Comparing Fem(in)cide: The Perils of Turning Data Into Action

Conceptualising frictions

While [femicide] has drawn attention to special ways in which women are selectively targeted, the definition has progressively become diluted and confused.

—Alvazzi del Frate, When the victim is a woman, 2011

The conceptual frictions between ‘femicide’ in English and ‘feminicidio’ in Spanish go beyond mere translations. The vast majority of research papers and articles, reports, transnational documents, dissertations, books and so on, address these differences to some extent (Corradi et al. 2016; Fregoso and Bejarano 2010; Howe and Alaattinoglu 2019; Toledo Vásquez 2009; Weil, Corradi, and Naudi 2018; UN Women 2014; UNODC 2015, 2018). Given that violence is conditioned by and bound to particular contexts it is crucially important to (re)cognise the situatedness of knowledge. However, and as the concepts of femicide/feminicidio/the gender-related killing of women fluctuate between bodies of knowledge and communities of practice, these particularities get diluted for the sake of creating a homogeneous (or ‘global’) picture. As Bueno-Hansen (2019) argues, generalisations of the circumstances that result in fem(in)icide ‘tend to gloss over complex details, weakening the international coherence of the campaign, making it necessary to ground its usage at every analytical level, from international to regional to national’ (2010: 308).

‘Femicide’ was publicly enunciated for the first time by Diana Russell at the International Tribunal on Crimes Against Women, held in Brussels in 1976. She defined it then as ‘the killing of females by males because they are females’. Twenty years later, due to the alarming escalation of the systematic violent murders of women at the border city of Ciudad Juárez (1) , the concept was repoliticised into ‘feminicidio’. The importance of this placebased epistemological conception (from ‘femicide’ to ‘feminicide’) lies in the fact that, in the Mexican context, the state played an important role by omitting, neglecting, and even colluding in these violent crimes against women. Digging deeper into these tensions and frictions is important as it might give us a good insight into the subsequent practices of quantification that resulted from the introduction of this political concept not only into the legal, penal, criminological and statistical realm but also into a further ‘global’ comparison.

In line with this argument is Silvana Fumega, who writes ‘[…] in the case of femicide specifically, there are often differences between judicial, prosecutorial, and statistical institutions of the state. In other words, the same database can admit different interpretations from different public and civil society organisations in different moments, with different additional sources of data, or in line with different definitions of femicide’ (2019). She reads this, however, as a strength since it triggers a dialogue across sectors. I agree with this view and further stress the impact that definitions have yielded on quantification practices.

However, the epistemological debate concerning the conceptualisation of fem(in)icide is not what concerns this paper. Nevertheless, and for the sake of the discussion, I believe it is important to share the definition of fem(in)icide I find most comprehensive:

‘Building on the generic definition of “femicide” as “the murder of women and girls because they are female” (Russell 2001), we define “feminicide” as the murders of women and girls founded on a gender power structure. Second, feminicide is gender-based violence that is both public and private, implicating both the state (directly or indirectly) and individual perpetrators (private or state actors); this encompasses systematic, widespread, and everyday interpersonal violence. Third, feminicide is systemic violence rooted in social, political, economic, and cultural inequalities. In this sense, the focus of our analysis is not just on gender but also on the intersection of gender dynamics with the cruelties of racism and economic injustices in local as well as global contexts. Finally, our framing of the concept follows Lagarde’s critical human rights formulation of feminicide as a “crime against humanity”’ (Fregoso and Bejarano 2010: 5).

After establishing the frictions that result from trying to define something so complex as the killing of a woman due to her ‘gender’, the next sections will engage with the logics of quantification, as a methodological practice where power and knowledge are mutually constitutive of one another. The need to use data on violence against women and girls (VAWG) in general and fem(in) icide in particular to ‘trigger action’ is discussed. Thereafter, I grapple with empirical ‘global’, ‘regional’ and ‘national’ efforts to compare fem(in) icide across time and spaces. The specific case of Mexico is zoomed in on to gain a closer insight into the challenges of comparing fem(in)icide at the national level.

Quantifying those frictions

Data are themselves part and parcel of that changing object of study. Quantitative data are always and necessarily manifestations of the process involved in describing a changing object of study; data do not exist in a vacuum independent of our knowledge any more than the objects they represent.

—Emma Uprichard, Dirty Data: longitudinal classification systems, 2012

Quantifying is a method, an act of measuring, judging, valuing, ordering, prioritising, (re)cognising. Quantifying not only describes but, perhaps more bluntly, produces the reality it aims to understand. Quantifying is a technical and social practice. ‘It is technical in that it involves measurements and social in that it involves agreement and convention, hence the now classic equation whereby to quantify = to agree + to measure’ (Desrosières and Kott 2005 in Bruno et al., 2016: 4). In this sense, practices of quantification involve a complex interaction between human expertise and material disposition (in the shape of tools, machines, and so on).

The work of Derosières et al. depends on the one hand, technical (pragmatic, empirical) knowledge and, on the other hand, conceptual (categorical) taxonomies of what needs to be counted. Paying special attention to the methodologies behind quantification can be useful for understanding the power dynamics which are embedded in the unequal global political economy. The question is: to what extent can these comparisons be made on a regional, national or ‘global’ level?

Turning ‘data into action’

A serious obstacle to progress is the inadequate and uneven data on various forms of violence against women and on how they affect different groups of women. The lack of data to evaluate the measures taken impedes informed analysis and policymaking, which are critical to developing the most effective responses.

—UNGA, In-depth study on all forms of violence against women, 2006

The trope of ‘turning data into action’ has become ubiquitous these days (2) and the field of violence against women and girls (VAWG) is no exception. A transnational consensus concerning the need to generate data on this matter is unquestionable, as the quote above suggests. From sexual harassment to fem(in)icide, we need to understand the scope extent of violence in order to generate effective policies and programmes. Yet, the consensus regarding how to quantify and with which parameters to use is far from being met.

This has been my particular research interest during the last three years. I have closely studied how fem(in)icide is conceptually resignified as it travels across time and space: from Diana Russell’s ‘femicide’ in 1976, to ‘feminicidios’ in México during the nineties, to the current ‘gender-related killing of women’ (UNODC 2015, 2018), and ‘intimate femicide’ (Walklate et al. 2019). I am intrigued by how fem(in)icide is being quantified and the perils of transnationally comparing such data.

Which data and what actions? Quantifying frictions across tiers: from global and regional to national

The need to collect ‘globally’ comparable data on all forms of violence against girls (VAWG) was stressed in the 2006 United Nations Secretary General’s In-depth study on all forms of violence against women (VAW) (). This study triggered a wide array of policies and practices destined to generate data and statistics on VAWG. Fem(in) icide was directly addressed in this document as a ‘form of violence against women in the community’ (3) (UNGA 2006: 127) and as a form of VAW that is under-documented (UNGA 2006: 223-225). Since then, the emergence and development of statistics on fem(in)icide has been exponential. Notable examples include the In-depth study on all forms of violence against women and UNODC’s 2019 Global Study on Homicide: Gender-related killing of women and girls.

What counts and who counts it matters at every scale of the analysis. Although the definition of fem(in)icide involves the killing of women from ‘both the state (directly or indirectly) and individual perpetrators (private or state actors)’ (Fregoso and Bejarano 2010: 5), UNODC’s 2018 study reduces it to ‘female victims of homicide perpetrated by intimate partners or family members […] ‘given severe limitations in terms of data availability’ (UNODC 2018: 10, 12). This means that even though 87,000 women were included in their report, these numbers belong to states which have a relatively stable statistical system that allows for the victim and perpetrator relationship to be reported.

This, however, is not the case in Mexico. In spite of having a strong statistical institution (INEGI), impunity and opacity hinder the inclusion of other variables to better understand homicides. More on this will be addressed in the next section, where the case of Mexico will be zoomed in on to strengthen the argument about the relevance of the contextual particularities when we seek to compare fem(in)icide. Framing fem(in)icide ‘globally’ relies on methods of comparison which ‘call relations of interdependence, connection, and disjunction into being’ (Choy 2012: 12). The idea of the ‘global’—and its comparison in data collection—becomes an important vector to think about the frictions of scale, position, perspective and/or depth (Lury 2018).

By the time UNODC’s document was published in 2018—and up until today—Latin America remains the only region with specific legislations designed to criminalise fem(in)icide. (4) This is important since legal implementation normally precedes and/or accompanies official efforts in data collection. The Gender Equality Observatory for Latin America and the Caribbean from the Economic Commission for Latin America and the Caribbean (ECLAC) offers the most comprehensive overview of the region. Their regional femicide indicator ‘corresponds to the annual total of gender-based killings of women. According to national laws, it is referred to as femicide, feminicide, or aggravated homicide due to gender’ (ECLAC 2019). The indicator started to expand in 2009 to include data from countries which quantified women killed by their current or former partners (Venezuela, Costa Rica, Colombia and Guatemala). The Gender Equality Observatory for LAC currently has official data from 34 countries with regard to three indicators: intentional homicide of women, femicide and intimate femicide.

Data from its most recent report, in 2019, compares numbers from 16 Latin American countries from 2019 and nine Caribbean countries from 2018, some of which only collect data from killings committed by the victim’s current or former partner (ECLAC 2019) (5) . Figure 1 shows the results in absolute numbers and rates per 100,000 women. Based on their results, the Caribbean seems to have the worst-off situation with Guyana at the head (8.8), followed by Saint Lucia (4.4), Trinidad and Tobago (3.4) and Barbados (3.4). In Latin America, El Salvador (6.8), Honduras (5.1), Bolivia (2.3), and Guatemala (2.0) have the highest rates per 100,000 women.

It is important to mention the caveats the ECLAC observes when presenting these data. First, they acknowledge that ‘regional averages cannot be calculated owing to the heterogeneity of national data sources and methodologies’ (ECLAC 2019). This means that, in spite of the legal and statistical national efforts to collect data on fem(in)icide for over 10 years we are still unable to fully grasp its regional extent. They further recognise other complexities in comparing data collection between countries since figures are produced by more than one institution (which, in many cases, are not standardised); some do not consider all killings of women by their current or former partners to be femicides; only a minority of countries document gender-related killings of transsexual women or female sex workers as femicides; there is insufficient updating of records (some do it yearly and their approaches widely vary); other victim’s characteristics (i.e. race, ethnicity, sexual identity or orientation, education level or place of residence) are not recorded (ECLAC 2019: 4).

The regional comparison of fem(in)icide, exemplified with the case of Latin America, shows how logics of quantification are often bound to a spatial condition of (re)production. Adding a longitudinal analysis complicates this further. As Uprichard suggests, we need to re-think how longitudinal quantitative research explores the qualitative and quantitative changes within each classification. ‘Capturing trajectories’ poses a big challenge when we try to bring together datasets that span across different timeframes. To capture the trajectory of fem(in)icide as a statistical category requires paying closer attention to the methodological issues related to longitudinal social research (Upichard 2012). This is particularly challenging when we are valuing or measuring categories that existed—in this case, the intentional violent killing of women—but were not labelled as fem(in)icide. Zooming in on the particularities and experiences of the Mexican context will serve as a poignant example of this.

Scattered data and dislocated actions: Mexico and the challenges of quantifying fem(in)icide

In Mexico, two bodies of knowledge produce statistics on sex-disaggregated homicides: the National Institute of Statistics and Geography (6) (INEGI) and the Executive Secretariat of the Public Security National System (7) (SESNSP), from the Secretariat of Public Security (SSP). The former publishes its reports yearly and it is based on administrative records (i.e. ‘alleged homicides’ death certificates). The latter presents a monthly report of crime incidence, which includes fem(in) icide. It is important to stress the particularities of the Mexican political system. Being a federal presidential constitutional republic, it renders independence to its 32 states under a central federal government. This means that each state has its own penal code with their own ways of typifying crimes. Fem(in)icide is no exception, evidencing different ways of defining and therefore quantifying fem(in)icide. This poses spatial and longitudinal frictions when comparisons are made.

For instance, if we only look at the SESNSP’s data—and render numbers as the main drivers of our argument—we could safely say that fem(in) icides in Mexico have more than doubled from 411 in 2015 to 983 in 2019. Implying that fem(in) icide has risen so dramatically in recent years, as many news reports infer (8) , ignores the historical contingency that precedes the categorical making of those numbers. Even though Mexico introduced feminicidio in its federal penal code in 2012, it was not until the end of 2017 that all 32 states collected data on this matter (9) . The bigger the scale of comparison, the more complex the process of aligning methods and methodologies becomes. In addition, in 2015, the National System for Public Security (SESNSP) implemented a new methodology to quantify the crime incidence rate. This means that the way in which crimes were typified, categorised and counted differ before and after this year. (10)

A focus on fem(in)icides in 2017, the same year that was analysed for UNODC’s Global Study on Homicide: Gender-related killing of women and girls, shows a disparity in the data. The SESNSP reported 741 women victims of fem(in)icide (11) with the INEGI recording 3,430 women being killed that year. This means that not all female homicides are categorised under fem(in)icide. Numbers differ further if we take into consideration the 2,205 intentionally murdered women documented by María Salguero, a digital activist who produces her own data on fem(in)icide based on media reports. Salguero argued that from those 2,205, at least 1,159 should be categorised as fem(in)icide in 2017 (Carranza et al. 2018), 55% more than what the SESNSP accounts. Neither the SESNSP nor INEGI’s official records are quantified in UNODC’s study since they do not specify the relationship between the victim and the perpetrator or the motive of the killing. Salguero’s database does and thus renders a deeper insight into the reasons behind and patterns of the killings.

Can we understand behavioural patterns in the killing of women in Mexico across time and between states if legal and statistical categories are constantly shifting and fluctuating? A remarkable report from Data Cívica, an NGO which focuses on producing data for human rights, offers an insightful approximation to this answer. In their Claves para entender y prevenir los asesinatos de mujeres en México (Keys to understand and prevent the killing of women in Mexico) (2019), they analyse, precisely, what has changed and what remains the same in terms of the intentional killing of women. Interestingly, they decided to use INEGI’s sex disaggregated homicide rates published instead of the SESNSP’s fem(in)icide data to compare and contrast patterns of violence across time. Data Cívica explains that, in spite of having a General Law on Women’s Access to a Life Free of Violence (12) (LGAMVLV), enacted in 2007— which Article 21 directly deals with ‘femicide violence’ (13)— the 33 penal codes in Mexico define ‘feminicidio’ differently. Some, for example, do not include impunity as a key variable to typify the killing as a femicide, as Article 21 suggests. Also, the ‘gender-related motives’ are not the same. If we add to these legal definitions, disputes over the ways in which each state’s attorney and prosecutor offices classify the intentional killing of women another layer of complication is added.

Furthermore, neither the INEGI nor the SESNSP include relevant variables to facilitate the classification of violence, such as: previous family violence complaints, record of kinship between the victim and the alleged aggressor, whether or not the victim spoke an indigenous language, the victim’s sexual orientation and gender identity, immigration status, whether or not they have a disability, where the body was found (not where the murder occurs), the number of the perpetrators, signs of aggressions prior to the murder and many other important elements.

Notwithstanding these limitations, their results provide strong evidence that patterns of what has varied and what has not. Contrary to the ‘global’ trend which suggests that fem(in)icides occur mostly inside the household, the last ten years have shown a distinctly different logic in Mexico. Figure 2 presents the differences across time in women’s homicides. While there is a slight variation in the killing of women in their households between 2000 and 2017, there is a significant increase in those committed in public spaces. The killing of women and men in the public space has doubled (14). The explanation is both political and contextual. In 2007 the former Mexican president Felipe Calderon, together with the U.S. government, launched a ‘War on drugs’, which immersed the country in a wave of rampant violence. María Salguero, in her recent work on fem(in)icide maps shows this pattern more clearly and there seems to be a distinct geographical correlation between increased rates of fem(in)icide and the activity of cartel groups As Figure 2 clearly shows, after 2007 violence exacerbated reaching its peak in 2012. Taken all together, the data analysed by Data Cívica provides evidence that the use of firearms to kill women has widely increased both in the public and the private sphere from 0.7 per 100,000 women in 2007 to 2.7 in 2017, an increase of 258.7% (Data Cívica 2019: 34). This has been the case for men as well. Nevertheless, and as Figure 3 reveals, other causes of death, such as choking or suffocation and those committed with a sharp instrument (arma blanca) are still more prevalent in the killing of women than the killing of men.

The results from Data Cívica’s report yield some interesting findings and shed light on the complexities of quantifying fem(in)icide across  time and space. First, that women in Mexico face a double risk when it comes to ‘feminicidal violence’: they are unsafe both at home and in the streets. While the intimate femicide is widely recognised, fem(in)icides perpetrated by strangers on the streets seems to have less impact in the global agora. 

One possible cause of this is the lack of resources available to provide access to justice and accountability of particular contexts. The frictions of comparing data between countries is not only methodological. It is alleged that in Mexico, nine out of ten homicides go unpunished, an impunity rate of 89%. Weak institutions, the lack of political will, a shortage of human, financial and technological resources, complex bureaucratic frameworks and the lack of transparency and accountability all play a role in the problems of quantification and in establishing which bodies count and which not in Mexico. The question here is not necessarily which ‘global realities’ are turned into numbers but rather how and under what conditions render these bodies visible. 

Moving forward: Reflexivity in data collection and the need to (re)cognise alternative ways of collecting data

From the 2006 In-depth study on all forms of violence against women to 2018 UNODC’s Global Study on Homicide: Gender-related killing of women and girls the need to generate comparable data on fem(in)icide is unquestionable. Based on content analysis of global, regional and Mexican data on fem(in)icide, this paper has shown how practices of quantifying it vary greatly from one setting to another. When the case of Mexico is positioned vis-à-vis a regional or global scale, we realise how specific contextual particularities, such as the increase of VAWG in the public space, get overlooked. These frictions are clear not only when we dig into the conceptual differences between femicide and feminicidio but even more so when we look into the numbers produced. 

In this sense, the general picture emerging from the data analysis is that specificities get diluted as fem(in)icide is rendered ‘comparable’ across time and spaces. As Merry argues, ‘the process of translating the buzzing confusion of social life into neat categories that can be tabulated or defined risks distorting its complexity’ (2016: 3). Therefore, the translation of a ‘global’ or ‘regional reality’ into concepts, indicators and numbers, involves an inevitable distortion. When these divergent concepts, indicators and numbers are equated, ‘they are inevitably stripped of their context, history and meaning’ (Merry 2016: 3).

Acknowledging this complexity and given that there is no infrastructure to build an alternative body of knowledge which independently collects data on fem(in)icide (15) with a flexible yet robust methodology yet, I suggest two possibilities to address these entanglements. The first one is ethical, as it sheds a light on reflexivity as a crucial element of feminist research ethic. This has been widely discussed in sociological literature (Ackerly and True 2008; Bourdieu 1990, 2004; Bourdieu and Wacquant 1992; Wooglar 1991) but has rarely permeated other epistemologies in knowledge production, such as international and nongovernmental organisations. Rather than providing an overview of what reflexivity is, I will make use of the set of questioning practices Ackerly and True (2008) put forward as feminist research ethics.

As mentioned along this paper, (re)producing reality into numbers entails an articulation of power and knowledge behind the ways in which meanings are enacted. Informed by a critical feminist perspective, Ackerly and True’s take on reflexivity begins by asking how we can ‘study power and identify ways to mitigate its abuse in the real world when […] researchers also participate in the powerful projection of knowledge in this world’ (2008: 694) and how our own subjectivities affect both researchers and the processes themselves? In order to respond to these questions, they stress the commitment to inquiry about how we inquire. In this sense, our research ethics need to be attentive to:

1. the power of knowledge: research processes reproduce power differences. Recognising that there are different ways of knowing ‘each opening and foreclosing certain understandings of what it means to know and to contribute to shared knowledge’ (2008: 695) enhances research practices. In this regard, moving beyond the ‘intimate femicide’ category might visibilise that other forms of fem(in)icide are also prevalent in certain settings.  
    
2. boundaries, marginalization and silences: this relates to the need to pay attention and to interrogate what we include or exclude from our research process: ‘we should, therefore, examine the function of boundaries and consider their effects on what is important for us to study and how we can study it’ (2008: 697). This is particularly important when producing data on sensitive information, such as fem(in)icide. Being weary of contexts with a high rate of impunity, which often marginalizes certain types of poor peripheral bodies, is key to staying reflexive in our research.
    
3. relationships and their power differentials: remaining attentive to the interconnections with others that inform and nourish our research. As the authors point out: ‘being attentive to relationships in our research means recognizing not only the power dynamics among research subjects (organizations or individuals) and the relationships of power in which they are embedded, but also between the researcher and their research subjects or between researchers’ (2008: 698). Triggering dialogues about positionality and/or intersectionality amongst our peers, colleagues on these issues becomes a valuable pathway for reflexivity as well.
    
4. the situatedness of the researchers: this echoes to Donna Haraway’s situated knowledges in the sense that objectivity is only possible if we ‘take responsibility both for our necessary situatedness, and for the recognition that we are located in and produced by sets of partial connections’ (1988: 69). This is important when we deal with any form of ‘global’ knowledge as it requires bodies in specific settings and translations in languages that might not be our own. Being aware of our own positionality is a form of self-reflexivity which is attentive to dynamics of power.

The second and final suggestion, which is the subject of my current PhD dissertation, is concerned with the incorporation of alternative practices of knowledge production in the makings of fem(in)icide data. Digital technologies have opened access to new ways of generating data beyond official statistics. Data activists—a term coined by Gutiérrez and Milán (2015)—from Mexico to Turkey and Australia are using different digital technologies and infrastructures as methods and vehicles to build their own datasets. This data often contains more precise patterns and categories which helps to generate a more sophisticated understanding of the violence women endure in specific settings. Looking into these practices not only allows us to include more local narratives as the scale of comparison broadens but also strengthens the reflexivity in our research.

An Article by Saide Mobayed

Comments

1. By this I refer to ‘las muertas de Juárez’ (the death women of Ciudad Juárez): in the mid-nineties hundreds of young women were abducted from the public space to then be sexually abused, battered, killed and disposed of in the desert of Chihuahua. Many of them (approximately one third according to JuliaMonárrez Fregoso) worked at the maquiladoras, ‘Americanowned transnational factories’ (Arriola 2010: 25). Up until today what happened in Juárez remains a puzzle that can only be put together when we look at the a) complex global political economy (e.g. the neoliberal politics of the NAFTA free trade agreement; the incursion of female labour markets ); b) a patriarchal and misogynistic culture rooted in historical and political factors; c) the geographic setting of a rapidly changing desertic ‘city’ at the border between Mexico and the United States; d) the geopolitical financial flow of drug trade. Up until today the crimes are unsolved. However, Gaspar de Alba and Guzman recount some of the following ‘hypothesis’ that have been pointed out: ‘serial killers; satanic cults; snuff films, organ harvesting; white slavery; the Egyptian chemist “mastermind” (arrested in 1995); Los Rebeldes (local gang arrested in 1996); Los Choferes (band of bus drivers arrested in 1999); corrupt Mexican police; well-protected sons of rich families; cartel killings; the victims were leading double lives as prostitutes; the victims dressed provocatively in short dresses and high heels; unemployed men resentful of women getting jobs” (Gaspar de Alba and Guzmán 2010: 67).
    
2. This has exacerbated with the use of Big Data.
    
3. The case of Juárez is mentioned in this section, which shows its transnational relevance and influence.
    
4. With a total of 18 countries (UNODC: 57-62).
    
5. Bahamas, Barbados, Belize, Chile, Colombia, Grenada, Jamaica, Nicaragua, Puerto Rico and Suriname.
    
6. Instituto Nacional de Estadística y Geografía.
    
7. Secretariado Ejecutivo del Sistema Nacional de Seguridad Pública.
    
8. El Financiero. 2019. ‘Feminicidios en México crecen 111% en los últimos 4 años’. Available at: https://elfinanciero.com.mx/ nacional/111-mas-feminicidios-en-mexico-en-los-ultimos-4- anos.
    
9. Chihuahua, the state which saw the waves of fem(in)icides in Ciudad Juárez, was the last one to do so.
    
10. 10  In 2015 the SESNSP underwent a series of highly important changes in their methodology (‘Tool for the Registration, Classification and Reporting of Crimes and Victims’ (Instrumento para el Registro, Clasificación y Reporte de los Delitos de las Víctimas CNSP/38/15)) with the objective of expanding the disaggregation of crime categories and specifying the criminal information. This means that 20% of the crimes that were classified as ‘Others’ have either their own category now or where expanded in this new methodology. Relevant for violence against women and girls are: feminicidio, gender violence (violencia de género), family violence (violencia familiar), breach of family assistance obligations (incumplimiento de obligaciones de asistencia familiar), sexual abuse (abuso sexual), corruption of minors (corrupción de menores), sexual harassment (acoso y hostigamiento sexual), human trafficking (trata de personas). Regarding the ‘feminicidio’ category, it is important to mention that the SENSP defines it as ‘the criminal conduct committed by someone who deprives the life of woman based on her gender’. The SESNP understands as ‘gender-based reasons’ a list of nine specific circumstances (SESNP 2018). For more please go to: http://secretariadoejecutivo.gob.mx/docs/ pdfs/nueva-metodologia/Lineamientos_registro_feminicidio_ CNPJ_aprobada_5MZO2018.pdf
     
11. It is important to notice that yearly numbers vary according to the National System for Public Security (SESNSP) publications, which are published every month. The most recent one (from April 2020) shows 741 women victims of ‘feminicidio’ in Mexico in 2017 while, the document published in February 2019 states that there were 736 ‘feminicidios’ and the one from January 2018 shows a total of 671. This exemplifies the not only the constant readjustments of the data but also the efforts of the SESNSP to keep up to date.
     
12. Ley General de Acceso de las Mujeres a una Vida Libre de Violencia (LGAMVLV).
     
13. ‘Femicide violence is the most extreme form of gender violence against women, produced by the violation of their human rights in public and private spheres and formed by the set of misogynist actions that can lead to the impunity of society and the State and culminate in homicide and other forms of violent death of women’ (General Law on Women’s Access to a Life Free of Violence 2007).
     
14. More men die than women. This is unquestionable. However, women are killed in different ways and these patterns of variation across time have changed as well, which is the focus of this paper.
     
15. The UN Special Rapporteur on Violence against Women and Girls, Dr Dubravka Šimonović, has been calling for the establishment of ‘Femicide Watches’ with a ‘flexible methodology that would be workable for all States to assist them in establishing [it] as a separate mechanism or mechanisms attached to existing national mechanisms attached to existing national mchanisms or observatories on violence against women’ (UNGA 2016: 11).

References 

Ackerly, Brooke, and Jacqui True. 2008. ‘Reflexivity in Practice: Power and Ethics in Feminist Research on International Relations’. International Studies Review 10(4):693–707.

Arriola, Elvia, 2010. ‘Accountability for Murder in the Maquiladoras: Linking Corporate Indifference to Gender Violence at the U.S.-Mexico border’ in Gaspar de Alba Alicia and Georgina Guzman, eds. Making a Killing: Femicide, Free Trade and La Frontera. Austin: University of Texas Press, 25–62.

Bigo, Didier, Engin F. Isin, and Evelyn Sharon Ruppert, eds. 2019. Data Politics: Worlds, Subjects, Rights. Routledge, Taylor & Francis Group

Bourdieu, Pierre and Loïc Wacquant. 1992. An Invitation to Reflexive Sociology. Cambridge University Press.

Bourdieu, Pierre. 1990. In Other Words: Essays toward a Reflexive Sociology. Cambridge: Cambridge University Press. Bourdieu, Pierre. 2004. Science of Science and Reflexivity. University of Chicago Press.

Bruno, Isabelle, Emmanuel Didier, and Tommaso Vitale. 2014. ‘STATACTIVISM: Forms of Action between Disclosure and Affirmation’. Partecipazione e Conflitto 23.

Bruno, Isabelle, Jany-Catrice, Florence, and Touchelay, Béatrice, eds. 2016. The Social Sciences of Quantification. From Politics of Large Numbers to Target-Driven Policies. New York: Springer.

Bueno-Hansen, Pascha. 2010. ‘Feminicidio: Making the Most of an “Empowered Term’ in: Bejarano, Cynthia and Rosa-Linda Fregoso (eds.). Terrorizing Women: Feminicide in the Americas. London: Duke University Press, 290-311.

Carranza, Max, Sanabria, Lucy, Espinosa, Gabriela, and Álvaro Cortés. 2018. Nos están matando. Feminicidios en México. Special report for Sopitas. com. Available at: https://www.sopitas.com/ reportajes/nosestanmatan

Choy, Tim. 2011. Ecologies of Comparison. Duke University Press.

Condit, Rich. 2020. ‘Infection Fatality Rate — A Critical Missing Piece for Managing Covid-19’ in Virology blog. About viruses and viral disease. Available at: https://www.virology. ws/2020/04/05/infection-fatality-rate-a-criticalmissing-piece-for-managing-covid-19/

Connell, Raewyn, Fran Collyer, João Maia, and Robert Morrell. 2017. ‘Toward a Global Sociology of Knowledge: Post-Colonial Realities and Intellectual Practices’. International Sociology 32(1):21–37.

Corradi, Consuelo, Chaime Marcuello-Servós, Santiago Boira, and Shalva Weil. 2016. ‘Theories of Femicide and Their Significance for Social Research’. Current Sociology 64(7):975–95.

D’Ignazio, Catherine, and Klein, Lauren. 2020. Data Feminism. The MIT Press. Data Cívica. 2019. Claves para entender y prevenir los asesinatos de mujeres en México. Available at: https://datacivica.org/assets/pdf/claves-paraentender-y-prevenir-los-asesinatos-de-mujeresen-mexico.pdf

Del Frate, Alvazzi. 2011. ‘When the victim is a woman’. Global Buren of Armed Violence. Geneva Declaration on Armed Violence.

Desrosières, Alain. 1998. The Politics of Large Numbers. A History of Statistical Reasoning. Harvard University Press.

Economic Commission for Latin America and the Caribbean (ECLAC) and Gender Equality Observatory for Latin America and the Caribbean. 2019. ‘Measuring femicide: challenges and efforts to bolster the process in Latin America and the Caribbean’.

Economic Commission for Latin America and the Caribbean (ECLAC). 2019. Measuring femicide: challenges and efforts to bolster the process in Latin America and the Caribbean. Gender Observator Equality Observatory.

Engle Merry, Sally. 2016. The Seductions of Quantification. Measuring Human Rights, Gender Violence, and Sex Trafficking. The University of Chicago Press.

Foucault, Michel. 1978. ‘Governmentality’ in Foucault, Michel, Graham Burchell, Colin Gordon, and Peter Miller, eds. 1991. : With Two Lectures by and an Interview with Michel Foucault. Chicago: University of Chicago Press, 87-104.

Foucault, Michel. 1980. ‘Questions of method’ in Foucault, Michel, Graham Burchell, Colin Gordon, and Peter Miller, eds. 1991. The Foucault Effect: Studies in Governmentality: With Two Lectures by and an Interview with Michel Foucault. Chicago: University of Chicago Press, 73-86.

Fregoso, Rosa Linda, and Cynthia Bejarano. 2010. ‘Introduction: A Cartography of Feminicide in the Américas’. in Terrorizing women feminicide in the Américas, edited by R. L. Fregoso and C. Bejarano. Durham: Duke University Press.

Fumega, Silvana. 2019. ‘Standarisation of femicide data requires a complex participatory process, but key lessons are already emerging’ in LSE Latin America and Caribbean blog. Available at: https:// blogs.lse.ac.uk/latamcaribbean/2019/08/08/ standardisation-of-femicide-data-requires-acomplex-participatory-process-but-importantlessons-are-already-emerging/

Gaspar de Alba, Alicia and Georgina Guzman, eds. Making a Killing: Femicide, Free Trade and La Frontera. Austin: University of Texas Press.

Hacking, Ian. 1980. ‘How Should We Do the History of Statistics? in Foucault, Michel, Graham Burchell, Colin Gordon, and Peter Miller, eds. 1991. The Foucault Effect: Studies in Governmentality: With Two Lectures by and an Interview with Michel Foucault. Chicago: University of Chicago Press, 181-196.

Haraway, Donna. 1988. ‘Situated Knowledges: The Science Question in Feminism and the Privilege of Partial Perspective’. Feminist Studies 14(3):575.

Howe, Adrian, and Daniela Alaattinoglu, eds. 2019. Contesting Femicide: Feminism and the Power of Law Revisited. Abingdon, Oxon; New York, NY: Routledge.

Lury, Celia, Rachel Fensham, Alexandra HellerNicholas, Sybille Lammes, Angela Last, Mike Michael, and Emma Uprichard, eds. 2018. Routledge Handbook of Interdisciplinary Research Methods. 1st ed. Routledge.

Milan, Stefania and Miren Gutiérrez. 2015. ‘Citizens’ Media Meets Big Data: The Emergence of Data Activism’. MEDIACIONES 11(14):120–33.

Salguero, María. Yo te nombro. Mapas de feminicidio. Available at: http://mapafeminicidios. blogspot.com/p/inicio.html

Schiffmann, Avi. 2020. World COVID-19 Stats. Available at: https://ncov2019.live/data

Schröttle, Monika and Ksenia Meshkova. 2018. ‘Data collection: challenges and opportunities’ in Weil, Shalva, Consuelo Corradi, and Marceline Naudi. Femicide across Europe : Theory, Research and Prevention. Bristol: Policy Press.

Secretaría de Relaciones Exteriores (SRE). General Law on Women’s Access to a Life Free of Violence. Available at: http://www.summit-americas. org/brief/docs/Law_on_access_to_a_life_free_ violence.pdf

Secretariado Ejecutivo del Sistema Nacional de Seguridad Pública (SESNSP). 2018. Instrumento para el Registro, Clasificación y Reporte de los Delitos y las Víctimas (CNSP/38/15). Available at: http://www.secretariadoejecutivo.gob.mx/ docs/pdfs/nueva-metodologia/Manual_Nuevo_ Instrumento.pdf

Secretariado Ejecutivo del Sistema Nacional de Seguridad Pública (SESNSP). 2020. Información sobre violencia contra las mujeres. Incidencia delictiva y llamadas de emergencia 9-1-1. Información con corte al 31 de marzo de 2020. Centro Nacional de Información (CNI).

Toledo Vázquez, Patsilí. 2009. Feminicidio. Oficina en México del Alto comisionado de las Naciones Unidas paa los Derechos Humanos.

United Nations Entity for Gender Equality and the Empowerment of Women (UNWomen) and United Nations Human Rights Council Regional Office for Central America (OHCHR) et al. 2014. Latin American Model Protocol for the investigation of gender-related killings of women (femicide/ feminicide).

United Nations General Assembly (UNGA). 2006. In-Depth Study on All Forms of Violence against Women. Background documentation for: 61st session of the General Assembly Item 60(a) on advancement of women. Secretary-General’s study on violence against women. Forthcoming as document A/61/122/Add.1

United Nations General Assembly (UNGA). 2016. ‘Violence against women, its causes and consequences. Report of the Special Rapporteur on violence against women, its causes and consequences’. A/71/398.

United Nations Office on Drugs and Crime (UNODC). 2015. Recommendations for action against gender-related killing of killing of women and girls.

United Nations Office on Drugs and Crime (UNODC). 2018. Global Study on Homicide: Gender-related killing of women and girls.

Uprichard, Emma. 2012. ‘Dirty Data: Longitudinal Classification Systems’. The Sociological Review 59(2_suppl):93–112.

Walklate, Sandra, Kate Fitz-Gibbon, Jude McCulloch, and JaneMaree Maher. 2019. Towards a Global Femicide Index: Counting the Costs. 1st ed. Routledge.

Weil, Shalva, Consuelo Corradi, and Marceline Naudi. 2018. Femicide across Europe : Theory, Research and Prevention. Bristol: Policy Press.

Women against Violence Europe (WAVE). 2019. Wave Country Report. The Situation of Women’s Specialist Support Services in Europe.

Wooglar, Steve. 1991. Knowlege and Reflexivity. New Frontiers in the Sociology of Knowledge. SAGE Publications.