for the perceiver to maximise congruency; in other words to make the impression fit the attribute even if this means some adjustment in the interests of creating a unified impression. One example of this is the so-called halo effect, whereby a positive feature of an individual is generalised and other attributes are made to fit this overall impression. The opposite is also observable when a negative feature generalises to attributions of behaviour to fit the negative impression. So, as Joan McGregor describes in the rape allegation made against William Kennedy Smith, the discussion of his accuser’s underwear cast her as a woman in search of sex, undermining the credibility of her complaint. Advice to defendants by their counsel is still likely to be ‘turn up to court looking presentable in a suit’, to give the impression that a good-looking and smart young man would have no need to force a sexual encounter.
Blaming the victim The idea of blaming the victim can be found in Victorian ideals which suggested that to have unsanctioned sex outside marriage was such a blow to respectability that it was a fate worth fighting literally to the death to preserve. The belief that chastity should be protected at all costs meant that failure to do so implied consent. The victim is blamed both for precipitating her own victimisation through her sexual attractiveness and also through her failure to resist when attacked. As Horvath and Brown (2010) point out, this places the victim between a rock and a hard place: greater believability is attached to notions of ‘real’ rape – the stereotype stranger attacks in the dark alley with a knife, which is actually relatively rare – whereas the more frequent, contested consent cases involving allegations against someone the woman knows tend to be more often disbelieved. In that latter instance, successful prosecution relies on scrutiny of the consent rather than the behaviour of the attacker (Stanko and Williams 2009). Blaming the victim and exonerating the perpetrator is a prevailing rape myth which experimental social psychological studies by, for example, Bieneck and Krah´e (in press) show to be attributable to schematic information processing, i.e. when people judge social information on the basis of their generalised beliefs and knowledge that are stored in their memories. They found that this tendency was more pronounced in cases of rape rather than other crimes such as robbery, when the issue of prior relationship and being drunk seemed significantly less related to the attribution of blame. Rape, they concluded, represents a special case. Brown and Horvath (2009) then argue that the operation of such culturally embedded myths acts to attenuate the perceived harm experienced by women and actually creates a reversal whereby men become the victims of sick, vindictive or vengeful women by virtue of belief in the number of false allegations that are thought to be made. There is a commonly held belief that many women offer only token resistance to sex and that when they say no they actually mean yes (Muehlenhard and Hollabaugh 1988). Young college women (N=610) completed a questionnaire designed to measure their explicit verbal behaviour and behavioural intentions whether to engage in sex. Muehlenhard and Hollabaugh report that 40 per cent of women indicated saying no to sex when actually they meant yes. The reasons for this were: fear of appearing promiscuous; feeling inhibited about sex; or more game playing. The conclusion drawn from this study was that although there was a substantial minority who did say no when they meant yes, this could be attributable to perceptions of the attitudes embedded in cultural expectations about sex. Most women did not engage in token resistance, thus ‘when a woman says no, chances are that she means it regardless of the incidence of token resistance; if the woman means no and the man persists, it is rape’ (Muehlenhard and Hollabaugh 1988: