In June 2000, Andrea Dworkin, an American feminist activist and author, published an account of being raped in a Paris hotel room a year earlier. The story was met with widespread disbelief, including from feminist readers. This article explores the reasons for this disbelief, asking how and why narratives of rape are granted – or denied – truth status by their readers. The article argues for understanding the conferral of belief as a narrative transaction involving the actions of both narrator and reader. It posits that Dworkin was widely seen as an unreliable narrator but argues that for ideologically charged narratives such as rape narratives judgements of reliability and belief inevitably draw upon the normative standpoint of the reader. I suggest that there are opposing criteria for establishing the truth of rape narratives; a ‘factual’ or legal model, which sees rape narratives as requiring scrutiny, and an ‘experiential’ model, located within certain strands of feminist politics, which emphasises the ethical importance of believing women’s narratives. The article finishes with a consideration of the place of belief within an ethics of reading and reception of rape narratives.

1. Introduction

In June 2000, an article by Andrea Dworkin, controversial radical feminist activist and author, appeared in the British newspaper The Guardian (Dworkin 2000b). The article, “They Took My Body From Me and They Used It” was re-published shortly after in the current affairs magazine, New Statesman with the title “The Day I Was Drugged and Raped” (Dworkin 2000c). The articles told the story of Dworkin being “drug raped” in a Paris hotel room in May 1999 by the hotel bartender and a waiter, and of the trauma which followed the assault.² The narrative was met with widespread public statements of doubt and disbelief, including by many feminist commentators. Dworkin claimed to be “saddened” but unsurprised by the reactions. “If the Holocaust can be denied even today,” she is reported to have said, “how can a woman who has been raped be believed?” (Gibbons 2000, 10)

Dworkin’s question is also mine. I repose the question, shifting it from a rhetorical statement of hopelessness to a topic for investigation that sees belief and disbelief as opposing potential outcomes of autobiographical narrative transactions. The varying statements of disbelief that emerged in response to Dworkin’s account allow for an investigation of the conditions under which belief is not granted to autobiographical narratives and the explanations that readers offer for this refusal of validation. I am interested therefore in ‘truth’ as a status that is granted by audiences’ conferral of belief, rather than as an ontological fact or “objective” measure of correspondence with extratextual realities. In keeping with Philippe Lejeune’s (1989, 19) notion of the “autobiographical pact” I argue that truth judgements can be viewed as a particular type of narrative transaction involved in factual and autobiographical texts. It is a transaction in which the narrator seeks to compel belief while the audience does or does not confer it. These questions are explored further in the opening section of the article.

Despite its status as one of the “master narratives” of Western thought, autobiographical self-fashioning has received little direct attention within narrative theory (Smith 1987, 6; Shen / Xu 2007, 44). Following the work of Dan Shen and Dejin Xu (2007), however, I suggest that the concept of narrative (un)reliability, when applied to autobiography, is closely tied to criteria of belief through reader assessments of narrative plausibility, narrator credibility and the narrative’s consistency with the reader’s understanding of extratextual realities. In autobiography, therefore, judgements of reliability are closely tied to notions of truth and belief. Further, judgements of autobiographical narratives always involve evaluations of the persons who tell them (Trinch 2003, 18). The concept of reliability is also useful for elaborating the link between evaluations of narratives and narrators. In the second section I make use of Shen and Xu’s (2007) work on the application of narrative (un)reliability to autobiographical texts to outline the ways in which Dworkin’s account was found to be unreliable.

Judgements of texts, such as whether they are reliable or true, are always influenced by what Ansgar Nünning (2008, 95) describes as “normative presumptions”. This is particularly true for narratives of sexual violence which are highly contested and ideologically charged. The third section contextualises the preceding discussion of reliability through a consideration of the application and functioning of the truth criteria that are applied to women’s narratives of rape. The period from the 1970s onwards, I argue, has seen a conflict between legal standards of “truth-telling” and the feminist project to enable and disseminate women’s narratives of rape. I suggest that feminists sought to alter the truth criteria applied to rape narratives and to change, and broaden, the conditions under which women’s accounts of sexual violence would be believed. Dworkin’s account, along with other rape narratives, can therefore be seen to sit between two different generic standards of “truth-telling”, a legal standard with an emphasis on factual truth and external verifiability, and a feminist one with an emphasis on subjective or experiential truth and markers of authenticity. Narrators and readers of these texts must, therefore, navigate between different and ideologically opposed criteria of believability, making evaluations of texts highly conditional on the “normative” presumptions that readers bring to them.

Theories of autobiography have emphasised the vulnerability and risks faced by autobiographical narrators as they seek to construct an autobiographical self and have that self recognised. In narratives of trauma this vulnerability is enhanced both by the difficulty of telling and the potential for the original trauma to be compounded if a narrative is disbelieved or rejected (Smith / Watson 2008, 364). This vulnerability is particularly enhanced for women who tell of experiences of rape given the multiple ways in which these narratives have historically been disbelieved, over-written and re-inscribed as other stories, from ‘fantasies’ to tales of consensual sex (Ehrlich 2001). The third section, therefore, focuses on the stakes and consequences of believing or not disbelieving accounts of sexual violence. I investigate debates amongst feminist readers of Dworkin’s narrative to argue that there is an ethics of reading or reception in relation to rape narratives that encompasses questions of belief. I conclude by asking what such an ethics should entail, considering how women’s stories of rape come to be believed but also more normative questions of why and under what circumstances they should be.

2. Doubting Dworkin: Truth, Belief and Autobiographical Narratives

Dworkin’s (2000b, 2) published account of the Paris rape opens with her sitting in the hotel bar drinking and reading. The action commences when her second drink “didn’t taste right” and “then I felt sickish or weakish or something”. She orders room service, recalling that the waiter “who had also made the drinks, had said: ‘It will be my great pleasure to serve you your dinner tonight’”. Upon reaching her room Dworkin immediately “conked out”, returning only briefly to consciousness when the waiter entered the room.

She woke with pain “deep inside” her vagina, vaginal bleeding, and bruising and gashes on her breast and legs.

The event at the heart of the narrative, the assault, is, therefore, initially reported through ellipsis. Following her account of her pain, however, Dworkin reconstructs what she believes “must have” happened, based on her assessment of her physical injuries:

These events occupy the first five paragraphs of a seventeen paragraph account. The remainder is devoted to the impact of this event on Dworkin’s life in the year between the events and publication of the narrative. She describes feelings of depression, helplessness and hopelessness, as well as a lack of sympathy from significant others, including her gynaecologist and her partner, John. The final paragraphs tell of the death of her father six months after the Paris assault and her own subsequent disintegration. Following her father’s death she was hospitalised for “bronchitis, pneumonia, cellulitis […] and blood clots” after being found wandering in New York in a “high delirium” (Dworkin 2000b, 2). The article concludes with the sentence, “I’m ready to die”.

Five years later she did die of a bone disease whose onset coincided with her father’s death and her own hospitalisation. In “Through the Pain Barrier,” written a month before she died, Dworkin states that, for her, the bone disease was the final consequence of the sexual assault that occurred in Paris. Her doctors, however, disagree:

The trauma of the rape was augmented by the scepticism and disbelief which greeted her account right from the beginning. Her gynaecologist responded to Dworkin’s phone call the day after the assault by saying it had convinced her she needed an unlisted number while John “abandoned her emotionally” and “looked for any other explanation than rape” to explain Dworkin’s subsequent ill-health. Six days after the publication of the Guardian article Dworkin’s book Scapegoat was published in the UK (Dworkin 2000a). On the same day The Guardian published an article by columnist Catherine Bennett (2000), “Doubts about Dworkin,” which directly questioned both the validity of the account of rape and Dworkin’s veracity, suggesting that the timing was highly convenient given the need for publicity for Scapegoat. What followed was, according to sympathetic columnist Julia Gracen, “an accusatory pile-on […] in the UK press and on the Web” (Gracen 2000). Even Gracen, however, stopped short of accepting the validity of Dworkin’s account. She suggested that while “there is no question that something happened to Dworkin last year” the narrative was more likely to have resulted from mental health problems than an actual sexual assault.

The controversy, Dworkin later stated, caused her to withdraw from public life. Commenting on the number of feminists who publicly doubted her account, Dworkin said it “was unbearable being disbelieved by my so-called sisters” (Bindel 2004). In 2002, she re-emerged with the publication of her memoir, Heartbreak, which dealt with her earlier life and omitted the events of Paris entirely (Dworkin 2002). In other accounts, however, the Paris rape remained the defining event of her final years, leading ultimately to her illness and death.

This conflict between Dworkin and her readers can be seen as a failure of the “autobiographical pact” which structures narrator and reader expectations of autobiography, and which centres on questions of truth and belief (Lejeune 1989, 19). Readers expect narrators to tell the truth and narrators expect their account to be believed. Within this interaction narrators therefore make a truth-claim for their story. They do this through generically framing it as an autobiographic account but may also attempt to buttress that claim, asserting the validity of their narrative through providing factual or verifiable details, as Dworkin does in the opening of her account:

The detailed specificity gives the impression of truth and testifies to the accuracy of Dworkin’s memory (Shen / Xu 2007). The final statements regarding drinking give a sense of veracity at the same time that they attempt to foreclose the suggestion that intoxication has affected the accuracy of her recollections.

For readers to understand an account as true they must accept its validity or factual truth and the narrator’s veracity, points implicitly addressed in Dworkin’s opening. They must also judge the text as corresponding to external reality, as internally consistent and consistent with other accounts given by the author and as presenting a plausible series of events and interpretations, which match readers’ understandings of the world (Smith / Watson 2008, 359). If these conditions are not met to the readers’ satisfaction they will reject the account, refusing to confer authority and belief on the narrator. Disbelieving a narrative such as Dworkin’s means deciding it didn’t happen, at least in the way in which it was presented, and rejecting the truth claim made by the author. The account may be judged to be unsupported or even contradicted by extratextual factors, for instance, the presence or absence of hotel records validating the fact that Dworkin stayed in Paris in May 1999. It may be judged to be internally inconsistent. Or it may be judged to be implausible, if a reader finds it unbelievable that no official report or complaint was made by Dworkin following these events. Such judgements are made on the basis of judgements both of the text and its narrator.

Dworkin’s response to her readers’ disbelief was to insist on their agency and responsibility, holding them ethically lacking for denying her account. She did this in a number of ways. In the quotation that refers to the Holocaust she refers to herself as “a woman who has been raped” and disbelieved, referencing the silencing, erasure and re-writing of women’s accounts of sexual violence that has been well-documented by feminists (e.g. Higgins / Silver 1991; Ehrlich 2001; Serisier 2007). The reference to the Holocaust extends this process of generalisation through an analogy to revisionist historians and their erasure and silencing of the voices and stories of Holocaust survivors (Lyotard 1988). The analogy both confers the moral legitimacy of victimhood on Dworkin and formulates the act of disbelief as normatively suspect. Her account of the disbelief of others is one in which she has been doubly wronged, both through a direct act of violence and the denial of her account of that violence. It is this second harm that makes the betrayal of her “so-called sisters” “unbearable” for Dworkin, emphasising both that readers have agency around questions of belief and that their judgements are inflected by social norms and relations of power.

3. Dworkin’s Unreliable Narration: Narrative Theory and Autobiography

The concept of narrative reliability was developed by Wayne Booth (1983) whose rhetorical approach defined unreliability as lying in a discrepancy between the norms and behaviour of a narrator and the norms of the text’s implied author. In Yacobi’s (2008, 109f.) constructivist approach, however, unreliability is conceptualised as a “reading hypothesis” that is formed to “resolve textual problems” such as internal inconsistency. In order to bridge the conceptual gap between text-oriented and reader-oriented models, Ansgar Nünning (2008) has recently outlined a synthesis of existing approaches, combining insights from constructivist and cognitive models with rhetorical and ethical concepts of narrative unreliability. This synthetical approach is compatible with Phelan’s (2005, 51f.) typology, where a narrator may wrongly or inadequately report, interpret or evaluate elements of a narrative, resulting in six types of unreliability: underreporting, misreporting, underinterpretation, misinterpretation, underevaluation and misevaluation.

While the question of reliability in autobiographical narratives is relatively undeveloped, Shen and Xu (2007) provide a useful model of how the concept – originally developed for the analysis of unreliability in narrative fiction – might be applied to non-fictional storytelling. They argue that the constructivist approach on the whole is more fruitful than the rhetorical one given that the distinction between the implied author and the narrator does not apply to autobiographical texts. As autobiographical narratives involve “direct telling from author to audience” the author and narrator are more usefully collapsed into a single figure of the author-narrator. Autobiographical texts, however, still possess problems, such as inconsistency which readers resolve using hypotheses which may include unreliability of the narrative and its narrator.

Catherine Bennett (2000), author of the original and most detailed critique of Dworkin’s narrative, can be understood as judging the text to be unreliable in various ways. She notes that the rape is under-reported as Dworkin did not directly witness it. Further, she complains that where “most people would expect to find facts […] Dworkin supplies inconsistency, absence of evidence, lack of support” (Bennett 2000, 9). This under-reporting leads Bennett to find the entire account implausible ultimately judging the rape to be a complete misreport, suggesting Dworkin may have even invented it to provide publicity for her forthcoming book. If the facts were true, she argues, surely the absence of the waiter and bartender would have been noticed. If the curtains had indeed been open, someone might have seen something. She also accuses her of misevaluation, either due to her mendacity or excessive naivety. Dworkin describes the rape as “foolproof”, concluding that prosecution would be impossible: “You can do this hundreds of times with virtually no chance of getting caught, let alone having anyone be able to make a legal case in any court of law” (Dworkin 2000b, 2). Bennett contests this evaluation, pointing out that physical evidence such as DNA from the bartender’s sperm could have easily been collected and records of unexplained absences of the bartender and waiter from their posts would have provided circumstantial support to Dworkin’s story and evidence in a court of law.

Bennett’s arguments draw attention to the role of external verifiability or truth in judgements of autobiographical reliability, recalling the autobiographical pact. “What really happened” is the “ultimate yardstick” in assessing autobiographical accounts and it is this that makes judgements of reliability and believability largely synonymous (Shen / Xu 2007, 56). The ambiguity of the text, therefore, is less likely to result purely from intratextual discrepancies. They also involve intertextual or extratextual phenomena, in that they conflict with other narratives, external facts or the reader’s wider knowledge of the world. The absence of extratextual verifiability and judgements of implausibility based on Bennett’s understandings of the world are both central to her doubts about Dworkin’s account.

The significance of extratextual knowledge in judgements of autobiographical reliability leads Shen and Xu (2007, 47) to divide readers into two categories: an “uncognizant reader” who “is not cognizant of the discrepancies between the textual story and the real personal experiences it is supposed to represent”, and a “cognizant reader” who compares the textual world created by the author-narrator with their knowledge of the extratextual world. While uncognizant readers have few resources with which to judge a narrative or narrator unreliable, cognizant readers such as Bennett engage consistently in judgements of reliability on the basis of the narrative’s consistency with intertexts by the same author and extratextual reality (Shen / Xu 2007, 57).

Cognizance is in part a function of the reader’s familiarity, culturally, geographically and historically, with the world of the author-narrator. As Chambers (2002) argues, such familiarity leads readers to believe that they can imagine themselves in the story, and know how they would act, and indeed, how an authentic protagonist of that story would act. Judgements of autobiographical reliability are thus also “comparative”, judging the behaviour of the narrator in relation to how comparable others, including the reader, would act in similar circumstances (Shen / Xu 2007, 59). This comparative logic is also a clear component of Bennett’s (2000, 9) scepticism. If Dworkin awoke with serious injuries surely she would have sought immediate medical attention rather than simply phoning a doctor in a different country. If she believed she had been raped and “poisoned”, surely she would have gone to the police.

Bennett’s criticisms of the text also involve questions about the real or historical Dworkin’s actions in the world beyond the narrative, another unique feature of autobiographical narratives, according to Shen and Xu (2007, 49). They argue that the collapse of the implied author and narrator roles gives the real author a more direct role in the narrative transaction.³ The actions of the narrator are not only compared to general reader beliefs regarding the outside world but also to their knowledge and understanding of the author (47). In autobiography, reader judgements of narratives, narrators and authors are inextricably linked, to the extent that perceptions of an author may lead readers to see any narrative produced by them as inherently untrustworthy. On the other hand, the causes of narrative unreliability are located in the “incompetence, untruthfulness, unawareness, [or] misjudgment” of the narrator and, by extension, the author (Yacobi 2001, 224).

In the narrative, Dworkin describes her life outside it: “I know a lot about rape. I study it. I read about it. I think about it […] I engage with prosecutors and lawyers and legislators” (Dworkin 2000b, 2). Bennett (2000) found the narrative’s account of the real Dworkin’s actions implausible in part because of her identity as an internationally renowned radical feminist who has made a career studying and decrying male violence. Dworkin’s own explanations of her inaction were inconsistent. Describing her expertise in the original narrative she claimed that knowing “too much” about rape had caused her paralysis. In a 2002 interview, however she explained her decision not to take legal action as a result of ignorance:

If this description of an isolated woman, confused and lacking knowledge, was deeply at odds with the understanding that cognizant readers such as Bennett had of Dworkin, the inconsistency reached its peak in what became the most controversial passage in Dworkin’s analysis. Discussing her feelings of self-blame, Dworkin goes through a ‘checklist’:

Even for Gracen (2000), one of Dworkin’s most sympathetic readers, this ‘checklist’ was “contrary to everything a rape expert should know”. The normative dissonance was so pronounced that readers who did not simply concur with Bennett that Dworkin was a liar, found themselves compelled to offer other explanations, such as Suzie Bright’s (2005) comment that the checklist was the point in the narrative at which “you know she has completely lost her mind”. For these readers, the only way to resolve the inconsistency between the narrator and their understanding of Dworkin was to hypothesise that her identity had been fundamentally altered through the effects of mental illness.

As Shen and Xu (2007, 78) point out, authors of autobiographical narratives are often public figures, well-known beyond the text. In Dworkin’s case especially readers came to the text not only with high levels of existing knowledge about Dworkin but often with strong and entrenched normative judgements of her personality and actions. Andrea Dworkin was a highly polarising public figure, seen as a heroic feminist militant by some and as an “anti-sex” villain by others (Serisier 2013). Readers in the former category were likely to come to the narrative with strong predispositions towards belief and sympathy. The attitude of the latter group was described by Suzie Bright (2005), a long-time political opponent of Dworkin’s: “Plenty of my peers would say that they are utterly cold to any misfortune that might befall her.” Such readers’ views of Dworkin would render them unwilling to engage with any account produced by her, no matter how traumatic the content.

4. Genre and Judgement: From the Law’s Lying Women to Feminism’s Embattled Truth-Tellers

The extreme nature of reactions to Dworkin indicates the importance of normative or ideological contexts for the evaluation of autobiographical texts. This context is broadly framed by the set of shared norms or beliefs that Roland Barthes (1977, 70f.) referred to as doxa. These beliefs structure our understanding of everyday happenings and our judgements of the reliability, both factual and normative, of autobiographical and other factual texts. Applied to rape, these beliefs include both common sense understandings of how and under what circumstances rape occurs and gendered beliefs about how women and men should and do behave. Hegemonic beliefs in this area are often referred to by feminists as “rape myths” because they often bear little direct relation to the realities of rape (Medea / Thompson 1974). For instance, there is a strong cultural belief that rapes are generally committed by violent strangers, whereas in reality most sexual assaults occur between partners, acquaintances and family members (Estrich 1987). These myths create an “alternative reality” which fractures and distorts women’s experiences and narratives until they are virtually unrecognisable (Kaspiew 1995, 350). For instance, Susan Ehrlich (2001, 66) has documented the ways in which women’s legal testimony of acquaintance rape may be rewritten as consensual sex. One way in which this is achieved is through applying an “ideological framing” of non-consent as requiring “utmost” physical resistance. This framing enables the categorisation of women’s verbal signals, such as asking an assailant to leave, as ineffectual or inadequate.

Ehrlich’s example also demonstrates the significance of generic and institutional contexts in evaluating narratives. The same narrative that is deemed inadequate as legal testimony may be believed when told in a different genre. While the autobiographical pact is an articulation of the process of granting belief within autobiography as a genre, many autobiographical texts are also or primarily located in other genres with their own truth criteria, such as the genre of legal testimony. Even within this genre, different types of testimony will be treated differently. A victim of theft, for instance, does not have his or her reliability undermined by a failure to demonstrate physical resistance when testifying. Whether or not a text is judged to be truth, fiction or a lie depends on the generic classification of the narrative and the understandings of that categorisation that readers bring to it. For this reason, translation between genres is a difficult and risky process as Shonna L. Trinch (2003) demonstrates in her discussion of the ways in which the “translation” of Latina women’s narratives of domestic violence from conversational accounts to legal documents in the form of applications for “Protective Orders” requires a series of erasures, substitutions and re-inscriptions. This risk is particularly pertinent for narrators, such as Trinch’s subjects who could be classified as “uncognizant” of the rules and truth criteria of a genre with which they are engaged. “Cognizant” narrators, on the other hand, could be defined as narrators who are aware of both the broad cultural doxa and the generic conventions that will be used to evaluate their narratives. These narrators will be better equipped to construct successful narratives and to move between different genres.

Historically, the primary institutional setting for telling and evaluating rape narratives has been the courts and their primary genre has been that of legal testimony with its demands of external verifiability and forensic truth. This remains a primary lens for reading rape narratives, as Bennett’s (2000, 9) demand that Dworkin’s narrative provide “evidence” and “facts” demonstrates. However, since the late 1960s, feminist traditions of “speaking out” have existed as a competing genre with a different set of truth-criteria. Rape narratives thus sit at the intersection of a number of different genres, and the truth-criteria that should be applied to them are contested. In this section I explore Dworkin’s story not only as autobiography but also as a rape narrative. The telling and reception of such narratives must be analysed, I argue, within the wider generic and institutional conflicts that surround accounts of sexual assault, and that these conflicts frame the construction and evaluation of the text. I first provide an account of these conflicts before exploring their operation in this example.

Within legal discourse, rape is considered to be among the most heinous of violent crimes. Women’s accounts of rape, however, have often been met with suspicion, summarised most famously in the warning by seventeenth century jurist Matthew Hale that rape is “an accusation easily to be made and hard to be proved, and harder to be defended by the party accused, tho never so innocent” (Ferguson 1987, 89). Modified versions of this warning, which in effect, encourage jurors to take a sceptical stance towards women’s testimony, persisted in many English-language jurisdictions until the 1980s (Larcombe 2002, 96). Legal discourses have been structured by, and perpetuate, cultural understandings of women’s narratives of rape as inherently untrustworthy. This has led some feminist theorists to describe the process of testifying as a “second violation” where not only a woman’s story, but also her veracity and rationality, are questioned (Grix 1999, 90).

Wendy Larcombe (2002, 99f.) argues that legal distrust of women’s narratives exists because they threaten the law’s self-conception as an “objective” arbiter of the truth on the basis of externally verifiable facts. Such facts are extremely difficult to establish in rape trials because there usually are no eye-witnesses and forensic evidence cannot determine if sexual activity was or was not consensual, the key point to be proven in the majority of rape trials. Further, it is generally not the “facts” that are in dispute but rather the “narrative glue”, the way that incidents and events are combined into a meaningful story, one which is titled “consensual sex” on the one hand and one titled “rape” on the other (Brooks 2008, 417). In the majority of trials the law must differentiate between two such stories, deciding whose narrative “glue” is most believable. As Ehrlich’s above example illustrates, in deciding who to believe, the law has few resources beyond existing gendered doxa which in practice work to construct acquaintance rape cases particularly as casual sex.

When second wave feminists turned their attention to rape in the 1960s they began to challenge these common sense understandings of sex and violence, as can be seen in Susan Brownmiller’s (1976, 369) response to the Hale warning: “Since four out of five rapes go unreported, it is fair to say categorically that women do not find rape ‘an accusation easily to be made’”. They also argued that, given the historical suspicion, denial and re-inscription that had greeted women’s narratives of rape, feminist readings should be based on an ethical presumption of belief (Smith / Watson, 364f.; Serisier 2007). Finally, they argued that the law, as it existed, was an inadequate site to determine the truth of rape. This critique resulted in legal reforms aimed at altering legal truth-criteria or at least limiting its pernicious effects. “Rape shield laws” which restrict the type of questions that lawyers were able to ask in cross-examination are a useful example (Ehrlich 1991, 23). This reform was an attempt to decentre the question of women’s ‘credibility’ and restrict the ways in which it could be undermined through tropes such as recklessness or promiscuity.

They also facilitated the production and dissemination of women’s narratives outside of legal forums through dedicated publications (e.g. Warshaw 1988) and events such as “Take Back the Night” marches (Serisier 2007). Dworkin’s decision to publish her narrative rather than testify in court could be placed in this tradition. This new feminist genre replaces restrictive legal truth criteria with the aim of valorising narratives of experiential truth. Rather than being subjected to doubt and suspicion, women’s autobiographical narratives are considered to be a privileged site for truth production. Veracity, sincerity and authenticity are privileged over validity, plausibility and verifiability, reversing legal hierarchies. These narratives are not intended to ‘prove’ a truth in a court of law but to allow the expression of an inner truth. A secondary function is to facilitate the narratives of other women who, without a reasonable expectation of being believed, would be deterred from telling their stories. The role of audiences in this model is not to arbitrate but to confer belief and legitimacy.

As the feminist model of “experiential truth” has become increasingly influential, however, critics have drawn attention to its limitations. Joan W. Scott (1992) has influentially argued that the model is based on a mistaken idea that women’s autobiographical narratives enable direct and transparent insight into the “truth” of their experiences. However, as theorists of autobiography have long argued, the act of narrating creates rather than uncovers the truth of the self. Or, as Leigh Gilmore (1994, 25) puts it, the subject of experience “is produced not by experience but by autobiography”. In other words, experiential truth is not pre-discursive but shaped through existing genres and discourses and their systems of evaluation and belief conferral. Experiential narratives are not, therefore, authentic if authenticity is taken to signify an absence of artifice or narrative construction. Rather, “authenticity” is a narrative effect produced through its own set of generic conventions which in their own way are as rule-bound and prohibitive as the conventions of legal testimony or any other genre.

As Dworkin’s example shows, failure to fulfil readers’ expectations may result in a narrative being read as unreliable or untrue. Telling a story that is recognised as authentic and sincere involves a number of compulsory elements. Some insight into these elements is provided by the arguments of prominent defenders of Dworkin such as Deborah Orr (2005) and Julia Gracen (2000). Orr (2005, 14) asserts the supremacy of sincerity over verifiability in her claim that the central “fact” of the case was that Dworkin “really believed this awful violation had been made against her, and understood that she was widely disbelieved.” She went so far as to say that “[w]hether Dworkin’s construction of events is true or not, is pretty irrelevant”. Authenticity, in contrast, is primarily achieved through what Judith Butler (1997, 144) has described as “compulsory affect”. Butler argues that genres which demand the “truth” of experience have a tendency to judge that truth in a narrator’s willingness and ability to communicate suffering to the reader. Dworkin’s success in this regard leads Gracen (2000) to argue that Dworkin should be believed because she is clearly suffering. Not only does she assert that “pain is pain” but it is her narrative’s expression of pain that demonstrates that “something” has happened to her.

Feminist criteria for belief do not depart completely from broader autobiographical measures of reliability. Experiential truth similarly requires a high degree of consistency within and between narratives and for readers to be able to imagine acting or responding in a similar way in the circumstances described. Dworkin’s level of intertextual consistency was very high, adding to readers’ conviction that she “really believed” her own account. The narrative of rape, its aftermath and the pain of disbelief was told and retold for the remainder of her life. However, the internal consistency and plausibility of the narrative were more frequently questioned. For instance, Bennett (2000, 9) writes of Dworkin’s decision not to report the rape that “the reluctance of a rape victim to be further violated by examination and questioning is understood”, referring to the way in which feminist criticisms of the legal system have come to shape common sense understandings of survivors’ actions. “However”, she continues, “if this is what prevented Dworkin from seeking help it does not seem consistent with her current decision to relive the ordeal, in vivid detail, for readers of the New Statesman”. For Bennett, as for Gracen, “pain is pain” and if trauma prevents a victim from speaking then this should be true across different sites of speech.

In the debates and doubts over Dworkin no single mode of truth was able to assert its hegemony or completely sideline competing truth criteria. Both experiential models of sincerity and factual models of validity were applied to Dworkin’s account with different participants making claims for the primacy of both perspectives. While Orr (2005, 14) insisted that Dworkin’s sincerity constituted the central “fact” of the case, Bennett (2000, 9) equally vehemently asserted that “a rape either occurred or it did not”. The legal and factual truths demanded by Bennett remain powerful judgement criteria for assessing these narratives, even when they are told in other settings than the court and other genres than legal testimony. But these criteria are haunted by their inadequacies, as demonstrated by decades of feminist critique. In what Gracen (2000) describes as this “brave new world”, or post-feminist era, women’s narratives of rape sit in a zone of contestation, between and among different generic criteria for truth and truth-telling. This means effectively that readers have options in terms of what truth-criteria they apply to the narrative in order to confer or refuse belief. I now examine the role of readers more closely, asking what is at stake in believing or not believing rape narratives.

5. The Ethics of Reading Rape Narratives: What is at Stake in Belief?

Autobiographical narratives do not merely tell a story but also work to construct a self, an ‘I’ who is a product of the story and its telling (Smith / Watson, 357). But this ‘I’ can only be constructed with the recognition of others, a recognition that follows from acceptance of the autobiographical narrative and belief in its story. This makes such narratives inherently risky, no matter how effectively the narrator deploys narrative and rhetorical strategies to compel belief. A successful autobiographical account that achieves recognition also constructs a particular kind of “I”, one who is judged to be rational and truthful, while an unsuccessful account may do the opposite (Trinch 2003, 22). If a narrator is seen as having failed to fulfil the autobiographical pact made with the reader, they are considered to be either bad or mad; a liar or unable to distinguish truth from fiction (Smith / Watson 2008, 359).

The consequences of a lack of belief may be highly damaging for the teller, particularly when the narrative is one of trauma and suffering (Smith / Watson 2008, 364). Narrators of traumatic experiences require legitimation of their narratives as part of the project of narratively reconstituting their subjectivity following a desubjectifying event. Audiences are asked to take an ethical stance of “witnessing” rather than an evaluative one (Smith / Watson 2008, 365). But there is no guarantee that an audience will respond ethically or recognise the story as true. For some the risk of failure or lack of recognition will be too great, rendering their story “untellable” (Norrick 2005, 136). Others will find that readers refuse to legitimate their narratives because they are too difficult to hear, because they are unwilling to take on the responsibility of witnessing or simply due to suspicion of the teller (Smith / Watson 2008, 366). If assessments of narrative reliability are always a “subjectively tinged value judgement projection governed by the normative presumptions and moral convictions of the critic” (Nünning 2008, 95), then this is especially true for autobiographical narrative of trauma where dis/belief may arise from a reader’s broader political convictions rather than any inherent property of the narrative text. This is exacerbated for rape narratives which are both highly ideologically marked and often lacking external markers of verifiability. In such a case, the reader may have little choice but to rely on their normative presumptions and moral convictions when evaluating the narrative.

Rather than automatically inferring a lack of reliability or truthfulness when confronted with a “textual problem” readers may, as Tamar Yacobi (2008, 110) explains, opt for an alternate “integration mechanism”, such as explaining narrative ellipses or absences as an understandable result of trauma or judging it more important that the narrator adequately conveys her affective response to events than that she provides a factually accurate narrative.⁴ While Shen and Xu (2007) outline the ways in which reader cognizance facilitates judgements of unreliability in rape narratives cognizance may therefore grant readers additional resources to justify belief in a narrative.

Feminist responses to rape narrative can be seen as a variation on Lejeune’s autobiographical pact, devised in recognition of the ethical responsibilities of readers. In its ideal form the pact requires that a reader accepts a narrator’s authority to tell her story without reference to factors such as external verifiability, factual reliability or plausibility. In short, sincerity obligates the reader to respond with belief and failure to do so may be a sign of ethical or normative inadequacy. For Julia Gracen (2000), for instance, the sceptical responses to Dworkin’s story signified a public sphere increasingly hostile to survivors of rape. She writes that in the immediate aftermath of second wave feminism women’s narratives were more likely to be heard and respected but “over time the default response to the charge [of rape] has changed. Instead of a tendency towards belief and sympathy, there is now considerably more caution and doubt.” Such a conclusion was buttressed by Bennett’s (2000, 9) common sense formulation that Dworkin’s “rape claim, like any other, seems to deserve scrutiny”.

As both of these approaches illustrate readers’ orientations towards a single rape narrative are informed by broader ethical and normative judgements applied to the genre as a whole. Consequently, responses to an individual narrative are seen to have effects on the conditions of tellability and reception for future narratives, a presumption that is central to the ethics of the feminist autobiographical pact. For Orr (2005, 14), Dworkin’s experience of disbelief connect her to women “the world over”. Further, rejections of Dworkin’s account were implicitly based on the misogynist fiction that “Dworkin was simply too old, fat and ugly to receive any sexual attention at all”. More perniciously the cultural reproduction of such tropes may inhibit other women from coming forward. Gracen (2000) argued that responses to Dworkin’s story contained an “ugly lesson”:

This construction makes clear that denial and disbelief can be acts of violence with egregious consequences, in this case not only for Dworkin but for survivors of violence more broadly.

Given Orr and Gracen’s insistence on the violence of disbelieving Dworkin the validation that they offered of her account was, however, strikingly limited. While both reiterated their belief that “something happened” neither was willing to affirm the factual validity of the narrative. Orr (2005, 14) instead asserted that the factual truth was “irrelevant” given that Dworkin both clearly believed in her narrative and was suffering from trauma regardless of the cause. Gracen (2000) was more explicit regarding her doubts, suggesting that Dworkin’s narrative, with its implausibility and incoherent elements, implied that it was a manifestation of some kind of trauma-related condition such as post-traumatic stress disorder. To believe that someone believes their account does not preclude disbelieving their account and, in the end, neither Orr nor Gracen was prepared to exchange a clear expression of belief for Dworkin’s own sincerity. While both Gracen and Orr rejected the characterisation of Dworkin as a “lying woman” they did not grant her narrative authority. Indeed, their position seemed to be that although Dworkin was not bad she might indeed be mad, a judgement echoed by other ‘sympathetic’ readers such as Bright (2005). The consequences of such a stance are paradoxical, particularly for Dworkin. While Orr could assert the irrelevance of the factual status of the narrative, acceptance of her story’s validity clearly mattered to Dworkin. She writes that her partner John’s inability to accept her account of the sexual assault destroyed their relationship (Dworkin 2000b, 2). To label a narrator as sincere but mad as opposed to a liar does not offer a greater legitimation of their narrative or the self that it constructs, and leaves the ethics question surrounding belief unresolved.

The paradoxical nature of Gracen and Orr’s position suggests that despite their ethical orientation they were unable to successfully apply alternate integration mechanisms for the elements of the text that they found unreliable. This was perhaps because accepting the authority of the account had certain wider normative consequences. If sceptical readers could be accused of implicitly drawing on misogynist tropes Dworkin (2000b, 2) makes explicit use of them in her narrative. The ‘checklist’ discussed above is run through to assert Dworkin’s lack of culpability but in doing so implies that had she been guilty of any of the actions on the checklist she would have been somehow culpable for the violence committed against her. Bennett (2000, 9) also questions the ethical integrity of Dworkin’s refusal to act in the aftermath of the assault. If her account is factually accurate, Bennett asks, then, “Is this bartender, with his accomplice, to be allowed to continue drugging and raping female guests?” This question could potentially be directed not only at Dworkin but at readers who believe or validate both the facts of the account and the interpretations and evaluations that Dworkin offers of them. While feminists have drawn attention to the ethical harms of disbelief this narrative demonstrates that belief may also contain its own ethical complications.

Possible reading positions that emerge from this debate include defining Dworkin as: a factually accurate narrator whose story ethically compels belief and legitimation, a factually accurate narrator whose normative outlook and actions in the extratextual world are questionable, a sincere narrator who tells an experiential truth of trauma but whose factual recounting and normative judgements cannot be granted authority or a liar. There are of course also possibilities of hybrid or ambivalent responses that contain elements of some or all of these. While the first position would validate Dworkin completely it would arguably also validate a series of “rape myths” and a lack of concern for future victims. The second position would confer belief on Dworkin only to censure her, making the ethical purpose of belief largely redundant. The third and fourth position refuse to verify the account as true, thus refusing to confirm Dworkin’s story and her identity as a traumatised survivor of rape, replacing it either with the identity of a confused subject of trauma or a liar. It is understandable that given these options, readers such as Gracen and Orr who wished to take an ethical position in relation to Dworkin, chose what they may have considered the path of least harm, a sympathetic reading that asserted her sincerity and their compassion for their suffering but that withheld the conferral of full narrative authority with its unpalatable normative consequences.

6. Conclusion: How Can a Woman Who Has Been Raped Be Believed?

Was Andrea Dworkin raped in a Paris hotel room by a waiter and a barman? She was unable to produce a narrative which compelled its readers to answer that question in the affirmative. The same story, told in a different way, and perhaps by a different author, might have produced a different answer. Such a conclusion may appear deeply unsatisfying and even dangerous given that the subject of the story is sexual violence. However, it is less dangerous than an answer that provides the comfort of false certainty, and it helps to explain the very real problems that our society has in responding to women’s narratives of sexual violence, both individually and collectively. Rather than continuing to seek to establish the ‘truth’ of rape narratives it may be time to devote more serious attention to how belief is called forth in readers and to developing an ethics of belief when responding to women’s accounts of sexual violence.

I have argued here that belief or the evaluation of an autobiographical narrative as ‘true’ is a transactional process, part of the functioning of the autobiographical pact. Truth, to paraphrase Nünning’s (2008, 94) discussion of reliability, can be usefully understood as a form of authority sought by the narrator and conferred by the reader. While a narrator seeks to compel belief in her reader and a reader is able to confer belief upon the narrator in practice the transaction generally requires both elements – a narrator who is cognizant of the appropriate truth criteria and a reader who is willing and able to recognise the narrative as true.

These transactions occur within wider social and generic sets of truth criteria, a fact understood by cognizant readers and narrators. Sets of truth criteria are not, however, ideologically neutral but rely on hegemonic ideas of how the world is and how it should be. For women seeking to recount experiences of rape the gendered doxa that surround their stories have often invalidated them, re-inscribing them as, for instance, narratives of consensual sex. An orientation of suspicion towards rape narratives continues in contemporary society, seen in claims such as Bennett’s (2000) that all rape narratives should be subjected to scrutiny. This ideologically charged cultural framework around rape narratives must be accounted for in any narrative attempt to compel or establish belief.

Narrators cannot entirely eliminate the risk that their narrative will not find a receptive audience. However, if they are cognizant of the types of truth criteria that are likely to be applied to their narratives they can seek to maximise the chances that they will believed through adopting specific textual strategies designed to compel belief. In a different autobiographical account, published two years before the events of May 1999, Dworkin (1997, 16) describes her approach as one which seeks to compel belief through the rhetorical and affective force of her writing: “My only chance to be believed is to find a way of writing bolder and stronger than woman hating itself – smarter, deeper, colder”. A risk with this approach is that it may place excessive faith in the ability of affect to overcome seeming factual inconsistencies or difficulties, resulting in the situation where even sympathetic readers grant that a narrative possesses sincerity but not accuracy. A second danger is that a narrative approach which consistently opposes dominant cultural beliefs may alienate as many readers as it compels, resulting in the highly polarised responses to Dworkin’s work and her public identity.

An example of a more prosaic, and successful, strategy to compel belief can be seen in Alice Sebold’s (1999) account of testifying as a rape victim. She describes her testimony as a deliberate manipulation of tropes of innocence and virginity to construct herself as a believable and authentic victim: “I represented an eighteen-year-old virgin coed. I was dressed in red, white and blue” (ibid., 172). While this approach may successfully compel belief it is not without its own risks. The narrator may erase or invalidate her own experiential truth in order to harness the authority of legal discourse. She replaces her own story with one that echoes the dominant narratives of the law, inadvertently legitimating the myths that the law has put at the centre of its truth of rape (Heinzelman 1990, 100).

As I have emphasised throughout this article readers also engage in ethical or strategic relationships to narrative. Readers of traumatic narratives, particularly, are called by narrators to engage in a complex ethical act that recognises the harm done to the narrator, acknowledges the risks and significance of the narrative, and legitimates the narrator as a rational and truthful subject. An ethics of reading these narratives requires an orientation towards belief, and such an orientation forms the basis of what I have described as the feminist autobiographical pact in relation to women’s narratives of violence. They must confront the cultural tendency to cast women’s narratives of sexual violence as inherently deserving of critical scrutiny.

An orientation to belief, however, is insufficient on its own. Conferring authority onto ideologically contested narratives requires critical engagement with doxa that operate to invalidate these narratives as inherently untruthful or irrational. These two ethical imperatives do not always align. This was seen in the discussion of Dworkin’s narrative but can also be seen in relation to Sebold’s story. In the case of Dworkin’s narrative, Gracen and Orr clearly sought to undertake an ethical reading of Dworkin’s narrative but refused to validate the factual accuracy of the account or its reproduction of harmful rape myths. As this example shows, an ethics of reading may not necessarily entail belief. The question of “how” a rape narrative can be believed needs to be accompanied by questions of under what circumstances and why such narratives “should” be believed. An ethical reception of rape narratives requires adopting a set of truth criteria that can be applied to individual cases without compromising the orientation towards belief. As the reception of Dworkin’s narrative shows, developing such an ethics remains an ongoing project.

Bibliography

Bennett, Catherine (2000): “Doubts about Dworkin”. In: The Guardian (08.06.2000), p. 12.

Booth, Wayne C. (2008): “Resurrection of the Implied Author: Why Bother?”. In: James Phelan et al. (Eds.), A Companion to Narrative Theory. Oxford, pp. 75-88.

Bright, Suzie (2005): “The Baffling Case of Andrea Dworkin”. In: Suzie Bright, Inspired by Andrea. New York [unpublished manuscript obtained from the author; originally published on suziebright.blogs.com (2000)].

Brooks, Peter (2008): “Narrative in and of the Law”. In: James Phelan / Peter J. Rabinowitz (Eds.), A Companion to Narrative Theory. Oxford, pp. 415-426.

Chambers, Ross (2002): “Orphaned Memories, Foster Writing, Phantom Pain: The Fragments Affair”. In: Nancy K Miller / Jason Daniel Tougaw (Eds.), Extremities: Trauma, Testimony and Community. Urbana.

Dworkin, Andrea (1997): “My Life as a Writer”. In: Andrea Dworkin, Life and Death. New York, pp. 3-38.

Dworkin, Andrea (2000a): Scapegoat: The Jews, Israel, and Women’s Liberation. New York.

Dworkin, Andrea (2000b): “They Took My Body from Me and Used It”. In: The Guardian (02.06.2000), p. 2.

Dworkin, Andrea (2000c): “The Day I was Drugged and Raped”. In: New Statesman (05.06.2000), p. R 9.

Dworkin, Andrea (2002): Heartbreak: The Political Memoir of a Feminist Militant. New York.

Dworkin, Andrea (2005): “Through the Pain Barrier”. In: The Guardian (23.04.2005), p. 28.

Ehrlich, Susan (2001): Representing Rape: Language and Sexual Consent. New York.

English, Bella (2002): “Feminist Outsider’s Work Finds its Way Inside Ivy Walls”. In: The Boston Globe (12.03.2002), p. E1.

Ferguson, Frances (1987): “Rape and the Rise of the Novel”. In: Representations (No. 20), pp. 88-112.

Gibbons, Fiachra (2000): “Child Sex Attacks ‘Merit Vengeance’”. In: The Guardian (16.08.2000), p. 10.

Grix, Julia (1999): “Law’s Truth and Other Lies: Women, Sexual Assault and the Criminal Justice System”. In: Australian Feminist Law Journal 12, pp. 83-93.

Heinzelman, Susan S. (1990): “Women’s Petty Treason: Feminism, Narrative, and the Law”. In: Journal of Narrative Technique 20 (No. 2), pp. 89-106.

Higgins, Lynn A. / Silver, Brenda R. (1991): “Introduction: Rereading Rape”. In:, Lynn A. Higgins / Brenda R. Silver (Eds.), Rape and Representation. New York, pp. 1-11.

Kaspiew, Rae (1995): “Rape Lore: Legal Narrative and Sexual Violence”. In: Melbourne University Law Review 20 (No. 2), pp. 350-369.

Larcombe, Wendy (2002): “Cautionary Tales and Telling Anxieties: The Story of the False Complainant”. In: Australian Feminist Law Journal 16, pp. 95-108.

Lejeune, Phillippe (1989): “The Autobiographical Pact”. In: Philippe Lejeune, On Autobiography, ed. and with a foreword by Paul John Eakin. Translated by Katherine Leary. Minneapolis, pp. 3-30.

Lyotard, Jean-François (1988): The Differend: Phrases in Dispute. Minneapolis.

Norrick, Neal R. (2005): “Conversational Storytelling”. In: David Herman (Ed.), The Cambridge Companion to Narrative. Cambridge, pp. 127-141.

Nünning, Ansgar (2008): “Reconceptualising Unvreliable Narration: Synthesizing Cogntive and Rhetorical Approaches”. In: James Phelan / Peter J. Rabinowitz (Eds.), A Companion to Narrative Theory. Oxford, pp. 89-107.

Orr, Deborah (2005): “Myth, Misogyny and Some Harsh Truths About the Gender Divide”. In: The Independent (16.04.2005), p. 14.

Phelan, James (2005): Living to Tell about It: A Rhetoric and Ethics of Character Narration. Ithaca / London.

Scott, Joan W. (1992): “Experience”. In: Judith Butler / Joan W. Scott (Eds.), Feminists Theorize the Political. New York / London, pp. 22-40.

Serisier, Tanya (2007): “Speaking Out Against Rape: Feminist (Her)Stories and Anti-Rape Politics”. In: Lilith: A Journal of Feminist History 16, pp. 84-95.

Serisier, Tanya (2013): “Who Was Andrea? Writing Oneself as a Feminist Icon”. In: Women: A Cultural Review 24 (No. 1), pp. 26-44.

Shen, Dan / Xu, Dejin (2007): “Intratextuality, Extratextuality, Intertextuality: Unreliability in Autobiography versus Fiction”. In: Poetics Today 28 (No. 1), pp. 43-87.

Smith, Sidonie (1987): A Poetics of Women’s Autobiography: Marginality and the Fictions of Self-Representation. Indianapolis.

Smith, Sidonie / Watson, Julia (2008): “The Trouble with Autobiography: Cautionary Notes for Narrative Theorists”. In: James Phelan / Peter J. Rabinowitz (Eds.), A Companion to Narrative Theory. Oxford, pp. 356-371.

Trinch, Shonna L. (2003): Latinas’ Narratives of Domestic Abuse: Discrepant Versions of Violence. Amsterdam.

Yacobi, Tamar (2001): “Package Deals in Fictional Narrative: The Case of the Narrator’s (Un)Reliability”. In: Narrative 9, pp. 223-229.

Yacobi, Tamar (2008): “Authorial Rhetoric, Narratorial (Un)Reliability, Divergent Readings: Tolstoy’s Kreutzer Sonata”. In: James Phelan / Peter J. Rabinowitz (Eds.), A Companion to Narrative Theory. Oxford, pp. 108-123.

Dr Tanya Serisier
School of Sociology, Social Policy and Social Work
Queens University Belfast
E-mail: t.serisier@qub.ac.uk

Bitte zitieren Sie nicht die HTML-Version, sondern ausschließlich die PDF-Datei / Please do not cite the HTML version but only the PDF file:

1 I wish to acknowledge the valuable and insightful suggestions of the two anonymous reviewers as well as the DIEGESIS editorial team who have greatly improved this article.

2 The articles were identical apart from the opening sentence. The Guardian version begins, ‘I was in Europe’ while the version in New Statesman is, ‘I was in Paris.’

3 In what follows I will distinguish between the narrator (incorporating the implied author) and the author, meaning the real or historical Andrea Dworkin.

4 Yacobi’s (2008, 110-113) discussion usefully outlines what she considers to be five major types of integration mechanisms. However, the full typology is not necessary for my argument here which requires only recognition that readers have resources to facilitate belief in a text just as they have to facilitate disbelief.