Femicide and Feminicide in Mexico: Patterns and Trends in Indigenous and Non-Indigenous Regions

Abstract

The killing of women in Mexico has attracted both national and international attention. Many of these homicides are regarded as feminicides, which are defined as the misogynistic killing of women for reasons of gender rooted in ideological and structural gender inequalities. This study examines changes and continuities in female homicides and femicides from 2001 to 2017 in indigenous and non-indigenous municipalities. Female homicides have increased at a higher rate than femicides, but the latter has increased at a higher rate in indigenous municipalities than in non-indigenous municipalities. This is associated with an increase in gender equality in a context of conflict and structural discrimination against indigenous peoples. Implications for future research and public interventions are discussed.

Keywords

femicide ethnicity measuring crime international issues homicide

Introduction

Since Felipe Calderón’s inauguration as president of Mexico in 2006, national and international news have been replete with stories of brutal murders happening in the country. However, by the end of the 1990s, the killing of females for gender reasons—also called feminicide—in Chihuahua had already captured both national and international attention (Monárrez Fragoso, 2002). Femicide is the most extreme expression of gender violence committed against women. The term was coined by Diana H. Russell in 1976 at the first International Tribunal on Crimes against Women and defined as “the killing of females by males because they are female.” Femicide was a term already used in the Anglo-Saxon lexicon, but “Russell added critical political meaning to it and placed it within a broader feminist politics framework” (Grzyb et al., 2018, p. 20). Afterward, Radford and Russell (1992) refined the definition to the misogynistic killing of women by men motivated by hatred, contempt, pleasure, or a sense of ownership over women. It is a phenomenon that needs to be investigated in “the context of the overall oppression of women in a patriarchal society” (Radford, 1992, cited in Corradi et al., 2016, p. 3). Since the 1990s, the definition of femicide has evolved to “the killings of females by males because they are females”¹ and “the killing of a female because she is a female.” This aimed at capturing the fact that some women might be killed by other women.

In Mexico, Lagarde (2006) borrowed the term “femicide” and transformed it into “feminicide.” In Spanish, femicide (femicidio) can be understood as the feminine of homicide (homicidio). To emphasize that many female homicides are perpetrated for gender reasons, Lagarde (2006, 2010) translated the term as “feminicide” and expanded Russell’s definition to include the impunity of these crimes and the neglectful manner in which the State addresses them.

Feminicidal violence is the extreme, the culmination of many forms of gender violence against women that represent an attack on their human rights and that lead them to various forms of violent death. In many cases, these forms of gender violence are tolerated by society and the State; at other times, citizens live feminicidal violence with powerlessness, for there are few channels available for the enforcement of rights (Lagarde, 2010, p. xxi)

Russell claims that the inclusion of the term “impunity” is problematic because “a sound definition must separate the phenomenon being defined from the response to it.” According to Lagarde’s definition, if the killing of a woman for gender reasons is punished, it would not be considered “feminicide.” Nevertheless, I use the term feminicide because it is broadly used in the Latin-American region (Toledo Vázquez, 2009).

In Mexico, feminicide has been addressed from different perspectives as a social and public health problem. Qualitative research on feminicide is centered on either the analysis of criminal files (Arteaga Botello & Valdés Figueroa, 2010), newspapers (Alcocer Perulero, 2014; Martin & Carvajal, 2016), or both sources combined with ethnographic data that offer an in-depth understanding of the phenomenon (Aguilar Gutiérrez & Aguilar Hernández, 2019; Luna Blanco, 2019; Monárrez Fragoso, 2000). Quantitative research uses either administrative records based on law enforcement cases investigated as feminicides or mortality records, which tend to consider all killings of women as feminicides (Bejarano Celaya, 2014; Incháustegui, 2014; SEGOB et al., 2016; Valdivia & Castro, 2013). These quantitative studies fail to reveal the true scope of the phenomenon because not all feminicides are investigated as such (Luna Blanco, 2019; Quintana Osuna, 2018), and not all female murders are perpetrated for gender reasons. Expanding on this, Torreblanca and Merino (2017) explain that, based on existent data, differentiating between female homicides and feminicides is complicated because available information on mortality records regarding contextual data (i.e., perpetrator and victim characteristics) tend to be insufficient in the identification of femicides. According to their estimates (in which they take into consideration where the body was found, the means used to commit the homicide, and the relationship with the perpetrator), 34% of female homicides can be labeled as feminicides, during the period 2004 to 2016 (n = 8,913).

An additional problem in the current research on feminicides is that it tends to essentialize the category “female” because it does not differentiate among those that might face additional vulnerabilities such as age and race/ethnicity (Figueroa Romero, 2019). In Mexico, according to the 2015 Intercensal Survey, 21.5% of the Mexican population, according to their traditions, culture, and shared history label themselves as indigenous. This percentage is clearly distant of the 6.5% (7.4 million) of the population that speak one of the 68 indigenous languages spoken in Mexico (Vázquez Sandrin, 2014). The indigenous population is mostly rural, experience poverty, discrimination, and social gaps in education, health, and service provision (Gutiérrez Chong & Valdés González, 2015; Roldán et al., 2017; Solís, 2017). Indigenous women endure even more social disadvantages than indigenous men (Rizo Amézquita, 2017). They are also, compared to non-indigenous women, more likely to experience gender-based violence such as partner violence (Casique, 2021) as well as some expressions of sexual violence during childhood and adolescence (Frías, 2021).

Current administrative records and official statistics seldom disaggregate homicide or feminicide data by race or ethnicity (Figueroa Romero, 2019; Ríos, 2021).² In addition, indigenous communities tend to be located far from main cities—wherein 21 feminicide specialized State Attorney offices are located—and have restricted access to medical services, which can obstruct the identification of feminicides as official data ignore cases registered by traditional authorities (Figueroa Romero, 2019). This has resulted in the invisibility of feminicide of indigenous women in this regard.

With the aim of overcoming some of the limitations of the current research noted above, this study has three main objectives. First, to examine female homicides by differentiating between those linked to gender reasons (feminicides) and those associated with other circumstances during the period 2001 to 2017 (2017 is the most recent year with microdata available).³ This will provide us with the data needed to empirically test the claims made by numerous Mexican NGOs regarding increases in the rate of feminicide over time. Our second aim is to assess the relationship between gender inequality, feminicide, and female homicides in Mexico. The third objective aims to examine feminicides and female homicides trends in indigenous and non-indigenous municipalities and to describe differences in homicides and feminicides in terms of victims’ sociodemographic characteristics and means employed to commit the crime.

Conceptualizing Feminicide in the Mexican Context

Conceptually, feminicides are inextricably linked to gender reasons. Gender-based violence is “rooted in the hegemonic masculinity model embodied mostly by men—but also by some women—exerted against women, and also against men who follow non-hegemonic masculinity models, both heterosexual and homosexual” (Oliver & Valls, 2004, p. 16). The term “gender-based” aims to differentiate between acts of violence rooted or shaped by differential gender roles and statuses that men and women carry in society (such as forced sterilization and partner violence) and acts of violence not associated with the social construction of gender differences (see Russo & Pirlott, 2006). Although women are most often the victims of gender-based violence, males might experience it as well; likewise, both males and females might be perpetrators of gender-based violence, including feminicides. This has been an issue of extensive debate because early conceptualizations of the phenomenon characterized femicide aggressors as solely male (Caputi & Russell, 1992).

As shown by Toledo Vázquez’s (2009) extensive review, the reasons of gender in feminicide are very diverse. The Latin American Protocol for Investigating Violent Deaths of Women for Gender Reasons (OACNUDH & ONUMUJERES, 2014) provides a comprehensive classification of feminicides: (a) Intimate feminicide; (b) Non-intimate feminicide, which is the death of a woman by either an unknown person or someone known by the victim that does not a have an intimate relationship with her (e.g., a neighbor); (c) Child feminicide, the death of a girl under 14 years of age committed by a male in the context of a relationship of power, responsibility, or trust; (d) Family feminicide, which the death of a woman in the context of an adoption, consanguinity, or affinity; (e) Feminicide by connection, which is when the death of a woman occurs in the same place where a male kills or attempts to kill another female friend, mother, daughter, etc.; (f) Sexual systemic feminicide, associated with the death of women previously kidnapped, tortured, and/or raped—it is disorganized if the woman is killed in a certain period of time and organized if the perpetrators act in a network of sexual feminicides; (g) Feminicide associated with sexual exploitation and prostitution; (h) Feminicide associated with female trafficking and sexual exploitation; (i) Transphobic feminicide, associated with deaths linked to the victim’s transgender or transsexual identity; (j) Lesbophobic feminicide, which is the death of a woman due to her sexual orientation, rejection, or hatred; (k) Racist feminicide, associated with the death of a women due to her phenotype or hatred of her racial/ethnic origin; and (l) Feminicide linked to genital mutilation, which is the death of a girl or woman due to genital mutilation (other classifications in Arteaga Botello & Valdés Figueroa, 2010; Monárrez Fragoso, 2000; Toledo Vázquez, 2009).

It is important to note that in the Latin American region, the concept of feminicide entails the subjective intention of killing a woman (i.e., through criminal behavior). It does not include the deaths of women resulting from gender discriminatory practices that are not crimes, such as preventable maternal deaths, gender-selective undernourishment, forced sterilization, and genital mutilation (Caputi & Russell, 1992). These are labeled as “passive or indirect feminicides” in the Latin American Protocol (OACNUDH & ONUMUJERES, 2014). Most studies, even those by Caputi and Russell (1992), refer to feminicides when the fact entails a crime –homicide–. It is clear, therefore, that identifying feminicide is a difficult task; the lines between the various classifications are somewhat blurry. In the indigenous regions of Oaxaca and Chiapas in Mexico, maternal mortality during pregnancy and childbirth has been extensively documented. Some of these deaths are the result of men’s control over female bodies because they and their families do not allow females to receive prenatal care (Freyermuth, 2003; Sesia, 2016).

Structural gender inequalities interact with sociopolitical conflicts in indigenous regions. In Mexico there is evidence of the murders of indigenous women associated with the defense of communal properties, conflict among indigenous communities, and violent acts of mass killing due to the militarization of rural areas and the presence of paramilitary groups (Castañeda Salgado et al., 2013; De Marinis, 2017; Hernández Castillo, 2019; Valladares de la Cruz, 2018). These murders are seldom labeled as feminicides “because there may be other motives involved that do not necessarily fit the specific legal definition” (Figueroa Romero et al., 2017, p. 11). Hernández Castillo (2019, p. 640) in regard to sexual violence against indigenous women claims that, “sexual violence is a message in patriarchal semantics to promote demobilization, and eventually, displacement and dispossession.” I argue that feminicide as the most extreme form of gender violence is also instrumental to these ultimate goals.

According to the Mexican Federal Criminal Code, feminicide involves the killing of a woman for gender reasons (Art. 325). It is noteworthy that this conceptualization includes not only homicides perpetrated by intimate partners and family members—the most common perpetrators of feminicide (Dawson & Gartner, 1998; Toprak & Ersoy, 2017)—but also other types of perpetrators such as strangers, acquaintances, and friends, among others. The operationalization of gender reasons in the Criminal Code is made by identifying seven specific circumstances, namely if (a) the victim experienced any sort of sexual violence; (b) the victim suffered ignominious or shameful injuries or mutilations either before or after her death or was the object of necrophilia; (c) the perpetrator had previously acted violently against the victim (in any of the following relationship contexts: family, school, or work); (d) the existence of a romantic or emotional attachment or bonds of trust; (e) evidence that the victim was previously harassed, assaulted, or threatened by the perpetrator; (f) the victim was put in solitary confinement before her death; and (g) the victim’s body was exposed or displayed in a public space.

The Federal Criminal Code does not explicitly define the perpetrators of feminicides as males. Nevertheless, the prevalent social representation is that they are indeed males. Conceptually, females can perpetrate feminicides if their crime is driven by gender reasons (Glass et al., 2004). Since 2012, when the killing of a woman for gender reasons was included in the Federal Criminal Code, 28 Mexican states include feminicide as felony. The number of feminicide cases that the State Attorneys’ offices investigate has increased from 411 in 2015 to 983 in 2019 (2.4 times higher). In addition to the plausible escalation of the phenomenon, there are at least three explanations for this increase: (a) Over time more states have included feminicide as a felony in their Criminal Codes; (b) The media attention to the phenomenon promoted by social activists that claim that all intentional deaths of women need to be investigated as feminicides; and (c) the State Attorney’s offices’ increase in cognitive and human resources with which to identify cases. I will use the case of the most populated state of Mexico, namely Mexico State, as an example. There were 59 of cases of feminicide under investigation in 2015; by 2019, this number had more than doubled to 123 cases. Of course, these figures correspond to the cases of feminicide identified by public authorities and might differ from the real prevalence of the phenomenon. This was the case of Mariana Lima, a 29-year-old woman killed by her husband in 2010 after enduring partner violence. Her husband was a police officer appointed at the State Attorney’s office. He colluded with other police officers to disguise Mariana Lima’s murder as a suicide. Her mother did not cease her outcry for justice until Mariana’s case was finally recognized as feminicide in 2015 (details in Quintana Osuna, 2018).

Gender (In)equality, Gender-based Violence and Feminicides

Gender inequality has been defined as “the departure from parity in the representation of women in key dimensions of social life” (Young et al., 1994, p. 54). It is grounded in and supported by patriarchal structures and beliefs that promote traditional gender roles and stereotypes (Frías, 2008). Between 1995 and 2018, the level of gender inequality in Mexico decreased by 53% (own calculations based on PNUD’s Gender Inequality Index). Low levels of gender equality are associated with discrimination against women and fewer opportunities in socially valued positions; therefore, women do not threaten traditional men’s status in society (Whaley et al., 2013). Based on two highly referenced feminist approaches, namely backlash and ameliorative, empirical evidence of the relationship between increases in gender inequality and female homicide seems to be contradictory (Rogers & Alsleben, 2020).

Backlash theory contends that as gender inequality decreases, men might feel vulnerable by women’s increased status in society (i.e., in political, educational, economic, and legal spheres) and gender-based violence against women (including female homicide and feminicides) might increase. Gender equality threatens men’s status in society; backlash or retaliation is manifested through social strategies aimed at controlling women and their achievements (Russell, 1975; Yllo, 1984). “Violence and the threat of violence are quite often effective means of social control” (Whaley et al., 2013, p. 733). In contrast, the amelioration approach argues that increases in gender equality will entail a decrease in violence against women (see Rogers & Alsleben, 2020) because men and women share power and responsibilities both in public and private realms. In other words, the distribution of power in the economic, political, educational spheres and familial and domestic responsibilities are distributed more equitably. This more equitable distribution is connected to a weakening of traditional gender roles and stereotypes that support gender-based violence against women.

Previous studies offer contradictory evidence on the relationship between gender inequality and female homicide. Rogers and Alsleben (2020) claim that this might be linked to heterogeneous definitions of gender inequality. It is also plausible that these inconsistent results could be associated with the failure of current research to differentiate between female murders rooted in gender inequality (feminicides) and female homicides unassociated with gender issues. In addition, contextual sociopolitical conditions, and conflicts, as well as other structural factors such as structural violence might be behind these inconsistencies.

Methods

This study employs mortality statistics published by the INEGI⁴ for the period 2001 to 2017 (the last year for which microdata is available) and population estimates drawn from the National Council of Population (Consejo Nacional de Población, CONAPO, 2019). As mentioned before, to assess whether a female homicide is indeed feminicide based only on mortality records is a difficult task. Inspired by the methodology developed by Torreblanca and Merino (2017) and used by others (Valdivia et al., 2020), I created a more comprehensive operationalization of the category of feminicide. However, this is still a proxy to the true prevalence of the phenomenon. We consider the death of a woman to be a case of feminicide when:

a) The victim experienced any sort of sexual violence: 0.24% of females died as a result of a sexual assault (causes of death Y050–Y059 in the Manual of the international statistical classification of diseases, injuries and causes of death).

b) The victim suffered ignominious or shameful injuries or mutilations either before or after her death or was the object of necrophilia: 0.76% of females died due to physical assault or because somebody threw corrosive substances at them (X860–X869 and Y040–Y049).

c) The perpetrator had previously enacted violence against the victim (in any of the following relationship contexts: family, school, or work). In 3.4% of the cases, the victim had previously experienced family and/or partner violence.

d) The existence of a romantic or emotional attachment or bonds of trust. As a proxy, we would consider all female deaths occurring in a household regardless of the cause of death as feminicide, as well as those associated with the neglect of parents, partners, friends, public authorities, or people known by the victim (33.3% of all intentional female deaths).

We coded these cases as feminicide independently of the sex of the perpetrator. The available data on mortality records do not permit us to ascertain whether the victim was previously harassed, assaulted, or threatened by the perpetrator; whether she had been put in solitary confinement before her death; or if her body had been exposed or displayed in a public space. All other female intentional deaths were classified as female homicides. According to our operationalization, during the period under study (2000–2017), a total of 301,849 people were killed in Mexico, 88.8% of which were male homicides, 7.3% female homicides, and 3.9% femicides. In other words, 35% of all female murders were feminicides.

The external validity of our operationalization was tested for 2017 (the most recent data point). We compared the number of cases being investigated as feminicides in the State Attorneys’ offices (N = 741) according to the official data⁵ with the number rendered by our operationalization (N = 973). These figures are very similar when one takes into consideration the level of inclusion of feminicide in Criminal Codes at the time and the existing evidence that not all feminicides are investigated as such, which might be the case of those occurring in remote rural communities (Figueroa Romero, 2019).

The mortality records do not allow us to accurately identify the victims’ ethnic characteristics. Until 2012, whether the victim spoke an indigenous language—a proxy for indigenous ethnicity adopted in official statistics—was not included on their death certificate. In 2012, whether the victim spoke an indigenous language was missing on about 56% of all death certificates, in 2017 the percentage decreased to 40.6% for male homicides, and 39% for female homicides (own calculations).⁶ For this reason, we take whether the municipality where the deceased body was found is classified as indigenous or non-indigenous according to the classification of Serrano Carreto (2006). According to his classification, a municipality is classified as indigenous when: first, more than 40% of the population speaks an indigenous language (60% of those who speak an indigenous language nationally); and second, those municipalities that nationally concentrate 32% of the population speaking an indigenous language if at least 10% of the population speaks an indigenous language. Further, five additional municipalities were included because they had highly concentrated areas with the population speaking a minority-used indigenous language. Of the 2,443 existing municipalities, 907 were classified as indigenous. According to this classification, in the so-called indigenous municipalities we can find both indigenous and mestizo (mixed race) individuals. Progressive disuse of indigenous languages and the discrimination and stigma associated to speaking an indigenous language suggest that this geographical criterion might accurately identify indigenous populations.

I conducted descriptive analyses from the available official mortality records. These are population data. For this reason, analyses to test for statistically significant differences between indigenous and non-indigenous municipalities were not required.

Homicides of Males and Females and Feminicides in Mexico (2001–2017)

The increase of homicides in Mexico since the inauguration of Felipe Calderón in 2006 has been broadly discussed. There have been several explanations for this escalation, some of which are associated with the recrudescence of violence and competition among drug cartels and the diversification of illegal activities they are involved in (Durán-Martínez, 2015; Fuerte Celis & Pérez Lujan, 2017; Trejo & Ley, 2018), the trafficking of firearms from the United States (Dube et al., 2013; Perez Esparza et al., 2019); high unemployment rates, lack of access to formal education, the military control of public security, decreases in the GDP per capita, and political competition (Mejía Escamilla et al., 2020; Morales, 2013; Trejo & Ley, 2018).

As observed in Figure 1, from the period 2001 to 2007, male homicide rates experienced a decreasing trend. After 2007, however, the rate began increasing to reach a peak during 2010 to 2011. The rates began to decrease again during 2012 to 2014 and reached its highest peak in 2017. Female homicide rates are much lower than those of males and the general trend mimics that of males. When one differentiates between female homicides and feminicides, one can notice that the rate of feminicides has remained more stable than that of female homicides. This is consistent with the theoretical origin of the phenomenon: gender inequality rooted in patriarchal traditions and beliefs. The overall rate average increase of feminicides for the period 2001 to 2017 is 2.2% yearly (being higher in indigenous municipalities than in non-indigenous municipalities as shown in Table 2), while that of female homicides is 6.0%. This percentage of rate increase is similar to that of males (5.9% per annum) (analyses not shown).

Figure 1.

Feminicides, male and female homicide rates per 100,000 (2001–2017).

In Figure 2, we graphically examine the association between female homicides, male homicides, and feminicides with lower levels of gender equality, measured by the UNDP Gender Inequality Index (GII). It reveals a lack of linear association between gender inequality and both male and female homicides for the period 2010 to 2017 (male homicides not shown in Figure 1). This lack of association is also found when feminicides and female homicides are merged. For feminicides, however, there is a significant negative association (cor = −0.92***), which suggests that on a national level, as the levels of gender inequality decrease, the rate of feminicide increases. This supports the backlash theory, which claims that women’s gains in gender equality are associated with a violent reaction aimed at maintaining the status quo. As expected, neither male nor female homicides are linked to women’s gains in gender equality. These might be related to other contextual and socioeconomic variables, as discussed in the previous section.

Figure 2.

Relationship between the Gender Inequality Index, feminicides and male and female homicides (2010–2017).

Feminicides and Male and Female Homicide in Indigenous and Non-Indigenous Municipalities

Figure 3 differentiates between male and female homicides and feminicides in indigenous and non-indigenous regions. Male homicide rates per 100 thousand males (bars) reveal a different pattern in both indigenous and non-indigenous regions. The sharp increase in male homicides during the period 2007 to 2011 in non-indigenous municipalities is not present in the indigenous municipalities. In the latter, we find a more sustained increase. In indigenous municipalities, feminicides and female homicides follow a similar pattern, but female homicides’ rates are higher. However, there is a small increase in female homicide rates after 2015. In non-indigenous regions, the feminicide rates have remained more stable over time but female homicide rates have increased since 2008 and follow a similar pattern to those of the males. Unfortunately, a municipal-level GII that would allow us to assess if there is a relationship between gender inequality and the death of females is not available.

Figure 3.

Feminicides, male and female homicide in indigenous and non-indigenous municipalities per 100,000 inhabitants, 2001 to 2017.

Table 1 shows that males were most affected by murders during the 2001 to 2017 period (89% in non-indigenous municipalities and 87.7% in indigenous municipalities). The percentage of female homicides is very similar in both types of municipalities (7.3% and 7.4%, respectively). Nevertheless, the percentage of feminicides among all the murders is higher in indigenous municipalities than in non-indigenous municipalities (4.8% vs. 3.8%). Similarly, 34.1% of female murders are feminicides in non-indigenous municipalities, which is considerably lower than that in indigenous municipalities (39.3%).

Table 1.

Feminicides, Male and Female Homicides in Indigenous and Non-Indigenous Municipalities (2001–2017). In Percentages.

	Non-indigenous	Indigenous
Male homicide	89.0	87.7
Female homicide	7.3	7.4
Feminicides	3.8	4.8
% Feminicides among female murders	34.1	39.3
N	255,348	45,419

Table 2 shows that, on average, the rate of feminicide in indigenous municipalities is higher than in non-indigenous municipalities (1.45 vs. 1.26 per 100,000 females). In contrast, both the rate of female homicides and male homicides is higher in non-indigenous municipalities (2.42 vs. 2.03 and 31.04 vs. 25.35, respectively). Table 2 reveals that there is more fluctuation in feminicide rates in indigenous municipalities than in non-indigenous municipalities. For example, in 2002 and 2005, the feminicide rate in indigenous municipalities reached 1.9 per 100,000 females, while the lowest rate corresponds to 2008 (1.1 per 100,000 females). In contrast, in non-indigenous communities, the rate ranged from 0.9 in 2007 to 1.6 in 2017. This suggests that feminicide in non-indigenous municipalities is a more stable phenomenon than in indigenous municipalities. The last line of Table 2 shows the average increase per annum. It reveals that the rate of feminicide in indigenous municipalities has increased by 0.03 yearly in contrast to 0.02 in non-indigenous. The average increase in the rate of female homicides is similar in both indigenous and non-indigenous municipalities.

Table 2.

Feminicide, Female and Male Homicide Rates in Indigenous and Non-Indigenous Municipalities per 100,000 Inhabitants (2001–2017).

	Indigenous			Non-indigenous
	Feminicide	Female homicide	Male homicide	Feminicide	Female homicide	Male homicide
2001	1.3	1.7	27.1	1.2	1.8	20.7
2002	1.9	2.4	26.4	1.1	1.5	19.1
2003	1.5	1.9	23.1	1.1	1.6	18.4
2004	1.7	1.7	21.0	1.1	1.2	17.0
2005	1.9	1.1	20.1	1.2	1.5	18.0
2006	1.6	2.0	20.3	1.0	1.4	18.2
2007	1.2	1.4	17.2	0.9	1.1	15.3
2008	1.1	1.4	19.3	1.2	1.6	26.0
2009	1.2	1.9	26.1	1.4	2.3	35.6
2010	1.3	2.1	26.2	1.4	3.2	47.7
2011	1.2	1.7	25.0	1.3	3.9	49.0
2012	1.6	2.3	28.3	1.2	3.8	44.4
2013	1.5	2.2	27.7	1.4	3.2	37.1
2014	1.7	2.4	25.6	1.3	2.6	31.1
2015	1.3	2.0	27.7	1.3	2.6	31.7
2016	1.4	2.5	30.0	1.5	3.2	38.6
2017	1.6	3.2	36.3	1.6	3.9	49.0
Period average	1.45	2.03	25.35	1.26	2.42	31.04
Year average increase	0.03	0.08	0.03	0.02	0.08	0.08

Source. Own calculations.

These results about the increase in the rates of feminicides must not be broadly linked to reductionistic cultural interpretations related to the acceptance and normalization of subordination and violence against women in indigenous communities. Indigenous communities are heterogeneous in terms of their traditional norms and the role that women play. In the last decades, indigenous women have made significant efforts to erode patriarchal norms and traditions that condone, normalize violence against them, promote their subordination, and prevent them from participating in community and political affairs altogether they have made clear demands for indigenous self-government (Gómez, 2014; Hernández-Castillo, 2002). In indigenous regions, structural and sociopolitical factors might be also behind the higher rates of feminicides such as poverty, marginalization, insecurity, progressive disintegration of the peasant economy, militarization of indigenous regions, forced displacement, resistance to land dispossession, historical racism and the war against drugs (De Marinis, 2016; Hernández Castillo, 2019; Olivera, 2006; Valladares de la Cruz, 2018).

Characteristics of the Crimes and Female Victims

Table 3 shows the marital statuses and average age of the victims of feminicide and female homicide. There are not many differences among those from indigenous and non-indigenous municipalities. Among the victims of feminicide, there is a higher percentage in both kinds of municipalities that were married or cohabited than among female homicide victims (50.7% vs. 36.6% in indigenous municipalities, and 46.4% vs. 34.3% in non-indigenous municipalities). By contrast, most female homicide victims in both types of municipalities tend to be single (39%). In non-indigenous municipalities, the percentage of femicides and female homicides among divorced and separated women surpasses that of indigenous municipalities. It is also remarkable that about 8% of all feminicides are committed against children under 12 years old.

Table 3.

Marital Status and Average Age Feminicide and Female Homicide Victims by Indigenous Municipality (2001–2017).

	Indigenous		Non-indigenous
	Femicide	Female homicide	Femicide	Female homicide
Marital status
Married	28.9	20.6	27.9	20.2
Cohabiting	21.8	16.0	18.5	14.1
Divorced and separated	2.8	2.1	4.0	4.0
Single	25.7	38.2	28.3	39.0
Widow	8.7	7.6	8.5	4.4
Under 12 years old	7.8	5.6	8.0	4.4
Unknown	4.3	9.9	4.8	13.9
Average age	38.7	33.9	37.7	31.9

Source. Own calculations based on INEGI’s mortality database.

For the period under study, the victims of feminicide tend to be older on average in both indigenous and non-indigenous municipalities (38.7 and 37.7 years, respectively). Victims of female homicide are significatively younger. Victims of female homicide were 33.9 years old in indigenous municipalities while they averaged 2 years younger in non-indigenous municipalities (31.9 years old).

Regarding the means used to commit homicide, over time, female homicides and feminicides in which a firearm was used has increased (analyses not shown). Table 4 shows the causes of female deaths in indigenous and non-indigenous municipalities during the period 2001 to 2017. The analyses show that female homicides in both types of municipalities tend to be committed with a firearm and a knife or machete. During the period under study, victims of feminicide are more likely than victims of female homicide to die from wounds inflicted by knives or machetes (21.6% vs. 16.7%% in indigenous municipalities, and 22.3% vs. 11.7% in non-indigenous). This type of death suggests that cruelty is more associated with feminicides than with female homicides. Similarly, victims of feminicide are more likely than female victims of homicide to die from the use of physical force. In indigenous municipalities there are not relevant differences in the percentage of women who died via suffocation or drowning; in non-indigenous municipalities, feminicides committed under these circumstances is higher (20.2% vs. 15.9%). Finally, in indigenous municipalities 32.1% of the murders were perpetrated by other means in indigenous municipalities (poisoning, sexual violence, burning or scalding, or as result of neglect, among others) versus 21.5% in non-indigenous municipalities. This data suggests that feminicides are perpetrated by different means and are more likely to entail cruelty and suffering for the victim.

Table 4.

Means to Commit Feminicide and Female Homicide in Indigenous and Non-Indigenous Municipalities (2001–2017).

%	Indigenous		Non indigenous
%	Feminicide	Female homicide	Feminicide	Female homicide
Firearm	26.9	36.8	31.2	48.0
Knife, machete	21.6	16.7	22.3	11.7
Physical force	5.0	2.1	4.8	1.5
Suffocated and drown	14.5	14.9	20.2	15.9
Other	32.1	29.5	21.5	23.0

Source. Own calculations based on INEGI’s mortality database.

Conclusions

Homicides and feminicides are a social problem; both male and female homicide rates have increased since 2007. The rates of female homicide and feminicide have increased over time, but the increase is much lower than that of males. The trend followed by female homicides resembles—on a smaller scale—that of male homicides. Feminicide rates have remained more stable in Mexico than female homicide rates. Nevertheless, from 2001 to 2017, feminicide rates increased yearly in both indigenous and non-indigenous municipalities (0.03 and 0.02, respectively). The share of feminicides in the overall female murder rate is higher in indigenous municipalities than in non-indigenous (39.3% and 34.1%, respectively).

The theoretical differences between feminicides and female homicides are empirically supported by the diverse sociodemographic characteristics of the victims and the ways in which the murders were committed. In both indigenous and non-indigenous municipalities, the rate of femicides in which a knife, machete, or physical force was used is much higher over time than the rate of female homicides in which these means were used. This suggests that feminicides are likely to be committed by means that inflict more suffering to victims.

It is necessary to differentiate between female homicide and feminicide because research oriented from a feminist perspective that does not distinguish between them might lead to misleading results. For example, female homicides associated with civilians being killed in the crossfire between military and drug cartels are not rooted in socially constructed gender inequalities between men and women. In contrast, the misogynistic homicide of a prostitute by a consumer or the killing of a girl who is sexually abused by an acquaintance is related to a broader pattern of gender domination in which women and girls are seen as the property of males as well as disposable and to-be-used bodies. Public policies aimed at preventing this most brutal expression of violence against women must differentiate between feminicides and female homicides; while the latter might require the development of public security-oriented programs, feminicides require interventions that tackle structural and ideological gender inequalities. In the short term, these interventions might increase feminicide rates if the backlash theory is correct. However, I hypothesize that this increase will not follow a linear trend; instead, it might follow an inverted U-shaped trend that would comprise both the backlash and the ameliorative perspectives (depending on the time frame elapsed since ideological and structural gender inequality structures begin to be significantly eroded).

In the case of Mexico, this study shows that, for the period 2010 to 2017, lower levels of gender inequality measured by the Gender Inequality Index are associated with an increase in feminicide but are not related to either female or male homicides. Due to the lack of available data, it was not possible to conduct an analysis on the plausible association between gender inequality and prevalence of female homicides and feminicides in indigenous and non-indigenous municipalities.

The rate of feminicide in indigenous municipalities has increased more than in non-indigenous municipalities since 2001 (0.03 vs. 0.02), whereas the yearly average increase of female homicide rates is the same regardless of the type of municipality. This suggests that public policies need to be specifically tailored to address feminicides and their particularities in indigenous municipalities. Recognizing the heterogeneity among indigenous cultures and without aiming to provide a stereotypical image of them, indigenous women tend to hold more traditional gender beliefs and gender roles than their non-indigenous counterparts (Casique, 2021). The progressive changes in ideological and gender structures that have been taking place in Mexico (Frías, 2017) may have triggered a stronger backlash reaction among men in indigenous municipalities than in non-indigenous municipalities. In addition, the current context of structural discrimination against indigenous peoples, militarization of indigenous regions, the war against drugs and land dispossession, women’s bodies have become territories to be invaded, mutilated, and discarded (De Marinis, 2016, 2017; Hernández Castillo, 2019). These explanations might be behind the increase of feminicides in indigenous municipalities.

Regarding the claims made by feminist and women’s organizations concerning the increase in the number of feminicides in Mexico, this research suggests that the number has remained more or less stable with a small tendency to increase, especially in indigenous municipalities. The investigation of feminicides as such, instead of as homicides, suicides and accidental deaths, has undoubtedly increased over time due to the changes in the criminal codes and the allocation of resources in State Attorney’s Offices to investigate feminicides, or at least register them. The use of new communication technologies and social media has promoted the visibility of feminicides (see Figueroa Romero, 2019). It is plausible that the alleged increase in the phenomenon could be confounded by an increase in its identification. Regardless of the trend, the Mexican State must tackle feminicides by strengthening public policies aimed at preventing all expressions of gender violence against women and girls, especially in those regional contexts in which poverty, social and political conflicts.

Previous studies suggest that most feminicides are committed by (ex-)partners and family members (Dawson & Gartner, 1998; Johnson et al., 2019; Toprak & Ersoy, 2017). Both in indigenous and non-indigenous municipalities, current prevention efforts aimed at empowering women and girls must be paralleled with programs intended to change men’s gender beliefs in order to control their retaliatory behaviors upon women’s increased statuses. These programs need to adopt an intercultural and intersectional perspective to acknowledge the needs and particularities of different social groups, such as indigenous populations.

This study has several limitations associated with the use of secondary data and the classification of female murders as feminicides and female homicides because no contextual information is provided in the data drawn from the death certificates. Nevertheless, as shown in this study, the number of feminicide cases in the States’ General Attorney’s offices in 2017 is very similar to that resulting from my operationalization. Because death certificates feature incomplete information (i.e., whether homicide victims spoke an indigenous language, which is a proxy for ethnicity), we determined whether the municipality wherein the homicide took place is classified as indigenous. This does not necessarily imply, however, that victims were of indigenous ethnicity.

Finally, this research highlights that future research addressing homicides from gender and feminist perspectives must differentiate between homicides and feminicides because the roots of both phenomena are different. Therefore, public interventions aimed at preventing them cannot be the same, and these need to incorporate contextual and cultural particularities, as well as structural inequalities that might put women and girls at risk in indigenous municipalities.

Footnotes

Declaration of Conflicting Interests

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

Funding

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

ORCID iD

Sonia M. Frías

Notes

Author Biography

Sonia M. Frías, PhD, is a researcher and professor at the Regional Center for Multidisciplinary Research (National Autonomous University of Mexico). She received her PhD in Sociology from the University of Texas at Austin. She conducts research on gender equality, violence against women and children from a gender perspective, and the role of the State in protecting women and children from violence. She is currently the associate editor of the Journal of Latin American Population (Revista Latinoamericana de Población).

References

Aguilar Gutiérrez

Aguilar Hernández

O. A

. (2019). Violencia de género en México, hacia un mapa del feminicidio en Tlaxcala. Cuestiones de género: de la igualdad y la diferencia, 14, 295–315.

Alcocer Perulero

. (2014). “Prostitutas, infieles y drogadictas”. Juicios y prejuicios de género en la prensa sobre las víctimas de feminicidio: el caso de Guerrero, México. Antípoda. Revista de Antropología y Arqueología, (20), 97–118. https://revistas.uniandes.edu.co/doi/10.7440/antipoda20.2014.05

Arteaga Botello

Valdés Figueroa

. (2010). Contextos socioculturales de los feminicidios en el Estado de México: nuevas subjetividades femeninas. Revista Mexicana de Sociología, 72(1), 5–35.

Bejarano Celaya

. (2014). El feminicidio es sólo la punta del iceberg. Región y Sociedad, 26(Especial 4), 13–44.

Caputi

Russell

(1992). Femicide: Sexist terrorism against women. In Radford

Russell

D. E. H.

(Eds.), Femicide: The politics of woman killing (pp. 13–21). Twayne.

Casique

(2021). Prevalencia y factores asociados a las violencias de pareja contra las mujeres indígenas en México. In Frías

S. M.

(Ed.), Violencias de género en contra de mujeres y niñas indígenas en México en contextos públicos, privados e institucionales (pp. 39–86). CRIM-UNAM.

Castañeda Salgado

M. P.

Ravelo Blancas

Pérez Vázquez

. (2013). Feminicidio y violencia de género en México: omisiones del Estado y exigencia civil de justicia. Iztapalapa, 74, 11–39.

CONAPO. (2019). Proyecciones de la Población. Nacional 2010-2050: INEGI. http://www.conapo.gob.mx/es/CONAPO/Proyecciones_de_la_Poblacion_2010-2050

Corradi

Marcuello-Servós

Boira

Weil

(2016). Theories of femicide and their significance for social research. Current Sociology, 64(7), 975–995. https://doi.org/10.26633/rpsp.2019.26

10.

Dawson

Gartner

(1998). Differences in the characteristics of intimate femicides: The role of relationship state and relationship status. Homicide Studies, 2(4), 378–399.

11.

De Marinis

. (2016). Mujeres indígenas ante los escenarios del miedo en México: (in)seguridad y resistencias en la región triqui de San Juan Copala, Oaxaca. Estudios Latinoamericanos, 37, 65–86.

12.

De Marinis

. (2017). Despojo, materialidad y afectos: la experiencia del desplazamiento forzado entre mujeres triquis. Desacatos, 53, 98–113.

13.

Dube

García-Ponce

(2013). Cross-border spillover: U.S. gun laws and violence in Mexico. American Political Science Review, 107(3), 397–417.

14.

Durán-Martínez

(2015). Drugs around the corner: Domestic drug markets and violence in Colombia and Mexico. Latin American Politics and Society, 57(3), 122–146.

15.

Figueroa Romero

. (2019). Políticas de feminicidio en México: perspectivas interseccionales de mujeres indígenas para reconsiderar su definición teórica-legal y las metodología de recolección de datos. Journal of International Women’s Studies, 20(8), 64–86.

16.

Figueroa Romero

Jiménez Estrada

Guimont Marceau

Rice

. (2017). Violence against Indigenous women in the Americas. https://poli.ucalgary.ca/manageprofile/sites/poli.ucalgary.ca.manageprofile/files/unitis/publications/1-8355408/CALACS_Virtual_Forum_Violence_against_Indigenous_Women.pdf

17.

Freyermuth

(2003). Las mujeres de humo. CIESAS.

18.

Frías

S. M.

(2008). Measuring structural gender inequality. Toward the construction of a gender equality index in Mexican states. Social Indicators Research, 88(2), 215–246. https://doi.org/10.1007/s11205-007-9193-4

19.

Frías

S. M.

(2017). Structural and ideological gender equality in Mexico. In Örtenblad

Raili

Vasiljevil

(Eds.), Gender equality in a global perspective (pp. 152–176). Routledge.

20.

Frías

S. M.

(2021). Violencia sexual en contra de las mujeres y niñas indígenas. Limitaciones y alcances de la ENDIREH 2016. In Frías

S. M.

(Ed.), Violencias de género en contra de mujeres y niñas indígenas en México en contextos públicos, privados e institucionales (pp. 257–298). CRIM-UNAM.

21.

Fuerte Celis

M. P.

Pérez Lujan

. (2017). ¡La violencia cambia! Securitización y configuración territorial del crimen organizado en México, 2007–2011. Journal of Latin American Geography, 16(3), 59–81.

22.

Glass

Koziol-McLain

Campbell

Block

C. R.

(2004). Female-perpetrated femicide and attempted femicide: A case study. Violence against Women, 10(6), 606–625.

23.

Gómez

(2014). Procesos organizativos y participación política de mujeres indígenas en México: voces de activistas y abordajes en la bibliografía. Mora, 20, 47–65.

24.

Grzyb

Naudi

Marcuello-Servós

(2018). Feminicide definitions. In Weil

Corradi

Naudi

(Eds.), Feminicide across Europe (pp. 17–31). Bristol University Press.

25.

Gutiérrez Chong

Valdés González

L. M

. (2015). Ser indígena en México. Raíces y derechos. IIJ-UNAM.

26.

Hernández-Castillo

(2002). National law and indigenous customary law: The struggle for justice of indigenous women in Chiapas, Mexico. In Molyneux

Razavi

(Eds.), Gender, development and rights (pp. 384–413). Oxford University Press.

27.

Hernández Castillo

R. A

. (2019). Racialized geographies and the “war on drugs”: Gender violence, militarization, and criminalization of Indigenous peoples. The Journal of Latin American and Caribbean Anthropology, 24(3), 635–652.

28.

Incháustegui

(2014). Sociología y política del feminicidio: algunas claves interpretativas a partir de caso mexicano. Revista Sociedade e Estado, 29(2), 373–400.

29.

Johnson

Eriksson

Mazerolle

Wortley

(2019). Intimate femicide: The role of coercive control. Feminist Criminology, 14(1), 3–23. https://doi.org/10.1177/1557085117701574

30.

Lagarde

(2006). Del femicidio al feminicidio. Revista de Psicoanálisis, (6), 216–225.

31.

Lagarde

(2010). Preface: Feminist keys for understanding feminicide. Theoretical, political and legal construction. In Fregoso

R. L.

Bejarano

(Eds.), Terrorizing women: Feminicide in the Americas (pp. xi–xxvi). Duke University Press.

32.

Luna Blanco

M. A

. (2019). Razones de género en los feminicidios: elementos objetivos y subjetivos en 5 casos de violencia feminicida en Chiapas. Cuestiones de Género: de la Igualdad y la Diferencia, 14, 253–272.

33.

Martin

P. M.

Carvajal

(2016). Feminicide as ‘act’and ‘process’: A geography of gendered violence in Oaxaca. Gender, Place & Culture, 23(7), 989–1002.

34.

Mejía Escamilla

M. G.

Spijker

Manetta

. (2020). Distribución espacial del homicidio juvenil masculino y desigualdad territorial en Ciudad de México y Estado de México (2000-2002 y 2010-2012). Revista Latinoamericana de Población, 27, 185–227. https://doi.org/10.31406/relap2020.v14.i12.n27.6

35.

Monárrez Fragoso

J. E

. (2000). La cultura del feminicidio en Ciudad Juárez, 1993-1999. Frontera Norte, 12(23), 87–117.

36.

Monárrez Fragoso

J. E

. (2002). Feminicidio sexual serial en Ciudad Juárez: 1993-2001. Debate Feminista, 25, 279–305.

37.

Morales

(2013). Factores que explicaron los niveles de homicidios y fallecimientos por rivalidad delincuencial entre 2007 y 2010 en México. Coyuntura Demográfica, 3, 17–22.

38.

OACNUDH, & ONUMUJERES. (2014). Modelo de protocolo latinoamericano de investigación de las muertes violentas de mujeres por razones de género (femicidio/feminicidio). OACNUDH & ONUMUJERES.

39.

Oliver

Valls

(2004). Violencia de género. Investigaciones sobre quiénes, por qué y cómo superarla. El Roure.

40.

Olivera

(2006). Violencia femicida: Violence against women and Mexico’s structural crisis. Latin American Perspectives, 33(2), 104–114.

41.

Perez Esparza

Johnson

Gill

. (2019). Why did Mexico become a violent country? Assessing the role of firearms trafficked from the US. Security Journal, 33, 179–209.

42.

Quintana Osuna

K. I

. (2018). El caso de Mariana Lima Buendía: una radiografía sobre la violencia y discriminación contra la mujer. Cuestiones Constitucionales, 38, 143–168.

43.

Radford

Russell

D. E.

(1992). Femicide: The politics of woman killing. Open University Press.

44.

Ríos

(2021). Disponibilidad de información sobre violencia de género contra mujeres y niñas indígenas en las entidades federativas. In Frías

S. M.

(Ed.), Violencias de género en contra de mujeres y niñas indígenas en México en contextos públicos, privados e institucionales (pp. 381–424). CRIM-UNAM.

45.

Rizo Amézquita

J. N

. (2017). Población indígena en cifras. Boletín CONAMED - OPS, 13, 7–13.

46.

Rogers

M. L.

Alsleben

M. M.

(2020). Gender equality and the usual suspects: A cross-national study on female and male homicide victimization. Homicide Studies, 25(1), 83–105. https://doi.org/10.1177/1088767920937765

47.

Roldán

J. A.

Álvarez

M. A.

Carrasco

Guarneros

Ledesma

J. Á.

Cuchillo-Hilario

Chávez

(2017). Marginalization and health service coverage among indigenous, rural, and urban populations: A public health problem in Mexico. Rural and Remote Health, 17, 3948. https://doi.org/10.22605/RRH3948

48.

Russell

(1975). The politics of rape: The victim’s perspective. Stein and Day.

49.

Russo

N. F.

Pirlott

(2006). Gender-based violence. Concepts, methods, and findings. Annals of the New York Academy of Sciences, 1087(1), 178–205.

50.

SEGOB, INMUJERES, & ONUMUJERES. (2016). La violencia feminicida en México: aproximaciones y tendencias 1985-2016. SEGOB; INMUJERES; ONUMUJERES.

51.

Serrano Carreto

. (2006). Regiones indígenas de México. CDI.

52.

Sesia

P. M.

(2016). Violencia y mortalidad materna en contextos indígenas de Oaxaca: Una mirada etnográfica. Géneros, 14(1), 53–84.

53.

Solís

(2017). Discriminación estructural y desigualdad social. CONAPRED; CEPAL.

54.

Toledo Vázquez

. (2009). Feminicidio. Naciones Unidas.

55.

Toprak

Ersoy

(2017). Femicide in Turkey between 2000 and 2010. PLoS One, 12(8), e0182409.

56.

Torreblanca

Merino

(2017). Una propuesta para contar feminicidios en México. Animal Político, (28 noviembre). https://www.animalpolitico.com/blogueros-el-foco/2017/11/28/una-propuesta-para-contar-feminicidios-en-mexico/

57.

Trejo

Ley

(2018). Why did drug cartels go to war in Mexico? Subnational party alternation, the breakdown of criminal protection, and the onset of large-scale violence. Comparative Political Studies, 51(7), 900–937. https://doi.org/10.1177/0010414017720703

58.

Valdivia

Castro

(2013). Gender bias in the convergence dynamics of the regional homicide rates in Mexico. Applied Geography, 45, 280–291.

59.

Valdivia

Castro

Rodríguez Luna

(2020). Análisis espacial de la dinámica de la tasa de homicidios por sexo y feminicidios en México (2001-2017). In Castro

Ríquer

(Eds.), Violencia contra mujeres. Sobre el difícil diálogo entre cifras y acciones de gobierno (pp. 47–100). CRIM-UNAM.

60.

Valladares de la Cruz

L. R.

(2018). El asedio a las autonomías indígenas por el modelo minero extractivo en México. Iztapalapa. Revista de Ciencias Sociales y Humanidades, 39, 103–131.

61.

Vázquez Sandrin

. (2014). La identidad étnica desde los estudios longitudinales. In Rodríguez Wong

L. L.

Antón Sánchez

(Eds.), La población afrodescendiente e indígena en América Latina (pp. 115–130). ALAP.

62.

Whaley

R. B.

Messner

S. F.

Veysey

B. M.

(2013). The relationship between gender equality and rates of inter-and intra-sexual lethal violence: An exploration of functional form. Justice Quarterly, 30(4), 732–754.

63.

Yllo

(1984). The status of women, marital equality, and violence against wives. Journal of Family Issues, 5(3), 307–320.

64.

Young

Fort

Danner

(1994). Moving from ‘the status of women’ to ‘gender inequality’: Conceptualisation, social indicators and an empirical application. International Sociology, 9(1), 55–85.