The gender Fallacy

William C. Dowling

My argument this afternoon will be that the problem of "men writing the feminine," to borrow the title of a recently-published collection of essays, involves a version of the genetic fallacy, what older logic texts explained as a confusion between a thing (an object of explanation) and its origins. As it happens, I think it is ultimately more useful to see the error as a kind of category mistake, an illegitimate jump between two separate universes of discourse. But there is a great deal to be gained, I think, by starting out with a sense that what is involved is in purely logical terms a fallacy, the sort of thing Wimsatt and Beardsley had in mind in speaking of the "intentional fallacy" some fifty years ago. Indeed, part of my point will be that the arguments Wimsatt and Beardsley raised then, in what we tend now to think of as an ancient episode in literary theory, pose objections to contemporary gender criticism that have yet to be answered in any convincing way. The genetic fallacy as I want to consider it is usefully exemplified in an analysis of Eloisa to Abelard undertaken a few years ago by Ellen Pollak in a feminist study of the poetry of Pope and Swift. Eloisa to Abelard poses a genuine problem for feminist criticism precisely because Eloisa has to so many generations of readers seemed so real "as a woman." Crying out from the lonely walls of the convent to which she has been exiled by their tragic love affair, speaking to an absent and imagined Abelard with all the passion that she must have felt when she was younger and they were together, Eloisa has always been a poetic voice behind whom it has seemed difficult -- or, what is perhaps the same thing, irrelevant -- to hear the voice of an "actual" Alexander Pope. Pollak, to her credit, does take into account the long tradition of response that has heard Eloisa's voice as wrenchingly "real," even noting the older tradition of biographical criticism that saw its emotional intensity as owing to Pope's own personal circumstances (specifically, his longing for the absent Lady Mary Wortley Montagu, with whom he was for a time wholly infatuated). But, Pollak concludes, given Pope's authorship of the poem, any appearance of emotional authenticity must be misleading . What is involved is necessarily ideological mystification, the male domination of women being carried out under the sign of a spurious empathy: "if anything is indulged here, it seems to me to be not the specificity of a woman's torment, or her display of erotic and emotional intensity, but rather a voyeuristic male appropriation of female eroticism in the service of a phallocentric ordering of desire in which both excess and lack are figured as female" (186). There are several logical problems here. The first is that it is hard to see how, insofar as it begins from a notion of male writers as in some necessary sense misogynist, a properly feminist interpretation of Eloisa to Abelard could come to any other conclusion. Pollak herself explicitly raises this issue as part of the question whether Pope or Swift might not, in this or that isolated instance, have registered a genuine empathy with female experience. But it is not a suggestion she wishes to entertain. Even "to argue whether or not the texts of the two poets are misogynist is," she points out, "already to entertain the possibility that they (or some of them) may not be." And, "since these texts are the products of a phallocentric culture and of its authorizing sign-systems and codes" (183), it is something like an a priori certainty that they will mirror their misogynist origins. Yet this raises a very great difficulty. Such a reading of Pope's poem will then have to depend in some strict and exclusive sense on its being, precisely, Pope's poem -- that is, a text produced by a male author in at least the minimal sense that the hand holding the authorial pen was anatomically connected to a male body. The easiest way to see why this reading rests on the genetic fallacy is simply to imagine what would happen in an imaginary scenario in which it had turned out, through some recently-discovered documentation, that Eloisa to Abelard had been written by not by Pope but by Lady Mary Wortley Montagu. The scenario: having composed the poem -- no great stretch of imagination called for here, Lady Mary having been a gifted poet in her own right, and one whose poetry includes Ovidian verse epistles like Eloisa to Abelard -- and being on the point of leaving for a long absence in Constantinople, she is moved by aristocratic modesty to conceal her authorship and deliver the poem over to her friend Alexander Pope for publication in his 1717 Poems. The problem for Pollak is obvious. Through an alteration of circumstances wholly external to the text, Eloisa to Abelard has suddenly gone from being a voyeuristic male appropriation of female emotional experience to (presumably) an authentic and moving expression of such experience. Yet not a word of the text has changed. To open an anthology and turn to Eloisa to Abelard is to find just the same poem as before. 2. "The Intentional Fallacy," wrote Wimsatt and Beardsley in 1951, "is a confusion between the poem and its origins, a special case of what is known to philosophers as the Genetic Fallacy " (21). In those more innocent days, their argument was against a historical or biographical criticism that wanted to move from information about an author, usually drawn from letters or journals or recollections of friends, to a claim about what some text bearing that author's name must mean, by virtue of embodying the traces of a unique personal or psychological history. On the most general or abstract level, their counter to this notion would be the theory of literary autonomy: novels and plays and poems understood as worlds in themselves, obeying their own laws and their own logic and subject to violence and distortion when made to answer to doctrines or ideologies external to themselves. The theory of literary autonomy as Wimsatt developed it in The Verbal Icon had a negative and a positive thrust. The positive idea was that of a speaker or narrator wholly internal to the work. Even the older biographical-historical critics against whom Wimsatt was writing had in certain obvious cases been compelled to make a distinction between author and speaker or narrator -- the distinction, say, between Mark Twain and Huckleberry Finn, or Nabokov and Humbert Humbert, or Browning and Fra Lippo Lippi. To do so, as Wimsatt saw, was already in some sense to understand the work as a sphere of reality existing independent of the writer who had composed it. Wimsatt's radical move in The Verbal Icon was to extend this distinction to anything understood as a literary work, to insist that the "Keats" who speaks in When I Have Fears is no less a voice within the text, created and sustained by its discourse and belonging to its world, than Huckleberry Finn or Fra Lippo Lippi. The "personal," on this account, even at its most intimate or self-revelatory, must always nonetheless be seen as an effect of language in its public aspect as literary discourse. This distinction between author and speaker is not central to my present argument about the genetic fallacy, but I do want to note in passing that it suggests an immediate solution to the problem raised by Pollak's reading of Eloisa to Abelard. For at the time The Verbal Icon was written the usual way of insisting that the distinction was radical and constitutive was to say that a valid interpretation could survive any change in an attribution of authorship. For instance, a reading of To His Coy Mistress that attributed to its speaker attitudes and opinions of the historical Andrew Marvell could not on this account survive the revelation, come to light in a newly-discovered diary, that the poem had actually been composed by one Jonathan Smedwick, an obscure clergyman and contemporary of Marvell's living in Cornwall. An interpretation taking the "I" of the poem as a mind or consciousness dwelling wholly within the text, on the other hand, would need no adjustment or alteration. The way this argument applies to Pollak's reading of Pope -- that is, to evidence unexpectedly transferring Eloisa to Abelard from Pope to Lady Mary Wortley Montagu -- does not, perhaps, need to be spelled out. Still, it was the negative thrust of Wimsatt and Beardsley's argument, deriving from a purely logical or analytic insight, that gave it so enormous an influence in its day. This was the point that the problem of intention is illusory. The way Wimsatt and Beardsley put it is worth recalling. Imagine, they say, that a poet intends something and we want to know whether or not that intention was fulfilled in the text. How are we to find this out? The answer: "If the poet succeeded in doing it, then the poem itself shows what he was trying to do. And if the poet did not succeed, then the poem is not adequate evidence, and the critic must go outside the poem -- for evidence of an intention that did not become effective in the poem" (4). This point comes to bear with some cogency on a certain kind of argument -- Pollak's "voyeuristic male appropriation," for instance, taken not as an argument about Pope but as a form of argument about literary meaning -- common in contemporary gender criticism. To bring the point to bear on the problem of gender in Eloisa to Abelard, we need only substitute "maleness" where Wimsatt and Beardsley speak of intention . If there is evident in Eloisa's voice some alien or dissonant element of maleness -- "evident" in this case meaning "discernible by a competent reader who had not been told whether a man or woman wrote the poem" -- a reading of Eloisa to Abelard as a voyeuristic male appropriation of female experience makes perfect sense. If, on the other hand, nothing of Pope's maleness has made it through into the text -- if, whether he or Lady Mary or somebody else altogether should happen to have written the poem, the "I" who speaks as Eloisa has been imagined in purely female or feminine terms -- then such a reading will be a construction imposed upon the text, a signal instance of the genetic fallacy. No better example of this point could be imagined, perhaps, than Diderot's La Religieuse as discussed by Beatrice Durand in the collection Men Writing the Feminine. For the circumstances in which Diderot's "novel" originated are so bizarre as to constitute a virtual parable of mystified or obscured authorship. In very brief terms, here they are. The Marquis de Crosmare, a member of Diderot's Parisian circle, in 1759 decided, to the regret of his friends, to retire to his estate in Normandy. Before leaving, the Marquis had taken up the cause of a young woman who had earlier been made to enter a convent against her will. He had intervened at court when she applied to renounce her vows. The appeal was lost. The Marquis departed for his estate, and Diderot and his friends, in an effort to entice him back to the capital, sent him a series of heart-rending letters purporting to be from the young nun. The Marquis, wholly taken in, answered these letters. The correspondence continued at such length that it would later provide materials for the "memoirs" Diderot would compose as La Religieuse. Here, one might reasonably suppose, we have hit upon a kind of Turing test of gendered textuality: a French nobleman, a literate and sophisticated man of the world wholly conversant with the codes and sign-systems of his society, has absolutely no intimation that he is reading texts composed by his own male friends. So sex or gender really do, in at least one signal instance to which we can point, leave no textual trace. Yet this is not the conclusion drawn by Durand, for whom "Suzanne" as either the female voice speaking in the original letters to the Marquis or the narrator of the subsequent "memoirs" is not, "even if her character is credible," a "woman who speaks and writes." Suzanne is, on the contrary, "a man assuming a mask, the false identity of a woman. She is the image of the feminine that a male author must project in order to disguise himself as as woman" (90). The great interest of La Religieuse thus lies, from this perspective, not in its relation to eighteenth-century French culture or other eighteenth-century texts but to the preoccupation of contemporary gender criticism with such matters as transvestitism and cross-dressing: "According to male transvestites who have been interviewed, one of the biggest satisfactions in cross-dressing is to 'pass,' to make oneself credible as a member of the opposite sex. . . . In conforming to the code of feminine voice and gestures, Diderot and his friends successfully 'passed': they were rewarded for their efforts by the response they received from the Marquis" (92). The fallacy on which such arguments rest has no agreed-upon name in standard logic texts. I shall refer to it as Ophalmos reasoning, borrowing the name from the episode in Father and Son in which Edmund Gosse's father, at once a brilliant natural scientist and a Christian fundamentalist, finds himself caught between his own deep religious belief and the unsettling new hypothesis of Darwinian evolution . This is the situation in which, famously, Philip Gosse, suddenly glimpsing what he thought was a way out, would write the treatise Ophalmos, arguing that God had indeed created the world during the six days described in Genesis, but had at the same time, to test the faith of a scientific age, strewn the geological record with fossil evidence to suggest a process of development requiring millions of years. The argument, in short, is either that (1) God exists and the world shows that he exists -- the standard physicotheological idea, summarized in works like Butler's Analogy, that nature reveals God's design -- or (2) he must exist because he was there to construct the world in such a way as to imply that he does not exist. The slide from genetic fallacy to Ophalmos reasoning evident in Durand's analysis needs no extended comment. The idea is that the more successful Diderot was at effacing from the text every trace of his maleness, the more male it in some ineffable sense turns out to be, until a complete absence of textual evidence becomes, through a sudden twist or reverse of logic, an overwhelming weight of evidence. 3. Gender criticism has not been unaware that there is a problem here, or even that the problem may be associated with the attempt to impose modern ideological categories on poems and plays and novels existing quite independent of those categories. Yet it is not clear that anyone working in gender criticism has yet grasped that the problem arises from a version of the genetic fallacy that is evident in literally hundreds, perhaps thousands, of recent books and articles and reviews in scholarly journals, or in empty or Ophalmos argumentation that derives from the genetic fallacy. This is why, one suspects, there has been a tendency to misunderstand Nina Baym's point, in "Why I Don't Do Feminist Literary Theory," in objecting to such theory precisely because it "requires sexual difference as its ground" (qtd. in Siegel 61). The response of feminist critics to Baym typically betrays, as does that of Carol Siegel in Men Writing the Feminine, a note of honest puzzlement. Since the very idea of feminist criticism is based on the need to expose the illegitimate domination of women by men, how could it not require sexual difference as its ground? There is, Siegel points out, a further consideration. Feminist criticism is not merely an area of intellectual inquiry -- a discipline or "field" as such -- but part of an ongoing political struggle: "As Toril Moi puts it, 'the feminist struggle must both try to undo the patriarchal strategy that makes 'femininity' intrinsic to biological femaleness, and at the same time insist on defending women precisely as women" (Siegel 82). The point that such a response misses is that Baym can be taken to be objecting not to any concern with gender in literary texts but to "sexual difference" as Ophalmos reasoning. In the Diderot case, for instance, if one has grounds to suppose that the young woman in La Religieuse was originally a creation of Diderot and his male friends, then the argument from "sexual difference" will be that the text is an example of literary transvestitism or cross-dressing. If evidence comes to light tomorrow in the Bibliotheque nationale that "Suzanne" was an actual woman whom Diderot was protecting from retribution by her religious superiors -- the originals of the letters to the Marquis turn up, let us say, in the handwriting of an eighteenth-century Frenchwoman whom scholars know to have been put in a convent against her will -- then the argument from "sexual difference" will be that La Religieuse has suddenly been transformed into an authentic expression of female suffering. If these letters turn out to be a literary hoax -- a clever forgery by a modern collector who has seen a chance to make a financial killing, for example -- then the argument from "sexual difference" will be that we have gone back to dealing with literary tranvestitism. Yet through all these alterations, as in the case of the Eloisa to Abelard scenario, not a word of the text will have changed. 4. The occasional attempts to deal with this problem have come from outside gender criticism. In a noteworthy Diacritics essay published in 1988, for instance, Jefferson Humphreys argued that much discussion of the body in feminist criticism has been a way of smuggling in an essentialist idea of gender that is, strictly speaking, forbidden to such criticism by its own postulates -- a more generalized version, in short, of the genetic fallacy as discussed here. To the extent that feminist criticism takes seriously the notion of the "female" as a cultural construction, Humphries argues, and to the extent that it is able to grasp textuality as a primary site of such construction, feminist criticism must by its own lights be guided by "a recognition that gender difference functions as a trope, subject to reversal and substitution." This is the sense in which the literary work, even for the most politically engaged feminist critic, "already contains the antidote to phallocentrism": its very textuality subverts the logic of essentialist or "natural" categories on which an ideology of domination must, on the account of gender criticism itself, be ultimately based. The answer to this sort of argument by feminist theorists has been that it, too, is a mode of male domination. Thus, for instance, Teresa de Lauretis: "This kind of deconstruction of the subject is effectively a way to recontain women in femininity and to reposition female subjectivity in the male subject, however that will be defined" (24). Even to talk about a logic of textuality in relation to feminist criticism is, in short, to lapse into precisely the sort of abstract "male" rationality against which feminist criticism is engaged in perpetual struggle. The point to which theory is brought by such a response resembles the controversy that in an earlier day raged around the Freudian concept of denial. The problem looks like this. Freudian psychoanalytic theory offers me an elaborate model of the workings of my own psychic economy, including the lawless sphere of instinctuality that is my unconscious. There is, as it follows from this, "denial," the absolute refusal of my conscious mind to register or acknowledge intolerable truths about myself. I survey the model and remark that it seems to me not a theory at all, or even a very interesting body of speculation about human psychology. It seems, rather, a compound of medical quackery and Greek myth and fin de siecle irrationalism attempting to pass itself off as a science. And the psychoanalyst, or the one who has been appointed to speak for psychoanalytic theory, says, of course, now that is denial, and you will find it fully analyzed by Freud in his case study on . . . There is, it may be, no way out of the impasse created by this move. To argue that contemporary gender criticism rests on the genetic fallacy is simply to make a theoretical point made long ago, as we have seen, by Wimsatt and Beardsley. But to be told that this sort of point is itself one of the ruses of masculinist or patriarchal domination -- that theory as such, as de Lauretis says in an influential formulation, is a "technology of gender" -- is abruptly to reach what at least feels like a dead end. As with the Freudian psychoanalyst, one does not see exactly how one could enter a meaningful objection to a system defended in this way. Wimsatt and Beardsley were theorists of literary autonomy. That theory, though it has been politically denounced by Terry Eagleton and others, has not yet been shown on any persuasive theoretical grounds to be mistaken. In recognition of that fact, perhaps, even though they passed from the scene long before the arrival of gender criticism, Wimsatt and Beardsley may this afternoon be permitted the last word: "The poem is not the critic's own and not the author's. It is detached from the author at birth and goes about the world beyond his or her power to intend about it or control it. The poem belongs to the public. It is embodied in language, the peculiar possession of the public, and it is about the human being, the object of public knowledge" (5) Works Cited de Lauretis, Teresa. Technologies of Gender: Essays on Theory, Film and Fiction. Bloomington: Indiana UP, 1987. Durand, Beatrice. "Diderot and the Nun: Portrait of the Artist as Transvestite." In Morgan 89-106. Humphreys, Jefferson. "Troping the Body : Literature and Feminism. Diacritics 18:1 (1988). Morgan, Thais, ed. Men Writing the Feminine: Literature, Theory, and the Question of Genders. State University of New York Press, 1994. Pollak, Ellen. The Poetics of Sexual Myth: Gender and Ideology in the Verse of Pope and Swift. University of Chicago Press, 1985. Siegel, Carol. "Border Disturbances: D.H. Lawrence's Fiction and the Feminism of Wuthering Heights." In Morgan 59-76 Wimsatt, W.K. The Verbal Icon. New York: Farrar, Strauss, 1964. Wimsatt, W.K., and Beardsley, Monroe. "The Intentional Fallacy ." In The Verbal Icon 3-20. Paper originally delivered at Twin Oaks theory seminar in May, 1995. Contents copyright (c) 1996 by William C. Dowling. No portion of the above may be quoted without permission.