Mafia organizations take many different forms. In this article I analyze three Italian mafia organizations, the Cosa Nostra (the Sicilian mafia), the Camorra (the Campanian mafia), and the ‘Ndrangheta (the Calabrian mafia) and identify two different organizational orders: vertical and horizontal. Mafias with a vertical organizational order are characterized by the presence of higher levels of coordination, centralized power and systemic decision-making processes. In contrast, mafias with a horizontal organizational order are characterized by the absence of higher levels of coordination, and the use of distributed power and clan-based decision-making processes.
The distinction between vertical and horizontal order is relevant to understanding conflict and violence in mafias. Namely, the presence of higher levels of coordination allows mafia organizations to better control conflict and reduce the number of “ordinary” murders committed. At the same time, mafias with vertical organizational orders are in a better position to fight state repression, thus carrying out a greater number of “high-profile” killings (e.g. politicians, magistrates, and other institutional members). However, vertical orders are more visible and therefore more vulnerable to the repressive action of law enforcement agencies. I provide evidence of this argument using a mixed-methods approach that combines a qualitative, organizational analysis of historical and judiciary sources, in order to reconstruct the organizational models and their evolution over time, with a quantitative analysis of assassination trends, in order to relate organizational orders to the use of violence.
This article applies an organizational approach to understand the use of violence in mafia organizations. Organized crime, and mafia in particular, control large and remunerative illegal markets, and also operate in legal markets, sometimes affecting the economic development of entire territories. Nevertheless, organizational studies have not devoted a great deal of time to studying this phenomenon (i.e., cf. Perrow Reference Perrow1986; Handel Reference Handel2003; Baum Reference Baum2005; Clegg, Hardy, Lawrence and Nord Reference Clegg, Hardy, Lawrence and Nord2006; Davis and Scott Reference Scott and Davis2007). Furthermore, the few studies of conflict within and between organizations have principally, if not exclusively, related to legitimate organizations (Ackroyd Reference Ackroyd2009). Sociological studies of conflict have analyzed a wide range of individual and relational factors that account for violent and criminal activities (Collins Reference Collins2010). In particular, recent studies of violence have brought to the fore the importance of moving from a focus on the individual determinants of crime, to the analysis of its relational and social networks basis. Gould (Reference Gould2003) argued that interpersonal violence is a property of relations, and most often emerges from disputes over dominance. He has demonstrated how conflict is more likely to occur in symmetrical relationships, in which there is ambiguity concerning the relative status of the contenders, than in hierarchical ones. Papachristos (Reference Papachirstos2009) convincingly applied this relational approach to the study of organized crime, explaining the spread of violence in gangs through the patterns of interactions and conflict among gang members. While this scholarship has successfully documented the role of interpersonal relationships and social networks in explaining violence in organized crime, in this paper I switch the focus to the formal organizational structure of criminal organizations—that is, their organizational order. By applying an organizational perspective to the study of mafias, I investigate the relationship between types of organizational order and the morphology and frequency of conflict. In more general terms, this article sets the basis for an organizational theory of mafias and organized crime.
This article is organized as follows. In the next section, I will clarify the nature of mafia organizations and how they differ from other forms of organized crime. I will discuss the most significant contributions made by the mafia literature, and in particular the concept of the mafia as a cultural phenomenon versus the mafia as an organization. I will then formulate two hypotheses regarding the influence of the organizational model of a mafia organization on: (1) the frequency of conflicts and the ability to manage them; (2) the ability to carry out unified strategies and high-profile assassinations. I will then analyze the three mafia organizations with the objective of highlighting their different organizational orders: vertical vs. horizontal. The two hypotheses will then be tested. Finally, I will discuss the main findings of the study, and bring out their implications in terms of understanding the relationship between mafia organizations and criminal behavior.
Mafia organizations
The name “mafia” first appeared in 1863, initially indicating the Sicilian criminal organization, which later took on the name Cosa Nostra (“our thing”), and now being used to indicate a specific type of organized crime. Schelling points out that organized crime is substantially different from crime that is organized (1971: 72). Both operate in the world of illegal markets, but the former includes crime that involves functional role division, planning, and cooperation. The latter attempts to actually govern them, by seeking to achieve a monopolistic control on illegal economic activity.Footnote 1 Paradoxically, the monopolistic tendency of organized crime may bring about positive externalities compared to disorganized crime, because the monopoly over criminality guarantees greater control over violence (Buchanan Reference Buchanan1980). Paraphrasing Adam Smith, “it is not from the public-spiritedness of the leader of the Cosa Nostra that we should expect to get a reduction in the crime rate, but from their regard for their own self-interest” (Buchanan Reference Buchanan1980: 132). All organizations, both legal and illegal, try to maximize their profits in the environments in which they operate. Mafias accomplish this goal by providing illegal goods and services as well as operating in legal markets through illegal activities.
Many scholars have argued that mafias are criminal groups that aim at controlling entire markets and territories. They are a form of governance that specializes in supplying different forms of private protection (Sabetti Reference Sabetti1984; Gambetta Reference Gambetta1993; Chu Reference Chu2000; Varese Reference Varese2001, Reference Varese2013; Hill Reference Hill2003): e.g. protection against extortion, theft, police harassment, and competitors; as well as protection of property rights, and debt collection. They also provide protection by settling a wide range of conflicts and disputes, and intimidating customers, workers, and trade unionists for the benefit of employers. Gambetta argues that the essential characteristic of the Sicilian mafia is that it is “a specific economic enterprise, an industry which produces, promotes, and sells private protection” (1993: 1). In line with this argument, Varese (Reference Varese2010; Reference Varese2013) identifies mafia groups as a subset of organized crime groups (OCGs) which specialize in protection. He argues that the term mafia can be used not only for the Sicilian mafia, but also as a general term for such OCGs that share the same core characteristics as the Sicilian mafia: the American Cosa Nostra (Reuter Reference Reuter1983; Gambetta and Reuter Reference Gambetta and Reuter1995), the Russian mafia (Varese Reference Varese2001), the Hong Kong Triads (Chu Reference Chu2000), the Camorra (Campana Reference Campana2011) and the Yakuza (Hill Reference Hill2003).Footnote 2
The three Italian mafia organizations, Cosa Nostra, Camorra, and ‘Ndrangheta, are involved in and provide protection through a variety of activities: extortion, drug trafficking, weapon trafficking, waste management, usury, cigarette smuggling, fraud directed at the European Community, robbery, illegal gambling, illegal immigration trafficking and a great deal more. These activities can be controlled: (1) directly by the mafia organization (e.g. extortion); (2) through sub-contracting groups and criminal cells that are not an organic part of the mafia organization, which benefits from a percentage of the earnings (e.g. drug dealing); or (3) in co-partnership with other criminal groups and/or other mafia organizations. However, the core business of mafia organizations is “market governance”, the monopolistic control of every relevant economic activity. Other criminals must ask their permission and pay for carrying out a specific task.
Rules exist both within a family, and between families: a system of formal regulations, sometimes in written form, of formal criminal constitutions. These rules discipline organizational life, from the recruiting phase on.Footnote 3 Initiation rites too are strictly regulated, in ways that differ from organization to organization and with varying levels of rigidity. The oath taken to be a member of the organization, as a man of honor, is binding for life,Footnote 4 and it requires a subordination of all allegiances to mafia membership. Honor and secrecy (omertà) are key inspiring principles of members’ duties.
Until the mid-1980s, two opposite schools of thought, both reductive and inadequate, characterized the debate concerning mafia organizations.Footnote 5 The first approach argues that the mafia is not an organization, nor a cohesive unit or a corporate group: rather, it is a cultural phenomenon (Hess Reference Hess1970), or a collective attitude (Hobsbawm Reference Hobsbawm1959). The mafia was considered a cultural attitude, a forma mentis, and a form of power with no corporate dimension (Arlacchi Reference Arlacchi1986). “The mafioso exists, but not the mafia”, stated Lestingi (1880: 292). The mafia is neither a sect nor an association, it has neither regulations nor statutes, claimed Pitrè (Reference Pitrè1889). The mafia was characterized then as a “mafioso spirit”, a social practice, a behavior and a power, and not as a formal, secret organization (Block Reference Blok1974; Ianni Reference Ianni1976; Schneider and Schneider Reference Schneider and Schneider1976). The second approach conceives of the mafia as a hyper-centralized organization, pyramidal and top-down, similar to a multinational firm, like an octopus with a head and a host of tentacles. There is a top level which makes all the decisions and a base that faithfully carries these decisions out (Cressey Reference Cressey1967).
Both approaches have their limits: the cultural view underestimates organizational aspects, while the hyper-centralized idea overvalues them. Between one highly fragmented and uncoordinated interpretation of mafia organizations, and one that is unified and rigidly hierarchical, in this article I provide evidence for a more realistic intermediary position. Namely, I show that mafia organizations are not monolithic criminal groups: rather, the autonomy of the families and clans and the coordination between these are structured within a complex network of territorial relationships and arrangements. Moreover, there is no single organizational model and, even within each mafia organization, this model can change over time.
Mafia organizational orders and the control of violence: Two hypotheses
In order to function as effectively as possible, mafias establish organizations with particular structures. To analyze the three Italian mafia organizations, I distinguish between two different organizational orders: vertical and horizontal (Gerlach Reference Gerlach2001; Kenis and Knoke Reference Kenis and Knoke2002; Kogut and Walker Reference Kogut and Walker2001; Powell et al. Reference Powell, White, Koput and Owen-Smith2005; Baldassarri and Diani Reference Baldassarri and Diani2007). Mafias organized according to a vertical order (such as the Cosa Nostra and the ‘Ndrangheta) are characterized by the presence of higher levels of coordination, centralized power, and systemic decision-making processes. In contrast, mafias organized according to a horizontal order (such as the Camorra) are characterized by the absence of higher levels of coordination, and they have distributed power, and clan-based decision-making processes. This distinction is relevant to understand the use of violence in mafia organizations, and in particular the number and the nature of homicides.
The theme of organizational orders and higher levels of coordination presents analogies with, but is not reducible to, that of organizational hierarchy, a classic theme in organization studies (Coase Reference Coase1937; March and Simon 1958; Thompson V. 1961; Thompson Reference Thompson1967; Williamson Reference Williamson1975, Reference Williamson1985; Wilson Reference Wilson1989; Miller Reference Miller1992). Selznick (Reference Selznick1948) identified stability in the lines of authority as one of the elements for the maintenance of an organizational system. Other authors have analyzed the evolution of various organizational configurations, from vertical and integrated forms to network organizations (Powell Reference Powell1990; Nohria and Eccles Reference Nohria and Eccles1992; Baker Reference Baker1992; Podonly and Page Reference Podolny and Page1998; Anand and Daft Reference Anand and Daft2007; Zaheer and Soda Reference Zaheer and Soda2009). Zuckerman (Reference Zuckerman2010) points out the advantages of the hierarchy having the right to speak on behalf of the organization. However, the organizational literature has, on the one hand, privileged the analysis of legitimate organizations and, on the other, made no distinction between hierarchical organizations (command tree) and vertical organizations characterized by higher levels of coordination, in which decision making takes place through negotiation among multiple actors (i.e. clans, or families). The latter is the case in some of the mafia organizations that are the object of this study. All mafias are vertical—and hierarchical—at the family level, but only a few are vertical above this level. If there certainly exists a hierarchy and command tree within a single clan-family, this does not mean that this exists between the clans-families that are part of a mafia organization. The decision-making processes between clans are more configurable in terms of a political coalition (March 1962; Cyert an March 1963).
I argue that the different organizational orders have consequences on mafia behavior, namely with respect to their capacity to govern conflict and contain violence. A common problem for both vertical and horizontal mafia organizations is conflict control and the use of violence, which is an instrument, a resource (Gambetta Reference Gambetta1993), and not the objective, of these organizations. As an instrument, this cannot be eliminated, but must be managed and contained in order to avoid: a) destructive conflict throughout the whole organization, b) greater repressive action from law enforcement agencies following the escalation of violence. As we will see, the management of conflict control and violence (both within and outside the organization) differs between the two organizational orders, both in quantitative terms (the number of crimes) and qualitative ones (the types of homicide) and this depends on the presence, or lack thereof, of higher levels of coordination. Along a similar line of argument, Leeson and Rogers (Reference Leeson and Rogers2012) argue that criminal organizations establish collusive inter-firm institutions designed to internalize the externalities of violent criminal activity and, in so doing, increase the criminal profits of participating mafia families.
Mafia groups that are based on a vertical organizational order, thanks to the presence of higher levels of coordination, should, in general, have greater capacity to speak with a single voice, to make collectively binding decisions, and to strategize, compared to mafia groups that present a horizontal organizational order. This should increase their capacity to govern conflicts and contain violence, and to better defend themselves from external enemies.
These considerations apply both to conflict within the mafia organization and to the use of violence toward the general population. The constitution of higher levels of coordination not only makes it possible to increase power within the organization (Ocasio Reference Ocasio2005), settling disputes, containing internal conflict (feuds between clans), and attempts to seize the position of leadership within a family. It also makes it possible to increase the power of the organization (Mizruchi and Yoo Reference Mizruchi and Yoo2005), exercised with regard to the external environment and in relation to other organizations. A mafia organization with a vertical order will have a strong influence on its territory, and will not need to use violence to control the local population. From this, the following hypothesis can be derived:
Hypothesis 1 – There exists a correlation between organizational order (vertical vs. horizontal) and the frequency of conflict and violence: the vertical model, due to the presence of higher levels of coordination, should lead to fewer conflicts and homicides.
As a response to repressive actions by law enforcement agencies, mafias sometimes resort to the killing of people pertaining to an institution, to politics, to the public sphere, whose activity in some way threatens the legitimacy and interests of the mafia organization. Different from the “ordinary” killings discussed in Hypothesis 1, these high-profile assassinations are more likely to be conducted by organizations with a vertical order. In fact, the decision to eliminate magistrates, policeofficers, politicians, trade unionists, priests or newspaper reportersFootnote 6 requires considerable strategic and operational capacity to be carried out, and organizational resilience to manage the consequences of the inevitable retaliation from the law enforcement agencies. The presence of higher levels of coordination provides mafia organizations with the ability to strategize, identify external enemies, and coordinate action in response to external threats. This brings me to my second hypothesis:
Hypothesis 2 – The presence of higher levels of coordination in the vertical order increases the capacity to respond to the threat of law enforcement agencies, resulting in a greater number of high-profile assassinations involving figures from institutions, politics, and the public sphere.
Taken together, these two hypotheses state that mafias organized according to a vertical model will be implicated in fewer “ordinary” murders, but more high-profile assassinations, than mafias with a horizontal order.
Data sources
Mafia organizations are difficult to penetrate: they are the quintessential secret organization (Simmel Reference Simmel1906). As with many criminal organizations, they cannot be studied as easily as any legitimate organization. Most research on mafia organizations is based on primary sources such as historical and judiciary documents, and the accounts provided by informants, as well as judges, law enforcers, and victims. It is harder to conduct interviews with the mob, and obtain reliable information about their ongoing activities, than with other forms of organized crime or gang members, as has been done, for instance, by Sanchez-Jankowski (Reference Sanchez-Jankowski1991) and Densley (Reference Densley2013). This is particularly true in the case of Italian mafias (Gambetta Reference Gambetta1993; Sciarrone Reference Sciarrone2009; Varese Reference Varese2011): since criminal activities connected to the mafia are prosecuted according to a harsher criminal code, mafia members have a greater interest in maintaining the secrecy of the organization, and their affiliation to it. Moreover, unlike other criminal organizations, such as gangs, mafias place an “emphasis on deceptiveness in public” (Collins Reference Collins2008: 239). Finally, it is technically illegal to entertain a relationship with known mobs, without reporting them to the police. In this research, I aim at both reconstructing the organizational models of the three Italian mafias and documenting their historical trajectory, as well as testing specific hypotheses concerning their criminal behavior. Different types of data, both qualitative and quantitative, are required for this type of enquiry. Several historical accounts are available covering the origins, in the middle of the 19th century, and evolution of the three mafia organizations (e.g. Barbagallo Reference Barbagallo2010; Ciconte Reference Ciconte2011; Dickie Reference Dickie2012; Lupo Reference Lupo2011a, Reference Lupo2011b; Marmo Reference Marmo2011). However, in most recent years, many more primary sources have become available, all of which have been used in this research. The following is a list of these sources (see appendix: sources):
1. Documents from the Italian Parliamentary Antimafia Commission, formed in 1962 and still active today.
2. Major judiciary investigations into mafia organizations and mobster trials (e.g. those conducted by judges Giovanni Falcone, Paolo Borsellino and the team of judges in Palermo into the Cosa Nostra in the 1980s; the “Operazione Armonia” and “Operazione Crimine” investigations regarding ‘Ndrangheta activities in the 2000s; the “Spartacus” trials involving the Camorra in the 2000s; etc.).
3. The testimony of informants. In the 1980s, the Italian Parliament introduced a witness protection program which constitutes a fundamental and innovative source of evidence, in particular with respect to our goal of obtaining a detailed and accurate view (from the inside) of mafia organizational structure.
4. DIA biannual reports. In 1991, the Italian Parliament formed the Direzione Investigativa Antimafia (DIA),Footnote 7 with the goal of coordinating different law enforcement agencies more efficiently in their fight against mafia organizations. Since then, the DIA has produced biannual reports on the criminal activities and evolution of Italian mafia organizations, including information regarding both organizational structure and homicides.
5. DNA annual reports. In 1991, the Parliament also formed the Direzione Nazionale Antimafia (DNA), a body of the Attorney General at the Supreme Court with the task of coordinating investigations relating to mafia organizations at a national level.
6. Data regarding homicides carried out by the three mafia organizations. This information regarding mafia homicides began to be recorded by police forces and ISTAT (the Italian Institute of Statistics)Footnote 8 in 1983. Only since 1982, in fact, has the legislature provided a definition of the concept of “mafia”, with the approval of a new article in the penal code (416 bis) which identifies a new type of crime: criminal association with the mafia. For this reason, data relating to homicides prior to 1982 are not always reliable, because mafia homicides were not clearly differentiated from other kinds of homicide.
7. Data regarding high-profile assassinations. Integrating different sources, from historical and official reports to web-sites, I have compiled an original dataset of high-profile assassinations, which goes back to the second half of the 19th century. People in this category include magistrates, members of the police force, politicians, trade unionists, priests and newspaper reporters. For each murder, I controlled across multiple sources to verify its nature, ensuring that the victim was a recognized mafia enemy.
To provide an analytical description of the different organizational models, their hierarchy of roles and power structure, inter- and intra-organizational relationships, and the presence of higher levels of coordination, I have relied mainly on judiciary, parliamentary and police sources, and on the testimony of informants (aka pentiti, Mafiosi who have turned state’s evidence). It can be argued that these sources have a “legal view” bias: the sources are oriented to highlighting what is legally relevant with regard to the objectives of the judicial enquiry. However, triangulating between different sources has allowed me to elicit the recurrent elements of mafia organizational structure, thus minimizing this bias. Another possible reason for concern may be “collaborator” bias: the pentiti may have a variety of interests in selecting the information they communicate. They may also lie regarding certain aspects that might be further incriminating in relation to their own position. However, informants have no particular incentive to lie in relation to the organizational aspects of mafia organizations, even though they may be keen to minimize their own involvement in criminal activities.
To test the hypotheses concerning the relationship between organizational order and criminal behavior, I have relied on an historical account of mafia wars, which are mostly reconstructed using historical and trial documents, and witness testimonies. I have also made use of the number and type of homicides, which I derived from DIA reports, ISTAT statistics, and other data sources. While conflict assumes different forms, homicides represent a reliable measure of the overall level of conflict (Gould Reference Gould2003), and there is a certain agreement in the literature concerning the fact that they are a form of crime that is less affected by under-reporting and other measurement errors (Fajnzylber et al. Reference Fajnzylber, Lederman and Loayza2002). Here, I use the overall number of killings as an indicator of the organizational capacity to control conflict. The number of high-profile assassinations, meanwhile, is used to measure the organizational capacity to target external enemies.
Cosa nostra, ‘Ndrangheta and Camorra: Organizational profiles compared
The comparison between the Cosa Nostra, the Camorra, and the ‘Ndrangheta is facilitated by the fact that they are not only similar with respect to their economic activities, but also because they are of similar size. They each include approximately 100-150 families (also called clans, or ‘ndrine), and more importantly they have a comparable number of members: around 5,000-6,000 each (see Table 1 for further details). Normally, the term “family” encompasses people who share a common surname, as in the case of the ‘Ndrangheta. With the Cosa Nostra and the Camorra, however, “family” is closer to “clan” than to “household”. The three mafias, however, greatly differ in their organizational structure, recruitment methods,Footnote 9 division of labor, and, above all, the presence of higher levels of coordination and the management of conflict and violence. The principal organizational characteristics of the three mafias are described below. To anticipate the major result of this organizational analysis, I find that, in general, the Camorra has a horizontal organizational order, while the Cosa Nostra and the ‘Ndrangheta tend to conform to the vertical model.
Table 1 Summary of the main characteristics of the three mafia organizations
The Cosa Nostra
The basic unit of the Sicilian mafia, the Cosa Nostra (“our thing”), is the “family”, a criminal group with a specific territorial base, which controls a zone of a city or an entire inhabited area from which it takes its name (e.g. The family of Porta Nuova, the family of Villabate, etc.). The term “family” denotes the fundamental importance given to the concepts of loyalty and honor: actual blood-ties between the members are not necessary.Footnote 10 The selection of members is extremely strict,Footnote 11 unlike the selection and recruiting mechanisms of the Camorra. The families are organized according to a pyramidal model, a hierarchy with subdivisions of power: from the picciotto (the “button man”) at the bottom, up to the family heads. This division of power is similar to the “decimal organization” (Keegan Reference Keegan1994) used by the army. The organizational structure of a family and the hierarchical chain of command work as follows: the base is formed by the men of honor, also known as “button men” or “soldiers”, who carry out the operational orders of the family. The “head of ten” (capidecina) oversees a crew of soldiers, numbering from 5 to 10 or 20, with a maximum of 30, according to the size of the family. The “representative” is the boss of the family, who is elected democratically (one person-one vote) in a secret ballot by the button men in specifically organized family meetings. For large-sized families, the heads of ten collect the votes from the men of honor, given the high risk of bringing together a large number of people in one venue. The vice-representative is nominated by the representative, and can make decisions in the absence of the representative, although this situation rarely arises. The advisors or counselors provide advice to the boss and also serve as liaison with the soldiers.
There is a clearly-marked division of labor: the representative makes the decision, the head of ten transmits the order and has it carried out, and the man of honor performs the action. It is essential for the Cosa Nostra to have complete control over the territory, both in economic and in criminal terms. In the territory controlled by a family, no illegal activity can take place without the agreement of its representative. This is also the case in relationships between families: no one can consider carrying out a criminal activity of a certain importance without the prior approval of the relevant family. Otherwise, sanctions and even death, would be incurred.
Until the mid-1950s, coordination among families within the Cosa Nostra was ensured by informal meetings involving the most important and influential men of honor from the principal families.Footnote 12 But from 1957 onwards, the Sicilian Cosa Nostra began to deploy higher levels of coordination. The heads of the families from the same province (Agrigento, Caltanissetta, Catania, Enna and Trapani) nominate a boss for the entire province, called the “provincial representative”. In the province of Palermo, however, (and sporadically in other provinces), three or more families with adjoining territories are organized into a mandamento (district). It is the job of the head of the mandamento to coordinate the families in operations involving the territories of more than one family, and also to resolve any eventual disputes. For the province of Palermo there is a “provincial commission”, consisting of the province’s various district bosses (18 altogether, representing 54 families), which elects a provincial representative, who represents the entire province. This representative is a secretary and a coordinator, however, and not a boss with the power to command the district bosses or to give orders to the members of the families.
Starting in 1975, the Palermo model, involving a provincial commission, was imitated on a regional scale, forming the highest level of coordination, the so-called Commission or Regional Cupola, which comprises the representatives of the six provinces where the Cosa Nostra operates (see figure 1).
Figure 1 The organizational structure of the Cosa Nostra
The ‘Ndrangheta
The ‘NdranghetaFootnote 13 (meaning “manliness” or “courage”) or “Honored Society”, founded in Calabria, has become the most powerful mafia in Italy (DIA 2004; CPM 2008; DNA 2010). It is estimated to have a membership of around 6,000 grouped into 155 gangs. The ‘Ndrangheta, unlike the Camorra and the Cosa Nostra, is based around blood parentage and the links between the various families are further strengthened through intermarriage. The members are primarily recruited on the basis of blood relationships. There is a strong hierarchy within each family, regulated by a code that involves distinct rituals for every moment of criminal life: from initial membership to the investiture of the new member, to solemnly taken oath, to the next level in criminal ranking, up until the trials which the “tribunal” can impose on members should they violate any of the society’s rules. The organizational structure is more complex than that of the Cosa Nostra, as visualized in figure 2.Footnote 14
Figure 2 The organizational structure of the ‘Ndrangheta
The basic organizational level consists of the ‘ndrina, made up of members of a family nucleus (this can be a few dozen) related by blood ties: a true endogamy.Footnote 15 Every ‘ndrina has its own boss, called the capòndrina, or capobastone (“zone boss”).Footnote 16 Several ‘ndrine in the same area form la Locale, a consortium of families that operate in the same territorial district. In order to be founded, an ‘ndrina must possess at least 49 members. The Locale (or Società) has a boss, called the Capo Locale (“family chief”), or Capo Bastone, who makes decisions regarding operations, calls meetings of the Locale, decides membership and promotion, settles disputes between members and, most importantly of all, directs criminal activity within the territory for which he is responsible. The Capo Locale is assisted by the Contabile (“bookkeeper”), who handles the finances of the locale, and by the Crimine (“Crime”), who deals with illegal activity. Together, these three roles form the Copiata. The Capo Locale also makes use of the Mastro di Giornata (“day master”) who circulates information and orders throughout the organization, reporting back with any problems or specific situations. For extra security, the Locale is divided into two separate organizational groups: the Società maggiore (“major society”) and the Società minore (“minor society”). In each society, seniority is measured in doti, which means qualities or gifts. In other words, the merit accumulated by a member during his career, which increases hand in hand with the crimes committed. Doti are the ranks in the organization’s membership hierarchy.
In order to contain a series of violent internal conflicts, in 1991 the ‘Ndrangheta started making use of higher levels of coordination for the first time. It was a process that was particularly expedited in Reggio Calabria and its province and it led to the creation of two superordinate levels of coordination: the Mandamento (which grouped together several locali), divided into three different areas: Ionian, Tyrrhenian and Centre, and the Provincia. The latter is a supreme coordinating body of criminal activity, able to intervene in internal matters relating to individual locali, and to avoid and/or contain conflict between the families.Footnote 17 The Provincia is headed by the Capo Crimine (“crime boss”), who is not the boss of the bosses, but the speaker of an assembly, whose major role is to enforce rules and settle disputes. He is assisted by three senior officers: Mastro Generale (“general master”), Day Master and Bookkeeper.
The Camorra
The Camorra was founded at the beginning of the 19th century in Campania, first in the prisons, then in the city of Naples. It then spread out to cover the entire provincial territory, where its main pursuit was the organized extortion of the majority of economic activities in the area. The Camorra was initially organized into small, independent groups (Memoria sulla consorteria 1861; Monnier Reference Monnier1862), and showed a high level of organized hierarchy, in particular in Naples itself, where there were 12 groups, one for every district in the city, commanded by a boss (the caposocietà). No trace remains nowadays of this original organizational structure. The Camorra today is a population of criminal organizations (clans) in competition/conflict with each other for governing power over the territory and control over the economic activities present therein. In contrast with the Cosa Nostra and the ‘Ndrangheta, the Camorra is not a unified organization and no single higher level of coordination exists (such as the family or group of families) able to coordinate the entire criminal system. This horizontal organizational configuration is characterized by distributed power in which various clans are in competition and conflict with each other.
The only attempts to organize the various Camorra clans into a vertical order were carried out at the beginning of the 1970s with Raffaele Cutolo’s Nuova Camorra Organizzata (NCO – New Organized Camorra), which echoed the structure of the 19th century Camorra. But this attempt at centralization triggered a strong reaction from a group of clans, who in 1978 created the Nuova Famiglia (the New Family—with the Bardellino, Nuvoletta, Alfieri and Galasso clans, and many more): a temporary federation of families with no centralizing objective who violently opposed Cutolo’s NCO. In the end, this resulted in exacerbating the Camorra’s “anarchic fragmentation”Footnote 18, and led to a bloody conflict that, from 1978 to 1983, resulted in approximately 1,500 deaths.
Other similar attempts were made with Carmine Alfieri’s Nuova Mafia Campana during the 1980s, but these too were doomed to failure. The NCO came to an end in 1983, hopelessly weakened by political alliances, attacks from adversaries and above all by severe repressive measures from law enforcement agencies. The Nuova Famiglia ran out of steam around the same time, while the Nuova Mafia Campana was an aspiration rather than an achieved reality.
The Camorra’s organizational order is fluid, polycentric, conflictual. The central unit is made up of the families. Alliances tend to establish confederations between criminal groups and forces, but these alliances are only partial and contingent; they do not give rise to the configuration of a compact organization with a unified, coherent direction. Conflict prevails over agreement, especially during those periods where there are no leaders characterised by special abilities to command. The majority of the clans (around 70%) therefore find themselves in this situation of competition and conflict. In the Naples area, the clans no longer directly carry out many illicit activities (such as drug dealing, robbery, smuggling, receiving stolen goods and counterfeiting). Instead, they prefer to farm these jobs out to a variety of criminal cells, which then pay them a percentage of their earnings.Footnote 19 In the provinces, however, these activities are run directly by the clans.
The most significant phenomenon in the evolution of the Camorra is the multiplication of clans and families. While there were 16 clansFootnote 20 in 1861, there are 99 today (one of which, the Casalesi, is a cartel composed of 10 clans), and 28 minor groups (DIA 2010b).Footnote 21 To organize activity and reduce conflict, some clans have established forms of coordination that are either tightly coupled (cartels) or loosely coupled (alliances, non-belligerence pacts). In particular, it is possible to identify three types of inter-clan dynamics among the 109 clans:
1. Clans that operates alone, in competition or conflict with the other clans in neighboring territories (about 70% of the total);
2. Operational agreements (e.g. in Salerno), non-belligerence pacts and ad hoc alliances (e.g. the Secondigliano Alliance), which are present, but unstable (about 20% of the total);
3. Cartels or federations of clans, with centralized decision-making and strategic co-ordination (about 10% of the total). The clans have their own representative in these groups and pay part of their earnings from illegal activities into a common fund, giving them the right to use the cartel name.
One might wonder whether the Camorra should be treated as an organization. Although the Camorra in recent years has lacked higher levels of coordination, it still constitutes an organization, with clearly defined boundaries. In contrast, for instance, from a collection of gangs, Camorra clans are characterized by strong historical continuity, and new clans form mainly as scissions of pre-existing ones. Moreover, even in the contemporary period of scarce coordination, almost 10% of Camorra clans are organized in a vertical order in the Casalesi cartel, demonstrating that the potential for the constitution of higher levels of coordination is still exists. Finally, it is important to note that even highly vertical mafia organizations, such as the Cosa Nostra, have experienced periods in which there was no inter-family coordination, for instance when conflicts between important families led to a break down of the Cosa Nostra commission. Even more importantly, despite its fragmented power structure and the absence of higher levels of coordination, the Camorra is considered an organization by law enforcement agencies and by its own members. Camorristi (members of the Camorra) call it “O Sistema”—the System (Saviano Reference Saviano2007)—and distinguish themselves from other criminals and from other organized crime groups on the basis of this identification.
Comparison of the three mafias’ organizational orders
Although they are very similar in terms of the activities carried out, the number of organizational units involved and the overall number of members, the three mafia organizations still differ in numerous important ways, such as organizational structure and the methods of coordination and power management. This is summarized in Table 1.
Mafia organizations have up to three inter-organizational levels: in addition to the families, clans, or ‘ndrine acting autonomously at the local level, they have also, in some cases, introduced higher levels of coordination at the (1) supra-local level (e.g. the mandamenti); (2) provincial level; and (3) regional level.Footnote 22 As shown in figure 3, mafia organizations differ both synchronically (between each other at a specific historical moment) and diachronically (within the same organization over time) in relation to the level of organizational verticality that they have achieved.
Figure 3 Comparison of organizational evolution
The Cosa Nostra (Figure 3a) has moved toward a vertical organizational model, developing higher levels of coordination, first, in 1957, by creating supra-local and provincial levels, and then, in 1975, by creating the regional level. Following severe repressive measures, it broke up the regional level and, in part, the provincial level. Recent enquiries by law enforcement agencies have indicated attempts by the Cosa Nostra to rebuild the superordinate levels of coordination. At its origins in the 1800s, the Camorra (Figure 3b) possessed a centralized command and coordination structure for Naples and its province; thereafter it assumed a horizontal order without higher levels of coordination. The Camorra had two failed attempts —1975-1983 and in 1992—to adopt a hierarchical structure of coordination, but has never been capable of doing so. In contrast, the ‘Ndrangheta has moved from a horizontal model to a vertical one (Figure 3c). The ‘Ndrangheta initially presented a horizontal order. Following the violent conflicts between 1985 and 1991, a supra local (mandamenti) and regional (La Provincia) structure was introduced which greatly reduced—although it did not completely eliminate—the feuds and gave the organization a more strongly unified governance.
In the following sections, we will test whether these different organizational orders have affected their capacity to contain internal conflicts, strategize, and react to external threats, and if vertical organizational models are therefore better able to contain violence and conflict.
Available historical accounts support this expectation. Table 2 reports the major conflicts that have occurred in the three mafia organizations since the 1960s, distinguishing between inter-clan conflicts and intra-clan conflicts.
Table 2 Types of conflict in mafia organizations (a selection of the most relevant)
Inter-clan conflicts include a) mafia wars which involve all the criminal groups in a bitter struggle between two opposing sides; and b) feuds, which involve two clans in conflict with each other for the control of a specific territory, but do not extend to include other clans and other territories. Intra-clan conflicts on the other hand are c) violent internal conflicts that may, in certain cases, lead to d) division. The Cosa Nostra has had three mafia wars (in 1962, 1981 and 1991) and very few other instances of conflict. The Camorra, in contrast, has had only one mafia war, but an extremely violent one (in 1979, resulting in 1,500 deaths). It has however had numerous instances of feuds between clans, internal conflict and divisions. The ‘Ndrangheta has had two mafia wars (1964 and 1985), and a number of clan feuds, though fewer than the Camorra, and practically no divisions.
The visible hand of organization: Different organizational orders generate different criminal behavior
Hypothesis 1 – There exists a correlation between organizational order (vertical vs. horizontal) and the frequency of conflict and violence: the vertical model, due to the presence of higher levels of coordination, should lead to fewer conflicts and homicides.
Mafia organizations have to deal with the problem of violence: to use it instrumentally outside the organization, and to contain it within the organization, both within the families themselves and between the families. The formation of higher orders of coordination makes it possible to control violence. Boulding (Reference Boulding1964) affirmed that one of the main roles of hierarchy is to establish a mechanism for dealing with disputes that cannot be resolved at lower levels, and whose resolution is required for the organization to act. In mafia organizations, the control of internal conflict, violence and homicides should depend, therefore, on the presence, or lack thereof, of higher-level organizational structures. The greater the intimidatory force and authority of a mafia, the less need it has to resort to violence.
My first hypothesis is that the type of organizational order, vertical, as in the case of the Cosa Nostra after 1975 and the ‘Ndrangheta after 1991, or horizontal, as in the case of the Camorra, impacts upon the nature and frequency of homicides. Strong evidence in this direction comes from figure 4, reporting the total number of homicides by mafia organizations. Since the three mafias are similar in size and number of members, we can directly compare the number of killings committed by each organization. In the period 1983-2012, the Camorra killed a total of 3,295 people, while the death toll for the Cosa Nostra was 1,917, and the ‘Ndrangheta 1,575. Of all the mafia related assassinations that have occurred in Italy since the 1980s, 49% were carried out in Camorra territory, while the remaining half were split between the Cosa Nostra (28%) and the ‘Ndrangheta (23%).
Figure 4 The total number of homicides of the three mafias 1983-2012
(Sources: the Ministry of the Interior, ISTAT; my elaboration).
Analyzing the data over time for each mafia organization provides additional insights that are in line with my hypothesis. Figure 5 shows the trend for the total number of assassinations for the Cosa Nostra. On average, the number of deaths per year is 64. There was a peak in killings in the early 1990s which is explained by two main events. First, there was an internal war within the Cosa Nostra itself, due to the attempt by one group, the Corleonesi, to take over control of the organization, eliminating their adversaries, as in the violent feud in Partanna (Trapani), which lasted from 1987 to 1991. Second, the Supreme Court’s confirmation in 1992 of the sentence of the Maxiprocesso Footnote 23 (“Great trial”) condemned many Cosa Nostra bosses to life imprisonment, which created a series of power vacuums. This peak was then followed by a steady decline, which was also due to severe repressive measures by the law enforcement agencies, culminating in the arrest of Totò Riina (the boss of the Corleonesi family).
Figure 5 Cosa Nostra homicides, 1983-2012
(Sources: the Ministry of the Interior, ISTAT; my elaboration).
In contrast, in the case of the Camorra, the absence of higher levels of coordination, the fragmentary nature of the organization’s clans and the fluidity of alliances made it impossible for the conflicts to be negotiated and managed in a non-violent manner within the organization. The principal enemy of a Camorra clan always remains the rival territorial clan, as shown by the high rate of homicides: 110 per year. The trend shows several peaks concomitant to various conflicts (see figure 6).
Figure 6 Camorra homicides, 1983-2012
(Sources: the Ministry of the Interior, ISTAT; my elaboration).
Finally, until the ‘Ndrangheta introduced higher levels of coordination––a command structure that was capable of controlling disputes––there was a great deal of internal conflict, with numerous feuds and deaths. In 1985, the second war began between ‘Ndrangheta clans, a war triggered by attempts to control contracts and tenders and by one clan’s desire for territorial expansion.Footnote 24 This conflict extended further, involving other clans and bringing about around 700 deaths amongst members. In 1991, to recover and reorganize in the aftermath of this conflict, a temporary superordinate unified structure was introduced (see figure 7).
Figure 7 ‘Ndrangheta homicides, 1983-2012
(Sources: the Ministry of the Interior, ISTAT; my elaboration).
The power and autonomy of each individual ‘ndrina remained unaltered, but this superordinate structure made decisions regarding the most important matters involving the entire organization. The special nature of this body lay in the balance between centralization and autonomy: on the one hand, the controlling structure forced all the ‘ndrine to respect its decisions; on the other, they were allowed total autonomy with regard to the rest of their mafia activity. This temporary solution then became a permanent one, beginning a process of progressive verticalization and centralization in the ‘Ndrangheta, with the constitution of the provincial Cupola, and in particular with regard to centralization relating to the most crucial and delicate decisions (DIA 2010b). The provincial level also became superordinate with respect to the structures operating outside Calabria. Thus, the process of centralization and the creation of superordinate bodies of coordination and conflict resolution had their origin in a violent clan war within the ‘Ndrangheta.
Thanks to this new organizational structure, the number of feuds and homicides (see figure 7) dropped sharply, from 680 homicides between 1986 and 1991 to 167 in the 1992-1997 period, and then proceeded to further diminish in the following years (1998: 15; 1999: 13; 2000: 1). Moreover, the leading body decided to do away with kidnappings, which generated a low level of income compared with the severe measures of repression they triggered from the state. The existence of a guiding structure taking responsibility for issuing and ratifying decisions has been confirmed by recent law enforcement enquiries (DNA 2011: 109).
While rates of homicides are the most compelling indicator of the capacity of vertical orders to contain violence, there are other considerations that strengthen the argument advanced in Hypothesis 1. Namely, the average age of the Cosa Nostra’s family heads is in general higher than for those of the Camorra (Gambetta Reference Gambetta1993). This is an indicator of greater organizational stability in the Sicilian mafia and suggests, according to Gambetta, that the Cosa Nostra bosses are older because their positions of power are acquired through a more organized selection process—one which discourages younger members from challenging the internal hierarchy. The main route for professional promotion in the Cosa Nostra therefore passes through internal channels, not through direct competition in the market, with the selection of top leaders taking place largely according to a hierarchical logic and not through the violence of the marketplace. The presence of superordinate bodies not only reduces conflict between clans, or families, but also within the individual clans themselves, discouraging in-family disputes and attempts by the dissatisfied or impatient to seize power. As one Cosa Nostra informant, Vincenzo Marsala, states:
When disputes of various kinds between members come to the attention of the representative, if the latter cannot settle the matter himself, he turns to the district boss, who intervenes to find an agreement between the members and reach a decision regarding the particular problem. When the matter is a serious one that involves the entire family territory, the representative seeks the help of the district boss and the latter intervenes through the provincial commission of Palermo (OSPA in Stajano Reference Stajano2010, p.102).
It should be remembered that in the case of the ‘Ndrangheta, the family is a criminal organization based on blood-ties and this relationship automatically dictates the chain of power, with a rigid succession through the direct male line. This is an important element of stability in the organization, given that matters relating to succession render mafia organizations vulnerable to conflict, fostering feuds, division and internal wars. In the case of the ‘Ndrangheta, the issue is resolved through the principle of the family law of dynastic male succession.
Higher levels of coordination come about for various reasons. Regarding the ‘Ndrangheta, such levels were introduced to control conflict between clans and violent feuds, especially in the 1985-1991 period. Where the Cosa Nostra is concerned, however, according to the informant Tommaso Buscetta, the commission was originally created in order to settle conflict within clans, and only thereafter to discipline the activities and possible disputes between families. The commission was an arena, states Calderone (Arlacchi Reference Arlacchi1993: 126 and thereafter), where it was possible to resolve such important matters as homicides, who to support at political elections, and to decide punishment for those who made mistakes and violated the rules. For example, if a Cosa Nostra family head disappeared for some reason, this would create a dangerous power vacuum that might generate conflict. The superordinate bodies would therefore intervene to make a decision regarding succession, bestowing stability upon the system. In the case of the Camorra, however, the disappearance of a clan boss would create a sense of opportunity both within the clan itself (with various members claiming power) and in neighbouring clans. This behaviour is not regulated by any superordinate force and can therefore break out into particularly violent conflict, as shown in figure 6.
For all these reasons, mafia organizations based on vertical organizational order, due to the creation of higher levels of coordination (e.g. the Cosa Nostra after 1975 and the ‘Ndrangheta after 1991) were, overall, more capable of containing conflict and homicides than the Camorra, as the history of the three organizations demonstrates.
Hypothesis 2 – the presence of higher levels of coordination in the vertical order increases the capacity to respond to the threat of law enforcement agencies, resulting in a greater number of high-profile assassinations involving figures from institutions, politics, and the public sphere.
Higher levels of coordination have been created to settle disputes and violent conflicts. However, in addition to conflict containment, the presence of higher levels of coordination in mafia organizations makes it possible to create organizational strategies and to identify and eliminate external enemies of the organization. This ability becomes significant when, for example, the state increases repressive measures in relation to mafia organizations. We therefore expect that, in case of intense investigative activity by law enforcement agencies or public attack by members of the civil society (e.g. politicians, priests, newspaper reporters, etc.), mafia organizations based on a vertical order will commit more high-profile homicides.
In theory, an alternative argument might be advanced: namely, that a well-coordinated vertical structure could represent a better partner for politicians and therefore foster collusion, thus reducing high-profile assassinations. However, this alternative hypothesis relies on the assumption that both actors involved, i.e., mafia organizations and the state, are unified actors, which act in a coherent and coordinated manner. While the Italian Cosa Nostra (after 1957 and 1975) and the ‘Ndrangheta (after 1991) were, in general, unified organizations, this has never been the case for the state (cf. Lupo Reference Lupo2011a; Dickie Reference Dickie2012). At any point in time, political and institutional actors (e.g. the state, magistrates, politicians, political parties, law enforcement agencies etc.) acted in different ways, with some parts preferring to collude with mafia organizations (by fear or interest), and other parts choosing to fight the mafia (e.g. Pio La Torre, Carlo Alberto Dalla Chiesa, Giovanni Falcone, Paolo Borsellino). As a consequence, mafias sometimes killed the latter and made agreements with the former. In sum, the fact that the state has always presented itself as a fragmented actor, makes this alternative framework unlikely to apply to the Italian case.
The Cosa Nostra and the Camorra have been targeted by law enforcement agencies and other institutions for a long period of time, while the ‘Ndrangheta has only been an object of attention in fairly recent years. If we take the value of the assets seized and confiscated from a mafia as an indicator, then we can see that, in the 1992-2011 period, assets worth 7,420,080 billion euros were confiscated from the Cosa Nostra, assets to a value of 4,799,346 billion euros were taken from the Camorra, while assets worth only 1,181,659 billion euros were seized from the ‘Ndrangheta (source: DIA 2012). The magistrate mentioned above, Giovanni Falcone (1992), believed that the ‘Ndrangheta, and to a certain extent the Camorra, given their horizontal organizational model, were less dangerous than the Cosa Nostra.Footnote 25 The ‘Ndrangheta was only really recognized as a dangerous mafia organization around the year 2000—the most dangerous, indeed (CPM 2008; DNA Annual report 2010 and 2011).Footnote 26 Only in 2010, was the ‘Ndrangheta included in the Italian antimafia law. Following this growing attention from the law enforcement agencies, the ‘Ndrangheta recently initiated a reaction with a series of attacks on, and intimidatory acts towards, well-known people (DNA 2010: 89).
My hypothesis is that the Cosa Nostra was able, in particular from 1975 onwards, with the constitution of the regional level, la Cupola—precisely because of its organizational model with higher levels of coordination—not only to reduce and contain conflict (as we have already seen). It was also able to identify its true adversaries in the state outside its organization, or, rather, in that part of the state or civil society which refuses to cooperate with it (magistrates, politicians, public administrators etc.). Table 3 reports the number of high-profile assassinations carried out by the three organizations. In Figure 8, the same information is presented, distinguishing between institutional, political, or public sphere killings. The Cosa Nostra is the organization that has killed the most high-profile people by far (231), the first dating back to the second half of the 19th century. The Camorra, in contrast to the Cosa Nostra, has carried out a significantly smaller number of high-profile assassinations (46, see table 3). In addition, more than half of them (26 out of 46) occurred in the eight years, from 1975 to 1983, in which the Camorra was organized in a vertical order. The fragmentary nature of the Camorra organization hinders the elaboration of unified strategies and makes it difficult to identify the entire criminal organization’s common enemies on the outside.Footnote 27 Moreover, the Cosa Nostra stands out as having killed a high number of magistrates, certainly the targets with greatest media impact, and also the most costly in terms of state retaliation.
Table 3 High-profile assassinations committed by three mafia organizations (my elaboration)
Figure 8 High-profile assassinations committed by three mafia organizations (my elaboration)
Finally, the ‘Ndrangheta, which until 2007-2008 was not the focus of attention of the repressive action of the state (cf. the Antimafia Parliamentary Commission 2008; DNA 2010), has limited itself to just 33 murders.
Further support for Hypothesis 2 comes from the analysis of the Cosa Nostra’s high-profile assassinations over time. In the 10 years before the creation of the Cupola (1965-1975), the Cosa Nostra killed “only” 9 prominent individuals, while in the following 10 years (1976-1986) the death toll of high-profile victims reached 58, almost a 600% increase. Moreover, while only one judge was killed in the decade preceding 1975, 7 were killed in the following 10 years. The same considerations hold if we look at larger time windows: if we focus on the 1957-1975 period, when the Cosa Nostra had a provincial structure of coordination, the overall number of high-profile killings reached 24, with only one judge among the victims. In the 18 years after 1975, however, there were 87 high-profile assassinations, including 13 judges.
The importance of having higher levels of coordination to carry out such notable killings has been repeatedly acknowledged. No murder of a significant person could be carried out without being authorized from on high. The informant Leonardo Messina has said:
If a normal person has to be killed, or if normal interests are involved, it’s enough for the town bosses to communicate this to the local mandamento. But if a journalist or magistrate has to be murdered, orders have to come from the region. To kill a “man of honor”, you need orders from the region […] Policemen, too, cannot be killed without orders from the region. To kill a “man of honor”, you need an order from the provincia. To kill a “head of ten”, you need orders from the region (CPM 1992: 517).
The informant Antonino Calderone stated:
If … an important man has to be hit—a politician or policeman or judge—the decision has to come from above, from the highest level, namely from the regional commission […] This is logical. A murder of this kind can cause harm to everyone. True, the killing is done in a given territory, but its consequences will be paid later by everyone (Arlacchi Reference Arlacchi1993: 36).
In general, the constitution of higher levels of coordination gives the Cosa Nostra and the ‘Ndrangheta’s leadership the right to speak, and act, on behalf of the organization (Zuckerman Reference Zuckerman2010). This does not only enhance their capacity to carry out high-profile assassinations, but also ensures several important organizational functions (see Table 4): strategic, making decisions regarding the interests of the whole organization; control and organizational, such as the control of territorial division between the families; conflict containment, containing conflict in progress and preventing any that might arise; and business, identifying business opportunities that exceed the capacity of an individual family and fostering economies of scale.
Table 4 Functions and activities of higher levels of coordination
Magistrate Giovanni Falcone stated (1992, 100):
The more the organization is centralized and clandestine, the greater the threat it poses, because it has the means to effectively control the market and maintain order in its territory, with only an extremely brief period between a decision being taken and the ensuing action. It’s a different kettle of fish when you’re dealing with a fragmented organization with a number of centers of power.
The Camorra, lacking higher levels of coordination and leading command structures, is not capable of systemic decision-making; nor can it make long-term strategies.
Discussion
There are several advantages related to a vertical organizational order: it increases the capacity to speak with a single voice, to make collectively binding decisions, to manage conflict, and to identify external enemies. In contrast, in the horizontal order there are no leaders who can speak on behalf of the whole organization; it is hard to manage and contain conflict, and promote systemic decision-making.
Given all these advantages, why do all mafias not converge towards a vertical organizational model? Vertical orders have their own downsides, the most important of which is their greater vulnerability to repressive action and collaboration of informers. Namely, having a hierarchical organizational structure makes it easier for law enforcement agencies to establish strategic priorities and to target its leaders. In addition, collaboration by informants can cause great damage by revealing crucial information about the organizational structure. In contrast, horizontal orders are less vulnerable to informant collaboration, and the arrest of members of one clan does not weaken the organization as a whole.
In the light of these considerations, we can argue that mafia organizations have to deal with an organizational dilemma, a “trade off”, between the benefits of one model and the other. The historical reconstruction of the organizational orders of the three mafias presented in this paper shows that mafias subsist in an unstable dynamic equilibrium, moving between a vertical model and a horizontal model, which gives greater autonomy to the individual families.
Analytically, this dynamic equilibrium is captured by the cycle of mafia organizational action presented in figure 9. This dynamic cycle goes (1) from individual organizational action (the family, clan, ‘ndrina), towards the search for forms of vertical coordination. This leads (2) to the creation of higher levels of coordination (based on the municipality, province, region), which makes it possible to reduce conflict and seize greater opportunities for complex business.
Figure 9 The cycle of mafia organizational action
If, on the one hand, vertical coordination increases the power of the organization (3), making strategy development possible, on the other, it engenders greater visibility and vulnerability, since law enforcement agencies can prioritize repressive action, with primary emphasis on the search for the higher-ranked members. In addition, (4) excessive centralization increases discontent and can lead to centrifugal forces, generating often violent clan conflict (mafia wars). The superordinate levels are thus delegitimized and/or undermined by repressive action (arrests) and are broken up the organization returns to a stage where individual organizational action prevails.
The process of vertical coordination and decisional centralization confers greater stability on an organization, allowing it to pursue long-term interests, since the capacity of the organization to persist over time ensures the participating actors with the guarantee that, assuming costs today, they will be compensated tomorrow. In contrast, in situations where processes of decentralization and autonomy prevail, short-term logic (“everything at once”) tends to prevail, since there is no incentive for the actors to postpone the satisfaction of their own interests.
It is worth pointing out that a vertical mafia organization—as with the Cosa Nostra after 1975, with the formation of the regional commission, and the ‘Ndrangheta after 1991 with the formation of the Provincia—does not imply a totally top-down decisional process. Rather, a vertical mafia organization is still composed of autonomous parts (the families, the minimal organizational units). Membership, in fact, is always passed through the families: a member pertains to a family, not to the Cosa Nostra or the ‘Ndrangheta in general. The Cosa Nostra and the ‘Ndrangheta have created higher levels of coordination with functions that work more in terms of controlling relationships and resolving disputes than of hierarchical chain of command. In these governance bodies, goals are not given but bargained and members operate in a situation of biased information, bounded rationality and opportunism (Simon Reference Simon1947, Williamson Reference Williamson1975). They are configurable rather as a political arena, with actors who are stakeholders with diverging objectives, establishing relationships with one another, either in terms of alliance or conflict (March 1962; Cyert and March Reference Cyert and March1963).
Historically, the experience of extreme hierarchical centralization seems destined to fail. Thompson (Reference Thompson1967) has pointed out that the pyramid headed by an all-powerful individual is possible only in simple situations. When Cutolo tried to impose a totally vertical model on the Camorra, assigning to himself the role of “boss of bosses”, the end result was to trigger the opposition of many families who formed a coalition in order to fight the centralizing project.Footnote 28 The same dynamic occurred in the American Cosa Nostra. Two bosses, Salvatore Maranzano and Giuseppe “Joe the Boss” Masseria, tried to become the boss of the bosses in New York City in the 1920s, giving rise to the Castellammare war in the 1930s. Maranzano won the war and tried to exercise hegemony over other mafia groups, arousing strong opposition from them. The role of “boss of bosses” did not last long and Maranzano was quickly eliminated by other families (Maas Reference Maas1968; Critchley Reference Critchley2006, Reference Critchley2009; Raab Reference Raab2006).
The hypothesis one might derive from the extant historical evidence is that mafia organizations are not suited to a hierarchical leadership. In fact, the existence of a “boss of bosses” is incompatible with an organizational order based on families, since it tends to compromise the interests of many and provide advantages to few; it increases conflict by triggering attempts to seize the leadership position by other families and by mafia leaders. The creation of higher levels of coordination through consensus and without the atypical “boss of bosses” role seems to have a greater possibility of success and longevity compared to the creation of “impositions”, with a boss who commands the entire organization as leader.
Conclusion
To understand conflict and violence in mafias, and more generally in organized crime, we need to understand their organizational orders. In this article I have argued that differences in criminal behavior can be traced to the specific forms that a mafia organization assumes over time. If mafias are based on a vertical order, they are more able to contain conflicts and violence, and tend to kill less ordinary people but more high-profile individuals. The opposite occurs if mafias are based on horizontal order. Higher levels of coordination are employed to reduce conflict (both intra- and inter-organizational), to carry out unified strategies and seize opportunities, to foster coordination and cooperation both between organizational units and individual people, and to improve career management and discourage opportunism. The broader goal of this article is to promote the use of an organizational perspective in the study of mafia organizations. This analysis can be extended in various directions: for instance, by considering additional mafia organizations or other types of organized crime, by analyzing single organizations in greater detail, and by systematically analyzing the strategic relationship between the different organizational orders and the repressive action of the state.
Acknowledgments
I am grateful to Delia Baldassarri, Paolo Campana, Giuseppe Delmestri, Diego Gambetta, David Greenberg, Federico Varese, and the EJS Board and reviewers for their helpful comments and suggestions.
APPENDIX: SOURCES
1. Parliamentary sources: CPM, Commissione Parlamentare d’inchiesta sul fenomeno della mafia e sulle altre associazioni similari, (Parliamentary commission of inquest on the Mafia) ( http://legislature.camer_bicamerali/_bicamerali/nochiosco.asp?pagina=/_bicamerali/antimafia/home.htm ).
CPM, (1992a), Annual Report, Doc. XXIII, n. 47, X Legislature, February 19.
CPM, (1992b), Audition of the collaborator with Justice Antonino Calderone, X Legislature, November 11.
CPM, (1992c), Audition of the collaborator with Justice Tommaso Buscetta, X Legislature, November 16.
CPM, (1992d), Audition of the collaborator with Justice Leonardo Messina, XI Legislature, December 4.
CPM, (1993a), Audition of the collaborator with Justice Gaspare Mutolo, XI Legislature, February 9.
CPM, (1993b), Report on relations between the Mafia and politics, Doc. XXIII, n. 2, XI Legislature, April 6.
CPM, (1993c), Report on relations between the Mafia and politics, Doc. XXIII, n. 2bis, XI Legislature, April 6.
CPM, (1993d), Report on relations between the Mafia and politics, Doc. XXIII, n. 2ter, XI Legislature, April 6.
CPM, (1993e), Audition of the collaborator with Justice Pasquale Galasso, XI Legislature, September 17.
CPM, (1993f), Report on the situation of crime in Calabria, Doc. XXIII, n. 8, XI Legislature, October 12.
CPM, (1993g), Minority report on the situation of crime in Calabria, Doc. XXIII, n. 8bis, XI Legislature, October 12.
CPM, (1993h), First Annual Report, Doc. XXIII, n. 9, XI Legislature, October 19.
CPM, (1993i), First Minority Annual Report, Doc. XXIII, n. 9bis, XI Legislature, October 19.
CPM, (1993l), Report on the Camorra, Doc. XXIII, n. 12, XI Legislature, December 21.
CPM, (1994), Final Report, Doc. XXIII, n. 14, XI Legislature, February 18.
CPM, (1998a), Report on collaborators with justice, Doc. XXIII, n. 11, XIII Legislature, June 30.
CPM, (2000c), Report on organized crime in Campania, Doc. XXIII, n. 46, XIII Legislature, October 24.
CPM, (2001), Final Report, Doc. XXIIII, n. 57, XIII Legislature, March 6.
CPM, (2006), Final Report, Doc. XXIII, n. 16, XIV Legislature, January 18.
CPM, (2008a), Report on the ’Ndrangheta, Doc. XXIII, n. 5, XV Legislature, February 19.
CPM, (2008b), Report on collaborators of justice, Doc. XXIII, n. 6, XV Legislature, February 19.
CPM, (2008c), Final Report, Doc. XXIII, n. 7, XV Legislature, February 19.
CPM, (2011a), Report on the economic costs of organized crime in southern Italy, Doc. XXIII, n. 5, XVI Legislature, February 9.
CPM (2012), Report on the first phase of the Commission's work, particularly with regard to the conditioning of the mafia economy, society and the institutions of the South, Doc. XXIII, n. 9, XVI Legislature, January 25.
2. Judicial sentences (alphabetical order)
CAV, Corte di Assise di Varese (1997), Tribunale di Varese, Sentence of the criminal proceedings n. 7/95 against Zagari Antonio + 125, November 13.
Corte di Assise di Appello di Napoli (2008), Tribunale di Napoli, Processo “Spartacus 2”, Sentence of second degree, June 18.
Corte di Assise di Appello di Reggio Calabria (2002), Tribunale di Reggio Calabria, Processo “Armonia”, Sentence of second degree, July 11.
Corte di Assise di Reggio Calabria (1999), Second criminal session, Tribunale di Reggio Calabria, Sentence of the criminal proceedings Olimpia, n. 46/93 R.G.N.R. D.D.A, n. 72/94 R. G.I.P. D.D.A, n. 3/99 Sent., n. 18/96 R.G. Assise, January 19.
Corte di Assise di Santa Maria Capua Vetere (2005), Tribunale di Santa Maria Capua Vetere – Caserta, Processo “Spartacus 1”, Sentence of first degree, September 15.
OSPA (1985), Tribunale di Palermo, Ordinance Sentence of the criminal Proceedings against Abbate Giovanni + 706, Palermo, November 8.
RPMNA (1986), Tribunale di Napoli, Indictment of the public prosecutor at the Court of Assizes of Naples against Abagnale + 101, vol. 1, April 23.
Sezione G.I.P. – G.U.P. (2012), Tribunale di Reggio Calabria, Processo “Crimine”, Sentence of the criminal proceedings against Agnelli Giovanni + 126, March 7-8.
Suprema Corte di Cassazione (2010), Prima Sezione Penale, Tribunale di Santa Maria Capua Vetere – Caserta, Processo “Spartacus 3”, Sentence of third degree, January 15.
3. Minister of Interior, Law and Enforcement Agency, Minister of Justice
Corte di Appello di Napoli, Annual report of the Attorney General of Naples Vincenzo Galgano on administration of justice in 2004 in the District of Corte di Appello of Naples, General assembly, 2005, January 15.
Dicastero degli interni e Polizia, Memory on the faction of the Camorra existing in the Neapolitan provinces, and Report on the Camorra, 1861 [in Marmo (Reference Marmo2011)].
Direzione Distrettuale Antimafia di Reggio Calabria, “Olimpia” Operation, 1994.
Direzione Distrettuale Antimafia di Reggio Calabria, “Primavera” Operation, 2000.
Direzione Distrettuale Antimafia di Reggio Calabria, “Armonia” Operation, 2003.
Direzione Distrettuale Antimafia di Reggio Calabria, “Crimine” Operation, 2010.
Direzione Distrettuale Antimafia di Reggio Calabria, Decree of detention of a suspect for murder c.p.p. Agostino Anna Maria + 155, Procura della Repubblica presso il Tribunale di Reggio Calabria, Operazione “Crimine”, 2010.
Direzione Distrettuale Antimafia di Reggio Calabria, Decree of detention of suspects of crime and seizure in urgency, N. 1364/2011 R.G.N.R. DDA, 2011.
Direzione Investigativa Antimafia (DIA), Minister of interior, Activities and final findings, 1998 – 2011 (a: first semester; b: second semester), Volumes 30, total pages 5518 (http://www.interno.it/dip_ps/dia/).
Direzione Nazionale Antimafia (DNA), Summary of the main forms of crime of Italian and foreign mafias: “Ndrangheta”, Annual Report, December 2001.
Direzione Nazionale Antimafia (DNA), Annual report on the activities of Procuratore Nazionale Antimafia and Direzione Nazionale Antimafia, as well as on the dynamics and strategies of organized crime, 2008 – 2010, Three Volumes, total pages 2841.
4. Other testimony of mafiosi who turned state’s witness
TB: Testimony of Tommaso Buscetta given to Giovanni Falcone, Giudice Istruttore di Palermo, et al., July-August 1984, vol. 3.
SC: Testimony of Salvatore Contorno given to Giovanni Falcone, Giudice Istruttore di Palermo, et al., October- 1984, June 1985.
AC: Testimony of Antonino Calderone given to Michel Debacq (Giudice Istruttore of Marseilles), and to the Commissione Rogatoria Internazionale, with Giovanni Falcone (Giudice Istruttore di Palermo) et. al., March 19, 1987 – June 25, 1988, vol. 4.
5. Data on general homicides
Website:
6.Data on high-profile assassinations
Website:
http://vittimemafi_bicamerali/index.php?option=com_content&view=category&id=35&Itemid=67.
http://progettolegalita.it/it/prodotti_sociali/elenco_vittime_della_mafia.php
Report:
S.M.S. Michelangelo Napoli (2009), Un nome, una storia, Centro di Documentazione regionale contro la camorra, Napoli.