Latin American Paper

The Criminal Subver Sion of m exi C an Demo C ra C y Andreas Schedler Andreas Schedler is professor of political science at CIDE in Mexico City. His most recent book is The Politics of Uncertainty: Sustaining and Subverting Electoral Authoritarianism (2013). At present, he is conducting public-opinion research on organized violence in Mexico.

O nce a century, it seems, Mexico stumbles into dramatic encounters with collective violence. The war of independence between 1810 and 1821 left around 200,000 dead, and the Mexican Revolution from 1910 to 1917 no fewer than a million. 1 Today, after decades of relative authoritarian peace and only two democratic presidencies, the country finds itself immersed in yet another epidemic of violence. In the 2000 presidential balloting,\� the victory of opposition candidate Vicente Fox of the conservative National Action Party (PAN) capped a long process of democratization by electio\�ns and ended seven straight decades of hegemonic rule by the Institutional Revolutionary Party (PRI). Yet even as Mexico’s fledgling democracy\� has been struggling to find its way, the country has slid—at first imperc\�eptibly, then dramatically—into civil strife. It has suffered a pandemic escal\�ation of violence related to organized crime. In 2006, after a close and contentious election, PAN’s Felipe Calderó\�n assumed the presidency amid a lingering security crisis. During Fox’s\� term in office, violent competition among drug-trafficking organizations\� (so-called cartels) had been provoking more than a thousand homicides per year, and the number was rising. Although it had not been an issue during the election campaign, President Calderón decided to make the fight against drug cartels the defining policy of his presidency, only to see that fight turn into his term’s defining failure. Relying heavily\� on the use of military force, Calderón intensified the unbalanced strategies\� that his predecessors had already tried. These approaches included bolstering\� the security apparatus without strengthening the justice system; drawing the military into police work without subjecting it to oversight; chasin\�g Journal of Democracy Volume 25, Number 1 January 2014 © 2014 National Endowment for Democracy and Johns Hopkins University \�Press 6Journal of Democracy down cartel leaders without dismantling cartel networks; pursuing drug trafficking while giving traffickers a license to kill one another; arre\�sting numerous suspects without being able to try them fairly and effectively;\� and seeking mass confiscations of drug money and arms while lacking serious strategies to stop money laundering and arms imports.

Policy incoherence permitted the lingering violence to become worse, both qualitatively and quantitatively. In qualitative terms, modes of assassination moved toward demonstrative cruelty, routinized and ritualized. In certain parts of the country, the public display of tortured, dismembered, and decapitated bodies became a regular feature of daily life. In quantitative terms, the number of annual homicides attributed t\�o criminal organizations shot up from around 2,200 in 2006 to more than 16,600 in 2011. In 2012, drug-related homicides declined for the first time since 2001, albeit remaining at a level (nearly 14,000) many time\�s higher than in the early 2000s. We do not yet know, of course, whether the 2012 dip constitutes the beginning of a trend. Moreover, the prob- lems that cluster around the task of compiling accurate data on the vio- lence are massive. Thousands of people have “disappeared” after be\�ing abducted. According to official figures, more than 26,000 individuals were reported “missing” during the Calderón years. 2 Sources: For 2001–2006: Attorney General’s Office, cited in Marcos Pablo M\�oloeznik, “Militarizing Mexico’s Public Security,” CHDS Regional Insights 11 (15 February 2009).

For 2007–2010: Presidency of the Republic, “Dataset of Deaths by Presumptive Criminal Rivalry.” For January–September 2011: Attorney General’s Office\�, “Dataset of Deaths by Presumptive Criminal Rivalry” (www.pgr.gob.mx). For October 2011–December 2012: Lan- tia Consultores, “Dataset of Violence of Organized Crime” (www.lantiaconsultores.com). Number of “Executions” Year Figure —A nnu Al n umber o F H omicides A ttributed to o rg Anized c rime in m exico , 2000–12 7 Andreas Schedler When confrontations between armed groups within a state cause more than a thousand “battle-related deaths” per year, academics s\�peak of “civil war.” At least since 2001, democratic Mexico has experie\�nced levels of “internal war” that surpass this conventional threshold.\� Yet the war is not one but many. Its major lines of conflict run between criminal enterprises. Many, perhaps most, acts of private coercion are hostile acts within a multilateral war among competing cartels. The Calderón administration routinely attributed 90 percent of drug-related assassina- tions to “score-settling” among criminal organizations. This figur\�e was merely impressionistic, not to say propagandistic. Only 10 percent of victims are innocent, it said; the rest are guilty. As a rule, their cas\�es have not led to prosecutions. While the so-called drug war entails various interacting “nonstate”\� conflicts, it also contains elements of “one-sided” violence that \�crimi- nals unleash against civilians. Profit-oriented participation in illicit\� mar- kets forms only a portion of organized crime’s activity. The drug car\�- tels are also massively engaged in predatory crimes involving unilateral\� violence against civilians. Organized homicides have only been the tip of the violent iceberg. As criminal organizations have diversified their\� activities, the country has seen the dramatic expansion of kidnapping, human trafficking, and extortion (mafia-like protection rackets). In a\�d- dition, insofar as the cartels wage a guerrilla war against state agents\�, they participate in a kind of criminal insurgency. In recent years, we have seen a constant stream of attacks against the state, such as the ki\�d- napping, torture, and murder of security officials and assaults on police stations using hand grenades and heavy weapons. Thus the Mexican state is a warring party, too. In theory, it has a monopoly on the wielding of legitimate violence. In practice, it commits\� criminal violence on a large scale. International human-rights groups agree that security agents have perpetrated “widespread” human-rig\�hts violations. In part, these violations are expressions of state abuse. They are the unintended but inevitable consequence of acting with brute force, little actionable intelligence, and no oversight in an “irregular war” characterized by endemic problems of information. In part, illeg\�al state violence is a symptom of partial state collusion. Between January 2008 and November 2012, more than 2,500 police officers and more than 200 military personnel were murdered by criminal organizations. 3 Yet in numerous instances, public officials have collaborated with crim- inal organizations. Sources of Violence How has Mexico turned into a “violent democracy” within just a few\� short years? Some might say that there is no puzzle here, for Mexico’\�s plunge into societal violence has been a process of Latin American 8Journal of Democracy “normalization.” Today, the country’s annual homicide rate of 1\�8.6 per 100,000 inhabitants is fairly close to the regional average of 15.6. 4 Be- sides, violence is not generalized, but territorially concentrated at en\�- try and exit points and along the transport routes by which drugs move transnationally. The states along the U.S. border (Baja California, So- nora, Chihuahua, Coahuila, Nuevo León, and Tamaulipas) as well as some states along the Pacific coast (Sinaloa, Jalisco, Michoacán, and Guerrero) have been the main theaters of the drug war. In recent years, however, orga- nized violence has been spreading to more states and municipalities. Still, between 2009 and 2011, less than 5 percent of Mexico’s municipalities experienced extreme levels of deadly violence (defined as an average annu- al homicide level of 100 or more per 100,000 inhabitants). 5 Many con- clude, therefore, that the current security crisis is bad, but not that \�bad.

In comparative perspective, it looks like a medium-sized problem, not a big one, and large parts of the national territory remain completely cal\�m.

Accordingly, Mexican officials as well as citizens often complain that the crisis draws excessive attention from the international community. This tranquilizing reading depends, however, on what we are prepared to accept as “normal.” To begin with, there is the reality that less than a decade ago Mexico’s homicide rate was only about half its current “normal for the region” rate. And then there is the exceptional le\�vel of violence in the region as a whole. According to Moisés Naím’s e\�stimate, Latin America has just 8 percent of the world’s population but 42 per\�cent of all its homicides. 6 By widening the comparative frame from region to globe we can better appreciate the extraordinary level of societal viole\�nce in Mexico as well as in other Latin American countries such as Brazil, Colombia, Honduras, and Venezuela. And even if we were prepared to habituate ourselves to a new level of “structural” violence, we wo\�uld still want to explain its recent surge. Most explanations rely on two bundles of causes: material resources and actor dynamics. “Resource-focused” accounts note that the material sinews of war have become more readily available. These include: Money: The trade in illegal drugs is a lucrative business whose largest single market lies just to Mexico’s north, in the United Stat\�es.

This business creates the wealth that permits criminal “oligarchs”\� to organize and equip themselves for violence. While the history of drug production and trafficking in Mexico reaches back to the latter part of The demand shock in the international cocaine market is what made the war ignite; the structural availability of money, arms, and personnel is what has made it feasible; and the fragmentation of actors is what has made it escalate. 9 Andreas Schedler the nineteenth century, the market received a massive expansionary shock in the 1990s, when cocaine-trafficking routes shifted from the Caribbean to Mexico. Illicit wealth sustains the organization of vio- lence. Yet the private organization of violence also produces wealth, and not only because the drug business is one in which market share is usually seized through deadly force. According to estimates, less than half the income of drug cartels now comes from actual drug sales. The rest comes from other violence-based illicit activities, some market- oriented, others purely predatory. 7 Arms: Since the late 1990s, Mexican drug cartels have been engaged in a kind of subnational arms race, expanding and professionalizing thei\�r structures of defense and repression. Given the porousness of the border\� and the free availability of small arms on the U.S. market (especially since the U.S. federal law banning “assault weapons” expired in 20\�04), they have enjoyed unlimited access to means of destruction. Personnel: According to one much-cited figure, the Mexican drug industry employs about half a million people. Their ranks include an un- determined number of professionals of violence who work in the para- military branches of criminal organizations as bodyguards, street fight- ers, kidnappers, torturers, and killers. 8 Clichés about poor young men who have nothing to lose suggest that the cartels’ proletarian reserv\�e army is unlimited. This may or may not be true. We know little about the\� identity and recruitment of killers. Up to now, though, labor supply for\� the Mexican killing fields has been abundant—even as rumors of forced\� recruitment abound and some foresee looming labor shortages. A second set of explanations puts actors at center stage. Both the state and organized crime have gone through processes of fragmentation. In the “good old days” of hegemonic peace, state officials and criminal organizations institutionalized corrupt exchanges. The former agreed to tolerate illicit enterprises, the latter to pay for official protection \�and to follow certain informal rules of conduct. These “state-sponsored prot\�ec- tion rackets” have now broken down. Both sides have been destabilized\� by the multiplication of actors. 9 On the one side, the spread of electoral competition has replaced hegemonic party discipline with party pluralism at all levels of the po- litical system. On the other side, the government’s strategy of leade\�r- ship decapitation has destabilized the entire system of criminal actors.

It has fractured all relationships: within cartels, among cartels, and between cartels and the state. It has, in short, made organized crime disorganized. In 2006, six major transnational drug cartels were operat- ing in Mexico. Four years later, there were twice as many. In addition, more than sixty local criminal organizations had sprung up, developing every kind of activity that organized violence can render profitable, from mass kidnapping to private protection. The destabilization and multiplication of violent actors intensified violence within cartels (s\�uc- 10Journal of Democracy cession crises), among cartels (market competition), against the state (self-defense), and against society (predation). 10 The demand shock in the international cocaine market is what made the war ignite; the structural availability of money, arms, and personne\�l is what has made it feasible; and the fragmentation of actors is what ha\�s made it escalate. Together, this bundle of factors explains why the war is unlikely to end any time soon. The Societal Subversion of Democracy In the comparative study of regimes, scholars have tended to look for the sources of democratic subversion from above, at the high lev- els of state power. In research on authoritarianism, we have examined dictatorial strategies of institutional manipulation, which are devised centrally at the heights of state power and backed by public coercion.

By comparison, we have tended to overlook the subversive powers that can arise from below and in a decentralized manner from armed actors within society. Outside the reach of state power, they are backed by private violence. While the “vertical” or “state-sponsored” \�subversion of democratic institutions by coercive governments has motivated an entire subdiscipline of comparative research, we know much less about the “horizontal” or “societal” subversion of representative institutions by coercive nonstate actors. On the shiny surface of Mexico’s democracy, things by and large seem fine. Regular elections take place for public offices from the fed- eral presidency on down; multiple parties peacefully compete for votes; plural media and a polyphonic civil society mold public debate; and all the needed democratic institutions are in place (including election\�- oversight and information-access bodies of worldwide repute). There is no dictatorship, and there is no antisystem party or insurgency battling\� to conquer state power. Yet there is internal warfare waged by criminal organizations. The generals and privates in this criminal war do not design elec- toral institutions, rig the vote, bribe electoral authorities, or shave vot- ing rolls. They have neither the means nor the intention to shape formal democratic institutions of electoral governance. But the practical effects of the criminal violence that they wield can be just as damaging to the democratic integrity of elections as the political violence that openly antidemocratic ideologues might employ. Here I focus on the damage that criminal violence does to democracy in the electoral arena. Free and fair elections are democracy’s minim\�al defining institution. Modern representative democracy must offer more than elections (even well-run elections that are inclusive, free, clean\�, competitive, and fair), but it cannot offer less. Criminal warfare dam- ages democratic elections in Mexico by limiting electoral rights and 11 Andreas Schedler liberties in the narrow sense. Yet even prior to that, it constricts the\� wider rights and liberties that nourish and protect democratic elections\�.

In particular, it subverts basic human rights, freedom of expression, and freedom of association.

The commission of violent crimes such as murder, torture, and kid- napping on a huge scale by private organizations reveals a massive fail-\� ure by the Mexican state to protect its citizens. In Mexico as elsewhere\�, the state’s failure to stop some citizens from wreaking systematic ha\�voc on others reflects both its inability and its unwillingness to do so. This is the iron law of lawlessness: When citizens oppress fellow citizens, the state is involved in the oppressive arrangement, whether by commission or omission. In the face of systematic societal violence, state agents often show a similarly systematic indifference. They are complacent about—or even complicit in—the criminal abuses that nonstate actor\�s commit. Contemporary Mexico is no different. Countless pieces of evi- dence point to a syndrome of state abuse, state collusion with crime, an\�d state indifference toward its victims. This syndrome coexists, of course, with state weakness, incapacity, and incompetence. 11 From 2008 to 2010, Mexico received a 4 (a score of 5 is the worst) on Reed M. Wood and Mark Gibney’s Political Terror Scale. Such a rat- ing implies that: “Civil and political rights violations have expande\�d to large numbers of the population. Murders, disappearances, and torture are a common part of life.” 12 Perhaps the most significant symptom of state failure has been the systematic impunity that violent criminals en\�- joy. According to figures collected by Human Rights Watch, between December 2006 and January 2011, Mexican authorities attributed about 35,000 homicides to organized crime. Of these, 2.8 percent led to for- Chihuahua Durango Nayarit Coahuila Sonora Baja California Baja California Sur Sinaloa Nuevo León Tamaulipas Oaxaca Guerrero Michoacán Jalisco Colima 2 3 7 4 Puebla5 Campeche Yucatán Quintana Roo Tabasco Chiapas GUATEMALA HOND.

BELIZE 1 Tamaulipas San Luis Potosí Zacatecas Mexico City 6 Veracruz m AP —m exico ’s stAtes 1. Aguascalientes 2. Guanajuato 3. Hidalgo 4. México 5. Morelos 6. Querétaro 7. Tlaxcala UNITED STATES Gulf of Mexico Pacific Ocean 0 200 100 400 km.

300 mi.

200 100 0 12Journal of Democracy mal criminal investigations, 0.9 percent led to formal criminal charges being filed, and 0.06 percent led to firm convictions. 13 For all practical purposes, the rate of successful prosecution is zero, which amounts to something we have seen in other places in Latin America: The de facto privatization of the death penalty. The state grants private actors (as\� well as its own agents) a license to kill. If democracy rests on the principle of popular sovereignty, and if (as Jürgen Habermas says) the public space is the institutional locus of\� pop- ular sovereignty, then democracy appears feeble and frightened across sizeable portions of Mexico. Analysts now habitually describe the coun- try as one of the most dangerous places in the world for reporters and media personnel. Between 2007 and 2012, at least 74 journalists and media-support workers were killed. Yet murder is only the most visible violation of media freedom. In 2012, the organization Article 19 docu- mented 207 “aggressions” against journalists, media workers, and m\�edia facilities. These included acts of intimidation, physical assault, force\�d abductions, the seizure of entire newspaper or magazine press runs, and even attacks on media buildings with hand grenades and machine guns.

Although criminal organizations are assumed to be responsible for the most brutal violations, Article 19 attributes 43 percent of all recorded\� aggressions in 2012 to state agents, thus identifying state and local of\�- ficials as the “main aggressor[s]” against media freedom. 14 In its 2012 report on media freedom in the world, Freedom House states that “drug cartels are behind the majority of the violence, but local political authorities and police forces appear to be involved in some cases, creating an environment where journalists do not know where threats are coming from or how to avoid the violence.” In the face of cross-pressures from multiple armed actors, many in the me- dia, particularly at the subnational level, have resigned themselves to self-censorship and silence. In some places, as Freedom House rightly observes, criminal organizations have even managed to deepen their influence “from imposed silence to active control of the news agenda.\�” They maneuver to capture, not just the state, but civil society too.

Overall, since 2011 “violence and impunity [have] pushed Mexico into the ranks of Not Free nations” in the realm of media freedom. 15 With a certain dry understatement, Freedom House’s 2013 Freedom in the World report notes that “nongovernmental organizations, though highly active [in Mexico], sometimes face violent resistance, includ- ing threats and murders.” 16 The strength that civil society has acquired in many places across Mexico is real, but it has been achieved despite manifold threats from both private and public agents. Civil society’s vibrancy does not reflect the strength of civil-liberties safeguards in Mexico. Instead, this vibrancy attests to the resilience that citizens h\�ave shown in the face of radical violations of their rights and liberties. During its first four years, the Calderón administration treated the 13 Andreas Schedler internal war’s victims with a mixture of indifference and disdain. In\� response to criminal violence as well as official neglect and abuse, a wide array of local movements have since arisen to defend the victims of violence. In 2011, the Movement for Peace with Justice and Dignity, headed by poet Javier Sicilia, served as a kind of reverse prism for the\� multicolored spectrum of local and regional movements, blending and focusing them into a beam visible to the entire nation. Its biggest suc- cess was to change the terms of public discourse regarding violence.

The movement shattered the generalized presumption of guilt that the government and its agents had been promoting with their suggestion that nearly all those caught up in the violence were combatants. It achieved the formal recognition of victims as victims. Subverting Electoral Integrity Even though the primary goals of criminal enterprises are nonpo- litical, their secondary goals do include political concerns. Just as vi\�- olent political movements easily slide into criminal activities, violent\� criminal organizations easily move into political activities. The political concerns of private violent enterprises are typically narrow. As illegal\� actors, their overriding concern is the criminal law and its enforcement.

Whether their primary economic activity is market-oriented or preda- tory, violent private enterprises can survive and thrive only when law enforcement is ineffective or incomplete. In this sense, they resemble not (armed) political parties that pursue broad policy agendas, but si\�n- gle-issue movements whose concerns are limited to one policy domain. In their ideal world, criminal enterprises would be able to build en- during monopolies of crime while enjoying the tolerance and perhaps even the protection of the state. But in the real world of simultaneous criminal and political competition (at various territorial levels), as in Mexico today, criminal enterprises in fact have a hard time building long-term cooperative relationships with state officials. Aside from set\�- ting up “crime-sponsored protection rackets,” these enterprises ne\�ed to deploy a broader arsenal of criminal survival strategies. In order to ne\�u- tralize law enforcement, they must strive to hide and evade the reach of\� the state (“concealment”), to colonize parts of it through intim\�idation or corruption (“capture”), or to face it down directly through i\�rregular warfare (“confrontation”). To use Jeffrey Winters’s term, the commanders of armed criminal enterprises are “warring oligarchs” who are able to defend their w\�ealth by private paramilitary means. 17 Their wealth sustains their violence, which in turn sustains their wealth. In relation to the state, they act \�like an armed lobbying group with a narrow, but real, interest in shaping the exercise of state power—and thus in influencing access to state power.

Under democratic conditions, this means that they have an interest in 14Journal of Democracy shaping the dynamics of electoral competition. They have a positive in- terest in seeing that cooperative candidates win elections, and a negative interest in seeing that uncooperative candidates do not. From a crimi- nal group’s point of view, the best candidates are those who offer th\�e prospect of discriminatory law enforcement, tolerating the group while combating its competitors. Naturally, the best candidates for one crimi- nal group are the worst for its adversaries. Criminal competition is thu\�s likely to translate into political competition. Luckily, Mexico has so far not seen the levels of political violence that shook Colombia in the 1990s. Yet episodic (and some systematic) evidence abounds regarding interference by criminal actors in electoral competition. This interference takes several forms. Candidate capture: Electoral processes at all levels in Mexico are now systematically contaminated by the suspicion that drug cartels coopt parties and candidates through campaign funding or personal cor- ruption. The assumption that criminal organizations regularly succeed in fielding friendly candidates is widespread. Naturally, hard facts are ha\�rd to come by. Only a handful of candidates or elected officials have been prosecuted and sentenced for their ties to organized crime. 18 Moreover, it is unclear how voters would be able to discern captive candidates, as\� they are likely to disguise their proximity to criminal actors by adopti\�ng aggressive “mano dura” (hard-line) stances on law enforcement. M\�ean- while, a quarter of the respondents to one 2011 survey declared them- selves willing to “vote for candidates related to drug trafficking in order to establish peace and security.” 19 Candidate cleansing: If the coopting of candidates is hard to detect, attempts to drive candidates out of electoral politics through intimida- tion and violence are disturbingly easy to observe. Innumerable candi- dates, along with their relatives and associates, have received threaten\�- ing messages or suffered violent attacks. The most prominent among these was Rodolfo Torre Cantú, who was close to being elected gover- nor of the northern Gulf Coast state of Tamaulipas when he was mur- dered just days before the election in a June 2010 roadside ambush. The next year, farther south and west in Michoacán, 51 candidates for loc\�al offices withdrew before election day. 20 We do not know how many more candidates have been coerced into withdrawing from electoral processes in Mexico’s violence-torn democracy. And we will never know how many have been dissuaded from ever running in the first place due to diffuse or specific threats of criminal violence. Agenda setting: The climate of violence shapes the electoral arena by distorting the field of competitors. In addition, it distorts the agenda of electoral competition. For candidates without criminal ties, the saf- est course is to remain silent. Since any public mention of crimes and criminals can have lethal consequences, silence is the best insurance. I\�n many locales, omert`a, the criminal code of silence, delimits the bounds 15 Andreas Schedler of permissible political discourse in electoral campaigns. You can talk about anything but them.

Voter intimidation: Violent criminals constrain the range of choices that voters enjoy in elections, and may even constrain the act of voting\� itself. Just as violence or the threat of violence may prevent potential\� candidates from running and actual candidates from talking about crime, it may also keep voters from voting. Emergent empirical studies of how violence affects voter participation tend to confirm that organized vio- lence reduces turnout. Aside from deterring participation, criminal orga\�- nizations have on various occasions made public efforts to tell voters f\�or whom they should or should not vote. If the number of voters is large, if the race is not close, and if rival violent groups exert cross-cuttin\�g pressures, such open criminal campaigns are unlikely to sway elections.

Yet even if it does not change outcomes, the very phenomenon of brazen criminal intrusion into the electoral arena jeopardizes the democratic spirit of free and peaceful political competition. In addition to depressing and distorting electoral competition, or- ganized violence corrodes another pillar of electoral integrity: deci- siveness. Through elections, citizens select the most powerful decision makers in the state. For this selection process to be democratic, it mus\�t be decisive, triggering an effective transfer of authority to the winner\�s.

De facto power wielders within state or society violate this condition when they remove certain policy areas from the effective decision- making power of elected authorities (tutelage) and when they prevent winners from taking office or dislodge elected officials from office (reversal). Criminal organizations in contemporary Mexico do both these things. In too many places, criminal enterprises exercise effective tutelage over local authorities. Not only candidates but serving elected official\�s cannot discuss crime. Local authorities know that they can govern (and stay alive) only so long as they keep their hands off the business of violent private actors. The shadow of violence is long. Between 2004 and 2012, 48 active or former mayors are believed to have been assas- sinated by killers acting on behalf of criminal organizations. 21 At least at the municipal level, organized criminals have proven their capacity t\�o reverse electoral outcomes that they find displeasing. The Politics of Silence Classical liberalism fought for the twofold liberation of individuals.

It strove to free citizens from violent impositions by their societies a\�s well as by their public authorities. When societal actors build private organizations of violence and wage private wars against rival organiza- tions, against the state, and against noncombatant citizens, we are force- fully reminded that the liberal agenda requires more than just the tamin\�g 16Journal of Democracy of the state: It also requires the pacification of society. Otherwise th\�e formal democratic promise of individual liberty risks suffocation, not by authoritarian state agents, but by authoritarian citizens.

The massive intrusion of freewheeling criminal violence into ordi- nary life and ordinary politics destroys the weight, autonomy, and in- tegrity of democratic politics and representative institutions. By chok- ing citizens’ rights and liberties and by curtailing the powers of el\�ected authorities, it damages what Larry Diamond calls “the spirit of democ\�- racy” to its core. Two sets of simple questions about the situation i\�n Mexico demand complex answers. The first set begins with the question: How bad is it? And how much does it matter for the overall quality of Mexican democracy? How ex- tensive and how deep are the harms to democracy caused by criminal violence? Are they limited to the subnational level? Should we think of criminal organizations as creating societal authoritarian enclaves at th\�e lo- cal level—what Guillermo O’Donnell once called “brown areas”\�—while at the national level democracy remains intact? 22 If national democracy is affected, how much so? Are we talking about problems of democratic quality or problems of democratic essence? Does it make sense to speak of democracy in the midst of self-reinforcing violence by multiple private armies? It is Mexican citizens who will have to struggle for answers. The second set consists of a single question: Have we seen the worst yet? Perhaps, or perhaps not at all. Organized criminal violence is a re\�- source that many actors can mobilize for their own purposes, be they pri\�- vate or political. We may well see a further diffusion of violence as well as its further politicization. The downward trend in homicides attribute\�d to organized crime that began in 2012 appears to have carried on through\� 2013. Organized violence seems as if it may be stabilizing—albeit at \�a level that only a few years ago would have seemed shocking or even un- imaginable. In his first year in office, President Enrique Pe~na Nieto, the young PRI governor of Mexico State who succeeded Calderón, has been ad- justing his policies against organized violence in subtle ways. He has maintained some policy corrections that his predecessor began. Chief among these is a shift in priorities away from prosecuting petty crimes (such as drug possession in small amounts) and toward containing vio- lent crimes (homicide, kidnapping, and extortion). The new president has also been centralizing the civil-security apparatus; like most of hi\�s predecessors, he plans to create a new federal police corps. He has sig- naled greater commitment to respecting human rights and the rights of victims. He has promised to investigate the thousands of disappearances that have been left unresolved over the past years and to reform the public prosecutor’s office, the Pandaemonium of corruption within the\� criminal-justice system. Overall, though, there has been much talk from the new administra- 17 Andreas Schedler tion about strategy but little clarity about its content. The biggest ch\�ange has been discursive: from the thundering war rhetoric of its predecessor\� to thundering silence. Beyond the few things sketched above, the new government has said little about crime and violence and by all appear- ances wants them off the agenda of public discussion. The president announces positive goals, invoking peace, security, and justice, and otherwise focuses on social and economic policies in such areas as en- ergy, education, and tax reform. It looks like a magic formula: Make the\� problem vanish by making it disappear from public debate. Behind the “magic” lies an implicit technocratic appeal: Trust me and my gene\�rals, and let us take care of this. By substituting the law of silence for pub\�lic debate, and by entrusting peace and justice to military and civil expert\�s, the new president is deciding not to tap a civilizing force that may be \�the only long-term remedy to Mexico’s ailments: civil society.

NOTES 1. See Enrique Krauze, “México: La tormenta perfecta,” Letras libres 167 (November 2012): 15.

2. See Paula Chouza, “El Gobierno mexicano reconoce una lista con 26.\�000 denuncias de desparecidos,” El País (Mexico City), 26 February 2013, 9. On disappearances and mass graves (“narcofosas”) related to organized crime, see also \�Cory Molzahn, Octavio Rodríguez, and David A. Shirk, “Drug Violence in Mexico: Data and \�Analysis Through 2012,” Special Report, University of San Diego Trans-Border Institute\�, 2013, 18–19.

3. Figures from Molzahn et al., “Drug Violence,” 30. On state abus\�e and collusion, see note 10 below. On information problems in irregular wars, see Stathis N.\� Kalyvas, The Logic of Violence in Civil War (New York: Cambridge University Press, 2006).

4. Data for 2010 is from Hemispheric Security Observatory, Report on Citizen Security in the Americas 2012 (Washington, D.C.: Organization of American States, 2012). I bor- row the term “violent democracy” from Enrique Desmond Arias and Da\�niel M. Goldstein, eds., Violent Democracies in Latin America (Durham: Duke University Press, 2010).

5. Author calculations with homicide statistics from the National System\� of Health In- formation (www.sinais.salud.gob.mx) and population data from the 2010 national census ( www.inegi.org.mx). Note that the World Health Organization considers violence to be “epidemic” once it surpasses 10 homicides per 100,000 inhabitants.\� 6. Moisés Naím, “La gente más asesina del mundo,” El País, 15 December 2012.

7. See, for example, Edgardo Buscaglia, “México pierde la guerra,”\� Esquire, March 2010, 95–110. On the cocaine boom, see Ioan Grillo, El Narco: Inside Mexico’s Criminal Insurgency (New York: Bloomsbury, 2011), ch. 4. For a critical revision of commonsensi- cal stories and figures on the transnational drug trade, see Fernando Es\�calante Gonzalbo, El crimen como realidad y representación: Contribución para una hi\�storia del presente (Mexico City: Colegio de México, 2012), ch. 5.

8. See Ana Arana, “Narco SA, una empresa global,” CNN Expansión\�, 20 July 2009, available at www.cnnexpansion.com/expansion/2009/07/17/narco-sa; and Alejandro Hope, “La milagrosa multiplicación de los sicarios,” Blog Anima\�l Político, 23 October 2011, available at www.animalpolitico.com/blogueros-plata-o-plomo/2011/10/23/la-mil- agrosa-multiplicacion-de-los-sicarios. 18Journal of Democracy 9. See Richard Snyder and Angélica Durán-Martínez, “Does Ill\�egality Breed Vio- lence? Drug Trafficking and State-Sponsored Protection Rackets,” Crime, Law, and So- cial Change 52 (September 2009): 253–73.

10. See, for example, Eduardo Guerrero Gutiérrez, “La raíz de l\�a violencia,” Nexos Online, June 2011; and Javier Osorio, “Democratization and Drug Violence in M\�exico,” Paper presented for the workshop on Order, Conflict, and Violence at Yal\�e, October 2012.

11. See, for instance, reports by human-rights organizations, such as Am\�nesty Interna- tional, Known Abusers, But Victims Ignored: Torture and Ill-Treatment in Mexico (Lon- don: Amnesty International, 2012); Human Rights Watch, Neither Rights Nor Security:

Killings, Torture, and Disappearances in Mexico’s “War on Drugs”\� (New York: Human Rights Watch, 2011); and Human Rights Watch, Mexico’s Disappeared: The Enduring Cost of a Crisis Ignored (New York: Human Rights Watch, 2013).

12. See www.politicalterrorscale.org/countries.php?region=NorthAmerica&country =Mexico.

13. See Human Rights Watch, Neither Rights Nor Security, 15.

14. See Article 19, Doble asesinato: La prensa entre la violencia y la impunidad: In- forme 2012 (Mexico City: Article 19, 2013), 10–17 (http://articulo19.org). Assassination figures are from Molzahn et al., “Drug Violence,” 30. The website \�“The Two of Us Met that Terrible Night” contains photographs and biographical notes of m\�urdered journalists. 15. See Freedom House, “Mexico,” Freedom of the Press 2012 (New York: Freedom House, 2012), www.freedomhouse.org/report/freedom-press/2012/mexico. In its 2013 World Press Freedom Index, Reporters Without Borders ranks Mexico 153 rd among 179 countries in the world. Its situation is classified as “difficult”\� for journalists, which is the second-worst category Reporters Without Borders uses to rate global medi\�a liberties. See http://en.rsf.org/press-freedom-index-2013,1054.html. For a narrative account of Mexican journalism under conditions of criminal war, see John Gibler, To Die in Mexico: Dis- patches from Inside the Drug War (San Francisco: City Lights Books, 2011).

16. For the Freedom House report, see www.freedomhouse.org. Under “websites for peace” (sitios por la paz), Sicilia’s organization offers links to like-minded movements.

See http://movimientoporlapaz.mx.

17. Jeffrey A. Winters, Oligarchy (New York: Cambridge University Press, 2011).

18. In one embarrassing 2009 episode, colloquially known as the “Mich\�oacanazo,” federal police and soldiers detained eleven mayors (plus sixteen other \�high-level officials and a judge) in the state of Michoacán on charges of colluding with \�organized crime. Two years later, all had been released for lack of evidence.

19. Raúl Benítez Manaut, ed., Encuesta Ciudadanía, Democracia y Narcoviolencia (Cidena) 2011 (Mexico City: Colectivo de Análisis de la Seguridad con Democracia,\� 2012), 57.

20. See Sandra Ley, “To Vote or Not to Vote: Elections in the Midst o\�f Violence,” 71 st Annual Conference, Midwest Political Science Association, Chicago, 11–\�14 April 2013.

21. Data for 2004 through 2010 come from the Drug Violence Dataset of th\�e Justice in Mexico Project, Trans-Border Institute, University of San Diego, avai\�lable at http:// justiceinmexico.org/data-portal/homicides. Data for 2011 and 2012 come from Molzahn, et al., Drug Violence, 29.

22. Guillermo O’Donnell, “The Quality of Democracy: Why the Rule o\�f Law Matters,” Journal of Democracy 15 (October 2004): 41.