that critical theory can abandonthe concept of the "self."
Instead, another question arises. How do we understandthe larger system within which these various self-functionsvie for control of thought, affect, and behavior? And further, what affects the disunified workings of any of these self subsystems in any particularway? Clearly, this force is termed many things by the differentdisciplines that feel a need to account for its functioning.But in the area of critical theory a good name for this force is rhetoric. We normally assume rhetoric is a force in languagethat manipulatesemotion in humans. But in a certain sense this force can never be in language;language(from a purely "objective" position) is only a collection of dumb signifierssubject to endless displacement. Rhetoric, then, is not in language. It is in selves. It is a force in the linguisticconstitutionof the self. To put the issue in better terms, rhetoric, as we discover it in texts, always bears witness to power whereby social and psychological forces in language"position" selves in relationto affect. Rhetoric is the umbilicusof a text, the point at which words come into being for the being of selves. Rhetoric, in this sense, has a presence prior to interpretive meaning. Rhetoric designates focal points for shared perceptions about texts, and readers share rhetoricalresponses to a text before they agree or disagree over what a text means. Further, since the forces of rhetorical organization found in texts reflect similar forces of rhetorical organization found in selves, the interactionsbetween text and self will always be reciprocal, dynamic, and ceaselessly ongoing as significationis ceaselessly consumed and displaced by self-functions. The concept of rhetoric, then, is a theoreticaljuncture where literarytheory, rhetoricaltheory, and psychoanalytictheory should converge. When we encounter rhetoric in texts, we encounter the forces attached to words that generate, employ, or "bind" emotion. This manipulationof emotion of amountsto nothingless than a manipulation the componentsof the self. Rhetoric designates the forces in the linguistic constructionof the self that structure affect. In reading, these forces transformprojective energies by verbally recontextualizing them. Through such verbal recontextualization, affect itself becomes susceptible to restructuralization. Further, since the unconscious strucof the self is built upon a particularity of words, rhetoric also depends turing power. Words in the particularity upon verbal particularityfor its restructuring of their visual and phonemic properties (punning, homophony, onomatopoeia, phonemic allusion, visual metaphoricity)and in their associative and referential possibilities become "rhetoricalperformers"for selves. As particularwords and particular signifying chains of words become animated by projective forces, words manipulate affect. But words work (just as ad writers and poets insist) precisely in their particularity. of It follows, then, that readingsattentive to the particularity words, attentive to words as rhetoricalperformers,can be especially rewarding.(I do not want to are suggest here that referentialapproximations non-functional.I merely want to emphasize what English teachers have always claimed-attention to the concrete text matters.) Throughsuch an attentiveness to the particularityof words, a reader may encounter new possibilities for perceiving and organizingthe disunificationsof the self. This possibility may be presented in several ways. First,
Rhetoric, Projection, and the Authority of the Signifier
the reader may encounter the particular signifiers that anchor his or her subjectivity, but encounterthese same signifierslinked to differentsystems of affect and cognition. In this case, the reader's self-system may not change, but values held by the self-system may. The self extends its approvalor understandingto new areas of experience. Second, however, a self-system may undergo radical change as deep stratas of affect and repression become evoked and then networked and anchoredto new verbal systems. In this case the self-structuremay itself change. This change, indeed, may be experiencedas a kind of "freedom." But critics believing in linguisticdeterminismmight well call this freedom not a choice but an effect producedwhen consciousness is subjectedto certainrhetorof ical pressures from the particularity signifiers. A broad understanding the psychoanalyticrole of the signifierin interpretaof tion should embrace both psychological and sociological perspectives. Sociological critics have, of course, long claimed that selves are products of the same processes that produce texts. From a sociological perspective, words, in the particularityof their effects upon consciousness, are culturalartifacts that constitute selves aroundsystems of culturallysanctifiedvalue. Generallyspeaking, however, sociological critics have avoided the complexities of psychoanalytic issues and denied personal, creative, and adversarialrelations between texts and readers. Fish speaks for many sociological critics when he asserts: If selvesareconstituted the waysof thinking seeingthatinhere socialorand in by and selves in turnconstitute texts according to ganizations, if these constituted thesesameways,thentherecanbe no adversary text between andself relationship becausethey are necessarily relatedproducts the samecognitivepossibilities. of (336) It may indeed be true that both "selves and texts" are produced "by ways of thinking and seeing that inhere in social organizations."But such productions should not rule out adversarialrelations. Sociological critical theory, by ruling out the presence of adversarialrelations, denies or oversimplifies a variety of importantquestions. How are readers ever persuadedof a better mode of interpretation?Whatfactors in texts can instigatepersonaland culturalchange? If, as Jameson, Lacan, and Kristeva suggest, texts are always conflictually heterogenous and the social constitution of the self is also always imperfect, then change can be explained as the result of a particularset of psychological forces-primary processes-responding to and in texts. Because primaryprocess thinking reveals the pressure of biological drives rebelling against social norms and constrictions, primaryprocess thinkingdoes not reflect the homogenous code of a homeostaticlinguisticand social self "system." Primaryprocess thinkingrebels against and deflects linguistic social norms. In this manner, such thinkinganimates the text with revolutionaryrhetoricalpotency. It is a style of thinking that exacerbates the social heterogeny of the self to experiment with different metaphors and different combinationsof linguistic effect. It explores different values in differentlinguistic effects, and throughthe linguistic dislocations it promotes, primaryprocess thinkingworks as a vital force in generating change. Further, since idealizing projection(the very mode of subjectivity that
processes and digests textuality) is influenced throughan upwelling of primary process affect, then personal change and culturalchange might well be a function of idealist projectionoperatingupon materialsignifiers. An acknowledgment of these conflicting forces at work behind all forms of idealizingprojectionin humanculture should improve sociological criticism. Idealizing projection explains how the social strands of conflict present in critical disagreementsreflect, but also and more importantlywork upon, components in conflict within the inner network of the self. Jameson's work already explores the psychological reflectionsof social heterogeneity.Jamesonargues in a general way that there are dissensions in the idiolects of every culture, and that these dissensions introducerupturesin the systems of words that bind selves together. of But we should bringmore conceptual elasticity to our understanding the way selves are formed and deformedby sociological structures.We might appreciate, for example, the psychological meaning of rupturesin linguistic self-functions. Such rupturesare not mere fracturesin the structureof language;they mark an often agonizing point of tension and conflict within a self. Such ruptures allow something that from another critical perspective might be called "interiority." Interiority,when it is representablein signifiers, can be a socially constructive mode of suffering.It can be "productive"in the full Marxist sense of the term. Throughthe materialsignifiersit produces, such interiorconflict offers a ground or impetus for both change and creativity. Changethus has its origin in the frictional and textual interfacebetween selves and culture. Often, change begins as a revolutionaryrhetoricalresponse to a culturallyprescribedrhetoricaleffect. Conflict, especially conflict in the overdeterminedmeanings of any particular word, is in a certain sense the raison d'etre of critical theory. Many schools of criticism, however, attempt to deal with conflict by outlawing certain possibilities of meaning. The consequence of this is that meaning becomes a funcshould avoid repression tion of repression. An adequatetheory of interpretation and privilege conflict (on both the social and the intrapersonal level) as the seminal principleof the hermeneuticendeavor. A psychoanalytic attention to the materiality and particularity of words as rhetoricalperformersoffers this possibility. For the psychoanalyst, hermeneutic conflict betrays the struggle of overdetermined meanings. Overdetermined meanings mark a point where the subsystems of the self suffer a failure of synthesis. Particularwords, chargedwith overdeterminedmeaning, reveal the combat of repressed value systems. These particularwords draw attention to themselves in "literary" ways and betray unconscious values that oppose the dominant values of the self's discourse. This betrayal may surface under standard interpretive procedures as a logical contradiction and appear quite puzzling. From a psychoanalyticperspective, however, this evidence of conflict and can overdetermination mark the beginningof useful realizationsabout opposing motives and values. The psychoanalyticperspective thus offers anotheremployment of words: words can probe and reveal the knots of conflict and contradiction within the self. Words bring into the purview of consciousness the ragged seams of linguistic self-constitution.
Rhetoric, Projection, and the Authority of the Signifier
Like psychoanalytic experience, reading can enable the registering in conbut sciousness of multideterminate particularizedsystems of signifiers exerting their unconscious pressure upon another overdeterminedsystem of signifiersthe self. Readingcan register in one textual "message" multipleand conflictual rhetoricalintent. More importantly,readingattentive to the pluralityof possible rhetoricalorganizationsinherent in systems of significationcan produce a frictional sliding in consciousness between disparate and unsynthesized units of rhetoric and intelligibility. In texts, as well as in selves, these unsynthesized units of intelligibilityor "meanings"normallyfind their proper subordinationin relationto dominantmasterterms that exert their various, conflictual,overdetermined, but nonetheless orderingforce across a variety of subordinatedsymbolic networks. These master terms normallyanchor the reader and anchor the text, thus subduing an otherwise chaotic strugglebetween dominantand adversarial culturalideology, dominantand adversarialpersonal idiolects. But an encounter with the conflictual seams (hitherto unnoticed) in the unsynthesized units of meaning in texts can dislodge the "meaning" of repression and liberate value from ideological denial. The materiality and particularityof words therefore not only constitute the self, but offer themselves in their particularityas tools (best used in moments when a breadthof vision has been achieved by an accumulationof reference)for redefining and renegotiating self-constitution. Thus, we might postulate that "good" readings of texts productively "defile" the reader's projections by means of the materialityof signifiers.The readerresponds actively to the objective and material signifiers of the text, seeking to rework the content of his or her projection. This activity is possible because, althoughwe do indeed always perceive throughthe decodingeffects of our own personalreadingstrategies, we also refer to material signifiers of the text in all their particular polyvalent qualities (and especially in their polyvalent qualities), seeking to modify, enlarge, and if necessary reorganizethe habitualdecoding strategiesthat normally allow us to "consume" and dispose of signifiers.Even if Hollandis rightin arguing that our reading strategies tend to be univocal and ideological as a result of defensive identity themes, it still remainstrue that the disciplineof textual study traditionallydemands, in the response to signifiers,more subtlety than the mechanical applicationof univocal decoding systems. We are encouraged to read reflectively. We are encouragedto recover from repressionsignifyingcodes that are latent and not dominantin ourselves and in the texts we read. This ethics of readingis not of course always successful, but it is nonetheless outspokenly idealized. This idealization defines the humanisticpromise of reading. It assumes that reading is not a simple process whereby the unity of a subject responds to the unit of a text. Instead, readingreflects the rhetoricalfriction producedat the intersection between the self (as a pluralityof textual codes) encounteringthe orientingand disorienting"order" of anotherpluralityof codes (the text). If the words of a text are "rhetoricalperformers,"manipulators affect that of bind selves to systems of value (and not merely representationsof a writer's or reader's experience), certain traditionalendeavors of literarycriticism assume a can be seen as legitimate role. First, literary canons (in all their arbitrariness)
collections of systems of signifierssituated in historicalmoments that either can have or have had significantrhetoricaleffect upon readersand cultural systems. Some texts, perhaps, for some historical periods may be better "performers" than others. Second, in terms of developing a student's rhetoricalexpertise and facility with language, it would be profitableto seek to know the various verbal strategies of any epoch's rhetoric. Third, it will be useful to seek to know an author's apparentintention-not in orderto know the meaningof the text as equivalent to the intention, but to understandmore fully the rhetoricalperformanceof the text in light of the professed intention.
III A reader who has found my extended discussion of projective idealizationplausible might at this point pause, reflect, and object that my originalclaims about the existence of an objective text must be illusory. Each person, after all, perceives the text differentlyin his or her idealization. Since texts come into being through the perceptions of readers, there can be no objective text. This objection, however, can be overcome if we make two distinctions:first, between perception of the material signifiers and interpretation of those same signifiers (which I have discussed at some length) and second, between unlimited, infinite sets of elements and limited (mutuallyexclusive), infinitesets of elements. An imprecise understandingof infinity has supported argumentsprecluding of textual objectivity. A clearer understanding infinitycan remedy this mistake. The objectivity of the text is an objective perceptualevent that produces an infiThe existence of an infinitenumberof cornite but limited set of interpretations. rect readings for a text does not mean that the text can mean anything. It only means that there is a certaininfinite,but limited, set of possible readings. The concept of a limited infinitymay, at first, be puzzling. But it shouldn'tbe. In mathematics, there are a variety of formulas that can designate a variety of infinitesets of numbers.The elements containedwithin each set of numbersdesignatedby a particularformulaare infinite. But two differentformulascan designate two very different, mutually exclusive, infinite sets of elements. The elements within the two infinite but mutually exclusive sets of numbers are quite different. Thus, it is possible to describe a set of elements that is both infinite and limited. The set of odd numbers,for example, is infinite, yet also limited: it does not contain the infiniteset of even numbers.It is infinitebecause it contains an infinite numberof numbers. But it is also limited because it does not contain the infiniteset of even numbers. The point of all this is that an infinityof differentreaders can produce an infinity of different readings of Othello, but such an infinite set of interpretations would exclude the infinite number of interpretations produced by readings of Moby Dick. Anyone who doubts such a claim should write a good, close thirtypage interpretationof Othello, change all the names to reflect names in Moby
Rhetoric, Projection, and the Authority of the Signifier
Dick, and try to teach or publish such an interpretationof Othello as an interpretationof Moby Dick. The result, of course, would be a failure in communication. Of course, there is always a level of abstractionat which any text will resemble any other text. Both Othello and Moby Dick reflect quests for truth. But such similaritydoes not make the two texts equivalent.High levels of textual abstraction should never be equated with textual meanings,for they ignore or subordinate the minute particularityof the text that gives meaning to the abstraction. The text, like a particular formula in math, offers infinite possibilities for meaning,but it also limits and excludes other possibilities. There are a variety of ways to talk about how interpretive possibilities are both limited and infinite, and I will review some of them. Although words can in principle refer to anything (and seem to legitimatean absolute infinityof possibility), words (and systems of words) cannot mobilize and organizeaffect in unlimitedways. Each cultural system legitimates and offers support for a relatively limited set of personality paradigms.Further,each culture offers distinct models for organizthat organiing affect and distinct metaphoricalorchestrationsfor manipulating zation. Each culturalsystem, we might say, possesses its own unique repertoire of rhetoric. It authorizesand endows with potency its own system of metaphors and thereby limits and preconditionscertain possibilities for the constitution of the self. Thus, although there are new and "creative" organizations of affect emergingin most cultures, affect is itself limitedby the biologicalconstitutionof the organismand the signifyingstructuresof affect made availableby a culture. The ability of projectionsand associations to be meaningfulfor other readers functions as anotherlimitationupon meaning. Usually the reader's power of association is taken as evidence for the impossibilityof definingan objective text. One could easily arguethe opposite. Althoughit seems possible to free-associate infinitelyin relation to any one word of the text, it is quite a differentmatter to tie clusters of free associations into a system that meaningfullyserves others as an interpretation a secondary system of words. The ability of free association of to produce meaningdoesn't make the text disappear;it instead limits the meanings the text can produce. First, meaningis the effect of the materialsignifier's filteringand networkinga reader'sprojection.Second, a reader's personal "rhetorical performance"of a text requiresa largercontext of meaningin orderto be intended for public consumptionwould, however meaningful.An interpretation a rhetorically effective contextual placing of individuindividualistic, require alistic responses into communalpatterns of discourse. This communal contextualization would be necessary for the rhetoric of the interpretiveperformance to have its effect. It would vouchsafe the possibility of somethingthat might be termed "intersubjectivetextual transactions": the potency of textual codes to modify the affective organization of another textual code, the self. Thus, the need for "intersubjectively transactive" patterns of discourse would restrain what might otherwise be a chaos of projectionand free association. Fish and Cullerare also partlycorrect in insistingupon the limitingand directive effects of interpretivecompetence. Readers are constrained in their interpretations by the interpretiveconventions of their culture. Meaning, clearly, is
meaningful only in a social and cultural context. But a responsible theory of reader response should acknowledgethat both readers and authors, though limited with respect to certain semiotic operations, can transcend the signifying horizons of their culture. Indeed, an original and idiosyncratic discovery of meaning on the part of an aberrantly signifying author can be so rhetorically forceful as to assume a dominantrole in culture. Many critics (e.g., Iser) assume that this is a prime function of the artist: to recover the repressed that has been excluded by conventionalmeaning. But if an authorcan invent a meaningfulyet "aberrant"signification,a readercan do likewise. Reading, like writing, can produce "private" or idiosyncraticmeaning, but an acknowledgmentand appreciationof this idiosyncratic generation of meaning does not mean that there is no objective text. Returning to the analogy with mathematics,we mightcomparea text to an algebraicformula.There are infinite ways of substituting numbers for the formula. But the formula's "objective" structure controls the meaningfulnessof the calculations derived from it. The formula, objectively considered, may mean somethingby itself, but this meaning is flat and empty. The "objective" formulais only a pure expression of a structure of relations. The full meaningof the formulabecomes practicallysignificant only when certainnumbersare fed into it to derive useful calculations. Texts are not objectively present as units of meaning, but texts are objectively present as meaningfulrelationshipsstimulatedand rhetorically sustained by material signifiers. By themselves, texts are only notations for relationships latent in signifiers.But like algebraicformulas,texts "work" when they are processed by another level of signification. Often, however, texts are not distinguished from the meaningsor rhetoricalperformancesthat they produce. Arguments about what a text is are conflated with arguments over what system of meaning should be used to process the text. But there have been attempts to counter these assumptions.Hallet Smith's "Introduction"to the TwentiethCentury Interpretations of the Tempest, for example, compares the play to a kaleidoscope and laments attemptsto freeze its many-facetedsignifyingpotency: so Becausethe plot is so simpleandthe characters far fromcomplex,the critics for to findit difficult account thegreateffecttheplayhasuponthem.In desperation inor philosophical, psychological religious, allegorical, they turnto biographical, of terpretations the play.(9) Biographical,allegorical,religious, philosophical,or psychological interpretations are systems of explanationfed into the mill wheel of the play's signifying structure, its algebraicformula. Throughsuch signifyingtransformations,broad and general explanatoryabstractionsbecome illustratedby particularelements found in the work. The result of these signifying conversions fulfill idealizing projections and provide stuffing for the pages of countless journals and books. But as Smith's metaphorimplies, none of these productions(and we can imagine an infinite numberof them reflectingan infinitenumberof theoretical positions) should be equatedwith the text. All of these productionsare the effects of projective idealizations. Projective idealization is a meaning effect, a rhetoricaleffect, produced by an angle of vision. Projectiveidealizationuses an angle of vision to produce a personal organization of the rhetoricaleffects of signifiers.Projective idealization, further, can
Rhetoric, Projection, and the Authority of the Signifier
rework the particulargivens of culturalsignifiersin an infinitenumberof ways. Reading and interpretation thus offer infinite possibilities for difference, and Holland is indeed correct in suggesting that the words of the text do not limit meaning--they proliferatemeaning. But if a text offers infinite possibilities for interpretation,this does not mean that the words do not limit, inform, or direct meaning. Smith's metaphor of the text as a kaleidoscope might be improved upon by comparingthe text to a crystal. The crystal, unlike the kaleidoscope, has three in dimensions. There is always a particularity its depth (the other side we cannot of see) that influences, by its manipulation light, the surface we do see. The ancrystal can be observed from many different gular faces of a three-dimensional angles and sketched as a two-dimensionalpattern. The crystal will authorize a differentsketch when seen as two-dimensionalfrom any of a variety of perspectives. Yet all the infinitelydifferent,two-dimensionalpatternsproducedby infinitely small changes in rotationsof the viewing angle have their groundin a single, stable three-dimensional structure. So also with criticism: to insure the systematic unity in argumentation, writers can write only from a twodimensional perspective; otherwise central concepts lose their clarity as they shift aroundthe rotationof changingangles of vision. Nevertheless, most critics can appreciate the beauty of things they discuss from three dimensions. They can see how other people get the ideas they do when they look at something from a "differentangle." Criticscan even appreciatethe "depth" perspective of the art object: they can appreciatethe beauty of it precisely in its ability to produce multiple, intricate,and compellingeffects from multipleangles of vision. The signifiers of a text, then, allow an infinity of different perspectives and different rhetorical performances, but this does not mean that the text has no distinctive qualities of its own. A good interpretation a text offers other readof ers a persuasive "network of projections" stimulatedand rhetoricallysustained by close attentionto the particularpropertiesof a referentiallyunstable, but not a perceptually unstable, verbal artifact. Fish's refutationof the objective text, then, is misguided. There is an objective text that "remainsthe same from one moment to the next." We cannot say "objectively" what the text means, but the objective text, like a mathematical formula, can be used to assess the propriety of the things we do with it. It can be used to evaluate the amount of projective occlusion and ungroundedfree association present in any interpretation. The language of a text, precisely in all its imaginablepossibilities, conditions and influences the set of associations the text can legitimately support. If words are rhetoricalperformers,the meaningof texts may be the reader's creation. But this creation cannot be legitimatelyproducedif the readerfails to hear all the performers.Just as music can be performedin differentkeys and different tempos, texts can be read in differenttones and from differentperspectives. The differences in performanceare often great, but the music performedmust reflect a set of musical relationshipssuggestedby the musical notes. If both good and acceptable interpretations depend upon a reader's organization of the force of rhetoricpotentiallypresent in texts and not upon the recovery of truth, then the rewards of interpretationindeed lie in the attainmentof
"narcissistic" gratification.Many people see this as a "sad" state of affairs and as characteristic of the "whole problem" of humanistic study. I would argue that this is not so much the "whole problem"of humanisticstudy as the "whole issue" of humanisticstudy, for this means that interpretiveactivity is involved not simply with the gratificationof suppressedwishes, but with a communalattempt to rework from within a deep structure of unconscious aspiration a culture's terms of idealization.Reading always explores ideals and seeks terms for value. But this is only anotherhighly generalway of statingmy central argument: readingis always an activity of projectiveidealization.Reading is thus always narcissistic. But this "narcissistic" activity does not destroy a text, and it does not deny the knowledgethat can be gainedfrom texts. It appropriatestexts for humanpurposes. Works Cited Alcorn, Marshall,and Mark Bracher. "Literature,Psychoanalysis, and the ReFormation of the Self: A New Direction for Reader-Response Theory." PMLA 100 (1985):342-54. Barthes, Roland.Image, Music, Text. New York: Hill, 1977. Bleich, David. SubjectiveCriticism.Baltimore:Johns Hopkins UP, 1978. Brenkman,John. "Narcissus in the Text." GeorgiaReview 30 (1976):293-327. Culler, Jonathan.ThePursuit of Signs. Ithaca:CornellUP, 1981. de Man, Paul. Allegories of Reading. New Haven: Yale UP, 1979. Edelson, Marshall.Language and Interpretationin Psychoanalysis. Chicago: U of ChicagoP, 1975. HarvardUP, 1980. Fish, Stanley. Is Therea Textin This Class? Cambridge: Foucault, Michel. "What Is an Author?" Language, Counter Memory, Practice. Ithaca: CornellUP, 1977. 113-38. Freud, Sigmund. "On Narcissism, An Introduction."Collected Papers. Trans. Joan Riviere. Vol. 4. New York: Basic, 1959.5 vols. 30-59. . "Instincts and Their Vicissitudes." Vol. 4. 60-83. Holland, Norman. Five Readers Reading. New Haven: Yale UP, 1975. --. "Psychoanalysisand Film: The Kuleshov Experiment."IPSA Research PaperNo. 1, Institutefor PsychologicalStudy of the Arts: U of Florida. --. "Recoveringthe PurloinedLetter." The Reader in the Text. Ed. Susan R. Suleimanand Inge Crosman.Princeton:PrincetonUP, 1980. 1."A Transactive Account of Transactive Criticism." Poetics 7 (1978): 177-98. . "Unity, Identity, Text, Self." PMLA 90 (1975):813-22. Iser, Wolfgang.TheAct of Reading. Baltimore:Johns Hopkins UP, 1978.
Rhetoric, Projection, and the Authority of the Signifier
Jameson, Frederic. ThePolitical Unconscious. Ithaca:CornellUP, 1981. Johnson, Barbara."RigorousUnreliability."CriticalInquiry11 (1984):278-85. Kristeva, Julia. Revolutionin Poetic Language. New York: ColumbiaUP, 1984. Lacan, Jacques. Ecrits. New York: Norton, 1977. . Le seminaire de Jacques Lacan. Ed. Jacques-AlainMiller. Paris: Edition du seuil, 1975-81. Livre 1: Les ecrits techniques de Freud, 1953-54. Livre III: Les psychoses, 1955-56. Quotations have been translated by MarkBracher. Skura, Meredith. The Literary Use of the PsychoanalyticProcess. New Haven: Yale UP, 1981. Smith, Hallet, ed. TwentiethCenturyInterpretations the Tempest. Englewood of Cliffs: Prentice, 1969. Steiner, George. "Narcissus and Echo." American Journal of Semiotics 1 (1980):3-16. White, Hayden. Tropicsof Discourse. Baltimore:Johns Hopkins UP, 1978.