Testimonial Smothering and Pornography: Silencing Refusing Sex and Reporting Assault

This paper defends the claim that there are two previously underexplored ways in which pornography silences women. These ways that pornography silences are (1) the smothering of refusal and (2) the smothering of sexual assault reports, and they can be expl (cid:258)(cid:349)(cid:374)(cid:286)(cid:282) (cid:349)(cid:374) (cid:393)(cid:258)(cid:396)(cid:410) (cid:862) testimonial smothering. (cid:863) Unlike the work of other writers in the pornography as silencing literature, my discussion of silenced refusal of sex deals with the cases where women have said yes to sex but would have said no if they had felt that they could have. I show that this, and cases where women do not report sexual assault, count as testimonial smothering through identifying rape myths as a species of (cid:862) pernicious ignorance. (cid:863) I make the connection to pornography in presenting evidence that pornography contributes to acceptance of rape myths. T(cid:346)(cid:349)(cid:400) (cid:410)(cid:258)(cid:364)(cid:286)(cid:400) smothering gives us a way in which pornography contributes to the silencing of women, by silencing their refusal of sex and their reports of sexual assault.

W focused largely ' L . 1,2 This paper defends the claim that there are two underexplored ways in which pornography silences women. These are (1) the smothering of refusal and (2) the smothering of assault reports, and they can be explained in part through Kristie D nt of 1 I will not discuss the influential account of silencing proposed by Langton for two reasons. First, it has already been discussed a great deal; see for example Jacobson (1995Jacobson ( B L H (2009 responses. " I A L arise. 2 Influenced by MacKinnon and Dworkin, e.g. MacKinnon (1987) and Dworkin (1981). testimonial smothering. 3 My discussion of silenced refusal of sex differs from L I deal with the cases where women in fact say yes to sex, but would, if they felt they could, have said no. These are two contexts in which women cannot speak that are so widespread that they call for explanation and move the debate from the confused ground of illocutionary silencing to the more tangible locutionary silencing, or literally not speaking. I do not claim testimonial smothering is the sole factor contributing to women consenting to unwanted sex and not reporting sexual assault, or that pornography is the sole cause of this testimonial smothering. I claim it is in these two cases and that pornography plays an important role in that testimonial smothering. Applying D demonstrating the broad utility of testimonial smothering as a concept, and in filling explanatory gaps regarding these two cases of women not speaking, where the existing literature is limited.
This paper is split into four parts. In part 1, I suggest two circumstances in women not reporting rapes and sexual assaults. In part 2, I explain an account from Dotson which describes similar kinds of silencing and categorizes them into testimonial quietening and testimonial smothering. In part 3, I apply D account of testimonial smothering to the silenced refusal of sex, and taking into account circumstances, I demonstrate that pornography contributes to this silencing. In part 4, I apply D pornography contributes in much the same way.
I do not commit to any particular definition of pornography. My arguments will hold under most definitions. They certainly hold under a radical feminist definition , sexually explicit subordination of women through pictures (MacKinnon 1987, 176). They will also hold under an everyday definition, such as sexually explicit material aimed at arousing the viewer, as long as most pornography is also oppressive and furthers harmful attitudes towards women. 4 I do not expect or intend my arguments to hold against feminist pornography or erotica, 5 or against pornography under an 3 In a surprising and encouraging instance of two people independently and simultaneously drawing the same conclusions, a similar argument to this is made in Beecroft (unpublished). 4 I am not assuming this; I spend the end of part 3 arguing that typical pornography does further harmful attitudes towards women. 5 Some feminists use this term for nonoppressive sexually explicit material aimed at arousal. everyday definition in a possible world where oppressive and harmful pornography is uncommon. 6 Whilst the area I am entering tends to treat pornography as speech, my discussion does not require commitment to this position. 7 The consequence is simply that where I think of pornography that silences, some will think of speech that silences.
Finally, I wish to make it clear that I will make no prescriptions in this paper as to what should be done about the problem of silencing. I leave it as an opportunity for other writers and researchers to discuss whether pornography can and should be restricted in order to minimize its silencing of women. I only show that pornography does contribute to the silencing of women in certain ways, though I do make the more modest claim that with this kind of silencing explained, we are in a better position to next establish what might be required in solving the problem.

Part 1: Two Failures to Speak
I suggest there are two very common circumstances in which women fail to speak, and that these should be characterized through a silencing account. These are: (1) women not refusing, and consenting to, sex that they do not want, and (2) women not reporting rape and sexual assault. In J. L. A are both instances of locutionary silencing, as they are not speech acts that failed in some way, but rather they are cases where women do not utter at all. 8 This means that they are a kind of silencing that all can take seriously, not just those A illocutionary silencing. This can help further our understanding of how pornography may providing an important alternative to the claims Langton makes.

Silenced Refusal of Sex
Many women do not refuse sex when they do not want it. Muehlenhard and Cook (1988, 64) found that 97.5% of women in a university had experienced unwanted sexual activity at some point in their lives, 9 and O " A 6 I I in this paper do not conflict with my pro-sex worker commitments; I am simply pointing to some ways that the pornography industry currently contributes to harms to women. These harms are not necessarily intrinsic to pornography. 7 For an excellent case against pornography being speech, and analysis of the related subordination debate, see Saul (2006). 8 Or, as in some cases, they do not utter refusal at all they utter consent. 9 In this study 507 men and 486 women filled in an anonymous questionnaire in a room with other respondents (seated at alternating desks to avoid being influenced by others) and a research assistant, all of the same gender. (1998,239) reported that 50% of women (compared to 26% of men) in relationships had consented 10 to unwanted sexual activity with their partner in the last fortnight. 11 In most studies, men who had unwanted sexual activity most often reported peer pressure as their reason for consenting, 12,13 whereas women more often consented out of fear or a sense of duty. Koss and Oros (1982, 456) found that 32.8% of 2,016 women students had unwanted sex because they feared that refusal was pointless as the man would be unable to stop. 14 In Muehlenhard and Cook (1988, 65), 56.6% of women reported having unwanted sex because they feared the man would terminate the relationship if they did not consent. Sandberg, Jackson and Petretic-Jackson found that 63% of women had unwanted sexual activity because they it would be inappropriate to refuse, (quoted in Walker 1997, 158). Shotland and Hunter (1995, 232) showed that 67% of women consented to unwanted sex because, as one woman said of her partner, I ; this was the most common reason given in this study. 15 Impett and Peplau (2002) reported similar findings, but additionally observed racial differences in results: while around two-thirds of all women in the study had consented to unwanted sex, 100% of the African American women in the study had done so. Though there is not space to cover it here, this shows that there is scope to address unwanted sex as a problem that not only disproportionately affects certain genders but also disproportionately affects certain races. 16 I have 10 I do not mean to suggest that nonrefusal and consent are equivalent; I mean to be talking about cases where women do not refuse, and in addition they may consent. 11 In this study, 80 men and 80 women who were in relationships kept diaries of their sexual activities for two weeks. 12 The exception is Muehlenhard and Cook (1988), who present the broadly sexual act T and 52.1% of men in engaging in unwanted sex for those reasons respectively. 13 Notice this implies the pressure comes from other men. 14 This study involved 3,862 students completing a survey on their sexual experiences. 15 In this study, 378 women students responded to a questionnaire asking how many times they had participated in certain sexual practices and why. 16 T D quietening, which demonstrates that black voices are taken less seriously than white voices.
shown here that women often do not voice refusal when they do not want to have sex, and that this is often because of fear or a sense of duty. It may be objected that the cases in which a woman says yes and where a woman says nothing are very different cases. And they certainly are. 17 However, as we will see, this does not mean there are different kinds of silencing at work. In both cases, a woman wants to say no, and that no is smothered for the kinds of reasons mentioned. This will not mean, as could be objected, that any speech that is not what the speaker would have preferred to say counts as this kind of silencing, because there are still three circumstances for testimonial smothering that need to be met, as we will see in part 2.

Silenced Reports of Assault
It is difficult to establish how many women do not report rape, precisely because they do not report it. Some women who do not report rape to the police may still rep Still we can gather information from studies and crisis centres, though we should be aware of the fact that we are potentially missing out huge numbers of women, and our figures for unreported rapes will always be too low. According to RAINN (sexual assault prevention charity and hotline) 68% of rapes and sexual assaults in the US are unreported (RAINN 2018), and according to Rape Crisis only 15% are reported in the UK (Rape Crisis England & Wales 2018).
So upwards of 68% of sexual assaults and rapes are never reported to the police; the 18 testimony on their experience is simply not spoken. These women fail to speak about their suffering, usually because of fear. Linda B two years, and found that the most common reason women had for not reporting crimes to the police was fear of retaliation and fear of police. The second most common reason was fear of family members finding out; this was mostly the case with underage women. The third most common reason was fear of not being believed, or being blamed, and fourth most common was fear of violence from their partner if they 17 The person who has sex with someone who has said yes and the person who has sex with someone who has said nothing have each done something quite different, and those differences should be attended to in discussions of that H silencing that affects what the woman can say. 18 I am aware that not all rape victims are women, but seeing as this is a paper I will continue talking about women rape victims. RAINN estimates 9/10 victims of rape are women, Rape Crisis claims 94% are women, and Koss and Oros (1982, 455) claim that according to the Law Enforcement Assistance Administration found out (Linda Belden 19791980. 19 McGregor and colleagues (2000,659) thus explaining why women raped by strangers and suffering obvious injuries are the most likely to report. 20 I have now shown that women not speaking about their sexual assaults and rapes is extremely common and often due to fears of violence and unsympathetic (fundamentally disbelieving) treatment.
It is surprising that so little attention has been paid to these cases in the W A act theory being so influential, writers on silencing have been more often concerned with the things women do say (or try to say), than the things they D illuminating exception to that rule and provides us with the tools to explain potentially many kinds of silencing.

Part D Testimonial Smothering
In Tracking Epistemic Violence, Tracking Practices of Silencing Kristie Dotson (2011) describes ways in which marginalized races, and black women in particular, are silenced. She terms this silencing a kind of epistemic violence, and separates this silencing into two kinds: testimonial quieting and testimonial smothering. I will briefly D discussing how it can be utilized in discussions of pornography. A D pernicious ignorance; this means a reliable ignorance that causes harm in a given context. This kind of ignorance is 21 (Dotson 2011, 238). It must also cause harm, but this will be different in each case (unless you subscribe to the view that all ignorance is harmful); physical and emotional harms will count, as well as social harms or harms to a movement. To give a very simple example: if some reliable ignorance causes a man to assault a woman, that ignorance is pernicious. Epistemic violence occurs in relation to testimony when a hearer fails in understanding a speaker due to pernicious ignorance. Dotson describes two kinds of silencing that fit this model of 19 The third reason listed here (fear of disbelief or blame) is also common in the study by Jones et al. (2009), but the other fears are not reflected in their study with the same frequency. This is because their study is severely limited by ruling out women who do not speak good English, who were drunk when assaulted, and who did not want photographs taken of their vaginas (418). It is hardly surprising that women who have been sexually assaulted and are afraid of further assault would prefer not to have a stranger poking around in their vagina for the purpose J 20 These authors collected data of 958 people who accessed the Sexual Assault Service over five years. 21 An epistemic gap is a lack of knowledge. epistemic violence. First, she describes testimonial quieting, when a hearer does not understand the speaker to be a person capable of knowing. For a speaker to give testimony, their audience must believe that they are knowers, otherwise their speech act will fail. When pernicious ignorance of a hearer causes them to not understand a speaker to be a knower, this constitutes testimonial quieting. The case that Dotson describes speech is not taken seriously because of ignorance about black women as knowers. I describe testimonial quiet D account and will not discuss this kind of silencing in any further detail. D testimonial smothering. This occurs when a speaker realizes that their audience may not react or understand their testimony in an appropriate way and so censors their own speech. Dotson (2011, 244) in orde audience demonstrates testimonial competence. Dotson urges us to understand these cases not as a free choice to speak (or not speak) about certain things, but rather as a coerced silencing; something about the hearer has forced the speaker to restrict their testimony. To show how this happens, Dotson provides three circumstances that tend to feature 22 in cases of testimonial smothering: 1) the content of the testimony must be unsafe and risky; 2) the audience must demonstrate testimonial incompetence with respect to the content of the testimony to the speaker; and 3) testimonial incompetence must follow from, or appear to follow from, pernicious ignorance. (Dotson 2011, 244; line breaks added for emphasis) Before explaining these circumstances, I must first explain what Dotson means by testimonial incompetence. An audience possesses testimonial competence if they have the ability to understand the speaker s testimony and to identify cases where they have not understood the testimony. An audience possesses testimonial incompetence if they would fail to understand the speaker s testimony and also fail to realize that they have not understood the testimony. 23 For an audience to understand 24 the testimony, they must find it clearly comprehensible and defeasibly intelligible (Dotson 2011, 245). This does not require that the audience clearly comprehend all of the testimony, but that the parts that the audience does not comprehend are apparent to them; get it. The example Dotson offers to illustrate this is herself attending a lecture on nuclear physics as a layperson: it will be pretty easy for her to tell when some material is not comprehensible to her (245 246).
The first circumstance refers to situations in which speaking a certain testimony may risk harmful consequences. Dotson (2011, 244 245) gives the example of women in African American communities wanting to report domestic violence, but being aware that there is a risk that hearers may understand their testimony as reinforcing racist associations between black people and violence. So, for black women, testimonies about domestic violence might be unsafe and risky because of the risk of harming attitudes to black people in general.
The second circumstance refers to situations in which the audience indicate that it is likely that they will fail to understand whatever it is that the speaker wants to say. How they demonstrate this misunderstanding can be understood very broadly, to include examples such as an eye-roll or the hearer s tone of voice indicating that they will be unsympathetic to the speaker s point of view, but in a way that relates to a particular lack of understanding. Whether this has occurred is left in the hands of the speaker: if they perceive 25 testimonial incompetence, this condition has been met (regardless of the intention). For example, if a man says to me, A indicated testimonial incompetence about periods. Testimonial incompetence has been indicated by his word choice ( really suggests disbelief) and by the need to ask the question at all; his asking shows that he has limited understanding of periods. I may then choose to silence myself, rather than risk him misunderstanding my testimony and thinking I am exaggerating or attention-W I D features, here I am persuaded by Beecroft that an amendment is necessary. Beecroft (unpublished,7 9) argues that the audience not demonstrating incompetence is not usually enough to enable a speaker to speak their testimony, as it is often reasonable for the speaker to come to expect incompetence until the audience demonstrates otherwise. Given a context where rape myths are widely believed, it has to be demonstrated by the believe rape myths, that is, that they are testimonially competent. One additional reason for accepting this amendment is that it more 25 I am going to assume that in all our cases, the perception on the part of the T while the speaker is not infallible, the speaker is well positioned to assess the level of understanding of their own speech, particularly when that speech is about something in regard to which their interlocutor is socially situated such that they are less likely to have direct access to the information (more on this later). See also note 31.
neatly parallels the third condition; for the third condition, pernicious ignorance must only appear s perspective. Since we prioritiz ctive for the third condition, it makes sense to do the same for the second. So the second circumstance can be restated as The audience must either demonstrate testimonial incompetence, or fail to demonstrate their competence, with respect to the content of the testimony to the speaker (Beecroft,9).
The third circumstance will refer to situations described in the second circumstance, but with the additional qualifier that the testimonial incompetence displayed appears to be ignorance. Remember, pernicious ignorance is a reliable ignorance that causes harm in a given context. So if the aforementioned man has a gap in his knowledge about periods, and that ignorance causes harm, it counts as pernicious ignorance. In this example, it causes harm by virtue of making me feel ashamed and making me silence myself; Dotson and I agree that someone silencing themselves can be understood as a harm in itself. 26 A less controversially harmful consequence would perhaps be that I am not granted sick leave I I not offered painkillers by him. In addition to pernicious ignorance, Dotson provides us with the concept of situated ignorance ; situated ignorance is ignorance possessed due to some social or epistemic position one holds. People in different positions have access to different knowledges, with epistemic advantages in some areas and disadvantages in others. 27 The man in my example is in the position of being male, and this position means he does not have the tools to readily/directly understand the experience of periods; he is epistemically disadvantaged in understanding female bodily functions. 28 In the same way, a white person will not have the tools to easily understand aspects of nonwhite communities, so they possess situated ignorance regarding nonwhite communities. Although situated ignorance is probably unconscious and not blameworthy, 29 it can still count as pernicious. Firstly, situated ignorance is very much reliable; white people will consistently lack information about life in black communities, and this lack of knowledge is predictable. Secondly, situated 26 At least when caused by a reliable ignorance. See Dotson (2011, 241). 27 This thought is rooted in Standpoint Epistemology. For an account of this, see Harding (1993) or Longino (1990Longino ( , 1999. 28 Again, I am aware that some men do have periods, but I am counting the man in my example as one who does not and am writing in this way for the sake of ease. 29 Although we may be considered morally responsible for making ourselves aware of potential situated ignorance that we might possess. ignorance can certainly cause harm. 30 My period-pain example is a clear instance of this; the man possesses situated ignorance about periods, and his failing to understand my period pain causes harm. Similarly, a black woman might suffer from a white person not understanding her testimony about experiences of racial injustice situated. Although situated ignorance is a kind of pernicious ignorance when it is harmful, there can also be situated ignorance that is not pernicious. It may appear that there is little difference between nonpernicious situated ignorance and everyday epistemic variations (e.g., Carli knows more about football than Dana because Carli is a footballer); however, the key difference is that situated ignorance results from broader social structures (e.g., race, class, gender, etc.) and is therefore much more difficult to overcome. 31 For this reason, situated ignorance is an important theoretical tool in understanding some types of epistemic variation.
I have now explained the three circumstances for testimonial smothering and presented one kind of pernicious ignorance: situated ignorance. I will next move on to apply these concepts to pornography, refusal, and reporting assault. Unsurprisingly, I will discuss situated ignorance men may possess with regard to information about life as a woman, as well as how pornography may contribute to testimonial incompetence.

Part 3: Silenced Refusal
For this discussion we will consider two cases of silenced refusal and D T decisive than the second, but I want to show that both count as silencing by pornography.
• Case (A): Edina does not want to have sex, but she consents to sex with Franco because she is afraid of the consequences of refusal; she has reason 32 to suspect Franco may be violent or end their relationship if she refuses. is going on in those cases. It may be that in cases where it is ambiguous whether there is a reason for belief or not, or good reasons for belief, it is more difficult to determine whether testimonial smothering is occurring, but this worry is beyond the scope of this paper.
• Case (B): Gina does not want to have sex, but she consents to sex with Harry because she feels like it is expected of her and refusing would be rude or inappropriate, 33 perhaps because she had had sex with him in the past.

I D
for testimonial smothering. 34

T
Edina certainly believes that speaking refusal would be unsafe/risky, and she is not alone. In addition to the evidence above that many women believe they are at risk of being raped or of having their relationship terminated if they refuse, women may also be afraid of nonsexual violence upon refusal. The website When Women Refuse 35 documents testimonies from women who have experienced violence after refusing sex, as well as news stories about women who were murdered, or had family members or pets murdered, as a result of refusing sex. W termination are not unfounded; there are plenty of data to show women being assaulted upon refusal. New stories appear on the When Women Refuse page every day, and 20% of women in one study had consented to sex because when they refused in the past their current partner had assaulted them (Basile 1999(Basile , 1050. 36 So the first circumstance has been met for case (A).
Although Gina may not consider herself at risk of violence to the same extent as Edina, she is risking other things. Gina thinks refusal would be inappropriate, so she is risking being seen as rude, being socially ostracized, and being in an uncomfortable situation where she must feel guilty for doing something inappropriate. I urge the reader to not be biased by E seeming greater; this is still a risk, albeit of a different kind. Again the data in part 1 show us G ot unfounded: many men believe that in certain situations a woman has a duty to have sex, 37 and that in some situations a woman has indicated she wants to have 33 By inappropriate, I mean something like seen as a surprising or disappointing way to behave, not meeting social expectations. 34

T , or fail to demonstrate their competence, with respect to the content of the testimony to
Testimonial incompetence with regard to women refusing sex seems like it would cover a broad range of attitudes. A man not understanding that she wants to sincerely refuse, that refusing can be frightening for women, and that the refusal is about her desires and not a tool to harm his ego, are all misunderstandings that men may demonstrate about refusal. More serious misunderstandings may also be demonstrated; sometimes men may give the impression that they do not understand women to be beings with beliefs and desires as complex and important as their own. In case (A), for Edina to have got the impression that refusing may be risky, Franco must have given some indication that he would not consider refusal acceptable (or at least, failed to indicate that refusal is acceptable). Whether this indication was deliberate or not, the salient factor is that Edina reasonably interpreted the situation in that way. 39 So the second circumstance has also been met. As in case (A), the second circumstance for case (B) will be met, as Gina must have understood Harry as demonstrating testimonial incompetence by virtue of the fact that she thinks he would take refusal to be inappropriate. Again we can posit the same kinds of testimonial incompetences as in case (A).

D F
ignorance? There are two likely scenarios under which it does. First, if Franco accepts rape myths, this is a clear example of pernicious ignorance. Rape myths are false beliefs about rape that tend to represent the victim in a more negative or blameworthy light and represent rapists in a more positive light, often denying that they are in fact rapists. I suggest that rape myths count as H have a right to sex (see Bouffard 2010), and showing that men should expect sex when women do certain things, like visit their home: see Bouffard (2010) and Burt (1978Burt ( , 1980. 38 F willingness to have sex (Burt 1978, 301). 39 If Franco has not given such an indication, but Edina has instead inferred testimonial incompetence from her experiences with other men, Franco has at the very least still failed to indicate testimonial competence. pernicious ignorance; 40 a false belief about rape victims is a kind of ignorance, and since rape myths are so commonly accepted, this ignorance is reliable. This reliable ignorance also tends to cause harm, as acceptance of rape myths correlates with committing rape as well as with poor treatment of victims (Burt 1980;Loh et al. 2005;Check and Guloien 1989, 170 171). 41 So belief in rape myths meets our definition of pernicious ignorance. If Franco accepts any rape myths (or Edina perceives that he does) E agency and refusal, then his apparent testimonial incompetence is a result of pernicious ignorance and thus meets our third circumstance. We can also consider Edina consenting to unwanted sex as a harm, 42 so even F ignorance was not contributing to wider harms, it has caused a harm here. A F pernicious ignorance is by virtue of the fact that he is a man in a society structured by gender hierarchy. Recall that situated ignorance is a kind of pernicious ignorance which one is more likely or better placed to possess by being a member of a particular group. As a man, Franco may well have situated ignorance in relation to information about life as a woman, including the kinds of experiences women have of refusing sex and having their desires respected (or not). Franco does not have direct access to the perspective of someone who may be afraid of refusing sex, afraid of violence, and accustomed to having their testimony taken less seriously. 43 Case (A) therefore meets the third circumstance in various ways, and we can now see how situated ignorance allows most cases to meet the third circumstance meet the third circumstance, provided there is also a harm involved, as there is here C A D  (A) and (B) are common D I suggest that real-world cases similar to (A) and (B) which meet the three circumstances will also be testimonial smothering. This is valua world experiences of consenting to unwanted sex and illustrating the breadth of D

Is this silencing by pornography?
My account provides good reason to believe that pornography is a cause of silencing. It thus helps us understand where pernicious ignorance can come from and provides a new direction for the pornography debate. 46 I first claim , and this makes pornography the source of pernicious ignorance in the cases we are considering. Then, to strengthen the connection between our case and pornography, I show that these rape myths correspond to reasons women silence their refusal. 44 Remember it is not impossible for Harry to gain this understanding; he is just not well placed to access it easily. 45 It could be objected here that this ignorance exculpates men who fail to ensure that the women with whom they are having sex really want to have sex with them; I do not mean to imply this. We should ensure the consent our partner gives us is enthusiastic and should attempt to identify and minimize any harmful ignorances we possess. 46 I do not claim pornography is the only contributory factor; there may well be I this is a singularly good reason to restrict pornography. I only mean to suggest that pornography plays a role here.
Multiple studies demonstrate pornography negatively affects attitudes to women and sex; a collection of evidence that pornography causes men to treat women more callously can be found in Zillmann and Weaver (1989). Zillmann (1989, 134 135) also found that both men and women consider rapists less blameworthy and deserving of jail time after repeated exposure to pornography. 47 Pornography-using m (Donnerstein, Linz, and Penrod 1987, 75 76;Malamuth and Check 1985). Many studies show that pornography use leads men and women to consider rape victims as less injured and rapists as less deserving of punishment. 48 All of this suggests that exposure to pornography means men are more likely to harbour attitudes less sympathetic to women and support rape myths. 49 Burt (1978) presented an influential study documenting how common belief in rape myths actually was. 50,51 Many of these rape myths contain content that is directly relevant to reasons women do not refuse unwanted sex. The best explanation of this similarity in content, I claim, is that the one influences the other. Where 50% of people believe the rape myth th a date implies willingness to have sex (Burt 1978, 301 302), a woman may not refuse unwanted sex on a date because she [she] had 52 (Shotland and Hunter 1995, 232 233), with recognition of the aforementioned rape myth informing that decision. Similarly, the 32.8% of 47 T pornography, moderate exposure (3 films a week for 6 weeks), and massive exposure (6 films a week for 6 weeks). Their attitudes were assessed one week, two weeks, and three weeks after the final film. In the third week after the final reliably recommended much less harsh sentences for rapists than those subjects of lower exposure. 48 For example, Zillmann (1989, 134 136) and Malamuth and Check (1985). 49 For additional evidence see Peter and Valkenburg (2011), Foubert, Brosi, and Bannon (2011) and Hald, Malamuth, and Lange (2013. 50 This study questioned 599 members of the public, 99 social workers who were likely to have worked with rapists, and 36 rapists, and asked to what extent they I se the general public results here. 51 This might seem dated, but Edwards et al. (2011) collect more recent data that show only fractionally better attitudes. For example, they found evidence that 25 35% of people agree with most rape myths and 66% of people agree with some (Edwards et al. 2011, 762). I use the Burt study as it is the most comprehensive. 52 Shotland and Hunter found 56% of women not refusing sex for this reason (1995, 232 233).
women who refusal would be pointless, as he would do it (Koss and Oros 1982, 456) have identified the prevalence (17%) of myths like, A woman who is drunk and has sex with a man at a party is fair game to be raped by others at the party (Burt 1978, 301 302).What these figures illustrate is that where a man may demonstrate testimonial incompetence by indicating acceptance of a rape myth, a woman may not speak her refusal because she has perceived this attitude. 53 Returning to case B: Gina H if a woman participates in other sex acts with a man, it is her fault if he rapes her, and since she has participated in other sex acts with Harry, she chooses not to refuse because . Though many cases will not have such a strong direct link between the content of his pernicious ignorance and her reasons for not refusing, this illustrates my hypothesis well; pernicious ignorance in the form of rape myths causes I have shown that in cases like (A) and (B) women experience testimonial smothering of refusal, and that pornography can contribute to a key source of pernicious ignorance: rape myths. If this is right, a change in focus in feminist discussions of pornography may be necessary, from speech act theory to testimonial smothering.

Part 4: Silenced Reports
We now turn to the second way women are silenced: silenced reports of sexual assault. In this section we explore two cases of silenced reporting of D othering to them. As in part 3, the first case will be more decisive than the second but I argue that both count as testimonial smothering.
• Case (C): Ida was raped. Ida would like to report the crime but does not because she suspects her rapist or her (male) partner may be violent if she says anything.
• Case (D): Jesse was raped. Jesse would like to report the crime but does not because she suspects she will be blamed or not believed.
Next, I D testimonial smothering. 53 It should be noted that while many of the studies into rape myth acceptance take both men and women in their samples, it has been shown that rape myth acceptance is significantly greater among men than women. See, e.g., Anderson, Cooper, and Okamura (1997) or Suarez and Gadalla (2010).

T
Ida certainly believes the content of her testimony would be unsafe or B (1979 1980) study recorded many women giving fear of retaliation as a reason not to report sexual assault; this is particularly common where the assailant is not a stranger (which is more likely anyway). 54 Again this fear is not unfounded; the assailant has already proven he is willing and able to use violence against a woman. As regards intimate partners, I could not find any data to show men being violent as a result of learning their partner has been However, given that 22% of women in Europe have experienced violence from their partners (European Union Agency for Fundamental Rights 2014, 21) and 34% of UK women have experienced violence in a previous relationship (European Union Agency for Fundamental Rights 2014, 29), the woman who gives her partner bad news (particularly news he may blame her for 55 ) has reason to fear. Belden (1979 1980, 12) describes one particular case where a , and she did not report it because before her violent and jealous husband got home. . . . She was terrified about what he would do if he found out she had been raped. I think this suffices to say that the fi I women can reasonably believe that reporting rape risks violence.
In case (D) Jesse considers the content of her testimony risky: she is risking being blamed or not believed. Jones and colleagues (2009, 420) report 73% of women chose not to report rape because they feared other people would think the women were responsible. This fear is justified; 53% of students believe (McMahon 2010). Forty-nine percent of women in J s study did not report rape because they thought some people would not believe them. Again, this fear is justified; many people believe that women lie about rape. 56 The likelihood of disbelief is greater when the victim knows the assailant, as is the case in 90% of rapes (Rape Crisis England & Wales 2018). The risk of being blamed or not believed is very real, and there is the associated risk of experiencing distress and relationship problems if the do not believe her, as well as the distress of police/courts not believing her.  1980,11). 55 When discussing case (D), we will see just how common victim-blaming is. 56 Fifty-six percent of people think women lie about rape when they are annoyed with a man (Burt 1978, 302).

T , or fail to demonstrate their competence, with respect to the content of the testimony to
If Ida is afraid that her partner will be violent, then she has taken him (rightly or otherwise) to have incompetence regarding her testimony, as violence is not the response her testimony calls for. If it is the rapist that Ida fears, then her silence might be testimonial smothering in the same way as Je she suspects the police, her family, or other hearers might not respond in the right way to her testimony. Alternatively, if Ida is only and has no concerns about how her hearers will respond, then this condition is not met, and what Ida is suffering is straightforward threats or blackmail, rather than testimonial smothering in particular. I will proceed as if it is her partner that Ida fears, for the sake of simplicity. In case (D), for Jesse to get the impression that the hearer would not believe her or would blame her, she must have perceived them as displaying testimonial incompetence. Further, testimonial incompetence is often displayed by police and attorneys regarding rape: Fortythree percent of attorneys showed moderate to high rape myth acceptance (Edwards et al. 2011), and police sometimes jeer at victims and do not file believe victims. 57 Policemen, and men in general, can be read as demonstrating testimonial incompetence in relation to rape victims. This B A a lifetime of living and learning about male attitudes, I had absolutely no (1979 1980, 10). In I J , the second circumstance has been met, as testimonial incompetence can be, and is, demonstrated by police and by other .

T
The partner who thinks violence is an appropriate response to a report of rape can be said to have pernicious ignorance. Firstly, if he accepts any rape myths, then this condition is met. For example, I s, with 53% 57 N B experience that the police never followed through with the report. Some quoted W (1979 1980, 10). Sadly, more recent studies report similar testimonies, for C K We know you were having an affair and you were having sex that morning and it all got a bit rough and you made all (Jordan 2008, 56). See also : Jordan (2011).
of students (McMahon 2010), 58 sexually assaulted, then that is a reliable ignorance that causes harm. It causes harm in this instance in particular, as Ida feeling unable to give her testimony is a harm, 59 and it causes broader harms, as acceptance of rape myths reliably predicts the likelihood of performing violence and sexual violence (Burt 1980;Loh et al. 2005;Lackie and de Man 1997;60 Check and Guloien 1989, 170 171). Secondly, the partner will possess situated ignorance, again by virtue of being a man. He ar violence from men and what it is like to be a victim. 61 As in Case (C), the testimonial incompetence in case (D) is rooted in pernicious ignorance in two ways: rape myths and situated ignorance. In cases where women are not believed or are being blamed, it is very likely that rape myths are being believed; all the examples I have drawn on here are rape myths (e.g., women lie about rape, women cause their own assaults, etc.). I will demonstrate the extent of these rape myths in the next section. Belief in rape myths counts as reliable ignorance that causes a harm not just the harm caused by rape myth acceptance in general, but also the harm for the individual who cannot testify about their assault. Secondly, policemen and other hearers who are men will possess situated ignorance by virtue of being men; that is, being people not accustomed to being taken less seriously whenever they speak, and being afraid of men. They possess the situated ignorance of being men, and not being victims, and perhaps in addition the pernicious ignorance of believing rape myths. Case (D) has now met all three of the circumstances for testimonial smothering.

Is this silencing by pornography?
I argued in part 3 that pornography influenced acceptance of rape myths. Given this, if there are common rape myths causing women not to report sexual assaults, this too has roots in pornography. In addition to the myths described previously, we can add some more with content that corresponds to reasons women do not report sexual assault. I again claim that the best explanation of this similarity in content is that the one influences the other: where 69% of people believe the rape myth that in most rapes the victim was promiscuous or 58 M M I ' M Acceptance Scale and the Bystander Attitude Scale. 59 As for Edina, it could be claimed that being unable to speak a testimony is a harm in itself. Also consider that allowing a rapist to get away with his actions and target future victims is uncontroversially a harm. 60 Lackie and de Man interviewed 86 men, assessing what factors affected their likelihood of sexual aggression. 61 Ferraro (1996) conducted 1,101 telephone interviews and found that women are much more afraid of violence than men are, and the difference is particularly strong with sexual violence.
B 1978, 301 302), a victim may choose not to report her experience people will think [she is] responsible 62 (Jones et al. 2009, 420), with her awareness of the former myth affecting this decision. The worry that people will think the women are responsible could be informed by many similar rape myths. For example, 71% of people believe that some women unconsciously want to be raped and may unconsciously bring the rape upon themselves, 33% believe hitchhikers deserve to be raped, and 48% believe that women wearing revealing cl B 1978, 301 302). The myths that women lie about rape to get back at men and that they falsely report rape for attention believed by 53% and 41% of people, respectively (Burt 1978, 301 302) on the entirely plausible belief [them] (Jones et al. 2009, 420): Forty-nine percent of women did not report sexual assault for this reason (Jones et al. 2009, 420).
These figures demonstrate that where a man possesses pernicious ignorance by accepting rape myths, a woman may read this testimonial incompetence and choose not to report sexual assault for that reason. As before, the content of the rape myth accepted by a man need not directly correlate to a ; the examples where they do correlate just illustrate the connection nicely, and give cases where the link will be particularly strong. Recalling case (D), imagine Jesse was wearing a short skirt when she was raped, but she wants to report the crime. Jesse may suspect that the police officer accepts the rape myth that women dressed revealingly trouble, so she chooses not to report the eople will think [she is] res In this case the testimonial incompetence Jesse perceives, resulting from pernicious ignorance, prevents her from speaking a T J D onial smothering. In cases like (C) and (D) women experience testimonial smothering of reporting sexual assault, and provided it is also the case (as I argued at the end of , pornography can contribute to the pernicious ignorance involved.

Conclusions
I have defended the claim that an account of pornography silencing women can D account of testimonial smothering. 62 Seventy-three percent of women did not report sexual assault for that reason (Jones et al. 2009, 420). This figure should be taken with a grain of salt; I suspect J ruling out women who had been drinking when they had been assaulted, women who did not want a vaginal examination, and women who did not speak good English.
In part 1, I introduced two very common ways in which women fail to speak: women not refusing and consenting to sex that they do not want, and women not reporting rape and sexual assault.
In part 2, I D testimonial smothering works: a speaker suspects their audience possesses chooses not to speak it at all. I explained testimonial incompetence and how this can be rooted in pernicious ignorance. In part 3, I gave two examples of women not refusing unwanted sex and demonstrated that they mee D I argued that refusing sex is risky, and that choosing not to refuse sex out of fear or out of a sense of duty counts as testimonial smothering where the choice not to speak is informed by the perceived testimonial incompetence and pernicious ignorance of the hearer. Additionally, I argued that rape myths are a species of pernicious ignorance, so a hearer of refusal of sex may possess pernicious ignorance in two ways: by believing rape myths, and by possessing situated ignorance. I ended the section by demonstrating that pornography contributes to the silenced refusal of sex by affecting rape myths. In part 4, I showed that women not reporting sexual assault also meets D s as silencing on that model. I illustrated that women may find reporting sexual assault risky because they may receive more violence and be treated as lying or blameworthy. I showed that police officers, partners, and other hearers often possess testimonial incompetence, and this may prevent a woman from speaking her testimony. I again posited that this testimonial incompetence can be attributed to two kinds of pernicious ignorance: rape myths and situated ignorance. Finally, I argued that as pornography increased rape myth acceptance in men, pornography can be seen as contributing to the silencing of women who are forced to silence their reports of sexual assault.
In sum, I suggest that D account of testimonial smothering provides a way in which pornography contributes to the silencing of women, by silencing their refusal of sex and their reports of sexual assault.