Criminal Networks and Trust

Paper presented at the 3rd annual meeting of the European Society of Criminology (ESC)
Helsinki, Finland, 29 August 2003
by Klaus von Lampe and Per Ole Johansen

(A short version of this paper has been published in an anthology edited by Semi Nevala and Kauko Aromaa, which is available in PDF; a revised version of this paper has been published in Global Crime, 6(2), 2004, 159-184)

PDF-Version

INTRODUCTION

In the organized crime literature it has become a truism to say that what holds "organized criminals" together are bonds of trust. There are, in essence, two perspectives from which the issue of trust and organized crime is addressed. On the one hand, there is the question of how to account for the fact that criminals cooperate in the first place. After all, there are particular risks involved in co-offending. Many of the institutional safeguards designed to compensate for the consequences of deceit and betrayal, such as the courts and insurance, are unavailable for illegal actors, and in light of the threat of law enforcement intervention and criminal sanctions, the consequences of disloyal behavior are likely to be much more severe than those to be expected in the legal sphere of society (McCarthy et al. 1998; Reuter 1983). Trust, it is said, reduces the uncertainty regarding the behavior of potential accomplices to a tolerable level and thereby stimulates the willingness to co-offend (Weerman 2003; Zaitch 2002). On the other hand, there is the desire to explain the perceived tendency of organized crime to be embedded in webs of kinship and ethnic ties. It is the requirement for trust, it is argued, which gives members of close knit communities a competitive advantage in the crime business (Paoli 2002).
While these notions have some merit to them, they cannot be taken at face value. The conventional wisdom regarding the relation between trust and organized crime lacks a comprehensive theoretical and empirical underpinning. Conceptually, often enough the term "trust" is being used in such a vague sense that it does not amount to much more than an empty phrase. Empirically, the notion of trust has been challenged by arguing that organized crime is better characterized by a lack of trust (van Duyne et al. 2001: 99, 127), as people who tend towards criminality are unlikely, in the words of Gottfredson and Hirschi (1990: 213), "to be reliable, trustworthy, or cooperative". Even life in the Mafia, as Diego Gambetta (1996: 152) has stressed, is fraught with uncertainty, distrust, suspicion, paranoid anxiety and misunderstanding. Before this background, considerable efforts have to be made to come to a better understanding of the importance of trust in the context of organized crime. What future research and theory construction will have to produce is threefold:
1. a clear conceptualization of what trust is and does for organized criminals;
2. a clear understanding of how trust emerges among criminals and how it contributes to the emergence and continued existence of patterns of criminal cooperation, and
3. valid and reliable data on the concrete presence of trust as a binding force in criminal relations.

The purpose of this paper is to tentatively explore where this endeavor may lead and how it can potentially contribute to a better understanding of organized crime in general. Drawing on the general sociological literature, we provide a preliminary conceptualization of trust in the context of organized crime that is centered around a fourfold typology along the micro-macro dimension, and we present some anecdotal evidence from our research on illegal markets for highly taxed goods in Norway and Germany to illustrate that there are different types of trust, that there are different consequences of the violation of trust, and, finally, that there are criminal relations that are not based on trust at all.

Conceptualizations of Trust

The issue of trust has received considerable attention in the social sciences over the past 10 to 15 years (see e.g. Fukuyama 1995; Gambetta 1988b; Laucken 2001; Misztal 1996; Seligman 1997). Still, trust continues to be "a very imprecise and confusing notion" (Misztal 1996: 9). Quite similar to the concept of organized crime, the use of the term "trust", as Seligman (1997: 7) has observed, "tends to be loose and imprecise".

Trust has to do with how people cope with risk and uncertainty. One person trusts another person where effective means to control the other's behavior are unavailable and where the other's behavior is potentially harmful in a broad sense of the word, and still the trusting person is confident that no harm will be done. Trust is seen as an essential component of social life. "Without trust", Niklas Luhmann argues, "only very simple forms of human cooperation which can be transacted on the spot are possible, and even individual action is much too sensitive to disruption to be capable of being planned, without trust, beyond the immediately assured moment. Trust is indispensable in order to increase a social system's potential beyond these elementary forms" (Luhmann 1979: 88).

The Rationality-Irrationality Dimension

There are at least two dimensions along which conceptualizations of trust vary. One dimension refers to the level of rationality or irrationality in the decision of a trusting person to trust. The other dimension, ranging from micro to macro level, refers to the social context in which trust emerges.
In the spectrum from rationality to irrationality, trust takes up a space somewhere between purely rational calculation of probabilities and irrational blind faith. Towards one end of the spectrum we find proponents of rational-choice theory like James S. Coleman, who assumes that a "rational actor will place trust (...) if the ratio of the chance of gain to the chance of loss is greater than the ratio of the amount of the potential loss to the amount of the potential gain" (Coleman 1990: 99). Outside the rational-choice paradigm there seems to be broad agreement that trust by definition involves some element of arationality or irrationality because the notion of trust, it is argued, can only have meaning in situations of uncertainty in which the trusting person has incomplete knowledge of the probabilities of the trusted person's behavior. Trust, then, is not exclusively based on the weighing of probabilities, but also on such factors as emotions, sentiments and culturally induced values (Dunn 1988; Fukuyama 1995; Gambetta 1988b; Misztal 1996; Simmel 1992[1908]). As Anthony Giddens (1990: 33) has put it: "All trust is in a certain sense blind trust!"

The Micro-Macro Dimension

The second major dimension in the conceptualization of trust, which may be termed the micro-macro dimension, pertains to diverging views on the allocation of trust in society. It ranges from trust in persons to trust in abstract systems (Misztal 1996: 72; Seligman 1997: 18). Some, namely proponents of rational-choice theory, discuss trust primarily in the context of face-to-face relations between a trustor and a trustee (see Coleman 1990). Others focus on what has been termed abstract or institutional trust. In these cases, trust is placed in the functioning of macro-level institutions (Giddens 1990; Luhmann 1979; 1988). Individuals are only recognized as agents who perform certain institutionally prescribed roles (Seligman 1997: 19). In between these two extremes, there are intermediate levels of observation with different forms of socially induced trust, for example trust that relies on generalizations about certain categories of people. Depending on whether or not trustor and trustee belong to the same social group, this kind of trust can be rooted either in stereotypes or in a sense of shared habits, norms and values. In the latter case, this can be simply a matter of inferring from generalizations to the likely behavior of a particular individual (the same would be true for trust based on stereotypes), but it has also been argued that belonging to the same social group can create moral obligations to place trust in others (Fukuyama 1995). Trust, then, "is a moral commandment to treat people as if they were trustworthy" (Uslaner 2002: 3).

Defining Trust

The most appropriate frame of reference for discussing trust in the context of organized crime, we would like to argue, is a network approach. We view trust first of all as a property of dyadic relations that form the basic elements of criminal networks. Criminal networks can be defined as sets of actors that are connected by ties which in some way or other support the commission of illegal acts (von Lampe 2001). Understood in this way, criminal networks constitute "the least common denominator of organized crime" (McIllwain 1999: 304) and should therefore be taken as the key empirical referent of the concept of organized crime (Coles 2001; Hobbs & Dunnighan 1998; Ianni 1975; Johansen 1996; von Lampe 2001; Morselli 2001).

The dyadic relations we are concerned about involve a trusting person P, the trustor, and a trusted other person O, the trustee. For reasons of simplicity, we specifically focus on the perspective of the trusting person while leaving aside the question of how the trusted person perceives, interprets and copes with the situation. Within this narrow framework, we define trust as the expectation of P, under conditions of uncertainty, that (1) O will not harm P, even though (2) O could harm P (see also Dunn 1988: 74; Gambetta 1988b: 219). This expectation can be based on a variety of more or less reliable perceptions, assumptions and attitudes that pertain to the trusted person and to the social environment in which trustor and trustee interact.

The quintessential situation in which trust is an issue in the context of organized crime is that of a collaborative criminal venture involving P and O as co-offenders. By saying that P trusts O in a cooperative criminal venture we are implying that P expects that
O will not violate the specific terms stipulated in the particular explicit and implicit agreements regarding the venture, that
O will not violate general expectations of considerate behavior, and that
O will protect the secrecy of the venture (with regard to law enforcement, as well as to competing and predatory criminals).

But co-offending is not the only constellation where the notion of trust may come into play. From an analytical perspective there is a broad spectrum of situations that have to be taken into account, including more passive relations, even to the extent that criminal actor P deals with O as a mere bystander. Here the trust P places in O relates to the expectation that
O will not interfere with P's actions, and that
O will not alert others (e.g. the police) to P's actions.

Between the two poles of cooperative relations on the one end of the spectrum, and passive relations on the other, there are several intermediate levels of interaction imaginable between trustor and trustee, for example, that relating to communication relations through which criminally relevant information is passed on between P and O.

A Typology of Trust in Criminal Relations

The first step in the analysis of a given criminal relation would be to examine whether or not trust plays a role at all. The potential alternatives to trust are a relation characterized by either a lack of trust or by mistrust. Lack of trust refers to a situation in which P perceives that O could harm P, but P has no basis on which to determine whether or not O will harm P. In contrast, mistrust exists when P perceives O to be capable and willing to harm P.

Provided there is a basis of trust in the relation between P and O, the crucial question is: on what grounds does P expect O to act in an agreeable way? We think that this question is heuristically tremendously valuable as it opens up the view to a complex of personal, relational and contextual factors that potentially contribute to the emergence and continued existence of criminal relations.

To systematize the analysis, we propose a typology modeled on the basic dimensions characterizing the sociological debate on trust. Our typology distinguishes different bases of trust along a micro-macro dimension, and within each category we consider variations in the rationality of the decision to trust.

Individualized Trust

The first category in our typology involves individualized trust. The expectation of agreeable behavior relates specifically to the trustee as an individual. The basis for individualized trust can be quite diverse. It may lie in previous observations of the trustee's behavior, characteristics and dispositions, and the more or less rational assumption that the trustee will act accordingly in the future (Misztal 1996: 76); it may lie in equally rational expectations of how the trustee reacts to sanctions (Coleman 1990: 115); or the motivation to trust a particular individual can be irrational in that it is the result of affections the trustor feels for the trustee (Gambetta 1988b: 232; Huemer 1998: 121; Misztal 1996: 21). Criminally exploitable individualized trust can primarily be expected to emerge from continuous interaction in delinquent peer groups or in prison settings, out of which affectionate bonds and a sense of predictability may develop.

Another aspect that needs to be taken into account is that trust, especially individualized trust, may be mediated. In this case no direct bond of trust exists between P and O. Instead, P and O are each connected to an intermediary who may take on the role of advisor, guarantor, or entrepreneur (Coleman 1990: 180). In the advisory role, the third person passes a judgment about O on which P relies. As guarantor, the third person will compensate P for losses incurred as a result of O's disloyalty. In the entrepreneurial function, the intermediary "induces the trust of several trustors and combines these resources, ordinarily placing them in the hands of one or more other actors who are expected to realize gains for" the trustors. In the latter two cases, the trustor places trust in the intermediary's performance capability and integrity (Coleman 1990: 180-2).

Trust Based on Reputation

The second category pertains to trust based on reputation. Here the trustor relies on publicly formed and held opinion about the trustee (Dasgupta 1988: 54). This type of trust hinges on the flow of information. Information may be dispersed in a context associated with illegality, for example the underworld "grapevine system" first described by Frederic Thrasher (1963[1927]: 285), but one can also think of criminal reputation being created by the media, possibly in concert with law enforcement agencies. It has been argued, for example, again drawing on the historical case of Prohibition-era Chicago, that Al Capone owed much of his (undeserved) reputation of being an all powerful underworld leader to the media (Bergreen 1994: 211).

From the point of view of rationality vs. irrationality, relying on reputation can of course be rather thoughtless and careless, depending on how sustained and how well tested the reputation is. On the other hand, reputation can be a valuable asset and from the trustor's perspective it may be quite rational to assume that the trustee will not want to put his or her reputation in jeopardy through disloyal behavior (Misztal 1996: 121, 126).

Generalized Trust

The third category, generalized trust, comprises constellations in which trust is linked to social groups rather than to a particular individual. The trustor places trust in the trustee based on the presumption that the trustee conforms to some more general norms or patterns of behavior, for example codes of conduct such as mutual support and non-cooperation with law enforcement that the trustee can be expected to share as a member of a delinquent subculture or a mafia-like fraternal association. Placing trust in another member of a deviant community can be quite rational, especially in the case of fraternal associations. Members of fraternal associations such as the Cosa Nostra can be expected to adhere to a set of mutually understood rules of conduct which in turn are ensured by selective recruitment combined with a period of testing and schooling, rigid enforcement of discipline, male bonding rituals, and secrecy (Haller 1992: 3-4; Jacobs 1994: 102). Secrecy, as Georg Simmel (1992[1908]: 448) has stressed, can have a binding force in and by itself by blocking members from outside interference.
Some of these factors can also bind members of a deviant subculture. Yet they will probably do so to a lesser degree. By definition, members of a deviant subculture share a set of norms and values and are likely to harbor feelings of solidarity in view of a hostile outside world. This has been noted, for example, in the case of the Western European drug cultures of the 1960s and 1970s (Ruggiero & South 1995: 134). Another classic example is found in the community of professional thieves described by Edwin Sutherland (1937: 202-6). Trusting another member of a deviant subculture, then, can be seen as a rational decision. On the other hand, placing trust out of a subculturally induced moral obligation to trust can be rather irrational. The same is probably true for relations between an outsider as the trustor who relies on more or less stereotypical expectations relating to an insider as the trustee.
On a more mundane level, but perhaps with the most rational justification to it, a trust producing sense of predictability may arise in routine situations that are governed by habit rather than by conscious deliberation (Garfinkel 1963). What Harold Garfinkel has emphasized about the relevance of unspoken rules in daily life could also be valid for everyday interactions under conditions of illegality, for example routine transactions of illegal goods.

Abstract Trust

The fourth category refers to trust that is placed in abstract systems that set and maintain certain basic conditions, for example the government, the monetary system or the medical system in society at large (see Giddens 1990; Luhmann 1979: 1988). While these institutions have no direct counterpart in the illegal spheres of society, it appears to be overly strict to say that, due to the lack of a public power, there is no systemic trust and therefore that under conditions of illegality "trust necessarily has only a personal basis" (Paoli 2002: 84). There are documented cases where criminal groups have exercised territorial control in the form that they imposed and enforced certain rules of conduct for criminals who operated in a given area (Abadinsky 1981; Gambetta 1996). It can be assumed that such a quasi-governmental framework shapes and influences behavior and thus contributes to a sense of predictability and the emergence of trust.

It should be noted that the four categories are not mutually exclusive. On the contrary, it can be expected that a given trust relation rests on different bases of trust (Misztal 1996: 21). Conversely one can think of social settings within which various trust building factors take effect.

Before the background of our own empirical research into the bootleg liquor market in Norway and the illegal cigarette market in Germany we want to briefly discuss four social settings with regard to their trust building potential: family, local community, ethnic community and business. Then we will look at some empirical constellations we have discovered that cannot be readily explained by conventional conceptions of trust. But first of all we provide a short overview of our objects of study which we explore using primarily interviews in the case of the alcohol black market in Norway (Johansen 1994; 1998; forthcoming) and judicial files in the case of the cigarette black market in Germany (von Lampe 2002; 2003).

The Basic Structures of the Alcohol and Cigarette Black Markets

The black market for alcohol in Norway and the black market for cigarettes in Germany, respectively, deal with commodities that are essentially legal but are subject to high taxation. Both markets can be divided into roughly three levels, procurement, whole sale, and retail. On the top level, the procurement level, alcohol and cigarettes, respectively, are obtained for black market distribution from low-tax or no-tax sources. These sources include manufacturers that sell their products for export and thus with no taxes added to the sales prices, or outlets in relatively low-tax countries such as, for example, Poland. In addition, some of the alcohol illegally distributed in Norway is also illegally manufactured domestically.
On the whole sale level, bulk loads are stored, divided up and distributed in more manageable consignments to retail dealers.
On the retail level, finally, the commodities are passed on to consumers. Retail distribution may take on clandestine forms, very similar to the distribution of, say, stolen goods, it may take place in the form of under-the-counter sales in otherwise legal outlet stores, as is the case in some countryside grocery stores that sell bootleg liquor in Norway, or retailing may take place in the open, as is the case in some places in Eastern Germany, where mostly Vietnamese asylum seekers sell contraband cigarettes as street vendors.

On each of these levels, within particular illegal enterprises, and between different levels of the market pyramid, actors cooperate on a more or less continuous basis and with more or less success, and in every instance the question arises: Is trust the binding force that brings and holds these actors together? And if so, on what basis does one criminal actor place trust in another?

Trust Producing Social Settings

Family

According to a widely held notion, criminal relations in general and within illegal enterprises in particular tend to be embedded in kinship ties. This view is supported by general assumptions about the link between family and trust. Anthony Giddens (1990: 101), for example, suggests that "kinspeople can usually be relied upon to meet a range of obligations more or less regardless of whether they feel personally sympathetic towards specific individuals involved". Trust in family members, it is argued, rests on familiarity and conformity, i.e. on individualized trust growing out of continuous interaction, and on generalized trust based on a sense of similarity and shared norms and values (Misztal 1996: 39, 157, 171).
However, the link between kinship ties and criminal relations via family based trust may not be as easily established as is commonly assumed. There is, first of all, the aspect of inner family conflicts. The assumption of family based trust hinges on a somewhat romantic notion of family harmony that does not necessarily live up to reality. In fact, some family based illegal enterprises may be more the product of considerations of conflict avoidance than an outgrowth of trust relations. As Pino Arlacchi in his analysis of Southern Italian Mafia groups has observed, "often relations between families of brothers are marked less by co-operation and solidarity than by disagreement and impulses towards mutual conflict". One strategy for dealing with the constant threat of disintegration is, according to Arlacchi, "the institution of forms of outright 'family communism', based on common ownership and control - by the members of the cosca's innermost nucleus - of a good part of the group's possessions and economic activities" (Arlacchi 1986: 135-6).
The other issue that needs to be addressed here both theoretically and empirically is the question of "borrowed loyalty": Under what circumstances do long-term relations with no illegal connotation, as is the case with familial ties, can become a basis of trust for criminal cooperation? In fact, our findings do not indicate a dominant role of kinship structures in the black markets we study. Among alcohol smugglers in Norway, the most common pattern of cooperation where kinship ties are apparent at all are father-and-son teams, sometimes structured on equal terms. And in some instances, smugglers have been found to receive moral and logistical support from their wives and families. One Norwegian bootlegger recollected in an interview: "My wife and I, we have always been together. I go nowhere or do nothing without her. We did our first deals in the 50s - went to the loan shark with our wedding rings to raise money for our first investment in cigarettes and booze... But my son, by the way, is a doctor." In the cigarette black market in Germany, likewise, cooperation between husbands and wives appears to be the most common type of family based criminal relations.

Local Community

Similar to families, local communities may produce trust through familiarity and conformity (Giddens 1990: 101; Luhmann 1988: 94). In fact, in the case of the illegal alcohol market, close-knit, local communities in rural Norway appear to be a more significant trust factor than the immediate family. In these communities moonshining is widespread and disloyal behavior would be directed not only against a business partner but against the entire community. Under such conditions, we hypothesize, family ties, to the extent they are criminally relevant at all, provide no added value in trust relations and are therefore no prominent feature of cooperative structures.

Ethnic Community

Perhaps more than family and local community, the role of ethnicity as a basis of trust has been stressed in the organized crime literature. The link between ethnicity and trust is fairly easy to establish where close-knit ethnic communities exist, because here the same notion of trust created by familiarity and conformity would seem to apply (Misztal 1996). Moreover, ethnic communities will tend to be characterized by a stronger sense of "we and they". This means that placing trust and cooperating within an ethnic community is not only a reflection of greater cohesion in response to outside pressure. It may also be a reflection of reservations and suspicions maintained by members of ethnic communities that work against communication and cooperation across ethnic boundaries. Members of ethnic communities may willfully withdraw from the larger society out of concerns that outsiders do not share their values and will therefore not treat them fairly (Uslaner & Conley 2003: 333). At the same time, this means that too much reliance on ethnic ties (or kinship or community ties, for that matter) can become a liability because it limits opportunities for cooperation (Pearson & Hobbs 2001: 28).

Where intra-ethnic relations are not embedded in close-knit communities, the trust building role of ethnicity is less clear. What would have to be assumed is that a sense of similarity is generally present in the interaction between people of the same ethnic background, and that their behavior will therefore be predicted with greater confidence (see Hardin quoted in Misztal 1996: 134).

With regard to the role of ethnicity as a trust producing factor under conditions of illegality, it needs to be stressed, first of all, that in many instances where bonds of trust are attributed to ethnic cohesion, the actual foundation may lie in social relations like kinship, friendship or community ties (Ianni 1975: 282; Kleemans & van de Bunt 1999: 25; Potter 1994: 121). However, ethnicity cannot be completely ruled out as a trust variable, considering that the marginalization of ethnic minority groups may eliminate moral precepts while at the same time creating internal solidarity. These factors may be reinforced by social norms of keeping matters secret from outsiders and by a high cultural value placed on loyalty (Bovenkerk 1998: 121-2). As such, a criminal actor may come to trust another simply on the grounds that they share the same ethnic background.

In our research we have found a significant difference in the importance of ethnicity between the bootleg liquor market in Norway and the cigarette black market in Germany. Whereas in Norway, members of ethnic minorities play only a marginal role at best, in the cigarette black market in Germany certain market levels are clearly dominated by particular ethnic groups. The procurement and whole sale levels tend to be occupied by Polish smugglers and dealers whereas the street sale of contraband cigarettes in East Germany, as mentioned, has for several years been almost exclusively in the hands of Vietnamese.

An illustrative example of the relevance of ethnicity is provided by the relation of Vietnamese dealers to other members of the Vietnamese community. As a legacy of East German politics, the Vietnamese community had been concentrated in large housing projects isolated from the German population. Within these dormitories, whole sale and retail dealers could openly store, transport and sell untaxed cigarettes without fear of being reported to authorities by their fellow countrymen. While in the instance of a particular apartment, the other occupants might have been loyal based on family and friendship ties, a general sense of solidarity seems to have prevailed among all Vietnamese living in these dormitories. No cooperation was sought with the German authorities despite the inconveniences connected with the frequent searches of the premises conducted by police and customs service.

Evidence could also be collected on how networks of Poles living in Poland and in Germany have been used to disperse information related to the smuggling and distribution of contraband cigarettes (von Lampe 2003: 58). Here, some of the factors that characterize the relations between ethnic minority community and host society may also characterize the asymmetric relations between Poland as a country of transition and Germany as an advanced capitalist country. This notion is supported by findings about cigarette smuggling ventures undertaken from Poland to Germany that have apparently been discussed and planned in Poland without concern for secrecy even among casual acquaintances.

The most striking aspect about the dominating role of certain ethnic groups in the cigarette black market, however, is not that they foster or at least facilitate criminal relations, but that there is an apparent ease with which ethnic cleavages are being bridged by criminal actors. Two constellations are noteworthy in this regard: the relation between Polish suppliers and Vietnamese dealers and between Vietnamese street vendors and German customers.

From the available evidence it seems that Polish cigarette smugglers have initially taken the risk of directly approaching potential Vietnamese customers without prior introduction by a third party or any other preceding connection. In the early 1990s, Polish suppliers typically established these contacts by more or less randomly soliciting customers in front of housing projects occupied by Vietnamese. Apparently, both sides were similarly confident that members of the respective other ethnic group would be trustworthy. In a legal business context, relations across ethnic and language barriers are generally believed to be difficult to establish and maintain (Neubauer 1997). There is great potential for gross misunderstanding where members of different cultures are oriented towards radically different sets of background assumptions and beliefs systems, but are unaware of these differences (Good 1988: 45-6).
Interestingly, however, quite the opposite may be true in the case of illegal business since the respective other's status as a foreigner minimizes the possibility of his or her cooperating with the authorities. Thus, the original relation between Polish suppliers and Vietnamese dealers may provide an example of trust based on ethnic stereotypes.

In the relation between German customers and Vietnamese street vendors the same mechanism may be in place. It seems plausible to assume that there is a general belief among Germans that these Vietnamese vendors can be trusted, particularly because they are unlikely to be covert customs officers, given the marginalized status of the Vietnamese community in East Germany (von Lampe 2002). Yet another trust building factor may be at work in the relation between German customers and Vietnamese vendors that should be mentioned in a comprehensive analysis, the routinization of the street sale of contraband cigarettes. These exchanges are publicly repeated in the same fashion over and over again so that a given customer will most likely not anticipate any deviation from this norm when he or she decides to approach a vendor. In the end one can hypothesize that the cultural and ethnic cleavages that separate German customers and Vietnamese vendors of untaxed cigarettes are bridged by a combination of trust based on ethnic stereotyping and trust in the continuity of a routine situation.

Business

While ethnicity cannot be ruled out as a trust building factor, at least in the case of the cigarette black market in Germany, there is one social setting that appears to have a much greater significance for the emergence of criminally relevant trust, but one that is seldom explicitly mentioned in the organized crime literature: legal business. Like the other settings discussed so far, networks of legal business relations, we assume, tend to be characterized by more or less intense cultural cohesion, patterns of repeated interaction, and transparency through social and geographical proximity. They are also likely to be linked by a well functioning internal communication system for the quick and easy dissemination of information. In such an environment, trust can be expected to be the result of a combination of factors like affectionate bonds, observations of personal conduct, reputation, and the reliance on shared norms and values.
Our findings suggest that this kind of trust can facilitate criminal cooperation. Both bootlegging in Norway and the trafficking in untaxed cigarettes in Germany are closely linked to legal business, namely the transportation sector. In our research we have found criminal relations growing out of existing or previous employer/employee relations, between employees of a legal firm, and between independent business partners. In several instances, we found that employees were gradually introduced to illegal activities taking place under the guise of legal business activities after they had initially been kept ignorant of the criminal background of their employers. In one case a businessman invited several of his workers to go into alcohol smuggling after his legal firm went bankrupt. An example for the exploitation of legal business contacts is provided by one interviewed Norwegian alcohol smuggler who stated that he could take advantage of the reputation he had gained as a legal entrepreneur for paying his debts on time for obtaining credit in the bootlegging business. A second example stems from a case involving the distribution of several container loads of contraband cigarettes in Germany, organized by a network expanding from the Netherlands and Germany to Poland and Lithuania. The illegal activities in this case apparently took place parallel to the cross-border trading of legal products such as salad oil.
Yet another essentially legal business relation with potential relevance for illegal transactions exists in Norway between customers and retail stores, e.g. grocery stores, which sell bootleg liquor on the side. Trust can here be expected to stem from experience, and from the perception that the retailer has a reputation to protect. For, if the grocery store were to sell poor quality liquor, this would have repercussions not only on future illicit sales but also on the sale of legal products, and on the reputation of the store keeper within the community.

Violation of Trust

It is a matter of further research to explore how the different social arenas relate to the emergence of trust relations and different levels of trust. What seems clear, however, is that no basis of trust is strong enough to rule out the possibility of betrayal (see also Zaitch 2002: 278). Accordingly, the analysis needs to focus not only on trust, but also on the conditions for, and the consequences of, the violation of trust.

No consequences

Our research, as well as organized crime research in general, suggests that the violation of trust can have very different consequences; which means that the overall picture becomes even more complicated. In fact, in some instances, the violation of trust may not entail any consequences. Consider a situation in which the trusting person P remains unaware of the disloyal behaviour of O, or a situation in which P is aware of some foul play but does not attribute it to O. One of the 'big-shots' of Norway's bootleg liquor business during the 1990s provides an illustrative example. Members of his network continued to cooperate despite some set-backs because they failed to realize that the reason for their failures was that the 'big-shot' himself occasionally informed on accomplices in order to fend off criminal investigations directed against him.
But even in cases where disloyal behavior is recognized, the transgression does not automatically trigger any response. The reasons for not responding can be manifold, including a lack of motivation, capacity or positive incentive on the part of P for seeking retribution. But it may also be the case that P has no alternative to continued cooperation with O. In the illegal alcohol market in Norway, where informing on others seems to be the exception rather than the rule, we were able to obtain information on market participants who have been known grassers for years and yet who remained untouched.

Responses

Likewise, when P does react to disloyal behaviour of O, the response does not necessarily consist of a termination of the cooperation with O, either by severing the ties to O or by eliminating O. Intermediate reactions include continued interaction with O on a lower level of risk, for example, by reducing the stakes in the cooperation. Otherwise, retribution may be aimed at ensuring a more successful cooperation with O in the future.

The level of violence is low in both the illegal alcohol market in Norway and the illegal cigarette market in Germany. In Norway, some cases of violence and threats of violence are documented in connection with debt collection and the sanctioning of police informers. In Germany, violence has played a role in the extortion of street vendors of contraband cigarettes. Other isolated cases involved rip-offs and the mock execution of a suspected police informer. Apart from that, only one case of violence could be found where disputes between market participants might have been an issue (von Lampe 2002: 157).
There may be several reasons for the relative absence of violence in these illegal markets. First, the share of violent-prone actors is likely to be relatively small among participants in illegal markets that display some overlap with legal spheres of business and enjoy a certain level of tolerance in society. Second, black markets for highly taxed goods tend to be characterized by low density, i.e. a high demand is met by a relatively small number of suppliers so that there is enough for everyone. Finally, the absence of structures that give rise to status conflict, such as the succession problem in mafia-like fraternal organizations, will contribute to a "non-violent" business culture. A Norwegian bootlegger seems to represent a common view by stating: "Make yourself a killer just because a guy turned you down on half a million? Stupid!"

Cooperation Without a Basis of Trust

Just as there are patterns of criminal cooperation that endure violations of trust, we also find cooperative relations among criminals that either seem to lack an initial basis of trust or appear to be characterized by outright mistrust.

Delimitation: Precarious Bases of Trust

At this point it must be stressed that trust as we conceptualize it, is a matter of subjective perceptions and expectations and therefore its presence or absence should not be judged from the angle of objective rationality. For the analysis of criminal relations this means that a sufficient basis of trust may exist even though it would seem too weak from an objective observer's point of view. One has to take into account that decisions under uncertainty tend to be based on biases and misleading intuitions (Tversky & Kahnemann 1982). No less, it seems safe to say, is true for trusting behavior (Gambetta 1988b, 232). As David Good has pointed out, "many superficial aspects of personal presentation can quickly lead to conclusions about the nature of another person's beliefs and sentiments. On first seeing someone, we immediately classify that person according to age, sex, and many other social categories. From this, and the briefest of details from our interaction, we rapidly draw inferences about his or her beliefs and the relationship between us" (Good 1988: 45). Thus, trust-based cooperation may occur on a very precarious basis and without much of a past history.

This may account, for example, for on-the-spot recruitments that have been documented on the whole-sale level of the cigarette black market in Germany. In these cases, illegal entrepreneurs recruited persons for the transport of contraband cigarettes in the course of chance meetings in bars (von Lampe 2003: 52). It can be hypothesized that these encounters are sufficient for actors to form an opinion about the other's trustworthiness based on some tentative assessments.

Trusting as Testing of Trustworthiness

These constellations notwithstanding, there seem to be instances where cooperation occurs without a basis of trust. There are essentially four types of cases that can be distinguished.
The first one does not readily fit in the categories of trust, absence of trust and mistrust. It involves decisions to engage in trusting behavior with no grounds for expecting the other's trustworthiness. Still, the trusting behavior appears rational because it serves to gain sympathy and to explore the other person's disposition. The trustee is given the opportunity to adopt a 'tit-for-tat' strategy by responding in kind and thereby demonstrating that he or she will not take advantage of the trustor's vulnerability. Actors who follow this path will typically make an initial cooperative move under circumstances where the risk is acceptably low (Gambetta 1988b: 234). Repeated cooperation on a low or only gradually increased level of risk may then eventually lead to individualized trust (Good 1988: 44).

Risking Trust Under Adverse Conditions

The second type of cooperation without a real basis of trust pertains to desperate situations from which a person cannot extricate him- or herself without help (Coleman 1990: 107-8; Gambetta 1988b: 223-4). In such adverse situations, criminals may feel that they simply have no choice but "to risk the trust that cooperation demands" (McCarthy et al. 1998: 174).

Fatalism and Thrill-Seeking

Finally, there are cooperative relations among criminals without a basis of trust or in the presence of outright mistrust that cannot be easily explained by any rational consideration. From the participants' perspective, it seems that the potential risks that lie in such fragile, cooperative ventures are often simply ignored, and where they are recognized they may be fatalistically accepted as an unavoidable fact of life or even welcomed in a gambler's thrill-seeking attitude (Adler 1985: 85).
Several interviewed Norwegian alcohol smugglers, for example, expressed a lack of concern for the dangers emanating from the collaboration with unreliable accomplices. "Sooner or later all crime operations will come to an end," one said, "but if you have a swell time before you are caught, it's worth it."
Many of these relations have a short time horizon, either because the participants lack the talent and patience to establish continuous, long-term operations, or because carelessness and outright dishonesty lead to failure. In fact, it would not come as a surprise to find most criminal ventures end in a fiasco. At the same time, it should be noted that cooperative criminal ventures undertaken by disloyal and careless partners can be successful against all odds, for example, because law enforcement agencies lack the capacity to follow up on leads.

Functional Alternatives to Trust

More risk-conscious actors take precautionary measures to safeguard against disloyal behavior. These functional alternatives to trust can be divided into three broad categories, measures that reduce the possibility of betrayal, measures that increase the costs of betrayal, and measures that aim at building trust.
Measures that reduce the possibility of betrayal include certain procedural arrangements like testing and counting merchandise before the transaction is made. Other common strategies in cases where actors cooperate without the security of pre-existing ties are anonymity and segmentation. In the study of the illegal cigarette market in Germany, as mentioned, a number of cases were found in which very coincidental acquaintanceships provided the basis for criminal cooperation, though mostly in connection with the assignment of subordinate tasks. In these cases, safeguards against deceit were anonymity of the recruiting person vis-à-vis the recruits, and sometimes a disguising of the true nature of the activity until the last moment. In one case, the recruits had been told they were hired to unload a truck full of vegetables. As it turned out, the vegetables only served as a disguise for a large shipment of smuggled cigarettes. Interestingly, persons that have been recruited on an ad-hoc basis prove to be willing witnesses, but the evidence they are capable of providing is typically insufficient to identify and prosecute those in charge, at least with the limited resources available to the German customs service.
Measures that increase the cost of betrayal include, most notably, the threat and use of violence. In our conceptualization, however, this is not so much a matter of alternatives to trust than of the reliability of different bases of trust, because we assume that the expectation of the effectiveness of the threat or use of violence is similar to any other expectation that the trustee will be loyal. At the same time it should be emphasized that assuring compliance by use of force is costly and in the long run is probably no adequate alternative to trust that is rooted in less drastic considerations (Gambetta 1988b; Reuter 1983).
Finally, one precautionary measure that criminals with a broader time horizon can be expected to adopt in the absence of trust, is to screen, test and school potential accomplices in an effort to establish a basis of trust for the future.

Conclusions

Going back to the truism of trust as the basis of criminal relations the conclusions from our observations are as follows. First, that trust is an empirically and theoretically significant variable for understanding organized crime, but it is a multifaceted phenomenon which stands in the way of easy explanations. Second, trust provides no exhaustive explanation because criminal cooperation may also occur in the absence of trust or even between mistrusting actors. Third, where trust provides a binding force for criminal relations, the foundations on which that trust is based may be quite diverse. Kinship, ritual kinship and ethnicity are only some of the aspects that need to be taken into consideration. Fourth, the violation of trust does not necessarily entail a drastic response in the form of a termination of the criminal relation or the elimination of the disloyal actor. Finally, from a researcher's perspective, we find the concept of trust heuristically valuable since it raises a number of questions that deserve closer attention. We believe that these questions require the combination of different perspectives on organized crime, including psychological, anthropological, and sociological approaches, since trust appears to be a function of both individual and social variables. For example, the expectations of a trusting person that another person shall be loyal can be influenced by factors such as the trustee's personality and previous actions, the norms and conventions that govern both persons' behaviour, and the trustor's ability and keenness to obtain and rationally process relevant information on the trustee. Of course, our discussion has been limited to only a fraction of the overall picture by focusing exclusively on the subjective view of one of two persons in a dyadic relation. A more comprehensive approach will have to consider both sides, as well as the immediate and broader socio-structural context.
Some of the key questions that remain to be answered concern the criminal exploitability of trust that has emerged in relations with no illegal connotation, the comparative strength and vulnerability of different bases of trust, the interplay of different levels of risk, hostility, adversity and trust, and the conditions under which criminal cooperation occurs despite a lack of trust, mistrust or a violation of trust, especially with regard to functional alternatives to trust.
It must be stressed that the illegal markets we are studying exist in environments that are characterized by relatively low degrees of hostility. We hypothesize that under favorable social conditions even negligent and foolish actors have a good chance to succeed in cooperative criminal ventures, whereas under hostile conditions the negligent and foolish are weeded out more quickly.
Regarding international comparative research, we believe it would be worthwhile to explore whether actors belonging to "low-trust societies" are better prepared for operating in a criminal environment than actors molded by "high-trust societies" such as those found in Germany and Norway (Fukuyama 1995). The underlying notion is that people socialized in "low-trust cultures" may be naturally, better endowed to cope with the absence of trust. On the other hand, while bonds of trust that do exist under these conditions could be particularly strong and strain resistant, the reluctance in "low-trust cultures" to place trust beyond the limits of kinship and local communities may prove a disadvantage in illegal markets where cooperation across ethnic and cultural cleavages is a necessity.

References

Abadinsky, Howard (1981), The Mafia in America: An Oral History, New York: Praeger.
Adler, Patricia (1985), Wheeling and Dealing: An Ethnography of an Upper-Level Drug Dealing and Smuggling Community. New York: Columbia University Press.
Arlacchi, Pino (1986), Mafia Business: The Mafia Ethic and the Spirit of Capitalism. London: Verso.
Bergreen, Laurence (1994), Capone: The Man and the Era, New York: Simon & Schuster.
Bonanno, Joseph, with Sergio Lalli (1983), A Man of Honor: The Autobiography of Joseph Bonanno. New York: Simon and Schuster.
Bovenkerk, Frank (1998), 'Organized Crime and Ethnic Minorities: Is There a Link?', Transnational Organized Crime, 4(3&4): 109-126.
Coleman, James S. (1990), Foundations of Social Theory, Cambridge, MS: The Belknap Press of Harvard University Press.
Coleman, James W. (1989), The Criminal Elite: The Sociology of White Collar Crime. New York: St. Martin's Press.
Coleman, James W. (1987), 'Toward and Integrated Theory of White Collar Crime', American Journal of Sociology, 93(2): 406-439.
Coles, Nigel (2001), 'It's Not What You Know-It's Who You Know That Counts: Analysing Serious Crime Groups as Social Networks', The British Journal of Criminology, 41(4): 580-594.
Dasgupta, Partha (1988), Trust as a Commodity, in: D. Gambetta, ed., Trust: Making and Breaking Cooperative Relations, 49-72. New York: Basil Blackwell.
Dunn, John (1988), 'Trust and Political Agency', in: D. Gambetta, ed., Trust: Making and Breaking Cooperative Relations, 73-93. New York: Basil Blackwell.
van Duyne, Petrus C., Pheijffer, Marcel, Kuijl, Hans, van Dijk, Arthur, and Bakker, Gerard (2001), Financial Investigation of Crime: A Tool of the Integral Law Enforcement Approach. The Hague, Koninklijke Vermande.
Fukuyama, Francis (1995), Trust: The Social Virtues and the Creation of Prosperity. New York: The Free Press.
Gambetta, Diego (1988a), 'Mafia: the Price of Distrust', in D. Gambetta, ed., Trust: Making and Breaking Cooperative Relations, 158-175. Oxford: Basil Blackwell.
Gambetta, Diego (1988b), 'Can We Trust Trust?', in D. Gambetta, ed., Trust: Making and Breaking Cooperative Relations, 213-237. Oxford: Basil Blackwell.
Gambetta, Diego (1996), The Sicilian Mafia: The Business of Private Protection. Cambridge, MS: Harvard University Press.
Garfinkel, Harold (1963), 'A Conception of, and Experiments with, Trust as a Condition of Stable Concerted Actions', in O. J. Harvey, ed., Motivation and Social Interaction, 187-238. New York: Ronald Press.
Giddens, Anthony (1990), The Consequences of Modernity. Cambridge: Polity Press.
Good, David (1988), Individuals, Interpersonal Relations, and Trust, in D. Gambetta, ed., Trust: Making and Breaking Cooperative Relations, 31-48. Oxford: Basil Blackwell.
Gottfredson, Michael R., and Hirschi, Travis (1990), A General Theory of Crime. Stanford: Stanford University Press.
Haller, Mark H. (1992), 'Bureaucracy and the Mafia: An Alternative View', in: Journal of Contemporary Criminal Justice, 8(1): 1-10.
Hobbs, Dick, and Dunnighan, Colin (1998), 'Glocal Organised Crime: Context and Pretext', in V. Ruggiero, N. South and I. Taylor, eds, The New European Criminology: Crime and Social Order in Europe, 289-303. London: Routledge.
Huemer, Lars (1998), Trust in Business Relations: Economic Logic or Social Interaction. Umea: Borea Bokförlag.
Ianni, Francis (1975), Black Mafia: Ethnic Succession in Organised Crime. London: New English Library.
Jacobs, James B. (1994), Busting the Mob: United States v. Cosa Nostra. New York: New York University Press.
Johansen, Per Ole (1994), Markedet som ikke ville d�: Forbudstiden og de illegale alkoholmarkedene i Norge og USA. Oslo, Norway: Rusmiddeldirektoratet.
Johansen, Per Ole (1996), Nettverk i gr�sonen: Et perspektiv p� organisert kriminalitet. Oslo, Norway: ad Notam Gyldendal.
Johansen, Per Ole (1998), 'Smuggling Alcohol: Organized Crime the Norwegian Style', in Institutt For Kriminologi Universitetet I Oslo, ed., �rsrapport 1998, 1-11. Oslo, Norway: Universitetet I Oslo.
Johansen, Per Ole (forthcoming), Herfra til Forbudstiden: Den illegale spriten 1914-1986. Oslo, Norway: Institutt for kriminologi og rettssosiologi.
Johansen, Per Ole, and von Lampe, Klaus (2002), Is there honor among crooks? On the importance of trust in criminal relations, paper presented at the annual meeting of the American Society of Criminology, Chicago, November 2002.
Kleemans, Edward R., and van de Bunt, Henk G. (1999), 'The Social Embeddedness of Organized Crime', Transnational Organized Crime, 5(1): 19-36.
Lacey, Robert (1991), Little Man: Meyer Lansky and the Gangster Life. Boston: Little, Brown and Company.
von Lampe, Klaus (2001), 'Organisierte Kriminalität unter der Lupe: Netzwerke kriminell nutzbarer Kontakte als konzeptueller Zugang zur OK-Problematik', Kriminalistik, 55(7): 465-471.
von Lampe, Klaus (2002), 'The trafficking in untaxed cigarettes in Germany: A case study of the social embeddedness of illegal markets', in P.C. van Duyne, K. von Lampe, N. Passas, eds., Upperworld and Underworld in Cross-Border Crime, 141-161. Nijmegen: Wolf Legal Publishers.
von Lampe, Klaus (2003), 'Organising the nicotine racket: Patterns of criminal cooperation in the cigarette black market in Germany', in P.C. van Duyne, K. von Lampe, J.L. Newell, eds., Criminal Finances and Organising Crime in Europe, 41-65. Nijmegen: Wolf Legal Publishers.
Laucken, Uwe (2001), Zwischenmenschliches Vertrauen: Rahmenentwurf und Ideenskizze. Oldenburg: Bibliotheks- und Informationssystem der Carl von Ossietzky Universität Oldenburg.
Luhmann, Niklas (1979), Trust and Power. Chichester: John Wiley & Sons.
Luhmann, Niklas (1988), 'Familiarity, Confidence, Trust: Problems and Alternatives', in D. Gambetta, ed., Trust: Making and Breaking Cooperative Relations, 94-107. Oxford: Basil Blackwell.
Marks, Howard (1998), Mr. Nice. London: Vintage.
McCarthy, Bill, Hagan, John, and Cohen, Lawrence E. (1998), 'Uncertainty, Cooperation, and Crime: Understanding the Decision to Co-offend', Social Forces, 77(1), 155-184.
McIllwain, Jeffrey S. (1999), 'Organized crime: A social network approach', Crime, Law and Social Change, 32(4): 301-323.
Misztal, Barbara A. (1996), Trust in Modern Societies: The Search for the Bases of Social Order. Cambridge: Polity Press.
Morselli, Carlo (2001), 'Structuring Mr. Nice: Entrepreneurial opportunities and brokerage positioning in the cannabis trade', Crime, Law and Social Change, 35(3), 203-244.
Neubauer, Walter (1997), 'Interpersonales Vertrauen als Management-Aufgabe in Organisationen', in M. Schweer, ed., Interpersonales Vertrauen, 105-120. Opladen: Westdeutscher Verlag.
Paoli, Letizia (2002), 'The paradoxes of organized crime', Crime, Law and Social Change, 37(1), 51-97.
Pearson, Geoffrey, and Hobbs, Dick (2001), Middle market drug distribution. London: Home Office.
Potter, Gary W. (1994), Criminal Organizations: Vice, Racketeering, and Politics in an American City. Prospect Heights, IL: Waveland Press.
Reuter, Peter (1983), Disorganized Crime: The Economics of the Visible Hand. Cambridge, MS: The MIT Press.
Ruggiero, Vincenzo and South, Nigel (1995), Eurodrugs: Drug use, markets and trafficking in Europe. London: UCL Press. Seligman, Adam B. (1997), The Problem of Trust. Princeton: Princeton University Press.
Simmel, Georg (1992[1908]), Soziologie: Untersuchungen über die Formen der Vergesellschaftung. Frankfurt/M.: Suhrkamp Taschenbuch Verlag.
Sutherland, Edwin H. (1937), The Professional Thief. Chicago: The University of Chicago Press.
Thrasher, Frederic M. (1963[1927]), The Gang: A Study of 1,313 Gangs in Chicago, Chicago: The University of Chicago Press.
Tversky, Amos, and Kahneman, Daniel (1982), 'Judgment under uncertainty: Heuristics and biases', in D. Kahneman, P. Slovic, and A. Tversky, eds., Judgment under uncertainty: Heuristics and biases, 3-20. Cambridge: Cambridge University Press.
Uslaner, Eric M. (2002), The Moral Foundations of Trust, paper prepared for the Symposium "Trust in the Knowledge Society", University of Jyväskylä, Jyväskylä, Finland, 20 September 2002. Online available at: http://www.bsos.umd.edu/gvpt/uslaner/working.htm
Uslaner, Eric M., and Richard S. Conley (2003), Civic Engagement and Particularized Trust: The Ties That Bind People to Their Ethnic Communities, American Politics Research, 31(4), 331-360. Online available at: http://www.bsos.umd.edu/gvpt/uslaner/working.htm
Varese, Federico (2001), The Russian Mafia: Private Protection in a Market Economy, Oxford: Oxford University Press.
Waring, Elin J. (1993), Co-offending in White Collar Crime: A Network Approach. Ann Arbor, MI: Dissertation Yale University.
Weerman, Frank M. (2003), 'Co-Offending as Social Exchange: Explaining Characteristics of Co-Offending', The British Journal of Criminology, 43(2), 398-416.
Zaitch, Damián (2002), Trafficking Cocaine: Colombian Drug Entrepreneurs in the Netherlands. The Hague: Kluwer Law International.