“Would your level of disgust change?” Accounting for variant reactions to fatal violence against women on social media

Abstract

The murders of Sarah Everard and Sabina Nessa, occurring in similar contexts in London over the course of 2021, prompted renewed public discourse around violence against women and the nature of stranger-perpetrated murder of women in British society. It also provided the opportunity to analyze our responses to such crimes as a community and, in particular, our expectations and assumptions about who is committing fatal violence against women. In this study, Facebook comments (n = 414) pertaining to the first identification of the alleged murderers in each of the above cases were analyzed for sentiment. This analysis revealed major differences in the levels of shock and/or surprise at Everard’s murderer (a police officer) being identified, compared with Nessa’s alleged killer (a migrant). The article assesses the divergent responses in each case and explores the reasons that allegations of migrant-committed crime appear to attract significantly lower rates of resistance than allegations of police crime.

Keywords

Migrant police crime sentiment analysis social media violence against women

For many people, and for a variety of valid reasons, the 2-year period of 2020/2021 will be remembered as a time of great upheaval. The catastrophic global impact of the COVID-19 pandemic aside, this period was also characterized by the public interrogation of race, power, and colonization raised by the 2020 Black Lives Matter protests and, in the United Kingdom, by the complicated arrival of the oft-delayed “Brexit” withdrawal from the European Union (Raymen et al., 2021). Set against this backdrop, in March 2021 yet another tragic event occurred that resonated with a significant cross-section of the British public and, especially, the nation’s women: as she was walking home from a friend’s home in the south London district of Clapham on 3 March 2021, 33-year-old Sarah Everard vanished. Six days later, Wayne Couzens—a 48-year-old serving officer with the Metropolitan Police Service (MPS)—was arrested and, later, charged with kidnapping and murdering Everard. He later pleaded guilty to charges including kidnap, rape, and murder and was sentenced on 30 September 2021 to a whole life term in prison (Dodd, 2021). Everard’s murder sparked widespread discussion in the United Kingdom around the safety of women and, specifically, the impact of gendered violence in society. The murder also led to a vigil on Clapham Common, near where Everard disappeared, on 13 March 2021, which garnered critical attention after being aggressively dispersed by MPS officers, just days after one of their own was arrested in connection with Everard’s death (Graham-Harrison, 2021).

Several months after Everard’s murder, the specter of violence against women (VAW) was reawakened as a result of the murder of 28-year-old teacher Sabina Nessa on 17 September 2021. Nessa was killed by 36-year-old Koci Selamaj in a park in south-east London, in circumstances that mirrored Everard’s death earlier in 2021 (Kwai, 2021). As with Everard, Nessa’s murder prompted discussion of women’s safety in the United Kingdom, and her accused killer became the focal point for robust public discourse around VAW. However, while the context of Nessa’s death was unquestionably analogous to Everard’s murder, public commentary on Selamaj attracted particular attention for a distinct reason: just as the rhetoric around Couzens was centered on his role as a serving police officer, discussion around Selamaj’s (then) alleged crimes was focused on his identity as an Albanian migrant residing in the United Kingdom (Thorburn, 2021).

Drawing on the conceptual lens of symbolic interactionism and (particularly) social role theory, this research aims to analyze the contours of public response through a content analysis of initial reactions on Facebook, via posts from a variety of news agencies in which the identity characteristics of the alleged murderers (police officer and migrant) were revealed for the first time to the public. Analyzing the comments across five articles pertaining to Couzens and Selamaj, posted by five news sources of diverse ideological and/or political persuasion, over 400 unique responses were able to be assessed through sentiment analysis. This sentiment analysis revealed a far greater defensive posture from the public in its reaction to Couzens being named as Everard’s alleged killer when compared to the identification of the migrant, Selamaj. Although the rate of negative comments in relation to Couzens’ role as a police officer was generally comparable to those regarding Selamaj’s status as a migrant, the number of comments doubting Couzens’ guilt were 3.5 times higher than those registered for Selamaj.

Furthermore, the sentiment analysis also accounted for Facebook comments which were neither positive or negative about role status, but nevertheless expressed shock or surprise at the fact that (a) a police officer or (b) a migrant would commit a crime such as the murder of Everard or Nessa. In this criterion, perhaps the most obvious result of the sentiment analysis was recorded: while 19% of comments about Couzens registered shock or surprise that a police officer would be involved in murder, not a single comment regarding Selamaj expressed the same reaction in relation to his migrant status. This sentiment analysis reveals clear distinctions regarding how the public responds to the revelation of a police officer involved in serious VAW versus their response when a migrant is involved. It offers insight into our sociocultural perspective on issues of identity and complicity, and has major implications not just for how the community perceives those accused of VAW, but who they perceive (and expect) those people to be.

Background and current research

In their longitudinal study (1994–2014) of the official Crime Survey for England and Wales (CSEW), Walby et al. (2016) identified a considerable increase in VAW beginning in 2009, and while much of this increase was constituted by domestic offenses, there was also a 20% rise in VAW committed by strangers, as in the Everard and Nessa cases. Recent data from the Office for National Statistics (ONS) from June 2021 revealed that 89% of British women reported experiencing harassment in a park or public space, which is particularly relevant given the geographical setting of Everard’s abduction (Clapham Common) and Nessa’s murder (Cator Park). This self-reported harassment has support in official crime statistics, with the ONS reporting that around 9% of rape offenses take place outside the victim or offender’s home—while a minority compared to assaults in a victim or offender’s home, this nonetheless reflects “public space” assault occurs at a rate that is not insignificant (ONS, 2021b). The gender disparity in this type of violence is reinforced by CSEW data from 2019 to 2020 which reflected that 31% of women reported feeling unsafe walking alone after dark; this figure is compared with 13% of men, signaling this as a key area of concern in terms of VAW (ONS, 2021a).

Liz Kelly (1988) argues for reimagining VAW not as behavior that is aberrant or episodic but, rather, a normative and functional aspect of patriarchal society that typifies “normative” gender relations between men and women. While severe cases of VAW such as murders committed by strangers unquestionably face social sanction, it is reductive to conceptualize these events as isolated cases of deviance when, in reality, they are simply extreme ends of a spectrum that has its origins on a more tolerated (even socially sanctioned) misogyny. In evaluating pervasive societal “rape myths,” Fanghanel (2017) notes that “women’s fear of crime has been described as the spatial expression of patriarchy . . . through which women are perpetually warned of their vulnerability’ to control their movements and, thus, agency in the public space” (p. 2). Ray Surette (1994) wrote that

the modern mass media have raised the spectre of the predator criminal from a minor character to a common, ever-present image . . . they represent a largely unquestioned set of beliefs about the world, a constructed reality that, as the aphorism “perception is reality” suggests, has the ability to shape the actual world to fit the media image. (p. 132)

As this paradigm suggests, public willingness to accept that an offender is accused (or guilty) of a stranger-perpetrated VAW offense is informed by limited data, mostly supplied by a mass media which places disproportionate and (often) sensationalist attention on a small set of cases.

Central to the comparative analysis of public responses to Couzens and Selamaj is the issue of role identity—specifically, in this case, the “role” of police officer (Couzens) and migrant (Selamaj). Although the social status of policing has been tarnished in recent years, the research affirms the contention that police officers themselves “generally enjoy fairly high levels of public support” (Moule, 2020: 1). There has been a “surprisingly thin British research literature on public confidence in policing” compared with similar research conducted in the United States (Jackson and Sunshine, 2007, 215). Jackson and Sunshine (2007) offer one of the few empirical studies of public attitudes, finding that elements like perceived procedural fairness, social identification with police, and perceptions of public trust are key indicators of positive opinions toward police; they surmise that “the public expects the police to both defend prevailing norms and embody local values . . . Confidence drops when they are perceived as failing to do so” (Kautt, 2011: 357). Paula Kautt (2011) argues that, after a series of corruption scandals and misconduct allegations in the 1970s, the British public adopted a more negative disposition toward the police, and “while support for the police in Britain has recovered somewhat in recent years . . . it has never regained the stellar heights once enjoyed” (p. 355). In all, the British-focused research on public opinions on police does not provide clear answers as to positive, or negative, sentiment: while problematic, this serves as a gap that this study can (in small part) help to fill, as we work toward a more holistic understanding of the British public’s relationship with the policing profession.

Compared to the deficit of literature on perceptions of police in the United Kingdom, a considerable research base exists that examines public opinions on migrants in the country—especially in a Brexit (or post-Brexit) environment. Eberl et al. (2018) observe a distinction in the treatment of migrants is carried out in the British press, with tabloids and midmarket newspapers engaging in “a deliberate use of the terms ‘migrant/immigrant’ to delegitimize the refugees’ or asylum seekers’ dire political and personal circumstances” (pp. 209–210). James Morrison (2019) refers to this “othering” campaign as one that cast the country as a “vassal state whose financial over-generosity and porous borders had left it vulnerable to invasion and exploitation by opportunistic migrants from parasitical poorer countries” (p. 595). In a national survey on public views on the “threat” posed by migrants, Stansfield and Stone (2018) found that narratives that migrants pose an “economic threat” did not have as much impact on public support for curtailing their rights as constructions of migrants as “criminal threats”; in their view, this indicated “that British citizens invoke deep rooted stereotypes about EU migrants as criminal . . . [which is] associated with support for more punitive criminal sanctions” (p. 592). As this study is concerned with the more vitriolic and punitive response to Selamaj, versus Couzens, these findings on the impact of criminal threat on public opinion of migrants are highly relevant.

Methodology

The contemporary digital landscape has resulted in social media platforms like Facebook assuming the position as primary sites for discourse in our society, where individuals who are “connected” with each other (either in real life or often solely via the social network) are able to share their views, beliefs, and opinions in a relatively public forum. Because of this function of social media, the medium is primed to provide content for sentiment analysis, a method where typically the polarity of a subject’s response to stimulus is measured and categorized as “positive” or “negative” (Gonçalves et al., 2013). In this case, the text analysis software Linguistic Inquiry and Word Count (LIWC) was utilized as a primary tool to determine both polarity and the presence of shock and/or surprise in relation to the alleged offenders’ social role in the Everard and Nessa cases. Beyond polarity, LIWC “evaluates emotional, cognitive, and structural components of a given text based on the use of a dictionary containing words and their classified categories” (Gonçalves et al., 2013: 29).

Using LIWC allowed not only for better categorization of sentiment, but also in filtering out comments that were designated “neutral” or not demonstrating any relevant sentiment, such as comments where the only content was tagging another user. To assess the dimension of shock and surprise, a bespoke LIWC dictionary was adopted to encompass relevant and otherwise key terms. While the entire list of terms would be too extensive to list, words correlated with positive sentiment included “innocent,” “unfair”/”fair,” “not guilty,” and “witch hunt.” Words associated with negative sentiment included “guilty,” “justice,” “hate,” “disgust,” and “shame.” The neutral category was analyzed separately, with posts displaying clear sentiment that were not detected by the LIWC process designated to the positive or negative category manually or, otherwise, confirmed as legitimately neutral. The utilization of any algorithm is not foolproof and requires manual analysis for “data cleaning.” On several occasions, there were instances where positive or negative sentiment was recorded and yet, on review, the comments in question were deemed to be irrelevant, or spam (Thakor and Sasi, 2015). There is also a need to cross-check the data to account for comments which displayed characteristics of sarcasm which may appear, on a superficial level, to express one emotion but, taken in context, are actually making a contrary point (Maynard and Greenwood, 2014).

The selection of comment sources was undertaken with the goal of choosing Facebook posts that best reflected the diversity of news sources (and, thus, news consumers) operating on social media. More than many other jurisdictions, the British media has a tradition of being polarized along ideological and/or political lines (Aldridge and Evetts, 2003). In recognition of this context, news outlets were chosen to reflect a representative balance: one outlet typically identified as left-leaning (The Mirror), two usually seen as right-leaning (The Times and The Sunday Times and The Daily Mail), and two that are considered centrist sources (The Metro and CNN International). The selection of CNN International—the sole non-British source—was purposeful: aside from providing another centrist perspective, as a global news source it offers a set of commenters who may be less partisan than other, more parochial British media followers and, thus, contribute a further layer of balance to the results.

Two social media posts were selected from each news outlet’s Facebook page: the first from March 2021 in which Couzens is named as Everard’s alleged murderer, and another from September 2021 wherein Selamaj is first identified. To limit off-topic content, the content included was limited to first-comments only (e.g. no replies to existing comments). The result was an approximately equivalent number of comments on each case—212 total comments on posts pertaining to Everard’s murder and 202 pertaining to Nessa. While there was variation in the total comments across news sources, a similar total was analyzed overall for each case, enhancing comparative reliability in a holistic sense. Overall, 414 comments were subject to sentiment analysis, serving as the sample informing the researcher’s observations.

Results

Data collection commenced with the selection of a total of 10 Facebook posts, published by a selected grouping of five “traditional” media organizations either based in the United Kingdom or with a British focus.¹ All were complemented by a traditional news article on the subject, with the URL included in the social media post. Text with posted articles was short in all instances (under 25 words, on average). In total, 414 comments responding to these posts were analyzed for sentiment (212 on the Everard posts; 202 on the Nessa posts). The spread of comments-per-post was variable, with some sources tending to dominate in terms of how many comments they received. For example, The Metro posts alone accounted for 171 of the comments analyzed here (41.3%/total)—and 48% (n = 102) of those regarding Everard’s murder alone. Total comments from CNN International posts made up 17.9% of the total (n = 74); a similar 15.7% of total posts came from The Mirror (n = 65). A slightly higher number of overall posts came from The Daily Mail, equating to 20.5% of the total (n = 85). The Times and The Sunday Times posts were far less statistically significant. In the case of The Times, a total of only 19 comments were included in this research, or 4.6% of the total.

Due to a lack of comparability between sources, it was determined that the best way to consider these results was to aggregate and take stock of public opinion holistically, rather than categorizing sources into silos. In each case, comments were only selected that offered some reference to the individual’s role identity as a police officer (in Couzens’ case) or a migrant (in Selamaj’s case). When aggregated, 10% of comments pertaining to Couzens were deemed positive and/or defensive (n = 21)—the vast majority of these positive and/or defensive comments came from The Metro (15 of 21, or 71.4%). Most of these comments referred in some capacity to media bias or prejudice, with commenters suggesting that The Metro (and other sources) risked condemning Couzens before his guilt was proven. A comment typical of this sentiment reads, “What ever [sic] happened to sub judice [sic] . . . My one question—IF he is charged in connection with the woman's disappearance, will he be able to get a fair trial?” (The Metro, 2021b).

Despite this (not inconsequential) group of commenters offering some level of support for Couzens, the majority of comments that expressed some comment on his role as a police officer were categorized as negative and/or condemnatory. A total of 18% (n = 38) of comments were in this category, with the most negative sentiment coming from comments on The Daily Mail, which accounted for 34.2% (n = 13). These comments tended to express a collective lack of surprise at police involvement in the Everard case, such as “this is exactly why my mom taught me to NEVER trust men just because they’re in a uniform” (The Daily Mail, 2021). The final category analyzed—shock and/or surprise—also accounted for a considerable portion of total comments on the Everard case at 9% (n = 19), roughly equivalent to the positive and/or defensive criterion. Most shock and/or surprise reactions came from CNN International (52.6%), perhaps indicating a less cynical interpretation of the police role emanating from what can be assumed to be a more diverse, international follower base. Overall, comments referring to Couzens’ role as a police officer made up 36.8% of the total. When adjusted to only consider comments referring to role identity (e.g. excluding posts not referring to Couzens’ police role), the sentiment percentage totals become clearer: positive and/or defensive (26.9%); negative and/or condemnatory (48.7%); and shock and/or surprise (24.4%).

Turning to the posts related to the Nessa murder, a total of 202 comments were recorded across the media publications’ Facebook pages. Of these, just 3% (n = 6) made some form of positive and/or defensive comment about Selamaj’s migrant status, compared with 10% in relation to Couzens’ police role. Positive and/or defensive comments were only identified on two sources, both centrist—four on The Metro and two on CNN International. Five of these six comments (83.3%) were centered on references to Selamaj’s race or migrant status by other commenters, with at least one commenter noting that “I see racism is alive and well still” (The Metro, 2021a). Another follower on The Metro post made explicit comment about the distinctions between public response to Couzens versus Selamaj, stating,

All the people focusing on the fact he is not English, did you have the same energy when the police officer killed Sarah Everard? Did you focus on his nationality? Would your level of disgust change if he had an English name? Shame on some of you!” (The Metro, 2021a)

When it came to negative and/or condemnatory comments about Selamaj, with reference to his migrant identity, this accounted for 23% (n = 47) of the total—5% higher than the total in the same criterion with regard to Couzens. Here, the results were clearly polarized based on political or ideological leanings: while centrists The Metro and CNN International, as well as The Times and The Sunday Times, all accounted for between 20% and 25% of total negative and/or condemnatory comments each, The Daily Mail (one of the more popular right-leaning publications) was responsible for almost 45% of all negative and/or condemnatory comments. Prominent left-leaning publication The Mirror was only the site for 6% in this category, potentially suggesting a more political polarization on the issue of migrant identity compared with that of the police role. Around 21.3% (n = 10) of negative and/condemnatory comments made reference to “deportation” to some degree, emphasizing a prevalent view that the British government bore ultimate responsibility for Nessa’s death as “another life that could have been saved if we hadn’t of imported her murderer” (The Metro, 2021a).

The gap between positive and/or defensive and negative and/or condemnatory comments on Selamaj is instructive—perhaps even more instructive, however, is the category that is absent from this section of the results: shock and/or surprise. Whereas around 9% of total comments (and 24.4% of “relevant” comments) on the Everard murder expressed shock and/or surprise at Couzens’ police status, not a single comment expressed the same sentiment on Selamaj’s migrant status. Despite being a null return, in some respects this is the most useful finding in this analysis. In all, 53 comments made reference to Selamaj’s migrant identity (26.2% of the 202 comment total), roughly 10% less than comments on Couzens’ police role. When filtered to only consider role-related (i.e. migrant-related) comments, positive and/or defensive comments made up 11.3%—15.6% less than the same category applied to Couzens. As there were no shock and/or surprise responses recorded, the remainder of comments (88.7%) were negative and/or condemnatory in relation to Selamaj’s migrant identity at a rate almost twice that of the negative and/or condemnatory comments leveled at Couzens in relation to his identity as a serving police officer.

Discussion

Accounting for shock in identity-oriented responses to VAW

Without question, the clearest result of the data analysis was the extreme disparity in the shock and/or surprise response to Couzens (as a police officer) murdering Everard, compared with the entirely absent shock and/or surprise response to Selamaj (a migrant) being charged with the murder of Nessa. There has been some suggestion that the disparity in public focus on each case has more to do with victims than perpetrators: Gopika Shaji Nair (2021) compared media coverage of Everard and Nessa’s murders, and asserted that Everard’s case may have attracted greater attention as a product of “missing white woman syndrome”—a term coined to describe the news media’s “bias toward white, conventionally attractive, middle-class women, as compared to women of colour, working class women or both.” While this may still be true, the results of this study do not support this finding: the number of comments related to Everard (a White woman) totaled 212, compared with 202 pertaining to Nessa (an Asian woman)—ultimately, a negligible variance. This does not necessarily mean that Nair and others who observed an intersectional racial and/or class dimension to public responses to the case are incorrect; however, it results in the substance of these comments taking on a more central focus, which is where disparities in reaction related to offender identity begin to emerge.

The data collected reflect a (relatively) elevated level of shock at the revelation that Couzens was a police officer: around 9% of total comments reacting to Couzens being charged expressed a shock and/or surprise sentiment, equating to almost a quarter (24.4%) of all comments that referred in some form to his role as a police officer. This finding is crucial to our understanding not just of responses to Everard’s murder, but also in terms of social responses to police crime in general. For some, that Couzens was a police officer only served to reinforce a pre-existing view of policing as an inherent patriarchal (even, misogynistic) profession that does not serve the needs of women effectively. However, for others, Couzens’ police status may have contributed to the phenomena Kate Manne (2017) describes as “himpathy”—a process which sees “the flow of sympathy [shift] away from female victims toward their male victimizers” (p. 23). Under the conditions of himpathy, men are treated as more sympathetic, and (privileged) male offenders are offered the “benefit of the doubt” over female victims; himpathy is augmented even further when intersectional factors are taken into account, like class or race. As a White, male British police officer, Couzens fits the clear archetype of an individual who may benefit from himpathy, with each element contributing to a characterization that could have mitigated the public’s response to his arrest and, later, conviction.

Kraska and Kappeler (1995) assert that “one would assume that police commit unjustifiable acts of violence only against men, and that women suffer no direct [emphasis in original] and systematic mistreatment at the hands of police officers” (p. 86). The weaker research base around police sexual violence (PSV) is attributed by some, such as Jasmine Sankofa (2016), to “gaps in data, underreporting, and fear of retaliation . . . [which] make it difficult to map its prevalence and the populations most vulnerable” (p. 655). These deficits in the research base are exacerbated by “the unique access officers have to victims combined with the unsupervised nature of patrol work . . . opportunity, power, authority, and isolation increases the likelihood of police sexual offending” (Rabe-Hemp and Braithwaite, 2012: 131). With so much of policework predicated on these elements, the picture of a covert deviance that is under-represented in official data begins to emerge. Since the Couzens case came to light, a far greater emphasis has been placed on PSV in the Metropolitan Police: data released under a Freedom of Information request in early 2021 revealed the force had upheld 119 cases of PSV against officers, out of a total of 594 complaints made between 2012 and 2019 (Townsend and Jayanetti, 2021); a subsequent investigation conducted by the Independent Office for Police Conduct determined that inappropriate sexual behaviors, among other forms of misconduct, were “not isolated or simply the behavior of a few ‘bad apples’ [in the Metropolitan Police]” (Dodd, 2022).

When it came to positive and/or defensive comments on the Couzens arrest, several responses adhered to the common “not all police” argument, calling on others to not “tar all police with the same brush” or arguing that the public “cannot be angry with the metropolitan police not all officers are bad [sic]” (The Metro, 2021a). Here, the comments in defense of the institution of policing align with a classic “rotten apple” construction wherein rogue actors within the profession commit offenses that lead to other, “good” police being tarnished unfairly. However, as Sankofa (2016) notes, “by characterizing officers who commit sexual offenses as ‘rogue’ actors, we fail to understand sexual assault as a structural violence” that the criminal justice system has, too often, failed to protect against (p. 657). Indeed, in response to the Couzens case, Metropolitan Police Commissioner Cressida Dick openly admitted that (on occasion) “a bad ‘un” slips through the net to become a police officer (Moloney, 2021).

When it came to Couzens, there was a level of surprise that “we can’t trust the police” and (in at least one instance) shock that he was a police officer when “some commenter said it was a migrant” (CNN International, 2021; The Mirror, 2021b). With this second comment, the real impact of role identity is played out in explicit term6s: whereas the commenter indicates that it would not have been a shock had the initial suggestion of migrant involvement been correct, it was a surprise that the perpetrator was a police officer. On its own, this could be treated as an isolated comment. However, when considered in line with the complete lack of shock and/or surprise sentiment in the later Selamaj case, the implication is that commenters are in some sense primed to believe that a migrant would commit violence like the murder of Everard or Nessa, at a rate far higher than they are primed to believe the same of a police officer like Couzens. As data released by non-government organization the Femicide Census reported in 2021, after Everard’s murder, the statistics on police violence contradict this widely held perception: the census revealed that 15 women were killed by serving or former police officers in the United Kingdom in the 12 years from 2009 to 2021 (Moloney, 2021).

Justice denied? Condemning the condemners

Aside from the shock and/or surprise differential, another notable finding that emerged in this research was the level of positive and/or defensive comments focused on Couzens’ revelation as Everard’s killer. When adjusted to filter out comments that did not refer to role identity, 26.9% adopted this sentiment in the Couzens-related posts, as opposed to 11.3% positive and/or defensive in response to Selamaj later in the same year. Deeper analysis reveals that this sentiment toward Couzens skewed primarily toward defensive postures and, in particular, the stance that due process was being denied to the (then) suspect as a result of the media publicly identifying him. Typical comments along these lines included “what happened to innocent before proven guilty” and “as much as this crime disgusts me and hopes when the person is found guilty rots in hell, no way should the person named till found guilty [sic]” (The Metro, 2021b).

In all, 61.9% of the positive and/or defensive comments referencing Couzens’ role as a police officer referred to the position that the media had injudiciously condemned him prior to his guilt being proven in court. This criticism is understandable, to some extent: the British laws on sub judice are generally aligned with best practice standards, and as such, there is an understanding among the public regarding the potentially prejudicial nature of reporting on a case that has yet to be concluded (Corker and Levi, 1996). Indeed, as Gauthier and Graziano (2018) determined in their research on public consumption of news media, “consumption of Internet news is related to negative attitudes about police” (p. 504). It is reasonable therefore to suggest that, perhaps even more than a typical offender, the identification of Couzens—a police officer—may have raised public concern around compromising the legal process had a jury trial occurred.

That said, the tenor of comments about the identification in this case did not adopt this perspective—rather than cautioning against prejudicing the judicial process, commenters (without exception) were more concerned with the lack of fairness toward Couzens himself, which is what caused these comments to fall under the positive and/or defensive categorization. The question of “fairness” was central to the defensive response in the comments about Couzens, with respondents seeming to neutralize the allegations against Couzens by, as Matza and Sykes (1957) called it, “condemning the condemners” who, in this instance, were identified as the media. Neutralization was common among the positive and/or defensive comments, even aside from references to the media. In several comments, the suggestion of other “explanations” for the allegations against Couzens were raised: one noted, “I am concerned about his bruised appearance . . . Did he have a confession beaten out of him?” whereas another questioned whether Couzens was “an embedded spy” with his arrest part of an elaborate undercover sting to catch the real murderer (The Mirror, 2021b; The Times and The Sunday Times, 2021).

The scale of neutralization observed in the Couzens comments was completely absent in comments referring to Selamaj: whereas, in that case, there were a limited number of defensive comments arguing against racial characterizations of Nessa’s alleged killer, no comments attempted to suggest innocence or mitigate Selamaj’s suspected involvement, as was the case in the Couzens comments (The Metro, 2021a). The propensity to criticize the perceived unfairness caused by the identification of Couzens was absent in the Selamaj comments, where defensive sentiment was targeted externally, supporting the migrant community rather than the alleged offender himself. Once again, the results reflect a willingness to accept Selamaj’s guilt at face value, at a rate far greater than observed in the Couzens case. Whether this was related to role identity is not entirely clear; however, when taken into consideration with the very different rates of shock and/or surprise related to the police versus migrant roles, it is one possible (even, likely) explanation for the results in this criterion.

Acceptance of guilt: A question of migration and belonging in Britain

The surprised reaction to Couzens being named a suspect in Everard’s murder can be contrasted with a seemingly tacit acceptance of Selamaj’s alleged participation in Nessa’s death; however, it is not only in this contrast that the contours of the response to Selamaj’s identity can be observed. Of the total comments about the Nessa case, 23% expressed negative sentiment directly related to Selamaj’s migrant status—adjusted to only include content relating to migrant identity, this rises to an overwhelming majority of 88.7%. Many of these comments (21.3% classified as negative and/or condemnatory) called for deportation, not just of Selamaj but “mass deportations” prompted by the perception that migrants are responsible for high levels of crime in the United Kingdom (The Mirror, 2021a). These results are in alignment with a survey conducted by Kings College London in 2019, which found that perceptions of the connection between immigration and crime were disproportionately high: while respondents believed (on average) that 34% of the British prison population was foreign-born, the real statistic was 12%, slightly under-representative of the 13% of the general population who were immigrants (Duffy, 2021).

The acceptance of migration “myths” was evident in other comments responding to Selamaj arrest, many of which referenced “another one I guess off a boat” and admonishing the government for “keep[ing] letting the dinghys in” carrying migrants across the English Channel (The Metro, 2021a; The Mirror, 2021a). Initial reports about Selamaj’s identity did not include further details beyond his nationality, and thus, any reference to arriving in the United Kingdom illegally or, indeed, via boat was entirely based on supposition and speculation fueled by an erroneous view on the nature of migration to the United Kingdom. There is little doubt this perspective is informed by high rate of boat arrivals in 2021 where 8000 prospective migrants had arrived using this method by mid-year (Ott, 2021); despite this record number of “illegal” arrivals, the truth is that most migration to the United Kingdom continues to occur via legal means—even then, 98% of boat arrivals to the United Kingdom go on to apply for legal asylum, with Refugee Council data showing that 52% are accepted on initial evaluation, with 46% of denied claims being accepted on appeal (Refugee Council, 2021). In the 12 months prior to March 2020 (the last pre-COVID figures), around 715,000 individuals migrated to the country through conventional pathways (Clark, 2021)—when the number of boat arrivals is compared with this legal migration statistic, the expectation (and acceptance) that Selamaj must have been an illegal boat arrival is called into question as a result more of a media panic, not a fear with legitimate basis.

It was not just Selamaj’s migrant identity that provoked comment, but his specific ethnic heritage as an Albanian-born migrant: 25.5% of negative and/or condemnatory comments referenced Albania or Eastern Europe, indicating a concern around migration organized around a specific narrative which perceives some migrants as more acceptable than others. This is, again, reflective of a pervasive misconception among the British population, that Albanian migrants are responsible for a “crime wave” in the country. This myth is, again, supported in part by statistics that seem (on the surface) to suggest serious concern is warranted. In 2020, Albanian nationals were revealed to be the largest foreign national group in the British prison system, making up 10% of the foreign-born prison population (Ford, 2020). Alongside this, Home Secretary Priti Patel began cooperation with the Albanian government in 2021 for the United Kingdom to aid in building a prison facility in Albania, with the express purpose of establishing a location which the country could deport “Albanian criminals” to in future (Morris, 2021). This unquestionably fed public concern that Albanians posed a tangible threat to public safety—enough to warrant the UK government constructing an entire foreign facility to house them in response.

Despite this, the National Crime Agency report that Albanians were just 0.8% of the nation’s organized crime operators, behind Romanians (1.5%), Pakistanis (1.2%), and Polish (0.9%)—more pertinently, all were significantly behind the rate of British-born organized criminals, which accounted for 61.6% of the total at the time the agency’s report was published (Weaver, 2017). It is reasonable to suggest that, regardless of crime data suggesting locally born offenders pose more of a risk in the United Kingdom, the public rhetoric around Albanian offenders in particular informed the response to Selamaj’s identification in the Nessa case, to some extent. When compared to the Couzens case, it could be argued that the over-representation of migrants (especially Albanians) in media reporting on crime, when compared with similar reporting on police crime, is a significant contributing factor to the very distinct social media responses to each of the offenders featured in this research.

Conclusion

The stranger-perpetrated murders of Sarah Everard and Sabina Nessa in 2021 prompted a renewed public discourse on the safety of women, and the prevalence of VAW, in the United Kingdom. As the nation grappled (and continues to grapple) with understanding the reasons for such tragic crimes, it is natural for public attention to turn to the offender and our response not just to what happened, but who was responsible. It is not the intention of this article to evoke sympathy for the perpetrators of these crimes, but rather to consider our reaction as a community when suspects are revealed who either meet or confound our societal expectations. What emerged from this research was an obvious distinction in the public reaction to a police officer (Couzens) being revealed as a murderer when compared to the identification of an Albanian migrant (Selamaj) under analogous conditions. Whereas the Couzens identification prompted a considerable proportion of responses which could be categorized as positive and/or defensive, the same “benefit of the doubt” was not afforded to Selamaj, who was accepted to (likely) be responsible for the crimes of which he was accused. So many of the Facebook comments in defense of Couzens were focused less on him as a person, and more on the sentiment that the media identifying him ahead of trial was in conflict with his judicial rights—while, at first, this seems a reasonably fair response, the complete absence of a similar rights-based defense in Selamaj’s case suggests that, in the mind of the public, justice is not something that is offered on equitable terms to all.

The defense of Couzens’ rights is potentially linked to the high rate of shock and/or surprise registered at his identification as a police officer accused of murder: the pervasive disbelief at his involvement may, in turn, underpin the position that justice must be served fairly in order to ensure that Couzens really was responsible for Everard’s murder. Where there is no shock and/or surprise, as in Selamaj’s case, there is seemingly a lesser need to adopt a presumption of innocence ahead of trial or, indeed, conviction. While the research highlights a greater defensive reaction, and greater levels of shock, at Couzens’ role in Everard’s death, the tenor of negative and/or condemnatory comments about Selamaj suggests that anti-migrant sentiment is more a factor in the reaction differential than Couzens’ identity as a police officer was. As noted, there was not a single comment registering shock and/or surprise at an Albanian migrant being accused of murder, in contrast with almost 10% of the total in Couzens’ case. This gap was filled instead by negative and/or condemnatory comments which revealed, on the contrary, an expectation that Nessa’s murderer would be a migrant and calling for penalties such as mass deportation in response.

Data show that the foreign-born population in the United Kingdom is under-represented in official crime rates; however, it must be remembered that Nessa’s murder occurs in a post-Brexit nation led by a Conservative government which has continued to place a firm focus on the “criminal threat” posed by migrants, perhaps accounting for the prevalent acceptance of this otherwise unsubstantiated perspective. The murders of Everard and Nessa are a societal tragedy, and the way we respond to their killers speaks to how we perceive such events in future. Although this is a relatively small-scale study focused on two cases (albeit with significant cultural impact), it stands as a contemporary reflection on initial societal responses to VAW offenders and contributes to current literature on how role identity is tied to the public’s reaction to (and engagement with) these events. While we remain more shocked and/or surprised that one “type” of person may commit VAW, and tacitly accepting of another “type” of person’s guilt, we can never fully understand the true nature of these offenses and, in turn, develop an effective community response.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Paul Bleakley

Notes

Author biography

Paul Bleakley is an Assistant Professor at the University of New Haven. His research focuses on police misconduct and sociocultural responses to ethical issues in law enforcement in the public sphere.

References

Aldridge

Evetts

(2003) Rethinking the concept of professionalism: The case of journalism. The British Journal of Sociology 54(4): 547–564.

Clark

(2021) Migration figures in the United Kingdom from 1991 to 2020. Statista, 3 June. Available at: https://www.statista.com/statistics/283599/immigration-to-the-united-kingdom-y-on-y/ (accessed 12 October 2021).

CNN International (2021) London police officer charged with murder of Sarah Everard. CNN International, 13 March. Available at: https://www.facebook.com/cnninternational/posts/10159173371704641 (accessed 9 October 2021).

Corker

Levi

(1996) Pre-trial publicity and its treatment in the English courts. Criminal Law Review September: 622–632.

Dodd

(2021) Wayne Couzens pleads guilty to murdering Sarah Everard. The Guardian, 9 July. Available at: https://www.theguardian.com/uk-news/2021/jul/09/wayne-couzens-pleads-guilty-murdering-sarah-everard (accessed 6 October 2021).

Dodd

(2022) Met officers joked about raping women, police watchdog reveals. The Guardian, 1 February. Available at: https://www.theguardian.com/uk-news/2022/feb/01/met-officers-joked-raping-women-police-watchdog-racist (accessed 8 April 2022).

Duffy

(2021) Britons are overwhelmingly swayed by myths about immigration—These are the five facts we often get wrong. The Independent, 6 January. Available at: https://www.independent.co.uk/voices/immigration-britain-channel-crossing-migrants-asylum-seekers-refugees-aid-a8713986.html (accessed 17 October 2021).

Eberl

Meltzer

Heidenreich

, et al. (2018) The European media discourse on immigration and its effects: A literature review. Annals of the International Communication Association 42(3): 207–223.

Fanghanel

(2017) How to break a rape culture: Gendered fear of crime and the myth of the stranger-rapist. In: Lee

Mythen

(eds) The Routledge International Handbook on Fear of Crime. Abingdon: Routledge, pp. 1–12.

10.

Ford

(2020) Albanians become largest foreign group in prisons. The Times, 1 May. Available at: https://www.thetimes.co.uk/article/albanians-become-largest-foreign-group-in-prisons-tw7d5rtt6 (accessed 1 October 2021).

11.

Gauthier

Graziano

(2018) News media consumption and attitudes about police: In search of theoretical orientation and advancement. Journal of Crime and Justice 51(5): 504–520.

12.

Gonçalves

Benevenuto

Araújo

, et al. (2013) Comparing and combining sentiment analysis methods. In: COSN ’13: Proceedings of the first ACM conference on online social networks, Boston, MA, 7–8 October, pp. 27–38. New York: ACM.

13.

Graham-Harrison

(2021) Police clash with mourners at Sarah Everard vigil in London. The Guardian, 13 March. Available at: https://www.theguardian.com/uk-news/2021/mar/13/as-the-sun-set-they-came-in-solidarity-and-to-pay-tribute-to-sarah-everard (accessed 2 October 2021).

14.

Jackson

Sunshine

(2007) Public confidence in policing: A neo-Durkheimian perspective. British Journal of Criminology 47: 214–233.

15.

Kautt

(2011) Public confidence in the British police: Negotiating the signals from Anglo-American research. International Criminal Justice Review 21(4): 353–382.

16.

Kelly

(1988) Surviving Sexual Violence. Cambridge: Polity Press.

17.

Kraska

Kappeler

(1995) To serve and pursue: Exploring police sexual violence against women. Justice Quarterly 12(1): 85–112.

18.

Kwai

(2021) Murder charge in case that renewed outrage over U.K. Women’s safety. The New York Times, 28 September. Available at: https://www.nytimes.com/2021/09/28/world/europe/uk-sabina-nessa-murder-charge.html (accessed 1 October 2021).

19.

Manne

(2017) Down Girl: The Logic of Misogyny. New York: Oxford University Press.

20.

Matza

Sykes

(1957) Techniques of neutralization: A theory of delinquency. American Sociological Review 22(6): 664–670.

21.

Maynard

Greenwood

(2014) Who cares about sarcastic tweets? Investigating the impact of sarcasm on sentiment analysis. In: Language resources and evaluation conference, Reykjavik, 26–31 May, pp. 1–6.

22.

Moloney

(2021) At least 15 serving or former police have killed women in UK since 2009—Report. The Guardian, 28 September. Available at: https://www.theguardian.com/uk-news/2021/sep/28/at-least-15-serving-or-former-police-have-killed-women-in-uk-since-2009-report (accessed 2 October 2021).

23.

Morris

(2021) Priti Patel signs deal to remove Albanian criminals and unsuccessful asylum seekers from UK more quickly. Sky News, 9 July. Available at: https://news.sky.com/story/priti-patel-signs-deal-to-remove-albanian-criminals-and-unsuccessful-asylum-seekers-from-uk-more-quickly-12352181 (accessed 23 September 2021).

24.

Morrison

(2019) Re-framing free movement in the countdown to Brexit? Shifting UK press portrayals of EU migrants in the wake of the referendum. The British Journal of Politics and International Relations 21(3): 594–611.

25.

Moule

Jr (2020) Under siege? Assessing public perceptions of the “War on police.” Journal of Criminal Justice 66: 1–12.

26.

Nair

(2021) Reporting Femicides: Is the UK Media Racially Biased? Media Diversity Institute. Available at: https://www.media-diversity.org/reporting-femicides-is-the-uk-media-racially-biased/ (accessed 11 April 2022)

27.

Office for National Statistics (ONS) (2021a) Estimates of proportions for how safe people feel walking alone after dark, by respondent sex, Crime Survey for England and Wales, year ending March 2020. Available at: https://www.ons.gov.uk/peoplepopulationandcommunity/crimeandjustice/adhocs/13019estimatesofproportionsforhowsafepeoplefeelwalkingaloneafterdarkbyrespondentsexcrimesurveyforenglandandwalesyearendingmarch2020 (accessed 11 April 2022).

28.

Office for National Statistics (ONS) (2021b) Nature of sexual assault by rape or penetration, England and Wales. Available at: https://www.ons.gov.uk/peoplepopulationandcommunity/crimeandjustice/datasets/natureofsexualassaultbyrapeorpenetrationenglandandwales (accessed 11 April 2022).

29.

Ott

(2021) Record number of migrants and asylum seekers reach U.K. By boat in a single day. CBS News, 20 July. Available at: https://www.cbsnews.com/news/uk-migrants-record-number-asylum-seekers-reach-england-by-boat/ (accessed 13 October 2021).

30.

Rabe-Hemp

Braithwaite

(2012) An exploration of recidivism and the officer shuffle in police sexual violence. Police Quarterly 16(2): 127–147.

31.

Raymen

Lloyd

Kuldova

, et al. (2021) The longest year: The future of crime, harm, and justice in the shadow of 2020. Journal of Contemporary Crime, Harm, Ethics 1(1): i–vi.

32.

Refugee Council (2021) An Analysis of Channel Crossings & Asylum Outcomes: November 2021. Stratford: Refugee Council.

33.

Sankofa

(2016) Mapping the blank: Centering Black women’s vulnerability to police sexual violence to upend mainstream police reform. Howard Law Journal 59(3): 651–704.

34.

Stansfield

Stone

(2018) Threat perceptions of migrants in Britain and support for policy. Sociological Perspectives 61(4): 592–609.

35.

Surette

(1994) Predator criminals as media icons. In: Barak

(ed.) Media, Process, and the Social Construction of Crime: Studies in Newsmaking Criminology. New York: Garland Press, pp. 131–158.

36.

Thakor

Sasi

(2015) Ontology-based sentiment analysis process for social media content. Procedia Computer Science 53: 199–207.

37.

The Daily Mail (2021) Police officer is arrested over disappearance of Sarah Everard in London. The Daily Mail, 9 March. Available at: https://www.facebook.com/DailyMail/posts/6778500988876258 (accessed 9 October 2021).

38.

The Metro (2021a) Man charged with murdering Sabina Nessa after body dumped in London park. The Metro, 27 September. Available at: https://www.facebook.com/MetroUK/posts/4865206130181303 (accessed 30 September 2021).

39.

The Metro (2021b) Pictured: Met police officer arrested over Sarah Everard disappearance. The Metro, 10 March. Available at: https://www.facebook.com/MetroUK/posts/4221530277882228 (accessed 13 October 2021).

40.

The Mirror (2021a) Man, 36, charged with murder of primary school teacher Sabina Nessa. The Mirror, 27 September. Available at: https://www.facebook.com/dailymirror/posts/10160368491644162 (accessed 2 October 2021).

41.

The Mirror (2021b) Policeman arrested on suspicion of murder of missing Sarah Everard in pictured. The Mirror, 10 March. Available at: https://www.facebook.com/dailymirror/posts/10159866716054162 (accessed 11 October 2021).

42.

The Times and The Sunday Times (2021) Wayne Couzens: Met officer in Sarah Everard murder trial in October. The Times and The Sunday Times, 16 March. Available at: https://www.facebook.com/timesandsundaytimes/posts/4397409583621623 (accessed 12 October 2021).

43.

Thorburn

(2021) Detectives probing murder of Sabina Nessa search woods in Kent: Met Police scour land near Tunbridge Wells as court hears she was “beaten with 2ft long weapon” in “premeditated attack caught on CCTV” as Albanian suspect, 36, is remanded in custody. The Daily Mail, 30 September. Available at: https://www.dailymail.co.uk/news/article-10045111/Albanian-national-accused-murdering-teacher-Sabina-Nessa-faces-court-time.html (accessed 1 October 2021).

44.

Townsend

Jayanetti

(2021) Revealed: The grim list of sex abuse claims against Metropolitan Police. The Guardian, 20 March. Available at: https://www.theguardian.com/uk-news/2022/feb/01/met-officers-joked-raping-women-police-watchdog-racist (accessed 8 April 2022).

45.

Walby

Towers

Francis

(2016) Is the rate of violent crime increasing or decreasing? A new methodology to measure repeat attacks making visible the significance of gender and domestic relations. British Journal of Criminology 56(6): 1203–1234.

46.

Weaver

(2017) Albanian gangs have significant control over UK cocaine market, says crime agency. The Guardian, 29 June. Available at: https://www.theguardian.com/uk-news/2017/jun/29/albanian-gangs-control-uk-cocaine-market-says-agency (accessed 2 October 2021).