Skip to main content
SearchLoginLogin or Signup

Chapter 1 – A Short Genealogy of Feminicide and Data Activism

Published onNov 02, 2022
Chapter 1 – A Short Genealogy of Feminicide and Data Activism

"We believe that the dead women are not the stigma nor should they ever have been "the problem"; silence, self-censorship, complicity, negligence – these are what anger and shame us."

– Rohry Benítez, Adriana Candia, Patricia Cabrera, Guadalupe de la Mora, Josefina Martínez, Isabel Velázquez, y Ramona Ortíz writing in the introduction of the Silence that Breaks the Voices of All Women, 19991

We are part of the hope for an internationalized insubordination-in-action. We reject the violence that saturates daily life and are projecting this challenge into public life. A foundational element of domination imposed through violence has been disrupted: the separation of public life and the private sphere; which can also be read as the rupture between production (of commodities and capital) and the reproduction of all life.

– Verónica Gago and Raquel Gutiérrez Aguilar, Women Rising in Defense of Life, 20182

On May 11, 2015, Argentine journalist Marcela Ojeda tweeted in frustration and outrage at the news of two separate murders from different parts of the country. Suhene Carvalhaes Muñoz and Chiara Paez, both young women, were beaten to death by their partners3. Ojeda tweeted: Women actors, politicians, artists, business women, social activists . . . women, all of us. . . are we not going to raise our voices? THEY ARE KILLING US.4 The call to action circulated through her network and far beyond and led to a massive uprising in the streets of Buenos Aires and across the country. Hundreds of thousands of people turned out at the seat of Argentina's national government and in more than 120 other cities in the country in early June 20155. They chanted and raised signs and placards that said "#NiUnaMenos" – Not one less (woman). The hashtag and the protests in public spaces traveled virally across Argentina, Latin America, and the world. The following year, public demonstrations spread rapidly to Uruguay, Chile, Mexico, Perú, El Salvador, Paraguay, Bolivia, Brazil, Spain and beyond, and were increasingly linked to gender rights issues such as street harassment, the wage gap, the legalization of abortion, and machista (sexist) culture6. Recognizing murder as the culmination of many forms of gendered violence, the movement also took up the slogan #VivasNosQueremos which translates to "We Want to Stay Alive" or "We Love Being Alive."7

A sign at a protest held up by a protestor’s arm with a drawing of a girl raising her hand and  #NiUnaMenos written on top. Behind the sign is a zoomed out crowd of protestors in front of a large neoclassical building made of gray marble and lit up in purple.
Figure 1.1 #NiUnaMenos mobilization in Buenos Aires in 2016. Courtesy of Fabian Marelli and La Nacion/Argentina.

These massive mobilizations were first and foremost challenging feminicide – the murders of women and girls for reasons relating to their gender and their gender subordination in a patriarchal society. That the protests traveled across Latin America and beyond echoes the fact that the concepts of femicide and feminicide have a long history of crossing South-North borders, undergoing translation, transmutation and adaptation to local feminist contexts.

The first documented use of the term "femicide" in English occured in an 1801 book, A Satirical View of London at the Commencement of the Nineteenth Century8 and referred to the killing of a woman. Linked to gender oppression, "femicide" was first used by the South African activist and scholar Diana Russell in testimony to the International Tribunal on Crimes Against Women in Brussels, Belgium, in 1976, to assert that the homicides of women and girls took a distinctly misogynistic form9. In her subsequent co-edited book with British scholar Jill Radford in 1992, the two authors provided a more explicit definition of femicide as "the misogynist killing of women by men."10 That is to say, women's murders were related to their subordinated gender status as women and girls. Femicide constituted a crime motivated by the victim's gender as a woman and takes distinctly different forms than the murders of men11. Men are not frequently violated and killed in their homes, for example, and men's bodies are not typically desecrated in brutal and sexualized ways. Included within Radford and Russell's formulation of femicide were explicit acts of murder against women and girls such as honor killings and female infanticide, as well as more widespread but equally extreme forms of violence such as the killing of a woman by her intimate partner. Femicide as a concept seeks to reframe fatal violence against women from a "private" matter to a structural phenomenon.

This chapter provides some background on feminicide and data activism as a base for reading the rest of the book. I trace a short genealogy of the terms femicide and feminicide, as well as describe other activist, feminist, Black feminist and Indigenous framings of structural violence against women. As we will see, data activism has long been integral to challenging gender-related violence. In recent years there has been a surge in both the practices of anti-feminicide data activism as well as consideration of data activism from literature on critical data studies.

Femicide to Feminicidio to Feminicide

Figure 1.2 A poster protesting feminicides in Juárez published by family-led activist groups. Translated into English, the text reads, "Not one death more! Not one less woman in Chihuahua!" Courtesy of the Esther Chávez Cano Collection located at New Mexico State University. Graphic design by Antonio Ochoa.

Just as Radford and Russell's book about femicide was published, the North American Free Trade Agreement (NAFTA) was signed and the now infamous disappearances and murders of women in the US-Mexico border town of Ciudad Juárez started to become publicly known. From 1994 until 2001, the homicide rate of men in Juárez increased by 300 percent, while it increased 600 percent for women12. Esther Chavez Cano, resident of Juárez and an accountant by training, was one of the earliest people to document the killings in Juárez by physically clipping articles and obituaries from newspapers and saving them in physical files. She co-founded the women's rights group Grupo 8 de Marzo as well as Casa Amiga, the first women's shelter in Ciudad Juárez.

These family-centered women's rights groups – such as Grupo 8 de Marzo and Nuestras Hijas de Regreso a Casa – were consistently denied access to justice for their daughters' murders and started building on Chavez Cano's documents. They began keeping records, lists of names and newspaper clippings. They organized protests, and lobbied the press and international community to pay attention13. Various groups banded together to form a social movement called Ni una más14. The coalition took its name from a 1995 phrase by the poet and activist Susana Chávez who was from Ciudad Juárez, "Ni una mujer menos, ni una muerta más" – Not one less woman, not one more death15. You can see this slogan in action in figure 1.2, a poster published by Nuestras Hijas de Regreso a Casa protesting the feminicides in Chihuahua, the Mexican state where Ciudad Juárez is located.

While the murders in Juárez have often been depicted with sensational, "true crime" types of storytelling that like to paint such violence as the exceptional and pathological work of serial killers, feminist investigations have pointed to the structural conditions that have fomented such gender-related violence. The earliest work to try to use data to document the scope of feminicides in Juárez was the 1999 book The Silence that Breaks the Voices of All Women. As quoted in the beginning of this chapter, the writers assert that the problem is not the women themselves. The problem is the "silence, self-censorship, complicity, negligence" that permits women and girls to be murdered.16

Early academic work by Mexican scholars Marcela Lagarde y de los Rios and Julia E. Monárrez Fragoso also pointed to the intersection of the Juárez murders with neoliberal economic policy, the rise of the maquiladoras and the feminization of their labor force, the migration of rural women, the on-going presence of intimate partner violence (still accounting for at least 30% of murders in Juárez between 1993-200717), as well as the predominating culture of machismo and subordination. Writes Lagarde y de los Ríos, "Feminicidal violence flourishes under the hegemony of a patriarchal culture that legitimates despotism, authoritarianism, and the cruel, sexist – macho, misogynist, homophobic and lesbophobic – treatment reinfored by classism, racism, xenophobia and other forms of discrimination."18

Lagarde y de los Ríos, an anthropologist by training, took this issue up when she ran for a seat in Mexico's federal legislature in 2003. She was elected and then proceeded to undertake one of the most thorough government-sponsored country-level studies of feminicide to date, a 14-volume report issued in 2006 which eventually led to the codification of feminicide in Mexican law19. In the process of translating "femicide" from the English language work of Russell and Radford, Lagarde y de los Ríos also introduced a significant theoretical shift. She chose "feminicidio" as the translation to mean not only the killing of a woman for gender-related reasons but additionally to link these killings and disappearances to human rights violations and to the climate of impunity created by state inaction. She frames feminicide as a crime of the state, an assertion which has been upheld in international courts of law20. This is to say that the state was systematically failing to ensure the most basic human right for more than half of its citizens.

This formulation has laid the groundwork for drawing feminicide out of the private and interpersonal sphere of life (typified by misogynist media framings which often depict them as "crimes of passion") and directly demanding public action and public accountability for the widespread, systematic discrimination they represent21. Whether a gender-related killing happens as part of domestic violence in the home perpetrated by an intimate partner or from a sexual assault perpetrated by narcotraffickers in public space, Lagarde y de los Ríos would say that both constitute feminicidal violence22. Both crimes violate women's human rights because they are gender-related crimes; both are made possible because of state negligence and because of widespread, systematic discrimination, including gender inequality and unequal access to economic opportunity.

Lagarde y de los Ríos' contribution also laid the groundwork for more intersectional elaborations of the concept of feminicidio and for its adaptations into contexts outside of Juárez and Mexico. Countries such as Bolivia, Paraguay and Colombia adopted legislation defining and criminalizing feminicidio in the 2010s. Yet Lagarde y de los Ríos' translation of femicide and theoretical shift around feminicidio is not a settled matter. Before and during this work, other Latin American scholars such as Montserrat Sagot and Ana Carcedo had translated femicide as femicidio and undertook important investigations of violence against women in the context of Central America, and many countries passed laws codifying femicidio (see Table 1.1)23.

In the past decade, activists, journalists and academics based in the US and Canada have taken note of the work on femicide and feminicide by their Latin American counterparts and traveled some of these concepts back into the English language. For example, in 2010, Rosa-Linda Fregoso and Cynthia Bejarano published an important edited volume called Terrorizing Women: Feminicide in the Americas which summarizes some of the Latin American theoretical innovations and brings them to an English-speaking audience. Fregoso and Bejarano land on feminicide as a term to mean the murders of women and girls founded on a gender power structure but also on a power structures that intersect with racism and economic injustice. Other scholars writing in a North American context have worked to further this translation of feminicide back into English24.

The rising tide of global feminist activism sparked by #NiUnaMenos in 2015 has also helped Latin American conceptions of femicide and feminicide to travel into the English-speaking world25. While #NiUnaMenos may have appeared to some people to arise suddenly, it was built on Argentina's long history of human rights organizing, the prominence of the Abuelas and Madres de la Plaza de Mayo silently protesting their relatives who were disappeared in Argentina's "Dirty War", the rising tide of the Buen Vivir movement linking gender rights to Indigenous rights and led by Indigenous groups such as Movimiento de Mujeres Indígenas por el Buen Vivir, as well as three decades of national "encuentros" - annual conferences and organizing events originally focused on convening women and now focused on women, trans people, non-binary people and travestis26. The transnational spread throughout the Latin American region was bolstered by decades of continent-wide organizing like Encuentros Feministas de América Latina y Caribe (EFLAC), Encuentros Lésbicos-Feministas de América Latina y Caribe (ELFLAC), las Cumbres Continentales de Mujeres Indígenas de Abya Yala and more. As Maria Florencia Alcaraz writes, there was "fertile ground" consisting of many social relations – physical and digital – and relational infrastructure laid over decades27. This powerful translocal infrastructure is what Gago and Gutiérrez Aguilar are referring to in their essay Women Rising in Defense of Life, quoted in the epigraph above. They write, "we are part of the hope for an internationalized insubordination-in-action."

#NiUnaMenos was also responsible for placing calls for better data and information at the center of political debates about feminicide. One of the five main demands of #NiUnaMenos in Argentina in 2015 stated: "Create a single Official Registry of victims of violence against women. Produce official and updated statistics on femicides. We can only design effective public policies by understanding the scope of the problem."28 This demand was only partially met in the form of the National Registry of Femicide from the Argentina Justice System, created in response to movement demands in 2015, but activists criticize it for the length of time it takes to publish information and the lack of inclusion of trans women and travestis29. Moreover, the Ministry of Women has struggled to retain public servants who are charged with implementing more comprehensive data collection of gender violence30. As we will see, the most complete national registries of feminicide in Argentina, as elsewhere, continue to be produced by activists.

#SayHerName, #MMIWG2, and feminicide activism in North America

In the US and Canada, in addition to femicide and feminicide, there are other important framings of gender-related violence which have been and continue to be the subject of intense organizing and data activism. Indigenous and Black women's groups have long been organizing against fatal gender violence, as well as linking this violence to multiple forces of structural domination, including patriarchy, white supremacy, settler colonialism and economic violence.

For example, following a spate of murders of Black women and girls in Boston in 1979, the Combahee River Collective organized a march of hundreds of people to memorialize their lives, protest their deaths and challenge the lack of justice which families had received, both from the judicial system and from the media. The Combahee River Collective kept records on the women murdered and published a pamphlet for self- and community-defense (figure 1.5). Here we see an early example of "counting feminicide." In this marked-up draft with notes, the numbers of Black women killed increase from six, to seven, then eight and finally eleven. While the numerical precision is important, the authors end the pamphlet by saying that it is not only about these eleven women but also about the "1000s and 1000s of women whose names we don't even know. As Black women who are feminists we are struggling against all racist, sexist, heterosexist and class oppression. We know that we have no hopes of ending this particular crisis and violence against women in our community until we identify all of its causes, including sexual oppression."31 Here, the Combahee River Collective insisted on an intersectional approach to understanding the root causes of gender-related violence. They refused a single-axis analysis that would attribute the violence in their community to either sexism or racism32.

Figure 1.3 Combahee River Collective pamphlet, “Eleven Black Women: Why Did They Die?” (1979). The Collective circulated more than 18,000 copies of the pamphlet in both Spanish and English. Courtesy of the Barbara Smith Collection at the Lesbian Herstory Archives in Brooklyn, New York. Pamphlet originally published by ​​Red Sun Press.

This early organizing in support of Black women finds an echo in contemporary campaigns like #SayHerName which seeks to uplift the stories of Black women and gender non-conforming people who are killed in police violence in the United States. Their lives and deaths are often invisibilized by the larger narrative around anti-Black police killings, which focuses on men, as well as pervasive underreporting in the media due to longstanding racism that constructs Black women victims as less worthy33.

In her work on Black femicide and the Black Lives Matter movement, political theorist Shatema Threadcraft traces the #SayHerName campaign back to "the United States’ most iconic necropolitical warrior, one Ida B. Wells."34 In the Red Record, published in 1895, Wells collected reports of Black lynchings published in the white press to systematically present the scope and scale of these brutal racialized killings and to challenge the white narratives that circulated about them35. Here necropolitics means mobilizing the politics of death, especially as they intersect with state power – who is targeted for death, who kills, and, most importantly, what their deaths mean36. This is a concept from postcolonial scholar Achille Mbembe. He was writing to counter (or maybe complement) Michel Foucault's concept of biopower, the idea that life chances are unevenly distributed by the state across social groups. Instead of life chances, Mbembe focuses more precisely on how death chances are distributed by the state – on how people from subordinated groups are marked for death and the ways in which their deaths are normalized. Necropolitics, then, involves not only the empirical fact of the disproportionate deaths of specific groups but also the performative and discursive element of interpretation around what those disproportionate deaths mean. Another way to think about necropolitics is to ask the question, "In a particular society, whose deaths matter?" and following Lorena Fuentes, what are "the discourses that visibilise, differentiate, and/or obscure the bodies of victims"?37

While Threadcraft describes the Black Lives Matter movement's necropolitical achievements in reframing Black police killings as unjust, she outlines tensions for #SayHerName to be able to do the same for Black women: "Activists concerned with stemming black femicide should reflect on the fact that the movement has relied on amplifying the spectacle of death in a context in which black women suffer from a severe spectacular violent death deficit. Femicide does not lend itself to being captured in the same way…what can stand in the place of this public spectacle, when what they are dealing with are a greater number of wholly private murders. What, then, will motivate people to rally around the bodies of our black female dead?"

Even studying the phenomenon is challenging because of missing data, underreporting due to media bias, and because Black women are killed in less public and therefore less visible ways than Black men38. As Crenshaw and Ritchie write, "there is currently no accurate data collection on police killings nationwide, no readily available database compiling a complete list of Black women’s lives lost at the hands of police, and no data collection on sexual or other forms of gender- and sexuality- based police violence."39 In response, advocates like Black Femicide US, Black Girl Tragic, the African American Policy Forum and Ritchie herself have created important databases of Black women and gender non-conforming victims of violence that they use in advocacy, support for families, consciousness-raising and movement building40.

Missing and murdered Indigenous women, girls and two spirit people – #MMIWG2

During the same time period that women's groups were bringing the feminicides in Ciudad Juárez to public attention, activism led by Indigenous women was surging in Downtown Eastside Vancouver, the unceded territories of the xʷməθkʷəy̓əm (Musqueam), Sḵwx̱wú7mesh (Squamish), and Sel̓íl̓witulh (Tsleil-Waututh) Nations in present-day Canada. On February 14th – Valentine's Day – 1991, a small number of people gathered to mourn the death of a beloved Coast Salish woman who had been sexually assaulted and murdered 41. Her mother, Linda Ann Joe, and other family members staged a small memorial to her life. This gathering became annual and has grown to encompass many thousands of participants in Vancouver as well as spread to more than 20 cities in Canada and the US42. It is known as the Women's Memorial March and its annual theme is: "Their Spirits Live Within Us."43 Each year, the march moves slowly through the Downtown Eastside neighborhood, and makes stops at sites where Indigenous women have last been seen or where they were killed. The march honors individuals, and family members often speak. But since the beginning, it has also challenged the structural conditions and cultural representations that produce such violence. This has involved keeping a count and publishing a list of disappeared and murdered women in the Downtown Eastside area. More than 970 names have been added to the list since the march started, and 75 names were added just in the year 2019 44. A pamphlet passed out at the 2001 Women's Memorial March read:


The Women's Memorial March, along with many, many other Indigenous-led efforts, laid the groundwork for the growing movement across North America now known by its hashtags – #MMIW, #MMIWG2, #MMIP46– and by its slogans – "No More Stolen Sisters." Indigenous organizers and scholars have been unequivocal in linking this violence to settler colonialism, and indeed in challenging the gender binary itself as a settler colonial construction of gender – a toxic ideological import from Europe – that has been wielded as a tool to disrupt Indigenous family relations and plunder Native communities, for generations47.

Figure 1.4 The Women's Memorial March, pictured here in 2017, has been held every year on Feb 14th since 1992 and spread to more than 20 cities across the US and Canada. Courtesy of Photo by Dan Toulgoet.

Contemporary Indigenous feminist scholars like Sarah Deer build on this legacy of linking gender-related violence to colonial violence and speak of the paths to multiple forms of sovereignty – land sovereignty, body sovereignty and soul sovereignty: "It is impossible to have a truly self-determining nation when its members have been denied self-determination over their own bodies."48 In parallel, Latin American Indigenous leaders have led the development of the concept of cuerpo-territorio. This is an analysis from Lorena Cabnal, Mayan Q’eqchi’-xinka healer, and other Indigenous feminists and community defenders, that links violence on the body (cuerpo) to violence on the land (territorio), because so many Indigenous women are killed in defending their land or protecting their water from destruction by man-camps, extractivist industry and toxic public-private partnerships49. These scholars and activists demonstrate how the gender-essentialist, single-axis frame of Russell and Radford of femicide as "women killed by men" is not expansive enough, nor precise enough, nor historical enough to appropriately describe this violence.

Much of what is demonstrated in scholarship is corroborated by reports from national inquiry commissions and advocacy groups, such as the prevalence of gender-related violence against Indigenous women at sites of resource extraction50. Still, the crisis persists and official data about MMIWG2 is widely known to be missing, poor quality, fragmented, misclassified or purposefully shielded from public view by the state. This has led think tanks such as the Urban Indian Health Institute to assert that there is also a "data crisis" and to a growing number of activist and civil society efforts to collect the data51. Annita Lucchesi, scholar and Executive Director of the Sovereign Bodies Institute, has stated that negligent and discriminatory institutional practices allow Native women "to disappear not once, but three times—in life, in the media, and in the data."52

Key terms

I have narrated a short genealogy of some key terms that describe fatal gender-related violence against women: femicide, feminicide, #MMIWG2, #SayHerName, Black women killed in police violence. All of these are insurgent necropolitical concepts aiming to reframe fatal violence against women, to move it out of the private sphere and place it into a structural, systemic, and public context. All of these resistant knowledge projects use data and data activism as one tactic in their larger struggles to prevent and eliminate gender-related, economic, territorial and racialized violence against women.

In this book, I will use "feminicide" most of the time for two reasons. First, as I stated in the introduction, the center of gravity of this book is anti-feminicide data activism in Latin America. Feminicide is the framing concept that the majority of the data activists and civil society groups that we interviewed and worked with use. Second, as elaborated by Largarde y de los Ríos and built on by Latin American feminist theorists as well as transborder scholarship like that of Fregoso & Bejarano's, it is a concept that, in contrast to femicide, can hold more intersectional consideration of how and why violent killing of women happens and persists. Feminist scholars and writers have theorized how feminicide is not only produced from patriarchal societies that oppress women, but also how this intersects with economic violence; with neoliberal capitalist regimes of production; with corporate investments in land dispossession and extraction; with the white supremacy, settler colonialism and racism that accompany these economic imperatives; with the complicity of the state in ignoring and neglecting such violence; with slow and everyday violence; and with machista culture. But my choice to use "feminicide" is not to castigate those who use "femicide" or any other framing of gender-related killing as politically incorrect. In Latin American feminist circles there are on-going debates about femicide vs feminicide, just as there are Latin American formulations of femicide that depart from the early gender essentialism of Russell and Radford53. These are open questions, concepts and discussions, not fully settled matters.

"Women" in this book is a political category that includes cisgender and transgender women. This is true for the vast majority of activists that we interviewed who produce data about feminicide. When I speak about gender-related violence that includes genders other than women, such as non-binary people, trans men, two spirit people, travestis, or others, I will use "feminicide and gender-related killing" or else speak about "trans killings" if the violence is directed specifically at transgender people. My intention is not to impose "feminicide" as a label on activists, nor to assert that all activists should be using the term. Grassroots groups and individuals know best what framing concepts to use for their work and, more often than not, data activists are themselves survivors of gender-related violence. When I speak about the data activism of a specific group, I will name the violence as they name it and draw from their framing concepts and motivations to describe the work.

Relatedly, you may have also noticed that I am using the term "gender-related violence" instead of the more common term "gender-based violence." This decision comes out of discussions with my collaborators on the Data Against Feminicide project. We decided to use this term to indicate that phenomena such as femicide, feminicide, MMIWG2, police violence against Black women, and LGBTQ+ killings are intertwined with gender oppression but cannot be solely explained by sexism, cissexism, homophobia, patriarchy and unequal gender relations, and their analysis requires an intersectional perspective. It is a term that is also favored by Rashida Manjoo, former UN Special Rapporteur on Violence Against Women54.

Laws and official data about feminicide

Feminicide is gathering energy in national and international law, policy frameworks, and governance. It has increasingly featured in public and policy debates, especially in Latin American countries. In the past two decades, public pressure from feminist- and women-led movements has led to the passage of legislation that criminalizes feminicide or femicide in 18 countries and one territory (see table 1.1)55.


Femicide/ Feminicide


Covers more than intimate partner violence?

Includes trans women?

Government publishes official data in some form (aggregated statistics, reports, etc)































Costa Rica






Dominican Republic












El Salvador















































Puerto Rico


















Countries in the Americas without laws about femicide or feminicide include the US and Canada, the South and Central American nations of Belize, Guyana and Suriname, and the Caribbean nations of Cuba, Haiti, Jamaica, Trinidad and Tobago, Bahamas, Barbados, Saint Lucia, Grenada, Antigua and Barbuda, Dominica, Saint Kitts and Nevis, St. Vincent and Grenadines, Antigua and Barbuda. Aside from Puerto Rico, no territories in the Americas have separate laws regarding femicide or feminicide.

Table 1.1 List of countries in the Americas that have laws about femicide or feminicide and its scope. The LAC region legislation database is available on the United Nations Gender Equality Observatory for Latin America and Caribbean websites.

Despite extremely high rates of female homicide, the US, Canada and most countries in the Caribbean do not have laws defining femicide or feminicide though they have begun to strengthen legal protections for related phenomena like intimate partner violence and MMIWG107. In the US, the most expansive federal law is the Violence Against Women Act (VAWA) which outlines provisions for services and programs to prevent various forms of gender-related violence but stops short of defining feminicide as a separate crime or the murder of women as a violation of their civil rights108. In fact this is one of the main challenges of legal reform: writing legislation that recognizes, defines and provides redress for feminicide as a structural, public problem rather than an interpersonal, private problem; a violation of civil rights and human rights, not (only) a personal dispute or "domestic" matter.

Many of these legal reform efforts include mandates for improved collection of data and publication of statistics. Often these are due to demands from advocates specifically requesting such provisions, as we saw in the case of #NiUnaMenos in Argentina in 2015 where one of the movement's five demands was the creation of a single, centralized registry for feminicide. For example, a 2021 law passed in Puerto Rico ordered the systematic collection of data about feminicide and transfeminicide stating: "If there are no reliable and comparable data collection mechanisms for a certain type of crime, there will be no appropriate ways to understand it nor effective strategies to combat it."109 All recent laws regarding Missing and Murdered Indigenous People in the US have provided extensive provisions for improving data collection and statistics following demands from families and tribal governments110.

These represent attempts to reform law and policy around data collection. But policy mandates get complicated when they run up against funding challenges like the neoliberal austerity measures increasingly being implemented in Latin America; or bureaucracy challenges like the fragmentation of agencies, each addressing a part of the issue or collecting part of the data with no agency fully responsible. Research by the Latin American Initiative for Open Data (ILDA) in Argentina that explored government data collection around feminicide corroborates this, "The strong fragmentation of this space in institutional, legal and occasionally political terms, makes it difficult to coordinate who is responsible for reporting and in what way on this particular type of crime." 111. Moreover, there is almost universally a lack of training and knowledge about the gendered nature of violence for public sector employees who handle and classify cases on the ground, such as police and medical examiners. MMIWG reports highlight the persistent racial misclassification of Indigenous women by the state112. Moreover, there is the added informational complexity of what needs to be known in order to determine whether a murder was motivated by gender, namely: the relationship between victim and perpetrator, the motive for the murder, and the types of prior violence that the person may have suffered113. This adds a significant time and resource burden per case, which has led some to suggest shifting the burden of proof such that the violent death of a woman is considered by default to be motivated by gender and the burden of proof lies on officials to prove otherwise. For example, in August 2020, Alejandro Gertz Manero, Attorney General of Mexico, recommended that all murders of women be investigated as feminicides114. And finally, there is the distance between what is written in law and what actually happens on the ground. Sociologist Mariana Mora studies feminicide in Costa Rica and states that the official party line is that state investigators should assume that a killing of a woman is a feminicide and later rule it out. She elaborates, "This was told to me several times by state representatives. But in practice we know that this does not happen."115

Figure 1.5 ILDA's flow chart to identify femicides. They propose that organizations collecting data undertake a quarterly review and an annual review of all cases, and then unify the results. Courtesy of ILDA. Graphic design by DataSketch. Graphic adaptation for this book: Wonyoung So.

As nations grapple with these difficulties, regional and international efforts have stepped in to try to help provide both data standardization and technical guidance to governments (and, in some cases, such as Mexico, to provide international accountability in the face of government inaction116). For the past four decades, transnational networks of feminist groups, international NGOs and supranational agencies like the World Health Organization (WHO) and United Nations (UN) have held convenings and released reports on the topic of femicide/feminicide117. These publications have framed the topic as a global issue of gender inequality and invariably highlight the lack of reliable, comparable data. Surveying this work, Sandra Walklate and colleagues have posited the potential value and challenges of working towards a "global femicide index", a set of standard methods to count, map and measure the prevalence of femicide across countries118. In 2015, Dubravka Šimonović, UN Special Rapporteur on Violence against Women, called for establishing a "femicide watch" – an observatory effort to count, collect and monitor data about gender-related killings – in every country 119.

In 2014, amidst this intense and growing international coordination on feminicide, the UN published an important technical guidance document about feminicide. Titled the Latin American Model Protocol for the investigation of gender-related killings of women (femicide/feminicide) – and from here on out I'll just call it "the model protocol" – it outlines a baseline set of definitions, data fields and procedures for official government data collection about feminicide for use by, among others, police, medical examiners, forensic experts, public sector data analysts and social workers. It is called the "Latin American" model protocol because the UN took a regional consultation strategy for its development and primarily worked with professionals in the region.

Some key contributions of this document include a working definition of femicide/feminicide as gender-related killings that are rooted in the structural subordination of women. It also elaborates two key categories of femicide. Active or direct femicides include intimate partner femicides, misogynist murders, so-called "honor"or dowry-related killings, female infanticide, hate crimes against lesbians, and more. Passive or indirect femicides include deaths from unsafe abortions, maternal mortality, deaths linked to organized crime, and deaths linked to negligence by public servants. In the latter category, the state is implicated due to its role in creating an unsafe environment for women which diminishes their life chances, reduces their reproductive autonomy and creates a climate of impunity where women are disproportionately killed and justice is not served. The model protocol also outlines a typology of fifteen kinds of femicide that the authors found to be common in the Latin American experience. These include intimate- and non-intimate femicide; racist, transphobic and lesbophobic femicide; child femicide; femicide due to stigmatized occupation (e.g. sex workers); and more. The definitions, categories and typology of femicide in the model protocol have been influential for federal governments and also, as we will see in Chapter 5: Recording, for data activists as well. But coming in at 200 pages of technical documentation, the protocol is dense and there are numerous institutional barriers to governments just jumping in and adopting it.

This led ILDA to initiate a project to understand how well the UN's protocol met the needs of its intended users: the professionals that work with official data about feminicide and gender-related killing. After exploratory workshops and consultations with public sector officials, ILDA released a 15-page guide that offers a streamlined version of the model protocol but in simpler form120. The ILDA guide has flow charts for data collection and case registration processes, and a series of 66 variables recommended to be collected about each case. Included is key information to be able to ascertain whether a murder constitutes feminicide, such as relationship between perpetrator and victim, whether there was sexual aggression, and whether there were prior complaints filed against the perpetrator. ILDA has piloted their guide with public sector employees in Argentina and Uruguay, and later expanded to Ecuador, Honduras, Jamaica, Panama and Paraguay. As ILDA works, they publish updates to their guide and recommended registration processes, such as the flow chart in figure 1.5 that serves as a visual tool to help identify a crime as feminicide and advises adhering to a quarterly and annual schedule for reviewing cases. Shifting processes and protocols in one country – not to mention a whole region – is slow work but Silvana is cautiously optimistic, "We’re seeing some governments taking what we put out as an ideal structure and comparing to what extent they have that in place or not, which is a good sign that at least it’s being taken as a reference. We know that what we’re suggesting is an ideal."121

Yet despite legal advances, technical guidance and persistent political pressure on feminicide, governments are still unable or unwilling to produce reliable public data about the phenomenon. As you can see in Table 1.1, even those countries with legislation on femicide or feminicide fail to publish open, disaggregated data about pending and closed cases. In Mexico, sociologist Julia Monárrez Fragoso , who documented 442 cases of women and girls killed in Ciudad Juárez between 1993-2005, notes that “national statistics do not document the reason for the murder, the relationship between victim and victimizer, nor the various types of violences that the women suffered prior to being murdered. […] In the face of such absences, it is necessary to find alternative means to understand femicides with greater precision.”122 And one of the main findings in Canada's multi-year MMIW national inquiry is that "There is no reliable estimate of the numbers of missing and murdered Indigenous women, girls, and 2SLGBTQQIA persons in Canada."123

Data activism and feminicide

Indeed the vast majority of scholarly works, commissions and advocacy reports about feminicide lead with the fact that data about this violence are either completely absent or else sparse, unreliable, unavailable, untimely, neglected, misclassified, or not public. This is a phenomenon called missing data that we will explore further in the next chapter. An increasing number of grassroots activists and civil society groups have turned to the production and circulation of counterdata not only to fill in missing data, but also to challenge state bias and inaction, to galvanize media and public attention, and to help heal wounded communities.

This work can be best characterized as data activism, a concept which foregrounds the use of data and software to pursue collective action and exercise political agency124. Stefania Milan characterizes data activism as an emerging social movement tactic and draws parallels between data activism and media activism. In both cases, citizens and people are taking advantage of widely available digital tools of production to "uncover stories of injustice or change."125 In Milan's framework, co-developed with scholars Lonneke van der Velden and Miren Gutiérrez, there are two types of data activism: reactive and proactive126. Reactive data activism takes datafication and massive data collection as its object of struggle, e.g. people who resist mass surveillance by developing encryption tools. Proactive data activism, on the other hand, may have any number of objects of struggle. What proactive data activists share is that they "create, mobilize, solicit, appropriate, or crunch data in view of supporting alternative narratives of the social reality, questioning the truthfulness of other representations, denouncing injustice and advocating for change."127 Grassroots counterdata production about feminicide falls in the latter category and undertakes all of these actions with data, with a particular emphasis on the act of creating datasets and databases that document cases of feminicide. These datasets are often manually crafted (row-by-row, case-by-case, person-by-person) and constitute counterdata, data production that happens outside of – and often in opposition to – mainstream counting institutions like governments and corporations.

Indeed, prior work has shown how one of the key goals of data activism is to challenge whether and how institutions see, count, measure and evaluate particular phenomena. The 2016 report Changing What Counts describes numerous case studies of citizen data action that attempt to shift official institutional measurement of various things: police killings, air pollution, water access, and government pardons128. Civil society actors in these cases produce their own data as well as collate and combine data from diverse sources in order to challenge public sector numbers and also to challenge public sector narratives around these phenomena. That is to say that datasets operate not as a bucket of disembodied facts but are in fact mobilized by activists as “a vector for the circulation of affective and emotional bonds.”129 Pine and Liboiron have discussed data activism as the deployment of charismatic data, data whose dramatic or spectacular nature impels different stakeholders to take action towards social change130. This means that datasets are – or can be when deployed strategically – acts of rhetorical and political communication. The act of registering diverse instances of gender-related killings into a dataset about feminicide is both an act of counting and an act of classification and a rhetorical act to assert the concept of feminicide as a valid and valuable thing to count. Sociologists Martin and Lynch frame this as numeropolitics, and discuss how acts of counting are also always acts of classification. To count feminicide is also to select events from the world and to classify them. It is to assert them as feminicide and also to assert that feminicide matters as a concept. As Pine and Liboiron state succinctly, "measurements make things."131

Recent work has examined the intersection of data activism and grassroots data practices around feminicide. In Terrorizing Women, Fregoso and Bejarano place feminicide counterdata production practices in Latin America in a long history of Observatorios Comunitarios (Citizen Observatories), which have served to engage communities in monitoring state agencies and using collective action strategies to hold them accountable132. Helena Suárez Val – interdisciplinary scholar, activist and co-lead on the Data Against Feminicide project – has compared activist and government data schemas about feminicide and showed how each presents a different data frame, bringing some factors about the woman, the perpetrator and the event into the frame, and leaving others out133. This makes some political actions possible and forecloses others. And in studying how feminicide data circulate on social media, Suárez Val has asserted them as affect amplifiers, digital cartographies that seek to translate feminist grief and rage into public action134. This resonates with Annita Lucchesi's scholarship around using an Indigenous decolonial approach to mapping MMIWG2, in which maps and stories are used in the service of "resilience and resurgence and not just of loss."135

Other work has sought to link anti-feminicide data activism to scholarly conversations around critical data studies. In a 2019 case study, Jean-Marie Chenou and Carolina Cepeda-Másmela narrated the demands of #NiUnaMenos for a national registry of feminicide in Argentina. For these scholars, the activist demands for data constitute both an appropriation of hegemonic technology as well as the production of alternative imaginaries around big data – namely, that data could and should be used in the service of gender justice136. This work emerged as conversations in the field are growing around decolonizing data. Movements around Indigenous data sovereignty emphasize that colonization – in the form of settler colonialism – has not ended and that decolonizing data must be in the service of Indigenous communities and cultures and lands137. Milan, Treré, Dutta and other scholars in critical data studies have called for "de-Westernizing" the field and situating many of the harmful effects of contemporary datafication as a continuation of Western European colonization and their (our) extractivist, violent knowledge regimes138. While numerous important concepts have been advanced for naming these regimes – among them, surveillance capitalism, the New Jim Code, automated inequality, data extractivism, data colonialism – critical data studies has the potential to do more than retroactively describe hegemonic data science practices. It has been and remains a generative site for developing counter-hegemonic approaches to data, knowledge and power.

In her 2019 paper, Data Epistemologies, The Coloniality of Power, and Resistance, media scholar Paola Ricaurte analyzes how present regimes of data power are violent. They are colonial in their methods for capturing and extracting value from human life. They pose a threat to humans, biodiversity, climate and life on Earth itself. Yet Ricaurte leaves open the possibility that, through collective resistance, “we can reverse extractive technologies and dominant data epistemologies in favor of social justice, the defense of human rights and the rights of nature.”139 She discusses María Salguero's map of feminicides in Mexico as an example of such resistance, enacting the use of grassroots data to further justice, memory and human rights in the face of economic and patriarchal violence. For Ricaurte, feminicide data activism represents a step towards the development of "alternative data frameworks and epistemologies that are respectful of populations, cultural diversity, and environments.”140 Such alternative epistemological approaches are flourishing in both scholarship and activism, including feminist data refusal, radical data science, decolonial AI, Indigenous data sovereignty, queer data, and more141.

Lauren Klein and I's own work on data feminism can be situated here. We highlight ways that hegemonic epistemologies get embedded into data science and develop seven principles for working with data science drawn from intersectional feminist theory and action (see the toolkit in Chapter 8 for a list of these principles). In our book, we describe María's map and counterdata production work as an example of the principle challenge power – appropriating hegemonic data science and mapping tools to visibilize violence that is systematically invisibilized by the state and its institutions142. As the Data Against Feminicide project has evolved – as our research team has conducted interviews with data activists and as we have come into community with practitioners – I see more resonances with data feminism's principles, along with some dissonances as well. Almost all of the work to source, verify and deliberate around counterdata is collective (embrace pluralism). Emotion, memory and care are often at the center of the work (elevate emotion and embodiment). While activists may draw from international data standards and typologies, they tune their data schema and categories to meet their local context (consider context), often producing significant theoretical innovations in the process (examine power). Yet activists do not always foreground their labor when they circulate their data (make labor visible), often preferring to leverage mainstream society's perception of data and numbers as being neutral. As I stated in the Introduction, throughout the book I will seek to draw out these resonances, as well as tensions or gaps that arise and may point to aspects of feminist praxis with data that data feminism did not consider.

All of this is to make the case that the real-world, already existing practices of data activists have much to offer work in critical data studies that seeks to support and sustain alternative, feminist, Indigenous, queer and decolonial epistemologies. In Chapter 8 – A Toolkit for Counterdata Science, I include short descriptions of these diverse data epistemologies, all of which are challenging hegemonic data science in important ways from different standpoints. Activist data practices also have much to offer practitioners – journalists, librarians, advocacy organizations, nonprofits, planners, community organizers – who seek to use data science in the service of social justice and liberation. This case is aligned with Milan's recent paper outlining what critical data studies as a field may learn from data activism as it was practiced during the COVID-19 pandemic143. Grassroots data activism about feminicide invites us to imagine a data science, epistemology and ethics that rigorously takes power and people into account; that understands how structural inequality produces missing and flawed data and develops creative counterdata strategies to mitigate that; that views data science not as a technosolutionist panacea but as, first, an intimate act of care, witnessing and memory justice; and second, as a vector for social change. This is not to romanticize the very difficult and fraught labor of anti-feminicide data activists, which we will be exploring in more detail in the rest of this book. Rather, it is to simply to offer that mainstream data practitioners and critical data studies scholars have much to learn from these reflective practices, particularly those of us who wish to mobilize data science in solidarity with movements for social justice.

Alessandra Jungs de Almeida:

I think it appeared in the book before as “Not one [woman] less”.