What is the problem? Representations of gender and violence towards politicians in UK parliamentary debates

Recent research demonstrates that there are important gendered dimensions in politicians’ experiences of violence. Less scholarship has examined the phenomenon’s political impacts, including how violence is represented as a policy problem by political actors themselves. This article applies feminist frame analysis to theories of gender and political violence to explore how UK members of Parliament frame their experiences of violence (including abuse, harassment and intimidation) in the eight House of Commons debates specifically on the topic between 2010 and 2021. Violence towards politicians, including an increasing focus on the experiences of women and social media abuse, has been politicised over time as a normative policy problem, being framed as an attack on the public good of democracy. This article develops theories of gender and political violence by demonstrating that the problem of gendered political violence can and should be conceptualised not only as a unique phenomenon but also as a policy problem.


Introduction
Globally, there has been burgeoning academic and political attention to gendered dimensions of violence, including harassment and abuse, towards politicians and other political actors.Contributions by feminist scholarship include evidencing differences in men's and women's experiences (Bardall, 2011;Herrick and Thomas, 2019;Bardall et al, 2020;Håkansson, 2020), and establishing the existence of a specific phenomenon of violence against women in politics ( Krook and Restrepo Sanín, 2016a;Krook, 2020).
Less scholarship has explored the political impacts of such violence.This article analyses how UK members of Parliament (MPs) represent the violence they experience in the agenda-setting space of House of Commons parliamentary debates.The UK has been a global academic and political focal point for the phenomenon of violence in 'stable' contexts following the murders of MPs Jo Cox in 2016 and David Amess in 2021, as well as other high-profile reports of abuse.A growing body of research indicates not only that both men and women MPs experience high levels of abuse but also that women, especially those of racially marginalised identities, face particularly aggressive forms (Dhrodia, 2018;Collignon and Rüdig, 2020;Palmer, 2020;Ward and McLoughlin, 2020;Harmer and Southern, 2021;Kuperberg, 2021;Collignon et al, 2022).The government has commissioned research into political abuse (Committee on Standards in Public Life, 2017) and taken some policy action, for example, establishing the Parliamentary Liaison and Investigation Team (PLaIT) following the murder of Jo Cox and the Independent Complaints and Grievance Scheme (ICGS) during the #MeToo movement (Krook, 2020;Raney and Collier, 2021).
This article uniquely applies feminist frame analysis to the study of gender and political violence to examine underlying normative principles, assumptions and effects, from 'diagnosis' to 'prognosis' of problem representation (Bacchi, 2012;van der Haar and Verloo, 2016).Following feminist researchers (for example, Kelly, 1987;2011;Krook, 2020;Restrepo Sanín, 2022a), this article understands violence 'expansively' to include a broad range of forms, including murder, harassment, intimidation and online abuse.The article asks: 'What is the problem framing of violence towards MPs in parliamentary debates between 2010 and 2021?'; and 'How is gender considered in the problem framing?' The analysis finds that violence (comprehensively understood to include abuse, harassment and intimidation) towards MPs has been increasingly framed as 'normal' or 'expected', with an increasing focus on women's experiences.Different forms of violence, especially online, are represented as harmful.The phenomenon is framed as an attack on the normative public good of democracy.The main policy problem represented is the lack of social media regulation, to be solved by the government and social media platforms.
The research contributes to theories of gender and political violence by indicating that there are gender differences in how men and women MPs experience violence, including evidence for a specific problem of violence against women in politics.Most importantly, this article develops the existing literature by demonstrating that the problem of violence towards politicians, including its gendered dimensions, has been politicised over time as a normative policy problem in the UK.The problem of gendered political violence can and should be conceptualised not only as a unique phenomenon but also as a policy problem.
The next section presents the two pillars of the conceptual framework: feminist approaches to political violence and feminist frame analysis.Then, the methodology is outlined, followed by the presentation of the empirical findings.The discussion and conclusion explore empirical and theoretical implications.

Feminist theorising of political violence
Feminist scholars contribute to political violence scholarship by 'making gender visible' (Bjarnegård, 2018).Crucially, this scholarship expands the understanding of political violence beyond physical force in conflict settings to a definition that includes sexual, psychological and online forms in 'non-conflict' or 'stable' contexts (Bardall, 2011;Inter-Parliamentary Union, 2016;Krook and Restrepo Sanín, 2016b;Biroli, 2018;Krook, 2020).The theory of 'violence against women in politics' argues that there exists a specific phenomenon in which women experience violence because they are women in the political space (Krook and Restrepo Sanín, 2016a;Biroli, 2018;Krook, 2020).Others compare the experiences of men and women in specific contexts (Piscopo, 2016;Bjarnegård, 2018;Bardall et al, 2020;Holm, 2020).
The focus of the scholarship has been on the phenomenon itself, with some growing research on political impacts (Bardall, 2020;Krook, 2020;Restrepo Sanín, 2022a;vawpolitics, 2022).A growing number of contexts have developed policies to address gendered political violence.Places including Bolivia, Catalonia and Peru have specific legal frameworks that criminalise violence against women in politics (Restrepo Sanín, 2018;2021;2022a;2022b;Krook, 2020;Albaine, 2021;Verge, 2021).In tandem with, and arguably influenced by, these political developments and efforts by feminist activism, global normative frameworks have adopted an understanding of the gendered nature of political violence (United Nations, 2018;Castaño, 2021;Restrepo Sanín, 2022a).Relatedly, the global #MeToo movement has led to sexual harassment policy reform (Berthet and Kantola, 2021).In the UK, some research has examined the mixed impact of the reform of parliamentary and political parties' harassment policies on supporting staff and changing organisational cultures (Collier and Raney, 2018;Miller, 2021;Raney and Collier, 2021).

Framing parliamentary debates
This article develops the gender and political violence literature to explore how the specific problem of violence towards politicians is framed in political discourse.Parliamentary debates are important agenda-setting spaces where representations about policy problems are (re)produced in a dynamic interplay of the agendas of individual MPs, parties, the government and the parliamentary institution (Proksch and Slapin, 2015).MPs use frames as representations of ideas in order to politicise an issue as worthy of policy attention (Ilie, 2010).The UK House of Commons is particularly interesting because of the opportunities for both the government and opposition parties to present their ideas in response to government speakers or as debate proposers themselves.That there are any debates specifically on a topic, and that the number of debates has increased over time, indicates politicisation.
This article focuses on the specific problem of violence towards politicians.In line with other feminist research (for example, Krook, 2020;Restrepo Sanín, 2022b), this article adopts a 'comprehensive' understanding of violence that includes sexual, psychological and online forms, and, as such, 'abuse', 'intimidation' and 'harassment'.While other political actors are outside the scope of this research, there is growing scholarship aimed at understanding the experiences, driven by gendered power dynamics, of a range of actors, including voters, volunteers and staff (for an overview, see Krook, 2020).Indeed, it was revelations of MPs abusing their power over staff and volunteers that led to the reform of harassment policies, including the ICGS (Julios, 2022).This policy development was discussed in other parliamentary debates, and a specific analysis of these would be a rich avenue for future research.
Focusing on politicians carries relevance in terms of positionality and representation.In parliamentary debates, MPs usually act as advocates, conveying problems that impact groups in the population (Phillips, 1994;Heywood, 2004: 233;Proksch and Slapin, 2015), and there is rich feminist scholarship on how gendered differences impact communication styles and problem selection (Lovenduski, 2014;Hargrave and Blumenau, 2022;Siow, 2023).Yet, in debates on violence towards politicians, all MPs take on an unusual positionality of themselves being the impacted population.This dual role of advocates and impacted population may influence the way in which they frame the phenomenon.Exploring how MPs strategically frame the violence towards them in political discourse over time develops our understanding of the phenomenon theoretically and empirically.
Following feminist researchers, this article understands gender broadly.In the debates, considering gender could mean an explicit or implicit reference to how the phenomenon is experienced differently by men and women, or a recognition of the impact of structural sexism. 1 Understanding gender as intersectional (Crenshaw, 1989;Choo and Ferree, 2010;Collins, 2015), the analysis considers if and how gender is presented as intersecting with other identities or oppressions, such as race or racism.This could involve attention to the experiences and impacts of those of a specific identity group, and/or recognition of the causal impact of structural racism/sexism.
Adopting a feminist frame analysis means specifically and critically examining how normative principles are reflected in the representation of policy problems.Bacchi's (2012) 'What is the problem represented to be?' approach encourages asking questions around the assumptions, silences and effects of the problem representation, which relates to the attention of critical frame analysis to underlying normative principles, from problem 'diagnosis' to 'prognosis' (Verloo, 2007;Dombos, 2012;van der Haar and Verloo, 2016).

Methodology
Informed by this framework, the article asks: 'What is the problem framing of violence (including abuse and harassment) towards MPs in parliamentary debates during 2010-21?'; and 'How is gender considered in the problem framing?'The data (listed in Appendix 1) are the eight House of Commons debates on the phenomenon of violence towards politicians between 2010 and 2021.The time frame captures significant events, including the two murders of MPs since the Northern Ireland conflict and general elections.Debates were extracted from the 2021 parliamentary record, Hansard UK Parliament (2021), using keyword searches (including 'violence + members of Parliament' and 'abuse + public life').Initially, a broader database of all 24 parliamentary debates that mentioned violence towards politicians was analysed, but a narrow database with debates exclusively on the topic allows exploration of the phenomenon's specific framing.This analysis considers the overall thematic problem representation of the phenomenon, with a specific focus on gender.Employing frame analysis (Verloo, 2007;Dombos, 2012), a combination approach of open/inductive and closed/ deductive coding was used to capture such elements as: 'voice'; 'diagnosis', including definition, causality and gender; and 'prognosis', including solutions.Codes were derived from the theoretical literature on gender and political violence (for example, 'elections') and aspects that emerged from the data (for example, 'print media').A preliminary content analysis was conducted on the larger universe of debates to develop the codebook (see Appendix 2), which was then applied to the smaller data set.Each intervention (speech by an individual speaker) was coded, and common frames among speakers were identified.

Findings
The findings are organised into three frames: normalisation, social media and gender.The main frame that emerged over time is that violence has become 'expected' in MPs' roles, which impacts democracy.A consistent frame over time is that violence towards politicians is largely an online problem and solutions include the regulation of social media platforms.The main gendered dimension is that women MPs are particularly impacted by abuse, which threatens their political participation and, as such, democracy.

A problem of normalisation: 'part of our job description' and 'attack on democracy'
In the first debate on the phenomenon, 'Internet abuse against members of Parliament' in 2014, the four speakers emphasised individual victims.Although this debate focused on MPs, the framing presents the problem of Internet abuse towards them as important because their experiences suggest that such abuse is experienced by others.For example, John Mann opened the debate by saying, 'I wish to raise the problems that Members of the House and many more people in our communities face from the abuse of social media', and Luciana Berger, whose abuse was one of the reasons Mann proposed the debate, said: 'we must have in mind those people to whom he alludes and who are not in this House and have to suffer in silence….We need to do everything we possibly can to tackle this issue for them.' Speakers performed the positionality as advocates for others rather than the impacted population.
From 2017, the phenomenon of violence towards MPs became explicitly framed as a problem itself.The July debate on 'UK elections: abuse and intimidation' coincided with then Prime Minister Theresa May's announcement of an independent review into 'intimidation in public life' (UK Government, 2017), with further debates in September, ('General Election Campaign: Abuse and Intimidation') and December ('Harassment in Public Life').Many speakers framed the 2017 election as distinctive from previous elections in terms of abuse.For example, Alex Chalk (September) declared: 'the tempo and tone of what happened in the 2017 election was of a different order of magnitude from what had gone before'.A few speakers, even if they agreed that the recent election was unique, suggested that such abuse was not entirely new.For example, Cat Smith said, 'abuse and intimidation have taken place during previous elections too' (July), and 'Women MPs have been speaking out about these problems with social media for years' (September).
By 2019, abuse was represented to have become expected.Nearly all the 15 speakers in the May debate explicitly framed abuse as worryingly normalised.For example, Helen Jones said: 'No one should be abused and intimidated simply for doing their job, yet it is now unusual to come across an MP who has not received a death threat….Apparently, tolerating intimidation and abuse has become part of our job description.'The driving representation is that violence towards politicians is an attack on democracy.Over half the speakers explicitly stated intimidation's impact on democracy.For example, Harriet Harman declared: 'This is not just criminality against individuals; even more importantly than that, it is a fundamental attack on our democracy.'This representation elevates the individual and group experiences of MPs to the systemic level of the normatively important political system.
Notably, while the focus remained on MPs, the later debate titles do not name MPs but use general terms, such as 'harassment in public life' and 'elected women representatives'.This may indicate the strategic negotiation of MPs' positionality as both advocates and impacted population by linking their own experiences with those of others in public life and, ultimately, democracy.In a typical example of such broadening framing, Chris Skidmore (July 2017) declared: 'We owe it to our democracy to make clear that intimidation and abuse have no part in our society, not only for candidates who stood at the recent general election but for future generations of men and women who are considering entering public life.' The debates paying tribute to the two recently murdered MPs provide clear evidence that the 'normalisation' frame emerged over time.In 'Tributes to Jo Cox' (June 2016), the 18 speakers focused on her as an individual.Only two referenced political violence.Leader of the Opposition Jeremy Corbyn lamented 'the hatred that killed her in what increasingly appears to have been an act of extreme political violence', and Nigel Dodds MP of the Democratic Unionist Party (DUP) said: 'We in Northern Ireland … have indeed lost friends and dear colleagues to men of violence, have felt the pain and anguish that those close to Jo are going through now'.These references indicate the frame that violence towards MPs is unique, at least in recent years in a 'peace' context.
In contrast, in 'Tributes to Sir David Amess' (October 2021), while much of the focus was on the murdered MP as an individual, there were many references to other forms of abuse.Out of the 31 speakers, including the leaders of the major parties, 11 referenced those in public life who were murdered or suffered near-fatal attacks. 2 Notably, these attacks were referenced alongside other forms of violence, with several sharing personal experiences and advocating for improved security for MPs.Leader of the Opposition Keir Starmer (Labour Party) even implored a focus on tributes: I know that politicians across the country and across this House have their own experiences of threats to their security.Today is a chance to remember David, but in the weeks and days to come we must confront the threats and violence that everyone faces in enacting this country's democracy.
This reference to wider violence, which was not present in 'Tributes to Jo Cox', indicates that the political discourse had moved to recognise a phenomenon of violence towards MPs.Again, the frame of violence impacting democracy was reiterated by many speakers throughout the debate.Importantly, two speakers specifically referenced the normalisation of online abuse.For example, Mark Francois said: 'we are now systematically vilified day after day … if we want to ensure that our colleague did not die in vain … take the forthcoming Online Safety Bill and toughen it up'.This indicates the framing recognising different forms of violence, from online abuse to murder.
A social media problem: 'the rise in the use of online media has turbocharged abuse' While this link between online abuse and other forms of violence emerged over time, there has been a consistent frame of a problem with social media.The 2014 debate focused on Internet abuse.Yet, in this debate, there was not a clear link between online abuse and other forms of violence, nor an explicit frame involving democracy.From 2017, online abuse was presented as harmful itself.There emerged a causal framing of social media as a driver of the increase in abuse.For example, Diane Abbott (July 2017) expressed: 'the rise in the use of online media has turbocharged abuse'.A further part of the logic is that the nature of social media platforms makes them ample spaces for abuse.Specific aspects of social media, especially the use of anonymous accounts and ease of posting, are cited as particularly causal for abuse.For example, Tommy Shepherd (July 2017) said: 'All they need to do is switch on their phone and they can instantly and anonymously direct the most vile abuse to whomever they want'.
Notably, some speakers emphasised the role of print media in encouraging and perpetrating abuse.This framing was particularly prevalent in the December 2017 debate in response to specific headlines related to the Brexit debate (see, for example, The Telegraph, 2017).Out of the 35 speakers, eight explicitly raised the print media's role.For example, Yvette Cooper urged: 'I must press the Home Secretary again on an issue that has been raised by Members from all parts of the House -namely, the need to challenge national newspapers if they do things that incite death threats or have an impact on the quality of our debate.'The causal role the media's language was made clear here and by other speakers.For example, Diane Abbott (July 2017) argued: 'we always see, at the very least, a spike in abuse after there has been a lot of negative stuff in the media'.While the print media may not be the main perpetrator, it has been presented as at least part of the cause for abuse.
Social media platforms are consistently presented as important actors in creating and solving the problem.Specific solutions presented, sometimes explicitly but often implicitly, include improved reporting mechanisms, removing abusive content and banning abusive accounts.The 2014 debate focused on the action of social media platforms.For example, debate proposer John Mann referenced the 'Best practices for responding to cyberhate', developed by the Inter-parliamentary Coalition for Combating Antisemitism and the Anti-Defamation League (2014), and criticised social media providers for such guidance 'being violated repeatedly' by them.A framing strategy in later debates involved giving examples of abusive content that was not acted upon by platforms, for example: '"put a bullet between his ears" -that is a comment that I reported to Facebook, which said it did not breach its terms and conditions' (Cat Smith, Jul 2017).
Another important actor represented as responsible for addressing abuse is the government.Notably, representations are differentiated between government and nongovernment speakers.While the government seemed to accept some responsibility, it framed its action as sufficient.A common refrain by nearly every government speaker was: 'what is illegal offline is illegal online'.This framing of policy sufficiency was implicitly and explicitly challenged by others, including, notably, MPs of the government's party.For example, Maria Miller (April 2021) echoed the government's language to point out: 'The Minister will no doubt say that what is illegal offline is illegal online….There is no definitive figure, however, on the number of reports, the number of cases taken forward, and the number of convictions.'There were further contestations about whether new policy is needed and/or if existing policy frameworks should be extended.For example, in 2014, John Mann expressed: 'We do not need new law.We need the current law to be used imaginatively to remove profiles from the internet.'Yet, in later debates, others argued for new laws.The policy intervention by Marc Francois in 'Tributes to Sir David Amess' suggested that the forthcoming Online Safety Bill, which would 'drain the Twitter swamp', be called 'David's law': 'What better way to ensure that a fine parliamentarian did not die in vain than to enshrine one of his last wishes in legislation forever, for the benefit of all those in public life?'This intervention points to the significance of the framing of government regulation of social media.In the event of his close colleague's murder, Francois advocates for online regulation and elevates this solution to a public good.
Political parties have also been represented as actors with responsibility for addressing violence.Notably, in the 2017 debates, some Conservative speakers blamed the Labour Party for abuse.In an implicit assignment of blame, debate proposer Simon Hart (July) expressed: 'it all seems so different, with swastikas on election boards and offensive slogans and language on posters'.This blaming may be aligned with the usual positionality of debates being used to criticise political opponents (Proksch and Slapin, 2015).In 2017, the politicisation of violence towards MPs could be at least partly interpreted as a political strategy on behalf of the governing party to criticise their political opponents.Some Labour speakers directly or indirectly challenged the frame of Labour as culpable, for example: 'We would be doing a disservice to the democracy we all believe in if we did not recognise that this is an issue for all political parties' (Cat Smith, September).By 2019 and 2021, there was less explicit blame of individual parties and a greater focus on the need for party-specific and/ or joint codes of behavioural conduct that address violence.For example, Simon Hart (May 2019) suggested: 'there are the party codes, which … have proceeded at a reasonably gentle pace … this is the time to put our foot on the accelerator a bit'.

A gendered problem: 'the greatest intensity of threats'
Aligned with the normalisation frame, over time, there was increased attention towards the specific experiences of women, which impacts democracy.By 2021, there was a specific debate: 'Elected women representatives: online abuse'.
In the 2014 debate, there were some references to individual women, though not explicitly linked to their gender group.John Mann referred to racism against Luciana Berger, and Parliamentary Under-Secretary of State for Justice Shailesh Vara noted the arrests made for abuse towards Stella Creasy MP 'after she had supported a feminist campaign'.While these interventions may suggest gendering, there was not an explicit framing of these women being targeted because of sexism.In contrast, there was a group or structural focus on anti-Semitism.For example, Mann expressed: 'The grotesque racist abuse from a whole range of people in the past few weeks aimed at Unauthenticated | Downloaded 11/01/23 06:26 AM UTC my hon.Friend the Member for Liverpool, Wavertree (Luciana Berger) has been a factor in my request to Mr Speaker to grant this debate.'Notably, there was not explicit reference to the intersectional sexist and racist abuse experienced by Berger that specifically targeted her as a Jewish woman (which has been identified in research [see, for example, Kuperberg, 2021]).
Even as women's group experience became a more significant part of the framing, sexism was usually framed as a singular oppression rather than intersecting with other abuse.On the rare occasions the word 'sexism' was mentioned, it was usually grouped with other types of abuse, for example: 'Intimidation, including death threats, criminal damage, sexism, racism, homophobia and antisemitism, has no place in our democracy' (Cat Smith, May 2019).When applied, intersectionality was usually addressed implicitly, framed as minority ethnic women receiving the most abuse, and often with specific reference to Diane Abbott.For example, Cat Smith (September 2017) expressed: 'research by Amnesty International [Dhrodia, 2018] found that my right hon.Friend the Member for Hackney North and Stoke Newington (Ms Abbott) … received half of all the threatening tweets sent to women MPs'.In the next debate, Home Secretary Amber Rudd declared: 'Everyone in this House condemns particularly the abuse and harassment received by the right hon.Member for Hackney North and Stoke Newington.'Notably, this intervention framed the abuse not explicitly as gendered and/or racialised but with a focus on an individual.Some interventions used a clear intersectional framing.For example, in 2017, Chi Onwurah (September 2017) argued: 'We all know that candidates are often targeted because of their gender, sexuality, class and/or ethnicity.'In the 2021 debate on online abuse towards women, there were several interventions focusing on an intersectional gendered group.Three speakers explicitly used the word, for example, Abena Oppong-Asare argued, 'Online abuse is disproportionately experienced by women from an intersectionality identity', and there were several other references to particular 'barriers' faced by women of colour.
The main gender dimension of the problem representation is that women MPs face particularly high levels of abuse, which impacts their political participation and, as such, democracy.The logic seems to be that diversity is normatively important to democracy.While this frame was often implicit, it was sometimes made explicit.For example, Chi Onwurah (September 2017) warned: 'We cannot allow abuse to prevent women and ethnic minorities from entering politics.This Parliament is the most diverse in history: a record number of women, LGBT [lesbian, gay, bisexual and trans] and ethnic minority MPs were elected … failure to act risks reversing the progress made.'Similarly, Maria Miller (April 2021) noted the 'huge progress' in the UK electing women and that 'Parliaments are at their best when they are diverse … equal participation of men and women strengthens our democracy.' In the debates, women MPs seemed to drive the gendered problem representation.Women MPs negotiate the positionality of being both advocates and the impacted population.For example, Diane Abbott (July 2017) shared her experiences of 'racist and sexist abuse', and argued: 'This is not about just politicians or even women politicians.Any woman who goes into the public space can expect that type of abuse.'Similarly, Vicky Ford (May 2019) cited Inter-Parliamentary Union (2016) research to argue: 'There is serious evidence that that puts women off standing for Parliament, thus directly impacting our democracy….This is a problem not only for our democracy but for democracies across the world.'Here, the problem was elevated to a wider, global normative ideal of democracy.In the April 2021 debate, nearly all the speakers referenced research including by the global Inter-Parliamentary Union (2016; 2018) and the UK Fawcett Society (2021).
While the gendered dimension seems to be primarily driven by women, some men also pointed to women's experiences, framing them as particularly egregious.For example, Tommy Sheppard (July 2017) reflected: 'I am extremely concerned that it seems that the majority of the perpetrators of such abuse are male and the majority of the targets are female Members -or at least the greatest intensity of threats is directed towards them.'This 'greatest intensity' idea was echoed by other men (including Martin Whitfield and Jim McMahon, September 2017).Mostly notably, in 'Tributes to Sir David Amess', Mark Francois said: 'he had become increasingly concerned about what he called the toxic environment in which MPs, particularly female MPs, were having to operate.He was appalled by what he called the vile misogynistic abuse that female MPs had to endure online'.That he used gender in his intervention indicates the prevalence by 2021 of the frame that women experience particularly aggressive abuse.

Discussion and conclusion
The analysis has established that the phenomenon of violence (including abuse, harassment and intimidation) towards politicians has been politicised.In parliamentary debates during 2010-21, the framing of the phenomenon shifted over time from one that was experienced by some MPs to a 'normal' part of MPs' jobs, especially for women.The framing that abuse has problematically become normalised reflects theoretical frameworks of gendered political violence stating that violence should not be 'the cost' of participating in politics (Bardall, 2011;Krook, 2017;2020;National Democratic Institute, 2018).Such violence is important not merely because of its impact on individuals but because of its impact on the political system.MPs elevate abuse experienced by themselves to a threat to democracy -an assumed public good.This elevation is likely a strategic way for MPs to navigate their dual role as advocates and impacted population.The problem is beyond themselves and risks the structural, normative public good of democracy.
While the normalisation frame emerged over time, a consistent frame was that violence in politics is predominantly an online problem.An assumption is that online abuse is harmful.There is evidence of MPs linking different forms of violence, including murder, harassment and online abuse, which supports the theoretical work on using a 'comprehensive' definition of violence (see, for example, Krook, 2020).These findings align with research, particularly from the 'online violence against women' literature, that online abuse is violent and impactful in its own right and that 'the new public square' of social media (re)produces gendered inequalities (Jane, 2017;Barker and Jurasz, 2019;Holm, 2019).As expressed in the debates, part of the cause of harm is the ease of abuse spreading, which is reflected in cross-disciplinary research about online abuse (Bardall, 2013;Kosmidis and Theocharis, 2020;Krook, 2020).
Crucially, over time, especially from 2017, the experiences of women became more salient in the framing.A diversity logic is employed, that is, that women's experiences of abuse are particularly important politically because of the impact on political participation, which risks democracy.MPs seem to refer to the theory that women's descriptive representation improves substantive policymaking (Phillips, 1994) and, as such, link diversity with the normative good of democracy.
It is likely that this diversity logic is used largely because of the efforts of feminist researchers and politicians.As others have pointed out (Campbell and Childs, 2013), feminist scholarship has a long history of influencing activism and policymaking.Among UK policymakers, there has been a growing understanding of the importance of diversity.For example, Parliament commissioned a gendersensitive review, resulting in 'The good parliament' report (Childs, 2016).Indeed, an explanation for the presence of gender dimensions in the debates could be the impact of the framing by feminist researchers and third sector groups, especially Amnesty International, the Fawcett Society and the Inter-Parliamentary Union.The use of such evidence by MPs reflects the growing research on, and global attention to, gendered violence in politics; such global politicisation seems to be at least partially explained by feminist activism (Krook, 2020;Restrepo Sanín, 2022a).In addition, the impact of the #MeToo movement, including the specific movement at Westminster (Julios, 2022), even though this was not focused on MPs as victims, could have contributed to more awareness and focus on gendered power dynamics and the experiences of women.
While there is some reference to other diversity groups and the intersection between them, especially by women politicians of colour in the April 2021 debate, overall, there is a focus on women in relation to diversity and repeated silences on intersectionality.This separation rather than connection between different group identities is perhaps unsurprising given the UK equalities political discourse, which tends to separate structural oppressions and group experiences (Christoffersen, 2021).
This article has focused on the language representation of the phenomenon but pays some attention to politicians' identities.For example, it seems to have been women driving the gendered representation of the phenomenon, and the focus on race largely centres on the experiences of Diane Abbott.A particularly fruitful avenue for future research would be to develop theoretical ideas outlined earlier around MPs' positionality and advocating for issues that impact them.Scholarship could build on the rich theoretical work on representation (Lovenduski, 2014;Celis and Childs, 2020;Hargrave and Blumenau, 2022;Siow, 2023) and integrate the theories on feminist institutionalism to theorise ways in which the gendered and raced 'rules' impact the problem framing (Hawkesworth, 2003;Puwar, 2004;Mackay and Krook, 2015).
The solution or 'prognosis' frame identifies social media platforms, the government and political parties, which reflects research on the need for multisectoral approaches to address political violence, especially its structural gendered dimensions (National Democratic Institute, 2019).A notable 'silence' was a specific policy solution on violence against women in politics, which has been the approach of a growing number of countries.Further research could adopt a comparative approach to interrogate policy problem representations from different contexts, especially how gender is considered.Specific country or comparative research could examine a wider range of political data, such as policy documents and media interventions, and more systematically examine the causal process of the problem framing.
The explanation for the phenomenon of violence towards politicians becoming politicised is likely multifaceted.The rise in social media use seems to have caused an increase of abuse through the nature of the platform (which allows anonymity and abusive content to spread easily), its lack of regulation and the ongoing expectation of MPs' presence on the platforms (Baxter et al, 2016;Ward and McLoughlin, 2020).Furthermore, discrete events are likely explanatory.The 2016 murder of Jo Cox evidently raised awareness of the risks of violence.Notably, it was not until 2017 that there was a specific debate on general abuse towards MPs, which suggests that it was the murder coupled with other factors that led to the politicisation.As shown by this representation of multiple causes, growing political divisions are represented as a causal factor for abuse.This echoes research from the US which finds that political polarisation is associated with increased violence towards politicians regardless of gender (Thomas and Herrick, 2023).Yet, a notable 'silence' is the referendum on leaving the European Union (Brexit), which is only explicitly referenced occasionally.Perhaps this is because MPs see this event as an obvious factor.For example, Andrew Percy (May 2019) said: 'Brexit is, of course, a particular source of such abuse.'The 2017 election is more of a focal point in the representation of violence towards politicians, with two of the three debates that year focused specifically on the election.Importantly, all these explanations reflect insidious and broader realities around political divisions and structural inequalities.
To conclude, this article has offered a unique analysis of how the phenomenon of violence towards MPs is framed by them as policymakers in parliamentary debates.The article has confirmed and developed emerging understandings that there are gender differences in how men and women MPs experience violence, including a specific problem of violence against women in politics.Crucially, the article moves beyond the evidence of the phenomenon to consider its political impacts.By applying feminist frame analysis to gender and political violence, this research demonstrates the importance of recognising this phenomenon in political/policy discussion.Violence towards politicians, including an increasing focus on the experiences of women, has been politicised over time as a normative policy problem in the UK.This article develops theories of gender and political violence by demonstrating that the problem of gendered political violence can and should be conceptualised not only as a unique phenomenon but also as a policy problem.

Notes
1 Gender, of course, includes more than the binary of men and women.However, gender diversity is unlikely to be considered in these specific debates about politicians as, until recently, there were no openly gender-diverse MPs.Jamie Wallis 'came out' as the first openly trans MP in 2022 (Allegretti, 2022

House of Commons debates specifically on the phenomenon of violence against politicians during 2010-21
). 2 The list includes: Jo Cox MP in 2016; Stephen Timms MP, who survived a stabbing in 2010; Councillor Andrew Pennington, who died in a sword attack directed towards Nigel Jones MP in 2000; Airey Neave, Robert Bradford, Anthony Berry and Ian Gow, MPs who were assassinated between 1979 and 1985 by the Provisional Irish Republican Army (IRA); and parliament police officer Keith Palmer, who died during the 2017 terrorist attack on Parliament. 3 Including Swinford, 2017.