Abstract

Why are some subnational states more dangerous for journalists? This exploratory article assesses the association of social variables with the murders of journalists within one single country, Mexico, where forty-one journalists were killed from 2010 to 2015. The article suggests that the violent deaths of journalists in Mexico's thirty-two states are more likely to happen in those subnational polities with high levels of social violence, internal conflict, severe violations of human rights, low democratic development, and economic inequality. The implications of this research and policy recommendations are discussed within the conclusion.

Introduction

On January 24th, 2015, the body of Mexican journalist Moisés Sánchez was discovered after twenty-two days of being kidnapped by armed men dressed in civilian clothing. He was the owner and editor of the weekly magazine La Unión in which he denounced local corruption, crime, and the poor quality of basic services. According to the official investigation, the journalist was killed on the same day of abduction following the order by the mayor of Medellín de Bravo, the hometown of Sánchez in the metropolitan area of the city of Veracruz.1 Sánchez's murder, which maintained impunity, not only constituted an extreme form of censorship and caused an expanded chilling effect among his colleagues, but also left his community without a critical voice from which the public could receive information of public interest.

As Sánchez's brutal homicide exemplified, lethal violence against journalists is a horrifying reality among many local journalists and investigative reporters working on corruption, crime, and human rights abuses at the political and metropolitan periphery in their home countries.2 However, as recent literature on the matter shows,3 lethal attacks against journalists and reporters are particularly rife in those political contexts (such as post-authoritarian countries or emerging democracies) in which journalists have certain incentives to pursue sensitive stories, but where the state has not been able to safely guarantee the exercise of journalism throughout the national territory, systematically failing to convict the perpetrators of those crimes.

In Mexico, Sánchez's murder was one of forty-one violent homicides of journalists that occurred between 2010 and 2015. One-third of these happened in the state of Veracruz, the deadliest place for journalists in the country.4 However, other Mexican states are almost as dangerous as Veracruz, such as Oaxaca or Tamaulipas, where five and four journalists have been killed, respectively, in this six-year period. In fact, even though journalists have been killed in fourteen Mexican states, six states made up more than 70 percent of the forty-one cases.5 Thus, the question that arises is the following: Why are some subnational states more dangerous for journalists in Mexico?6

Following special literature on this matter,7 it is suggested that the lethal danger to which local journalists are exposed in Mexico's states increases in those polities with internal conflict, high levels of human rights abuses, and societal violence, as well as poor levels of democratic and economic development. Thus, the purpose of this article is twofold. First, to identify within which subnational states the major rate of violent deaths of journalists is concentrated. Second, to explore to what extent the social determinants associated with the killing of journalists, found in the international literature, apply at the subnational level in Mexico.

Thus, the current situation in Mexico offers a good opportunity to explore the determinants associated with the killing of journalists at the subnational level, where the vast majority of lethal violence against the press occurs. For this reason, this work follows a subnational comparative research strategy, which has been widely used in political science and economics.8 This research design allows analysis of the research phenomenon and tests hypotheses about the social determinants associated with the killing of journalists in many cases (thirty-two subnational states) instead of just one (Mexico).

This article is structured as follows. The first part reviews the social determinants related to the killing of journalists. Then, the case of Mexico is presented. Further, in order to evaluate the validity of our hypothesis, we present the dependent and independent variables as well as the results of the logistic regression analysis and the Poisson models. Finally, the implications of this research and policy recommendations are discussed.

Setting the Stage: Journalists Killed in Non-Regular Wars

The violent deaths of journalists are a very critical issue worldwide; they represent not only a public health problem but also a major threat to the establishing of open societies. According to the Committee to Protect Journalists (CPJ), a well-respected and independent news safety organization based in New York, more than 1,300 journalists and media workers worldwide have been killed when performing their duty in the last two decades (see Figure 1). However, the figure shows that in the last twenty years this extreme form of censorship has been increasing worldwide. Among the possible reasons related to these trends, Simon Cottle9 suggests the following three: the loss of neutrality status in the journalistic field, the growth of global organized crime, and the use of new technology that increases exposition and creates digital concerns.

However, international figures also show that the deaths of journalists are not limited to regular conflicts or civil wars, as is primarily explored in the literature on war reporting.10 In fact, under certain circumstances, journalists working on reporting about local politics or crime can find themselves with their lives at lethal risk, sometimes as risky as in war situations.11 Indeed, specialized literature shows that this trend is common in countries that do not experience conventional wars or regular armed conflicts, like Mexico, but where the state is unable to safely guarantee the exercise of journalism throughout the national territory, and systematically fails to convict the perpetrators of crimes against the press.12

Furthermore, this lethal context for journalists is exemplified in international figures, which show that more journalists have been killed in countries that are not directly associated with ongoing conflict between sovereign parties or with regular civil wars.13 In fact, from 2008 to 2015, according to the International News Safety Institute (INSI), 60 percent of journalists and media workers were killed in peacetime14 rather than in regular wars or internal civil conflicts (see Figure 2).

In general, the majority of those journalists killed in “peacetime” (to use INSI terminology) were local journalists (or photojournalists) and investigative reporters working on politics, corruption, human rights abuses, or crime at the time of their murders.15 Further, some recent evidence rightly suggests that the violent deaths of journalists in non-regular wars are more likely to occur in those societies that enjoy some degree of press freedom, which foster conditions for local journalists to pursue sensitive news stories in their communities, but where the state institutions systematically failed to guarantee the exercise of journalism and convict the perpetrators of those crimes.16 Further, the literature also rightly points out that the particularly dangerous societies are those post-authoritarian societies or emerging democracies, like Mexico, in which the breakdown of the ancient regime and the internal instability and rife criminality accounts for why the press is the target of lethal violence. However, even when this worrisome situation has produced multiple headlines, it has not elicited a great deal of scholarly attention, and specialized literature has barely explored systematically the risk factors associated with those murders. Among the few empirical studies addressing the risk factors associated with the murder of journalists, the literature suggests that high levels of societal violence, internal conflict, and human rights abuses, as well as poor levels of democratic and economic development, all played a role in explaining the occurrences of these crimes.

Some Determinants to Journalist Killings

In general, social violence and lethal conflicts have important impacts on violent deaths of selective targets or populations.17 In regards to antipress violence, specialized literature commonly suggests an association between the killing of journalists in a country and the country's level of violence or violent conflicts.18 In this account, violence against the press is, in part, a result of the widespread violence at the macro level. Silvio Waisbord comments on this: “More than attesting how the press or public discourse is controlled and suppressed, anti-press violence reflects the impossibility of any institution of the public sphere to exist outside the spiral of violence.”19 In fact, the very few scholars that have empirically explored this assumption agree that, under certain conditions, the number of violent deaths of journalists in a country is highly associated with the country's levels of violent civilian deaths per capita.20

H1: Journalists are at greatest risk of being murdered in subnational polities with higher levels of social violence.

Highly pertinent to the above hypothesis is that the literature also suggests that another factor often associated with the lethal attacks of any selective (or nonselective) population are the widespread violations of human rights by the state actors.21 In regards to the factors that put journalists at lethal risk, following this line of thought, scholars argue that the most dangerous places for journalists are societies with a widespread context of violations of human rights and poor performance in defending human integrity.22 Victor Asal and colleagues offer some evidence for this: “we find that journalist killings are more likely in countries where there is civil conflict or more violations of personal integrity rights.”23 Thus, even when the association between the numbers of journalists killed in a given society and the numbers of human rights violations by state actors appears intuitive and is commonly mentioned in the literature, it has been rarely explored in empirical analysis. However, the few recent studies referring to this association have found a significant relationship between the number of journalists killed in a country and the country's low levels of physical integrity rights.24

H2: Journalists are at greatest risk of being murdered in subnational polities with more violations of human rights by the state.

In relation to the previously mentioned hypothesis, the literature often states that one factor associated with the violent murder of any population is the presence of internal armed conflict and/or internal conflict.25 Following this line of thought, as we mentioned earlier, international evidence suggests that journalists often suffer lethal violence, or drastically increase their risk of being murdered, when reporting from societies with internal conflict.26 However, this situation is not only a lethal threat to war correspondents, as it is also a rampant reality among many local beat journalists and investigative reporters working in developed regions, where the state's efficacy to monopolize the means of violence face severe limitations imposed by the presence of organized crime, the paramilitary, or guerrillas (what Waisbord calls “statelessness,” a condition which affects large regions in the Global South).27 Further, Riddick et al. add some empirical evidence to this idea: “When considering the rate of homicides of media workers within all countries, it appears that the ability of governments to control armed groups may be [a] relatively important factor on the extent to which governments can establish safe working environments for the media.”28 In fact, recent evidence on this matter in Latin America suggests that this is a reality among many local beat journalists and investigative reporters working in “low-intensity conflicts” in the region (like Mexico), where the violent presence of armed groups, like violent drug cartel cells, represent one of the major threats among local journalists and national correspondents.29

H3: Journalists are at greatest risk of being murdered in subnational polities with internal conflict.

Further, social scientists also suggest that journalists are at more of a lethal risk in developed societies with low performance of democratic institutions30 and/or poor “government quality.”31 Thus, some authors suggest that the killing of journalists is more likely to happen in those developed societies (especially new or emerging democracies) that experience poor performance in democratic institutions, and that suffer from a widespread culture of corruption and impunity, as well as a generalized weakness in the Rule of Law.32 In this line of thought, recent studies have found a relationship between these two variables (the violent murder of journalists and low democratic development).33 In their analysis, Asal et al. conclude that the country's poor levels of government functionality increase the lethal risk for journalists.34 Further, in Latin America, some scholars suggest that the uneven development of local democracy within subnational frontiers has a negative impact on the democratic function of the press35 and, under certain circumstances, may lead to antipress violence.36

H4: Journalists are at greatest risk of being murdered in subnational polities with low democratic development.

In order to explore more broadly the issue of public health, and particularly the outcome of violent deaths in a given population, social science also suggests reviewing other social variables as possible determinants, such as economic inequality and development.37 Indeed, when it comes to the study of the violent deaths of journalists, some scholars often mention economic development and poverty as social variables associated with the occurrences of antipress violence, especially in developed regions of Latin America.38 In this account, violent homicides of journalists are more likely to happen in remote areas away from the big cities—“brown areas,” says Guillermo O'Donnell39—where the press is more vulnerable to financial and political pressures. A recent study of journalists working on violent environments in Mexico addresses the influence of structural conditions (such as economic inequality and poor economic development) as an important external determinant of their professional autonomy.40 However, the few empirical studies researching the association between the violent deaths of journalists and a country's development indices have not found significant relationships in this matter.41

H5: Journalists are at greatest risk of being murdered in subnational polities with more levels of economic inequality.

H6: Journalists are at greatest risk of being murdered in subnational polities with less economic development.

The Mexican Case

Over the past decades, the number of journalists targeted with death has increased considerably in Mexico (see Figure 3). Indeed, according to international figures, there were sixteen journalists killed over the last twelve years of the one-party dictatorship (from 1988 to 1999);42 this increased to twenty-six during the first six years of the first federal government elected via democratic and free elections (2000–2006). From 2007 onwards, when the government launched an aggressive strategy against drug cartels, lethal attacks on the press increased even more. Since then, as many as seventy-nine journalists have been killed, according to Article 19 Mexico.43

Further, as the literature has suggested,44 the increasing number of journalists killed in Mexico, and the lethal conditions to which local journalists are exposed, coincide with large-scale social transformations, which altogether account for the increasing number of journalists killed after the demise of the long-term authoritarian regime in 2000. These social transformations include: dispersion of political power from the center to the periphery; widespread corruption among all levels of government; general weakness in the Rule of Law, which turns into impunity in crimes against the press (that is about 89 percent, according to the CPJ);45 and the frontal strategy against organized crime launched by the government one decade ago, which destabilized and multiplied the source and intensity of violence against drug cartels, among cartels, and against society.46

However, in addition to all these factors, recent literature on antipress violence in Mexico47 also suggests that when it comes to understanding the increasing number of journalists killed in the country, it is important to consider certain elements within the Mexican journalistic field that increase the lethal risk to which local journalists are exposed. In general, these authors suggest that high levels of instrumentalization and passive reporting, a lack of professionalization in gathering crime and corruption news, and an extended precariousness of working conditions all play a role in understanding the increasingly lethal risk to which journalists are exposed.

Thus, as all these social and media-related compounders are presumably associated with the high number of journalists targeted in Mexico, research on this phenomenon must be discussed at different levels of analysis, and from different disciplines and approaches.48 Nevertheless, in this article the association between the violent deaths of journalists and some social determinants are the main points of discussion.49 Specifically, it is focused on exploring the extent to which the social determinants of the killing of journalists identified by the international literature are associated, or not, with the killings of journalists in Mexico's states.

The Subnational Research Strategy

As in other countries of the region,50 in Mexico the majority of assaults against the press occur locally, where journalism is more vulnerable and the law is weaker. For this reason, this work follows a subnational comparative research strategy that has been widely used in political science and economics.51 This research design, which takes states or regions as a unit of analysis, allows the examination of the research phenomenon and tests hypotheses about the causes and implications of violence against the press in many cases (thirty-two states) instead of just one (Mexico). Moreover, a subnational perspective makes it possible to develop more valid findings within homogeneous cases than in cross-national studies, which are determined by problems of heterogeneity among their units of observation. Further, this research approach also allows me to address recent suggestions concerning the validation and exploration of social determinants related to the violent deaths of journalists in one single country.52

Independent Variables

In order to explore the relationship between social determinants and lethal violence against the press, in this section we present our independent variables, which correspond with our six hypotheses mentioned previously (see Table 1).53

Following previous studies on the association between societal violence and lethal violence against the press,54 the average number of intentional homicides by state is included, per 100,000 population per year (from 2010 to 2015), according to the National Institute of Statistics and Geography (INEGI). This measure is used as a proxy to the subnational level of social violence. This measure (which varies from two intentional homicides per 100,000 people in the southeast state of Yucatan, to sixty-one in the northern state of Chihuahua) is used as a proxy to the subnational level of social violence. Even when this measure indicates the level of lethal violence in each state, it is not indicative of the motive and intention of such homicides.

Then, in order to explore the relationship between violations of human rights and violent killings of journalists, we used as an independent variable the number of human rights violations perpetrated by federal security bodies, per 100,000 people, per state in 2015, according to the Mexican National Human Rights Commission (CNDH).55 This proxy summarized the number of complaints made to the CNDH in relation to human rights violations perpetrated by federal security bodies in Mexico's states. Among the federal security bodies incorporated in this measurement, the CNDH includes, but is not restricted to, the following: the military and the navy, the federal police, and the attorney general's office, among others. This measure varies from 0.3 human right violations per 100,000 people in the state of Puebla to 10 per 100,000 in the state of Tamaulipas.

In order to evaluate the presence of internal conflict at subnational level, we have used the presence (or absence) of any cases of internal forced displacement per state from 2014 to 2015, according to the Special Report on Internal Force Displacement (Informe Especial sobre Desplazamiento Forzado Interno), published by the CNDH in 2016.56 As has been documented elsewhere, internal forced displacement constituted a severe violation of a wide range of human rights and may be an indicator of internal conflict.57 In Mexico, specialist literature on the matter,58 along with the comprehensive report recently published by the CNDH, have documented that the most important cause associated with the wave of internal forced displacement is the human and social impact of the drug cartels' violence in Mexico's states. According to the CNDH special report, from which we gathered the data to construct this variable, the number of documented forced displacements between 2014 and 2015 varies from twenty forced displacements per 100,000 people in the state of Tamaulipas to virtually none in twenty other Mexican states. Thus, in order to mitigate the large number of zeros in the database, as well as the concentration of cases in a few states (just twelve states presented cases of internal forced displacement between 2014 and 201559), this article used a dichotomous measure of internal forced displacement per state (any case of internal force displacement versus no cases). Overall, we suggest that the presence of any case of forced displacement captures, to some extent, the presence of internal conflict within Mexico's states.

Furthermore, to assess the level of democratic development and good governance, the index of Democratic Development (IDD-Mexico) per state is included. For the purposes of this study, an average ranking from 2010 to 2014 per state is used. To our understanding, this index is one of the more complete, rigorous, and systematic measurements on the democratic performance of Mexico's states. Political scientists and civil society organizations have compiled the IDD-Mexico annually since 2010.60 In this sense, by using a great variety of sources and subindexes, the IDD-Mexico aims to consider a wide range of dimensions, all associated with the quality of democracy and good governance in Mexico's states. According to the information provided in their yearly reports,61 the index includes the following four dimensions: (1) the effective exercise of power and political participation; (2) the respect for political rights and civil liberties; (3) the institutional quality and political efficiency; and (4) the exercise and effectivity of the power to govern. The index runs from 0 to 10, where 10 represents a highly developed democracy. The lowest value was 2.25 (Guerrero), while the highest was 9.1 (Mexico City). Thus, this index encompasses the more relevant dimensions of democratic performance and, from our point of view, nicely captures the degree of democratic development and good governance in each Mexican state.

Finally, in order to assess the impact of economic inequality at local levels, the average mean of Gini index per state for 2012, according to the INEGI, is included. The Gini index is usually used as a proxy to economic inequality. The Gini index is a measure “of the deviation of the distribution of income among individuals or households within a country from a perfectly equal distribution,” according to the United Nations Development Programme.62 In the index, a society that scores 0.0 on the Gini scale has perfect equality in income distribution; therefore, the higher the number over zero, the higher the inequality; 1 represents the maximal inequality among values. In this study, the lowest value was for Tlaxcala (0.407), while the highest and therefore the most socially unequal was Zacatecas (0.500). Further, in order to assess the impact of economic development in Mexico's states, we included the average mean of each state's gross domestic product (GDP) per capita, between 2010 and 2014, according to the INEGI. The GDP per capita is often used as a proxy to explore the relation between the homicide rate in the general population and economic development.63 This measurement ranges from 47,707 Mexican pesos ($3,807 US) in Chiapas to 767,226 Mexican pesos ($55,816 US) in the state of Campeche.

Dependent Variable

In order to explore the validity of our six hypotheses, the dependent variable was created by using data from news safety organizations on journalist killings in Mexico's states from 2010 to 2015. However, as other studies on the matter recognize, defining who is a journalist has been a particularly difficult task among news safety organizations and scholars. Nick Mosdell acknowledges this issue in his analysis of the patterns of journalists killed worldwide: “Fundamental to all, though, is the fact that a definition of a traditional journalist is difficult to pin down, particularly when trying to decide if that person was killed as a direct result of their work or for other reasons that may be directly or indirectly associated with it.”64 However, in order to mitigate such identification problems, in this article Riddick's65 and Lucie Collinson's66 suggestions are followed to collect the cases that are only included in more than one national or international database. In these regards, three independent and internationally recognized databases are relied on to identify homicides of journalists in Mexico from 2010 to 2015. The criteria required that a death should appear in two or more data sources to be included in the analysis. As such, data on the murders of journalists were systematically collated from three international databases, from 2010 to 2015:67 the Committee to Protect Journalists,68Article 19 Mexico,69 and the Infoamerica project.70 The collection did not include media workers or citizen journalists, and was just concentrated on deaths of journalists, thus did not include cases of missing journalists. Cases were therefore only included where they were mentioned in two or more of these three databases.71

This analysis identified forty-one violent deaths of journalists in Mexico from 2010 to 2015 that met our parameters. This compares with the thirty-five cases identified by the CPJ (excluding media workers), forty-one cases identified by Article 19 Mexico, and seventy-three cases identified by the Infoamerica project.72 The major reason for exclusion was that the journalist's death was only recorded in one of the three databases. From the forty-one cases included, six states comprised 72 percent of all cases, namely: Veracruz (29 percent); Oaxaca (12 percent); Tamaulipas (9 percent); and Chihuahua, Guerrero, and Sinaloa (with 7 percent each).

Further, all forty-one journalists murdered were Mexican journalists. Eighty-two percent used to work for a local print or online newspaper, 12 percent for a national media outlet (including national correspondents), and 5 percent were freelancers. Sixty-three percent were print journalists, 25 percent worked for a radio or TV station, and 20 percent worked for online news media. Fifty-six percent covered crime beats, 41 percent covered local politics, and 12 percent covered human rights abuses. An additional 12 percent used to work in other sources (including culture and sports).

Data Analysis

As the international literature suggests, violent homicides of journalists represent some methodological challenges,73 as we have a problem with the large number of zeros in the row data (in our case eighteen states do not present any murders of journalists during the period of analysis). Further, another methodological issue is raised when we acknowledge that the vast majority of states where journalists were murdered (eleven of the fourteen states) presented very low numbers of fatalities (between one and three casualties per state in the period of analysis). Because of this, and inspired by the international literature, we decided to analyze the data using two methods. First, following Asal et al. and Riddick et al., we used a binary logit analysis with robust standard errors, with the dependent variable coded as one if a journalist was killed in a state, and zero if not. Thus, we analyzed the relationship between the occurrence of violent homicides of journalists (any homicide versus no homicide) per state and our six independent variables (see Table 2).

Second, following Riddick et al. and Collinson et al., we calculated the log of the homicide of journalists per state and created a six-year rate for homicide of journalists. In this sense, as Patrick M. Krueger mentioned, the ideal denominator for this six-year rate would have been the number of media workers in each subnational state in the years examined;74 however, as has been mentioned elsewhere, that data is not readily accessible.75 In order to mitigate such a problem, the second best option was to control the absolute number of journalists killed in Mexico with subnational population size, as is also suggested by some authors.76 By doing this, a six-year rate of homicide of journalists in Mexico's states per 10,000 people was created for this study (see Table 2). Then, we conducted additional analysis between the log of the six-year rate of homicide of journalists and our six independent variables using Poisson regression models with robust standard errors.

Thus, logistic regression analyses were used to explore the relationship between the occurrence of any homicide of journalists in a state and the six hypotheses mentioned previously. As is shown in Table 3, in the logistic regression analysis the relationship between our dependent variable and the following three independent variables were significant: intentional homicide per population (p < 0.083), internal forced displacement (p < 0.026), and violations of human rights abuses (p < 0.024). However, there was no significance association identified between the dependent variable and the rest of the variables (the index of democratic development (p < 0.149), the economic development indicator (p < 0.519), and the economic inequality measure (p < 0.289). Thus, the logistic regression analyses provide rather weak evidence to support the first three hypotheses.

Further, the results of the Poisson regression analyses identified a highly statistical significant (all at p < 0.001) association between the log of the state homicide rate and the following four variables: intentional homicide per population, internal forced displacement, violations of human rights, and the index of democratic development (see Table 3). Further, the Poisson regression analysis between the log of the journalists' homicides and the economic inequality measurement was also significant (p < 0.041). However, there was no significant relationship identified between the six-year rate and the economic development indicator (p < 0.199). Thus, overall, the Poisson regression analysis provides some evidence to support the first five hypotheses, but does not provide any support for the last one.

Discussion

Relationship between journalists' risk of being murdered and societal violence

According to the aforementioned results, we find evidence to support the first hypothesis, which suggests that journalists are at greatest risk of being murdered in subnational polities with higher levels of violent deaths per population, as has been widely suggested by the international literature.77 However, even when there is a clear correlation between these two variables, the results in at least one of the regressions is rather weak. Nevertheless, the association between the log of the six-year rate of homicide of journalists and the violent deaths per state is highly significant. Furthermore, in the period of analysis of this article, between 2010 and 2015, the association between lethal violence against journalists and violent deaths may be exemplified in some Mexican states, such as Chihuahua, Guerrero, or Sinaloa, where at least three journalists were killed per state, and which presented more than twenty violent homicides per 100,000 people. Thus, as has been referenced by recent literature,78 one plausible explanation is that the association between the increasing number of journalists killed is related to the criminal violence derived from the frontal strategy launched by the government against drug cartels one decade ago. However, in contrast to some authors79 who have argued that criminal violence between drug cartels cells is the main predictor for the high rate of journalists killed in Mexico's regions, it is pertinent to consider, although not very often recognized in the academic literature, that not all journalists killed in Mexico were murdered during disputes between cartels, or between cartels and the government. In fact, news safety organizations working in Mexico have highlighted80 the fact that some of the journalists killed between 2010 and 2015 were targeted with political violence (not just criminal violence). Indeed, between 2010 and 2015, the CPJ has documented some cases of journalists killed81 in which the suspected perpetrators involved government officials and local police forces.

Relationship between journalists' risk of being murdered and severe violations of human rights by the state

Furthermore, according to our results, we find some support for the second hypothesis, which suggests that journalists' risk of being murdered increases in states with violations of human rights perpetuated by the state. According to the results, the risk of being targeted with murder increases by 1.7 times in states with higher numbers of human rights violations by the federal security forces. Further, the association between the violent killings of journalists and the number of human rights violations perpetrated by public security bodies adds some evidence to the idea that antipress violence is, at least, related to the uncheckable performance of the military, the navy and other federal security forces in some Mexican states. This idea is often referred to in news safety reports82 but has barely been explored in the academic literature, and deserves further analysis. Overall, these results are in concordance with the latest report published by Mexico's National Human Rights Commission in which the autonomous body raised awareness in regards to the association between antipress violence and a general climate of human rights violations in Mexico—in fact, the Commission raised awareness in states like Guerrero, Tamaulipas, and Veracruz,83 where almost the half of the forty-one journalists killed in Mexico between 2010 and 2015 were concentrated, according to the data compiled in this article.

Relationship between journalists' risk of being murdered and internal conflict

Our results provide evidence to suggest that journalists are at greatest risk of being murdered in subnational polities with internal conflict. Indeed, this relationship proved to be statistically significant in both the logistic regression and the Poisson regression models. According to these results, a lethal attack against journalists is seven times more likely to happen in subnational polities with cases of internal forced displacement documented by the CNDH, between 2014 and 2015. Furthermore, following the information reported by the CNDH, the association between forced displacement and the killing of journalists may be exemplified in states like Tamaulipas and Guerrero. Thus, these two states presented both a high number of journalists killed (according to the six-year rate of journalists killed), on the one hand, and dozens of forced displacement cases documented by the CNDH, on the other. Further, many of these forced displacements in Mexico's states, as the Commission suggests, were the product of the extreme forms of violence and abuses perpetuated by drug cartels (sometimes in collusion with state or municipal security forces) between 2010 and 2015. In fact, there is exhaustive evidence84 that the frontal strategy against drug cartels launched by the Mexican government in 2007 was an important catalyst for the intensification of violence between criminal organizations, especially in geographical areas holding a high concentration of and rivalry between cartel cells.85 However, we suggest that the association between journalists' violent murders and the presence of internal forced displacement tested in this article cannot be generalized over time and space among Mexico's states. In fact we believe further research is recommended, especially to investigate the relationship between antipress violence and other measurements of internal conflict and/or indices of state failure at a subnational level.

Relationship between journalists' risk of being murdered and democratic development

Further, the previously mentioned results provide rather weak evidence of the relationship between the violent deaths of journalists and the degree of democratic development in Mexico's states. The results (especially the Poisson model) offer certain evidence suggesting that journalists are at greatest risk of being murdered in subnational polities with less democratic development. This association is exemplified by some of Mexico's states, like Guerrero, Veracruz, and Oaxaca, which, in the period of analysis, presented both the highest rates of violent homicides of journalists, on the one hand, and the lowest rates in the IDD-Mexico, on the other. Furthermore, even when we acknowledge that democracies are not a “safe haven for journalists,” as Asal et al. have demonstrated, it is important to mention that our results also suggest that journalists' risk of being murdered decreased in 30 percent of those states with better performance in the IDD-Mexico. One way to read these findings is by suggesting that a better performance in democratic institutions may play a role in providing a safer atmosphere for local journalism (rather than states with low democratic development). However, we acknowledge that these findings remain in an exploratory phase. Indeed, as in all the previously mentioned hypotheses, further research is needed in order to better explore to what extent a better performance of democratic institutions, or the quality of local democratic government, may have a positive impact on journalists' safety.

Relationship between journalists' risk of being murdered and economic indicators

Finally, according to the results, there is some evidence (rather weak) suggesting that the higher the economic inequality in the state, the higher the rate of violent homicides of journalists. Thus, the results offer some evidence to our fifth hypothesis, indicating that journalists are at greatest risk of being murdered in subnational polities with more levels of economic inequality. This trend is exemplified by states that have high rates of journalists killed, such as Veracruz (twelve journalists killed), Oaxaca (five), and Guerrero (three), and where social inequality is particularly rife (all these three states present between 0.46 and 0.49 rate according to the GINI coefficient). Indeed, the findings suggest that in developed societies the material conditions in which journalists worked need to be considered in order to better understand the occurrences of antipress violence.86

Further, neither the logistic regression nor the Poisson regression model provided any evidence to validate the sixth hypothesis, which states that journalists are at greatest risk of being murdered in subnational polities with low economic development. In fact, the results show that journalists' risks of being murdered neither increases nor decreases in regards to the degree of economic development per capita per state. However, we suggest more research exploring this relationship is needed. It may be especially useful, we believe, to not only test other indicators associated with economic performance, poverty, and inequality, but also to shift the scale of analysis and research to the association of antipress violence and journalists' income and working conditions at a subnational level.

Limitations of the Study

This study presents some limitations. First, we are reliant almost entirely on the accuracy of the databases used to build our dependent variable. However, even when we tried to mitigate this issue using a range of three databases (including some of the most respected in the academic community), further research may include other selection strategies—such as adding more databases (including governmental information), or (wherever possible) controlling for population of journalists, rather than for state population. Second, we acknowledge some limitations related to the issue of time. As such, as antipress violence is a contextual and contingent phenomenon, the rate of violent murders of journalists is not a stable and consistent measurement over time.87 So, for instance, by choosing a different period of analysis we would have included different states. Thus, further research may include additional controls to deal with the time issue, both in the construction of the dependent variable (i.e., adding a time control per year), and in the period of analysis (i.e., adding more years to the analysis). Third, this research does not include any variable to explore the relationship between the violent deaths of journalists and certain aspects of the journalistic field that may increase the lethal danger to which journalists are exposed in subnational media systems (i.e., years of experience working on hazardous reporting, type of newsbeat, type of news media—local versus national—and organizational support and training to cover violence, among others). Further research could integrate some of these “field” dimensions into the analysis. Finally, this research explored the relationship between certain social determinants with the violent homicides of journalists; however, we do not advance a model to test all these variables at once, as we believe more sophisticated research designs may advance more robust explanations around this issue.

Conclusion: Theoretical and Policy Implications

Why are some subnational states more dangerous for journalists than others? In line with the international literature, this article finds some evidence suggesting that Mexican journalists are more likely to be lethally injured in those states with high levels of social violence, internal conflict, severe violations of human rights, and low democratic development. Thus, this article offers one of the first systematic analyses assessing the association between some social determinants and the killing of journalists in one single country. However, as we acknowledge in the body of this article, these findings—which demonstrate coherence with the proposed theoretical framework explained at the beginning—just offered a first exploratory assessment of this issue and, as discussed, need to be tested again. Indeed, we think further research may assess the impact of the social determinants mentioned here and antipress violence not only within Mexico's states but also in other scenarios. That being said, this article offers both theoretical and policy implications, which we will develop in the conclusion.

In the theoretical dimension, first, this article offers a systematic overview of the kinds of hypotheses that could be relevant when exploring the association between the lethal injury of journalists and social determinants. Specifically, we think that the recognition of certain dimensions not often mentioned (and assessed) in this kind of research constitutes, itself, an academic contribution (just like the impact of human rights violations by the state, democratic development, and economic inequality fostering antipress violence). In fact, we suggest that the inclusion of economic inequality variables in our theoretical framework is a particular improvement in our understanding of antipress violence. This is especially true within very unequal societies in Latin America, like Brazil or Central America, where are trends and patterns of journalists killing similar to the Mexican case.

Second, this article also calls for the theoretical inclusion (and methodological testing) of comparisons beyond the national state, between regions, states, or provinces, to explore issues related to media policy, press freedom, and antipress violence. In fact, we believe the subnational comparative approach in media, communication, and journalism studies remains underdeveloped and, as we suggested in this article, offers (potentially) strong explanatory power that other approaches often overlook (for instance, those approaches that take the national state as a default category for analysis). Indeed, we think this theoretical framework and research design may inform subnational comparisons in other regions and states, especially in very diverse, unequal, and heterogeneous countries, which experience a systematic and unprecedented cascade of antipress violence.

The third and final theoretical remark states that, under certain conditions, new democracies seem to offer a lethal trap to journalists, who enjoy certain incentives to perform more critically than under previous authoritarian regimes, but where the state is unable to safeguard the exercise of journalism all over its territory. This idea, just very recently explored in the literature, acquires relevance when we try to explore the high rates of lethal homicides of journalists in post-authoritarian societies with democratic institutions still under construction (or captured by authoritarian incumbents), like Russia, Pakistan, and Mexico (especially at a subnational level).

Further, the findings suggest that the issue of antipress violence and impunity mirrors a major problem in the functionality of the state and political institutions at large. Thus we believe that a major qualitative change in the functionality of the state and democracy may improve, considerably, the issue of safety of journalists and impunity. By saying this, we believe that any substantial development in journalists' safety and the ends of impunity against them is a consequence of a qualitative improvement in the protection of the general population, human integrity, and better exercise of the political and judicial institutions. Thus, we believe that violence against the press reflects, in part, the impossibility of any social group to escape the spiral of violence in society, as Waisbord notes. However, as Asal et al. suggest, even when certain features of democratic societies may encourage antipress violence, under certain circumstances democratic institutions may mitigate against them as well. As such, by focusing on the findings of this work, we propose some policy-oriented implications.

First, the results show that policy initiatives need to be rooted in the social, political, and civil society contexts at the local level, rather than traveling from the international experience to the community. In this sense, our findings suggest that any policy needs to consider not only the local social, political, and criminal environments—which encourage aggressions against the press—but also integrate further dimensions (like the quality of local democracy and civil society ties, among others) that, potentially, may offer a more protective framework for the exercise of journalism.

Second, this article suggests that more focused and specialized contingency plans and policy initiatives need to be implemented, mainly concerning the critical subnational states (those in which the major rates in violent homicides against the press are concentrated). In this sense, even when antipress violence is a rampant reality all over Mexican territory, our findings suggest that the violent homicides of journalists are concentrated in a few states. In those critical cases, major efforts conducted by different state-holders may raise awareness, create better contingency plans and, ultimately, provide a more robust protection apparatus for local journalists.

Third, we believe that better professionalization in the gathering of news and in organizational support and monitoring of special assignments may improve the safety of local journalists. Thus, the vast majority of local and national news media houses have abdicated their responsibility to protect their journalists. In a country where local journalists targeted with violence have experienced as many risks as war correspondents, news media organizations need to become more aware of the perils and dangers to which local journalists are exposed.

Footnotes

I would like to thank the two anonymous reviewers from the Journal of Information Policy for their useful suggestions and recommendations. Also, I would like to thank the journal editors, Amit Schejter and Krishna Jayakar, as well as Petros Iosifidis, Keren Tenenboim-Weinblatt, and Jairo Lugo-Ocando, for their thoughtful comments on an earlier version of this paper. This research was possible thanks to a grant from the Center for Global Communication Studies at the Annenberg School for Communication at University of Pennsylvania, and to a scholarship granted by the National Council of Science and Technology of Mexico (CONACYT) for doctoral studies abroad.

1.

Committee to Protect Journalists (CPJ), “Journalist Kidnapped.”

2.

The murders of journalists are the final and most extreme form of censorship, which just skims the surface of the dangers to which many local journalists are exposed in many post-authoritarian states or new democracies. From 2010 to 2015, for every journalist killed per year in Mexico there were at least thirty-five aggressions against the press. Article 19 Mexico.

3.

Asal et al.; Bjørnskov and Freytag.

4.

Reporters Without Borders; Del Palacio.

5.

The six states are: Chihuahua, Guerrero, Nuevo León, Sinaloa, Tamaulipas, and Veracruz.

6.

My research question is largely inspired by a recent study that asks: What countries are most dangerous for journalists? Asal et al.

7.

Asal et al.; Bjørnskov and Freytag; Cottle, Sambrook, and Mosdell; Lisosky and Henrichsen; Riddick et al.; Waisbord, “Antipress Violence,” “Democratic Journalism.”

8.

Moncada and Snyder; Snyder.

9.

Cottle.

10.

Tumber.

11.

Sambrook; Feinstein

12.

This dangerous situation becomes even worse when we consider that the impunity rate in all those cases is as high as 85 to 90 percent. CPJ, “Getting Away with Murder.”

13.

International News Safety Institute (INSI).

14.

According to the INSI, international conflicts or regular civil wars are “international armed conflicts involving two or more states, or national armed conflicts, where one of the participants may or may not be the internationally recognized sovereign power.” On the other hand, the Institute defines peacetime where “there is no internal conflict, but where there is persistent criminal or political violence and questions about the adequacy of existing legal and normative protection for freedom of speech and the safety of journalists and other news media staff.” INSI.

15.

CPJ, “Getting Away with Murder.”

16.

Asal et al.

17.

Kalyvas.

18.

Díaz-Nosty; Foerstel; Lisosky and Henrichsen; Riddick et al.; Saul; Taback and Coupland; Waisbord, “Antipress Violence,” “Democratic Journalism.”

19.

Waisbord, “Antipress Violence,” 93.

20.

Collinson, Wilson, and Thomson; Holland and Rios.

21.

Kaldor.

22.

Asal et al.; Riddick et al.; Taback and Coupland; Waisbord, “Democratic Journalism.”

23.

Asal et al., 14.

24.

Loc. cit.

25.

Kalyvas. According to the Article 3 of the Geneva Conventions of 1949, internal conflict (or non-international armed conflict) includes those armed conflicts “between governmental forces and non-governmental armed groups, or between such groups only.” International Committee of the Red Cross.

26.

Sambrook.

27.

Waisbord, “Democratic Journalism,” 119.

28.

Riddick et al., 686.

29.

Holland and Rios.

30.

Diamond.

31.

In this article we refer to “government quality” as the “levels of corruption and the state's ability to impose law and order.” Asal et al.

32.

Hughes and Lawson, 10.

33.

Asal et al.; Bjørnskov and Freytag; Riddick et al.

34.

Asal et al., 14.

35.

Durazo Herrmann.

36.

Waisbord, “Antipress Violence.”

37.

Krueger.

38.

Boas; Díaz-Nosty; Hughes et al.; Lugo.

39.

O'Donnell.

40.

Hughes et al.

41.

Krueger.

42.

The literature suggests that even when lethal attacks on journalists (and antipress violence in general) also occurred during the single-party regime, this type of lethal aggression did not constitute a common threat to the press. In fact, as in other closed regimes, Mexican autocrats preferred co-optation and clientelism as the main form of press-control. Hallin.

43.

Article 19 Mexico.

44.

Estévez; Farah.

45.

When it comes to crimes against journalists, following the CPJ, 89 percent of the homicides committed since 1992, specifically against journalists as a result of their jobs, have seen the killers go unpunished. The high impunity rate remains despite the fact that in 2013, federal authorities created the Special Prosecutor's Office for Crimes Committed against Freedom of Expression, which allows federal jurisdiction to prosecute crimes against freedom of expression in the entire country.

46.

Schedler.

47.

Márquez; Del Palacio; Hughes et al.; Relly and González de Bustamante; Hernández Ramírez;

48.

González de Bustamante and Relly.

49.

Thus, it is recognized that the type of news and the journalistic profile (local vs. national) have an association with the increasing lethal risk to which journalists are exposed. However, in this article only the social variables (or the exogenous factors to the field of journalism) are analyzed.

50.

Hughes and Lawson; Waisbord, “Antipress Violence.”

51.

Gibson; Moncada and Snyder; Snyder.

52.

Krueger.

53.

According to the Spearman's correlation, none of our independent variables present a high correlation within them, or between them and our dependent variables.

54.

Collinson, Wilson, and Thomson.

55.

All information about human rights violations perpetuated by public security forces were gathered from the Sistema Nacional de Alerta de Violaciones a los Derechos Humanos (National System on Human Rights Violations). Comisión Nacional de los Derechos Humanos (CNDH), “Indicadores por sector seguridad.”

56.

CNDH, Informe Especial.

57.

Bayefsky and Fitzpatrick.

58.

Cantor.

59.

According to the CNDH, the states with any case of internal forced displacement between 2014 and 2015 were the following: Chiapas, Chihuahua, Durango, Guerrero, Michoacán, Oaxaca, Sinaloa, Tamaulipas, and Veracruz.

60.

Among the organizations that participated in the yearly elaboration and support of this index are the following: El Colegio de Mexico, the Employers Confederation of the Mexican Republic (COPARMEX), the Konrad Adenauer foundation, and the political consultant firm Polilat.

61.

All yearly reports and details are available on the IDD-Mex website. IDD-Mexico.

62.

United Nations Development Programme (UNDP).

63.

Butchart and Engström.

64.

Mosdell, 52. The main barriers in regards to identification rely on many factors, including but not restricted to the following three: (1) enough reliable information about the circumstances of the homicide (especially at the community level); (2) the thin, porous and very blurred line between activists and journalists; and, increasingly, (3) the wide use of digital technology which allows anybody with a smart phone to become a citizen journalist.

65.

Riddick et al.

66.

Collinson, Wilson, and Thomson.

67.

Even when we have access to information for more than six years, we concentrate on this period as we face limitations in the availability of data to our independent variables.

68.

CPJ, “Journalists Killed.”

69.

Article 19 is a British human rights organisation, based in London, with a specific mandate and focus on the defense and promotion of freedom of expression and freedom of information worldwide. The information used in this article came from this office in Mexico City. Article 19 Mexico.

70.

The Infoamerica Project is an independent website financed by UNESCO and administrated by the Universidad de Málaga (Spain) that is devoted to the analysis of communication process throughout the Latin American region. Infoamerica,

71.

Official information about the violent deaths of journalists is not included in this article because the personal details (name, place, and date of death, etc.) of journalistic casualties are not made public by the governmental institutions in charge to collect this information. So, it is almost impossible to compare official information with other databases and, ultimately, to determine to what extent those casualties meet our requirements.

72.

The seventy-three cases identified by the Infoamerica Project also include missing or disappeared journalists.

73.

Asal et al.

74.

Krueger.

75.

Riddick et al.; Krueger; Taback and Coupland.

76.

Bjørnskov and Freytag.

77.

Díaz-Nosty; Foerstel; Lisosky and Henrichsen; Waisbord, “Antipress Violence.”

78.

Estévez; González de Bustamante and Relly.

79.

Holland and Ríos.

80.

Article 19 Mexico.

81.

According to the CPJ, there is documented evidence of the implication of local government officials and local security forces in the violent killings of journalists during our period of analysis. Some of those casualties include journalists Moises Sanchez Cerezo (killed in 2015) and Regina Martinez (killed in 2012) in Veracruz, as well as Humberto Millan (killed in 2011) in Sinaloa, among others.

82.

According to Article 19 Mexico, in 2012, the military, the army, and the police were directly implicated in around the half of all attacks against the press in Mexico. Lakhani.

83.

CNDH, “Recomendación general No. 24.”

84.

Osorio.

85.

Holland and Rios; Osorio.

86.

Hughes et al.

87.

Díaz-Nosty; Garza.

Bibliography

Article 19 Mexico
. “Miedo,” edited by Article 19 Mexico. Mexico City, 2016.
Asal, Victor, Matthew Krain, Amanda Murdie, and Brandon Kennedy. “Killing the Messenger: Regime Type as a Determinant of Journalist Killing, 1992–2008.” Foreign Policy Analysis (2016).
Bayefsky, Anne Fruma, and Joan Fitzpatrick. Human Rights and Forced Displacement. Vol. 4. The Hague, Netherlands: Martinus Nijhoff Publishers, 2000.
Bjørnskov, Christian, and Andreas Freytag. “An Offer You Can't Refuse: Murdering Journalists as an Enforcement Mechanism of Corrupt Deals.” Public Choice 167, no. 3 (2016): 221–43. Available at SSRN 1790384.
Boas, Taylor C. “Mass Media and Politics in Latin America.” In Constructing Democratic Governance in Latin America, edited by Jorge I. Domínguez and Michael Shifter, 48–77. Baltimore: Johns Hopkins University Press, 2013.
Butchart, Alexander, and Karin Engström, “Sex-and Age-Specific Relations between Economic Development, Economic Inequality and Homicide Rates in People Aged 0–24 Years: A Cross-Sectional Analysis.” Bulletin of the World Health Organization 80, no. 10 (2002).
Cantor, David James. “The New Wave: Forced Displacement Caused by Organized Crime in Central America and Mexico.” Refugee Survey Quarterly 33, no. 3 (2014): 34–68. hdu008.
Collinson, Lucie, Nick Wilson, and George Thomson. “Violent Deaths of Media Workers Associated with Conflict in Iraq, 2003–2012.” PeerJ 2 (2014): e390.
Comisión Nacional de los Derechos Humanos (CNDH)
, “Indicadores por Sector Seguridad.” México City, 2016. Accessed December 10, 2016. http://appweb.cndh.org.mx/SNA/ind_Sector_SM.asp?Id_Sec=2.
Comisión Nacional de los Derechos Humanos (CNDH)
. Informe Especial sobre Desplazamiento Forzado Interno. México City, 2016 Accessed December 10, 2016. http://www.cndh.org.mx/sites/all/doc/Informes/Especiales/2016_IE_Desplazados.pdf.
Comisión Nacional de los Derechos Humanos (CNDH)
. “Recomendación general No. 24. Sobre el ejercicio de la libertad de expresión en México.” México City, 2016. Accessed December 10, 2016. http://www.cndh.org.mx/sites/all/doc/Recomendaciones/generales/RecGral_024.pdf.
Committee to Protect Journalists (CPJ)
. “Getting Away with Murder.” News release, October 8, 2015. Accessed December 10, 2016. https://cpj.org/reports/2015/10/impunity-index-getting-away-with-murder.php.
Committee to Protect Journalists (CPJ)
. “Journalist Kidnapped from His Home in Mexico.” News release, January 5, 2015. Accessed December 10, 2016. https://cpj.org/2015/01/journalist-kidnapped-from-his-home-in-mexico.php.
Committee to Protect Journalists (CPJ)
. “Journalists Killed in Mexico.” New York, 2016. Accessed December 10, 2016. https://cpj.org/killed/americas/mexico/.
Cottle, Simon. “On the Violent History of the Globalised Present.” In Cottle, Sambrook, and Mosdell, Reporting Dangerously, 61–87.
Cottle, Simon, Richard Sambrook, and Nick Mosdell, editors. Reporting Dangerously: Journalist Killings, Intimidation and Security. London: Springer, 2016.
Del Palacio, Celia. “Periodismo impreso, poderes y violencia en Veracruz 2010–2014. Estrategias de control de la información.” Comunicación y Sociedad 24 (2015): 19–46.
Diamond, Larry. Developing Democracy: Toward Consolidation. Baltimore: JHU Press, 1999.
Díaz-Nosty, Bernardo. “Medio Siglo De Agresiones a Periodistas. Síntesis Estadística De Asesinatos Y Desapariciones En América Latina (1970–2015).” Contratexto 24 (2016): 171–201.
Durazo Herrmann, Julián. “Media and Subnational Democracy: The Case of Bahia, Brazil.” Democratization 24, no. 1 (2015): 81–99.
Estévez, Dolia. “Protecting Press Freedom in an Environment of Violence and Impunity.” In Shared Responsibility: US-Mexico Policy Options for Confronting Organized Crime, edited by Eric L. Olson, David A. Shirk, and Andrew D. Selee, 271–90. Washington, DC: Woodrow Wilson International Center for Scholars, 2010.
Farah, Douglas. Dangerous Work: Violence against Mexico's Journalists and Lessons from Colombia. Washington, DC: Center for International Media Assistance, 2012.
Feinstein, Anthony. “Mexican journalists and journalists covering war: a comparison of psychological wellbeing”, Journal of Aggression, Conflict and Peace Research, 5, (2013): 77–85.
Foerstel, Herbert N. Killing the Messenger: Journalists at Risk in Modern Warfare. Westport, CT: Greenwood Publishing Group, 2006.
Garza, Javier. “Violencia Contra Periodistas En México: De Norte a Sur.” El Pais,
July
7
, 2015.
Gibson, Edward L. Boundary Control: Subnational Authoritarianism in Federal Democracies. New York: Cambridge University Press, 2013.
González de Bustamante, Celeste, and Jeannine E. Relly. “The Practice and Study of Journalism in Zones of Violence in Latin America: Mexico as a Case Study.” Journal of Applied Journalism & Media Studies 5, no. 1 (2016): 51–69.
Hernández, Maria Elena. “Periodismo regional en México y violencia del crimen organizado: un diálogo académico.” Comunicación y Sociedad 27 (2016): 281–294.
Hallin, Daniel. “Media, Political Power, and Democratization in Mexico.” In De-Westernizing Media Studies, edited by James Curran and Myung-Jin Park, 97–110. Abingdon: Routledge, 2000.
Holland, Bradley E., and Viridiana Rios. “Informally Governing Information How Criminal Rivalry Leads to Violence against the Press in Mexico.” Journal of Conflict Resolution (2015): 1–25. 0022002715600756.
Hughes, Sallie, Miguel Garcés, Mireya Márquez-Ramírez, and Jesús Arroyave. “Rethinking Professional Autonomy: Autonomy to Develop and to Publish News in Mexico and Colombia.” Journalism (2016): 1464884916659409.
Hughes, Sallie, and Chappell Lawson, “The Barriers to Media Opening in Latin America.” Political Communication, 22, no. 1 (2005): 9–25.
Index of Democratic Development México (IDD-Mexico)
. Accessed December 10, 2016. http://www.idd-mex.org/2015/.
Infoamerica
. “Periodistas muertos y desaparecidos, México.” Málaga, 2016. Accessed December 10, 2016. http://www.infoamerica.org/libex/muertes/atentados_mx.htm.
International Committee of the Red Cross
. “How is the Term ‘Armed Conflict’ Defined in International Humanitarian Law?” Geneva, 2008. Accessed December 10, 2016.
International News Safety Institute (INSI)
. “Killing the Messanger.” UK: Safety Institute into the Protection of Journalists, 2007.
Kaldor, Mary. Human Security. Cambridge, UK: Polity, 2007.
Kalyvas, Stathis N. The Logic of Violence in Civil War. Cambridge, UK: Cambridge University Press, 2006.
Krueger, Patrick M. “The Social and Political Functioning of States and the Homicide of Media Workers.” Journal of Epidemiology and Community Health 62, no. 8 (2008): 666–67.
Lakhani, Nina. “Mexican Journalists Targeted.” Al Jazeera,
August
2
, 2013. Accessed December 10, 2016. http://www.aljazeera.com/indepth/features/2013/07/20137299103914323.html.
Lisosky, Joanne M, and Jennifer Henrichsen. “Don't Shoot the Messenger: Prospects for Protecting Journalists in Conflict Situations.” Media, War & Conflict 2, no. 2 (2009): 129–48.
Lugo, Jairo. The Media in Latin America. Maidenhead, UK: McGraw-Hill, 2008.
Márquez, Mireya. “Estudio introductorio.” In Violencia Y Periodismo Regional En México. México: Juan Pablos, 2015.
Moncada, Eduardo, and Richard Snyder. “Subnational Comparative Research on Democracy: Taking Stock and Looking Forward.” Comparative Democratization Newsletter. American Political Science Association 10, no. 1 (2012).
Mosdell, Nick. “Mapping the Parameters of Peril.” In Cottle, Sambrook, and Mosdell, Reporting Dangerously, 36–58.
O'Donnell, Guillermo. “On the State, Democratization and Some Conceptual Problems: A Latin American View with Glances at Some Postcommunist Countries.” World Development 21, no. 8 (1993): 1355–69.
Osorio, Javier. “The Contagion of Drug Violence Spatiotemporal Dynamics of the Mexican War on Drugs.” Journal of Conflict Resolution 59, no. 8 (2015): 1403–32.
Relly, Jeannine E., and Celeste González de Bustamante. “Silencing Mexico: A Study of Influences on Journalists in the Northern States.” The International Journal of Press/Politics 19, no. 1 (2014): 108–31.
Reporters Without Borders
. “Veracruz: Los periodistas frente al Estado de miedo.” edited by
Reporters Without Borders Mexico
. Mexico City 2017. Accessed February 13, 2017. http://www.rsf-es.org/news/mexico-veracruz-los-periodistas-frente-al-estado-de-miedo-nuevo-informe-de-reporteros-sin-fronteras/
Riddick, L., George Thomson, N. Wilson, and G. Purdie. “Killing the Canary: The International Epidemiology of the Homicide of Media Workers.” Journal of Epidemiology and Community Health 62, no. 8 (2008): 682–88.
Sambrook, Richard. “Reporting in Uncivil Societies and Why It Matters.” In Cottle, Sambrook, and Mosdell, Reporting Dangerously, 17–35.
Saul, Ben. “The International Protection of Journalists in Armed Conflict and Other Violent Situations.” Australian Journal of Human Rights 14, no. 1 (2008): 99–140.
Schedler, Andreas. “The Criminal Subversion of Mexican Democracy.” Journal of Democracy 25, no. 1 (2014): 5–18.
Snyder, Richard. “Scaling Down: The Subnational Comparative Method.” Studies in Comparative International Development 36, no. 1 (2001): 93–110.
Taback, Nathan, and Robin Coupland. “Security of Journalists: Making the Case for Modelling Armed Violence as a Means to Promote Human Security.” In Thinking Outside the Box in Multilateral Disarmament and Arms Control Negotiations, 191–206. Geneva: UN Institute for Disarmament Research, 2006.
Tumber, Howard. “The Fear of Living Dangerously: Journalists Who Report on Conflict.” International Relations 20, no. 4 (2006): 439–51.
United Nations Development Programme (UNDP)
. “Income Gini Coefficient.” Accessed December 10, 2016. http://hdr.undp.org/en/content/income-gini-coefficient.
Waisbord, Silvio. “Antipress Violence and the Crisis of the State.” The Harvard International Journal of Press/Politics 7, no. 3 (2002): 90–109.
Waisbord, Silvio. “Democratic Journalism and ‘Statelessness.’” Political Communication 24, no. 2 (2007): 115–29.
This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.