“Too late for tears, dear sister”: Constructing victims and perpetrators of rape in the advice column Dear Dolly from 1984 to 2004

This article reports on the ways in which the rape of women by men is constructed in the advice column Dear Dolly, published in the South African periodical Drum Magazine. The data collected for the study spans from 1984 to 2004, encompassing both 10 years before and 10 years after the onset of democracy in South Africa. The article uses critical discourse analysis (Fairclough 2003) as main analytical tool, but also draws on critical feminist theory (Bourke 2007). The findings suggest that there has been a decrease in explicit victim blaming after 1994, but that subtle and opaque victim blaming is still evident in readers’ letters and in the responses. These rape discourses presented in Drum after 1994 are, as Bakhtin (1981) suggests, made up of multiple voices articulating different gendered discourses. In this article, we argue that even though the use of less explicit victim blaming might seem like a positive move in the representation of rape and gender, this is not always the case. The more subtle forms of victim blaming avoid contestation and consequently often go unchecked (Fairclough 2003: 58). Additionally, new rape myths are created to mitigate the responsibility of males. These processes of subtle victim blaming and new myth-making manufacture consent and make it more difficult to counteract dominant discourses.


Introduction
During 2013/2014, a total of 46 253 cases of rape were reported in South Africa (Africa Check 2014).It is important to note that these figures refer only to reported rapes, and that it is estimated that only 1 in 9 rapes are reported to the police (Africa Check 2014).Statistics like this have prompted Moffett (2006: 129) to suggest that there is an "unacknowledged gender civil war" in South Africa.Despite efforts to address this social problem, South African women continue living in fear of sexual violence.While various psychosocial research projects have attempted to explain the dynamics behind this perpetual state of affairs, there is very little work that has explicitly targeted the societal norms and ideologies expressed in South African media discourses as their main focus (see however Bonnes 2013).It is exactly this gap in scholarly research that is addressed by this article.The focus of this article is an analysis of how victims and perpetrators of sexual violence are discursively constructed in the advice column Dear Dolly, published in the South African periodical Drum Magazine.This iconic South African magazine was selected for this study for its cultural significance and the role it played in shaping African values and traditions (Odhiambo 2006: 165).The decades' worth of archived Drum periodicals contain material so rich in cultural and ideological identification that its close analysis, in light of its immense readership, may shed light upon the dynamics of contemporary South Africa (Odhiambo 2006: 171).
The data collected for the study spans from 1984 to 2004, encompassing both 10 years before and 10 years after the onset of democracy in South Africa.The focus on the advice column is deliberate, as this particular genre is generally underrepresented in studies within the theoretical framework of critical discourse analysis (CDA).Advice columns simultaneously (to some extent) represent the voices of the readers and the editorial team.Bakhtin (1981) proposed that all texts are multidiscursive and dialogic, but one could argue that the multidiscursivity in an advice column is explicit.This makes it an appropriate genre to analyse in order to gain theoretical insight into how contradictory or clashing voices might be presented in public discourse.
This study is located within two broad and related theoretical frameworks, namely CDA and Critical Feminist Theory (CFT).CDA is built on the assumption that language projects social relations and structures as desired by (generally more powerful) participants.No linguistic form is ultimately neutral -all linguistic forms must express a stance, inflection or modality of that which is represented (Kress 1990: 90-92).The founding feature of CDA is to uncover such power imbalances in the most ordinary and unremarkable discourses, not only in those which explicitly declare their position of power or discrimination (Kress 1990: 84).One of the most prominent CDA researchers, Norman Fairclough, attributes two causal elements to the production of text, namely social agents on the one hand, and social structures and practices on the other (Fairclough 2003: 22).Fairclough (2003: 23) describes a social structure as an abstract entity that defines a potential or set of possibilities within a particular social domain.Social structure is the framework around which social practices and events are constructed, making its evaluation and potential amendment the chief object of CDA.The relationship between social structures and events is, however, not in any way direct or linear.The potential denoted by social structures is "filtered" by intermediate organisational entities to shape social events.These entities are known as "social practices" and are responsible for controlling the selection or exclusion of certain structural possibilities and their perpetuation over time (Fairclough 2003: 23-24).Social practices may, over time, also exert influence on social structure.In addition to social structures and practices, social events and their meanings are shaped by social agents.As individuals, each with unique personal influences and agendas, social agents contribute meanings to events that are not reducible to the causal powers of social structures and practices.Despite this agency, social agents -as members of a particular society -are socially constrained and invariably operate within existing social structures (Fairclough 2003: 22).
In a similar vein, in CFT, social structures and practices are seen to shape beliefs.Sexual violence perpetrated against women is regarded as a consequence of social traditions in which men have reigned over virtually all political and economic activities.This type of violence is also viewed as a medium through which male dominance and female submission are enforced and maintained (Ackerman 1995: 99).Rape is therefore regarded as a direct result of women's lack of political and economic resources, relative to men (Ackerman 1995: 99-100), allowing men to use sexuality to establish and maintain dominance and control over women.According to Ackerman (1995: 85), patriarchy recognises the potential power of women and aims to destroy women's consciousness of their power in order to protect the appropriation of female sexuality and reproductive capacity.The connection between the social structure of patriarchy and the discursive means which keep the structure intact can be seen in ways of talking about rape, especially the circulation of rape myths.These myths are unfounded beliefs held by society about sex, sexuality and gender roles that justify rape and even blame the victims for their own violation.Bourke (2007: 48) defines these rape myths as "strategies by which less powerful members of the community could be even further marginalised".
Myths are used to convert atrocities into "flaccid catchphrases that seem clear and self-evident", allowing perpetrators to contextualise sex crime in a recognised sociolinguistic framework of attribution (Bourke 2007: 24).Comments such as "She was wearing a short skirt", "What was she doing out so late?" and "Why did she go to his house?" are examples of rape myths commonly recited.According to Ackerman (1995: 198), rape myths have two fundamental functions: to accuse the woman and to excuse the man.
The aim of this article is to draw on both CDA and CFT in order to investigate how victims of rape, perpetrators of rape, and the act of rape are constructed by the selected texts.

CDA and media analysis
Media discourses have been the focus of CDA since the inception of the field (see, for example, Fowler 1991, Van Dijk 1991, Fairclough 1995).Public media are often held responsible for the cultivation and perpetuation of consciously shared perspectives.Van Dijk's (1991) Racism and the Press, for example, detailed the way media discourses, along with other forms of public discourse, construct ethnic minorities and other out-groups as "other" or inferior to the in-group.Although some theorists have argued recently that CDA fails to account for the social practices and content of media discourses (see Philo 2007), CDA continues to be a popular method of analysing media representation.Others have continued to point out the advantages that a CDA approach can offer (see Ainsworth and Hardy 2004).Ainsworth and Hardy (2004: 247) have argued that, although CDA needs other theoretical approaches, other approaches also need CDA (especially those with positivist orientations) because of CDA's criticality.Scollon (1998) argues that media discourses invariably rely upon at least two parties -newsmakers (those whom news stories are about) and the journalists who construct reports around them.Parallel to this interdependence exists a tension between authorship and principalship, the former of which Goffman (1981, in Scollon 1998: 231) defines as the role of verbalising thoughts and ideas (of newsmakers), and the latter which is defined as the role of taking ownership of that which those words convey.News journalism invariably takes advantage of this tension between authorship and responsibility to represent newsmakers and news reports within certain topical frames under the guise of neutrality.Simply put, the newsmaker of any given news story is merely a figure subjectively constructed from the characterisations and evaluations of the journalist, and it is this subjectivity the reader is presented with as fact.Along with the disparities between authorship and principalship in journalistic discourse, news journalism is governed by a set of topical frames to which events must adhere to be considered newsworthy, usually referred to as "news values" (Harcup and O'Neill 2001).It is therefore undeniable that news stories are invariably subjective reconstructions of real events.

Magazine exceptionalism
Although magazine discourse is largely removed from news discourse, it is similarly governed by journalists' constructs within certain newsworthy frames.However, unlike fact-driven media platforms such as newspapers or television broadcasting, magazines are essentially a product of their social and cultural environment and a potential catalyst for social change, a distinction which Abrahamson (2007: 667) identifies as "magazine exceptionalism".This distinction suggests that magazines -as sources of pleasure, interpretive information and identity formation of more or less clearly identifiable communities -serve as exceptionally useful indicators of any particular sociocultural reality, meriting a cultural importance far greater than their facile literary image would suggest (Abrahamson 2007: 667, Holmes 2007: 510-511).Abrahamson (2007: 669-670) attributes magazine exceptionalism to several interrelated causes.The most prominent of these are the fact that the subject or genre of a magazine identifies a particular interest-driven community, forging a uniquely intimate relationship between journalists and readers, and that magazine journalism typically involves the reader on a personal level by inspiring some sort of self-improvement.However, the theory of magazine exceptionalism remains a challenge to substantiate, with particular regard to the concept of 'causality'.It is difficult to prove whether magazine journalism initiates changes in society or vice versa.Nevertheless, despite the variety of academic approaches to the study of magazines -ranging from Frank Luther Mott's (1938) A History of American Magazines 1850-1865 to second-wave feminist analyses of the subliminal messages behind magazines' portrayal of women (see Friedan 1963, Ferguson 1983) -magazine analysis invariably revolves around the tension between the content of a given periodical and the corresponding sociocultural reality of its readers.Magazine journalism therefore proves particularly valuable for social research as it has largely been unique in its function as a simultaneously reflective and transformative cultural entity, rivalled in this respect only by the relatively recent boom of internet-based social media (Abrahamson 2007: 667).et al. (2000et al. ( , in Smolej 2010: 71) : 71) consider the foregrounding of victims of crime in news media as one of the most important qualitative developments in the representation of crime following the Second World War.Victims have evolved from having no more than a functional role in crime narratives to holding a focal position, with their trauma becoming the subject of the narrative.In this way, victims have become relatable characters whose experiences are expected to be representative of the general citizen, as opposed to atypical and unique (Garland 1999: 355).The development of tabloidisation or "infotainment" ensued, relying on emotional responses to attract audiences (Smolej 2010: 71).

Reiner
Media reporting of rape increased dramatically during the 1970s and was established as a routine feature by the mid-1980s.Soothill andWalby (1991: 22, in Berrington andJones 2002: 311) attribute this phenomenon directly to commercial imperatives, with news media using sexually titillating reports of sexual violence as "soft pornography" to hike sales.Reports of sexual abuse have essentially become voyeuristic and misogynistic, reflecting and reinforcing patriarchal values.Stories of sexual murders, for example, act as a warning to female readers not to stray too far from the safety of male protection.With reports becoming more eroticised, a certain level of male sexual violence against women is normalised as an expected part of life.Consequently, boundaries continue to be pushed to shock audiences with new angles and heightened levels of suffering in a continuum of depersonalised depictions of sexual assault and the women whom it befell (Berrington and Jones 2002: 311-312).
Recent research on sexual violence has pointed out that old rape myths are recirculated and new ones created in many contemporary reports about rape (see, for example, Meyer 2010, Bernhardsonn andBogren 2012).This phenomenon of the perpetuation of rape myths occurs especially within the context of sexual violence in relation to alcohol use by female victims.In an analysis of discursive strategies used by the Daily Mail to connect binge drinking and rape in the UK, Meyer (2010: 28) found that "rape involving alcohol is reconfigured as a problem of female binge drinking, rather than male rape".This reconfiguration is achieved by three discursive strategies: the use of direct statements, "a focus on women's drinking and omission of male violence" (Meyer 2010: 28), and a repetition of a supposed connection between female binge drinking and "vulnerability to rape".These discursive strategies lead to a regendering of binge drinking (constructed as a female problem) and a degendering of rape (downplaying rape as a male problem).Ultimately, these processes of regendering and degendering lead to the reinforcement of old rape myths and the creation of a new myth "that the problem is one of female binge drinking instead of male rape" (Meyer 2010: 31).This recent research shows that rape discourses in the media is still an area requiring academic analysis, and Meyer (2010: 32) proposes that more research is needed "to establish how wider public opinion connects to the discourses (re)produced in cultural institutions such as the media and the law".

Advice columns
According to Liao and Liao (2009: 307), giving advice has been recognised as a common discourse function with the purpose of expressing opinions to influence others' behaviours or decisions.Readers look to advice columns for authoritative yet non-threatening emotional support (Moran 1989: 119).In the study of written advice discourse, advice columns in magazines have received primary attention, with particular focus on the subliminal communication of norms and values, contributing in particular to the fields of psychology and sociology (Currie 2001, Stoll 1998).It has been determined that of all aspects of media, advice columns, together with advertisements, have the greatest intention to influence readers directly (Moran 1989: 120).
According to Locher (2010: 47), advice columns have the following features: firstly, a written exchange takes place between advice-seekers and advice-givers.This exchange typically takes place in the form of a "problem letter" and a "response letter", and in some cases incorporates features of traditional letters such as salutations like "Dear Dolly" or a farewell greeting.One of the most important features of advice columns is that the exchange is made public and although the genre seems like a personal exchange, the publication thereof means that it will reach a larger audience.Specific advice columns also develop a particular voice in response to their target audience.
Advice columns have been quite extensively investigated from the perspective of feminist theory (cf.Gudelunas 2005, Reviere and Byerly 2013), with many of these studies drawing on some form of CDA.Mutongi (2000) investigated the same advice column that this article focuses on and found that the column dealt predominantly with readers' questions about romance and sexual relationships.These columns therefore provide a more or less direct insight into the ideological perspectives of the magazine's readers and editors on sex, courtship and sexual autonomy.Gill (2009: 346), who investigated the ways in which Glamour magazine provided sex and relationship advice, found that there was a "distinctive postfeminist sensibility" in the way in which the magazine approached heterosexual relationships.Although the magazine often used discourses of female empowerment, the values they espoused were traditional values.Women were encouraged to become empowered, to get to know their bodies and their sexualities in order to find and/or keep a man.Furthermore, sex and the labour associated with becoming better at sex were often provided as the means to become empowered (Gill 2009: 365).

Dear Dolly
The fictional Dolly was initially voiced by a group of the magazine's mostly male employees, to be replaced by future editor-in-chief Liz Khumalo in 1980.The criteria by which letters were selected for print were not indicated, but the publishers of the magazine claimed that the majority of letters appearing in Dear Dolly dealt with easily relatable dilemmas such as unreciprocated love, interethnic relationships, homosexuality and premarital sex, of which the latter enjoyed most of Sis Dolly's attention (Mutongi 2000: 3).Of these questions, half were written by young men seeking advice on how to seduce women and have sex with them, and the other half by young women wanting to escape sexual advances without being considered prudish or condescending.The pressures experienced by these women to have sex were challenged by many men's insistence on dating or marrying "virtuous" women.Unplanned pregnancy was another factor for female letter-writers -for many young women, having a baby would be detrimental to their schooling and future relationships (Mutongi 2000: 9-11).In her study of the East, Central and West African editions of Drum, Mutongi (2000: 2) finds that Dear Dolly upheld these sexual double standards in that letters were replied to in a genderspecific manner.Questions were generally answered within one of two rhetorical frames -a didactic, moralistic frame for female letter-writers, and a light-hearted, humoristic frame for males.Young men lamenting women's rejection of their unwanted sexual advances were typically chided in playful or even titillating tones for their "naughtiness", while the young women who fell victim to such advances were solemnly advised to be cautious, responsible and respectable in their dealings with lecherous youths.It was clear that the editors believed that women, and not men, were to blame for sexual coercion, pregnancy and potential consequences such as illegal (sometimes fatal) abortions, loss of honour or the disgrace and lifelong financial strain of single motherhood (Mutongi 2000: 11, 13, 15).The analysis of Dear Dolly is therefore essential to an understanding of the beliefs about sexual violence held by the readers and editors of Drum during the selected time frame.

Method
Archived material from Drum was collected manually at the National Library of South Africa in Cape Town over a period of four months, as this data is not available in electronic format.Of the archived material, data was selected according to date of publication and theme.Data initially selected included all articles and advice columns published in Drum between 1984 and 2004 which focused on sexual violence perpetrated by men against women 1 .The data was first organised into salient themes and topics by means of content and thematic analyses.After this initial categorisation, the data was analysed by focusing on discursive strategies of reference, predication and argumentation, as well as through intensifying or mitigating frames (Reisigl and Wodak 2001).This was done specifically to analyse how participants were discursively constructed.Furthermore, since advice columns by definition rely on the dialogic qualities of text, Bakhtin's (1981) work on multidiscursivity was drawn on in the analysis and discussion.

Analysis
The analysis of the selected letters will be centred around the following three questions: i) How are victims and perpetrators constructed by the fictional Sis Dolly?ii) What kind of construction is created of "Dolly"?iii) How is the act of rape constructed in the problem and response letters?
From these questions and the investigation of the data, more fine-grained themes were selected, and subsequently the analysis will be organised in the following way: letters from family rape victims seeking abortions, letters from family rape victims seeking to end abuse, letters from perpetrators of marital rape, and letters from men looking to reclaim victimised girlfriends from rapists.In the time period selected, 91 letters appeared which either directly refer to or allude to sexual violence.From 1984 to 1994, only 10 letters appeared addressing sexual violence; after 1994, there is a marked increase in such letters.This in itself is significant as it shows either the ideological nature of the magazine changing towards a general tendency to discuss sexual violence somewhat more openly (see Table 1), or perhaps, as Berrington and Jones (2002: 311) suggest, the discussion of rape has become a form of soft pornography which will boost sales based on the old adage of "sex sells".More significantly, Dear Dolly shows definite ideological changes regarding the responsibility for sexual violence over the selected period of time, specifically when comparing columns published before 1994 to those published during the new democratic administration (see Table 1), with explicit victim blaming occurring more frequently before 19942 .Blame for sexual violence is mainly distributed between victims and perpetrators.A significant shift in the attribution of blame is evident over the selected period.Between 1984 and 1993, victims receive primary blame.Between 1994 and 2004, perpetrators are blamed primarily, albeit not exclusively.The following sections will analyse and compare, according to subject, all seven relevant letters and replies printed in Dear Dolly before 1994 to submissions with similar queries and their replies printed during and after 1994.The discourse strategies of intensification and mitigation, as well as referential strategies and mitigation strategies, will be identified and examined to determine (where applicable) the attribution of responsibility for sexual violence.

Letters from family rape victims seeking abortions
Between 1987 and 1988, two letters (Too late for tears (April 1987: 54) and My half-brother raped me (September 1988: 103)) appeared in Dear Dolly in which the letter-writers sought advice on obtaining an abortion after being raped by family members.In both cases, Dolly attributes at least part of the responsibility for the rapes and consequent pregnancies to the victims.

Letter 1: Too late for tears
The writer of Too late for tears (April 1987: 54) seeks Dolly's advice on how to obtain an abortion, having been raped and impregnated by her half-brother two months prior.Dolly mitigates the assault and resulting pregnancy within a reproachful frame.The woman's desire to have an abortion is dismissed as "very cruel", as she "should have known better from the start, that keeping quiet about such a nasty need, will land [her] into hot soup one day".Although the victim is not explicitly blamed for the assault, Dolly's claim that she "should have known better" implies that the woman has acted foolishly and is therefore responsible for her pregnancy.This inference is reinforced by Dolly's dismissal of the woman's victimhood with the reproach, "it is already too late for tears, dear sister".
A more explicit instance of victim blaming follows, as Dolly infers the victim's compliance during the assault from her reluctance to report the crime -a common rape myth: "I take it you were also impressed with what your half-brother did to you -otherwise you would have reported the matter to the police immediately".Dolly then repeats her warning against an abortion, this time claiming that "[risking] your life for an abortion won't be worth it", and that the woman's parents "won't be keen on an abortion either".Dolly concludes her reply with encouragement to take responsibility for the child: "Sister Dolly's advice is that you should take it upon yourself that you are going to become a new mother and look forward to bringing up your 'baby' with affection.Learn to love it now and do everything in your power to save its life."The referential expression "Sister Dolly" disguises these admonishments as the guidance of a caring relative.It is clear that the editors responsible for Dear Dolly had little sympathy for this particular victim of sexual violence, and placed the life of her unborn child above her victimhood and autonomy.

Letter 2: My half-brother raped me
In My half-brother raped me (September 1988: 103), Dolly's response mirrors the discrimination illustrated in the previous example.The advice-seeker in this case is a woman in need of guidance, having been raped by her half-brother and in need of an abortion.As with Too late for tears, Dolly responds reproachfully.She states that she's "just not in a position to encourage abortion" and explicitly blames the victim for the pregnancy, and, therefore, the rape: "In short, you are also to blame for taking this abuse and sparing him.You should have reported him to the police immediately" The woman is then advised to relate the issue to her parents and seek further advice from social workers.

Letter 3: Pregnant after uncle raped me
As in the pre-1994 examples discussed above, the letter-writer for Pregnant after uncle raped me (May 1994: 70) seeks advice on how to obtain an abortion after being raped and impregnated by a family member.Although Dolly similarly advises the letter-writer not to abort the pregnancy and to seek counsel, her response is less reproachful.As before, Dolly rejects the possibility of an abortion.However, instead of dismissing abortion as cruel, she simply states that "[she] personally [doesn't] encourage abortion".She also chastises the victim for not reporting the assault sooner.However, instead of blaming the victim's foolishness, she presents a more neutral reprimand, claiming that the victim "shouldn't have suffered in silence to protect a rapist".The victim is then advised to discuss the matter with her parents and seek further guidance from local social workers.Although the latter advice illustrates a degree of sympathy, the critical tone induced by the expression "you shouldn't have", paired with the relatively dispassionate content and style of the response as a whole, mitigates the victim's experience of the violence and reflects a matter-of-fact perspective on rape with little consideration for its causes or effects.

Letters from family rape victims seeking to end their abuse
Prior to 1994, two submissions discussing family rape (excluding those examples surrounding the obtaining of abortions) were printed in Drum's Dear Dolly column.In both these submissions, the first entitled My father rapes me (February 1992: 96) and the second entitled I'm better off dead (November 1993: 94), Dolly affords sympathy to the victims of the respective assaults with underlying suggestions of victim blaming.

Letter 4: My father rapes me
The writer of My father rapes me (February 1992: 96) seeks counsel on the increasing sexual demands of both her father and her boyfriend, the former whom has been raping her since she was eleven years old.Her reason for writing the letter is not to report the abuse, but to seek a solution to her inability to "cope with both their sexual demands".Dolly's response starts with the following assertion: "[Y]ou shouldn't subject yourself to such abuse.It is your right to say 'No!'".Despite the letter-writer's concern of managing the sexual demands placed on her, Dolly turns her focus to the victimhood of the letter-writer at the hands of her father.She claims to be "disgusted about what [the victim's] father makes [her] go through and [Dolly is] not going to say 'let bygones be bygones'.No way!"She then advises the letter-writer to "[take] this matter further, report him to the highest authority![…] He's an animal!"Only then does Dolly shift her focus to the victim's boyfriend.Although the letter-writer reported no abuse from her boyfriend, only that he "is very fond of sex", Dolly warns her that her boyfriend is "out to use [her] as much as he can and when he's ready to settle down he will go for a girl who is able to say 'No!'" Despite her focus on the letter-writer's victimhood, the referential phrase "a girl who is able to say 'No!'" implies that Dolly blames the letter-writer for not resisting what she (Dolly) perceives as abuse from the victim's boyfriend.This attribution of blame indicates that Dolly attributes at least a portion of the blame for sexual abuse to the victim's lack of resistance.

Letter 5: I'm better off dead
The letter entitled I'm better off dead (November 1993: 94) is a rare example of the attribution of blame to an external party for a particular instance of sexual violence.The letter-writer in this case is a sixteen-year-old girl who has been sexually abused by her uncle for two years and whose ignored cries for help to her mother are causing her to consider suicide.She "feels dirty and used" and has "done all [she] can to alert [her mother] to the situation".
In response, Dolly chastises the letter-writer's mother for not maintaining an open relationship with the victim.She claims that if the victim and her mother "were close enough, [the victim's] uncle's abuse wouldn't have gone this far".The victim's mother is blamed explicitly: "The blame lies squarely on your mom's shoulders and it's time she became a real mother.She has to protect you".The perpetrator is condemned for his transgression, and is said to "[deserve] a long jail sentence for his filthy deeds".A contradiction follows -whereas the letter-writer is advised not to "let this happen to [her] again," implying that the victim is allowing the abuse, Dolly also asserts that "it's not [the victim's] fault".One may conclude from the latter quote as well as the referential expression "filthy deeds" that Dolly intends to aid the letter-writer in escaping the abusive situation.Contrarily, Dolly's warning not to "let this happen to [the victim] again" reveals an underlying ideology of victim blaming.

Letters from perpetrators of marital rape
Only one letter from a perpetrator of marital rape was printed in Dear Dolly before 1994.The letter entitled I am forced to rape her (April 1988: 66-67) is a rare example of a perpetrator of sexual violence looking to Drum's Dolly for advice on his sexual predicament.

Letter 6: I am forced to rape her
I am forced to rape her (April 1988: 66-67) details the letter-writer's wife who has "started giving [him] a number of problems," including her sudden refusal to cook, clean, bathe before bedtime and have sex.The perpetrator complains that his wife "also won't let [him] have sex with her.[He has] to battle to get her to such an extent that [their] double-bed once broke down [sic] during [their] struggle.It's more of a rape than lovemaking".When questioned about her behaviour, his wife orders him to "go and pay the balance of the lobola3 to her parents".He then argues that he should not have to meet his wife's demands when "she has already shown [him] her true colours".
Despite the letter-writer's admittance to raping his wife, his crime goes unrecognised in Dolly's response.Instead, he is advised to fulfil the prerequisites of the lobola custom in order to "bring any relief" to his "dicey" problem.Dolly suggests that the man's wife "has been told [by her family] to discipline [him], so that it becomes a constant reminder to [him] that [he] must complete the required [lobola] procedure first".She then reassures the husband that he won't be "[wasting his] trouble on something that is no longer worth the trouble [since he has] nothing to lose because [he] already [has] a child with this woman".Dolly's failure to recognise the man's actions as immoral and illegal reveals the editors' assumption that rape, or at least marital rape, is not only excusable, but problematic only if the rapist's needs are not met.

Letter 7: Does my wife have the right to say no?
The response to the letter entitled Does my wife have the right to say no? (December 1997: 40) may be compared to the pre-1994 example discussed in the previous section, as both are written by the perpetrators of marital rape.In this instance, the husband writes that "[everything] is fine except she often isn't in the mood to have sex -and when she says no she means it".The wife is reported to have asserted her "legal right to refuse to have sex with [him]" and to have threatened to have her husband arrested for rape.The writer concludes his letter with the declaration that "[if] a man says please he should at least be considered".
Dolly's reply is morally ambivalent.While Dolly states that "[neither] of [them] has the right to force the other to have sex," Dolly empowers the perpetrator by adding that "neither of [them] has the right to refuse sex all the time either.Making love always has to be a reasonable compromise".The advice-seeker is advised to evaluate his relationship with his wife as a possible reason for her rejecting him.He is also advised not to "[force] her to have sex".
Contrary to her response to the inquiry about marital rape in 1988, Dolly acknowledges that marital rape is unacceptable, although the word "rape" is substituted by the more impassive referential expression "forced sex".The illegality of the attempted rapes is, however, not recognised.Instead, the sexual violence is treated as a private matter.In addition to this mitigation, the happiness of the perpetrator is prioritised along with that of the victim with the suggestion that he reveal to his wife his "lack of joy in [their] marriage" in order to resolve the potential marital problems blocking his sexual access.

Letters from men looking to "reclaim" victimised girlfriends from rapists
This category of letters to Dear Dolly introduces yet another group of letter-writers -a third party seeking counsel on a sexually abusive situation which involves a love interest or partner.Such letters printed before 1994 exhibit an unexpected query from the advice-seekers.Rather than seeking help for the rape victims, these men are concerned about how the sexual violence, construed as consensual sex, might affect their relationships with their abused partners.Although Dolly's responses include discussions of the criminal aspects of the abuse, her language use mirrors the perspectives of the men that write in and infers the main issue as the state of the relationship rather than perpetration of sexual violence.

Letter 8: Must I leave her?
Must I leave her? (June 1988: 68), written by an admirer and possible love interest of a victim of sexual assault, seeks advice on how to re-establish his relationship with the victim after she "fell in love with another guy" who she claims "forced himself onto her by beating her up".The advice-seeker is "so scared because this man threatens to give [the advice-seeker] a hiding if [he doesn't] leave this girl alone".He adds that he is "not prepared to get involved in fighting over a woman," yet does not want to end a "four-year-old affair because of a guy who only entered the scene a couple of months ago".From the predicational expressions "fell in love" and "fighting over a woman", and the referential expression "a guy who only entered the scene a couple of months ago", it is clear that the advice-seeker interprets his partner's rape as a consensual sexual relationship which would nullify their four-year-long relationship.
Dolly's response continues this assumption as she warns the advice-seeker that "behaving like a thug and fighting over women has always been wrong [as it would] put [the advice-seeker's] dignity and life at risk over a hooligan".She then asserts that "this issue doesn't involve [him] in any way [and] it is up to [his] girlfriend to see to the disciplining of this guy, long before deciding on giving in to his nasty demands".Here, the expression "deciding on giving in" perpetuates the myth that rape is impossible as it would require the victim's cooperation, or that protestation of sexual advances is merely a show of modesty or sexual inexperience and a natural part of the courting ritual (Bourke 2007: 24, 50).The victim is then reproached for "not [putting] enough effort into getting this matter corrected," attributing implicit blame to the victim for the assault.
The abovementioned rape myths are explicitly promoted with the claim that "[i]t's simple, she must approach this guy and tell him it's over, so that when he reacts by beating her up, she can easily report him to the nearest police station".This strategy to have the victim beaten up illustrates that the letter-writer's happiness is prioritised over the victim's safety and wellbeingagain, the victim's victimhood is questioned.Dolly advises the advice-seeker to "be very careful, because [her] instincts tell [her that the letter-writer's girlfriend] might be in love with the guy and [she is] only fooling [the letter-writer] with those excuses.This [he] can only detect if she doesn't agree to tell this guy off".

Letter 9: She was sexually abused
The letter entitled She was sexually abused (April 1991: 96) is written by a young man who wants to reconnect with his estranged girlfriend who was repeatedly raped by a housemate and gave birth to her rapist's child.The letter-writer is unsure of what to do for fear of "the child's father [refusing] to let go," and states that he "does not want to use force".Dolly starts her response by questioning the victim's credibility.She advises the letter-writer as follows: "[H]ear both sides of the story.If your girl wasn't being forced, why didn't she report the man to the police?"She then justifies this assumption by saying, "Yes, I know you might say she was stranded and couldn't take the risk".Dolly suggests that the letter-writer goes to "one or two people who were close to her, or neighbours, [and he] might come up with something solid".He is also told not to use force and endanger his life, and only to "decide on steps to take after clarifying her relationship with the other guy.And if all what [sic] she says is true, don't hesitate to report the matter to the police".As Dolly preoccupies herself with the legitimisation of the victim's claims of assault, rather than dealing with the injustice which the advice-seeker describes as fact, one may infer that sexual victimhood does not concern her as much as the advice-seeker's relationship dilemma.

Letter 10: Teachers snatch our girls
This letter (April 1994: 56) is the only post-1993 letter which is comparable to Must I leave her? of 1988, as it is also written by the third-party partner of a victim of sexual violence.The writer of this letter is a schoolboy lamenting that "every time [he gets] involved with a girl, a teacher comes along and snatches her".The possibility that the term "snatches" may be metaphorical or hyperbolic is dismissed with the complaint that "[his] current girl believes it will be best for her to leave school because the male teachers are harassing her".Dolly initiates her response with the declaration that it is "a disgrace that teachers harass schoolgirls.
[He], and this girl, can put a stop to this problem".She advises the boy to "take the matter up with the headmaster".Thereafter, Dolly again mitigates sexual violence as consensual sex and discounts the victim's safety in favour of the advice-seeker's academic achievement with the warning that "[she is] not suggesting that [he] should start competing for girls with the teachers […] [he should] forget parties and shebeens and concentrate on [his] studies".

Discussion and conclusion
The analyses of letters and responses in the previous sections illustrate not only how discourse strategies were employed to attribute blame to the victims of sexual violence, but how the distribution of such blame shifted dramatically in 1994.A possible reason for this shift may include the overall transformation of the media in democratic South Africa.In 1994, the South African media became an instrument of change by providing a platform for socio-political transformation and the construction of post-apartheid identities (Wasserman and De Beer 2009: 383).After 1994, non-governmental media watchdog organisations such as the Media Monitoring Project, Genderlinks and the Freedom of Expression Institute, also worked towards maintaining the transparency of South African media (Wasserman and De Beer 2009: 385).
Although there is a shift in explicit victim blaming, implicit blaming remains.Perpetrators were generally absolved from blame; in fact, on only one occasion in the letters analysed here was the perpetrator referred to by Dolly as a "rapist".On one other occasion the perpetrator was referred to as an "animal".The fact that perpetrators were rarely blamed can also be seen in the use of words that refer to the act of sexual violence, with "rape" not being used once.Instead, referential or predicational expressions such as "filthy deed", "dicey problem", "forced sex", "nasty demands", "abuse", "nasty deed" and "to snatch" are used.Dolly also seemed to construct herself as a family member giving kind and sensible advice.This is achieved by either addressing letterwriters in a familiar tone ("Dear Sister"), or by referring to herself as "Sister" (for example, "follow Sister Dolly's advice").The post-1994 Dolly is loaded with contradiction -on the one hand, explicit victim blaming is less prominent, yet on the other, conservative ideologies remain about women and their autonomy over their sexuality.As Gill (2009) cautions, it is not enough to simply point out contradictions in magazine discourses; one also has to ask what these contradictions mean, and what is achieved by these discourses.
Although it might seem as if a decrease in explicit victim blaming after 1994 might be a positive move, implicit victim blaming is equally problematic.Such assumed meanings are of particular ideological power as they avoid contestation and consequently often go unchecked (Fairclough 2003: 58).These meanings often become incontestable through a process of erasure.Irvine andGal (2000: 37, in French 2012: 344) define "erasure" as "the rendering of some social phenomena invisible in ways that simplify a social field".Discourses deconstructing social structures that render women powerless in public spaces are rendered invisible.Instead, these more complex discourses are simplified and presented as "commonsense".Women are constructed as having the power to decide who they can have sex with, and having the power to decide to lay charges without sanction and fear.However, the conditions which very often force women not to report rape are not scrutinised, and are in fact reinforced by rape myths such as "If it was rape she would have reported it immediately after it happened".Similar to Meyer's (2010) findings, rape is degendered as not a male problem.Whereas in Meyer's (2010) study women were constructed as responsible for their rape because of binge drinking, here women are blamed for not reporting rape.Not only does this shift the focus away from rape as a male problem, it also fails to highlight structural inequalities (for example, the way rape victims are handled by the police) which make it very difficult for women to report rape.
Advice columns provide a glimpse into the dialogue between the editorial team and the readers of a publication.Dolly not only responds to the readers but also draws on other social discourses, such as health and safety or punitive discourses, while readers' letters very often reflect prevailing dominant discourses, such as rape as "rough sex" rather than a violation.As Scollon, Tsang, Li, Yung and Jones (2004: 174) state, "all texts are uttered in multiple voices in response to multiple voices and in anticipation of polyvocal responses".However, where the editors could have used the dialogic qualities of the advice column to "talk back to dominant ideologies about rape" (for example, by educating readers about what the legal definition of rape is through their response letters), they instead respond to the readers with the same kinds of ideologies that the readers themselves use.Instead of becoming a true dialogue then, opposing ideologies are rendered invisible, and the multiple voices in the advice column do not evoke polyvocal responses but rather multiple similar responses.
Even though some strides have been made in the way in which victims and perpetrators of sexual violence are portrayed in the media, such as a decrease in explicit victim blaming after 1994, subtle victim blaming is still rife.As it is generally acknowledged that the media can shape public opinion, this is an important finding.This article has focused attention on advice columns and has shown that through response letters adherence to dominant ideologies is ensured.Advice columns could be a space for changing public opinion about rape, but instead become a hegemonic device to reinforce old rape myths and create new ones.

Table 1 :
Attribution of blame in Dear Dolly