Fences Outside Fences: The Uses of Heroic Marginality in Ethical Behavior

The Next Phase of Business Ethics: Celebrating 20 Years of REIO

ISBN: 978-1-83867-005-4, eISBN: 978-1-83867-161-7

ISSN: 1529-2096

Publication date: 4 September 2019


Despite the existence of a variety of approaches to the understanding of behavioral and managerial ethics in organizations and business relationships generally, knowledge of organizing systems for fidelity remains in its infancy. We use halakha, or Jewish law, as a model, together with the literature in sociology, economic anthropology, and economics on what it termed “middleman minorities,” and on what we have termed the Landa Problem, the problem of identifying a trustworthy economic exchange partner, to explore this issue.



Mitnick, B.M. and Lewison, M. (2019), "Fences Outside Fences: The Uses of Heroic Marginality in Ethical Behavior", Schwartz, M., Harris, H. and Comer, D.R. (Ed.) The Next Phase of Business Ethics: Celebrating 20 Years of REIO (Research in Ethical Issues in Organizations, Vol. 21), Emerald Publishing Limited, Bingley, pp. 103-156. https://doi.org/10.1108/S1529-209620190000021011



Emerald Publishing Limited

Copyright © 2019 Emerald Publishing Limited

The sage, Shmuel, was helped on a ferry by a stranger. Shmuel inquired, “What is your business here?” The man replied, “I have a lawsuit to present.” Shmuel responded, “I am ineligible to be your judge.”

The sage, Ameimar, was sitting as a judge. A feather settled on his head. A man in the court came over and removed the feather. The sage asked, “What is your business here?” The man responded, “I have a lawsuit to present.” Ameimar immediately declared, “I am ineligible to serve as your judge.” (Talmudic stories from the Babylonian Talmud, Ketubot 105b.)1

In this article, we develop a theory of what we term heroic marginality in the production of ethical behavior. Heroic marginality occurs when agents appear to take extreme precautionary behavior – create “fences outside fences” – to avoid behaviors of noncompliance, such as violations of obligations under agency.2 In other words, they go “over and beyond” the nominal requirements for ethical performance in order to absolutely ensure that performance on behalf of the principal. The behavior is “heroic” in the sense of taking extreme or difficult steps or overcoming obstacles in order to realize the precautionary behavior. It is marginal in that it extends the compliance activities at the margin beyond where they might have stopped. Heroic marginality involves more than taking exemplary care, that is, meeting the highest standards while complying. It involves going beyond nominal compliance to achieve a condition of failsafe precaution. To use the metaphor, a fence is placed around the fence. In what follows, we shall propose some elements of a theory of heroic marginality in this precautionary context.

Although the literature on ethics in professional and organizational settings, and in business relationships generally, has produced many promising theoretical works that suggest approaches to assessing and maintaining ethicality in behavior (e.g., among many, Ashforth & Anand, 2003; Calton & Lad, 1995; Donaldson & Dunfee, 1994; Frederick, 1995; Gatewood & Carroll, 1991; Greve, Palmer, & Pozner, 2010; Jones, 1991, 1995; Palmer, 2012; Tenbrunsel & Smith-Crowe, 2008; Treviño & Weaver, 2003), the problems of engineering systems to produce specified ethical outcomes are less well understood. One such class of dilemmas may be termed the general problem of organizing systems for fidelity. The issues here include the design of mechanisms to inspire fiduciary-like behavior in individual agents as well as the design of organizations and higher-level systems to produce desired organizational as well as societal outcomes involving ethical behavior. This article attempts an innovative approach to understanding the production of fidelity by examining how commitment to deeply-held, genuine beliefs can lead certain societal groups to exhibit particularly high levels of such behaviors. Such groups may then serve as models for the construction of settings more likely to be successful in producing behaviors of fidelity.

We begin by identifying a key research question: What difference does it make when social actors are driven by genuinely- and deeply-held beliefs regarding 
fundamental and plenary value systems? We consider the roles of terminal values and what we call deep codes.

The article then proceeds in two parts: We will first establish the criticality of genuine (as against instrumental) belief in the law as a producer of integrity in commercial and other representational, usually contractual behaviors. This is done via an examination of two related literatures: First, we consider that on “middleman minorities,”3 including claims that certain mechanisms surrounding the nature of the actors and their settings explain the production of compliant behaviors. According to these scholars, including Bonacich (1973) and especially Landa (1981), the cultural homogeneity and kinship ties in such middleman groups are said to produce levels of trust, reliability, and fidelity such that these groups find widespread employment in their societies in the roles of commercial intermediaries.

Second, we examine Avner Greif’s explanation for the relatively low level of cheating reported among Jewish merchant traders of the eleventh century, the Maghribi traders of the Mediterranean. Greif (1989) argues that compliant behavior could be obtained as the outcome of rational defense of reputation within a “coalition” of recognized agents for which violation of agency performance had a high likelihood of negative sanctioning.

We question whether these analyses provide adequate explanation. We develop a substitute, if potentially complementary explanation for this behavior based in Jewish law, halakha.4 We allow that explanations from all three sources may operate in tandem, but that it is impossible to discount and/or largely ignore, as is true of other works, the effects of mandatory instructions associated with genuinely-held belief.

Second, we develop an explicit, propositional theory of heroic marginality, using the case of Jewish law, halakha, as a model. Because of its wealth of relevant cases and analytics, and its status as a robust and reasonably comprehensive legal system that features such deep commitment, halakha is a particularly good source for suggestive and supportive examples and analysis. The theory developed is not, however, limited to the case of Jewish law.

The article concludes with a discussion of the limitations as well as some practical contexts and applications of the theory we have developed. Not only do organizations and other societal relations routinely construct circumstances promoting heroic marginality, but societal institutions may be deliberately constructed to take advantage of the benefits such marginality may provide.

The Importance of Terminal Values and 
Deep Codes

Terminal Values

Consider first a question that becomes central to this study: What difference does it make when social actors are driven by genuinely- and deeply-held beliefs 
regarding fundamental and plenary value systems? Though it might seem self-evident that actors who genuinely believe in legal mandates instructing moral behavior in commercial relations will act this way (and believe that others in the same position will behave similarly), in fact explanations for the actual occurrence of such behaviors have tended to rely on other factors.

Modern social science has tended to avoid asking about deeply normative drivers of human behavior. The years after World War II saw a reaction against social sciences that were perceived as overly institutional and descriptive in character, and that provided insufficient attention to the actual behaviors of individuals. Disciplines stopped asking what drives people to act and through what institutions/institutional mechanisms. Instead, they focused on elements of action, such as the characteristics of decisions and the human limitations of making decisions; an example is Simon’s (1957) “bounded rationality.” In behavioral psychology, 
B. F. Skinner rejected attempts to measure desires, and, instead, tracked the effects of presentation of reinforcements as they impact the consequences of previous conditioning; humans were black boxes whose behaviors could be predicted by studying patterns of reaction to patterns of stimuli. Scholars in economics argued that what drove people may not be measurable and, at any rate, that human drives could not be directly compared. Hence the emphasis shifted to the choices that people make as providing inference to what they wanted – while never actually directly measuring what they want. Economists spoke of “utility” and its maximization rather than of actual desires. The implicit revolt against description climaxed in Milton Friedman’s (1953) influential claim that economics as a science need not describe (i.e., have realistic assumptions) as long as it predicts well.

Moreover, we accepted social/group affiliation, association, and/or identity as substitutes or proxies for measuring and assessing actual terminal motivation. Thus, as we shall see, we could develop explanations of economic behavior based on what are essentially signaling models of ethnic association and recognized identity – without ever asking what exactly do these subjects really value.

In essence, although enormous literatures continued to examine such things as perception/cognition and decisions, including developing the realistic descriptions of prospect theory regarding how human errors in perception and biases in assessment influence choice (Kahneman, 2011), we stopped asking what terminally drives people to do what they do. It is only recently that scholars such as Steven Reiss (2000, 2004, 2015; Reiss & Havercamp, 1998) and Shalom Schwartz (1994, 2012; Schwartz & Boehnke, 2004) and their colleagues began to revive an age-old, fundamental question in the study of human behavior: Can we systematically identify the terminal values/desires of humans and measure them? In essence, can we identify what people really want and value?

Reiss’s work is particularly important because it empirically validated a set of terminal values; Schwartz’s work has had a similar aim, but his articles present analyses that introduce instrumental relationships among the values, leading to questions about whether his typology really identifies terminal values. The question of terminal value is absolutely central here if we seek to truly understand what drives people. Reiss (2015) even related his model to religious belief, exploring how the pursuit of terminal values, which differ among people, affect their relationship with religion. Thus, Reiss’s work can help us understand the bases of belief.

We need to understand people not only as social actors, but as believers. Because individuals find satisfaction of their varied terminal values via religious prescriptions such as those in halakha, genuinely-held belief in religion becomes a fundamental driver of behaviors. As we shall argue, the Maghribi traders of the medieval Mediterranean would refrain from cheating their principals and agents because they practiced a religion that satisfied their deepest, most fundamental values and in which doing wrong to another was forbidden.

Deep Codes

Our focus in this article is on the effects of deeply-held, genuinely-believed, terminally-valued, and widely-shared belief systems. We thus argue that there are five critical characters of such a deep code: First, it must be perceived as underlying, rationalizing, primarily motivating, and/or referentially driving important behaviors; it is the deep driver, so that behavior can be rationalized or explained with reference to it. Second, it is subject to genuine, whole-heartfelt, committed, all-encompassing belief; it has no competitors in the basic motivational calculus of actors. Third, its value support is anchored at the level of terminal values – it is inexorably linked to the particular, identified set of values that the actor prefers above all others. Fourth, it is widely-shared – it is practiced by a community, not by isolated individuals. And, in sharing, it is reinforced, continually re-examined, and social actors’ beliefs in it are continuously re-validated. Finally, it is systematic. It is a code – it has a set of rules or patterned instructions that can be communicated to others and understood similarly by others.

Although it is perhaps easiest to understand a deep code by reference to religion, religion is not required to have such a code. Covenantal, cultural, club-like, ethnically-homogeneous, military, professional, and other communities are examples of settings that have the potential to achieve or approximate a deep code, and hence may exhibit the high compliance features we claim for it. Given the five characters above, it is not the same as culture, though some perspectives on culture center on values (but terminal values are not definitionally central to culture; Giorgi, Lockwood, & Glynn, 2015); and it is not the same as a belief system, which also has a very large literature. The pervasive control of human behavior potential when such codes are operative has an obvious dark side as well (see Goffman, 1961 on “total institutions” for a related but different perspective). A task for future research will be to expand our understanding of the contingencies and contexts that support deep codes and the means by which they can shape desired, positive behaviors. The present article uses the case of Jewish law to focus on the role of religion as a model of a deep code.

The Problem of Explaining Fidelity in Agents

Conflict of Interest and Agency in Commerce

The behaviors of commercial representatives, whether studied as members of culturally homogeneous “middleman minorities” or as Jewish merchant traders, represent instances of agency, in which the agent is acting on behalf of a principal in a mutually beneficial relationship that is designed to achieve certain performance and outcome goals for the principal. Such relationships are understood to be fraught with potential conflicts, including conflict of interest between agent and principal. There exist a variety of issues, including questions of clear and adequate communication between agent and principal regarding tasks, effort levels, competence issues, the costs and adequacy of monitoring of the agent’s performance, and so on. It is clearly in the interest of both parties – absent successful cheating – to reduce the costs of such agency problems, if the relationship is to indeed be mutually beneficial.

The regulation of conflict of interest in agents, or prospective agents, is a commonplace feature of both formal and informal legal codes (on conflict of interest in public sector contexts, see, e.g., Frederickson, 1993; Roberts, 1988; Vaughn, 1979; for one interesting case of conflicts of interest among agents, see, e.g., Kurland, 1995b, 1996b; there is a voluminous literature on such conflicts in a variety of societal, sectoral, organizational, and legal settings). For example, the strictures placed on a fiduciary are emphatic: requirements of service to the principal without regard to agent self-interest and the duties of best use of competence, avoidance of any appearance of conflicting interest, and full disclosure to the principal. These requirements are strongly prescribed in the common law of agency, trusts, and contract (the major codification appears in American Law Institute, Restatement of Agency, Third (2006); on the legal concept, see, e.g., Clark, 1985; Cooter & Freedman, 1991; Finn, 1977; Frankel, 2011; on the fiduciary concept as a social norm, see Leib & Galoob, 2016; Mitnick, 1973, 1974, 1975a, 1975b; Stinchcombe, 1986; cf. Landa, 2005). Chief Judge Benjamin Cardozo’s florid description in Meinhard v. Salmon, 164 NY (1928) 548 (https://www.nycourts.gov/reporter/archives/meinhard_salmon.htm) of the standard to 
which a fiduciary must be held is widely cited (for a classical analysis of the fiduciary, see Scott, 1949):

Many forms of conduct permissible in a workaday world for those acting at arm’s length are forbidden to those bound by fiduciary ties. A trustee is held to something stricter than the morals of the market place. Not honesty alone, but the punctilio of an honor the most sensitive is then the standard of behavior.

Although the secular Anglo-American common law is quite clear on the existence of fiduciary obligations, and although there exists a vast case law as well as legal analytical literature on such obligations, including the Restatement of Agency, Third (2006), many gray areas remain. Indeed, it is typical to observe the presentation of vague hortatory language like that of Cardozo followed by a specifically contextual legal principle with an elaborate list of specific cases or specific contingencies illuminated by cases; this is not an area with a developed set of normative generalizations that may be easily grounded in or related to general types of social behaviors and linked directly to social theory.

The fiduciary principle applicable to many situations in the law is, however, mirrored in the existence of a fundamental social norm, the fiduciary norm. The fiduciary norm is triggered in conditions of high social dependency, whether or not the relationship is formally or informally contractual, that is, whether or not there is a mutual agreement between agent and principal on agent performance. The norm places similar burdens on the agent: requirements of service to the principal without regard to agent’s self-interest, action with highest competence, avoidance of the appearance of conflict of interest, and full disclosure to the principal. The fiduciary norm was first proposed as a common social norm by Mitnick (1973, 1974, 1975a, 1975b), and was independently proposed as well by Stinchcombe (1986; Stinchcombe’s work was actually done over a decade before the date of his book). Whether the context is formally structured, or the norm is informally triggered as the result of perceptions of high dependency, the prescription of service to the principal, although strict, provides no clear boundaries to action.

Thus the issue of where obligation stops remains a central and unresolved concern, if typically dealt with at least in the Anglo-American legal literature by analysis of case law and sometimes the derivation or implication of general legal principles, rather than via the generation of clear and unambiguous behavioral instructions (usually the realm of legislatures and regulatory agencies). The Code of Conduct for United States Judges states that a judge “should act at all times in a manner that promotes public confidence in the integrity and impartiality of the judiciary” (http://www.uscourts.gov/judges-judgeships/code-conduct-united-states-judges). It is hard to imagine, however, a magistrate in a secular court at any level recusing him or herself from a case because the defendant or plaintiff had brushed a feather off his or her head, or given him or her a helping hand up to the bench.

At the same time, it can hardly be argued that fiduciary obligations are universally observed and universally effective in all the settings of agency, trust, and contract in which they are prescribed. The business press seems liberally sprinkled with descriptions of cases of breach of fiduciary obligation in a wide variety of commercial settings, including securities dealings, corporate governance, trust administration, professional conduct, and so on. While we would hesitate to declare the failure of fiduciary obligation – we could undoubtedly compile a much longer list of contexts in which securities are traded without manipulation, directors and corporate officers perform flawlessly for the firm and its stakeholders, trusts operate successfully, physicians and attorneys serve only their clients’ best interests – it must remain a matter of serious concern that significant failures do persist.

Thus, fiduciary rules in the law are designed to provide assurance that agents will act with fidelity, and the fiduciary norm – a social norm outside the realm of formal law – can be triggered in social relationships of high dependency. But neither is adequate to explain why commercial agents in a world without an effective civil law of agency, such as the world of ethnic agents in societies without effective enforcement of the rule of law, or medieval merchant traders, still act with high fidelity. Although high dependency may exist, it is questionable whether even a widely-supported social norm would be sufficient to curtail cheating among unobservable commercial agents operating far from their 

Solution via Use of Trustworthy Culturally Homogeneous Agents: 
“Middleman Minorities”

In contrast, there have existed in a variety of cultural settings, in both Eastern and Western societies, ethnic or religious groups that have displayed extraordinary integrity and reliability in commercial and other transactions. In general, this trustworthy behavior has been accompanied by relative economic success. The groups displaying this behavior are a subset of a larger class of immigrant groups labeled “middleman minorities,” “immigrant entrepreneurs,” or “ethnic entrepreneurs.” Examples of immigrant entrepreneurial groups include Chinese, Japanese, Armenians, and Greeks outside their homelands; East Indians in Uganda; the Ibo in Nigeria; Jews; Jains and Parsees in India; and Quakers in England in the eighteenth century (Light & 
Bonacich, 1988, p. 17). A relatively small, if growing literature, largely in economic anthropology, law and economics, and sociology, has documented the commercial behavior of this larger class and sought to explain it (see, e.g., Blalock, 1967; Bonacich, 1973; Landa, 1981, 1994, 1996, 2016; Light & Bonacich, 1988; Waldinger, Aldrich, & Ward, 1990; on minorities, cf. Laitin, 1995; cf. Grief, 1989, 1993, 1994, 2006, on the Maghribi traders). For our purposes, the major goal in studying such groups is to try to understand how successful economic activity may be accompanied by and perhaps even promoted by trustworthy behaviors.

The roots of this focus on the study of trust and economic success may be traced to Banfield’s fascinating classic study of the reasons for the failure of commercial activity to thrive in southern Italy (Banfield, 1958; cf. Putnam, Leonardi, & Nanetti, 1993, which largely supports Banfield’s general thesis; see also Coleman, 1988 on “social capital”; cf. North, 1990 on the importance of institutions and noneconomic factors in economic success; see also, e.g., Bradach & Eccles, 1989; Greif, 1994; Kreps, 1990; Noreen, 1988; North & Weingast, 1989). The role of trust, however, is far more subtle than is commonly treated. Hardin’s (1992, 2002) concept of “encapsulated trust,” which involves taking the interest of the other party into account, and to which we return below, implicitly raises important questions about the necessary features of middleman agency. Indeed, there is now a voluminous literature on trust (e.g., Calton & Lad, 1995; Cook, 2001; Cook, Hardin, & Levi, 2005; Cook, Levi, & Hardin, 2009; Hardin, 1992, 1996, 2002, 2004, 2006; Hosmer, 1995; Husted, 1994; Jones, 1995; Kramer, 2006; Kramer & Cook, 2004; Kramer & Lewicki, 2010; Kramer & Tyler, 1996; Ring & Van de Ven, 1994; Shane, 1994; Mayer, Davis, and Schoorman, 1995; see also Ouchi, 1980, on clans and Alvesson & Lindkvist, 1993 on control via clan).

The mirror image of the ethos of a “backward” society, as Banfield (1958) termed it, is that which appears to drive successful commerce. Banfield compared regions in Italy characterized by profound differences in social and, relatedly, economic behavior, including degree of willingness to trust in contract; he was not addressing the case of an ethnic minority embedded in a larger society. He ascribed systemic commercial failure to a shared “ethos” supporting lack of trust in agreements made with others. Trust could extend no farther than the family; hence “amoral familism” dominated. It would be quite ironic were we to find that family and kinship ties, by promoting reliability in behavior, could be the general engine of commercial success for middleman groups.

The literature on middleman minorities extends over sociology, economics, anthropology, and political science, with occasional stops in social psychology. In short, here we have a phenomenon studied virtually across the social sciences. Yet it is not unusual for the scholarship in one field to ignore that in another. There is clear disagreement on the definition of “middleman minority” or “ethnically homogeneous middleman group” (“EHMG”). Blalock (1967) appears to have originated the term “middleman minority;” his usage was accepted by Bonacich (1973), a major theorist in this context, though Bonacich disagreed regarding grounds for inclusion of a group in this category. Indeed, rather than a definition we see a description of the activities they engage in. Middleman minorities are seen as immigrant ethnic minority groups that engage in “intermediate” rather than “low-status” occupations, usually in trade or commerce,

“but also other ‘middleman’ lines such as agent, labor contractor, rent collector, money lender, and broker. They play the role of middleman between producer and consumer, employer and employee, owner and renter, elite and masses.” (Bonacich, 1973, p. 583; on the distinction of “immigrant entrepreneurs” from “ethnic entrepreneurs,” see Light & Bonacich, 1988, p. 18)

Landa (1981) describes the “EHMG” in terms of a group engaged in trade and so also uses the term “middleman;” the subjects of her analysis have the characteristics of her label. It is not clear, however, exactly how we are to understand the boundaries of the term “middleman,” except that trading with other traders – 
representing one party to another party – is clearly part of the intended role.

Waldinger, McEvoy, and Aldrich (Waldinger et al., 1990, pp. 119–120) identify “middleman minorities” as “ethnic groups that specialize in trading with the majority population.” Thus, like Landa’s EHMG, the group in this case is composed of traders, but with the added qualification that the trade is with the majority population. The context of the Waldinger et al. (1990) discussion is that of “ethnic entrepreneurs;” the traders tend to be small entrepreneurs in business for themselves. In this analysis it is less clear what “middleman” refers to; Korean shopkeepers in Los Angeles are merely small business owners/operators, not intermediaries among, or agents of, members of the majority population.

Greif (1989, 1993, 1994, 2006a) studied documents related to the activities of medieval Jewish traders in the western Mediterranean, the Maghribi traders. The documentary trove had been found in the Cairo Geniza (Geniza is capitalized when reference is made to the Cairo Geniza; Goitein, 1974, p. 5). Documents that either contained, or were suspected of potentially containing, the name of God, or were simply written in Hebrew and/or its predecessor languages (or even in Arabic using Hebrew letters) that were then considered the language in which the written Torah was given were put into a geniza rather than destroyed. Note the relevance to the “fences outside fences” argument in this article: The richness of the documentary record left in the Geniza regarding the commercial activities of the Maghribi traders may have reflected their concerns with being certain to include documents that conceivably, even with very low probability, might contain the name of God. We argue that this practice can be included among the evidence that the Maghribi traders were strictly observant Jews who followed the rules of halakha (see below) as it was then formulated.

Greif’s Maghribi traders were merchants and agents of merchants; the Maghribi merchants used agents who were also Maghribi Jews, and also partnered with other Jewish Maghribi merchants. At the time, northern Africa and southern Spain were under Islamic rule, so the Maghribi were indeed minorities. But they did not act as merchants for non-Jews, and so were not like the middlemen described in other studies of ethnic groups that acted as commercial intermediaries for wealthy members of the majority population. As in many of the other cases, however, the rate of cheating in commercial transactions among Maghribi traders was very low. The research question thereby raised is, why.

Such variations in treatment of middlemen minorities pose serious dilemmas in identifying whether the particular group and/or status of “middleman minority” is distinct and exhibits characteristic behaviors. In Table 1 we have sorted and summarized the definitional and situational descriptors offered in these works. It is not clear where definition stops and situation, context, or consequence begins. In brief, it would appear that middleman minorities may be: homogeneous ethnic actors, immigrant and minority, populating occupations of intermediate status, engaged in entrepreneurial trading or intermediary work, on behalf of majority population actors and/or providing services to or interacting with majority actors or agents of majority actors, and (probably as a consequence, but only in certain cases) producing activity qualities that include, with respect to majority actors engaged in similar activities, relative integrity, reliability, and care-taking. Given the behavioral discussions in the literature, it would seem that, at minimum, they are immigrant, minority, homogeneous, ethnic actors; the other elements are consequential activities or outcomes that may well depend on other conditioning factors. Thus, Native Americans in the United States do not qualify in this class, by definition. On the other hand, it is apparent that not all immigrant, minority, homogeneous ethnic actors produce the behaviors typically described (see Waldinger et al., 1990 for descriptions of significant variations in entrepreneurship among such groups).

Table 1.

Definitional and Situational Descriptors of a “Middleman Minority.”

  • Ethnic
  • Homogeneous
Relation to setting:
  • Immigrant
  • Minority
  • Intermediate status occupation
Activity type or role:
  • Trader or intermediary
  • Entrepreneur (in business for self)
Agency status/activity objects (for whom and/or toward whom):
  • For majority actors
  • With or toward majority actors or agents (e.g., traders) of majority actors
Activity qualities:
  • Behaviors of relatively high integrity, reliability, and care-taking

Moreover, in the key roles in which the tenth and eleventh century Maghribi traders discussed below act, they act as agents for other Maghribi traders, who are minority, not majority members of the community. Of course, the Maghribi traded in the larger majority communities of the Mediterranean.

One of the most thoughtful discussions of these issues is by Bonacich (1973; see also Light & Bonacich, 1988), who presents an intriguing and relatively coherent theoretical model. She notes that earlier scholars had ascribed the behaviors of such a group either as a response to discrimination and hostility directed by the host society against the group, or as an expedient or exploitive use of the group made possible by sharp status differentials in the host society. In the first case, the group is forced to be marginal and, with the help of community solidarity and bootstrap efforts, pulls its way up in the marginal areas of economic activity available to it. In the second case, the minority does not share the status “hangups” of the larger society and is thus free to deal among all groups. Without “familistic” ties to other groups in the society, the group’s members can be “‘objective’ in the marketplace.” And, finally, they can serve as a buffer against the hostility of the lower status groups, protecting elites by serving as their agents (Bonacich, 1973, pp. 583–584).

Bonacich argues that the first explanation does not explain why groups succeed rather than are hurt by the societal hostility; the second explanation cannot deal with the fact that these middleman groups are found both in earlier and in modern settings that do not feature such stark status gaps. Incorporating some of the previous work, she presents a theory (see her Fig. 1; Bonacich, 1973, p. 584) anchored in the argument that the groups in question are fundamentally sojourners in their host society. They are not just minorities, but immigrants that retain the orientation of a transient toward the host society, whether or not there are immediate prospects of return to the homeland. As transients, “the more general occupational characteristic of these groups is liquidity” (Bonacich, 1973, p. 585). As sojourners, the members of the group engage in economic activities that produce transferable resources, and they work hard to build up the liquid capital that would support return. They therefore concentrate in certain occupations that, given the opportunities available to them as well as the support available from the group’s community, provide this liquidity. They have no reason to build strong ties to the host society, yet they have compelling reasons to build strong internal group ties, such as those inherent in active ethnic associations, that provide assistance to group members, integrate new arrivals into the group, and support activities oriented to return.

Fig. 1. 
The Landa Problem – The Problem of Identifying a Trustworthy 
Economic Exchange Partner: Three Mechanisms for Solving Landa Problems
Source: Adapted from Mitnick (2018b), “The Landa Problem.”

Fig. 1.

The Landa Problem – The Problem of Identifying a Trustworthy 
Economic Exchange Partner: Three Mechanisms for Solving Landa Problems

Source: Adapted from Mitnick (2018b), “The Landa Problem.”

The group ties and the common difficulties faced by group members promote trust among group members; there is a “‘primordial tie’ of blood” that “provides a basis for trust,” backed by the group’s ethnically based community associations (Bonacich, 1973, p. 586). The “family, regional, dialect, sect, and ultimately ethnic ties” become the basis “for preferential economic treatment” as the group organizes to provide its members with special forms of economic support (e.g., easier credit, skill training, temporary living arrangements, and access to entry-level jobs).

Bonacich argues that sojourning is a necessary, if not sufficient characteristic of these groups; many sojourning groups do not enter entrepreneurial small business. Indeed, she suggests that “culture of origin is an important contributing factor” (Bonacich, 1973, p. 588). But she does not isolate what cultural features may also be necessary to produce the middleman behaviors. Furthermore, apart from noting factors that might produce in-group trust, she does not explain the production of trustworthy out-group behaviors. Why do certain middleman minorities display exceptional integrity and reliability in their economic transactions with the majority group? That is, why do these minorities display trustworthy agency behavior in commercial interactions with parties outside the group?

Bonacich’s claim of the importance of sojourning has been challenged. Why would sojourners risk their hard-won capital (built up in employee roles) in highly risky entrepreneurship? Sojourners may have to divert their liquid capital to send home to support relatives, and, if relatives are still back home they are not available as cheap labor to help run any small business activity in the host country. Furthermore, Aldrich could not discern effects of sojourning in his empirical studies of these groups (see Waldinger, Aldrich, and Ward, 1990, pp. 43–44; Aldrich, Cater, Jones, and McEvoy, 1983, p. 16), though sojourning is retained as a factor in the Waldinger, Aldrich, and Ward volume (1990, p. 132, in a chapter with a different mix of coauthors).

In a fascinating, broad-based examination of “ethnic entrepreneurs,” Waldinger, Aldrich, and Ward (1990) lay out a five factor, interactive model of ethnic entrepreneurship (see their Fig. 1.1, p. 22). Opportunity structures, including market conditions (such as opportunities to market ethnic products and degree of openness or access to nonethnic markets (e.g., as entrants in markets abandoned by the majority)) and conditions of access to business ownership (e.g., that take advantage of business vacancies or abandonments and of government support policies) interact with such group characteristics as a variety of predisposing factors (blocks to mobility in the labor market, migration to the host society by more able and risk tolerant individuals, and higher “aspiration levels”) and resource mobilization factors (such as close and supportive in-group ethnic ties, facilitative ethnic social networks, and government policies). This interaction affects choice of ethnic strategies in attempting successful entrepreneurship. (See Waldinger 
et al., 1990, pp. 21–47 for a discussion of the model.)

As it is for Bonacich, trust among ethnic group members is important in the model’s discussion, as when the known characteristics of group members (especially kin) make it easier to hire relatively unskilled labor and risk training them, or when ethnic group acquaintances in business networks provide reliable business or competitive information. Indeed, “the cost of transactions between firms is likely to be lower when they have a basis for trusting each another” (Waldinger et al., 1990, p. 39; see also pp. 133–137). Later in the volume, the “predisposing factors” and “resource mobilization” factors are collectively termed “ethnic resources”; and “cultural endowment,” a “sojourning orientation,” and other factors are added to the predisposing factor list (p. 132; cf. Light, 1984, pp. 201–204; Light & Bonacich, 1988, pp. 18–19, on “ethnic resources” and “class resources”). Differences among ethnic groups are held to be relevant in explaining business success; the factors include pre-migration group characteristics (e.g., labor skills), “circumstances of migration” (e.g., commitment to permanent settlement), and a variety of post-migration factors, such as facility in the local language. The choice of “ethnic strategies” provides alternative pathways to entrepreneurial activity, and, combined with the earlier factors, helps explain the outcomes of such effort (Waldinger et al., 1990, pp. 40–47).

Although the Waldinger et al. (1990) volume provides much grist for analysis of what makes for entrepreneurial success among ethnic groups, it does not tell us directly what might produce unusually trustworthy and reliable behaviors among the entrepreneurs. As with Bonacich’s sojourning minorities, they may use trust with one another to profitable end, but how and why would they evidence such behavior in their external interactions? Again, it is of considerable interest to learn this in order to be able to understand and perhaps engineer economic institutions with such characteristics.

The Landa Problem

Interestingly, the work by Waldinger, Aldrich, and Ward and their associates reported in the 1990 volume makes no reference whatsoever to an important stream of work in economics on middleman groups. As we noted, Janet Landa argues that ethnically homogeneous middleman groups (EHMG) are a vehicle for promoting contract compliance in environments where the legal system that could enforce such compliance lacks development or does not exist (1981; see also Landa, 1994 and Landa, 2016 which compile her work; Carr & Landa, 1983; Cooter & Landa, 1984; Landa, 1996; Posner, 1980; Landa has a number of other works in this general area as well, including work on social classification, e.g., Landa, 2005; she is also a pioneer in the related area of bioeconomics).

In Landa’s approach, a variety of social and cultural factors can serve as indicators of the likelihood that there are low risks of agent cheating. As with Banfield (1958), family — kinship relations — are taken as the most compelling assurance of lack of relationship exploitation. From there, Landa in her original 1981 paper traced circles of social distance, with the most distant being the least likely to be acceptable as agents. Thus identifiers of closeness of in-group membership serve, as noted, as low cost screening devices. The approach assumes that the central actor can indeed recognize (identify) actors that sort into the classes exhibiting different levels of reliability, and that the actor can distinguish between reliable behaviors and those that are not. This is a real issue when the agent is hired not just because of the need to act at a distance but because the principal lacks the expertise to do the task himself or herself and cannot evaluate whether the agent is actually performing optimally, that is, when adverse selection may exist (on such agency problems, see Arrow, 1985; Mitnick, 1996). Landa more or less assumes that these concerns are managed, that is, that the actor knows reliable behavior when he or she sees it.

In general, the middleman group reduces transaction costs and assures compliance in a number of ways. Middleman traders will prefer to exchange only with other traders known to be trustworthy (on trustworthiness, see Hardin, 1996, 2002). Thus, the size of the trading group is bounded by the fact that contract enforcement costs rise as the circle of traders widens and potentially more heterogeneous and less reliable trading partners are included (Cooter & Landa, 1984). The ability to identify reliable trading partners that is assisted by their membership in a recognizable kinship or status or social or ethnic group may be reinforced by support for and employment of an often, though not always unwritten, code of normative instruction (including codes of ethics or conduct and/or recognized sets of social norms) regarding trading. Entry into the favored trading group is difficult for outsiders who lack the necessary identity as well as the ability to know and consistently follow the recognized codes of conduct or social norms. Because factors such as kinship or ethnic membership are not easily replicable, the use of such factors protects the reliability of the identification method (group membership) as well as the economic returns of the traders who do belong to the group.

It is efficient to use group affiliation as a “low-cost screening device” (Landa, 1981, p. 359; on the use of symbols as signaling artifacts that perform as efficient screening devices, see, e.g., Carr and Landa, 1983; Pondy, Frost, Morgan, & Dandridge, 1983); it would be far more costly to directly establish that behavior is likely to be trustworthy via building and defending an idiosyncratic reputation linked only to the single trader (and hence more vulnerable than the reputation of an entire middleman minority group of traders).The group can thus perform effectively without excessive reliance on potentially costly contractual mechanisms of enforcement and compliance. This allows the members to achieve contract performances in an especially reliable and efficient fashion, leading other societal groups to seek out the “middleman group” to perform commercial transactions on their behalf. Thus, although reputation management is a feature of the Landa model, it is done via screens or proxies (cf. the reputation mechanism in Greif’s model).5

Note that the adoption of the code of ethics is entirely instrumental; its particular content is irrelevant as long as it supports contract compliance. Because the trading group is built around the screening device, kinship or other particular ethnic status is paramount, and is the basis of monitoring. If threats arise to the performance of the trading group, they would be threats to the kinship or ethnic status community, not to its code of ethics, which is merely an instrumental tool. It is conceivable that codes of ethics in such circumstances, since they are not linked to the essential identity of the group, would be fungible and expediently modified. The Landa middlemen groups support ethical systems that are essentially procedure-based. As a result, the Landa theory would be equally explanatory of the behaviors of entrepreneurial Mafia groups or “families” (Landa, 1981 offers this example, p. 361).

Note also that the explanation for middleman success given above relies essentially on some basic homogeneity in the group – the term “homogeneous middleman group” is sometimes used. Such groups are not seen as self-regulatory in the deliberate sense; the informal social norms, common social and cultural practices (dress, speech, and behaviors), and shared cognitions of the social world, and perhaps the solidarity binds of kinship, produce the recognizable homogeneity viewed as essential to facilitate the group’s commercial role (on group culture, see Alvesson & Berg, 1992; Alvesson & Lindkvist, 1993; DiTomaso, 1987; Giorgi et al., 2015).

Landa’s work makes a fundamental contribution to the understanding of market existence and functioning; she introduces with more clarity than anyone had previously the problem of identifying a trustworthy economic exchange partner. Thus, we term that problem, the “Landa Problem.”

Greif’s Mechanism and the Landa Problem

Reliance on the civil legal system is minimal or nonexistent in the middleman minorities literature; the remarkable nature of Landa’s insight is how reliable functioning of agency occurs without the intervention of the legal system – which in the contexts studied often was unavailable anyway. Such a legal system did in fact exist in the world of the Maghribi traders; there were both Jewish and Muslim courts. But Greif (2012) argues based on available evidence from the Geniza documents that use of the legal system by the Maghribi was minimal; he uses that fact to support his argument (and to respond to critics) that it must have been the self-enforced reputational system that was the key mechanism, made possible by the establishment and recognition of the Maghribi trader in-group, which he terms a “coalition.” Note of course that the absence of use of one ­coercively backed dispute settlement system, the courts, does not necessarily mean that a second sanctioning system – threats from damage to one’s multilaterally-held reputation in the “coalition” of traders – must have been operative, especially if, as we will argue, another system was in place to encourage compliance.

Edwards and Ogilvie (2012) argue that Greif lacks adequate documentary support to make his claims, and that use of the legal system was greater than claimed; Greif (2012) disputes that, at length, accusing Edwards and Ogilvie of shifting their arguments and misunderstanding the existing records (cf. Greif’s response Greif, 2012).

But, let us assume with Greif that the courts were relatively little used by the Maghribi traders. How, then, did they achieve what is not doubted – a very high level of absence of cheating – a level much higher, in fact, than traders of other groups and other national settings, for example, the Genoese/Italian traders also discussed by Greif? Greif might argue that the Genoese traders did not have the cultural practice of within-group agency – recall that the Maghribi traders, both merchants and agents, used only other Maghribis as agents/merchants. But is that sufficient to see such a large difference in compliance? The key for Greif is the absence of the “coalition,” that is, the use of traders only within the recognizable group so that the reputational mechanism he posits can operate. Moreover, Greif does not use the standard arguments from the middleman minorities literature; he is not claiming as well that factors such as social distance, kinship, and recognition of common cultural aspects or of ethnic group membership as signals of likely reliable behaviors, are the active factors. His model, apart from identity recognition as Maghribi coalition members, is intended to be based on rational expectations of behavior to retain reputation and avoid negative consequences, such as being ostracized or becoming a target of cheating. That is, it is a closed system that is self-enforcing almost purely on self-interested grounds.

Although the costs of reputation maintenance can be considerable, reputation can be an effective policing mechanism. Bernstein (1992) describes how diamond merchants in New York relied on the actual posting of reputation bonds, in diamond exchanges, as well as reputational knowledge, to police cheating (see also Richman, 2004, 2006; on the breakdown of the reputation mechanism in the diamond industry, see Richman, 2017). But according to Greif, the Maghribi traders operated on the basis of that self-enforcing system of reciprocally honest behavior without reputation bonds. A trader found to have been cheating would be subject to cheating by those with whom he in turn dealt with; those who refrained from cheating behavior could expect that others would respect their reputations regarding the absence of cheating and also refrain from cheating them. Greif also does not explain this on the basis of any social norm, such as reciprocity, even though the behavior is reciprocal. As noted, he views the behavior as strictly rational and self-interested.

Clearly, a key factor is that the Maghribi were identifiable to other Maghribi as members of the same “coalition” of traders. That permitted traders to know with whom they could expect honest behavior, unless shared information on trader performance damaged one’s reputation. Left unclear and largely undiscussed, however, is how traders came to know about and assess reputation-related performance, given the far-flung nature of the trade and the absence of adequate communication devices. How indeed would anyone find out if a trader had cut corners on the other end of the transaction, when, according to Greif, it could take weeks to even months to reach destinations around the Mediterranean (although there appear to be some differences regarding transit speeds, Goitein (1967, i, 3, p. 69) refers to the “precariousness of communications” and mentions transit times of weeks)? And how would traders assess the quality of reports of such behavior? Envious traders might well reduce competition by lying about the performance of colleagues – a lie that would be difficult to overturn via the presentation of distant witnesses (cf. the difficulties of establishing a “community responsibility system” as described by Greif, 2006b).

An Alternative Explanation: Jewish Law

While a mechanism such as Greif’s may well have contributed to the reliability of Jewish agents and merchants in commerce, there appears to be an additional factor of critical importance that is not given prominence – the presence of a comprehensive and elaborated legal code in the form of a set of rulings, anchored in the Torah, that prescribed moral behavior (see Tamari, 1987, 1991). The authority of that law – “halakha” – is absolute, since it is based not in a design of humans, but is considered as an application of divine instructions. Thus, the direction of causality is potentially reversed in any explanation of commercial reliability or success – it is not the informal social and cultural underpinnings of the group that permit successful participation (via serving as a screening device) in formalized economic transactions; rather, it is a legal code that dictates the moral behavior so valued in commercial exchange. The social and cultural context is as well influenced, and operates to buttress the primary influence of the law, halakha.

That is, Jews known in the community as observant, and, hence, compliant with Jewish law, will be rewarded with respect, as well as expected to behave both in their business and personal lives, with consistency. Thus, simple observations of pious compliance with Jewish law, even in settings with no relevance to commerce, will be followed by expectations of compliance to the law’s moral dictates in other settings, including commerce. And such individuals will be provided with the social and status rewards bestowed on individuals who visibly value the law – 
because the community values the law. This is in part a reputational argument, too, but it has a completely different basis from that of Greif.

Moreover, in the Jewish case, the central argument is that the extraordinary moral dictates of their legal system are genuinely believed and supported by the actors, and produce unusually reliable behaviors that would be attractive to whoever had need of access to reliable agents in commerce. Why might attention to the prescriptions of Jewish law have been especially prominent during this period? Before we discuss the nature of Jewish law itself, it is important to note that Jewish law, and not the mostly absent secular legal settings of the regimes around the Mediterranean, was most often recognized as the governor of both personal and commercial relationships. Goitein (1967) is very clear on this:

It is hardly necessary to quote Geniza texts to prove that the same law was applied to the members of one religious community throughout the Mediterranean area, irrespective of the country in which they happened to live. (i, 3, p. 68)

At the root of all this was the concept that law was personal and not territorial. An individual was judged according to the law of his religious community, or even religious “school” or sect, rather than that of the territory in which he happened to be. Perhaps it would be even more exact to say that, with the exception of some local statutes, promulgated and abrogated from time to time, the states as such did not possess any law [emphasis added]. These were the preserves of the religions and their various subdivisions. (Goitein, 1967, i, 3, p. 66)

Goitein’s observations suggest two points regarding Greif’s approach: First, for the Jewish population described by Goitein as governed everywhere, and personally, by Jewish law, without access to civil enforcement agents, the absence of reports of legal cases in secular settings is hardly unexpected (on the absence of law, especially the absence of impartial law, see Greif, 2006a, p. 318). Disputes that turn on applications and interpretations of religious law would be referred to Talmudic authorities and/or rabbinical courts. In the twelfth century, that included Maimonides himself (see below), who wrote thousands of responsa, responses to questions posed regarding the applicability or interpretation of Jewish law. Moreover, as religious law, halakha is based in deeply- and genuinely-held personal beliefs that require compliance largely without ready external coercion. Second, while the reputation mechanism that Greif describes may well operate, his model may be ignoring the most important reason for the low level of merchant and agent cheating: Such cheating would be a violation of halakhic law, as well as of the specific terms of agency (cf. Richman, 2006, who notes the important role of Jewish law in the diamond industry). As we will argue, there can be multiple and reinforcing factors at work here.

Thus, whether our approach can claim to be a viable alternative to those of Landa and Greif depends on the status of Jewish law during the era, and whether genuine belief in the precepts of Jewish law was common and considered binding. To explore this, we will first provide a brief description of Jewish law, or halakha, and briefly describe some of the major codifications that occurred during that era as well as important codification that followed and provided the basis for Jewish law today. Then, finally, we will describe the relationship of the Maghribi traders to religion in the eleventh century: Was it likely that the Maghribi traders really did hold a deep and genuine belief in Jewish law – a belief sufficient to affect the way they did business?

Nature of Jewish Law: Halakha

While the ultimate authority for Jewish law rests in the Torah, Jewish law itself is spread over a variety of sources (for an accessible brief description of halakha, Jewish law, see http://www.jewfaq.org/halakhah.htm; for reviews of the historical development of Jewish law, see, e.g., Elon, 1994; Feldman, 1978; Hecht, Jackson, Passamaneck, Piatelli, & Rabello, 1996).

The Torah of Judaism is a dual one: The revelation at Mount Sinai consisted of one part that was given in written form (the five books of the “Pentateuch,” the beginning of the “Old Testament”), while the other part was meant to be transmitted orally and be memorized for successive generations (Neusner, 1986, p. vii). The Mishnah, the written form of the Oral Torah that became the basis for the Babylonian and Jerusalem Talmuds, is even written in a mnemonic form that enables memorization (Neusner, 1984). After the destruction of the Second Temple in Jerusalem in 70 ce and the expulsion of the Jews from Palestine by the Romans, Yehudah haNasi [130 ce –220 ce] wrote down the Oral Torah, ensuring its survival. He feared that exile, persecution, and restrictions on learning would inhibit oral transmission and lead to the Oral Torah being forgotten (Krupnik, 1981, 
p. 15; Shmuel haNagid, cited in Carmell, 1988, p. 68). The care taken in preserving and protecting the elements of the Torah reflects the transcendent importance of the prescriptions of Jewish law in the conduct of Jewish life.

These works make up the basis for all of Jewish law and its practice (Donin, 1972, p. 29). Furthermore, the central tenet of the Jewish legal system is that the law is of divine origin. One of the central theological principles of Judaism is that the Torah was given by God to Moses at Mount Sinai (see, for brief discussions and citations of further work, Donin, 1972, p. 21; Telushkin, 1991, p. 53). This is among the most fundamental beliefs in the Jewish religion.

A consequence of the original structure of the Torah is that legal rules that can be seen as part of the Torah (d’oraita) are given strict priority over rules deduced from commentary or other rabbinic sources (d’rabbanan); they are treated as absolutely binding (on the problems of distinguishing rules that are d’oraita from rules that are d’rabbanan, see Roth, 1986). Maimonides (see below) listed 613 mitzvot or commandments from the Torah in his work, the Mishneh Torah. An example relevant to merchant traders is, “When you sell property to your neighbor, you shall not wrong one another” (Lev. 25:14; see, e.g., among a great many online citations of the Hebrew text, https://
www.sefaria.org/Leviticus.25.14?lang=bi&aliyot=0; for a modern era sorting of the 613 mitzvot, see, e.g., http://www.jewfaq.org/613.htm). Thus, the observant Maghribi merchant traders would have perceived, personally – not merely as a civil law which they were obliged to obey, or as a matter of societally-supported ethics – a divine commandment to not do wrong in their commercial transactions. Not all of the 613 can be followed today, nor even in 1100 ce (e.g., references to sacrifices at the Temple, destroyed in 70 ce), but the bulk of the 613 mitzvot were very real and mandatory commandments to Jews in the eleventh century.

But despite the primary position of rules that are d’oraita, by far the majority of the thousands of pages of commentary represent rabbinic opinions and rulings that are d’rabbanan, that is, sourced back to rabbinic views. Although rulings that represent “fences-outside-fences” can be d’oraita, derived from statements in the Torah, the bulk of such precautionary rules are d’rabbanan. Rules that are d’rabbanan also include rabbinically recognized rules based in custom (minhag).

Jewish law dictates proper behavior in virtually every conceivable area of human existence (Donin, 1972, p. 29); there is constant (and often contentious) commentary on appropriate additions or interpretations to handle societal changes. The recognition of the need to make formal adaptations, modifications to ensure compliance, and/or exceptions is not recent, and dates back to the 
biblical era, many centuries before the Maghribi. A takkanah is a rabbinically-authorized rule to permit specific practices that might not be possible under a strict interpretation of the commandments that are d’oraita (divine law). It is variously a societally-driven adaptation or a practical change necessary to fully observe other commandments; it is not a change done to make compliance less binding or more lenient in general. A gezeirah is a prohibition of behavior that might threaten to violate divine law (sometimes the term takkanah is used to cover, generically, both cases).

Of special interest for our approach is the very presence of gezeirah in the religious practices of observant Jews, which would include the Maghribi. This type of legal safeguard typifies the “fences outside fences” emphasis in Jewish law. The opening of a famous and heavily-studied piece from the Mishnah (hence included in the Talmud), Ethics of the Fathers (Pirkei Avos; also sometimes Ethics of Our Fathers), includes the lesson, “… make a fence for the Torah” (Pirkei Avot 1:1 (https://www.sefaria.org/Pirkei_Avot.1.1?lang=bi&with=all&lang2=en)):

Moshe received the Torah from Sinai and transmitted it to Yehoshua, and Yehoshua to the Elders, and the Elders to the Prophets, and the Prophets transmitted it to the Men of the Great Assembly. They said three things: Be deliberate in judgment, raise up many disciples and make a fence for the Torah.

Cases that feature extra precautions – fences outside fences – are quite common in rabbinic prescriptions for behavior.6 Take, for example, the prohibition of writing on the Sabbath. The Torah prohibits work on the Sabbath, but “work” is defined very specifically as those particular activities that were required to build the Holy Temple in Jerusalem (see, e.g., Chait, 1992; Donin, 1972, p. 90; and many others). Writing is one of the activities that is prohibited on the Sabbath (Donin, 1972, p. 91). It is also forbidden, however, to even hold a writing implement. An outside observer may find this additional prohibition extreme, thinking that the simple prohibition against writing is a sufficient protection in and of itself. In fact, the rule against holding a pen or pencil is an additional rabbinic restriction placed upon the Sabbath observer. The additional restriction was added so that the actual forbidden activity (writing) was given less opportunity to occur, whether by accident or because the pencil holder simply forgot that it was the Sabbath. Notice that the original prohibition against work on the Sabbath, from which the writing restriction is derived, is d’oraita – it is from divine commandment. Hence, there is a strong basis for going out of one’s way to be certain of compliance.

Or take, for instance, the case of the prohibition on using the Chanukah lights for anything other than their nominal purpose, to remind one of the miracle said to have occurred at the time of the Second Temple (Eider, 1980, p. 15). Special care must be taken so as not to use them for another purpose. Thus one may not eat by the Chanukah lights, even sitting at a distance in the same room (Eider, 1980, p. 16). Some rabbis rule that it is permitted to use the Chanukah lights to fulfill a commandment, such as learning Torah, but most rabbis rule that even this use is prohibited. This requirement is d’rabbanan, but the rabbinic source is still sufficient to mandate the special care to avoid other uses.

Consider as well the case of shatnez. Deuteronomy 22:11 states the following: “You shall not wear cloth combining wool and linen” (Jewish Publication Society, 1985, p. 308).7 According to halakha, if one is walking along and finds that s/he is wearing a garment consisting of the forbidden mixture (known as kilayim, or, alternatively, as shatnez), that person must remove the garment immediately, even if they are in a public place (Lehmann, 1985; Roth, 1986, p. 27). Furthermore, according to the Chafetz Chayim (the popular name of Rabbi Yisrael Meir Hakohen; it is the title of one of his books and translates to “Choose Life”),

If a person wears kilayim even by chance, temporarily, and even if it is over ten garments, so that it does not benefit him at all, and even if it is [only] to get it past the customs inspection – he violates this prohibition. (Hakohen, 1990, p. 265)

This, according to the Chafetz Chayim, is d’oraita, a law directly attributable to the Torah. It is even forbidden to use kilayim wrapped around oneself as a blanket. In addition, the Chafetz Chayim reports, rabbinical law places still further restriction upon the use of shatnez. One may not even sit upon mats or spreads of the forbidden material. “Even if there are ten soft spreads, one upon the other, and the bottommost one of them is kilayim, it is forbidden to sit on the uppermost one” (Hakohen, 1990, p. 265). In the case of shatnez, Jewish law insists that the princess does not tolerate the pea.

The nature of Jewish law is to create an atmosphere in which even the appearance of wrongdoing is anathema: “Whenever the sages forbid anything for appearance’ [sic.] sake, it is forbidden even in one’s innermost chambers” (Babylonian Talmud, Shabbat 64b; cited in Bialik & Ravnitzky, 1992, p. 465). Rabbinic law (i.e., “d’rabbanan”) fills the path to sin with stumbling blocks by putting additional strictures upon behavior so as to ensure adherence to Biblical law (“d’oraita”) (see Roth, 1986 for a thorough discussion of the processes of rabbinic legislation). Sometimes, rabbinic requirements may seem to go to extreme lengths, but they are “… predicated upon a chain of authority that is ultimately derived from the Torah” (Appel, 1975, p. 5).

Halakha provides very extensive coverage of commercial relations, including business ethics, of course (see, e.g., Dorff, 1997; Green, 1997; Levine, 1980, 1987; Tamari, 1987, 1991, 1997; Wagschal, 1990; Zipperstein, 1983). Halakha has been applied to consider the obligations of owners in the case of employee layoffs, noting explicitly the question of whether to go beyond what is required (Van Buren, 1999). Jewish law prescribes not only how to pray and worship, but also how to mourn (see, e.g., Lamm, 1969), how to fast (see the Talmudic tractate Taanit), how to rest on the Sabbath (see, e.g., Chait, 1992), how to conduct sexual relations with one’s spouse (see, e.g., Boyarin, 1993), and even how to behave when using the lavatory (Ganzfried, 1987, pp. 11–17). No human activity is too minor for inclusion in Jewish jurisprudence.

Codifications of the Law During and After the Maghribi Era

Moses Maimonides (1135 ce–1204 ce), also known as the Rambam, author of the 14 books of the Mishneh Torah, one of the great codifications of Jewish law, both was born during the Maghribi era, which ended before 1200 ce, and was born in and grew up in the western Mediterranean region in which the Maghribi traded. The family of Maimonides fled eastward when he was a young adult, eventually settling in Cairo. His adult life was spent there, where the Cairo Geniza indicates the Maghribi traders also worked, even before Maimonides arrived in Cairo. Of relevance to our argument, the Cairo Geniza contained some documents written in the handwriting of Maimonides, implicitly linking him with them (Goitein, 1967, reproduces an image). Maimonides later became the head of the Jewish community, served as a revered physician, and was a legal scholar and philosopher. During his maturity, he was renowned throughout the Mediterranean as a scholar of Jewish law, and continually consulted. It is hard to see how could the observant Maghribi traders could have ignored the prescriptions of Jewish law extant in the eleventh and twelfth centuries, and, eventually, the presence and enormously influential writings of the great Maimonides himself.

One cannot over-emphasize the drive to develop useful and systematic tools to interpret and apply the commandments of divine law. Perhaps the most significant compilation of Jewish law is the Shulchan Arukh, a treatise prepared by Joseph Caro (1488 ce–1575 ce). Alarmed by a decline in scholarship and a lack of uniform observance among the dispersed Jewry of his time (Twersky, 1967), Caro undertook a comprehensive codification of Jewish law. Unlike Moses Maimonides, whose own famous codification (Mishneh Torah [1180 ce]) was limited to final legal rulings, and who deliberately omitted references to his sources, Caro clearly referenced and discussed the Talmudic sources from which the rulings in his code were derived. He also included alternative legal views and differences in opinion, again fully citing their Talmudic sources. The encyclopedic work was meant to be a study guide for all of practical Jewish law, giving full review of all arguments and then offering a final normative conclusion for each ruling (Twersky, 1967). The result of Caro’s monumental efforts was the Beit Yosef (1542), or House of Joseph.

Later, Caro wrote his own abridged version of Beit Yosef, entitled Shulchan Arukh (1564 ce), or The Prepared Table. Combined with a set of glosses written by Rabbi Moses Isserles (the “Rama;”1530 ce–1572 ce), this abbreviation of the Beit Yosef became “the standard legal code of Judaism” (Telushkin, 1991, p. 203). Because Caro’s work omitted much of the practice of Ashkenazic Jewry (those Jews living throughout Europe) and concentrated on the practice of Sephardic Jewry (those Jews living in North Africa and the Middle East), the glosses and dissents of Isserles (an Ashkenazic rabbi) were added in italics in published versions of Shulchan Arukh. The Shulchan Arukh is so widely-read and consulted within Jewish legal studies that, when one refers to ha-mechabber, or “the author,” one is known to be referring to Rabbi Joseph Caro (Werblowsky, 1977, p. 2). Shulchan Arukh is generally the first text that is consulted when rabbis are asked to judge on a piece of halakha (Telushkin, 1991, p. 203).

The continued development of the law to the present day has meant the creation of literally thousands of pages of commentary, as well as further efforts to provide both systematization and easy access to both d’oraita and d’rabbanan rulings.

Mechanisms for Resolving the Landa Problem

Any effort to understand how the Maghribi – or any observant Jews in medieval times – would behave must take note of how the immense effort involved in the historical evolution of Jewish law reflects its perceived importance in Jewish life. Indeed, the argument may be made that a Maghribi trader would not be permitted by his religion to even consider choosing to protect his reputation – Greif’s central motive – if the actions required to do so conflicted with the commandments of Jewish law.

The Mishneh Torah of Maimonides used sources from centuries of scholarship and legal rulings, that is, its legal and religious sources included sources that predated the Maghribi period. As just noted, during the latter part of the Maghribi era (eleventh to twelfth centuries; most of the Geniza documents studied by Goitein, Gil, and other scholars are from the mid-tenth through eleventh centuries), Maimonides’s fame was such that the traders would have certainly been aware of him. The Maghribi were observant Jews – the fact that they felt compelled to make use of a geniza, with their documents ending up in the Geniza in Cairo, is circumstantial evidence of that – and so they were very likely compliant with halakha, Jewish law. Moreover, as traders they would often require legal guides, and we know, per Goitein’s observations quoted above, that the same law – 
Jewish law – would govern them wherever they traded in the Mediterranean. As we noted, halakha is not like Western codes of law – it extends into every area of life, including business life (e.g., Dorff, 1997; Green, 1997; Tamari, 1997). The Maghribi would not have segmented the law that guided their business lives from that which pervasively governed their personal lives.

Moreover, one cannot separate the alternatives into situations of voluntary compliance, versus resolution in the courts, as the debate between Greif and his critics seems to do (on alternative mechanisms for “private ordering,” including reputation, see, Richman, 2004). Commercial activity did not proceed without dispute, and apart from the specific issues reflecting the eleventh century context, the kinds of disagreements often seem modern in character – failure to adhere to agreements, failure to fully compensate, losses from failure to take care, failure to execute agreements, cheating and lying about compliance to contracts, and so on (see Gil, 2003, pp. 305–314; see also Goitein, 1967). Both Jewish and Muslim courts existed and, as is evident from the Geniza documents described by Goitein (1967) and Gil (2003), the Jewish courts were in fact used both in dispute settlement and in the creation and certification of various legal relationships. A system of civil-like authority, if based in Jewish law, including appointed judges, was administered by the “head” of the Jewish community in Egypt, who came to be called the “nagid” (Cohen, 1980; see also Goitein, 1967). Both Moses Maimonides and his son Abraham served as head of the community. Gil (2003, p. 310) writes, based on the Geniza documents,

It may easily be argued that the normalization of trade relations was enforced by the Jewish courts, and that they were the entities which, according to Jewish religious law, attempted to resolve the conflicts and expose acts of unfairness.

There is also reference to appeals being made to the Muslim authorities to send soldiers to enforce agreements (e.g., Gil, 2003, p. 309) and to use of Muslim courts (Greif, 2006a, p. 63, note 13).

The issue is really whether the reputational enforcement system posited by Greif substantially reduced this legal activity – a claim that cannot be verified based on the evidence. The sample of letters from the Geniza is important, but only a sample, and scholars can only offer informed guesses as to what ended up in the Geniza. Indeed, we might speculate that what got into the Geniza may represent the activities of the most observant in the population – those who would take the trouble to deposit their documents. Such a population might conceivably be more concerned about compliance with Jewish law, as well as more concerned about using available rule systems to settle disputes – in essence, a more law-abiding part of the population. And it might thus be a component of the population that is more likely to be concerned about their reputations both as observant Jews and as honest businesspeople, and less likely to cheat. So we need to be concerned whether we are building models that are not really representative of behaviors at the time.

Greif (2006b) emphasizes the importance of what he terms the “community responsibility system.” In other words, he argues that compliance was chiefly self-enforced via concerns regarding the protection of multilateral reputations. Greif argues that actions by trading partners when infractions did occur, for example, ostracism or counter-cheating, were the important mechanisms used to enforce compliance (cf. Richman, 2006 and the sources of breakdown in enforcement in the diamond industry; Richman 2017). In contrast, however, we conclude that even though the Mishneh Torah did not appear until around 1180 ce, the influence of the legal rulings and interpretations on which Maimonides relied, even if debated and imperfectly understood among observant Jews of the eleventh and twelfth centuries, must have permeated the world of the Maghribi of this period as well. Hence, although no one would claim that it produced perfect compliance or completely obviated the need for the mechanisms cited by Greif, or for the availability of Jewish courts, it is hard not to conclude that Jewish law would have had a major role in ensuring honest and compliant behavior in trade – as it did elsewhere in everyday life for the Maghribi.

The arguments and circumstantial evidence from the development of Jewish law and its place in Jewish society are strong. But what direct evidence exists from the Cairo Geniza that the traders were influenced in their business by their religion? Goitein (1974, p. 8; Goitein’s essay in this 1974 conference volume draws on material from volume 3 of his classic work, Goitein, 1967), writes,

The religion of the geniza people was a stern, straightforward, Talmudic type of piety, concerned with the strict fulfillment of the commandments [emphasis added] and with the pursuit of the study required for their knowledge.


When the writer of a Geniza letter says to his business friend, “You do not behave like a merchant,” he means: you do not live up to the standards accepted in the international community of great traders. (Goitein, 1974, p. 9)

In other words, he is chiding his friend for not behaving with the high standards of the great traders – presumably, the Maghribi.

Gil (1973, p. 310) quotes a letter of Nathan b. Nehorai, after receiving what he considered an unjust demand for payment: “And today is the third day that I have been fasting all the time, because of the cheating.” He describes his antagonist as “a madman, with no intelligence or fear of God.” And he “cites Psalms 69:5: ‘What I have not taken, then I shall return.’” Can we imagine Jamie Dimon, chief executive officer (CEO) of JPMorgan Chase, fasting for three days because of what he considered an unjust demand from Citigroup for payment, and quoting the bible regarding it?

Goitein (1974, p. 10) notes that in contracts “the religiosity of the contractors was the safeguard for the proper observance of the conditions specified” [emphasis added]. Moreover,

In deeds of release, the parties would free one another from responsibility in this world and in the world to come, which proves that a clean account was a very great concern for anyone who believed that one day he would be summoned before the heavenly judge.

Reputation did enter, but not entirely as Greif would have it: “Whenever something went wrong in business, the other party’s … piety… was invoked …. his reputation as a gentleman, was constantly cited” (Goitein, 1974, p. 10). In this context, reputation is linked to piety – that the merchant would become known as someone whose piety is either not genuine, or insufficiently practiced. It is not necessarily reputation for cheating per se that is of concern, but reputation as an observant Jewish man. Note that the Greif arguments may still work with this twist, but his argument becomes an accompaniment to the effects of genuine belief in and commitment to one’s religious practices and prescriptions, that is, Jewish law (cf. Richman, 2006).

Goitein remarks on the

constant awareness of God as inspiring our actions and innermost thoughts …. When we read the halakhic disquisitions written by those learned merchants on the reverse sides of business letters received by them, we notice they were well-versed in Talmudic law. (Goitein, 1974, p. 16)

Botticini and Eckstein (2012) argue convincingly that a critical characteristic of the Jewish diaspora that strongly affected their economic behavior over centuries was the absolute requirement of literacy in order for Jewish men to practice their religion. That is, Jewish men had to be able to study the Talmud. We note that the enforcement of literacy based on this reason also has the consequence of recognizing and enforcing the other aspects of Jewish law that, being divine in origin, mandated compliance. One cannot pick-and-choose here; it all comes as one. The Jewish traders who would make notes about halakha on the backs of business letters were also likely to abide by the ethical instructions in Jewish law.

Finally, Goitein (1974) notes,

Despite the harrowing shortcomings of the realities of their existence they had a clear vision of the ultimate good, because in their everyday life they were convinced they knew what was good and what was not. (p. 17)

Thus, the documentary record of the Cairo Geniza clearly suggests that the Jewish Maghribi traders were certainly observant – indeed, potentially model businessmen in that regard. So the argument that their behavior was policed successfully solely or largely because reputational penalties for cheating could be enforced in the closed “coalition” of traders, does not seem compelling. It was not self-interest alone, but heavenly interest that drove them.

Besides compliance on the basis of individual belief in mandatory compliance with Jewish law, our argument has two other potential components. First is the receipt of approbation from community members who perceive the individuals as compliant with Jewish law. Of course, membership in the group is easily identifiable to other members of the minority group, an aspect of the rationales from the middleman minorities model. Traders who cheated would be seen as violative of the law and shunned, consistent with Greif’s approach, but for an entirely different reason.

Second is the likely presence of what Hardin (1992, 2002) calls “encapsulated trust.” Hardin argues that trust exists when the first party believes that the second party includes the first party’s interest among the second party’s interests, and thus that the second party will take the first party’s interest into account when making decisions. Because the Maghribi traders were all observant Jews, a fact known of course to all the Maghribi merchants and agents, and because they knew their own committed positions regarding the mandates of Jewish law in commercial behavior, they could expect that their Maghribi counterparties would understand and take the counterparties’ interests into account. To do otherwise would be to assume that support for the law did not exist in the counterparty, and that it was acceptable to proceed as if the other party did not care about the law. Thus, the Maghribi trading network was in effect, if not necessarily in intent, structured to recognize and support encapsulated trust.

A condition necessary to perceptions of encapsulated trust, however, is the provision of information about the status of other actors in regard to their interests and how they regard the interests of others. Obviously, traders observe the behaviors of other traders, and can potentially observe whether the interests of others are factored into their decisions. In other words, traders depend on obtaining reports of past behaviors (cf. regarding sustaining the trading system, see the argument in Milgrom, North, & Weingast 1990). These are treated as intrinsically more reliable than claims of current behavior, and more reliable than predictions of future behavior. The theory of testaments (Mitnick, 1999a, 1999b, 2000, 2009, 2018c) argues that, under uncertainty, reports about the past will be treated as more credible than claims about the present, and claims about the present will be treated as more credible than predictions of the future. Hence, it is in the interest of the Maghribi traders to let other merchants know that they have been reliable agents in the past. The fact that the world of the Maghribi traders is bounded obviously promotes such exchange of information.

It is possible for the reputational mechanism offered by Greif for the Maghribi traders to also function, as a fourth effect, reinforcing the other mechanisms. But we feel a much more prominent explanation is that the traders were simply observant Jews who felt strong obligation to follow the dictates of the moral code embedded in their orthodoxy – indeed, how could we consider them observant without attention to the code of prescribed religious behavior that they perceived as mandatory – whose belief was not instrumental merely as a means to retain their reputations as observant Jews?

In essence, we have turned the argument on its head. Primacy is given to law, which then supports continued social and cultural homogeneity (which is seen as facilitative of and responsive to the dictates of the law). Because the law is primary and not instrumental, its content is critical to exchange outcomes (cf. Donaldson & Dunfee, 1994 on “hypernorms”). Where in the model of the Landa ethnically homogeneous group the outcomes of exchange are irrelevant as long as contract compliance was attained via agents performing as directed (they have been screened to be reliable agents), for the case of Jewish merchants observant of the law, both intents and outcomes, as well as procedures, are important. It is inconceivable to think of Jewish law serving as the code of ethics for the operation of a criminal group. Thus it matters not only that they be reliable, but also that they be honest.

Because such qualities matter in Jewish law, and because the basis of the law is in divine instruction, behaviors of heroic marginality follow. There is no reason for a Landa trader or Greif’s Maghribi trader to go above and beyond what is required of contract as long as he abides by the expectations of trading and forgoes cheating. Greif’s reputation-based self-enforcing coalition of traders will be concerned with meeting their contractual obligations so as to maintain their reputations. Doing more than that would be costly and lead to heightened expectations during the repeat engagements that are likely to occur; similar achievement of above-and-beyond results cannot be guaranteed under the highly risky circumstances of medieval trade (Gil, 2008) – indeed, the opposite is the most likely outcome. Thus, there is little to be gained by spending more on agency in order to achieve results better than the principal has contracted.

But, consistent with our arguments about the role of “fences outside fences” in Jewish law, there are many reasons and circumstances in which going above and beyond could be mandated in order to be in compliance with Jewish law, that is, divinely-mandated compliance. The result is that we would expect the observant Jewish Maghribi trader to go above and beyond in performance because doing so would be consistent with creating fences-outside-fences – that is, guaranteeing that the promised contract terms would be met in the highly risky world of Mediterranean trade. Thus, it is compliance done not only with calculation to meet the secular terms of a contract, but compliance that must be achieved with mandated assurance, because violation of the promises of a contract would also violate Jewish law. We conclude that the Jewish Maghribi trader would be far less likely to engage in cheating than traders who are not subject to such facilitative religious beliefs. And, as noted, Greif reports that the rate of cheating is far lower than for other merchants, such as merchants in Genoa (Greif, 2006a, p. 63).

In Fig. 1, we re-sort the mechanisms for solving the Landa Problem suggested by the full literature on middleman agency, including the case of the Maghribi traders, redacting the mechanisms into three literature groupings. For all three mechanisms, we treat the dependent variable as the production of ethical and reliable commercial agency behavior.

The first mechanism displays the basic logical structure of the middleman minorities literature in sociology and economic anthropology, including some aspects of Landa’s work. Ethnic and sect ties and “ethnic resources,” regional effects and perceived social distance, and the ability to distinguish on the basis of group solidarity and identity as a group member, permit economic principals to identify trustworthy economic agents, and to distinguish them from those not so viewed. In addition, agents in kinship relations are a special class and 
are accorded special status as trustworthy. The selection of economic agents on these bases leads to ethical and reliable commercial agency.

The second mechanism is constructed of elements taken largely from the literatures in behavioral economics and social cognition. The theory logics typically involve economizing behavior limited by cognitive factors, and instructed by perceptions of social norms. Overtly cultural and sociological factors, other than norms, are not as prominent as in the first mechanism. Here principals make use of economizing screening devices to select among agents; membership in certain groups and/or symbols of status are seen not in terms of the group’s ethnicity or other social characters, but as a shorthand for likely behavior in agency. Self-enforcing reputational “coalitions,” as in Greif, police conduct via efforts to maintain reputations and avoid negative sanctions for discovery of cheating; essential to the task is that the members of the participating group are recognizable to others in the coalition, all of whom are potential participants both in reputation maintenance and reputation sanctioning. Particular ethnic or social statuses are irrelevant. Predictions of behaviors are informed by the rational recognition of encapsulated trust; trust is not based on deep or historical relations, or cultural or normative factors. The final factor recognizes dependency of the principal and/or discretion in the agent, triggering the fiduciary norm. As a general social norm, however, it is situationally relational and independent of ethnic, regional, or cultural factors.

The third mechanism sees the production of ethical and reliable agent behavior as the consequence of genuine belief in and adherence to law or to other deep codes. Compliance is not calculative and not done on the basis of self-interest. The plenary nature of the legal code often affects a wide range of daily behaviors, which in turn reinforce automatic compliance with the code when agency is the behavior at issue.

It is conceivable to have social settings in which all three basic mechanisms coincide (even if only the third is sufficient to ensure compliance), for example, when ethnic and kinship relations provide assurances of agent fidelity, principals employ symbolic or representative screening devices to select among agents, reputational self-enforcing systems are set up, agents recognize principal dependency, triggering the fiduciary norm, and the applicable legal code is, at any rate, followed as a matter of genuine belief. Still, it is apparent that the primacy of law in the Jewish case suggests the possibility of a fundamentally different causal structure than those yet advanced in the literature on such commercial intermediaries.

A Theory of Heroic Marginality

The preceding discussion implicitly suggests a number of propositions regarding the occurrence of heroic marginality, as well as about its consequences. The theory logic in what follows makes a number of assumptions about the actors and their contexts. It is essentially an institutional theory in which the actors socially construct the reality presented to them (e.g., Berger & Luckmann, 1966; Powell & DiMaggio, 1991; Scott, 1995; Scott & Meyer, 1994; Thornton, Ocasio, & Lounsbury, 2012). Moreover, the actors’ ability to construct that reality is limited, flawed, and thus subject to distortions (e.g., Fiske & Taylor, 1991); with perfect knowledge, behaviors of heroic marginality would be sharply reduced unless they were subject themselves to noncalculative institutional support (e.g., mimetic in performance).

Central to the theory is the assumption of the social actors’ support for, and genuine belief in the law, using the model of belief in the divine basis of Jewish law to model cases consistent with what we have called deep codes. But the theory is intended to be generalizable; it is not a theory of Jewish law. Though the extent of support for law can obviously vary, and can be subject to empirical study, as in the political science and law-and-society literatures, in the present theory we assume that such support occurs. We leave to another time the development of theory about the nature and sources of any variable support occurring in a society in which support is divinely mandated. The problems and process of normative referencing (Mitnick, 1995) are present in any system in which actors serve normative objectives – what norms should guide this agency? – but in the case of halakha, the guide is clear.

The dependent variables relate to the production of prescriptions of behaviors of heroic marginality, and the occurrence of such behaviors. The behaviors themselves are assumed to increase the likelihood of production of excessive returns to the principal’s interest at levels above those specified as acceptable by the principal. Or the returns may exceed levels inherent in contractual provisions, or commonly-accepted practices, or what outside, disinterested observers not in the agency role might view as appropriate in the societal context, given the apparent circumstances of agency. The prescriptions for such “fences-outside-fences” behaviors may take the form of instructions to perform elaborate precautionary behaviors that take extreme care to avoid violation of obligations under the agency role established in implementing the law. The theory features the logical sequence from prescriptions to behaviors to outcomes of heroic marginality, with the last logical relation, between behaviors and outcomes, left for future work.

The term “bindingness” is used below to refer to the quality of instruction in a law that requires response. The more binding a law, the more that the subject actor is obligated to abide by it. Another way of looking at this is to specify that the actor’s discretion of whether or not to respond is increasingly limited. But the specific actions, that is, the means and timing of responsive actions to the instruction that are selected to implement compliance may or may not be clearly detailed. For example, the secular law’s requirement that taxes must be paid is absolutely binding; there is no choice as to whether or not to pay taxes. But the tax code often permits choice as to how the payments may be made and what exemptions may be claimed. Thus, even with binding laws, there can be residual uncertainty regarding which actions provide the most appropriate responses to the law’s binding requirements. Action is mandated, but exactly which response depends on how specifically the law itself describes what is to be done. Thus, whether or not actors achieve given levels of response, or go beyond that level, may (or may not) be discretionary.

Discretion is defined here as the degree of choice permitted the agent that can make a difference to the principal. That is, we treat discretion as relational, not merely as unanchored choice. Having discretion in the agent is meaningless in the contexts we are considering if there is no effect on the principal.

The term “prescription of behaviors of heroic marginality” is used to refer to implicit or explicit descriptions in materials based on or derived from the law that elaborate failsafe, precautionary compliance behaviors. In essence, they encourage and may describe in detail behaviors that go “above and beyond” by requiring the use of “fences-outside-fences” to ensure compliance with the basic law or legal principle.

The independent variables in the theory divide into four groups: qualities of the source of the law and/or of the lawgiver; content of the law; consequences of noncompliance with the law; and aspects of the relation of the agent (the law-follower) to the principal. The qualities of the source of the law and of the lawgiver include the level of the law’s reputed source and/or lawgiver, the credibility of the law’s source, and the credibility of the lawgiver. The content of the law variable includes the degree to which the law concerns subjects central to the religion of the community. The consequences-of-noncompliance category includes the degree to which noncompliance may result in threats to life, health, and safety. Finally, the aspects of the agency relation include the degree of dependence of the principal on the agent and the degree of uncertainty regarding appropriate compliance activities by the agent.

The dependent variables are the prescription of behaviors of heroic marginality, and the behaviors of heroic marginality (where we will posit that prescription is positively associated with consequent production of the behaviors).

Obviously, we can specify additional independent variables in these categories, as well as additional categories. Our purpose in choosing these variables is to sketch some of the most important and characteristic patterns of behavior in a theory of heroic marginality.

The theory of heroic marginality is depicted in Fig. 2.

Fig. 2. 
Theory of Heroic Marginality

Fig. 2.

Theory of Heroic Marginality

P1. The greater the prescription of behaviors of heroic marginality, the greater the behaviors of heroic marginality.

The theory provides that actions do indeed follow from the instructions that the law gives its agents; after all, the aim is to understand how certain behaviors of fidelity can occur more frequently. This is not by any means self-evident; the literature on public policy implementation, for example, makes clear that actions to implement public policy prescriptions do not necessarily flow as requested (see, e.g., Mazmanian & Sabatier, 1983; Mitnick & Backoff, 1984).

Of course, there is also a voluminous literature in social and cognitive psychology on the problematic link between attitude and behavior (e.g., Fishbein & Ajzen, 1975 is one classic work in this stream; for an application and discussion, see, e.g., Kurland, 1995a). A key element of the context of Jewish law, however, is the genuinely-preferred or terminal as against instrumental character of preferences shaped by such law. The ultimate source of the law is divine, and the law is central to the religion. Thus compliance, or attempted compliance, to the law’s specifications may be judged more likely (on the tie between moral obligation and behavioral intention, see Kurland, 1996a). As noted above, this genuinely-based belief in the law is critical to the logic of this theory.

Because the dependent variables are prescriptions and behaviors of heroic marginality, we have not included the logical final relationship of behaviors of heroic marginality producing outcomes of heroic marginality, that is, outcomes of excessive returns for the principal. Obviously, the theory is developed with the assumption that such outcomes may indeed be produced. If behaviors of heroic marginality had no consequences then the theory itself would be of considerably less interest. The theory needs to be extended to cover the nature and management of excessive returns.

P2. The higher the level of a law’s reputed source (and/or lawgiver), the greater the bindingness of the law.

P3. The greater the bindingness of a law, the greater the prescription of behaviors of heroic marginality.

Corollary. The more that law is ascribed most directly to the highest level source, for example, divine instruction, the greater the prescription of behaviors of heroic marginality.

In other words, when laws are constructed to be maximally binding, as happens when they are judged of divine instigation, actors that are under binding conditions would of course certainly respond, and will now use their residual discretion to choose behaviors that leave no doubt as to meeting the goal – by exceeding the goal, that is, by going above and beyond.

Jewish law is more lenient with regard to laws that are rabbinic in origin as opposed to law that is of biblical origin. We have already observed that, according to Torah law, if one discovers that s/he is wearing kilayim, a forbidden wool and linen mixture, s/he must remove the garment, even if it is in a public place. In this case, a man’s (or a woman’s) public honor is subjugated to the Torah injunction. Now, let us look at a corresponding leniency in rabbinic law. There is a Torah commandment in Judaism that one wear special fringes (tzitzit) on the corners of one’s garment (Numbers 15:38-39; Appel, 1975, p. 94; Hakohen, 1990, p. 23; and many others). If one tears his fringes (this commandment is only incumbent upon Jewish men), then halakha considers it as if he is no longer wearing the garment, but rather carrying the garment. Yet carrying is a prohibited activity on the Sabbath. What is prohibited, specifically, is the carrying of an object from a public to a private domain, and vice versa.

A karmelit is a type of space that is considered a public domain by rabbinic authority, but is not considered as such by biblical authority law (Roth, 1986, p. 27). Since carrying from the public to the private domain is forbidden on the Sabbath, and since wearing the unfit tzitzit is considered as carrying them, then one would expect that rabbinic law prohibits a person from carrying/wearing the offending garment from a karmelit to one’s home. On the contrary, the rabbinic law indeed permits the man to continue to wear the unfit fringes on his garment and walk from a karmelit to his home (Roth, 1986, p. 27). Since the prohibition against carrying/wearing the offending fringes is a rabbinic one, halakha chooses the more lenient ruling and protects the garment wearer from embarrassment. In the case above, where the prohibition against wearing the offending garment was biblical, halakha chooses the stricter ruling with no regard to the individual’s embarrassment.

Wearing kilayim or shatnez (described earlier in our discussion of the nature of halakha, including the differential effects of status as d’rabbanan and d’oraita) is as serious a crime in Judaism as eating non-Kosher food (Lehmann, 1988, 
p. 5). According to Lehmann (1988, p. 5), the Chafetz Chayim (Rabbi Yisrael Meir Hakohen) instructed Jewish soldiers in the Czar’s army to wear army uniforms which contained kilayim only if the penalty for noncompliance consisted of capital punishment. He also advised these Jewish soldiers that it was better to go bareheaded than wear military headgear containing shatnez if such headgear was not obligatory.

Today’s military headgear serves a more practical protective function than did military headgear in Czarist times. A modern rabbi might possibly rule differently on the headgear question, although this is not necessarily the case. In any case, by this example, we see that an unexplainable biblical injunction has a greater bindingness than Czarist military law, and it may have greater bindingness than simple common sense (that pertaining to wearing protective headgear in hostile conditions).

P4. The greater the credibility of the legal source and of the person of the lawgiver, the greater the prescription of behaviors of heroic marginality.

The Jewish people hold their forebears in a position of the highest regard. As years go by in Jewish history, each successive generation is further removed from the generation that experienced divine revelation at Mount Sinai (see Exodus 20 and thereafter). The notion of honoring those who have come before is expressed in various biblical commandments, such as those commanding one to honor one’s parents (Exodus 20:12) and commanding one to rise up before an aged man (Leviticus 19:32). Similarly, earlier generations of Torah scholars are given more weight regarding halakhic decision making. The Mishnah, which makes up the core of the Talmud, and the other teachings of the authors (collectively known as the Tannaim) of the Mishnah, are treated as incontrovertible fact. Earlier scholarship is always relied upon as having greater weight because greater consideration is due to those who were nearer in time to the primary objective source of truth, namely divine revelation to the generations that lived in Sinai (Steinsaltz, 1989, 
p. 4). Of course, the opinions of later generations can result in refutation of earlier legal judgments (see Berkovits, 1983 for a full discussion of the nature of halakhic authority), but the “ … essential character and principles of fundamental laws …” are drawn from the earlier sages. Thus, the further back in time we journey through Jewish scholarship, the more serious and meaningful the teaching and advice.

Thus, the more that agents perceive the instruction as coming via actors closer to divine origination, the more the instructions are believed to be genuine and so credible – and, hence, binding. Thus, via the logic of P2 and P3, the more that behaviors of heroic marginality are prescribed.

P5. The more that the law concerns subjects of central significance to the religion [deep code] of the community to which it applies, the greater the prescription of behaviors of heroic marginality.

Without a doubt, study of Torah (which translates to “law”, or “doctrine”) is one of the most basic and fundamental forms of observance of the Jewish religion. Learning Torah is itself a commandment of the Torah (Deuteronomy 6:7). It has been argued that, within Judaism, study of Torah is even more important than the act of prayer itself (Lamm, 1991, p. 317), and that study is, in fact, the highest form of worship. One can hardly argue with the notion that the bulk of Jewish culture and history is based upon the Jewish people’s study and absorption of the Torah (see Botticini & Eckstein, 2012 re the long-term effects of requirement of such study).

So strong has been the emphasis on study and interpretation of the Torah that for hundreds of years it was forbidden to accept payment for teaching the Torah; it was feared that acceptance of compensation for teaching would taint the teaching. Thus, until the fifteenth century, the greatest Talmudic scholars all pursued other occupations in order to earn their support (Hertz, 1960, p. 757).

Given that Torah study is so central to Jewish religion and observance, it is not surprising that heroic marginality in the study of Torah is a prescribed behavior in Judaism. Jews are required to study Torah every day for their entire lives, both day and night. They must, in addition, teach Torah to their children, and the commandment to study has no regard for whether one has wealth or lives in poverty (Hakohen, 1990, pp. 27–28). Torah study is so fundamentally important in Judaism that no interruptions are tolerated: “Rabbi Levi said: A man who breaks off the study of Torah to engage in idle talk will be made to eat cinders of the broom tree,” whose hot cinders never die out (Babylonian Talmud, Chagigah 12b, cited in Bialik & Ravnitzky, 1992, p. 406). The sages in Ethics of the Fathers (Pirkei avos 3:9) even taught that if one is walking down the road reviewing a Torah lesson and stops to remark on the beauty of a tree, then that person has incurred a guilt that is only removable by death (https://www.sefaria.
org/Pirkei_Avot.3.7?lang=bi&with=all&lang2=en). Some of Judaism’s greatest heroes are those that maintain Torah study even under the harshest of conditions:

It was said of Rabbi Judah ben Ilai’s generation: [Their poverty was so great that] six disciples had to cover their bodies with but one cloak. Nevertheless, they occupied themselves with Torah. (Babylonian Talmud, Sanhedrin 20a, cited in Bialik & Ravnitzky, 1992, p. 407)

Jews are entreated to never desist from Torah study:

Ben Bag Bag says: Search in it and search in it, since everything is in it. And in it should you look, and grow old and be worn in it; and from it do not move, since there is no characteristic greater than it. (Pirkei avos 5:22, https://www.sefaria.org/Pirkei_Avot.5.22?lang=bi&with=all&lang2=en)

P6. Where noncompliance to the law may result in threats to life, health, or safety (as against threats to economic status or security), nonobservance of the law’s prescriptions is permitted. But heroic steps must be taken to insure that nonobservance is minimized, with compliance restored once the life or health crisis is over.

“Violate one Sabbath on his behalf that he may observe many Sabbaths” (Babylonian Talmud, Yoma 85b, cited in Bleich, 1977, p. 129).

Even within the seemingly absolutist realm of Jewish observance, there is a hierarchy of values (Bleich, 1977, p. 130). When it comes to the protection of human life, there is little, if anything, that takes precedence in Jewish law. Even the inviolable Sabbath becomes a negotiable commodity when it comes to matters of life and death. As expressed in the Talmudic passage above, the ultimate goal of maximizing observance to divine law implies that certain regulations can be transgressed when the result is future obedience. Rabbi J. David Bleich (1977) describes this phenomenon as something of a “… calculus of observance” (p. 130) that allows temporary suspension of the law in exchange for future conformity with it.

Take, for the example, the prohibition against using the automobile on the Sabbath. This prohibition is suspended when someone’s life is in danger insofar as the operation of an automobile will assist the patient. The automobile may be used to transport either the patient or a physician (Bleich, 1977, p. 135). The use of the automobile, however, must be conducted in such a way as to minimize transgression of the Sabbath.

Rabbi Bleich (1977, pp. 135–140) outlines a series of suggestions by Rabbi Joshua Neuwirth that deal with this problem. Among other things, Rabbi Neuwirth suggests that the shortest route possible be taken to the hospital in order to minimize the desecration of the Sabbath, although a hospital that is farther away may be chosen as the destination if the care there is superior. He also proposes that the brake pedal should be applied only to avoid danger, as the brake lights are illuminated when the pedal is depressed (turning lights on or off violates another Sabbath prohibition). When possible, the hand brake should be used instead, and hand signals should be used instead of directional signals if possible. If it prevents the dome light from turning on, a person should enter the car from a back door and climb over the seat to drive. Unless the car is needed for life-saving immediately afterwards, headlights should not be switched off at the end of the trip. A non-Jew should be asked to turn off the ignition. The keys may not be removed if in a place where no carrying is permitted, and so on. This is only the briefest of summaries of these rules. Rabbi Ephraim Eliezer Padawer (1990, pp. 126–155) also details a similar set of rules in the case of childbirth occurring on the Sabbath. Rabbi Padawer actually rules that it is preferable to call a non-Jewish car service in order to transport the patient.

Another example of halakhic in the case of health pertains to fast days. One who is ill or pregnant is not necessarily required to fast, even if there is Torah commandment to do so (some fasts are rabbinically ordained). On Yom Kippur, the holiest day of the year and the most important Jewish fast day, it has been ruled that “… if someone is ill, even if there is no absolute danger to his life, it is well to give him food and drink bit by bit …” (haLevi, 1988, p. 25).

There are only three cases where death is preferred over the violation of a precept of Judaism: idolatry, murder, and some forms of sexual immorality (Bleich, 1981, p. 159).

P7. The more dependent the principal on the behaviors of his or her agent, the greater the prescription of behaviors of heroic marginality to the agent.

Behaviors of heroic marginality are clearly instructed in the case of the commandment to return a lost object to the Jew who has lost it (Deuteronomy 22:1). Obviously, owners are completely dependent on the unknown party who recovers the lost item. Jewish teaching urges one to go to great lengths to fulfill the commandment of returning lost objects (for a discussion, see, e.g., https://www.ou.org/torah/mitzvot/taryag/mitzvah538/). In Sifre Devarim, a Tannaitic commentary on Deuteronomy (recall that the Tannaim were the authors of the Mishnah, the core of the Talmud, and lived from 20 ce to 200 ce), it is taught:

If one returns the beast, and it runs away again, and when he once more returns it, it runs away a second time, even if this goes on five times, he must return it each time …. (Hammer, 1986, Piskha 222, 234)

Sifre Devarim also says that if one returns the object or animal to a place where others can watch it, and it is then stolen or lost, the person is still responsible for it, up until it is returned to its rightful owner (Hammer, 1986, Piskha 222, 234).

Rabbi Abraham Chill (1974) points out that the commandment entails caring for the possessions of others in every respect (p. 453). This is illustrated by a story in the Palestinian Talmud (Demai, I:3, 22a; Sarason, 1982, p. 44) about two poor men who deposited two seahs (the volume of approximately 288 eggs; see Steinsaltz, 1989, pp. 286–288) of barley with Rabbi Pinchas ben Yair for safekeeping and then forgot about the grain. He sowed and reaped it in their absence and when the men finally remembered the grain and returned, he said, “Bring camels and asses [for lading] and take back your barley.” Rabbi Chill (1974, p. 453) says that the commandment is so comprehensive regarding others’ property that one who “… sees that an onrushing river is about to demolish the home of another must work feverishly to put up a dam or dike ….”

Dependency as an element triggering behavior featuring exceptional levels of caretaking is actually a well-discussed feature of several areas of English common law. As we discussed earlier, it is a feature of the instructions surrounding the fiduciary principle, with exceptional caretaking prescribed in many contexts in the laws of agency, trust, and contract. In general, agents acting for principals who are highly dependent on the agent’s activities (e.g., where the principal is an incompetent party for whom the agent is acting as trustee) must take extreme care to act in the principal’s behalf (see, e.g., Finn, 1977; Frankel, 2011; Scott, 1949; and the sources cited earlier; on duties of professional fiduciaries to take care to prevent harm, see, e.g., Frankel, 2011, pp. 160–164). As noted earlier, the fiduciary norm proposed by Mitnick (1973, 1974, 1975a, 1975b) and Stinchcombe (1986) is said to characterize many social relations of “acting for;” it is instructed in agents acting with discretion with respect to their principal’s interest, and it is especially prescribed where the principal is highly dependent on the agent’s actions (see also Clark, 1985; Easterbrook & Fischel, 1993). When dependency is high, agents guided by the fiduciary norm should rationally take precautionary action that ensures that the return to the principal will not fall below the expected level of benefit. Such action may tend to produce returns that exceed the targeted level, assuming of course that higher returns are at any rate preferred.

Uncertainty regarding outcomes enhances the prescription to take precautionary action, given the bindingness of the law. The fiduciary obligation in Anglo-American law to take care has obvious similarity to the “hashavas aveidah,” the mitzvah (commandment) to, in essence, take care of and return lost objects. Note that in the law of agency, however, the obligation to serve only the principal’s interest originates with contract; the agent is acting under orders and/or under a pre-existing contractual agreement, as would a trustee for an incompetent person, as noted above. The obligation in Jewish law is not contractual; it is triggered by events, hence is closer to how the fiduciary norm as a social norm operates. Both the Jewish mitzvah and the social norm are driven by perceptions of dependency, and both require the fiduciary to go to great lengths to serve the dependent party. The production of the precautionary behaviors that we label heroic marginality is facilitated by the asymmetric character of acting for another under uncertainty with the requirement of a protective outcome for the dependent. The agent cannot merely make a choice of action using a reasonable calculation under risk – the outcome cannot be allowed to fall on the negative side. Hence, as we noted, given uncertainty, an overproduction of effort may be required to guard against a negative, if unintended outcome. The result is behavior of heroic marginality.

P8. Where the bindingness of the law is high: the greater the uncertainty regarding the activities that would constitute compliance, the greater the prescription of behaviors of heroic marginality.

Thus, Fig. 2 shows that bindingness functions essentially as a mediating variable in the theory in the case of P2 and P3, and a moderating variable in the case of P8.

Because of respect for the law as a whole, given that it is taken as having divine origin (whether or not given rules are d’oraita or d’rabbaban), cases of uncertainty are always resolved via the default position of taking extra care. While specific actions may be debated, the actions themselves must be designed to guarantee compliance with the law.

One case of uncertainty leading to stricter watchfulness is that of the second festival day. The Jewish calendar is a lunar calendar, with the end/beginning of the month landing at the new moon. The Great Sanhedrin in Jerusalem, consisting of 71 judges, all rabbis, would listen to and question the testimony of witnesses who had claimed to see the new moon. With the confirmation of two witnesses to the new moon, the beginnings of the months were proclaimed by the judges. That is, an extra witness was required. This was important because the Torah often specifies observance of a festival on a fixed day of the month. Bonfires were lit to announce the news of the new month to distant communities. But decoy bonfires were lit by hostile Samaritans. So messengers were used instead of bonfires, but they could not reliably reach distant communities in time. To ensure that the festival would be celebrated on the biblically specified date, a second day would be added to each festival. After a fixed Jewish calendar was developed in the fourth century ce, the correct date was known in advance to all, but the Sages determined that the practice of adding a day to the festival had become hallowed through long practice. What had until then been observed because of doubt became fixed as rabbinical enactment (takkanah). The rabbinic decree of adding a second day has been preserved as if it had the same sanctity as a d’oraita or biblical ruling even though the original doubt has been resolved (see Donin, 1972, pp. 210–212).

Limitations and Some Potential Applications and Extensions

Obviously, further development of this approach as descriptive theory may be helpful in providing explanations for certain agency-like behaviors. But the theory of heroic marginality also suggests a variety of practical, if speculative, contexts and applications. Given the length of argument in this article, we will only sketch a few here. These speculative suggestions for practical design are summarized in Table 2.

Table 2.

System Design Features Suggested by the Theory of Heroic Marginality.

• The development and use of deep codes, genuinely-preferred by the community
• Support for the myth of the perfect agent
• Use of pantheonic directorates
• Identification of permitted behaviors of heroic marginality to avoid potential harm
• The creation of self-enforcing professional communities employing deep codes

In the case of halakha, the prescriptions, behaviors, and outcomes of heroic marginality depend on the existence of genuine and strong belief in the law, anchored in terminal values. If halakha can serve as a model suggesting effective means of producing behaviors of extreme fidelity, then ethical codes that seek the same ends must be based on genuine, deeply-held belief systems. It is not sufficient, for example, to merely promulgate corporate codes of ethics, or to train the code, advertise the code, and even reward with respect to the code (cf. Ford & Richardson, 1994, p. 216; Kurland, 1996a, pp. 69–70, 1996b, p. 303). To be most effective, ethical codes must be genuinely, not instrumentally preferred. We have suggested, moreover, that status as a deep code is critical, so must feature, besides genuine belief, status as “deep driver” or primary rationalizing source; use of terminal values; wide support; and systematic organization as a code.

Consider now each of the four categories of independent variables in turn (see Fig. 2): qualities of source of law and lawgiver, content of law, consequences for noncompliance, and aspects of the relation between agent and principal.

One obvious concern is what we will call the problem of the lawgiver. By using the model of a particular religion, we have in essence assumed away this problem: The lawgiver is obviously taken as divine, and it is divinity that makes compliance so mandatory.

The source (which may or may not be coterminous with the lawgiver), if not divine, must be perceived as being at the highest authoritative level in the setting and must possess the highest credibility to the agent, as must the lawgiver. This is consistent with the standard prescription in the implementation of corporate compliance activities, that is, that top management support is essential (see, e.g., Ford & Richardson, 1994). We have spoken of the “source” of the law and of the “lawgiver.” Obviously, it is not likely that the CEO will come down from the C-suite holding two tablets with the corporate code of conduct on them. In organizational settings, the lawgivers are, in essence, the authorities who promulgate the rules and enforce compliance.

But the recommendation must extend further, to what has been termed the myth of the perfect agent (Mitnick, 1985, 2009; cf. Meyer & Rowan, 1977). The argument is that organizational agents are likely to maximize their commitment to the organization and their performance for it if they believe that top managers (especially the chief executive, who, in the context described here is also the “lawgiver”) will serve as a perfect agent of the organization’s espoused values whenever the employee encounters conflicts with or threats to those values. Moreover, the chief executive will behave without bias, producing objectively fair and responsive decisions. Though such an ideal may fall short of divine, and be difficult to obtain in practice, if only because individual assessments of situational fairness and responsiveness to organizational formal norms differ, it does suggest that governors of the organization, for example, the board of directors, should insist on spotless behavior in the executive office. Perceptions of self-interested or opportunistic behaviors (e.g., excessive top-level rewards even when organizational performance is mediocre or worse) will be at odds with this approach.

Moreover, members of an organization practicing under a deep code are motivated by well-understood cognitive processes to project that perfection backward onto those in authority: The consistency of the code produces expectations that the individuals directing behaviors under it will also behave with consistency to the code. The danger, of course, is that such expectations will be misplaced. The appearance of perfect agency can also be a cloak for self-interested abuse.

In addition, the perceived identity of top-level organizational governors makes a difference. While divinity in chief executives may be suspect, ascription of heroic status to members of the board and to top managers may be more credible. It has been suggested that all organizations tend to create pantheonic directorates at their peaks (Mitnick, 1985, 2009). These groups of people with heroic status in the society serve a variety of functions for the organization, including making implementation of the myth of the perfect agent more credible. Thus boards are typically composed of people of high ascribed status and achievements; the popularity of CEOs and former CEOs, former senior government officials such as government department secretaries, astronauts, university presidents, and even former US presidents as corporate directors are among countless other examples. When organizational law is said to flow from such provenance, it is likely to be perceived as more binding and to contain more prescriptions of behaviors of heroic marginality, according to our theory.

The content of the law may relate to or incorporate value elements of predominant religions in the community but at any rate should be supported by or incorporated in deep codes that serve in that role and must reflect terminal values of the community.8 Thus codes by definition must reflect deeply- and genuinely-held community beliefs responsive to those terminal values, and these beliefs will generate and/or prescribe behavior that is held as acceptable (on the fundamental character of the normative base of the community, cf. Donaldson & Preston, 1995; Frederick, 1995; Mitnick, 1995; Wood, 1991). Thus, if, for example, the context is corporate codes, then such codes must be seen not merely as procedural guidelines for permitted and not-permitted operational behaviors, but as fundamentally obligated things to do that are anchored in the deepest beliefs and terminal values, widely-shared in the relevant community.

Of course, the design, implementation, and enforcement of deep codes are nontrivial tasks. It borders on hand-waving to argue that deep beliefs will be obtained magically to complete the design of a compliance system. Yet such belief systems are not so improbable nor so uncommon; we have used the case of Jewish law, halakha, as a model for this. And, as noted above, Jews are not the only group that have performed as trustworthy agents. Still, we leave the very critical questions of where to obtain and how to implement deep codes for future work. Indeed, we believe that these questions are of such fundamental interest that they may be seen as a component of what we call behavioral religion, the study of the behavioral patterns and consequences in organizations and in wider societal settings associated with the functioning of deep codes (on the economics of religious belief, see Hardin, 1997; see also Margolis, 1997; there is of course a significant literature in the tradition of Weber (1958) on religiosity and economic behavior/economic development/corporate behavior, see, e.g., Blau, 2017; Cai & 
Shi, 2017; Chintrakarn, Jiraporn, Tong, & Chatjuthamard, 2017; Conroy & Emerson, 2004; Longenecker, McKinney, & Moore, 2004). The lessons of behavioral religion would extend as well to the design of human systems in which the planned functioning of deep codes in alliance with other design features and in particular societal settings can generate desired outcomes.

Note also that deep-code-like systems can exist and function effectively, that is, what we earlier termed, organizing systems for fidelity, without using ethical – 
that is, right–wrong identifying – rule systems, though perhaps we would most often see such systems. We did not require that in our definition of a deep code. Richard White (2003) observed that business managers in the late-nineteenth century, a period of widespread corruption, could operate effectively by following a code based on “character.” In this sense, character (and, as well, trust) was not related to what we would otherwise consider behaviors of honesty and integrity but simply behaviors of promise-keeping. One could do business, and repeat business, with someone who kept his promises – even if the business itself involved corrupt manipulations.

It was really this sense of “character” that J. P. Morgan was prioritizing in his famous remarks to Samuel Untermyer before the Pujo Committee (1912–1913): “Because a man I do not trust could not get money from me on all the bonds in Christendom.” (Morgan’s testimony, which is in the public domain, came on December 18–19, 2012, in a Congressional investigation of the “money center” said to corruptly dominate American business; https://fraser.stlouisfed.org/title/80/item/23671, pp. 1083–1084). Thus, the Landa Problem – the problem of identifying a trustworthy economic exchange partner – in this case, one who would keep his promises – was of high relevance for the industrialists of the era.

Morgan’s notion of “character” raises a problem for the design of systems employing deep codes: It does make a difference what’s in the code. The terminal values in the code that Morgan seemed to espouse did not include modern notions of integrity. We have used religion as a model partly because most religions include codes of ethics, that is, rules for distinguishing right from wrong behaviors. Our model of a deep code requires such a systematic rule set or code, if not necessarily a code of ethics per se.

In the case of the category of consequences for noncompliance in our model, it may be helpful to construct situations that produce permitted behaviors of heroic marginality directed at preventing negative consequences whenever compliance with organizational directives would have serious, debilitating effects. When society has a clear notion regarding preferred outcomes, such as preservation of life and health, noncompliance with the law is to be more than merely permitted; it is instructed, though with strong preferences to hew as closely as possible to the law’s provisions. In essence, to put it in a common context, the theory suggests that whistle-blowing is to be more than permitted; it is essential once threats to life or health are established and other routes to redress within the procedural law of the corporation are blocked. Indeed, because the values of life and health, and not the derived instructions of the law, are primary here, behaviors of heroic marginality to realize those values, and avoid and remove the threats to life and health, are clearly expected. Those efforts must remain as far as possible within the law, but must not lose sight of the basic values of the community.

Perhaps one reason for the attractiveness of the professions to Jewish people is that the governance of professions is typically provided by a mandatory code of ethics that contains prescriptions of preferred outcomes (for a compelling explanation that relies heavily on the role of halakha and the literacy it required, see Botticini & 
Eckstein, 2012; for a good summary of their thesis, see https://www.pbs.org/newshour/economy/the-chosen-few-a-new-explanati; for the status at the end of the nineteenth century, see http://www.jewishencyclopedia.com/articles/12379-professions-statistics). Hence the professional code mimics in some ways the mandatory ethical prescriptions of halakha anchored in terminal values. 
The physician must “do no harm”; the lawyer must not only do his or her best for the client but is also an officer of the court and ultimately an implementer of public policy regarding fairness and due process in our legal system. These requirements are mandatory because they are linked to terminal values. It is not acceptable to mostly who do no harm; one cannot mostly represent a client nor get most of justice. The downside result is not acceptable according to the prescriptive norms, however often it may occur in practice in an imperfect world.

Thus physicians (and, presumably lawyers) must take extreme steps to avoid the possibility of harm creation. In the case of the law, especially in representative roles in which clients are dependent, the agent must be a pure fiduciary and, even after contracting for his/her own compensation for performance, take no account of personal reward, only client reward, as part of action in the fiduciary role (on professional communities and normative control, see, e.g., Barley, Meyer, & 
Gash, 1988; see also Kurland, 1996b).

Often, the possibility of harm generation is uncertain, yet that uncertainty coexists with the strong instruction to avoid harm. The consequence is prescription, and behavior, of heroic marginality: If the possibility of generation of harm is imperfectly known (hence, it remains a possibility), then precautionary action must go sufficiently far to avoid that possibility; a certain fence must be erected outside a likely fence. Thus, like the judges Shmuel and Ameimar, though perhaps not with the same extreme sensitivity to appearances, modern judges recuse themselves in cases in which they have even nominal ties to the parties before them, and public administrators remove themselves from regulatory cases if they own even small amounts of stock in a corporation subject to regulation.

Hence the creation of systems of dependency where effects are uncertain, backed by the societal prescription of deep codes of harm avoidance, may produce extreme care in compliance behaviors. And, in general, in complex societal settings, effects are going to be uncertain. The idea of such a design is to let the agent choose more extreme levels of harm avoidance than explicit regulatory instruction would generate. Thus the law as a deep code, like halakha, would function via individually-chosen precautionary behaviors to produce outcomes more desirable for the organization and for society than explicit, external regulatory instructions would obtain.

The costs of heroic marginality are, of course, apparent; this is not a panacea for the production of ethical behaviors. Taking precautions that may not prove necessary is costly. The possibility exists of Merton-like goal displacement (Merton, 1957) as actors lose sight of the basic norms instructing the compliance behaviors. The problems of professional communities are well-described, including economic protection and collusion, resistance to innovative outside influences, protection for community members who violate norms of service to outsiders, and so on (see, e.g., Van Maanen & Barley, 1984). And it is not clear that in all professional communities we see the kind of deep, genuinely-preferred system of governing norms that we posit here. But, on the positive side, those norms are more often followed than not. Thus it is possible that heroic attempts to comply, encouraged by a deep code and monitored by other practitioners, would actually result in compliance.

The model we use, Jewish law, was for centuries exclusively controlled by and practiced by men, even as its prescriptions extended to both men and women and to virtually every corner of everyday life. Differences remain to this day among Judaism’s varied sects. But our approach looked at how Jewish law functioned with respect to social compliance in the Middle Ages, and generalizes its compliance-related features (see Fig. 2). Our approach does not in any way assume that deep codes must mimic the specific contents of Jewish law, including in regard to such gender issues.

Indeed, we consider the theoretical arguments to be at least intendedly gender neutral at the level we present them. For example, as noted, reference to “heroic” in no way references, say, mythological figures who were predominantly male, such that heroic behavior was interpretable as male in character. Our use of “heroic” has no special relevance to that whatsoever. But the effects of gender differences are likely to be highly relevant in the design and management of successful deep codes as we go beyond the major structural features of such codes to create appropriate and effective codes for particular settings.

One concern regarding the model, however, is whether it requires homogeneity among the population of relevance. Greif requires that in his reputation sanctioning model. The EHMG model of Landa requires it simply because it is the key screening device for identifying reliable agents. But Jewish law, as Goitein (1974) points out, requires each person to meet its commandments, without an intermediary. Although we point out potential collective enforcement mechanisms, such as approbation for abiding by Torah teachings, Jewish law requires individual study and assumes individual compliance. Still, taken as a whole, of course, the divinely-originated law is rigid, and adaptable (e.g., takkanah) only in a limited respect. Thus, practitioners are likely to have some features in common, if only because the uniform law is aimed at uniform compliance behaviors.

So, when the approach is translated to aid in constructing a deep code, we are in fact assuming that the code would be reasonably stable and consistently applied. The issue would be whether such a code could be effective if its subjects were diverse. It is probably true that the argument works best when practitioners understand and employ the code in parallel ways: They are all, say, professionals in the same area, or members of a military unit, or even members of an ethnically-homogeneous group. Such uniformity enables encapsulated trust.

But the logic does not require homogeneity, even as homogeneity is undoubtedly facilitative. The issue is creation and operation of a successful deep code, and that does require the presence of a rationalizing deep driver so that specified behaviors can be explained; genuine, committed belief; support via a set of identified terminal values; wide sharing and practice in the community, with reinforcement from others; and systematics in the code itself. The manner in which such a code does or does not interact successfully with populations that are diverse in different ways is an area for future study.

If the model of the professions is useful here, then one potential normative application of the approach – suggested by our arguments regarding the mechanisms by which the Landa Problem may be addressed – would be the development of revered normative systems set up to guide communities of endeavor in which the outcomes for failure of performance reflect serious societal consequences. In essence, the goal is to establish “self-enforcing communities” anchored in a terminal values-anchored, outcome-oriented law among a set of practitioners whose identity, as well as performance outcomes, were observable to others in the community (cf. the shared and both mutually and self-enforced cultural community of Herbert Kaufman’s Forest Service; Kaufman, 19609). Where divinity and similar authoritative constructions are absent, the normative systems themselves can become reified so that the lawgiver is emergent from the system; social actors see themselves as being guided by the spirit or essence of the community. An example is the military unit whose members are willing to engage in extreme, personally dangerous behaviors in order to maintain the expectations of the community-as-lawgiver; for example, one must uphold the values of the Marines.

In conclusion, we speculate where the lessons of the Maghribi might lead us today, were we to apply them to the modern corporation.

For example, let us assume that a new MBA degree created not only an identifiable credential and a toolbox for skillful management, but recognizable membership in what we’ll call a PMC: professionalized management community. The values held for and pursued by such a professional community would be more than procedural. In this perhaps utopian conception (we are describing a speculative system, and ignoring the thorny question of how to establish such a system), many basic values would undoubtedly conflict – efficiency and wealth creation may not always find themselves consistent with, say, protection of the natural environment as an end (i.e., a terminal value). But the new business law would attempt to identify, reflecting terminal values and strong commitment to genuinely-held beliefs, a set of statements regarding consensual, directed outcomes, from prohibitions against corruption to the production of safe products that would constitute a deep code for the members of the PMC.

Note that the identity of the terminal values – what are valued as ends – makes a difference, and the manner in which those values relate to the rationalizing component of the code as well as support the rules of the code are critical; there are plenty of examples of perverse organizational cultures, and this design is intended to avoid such an outcome. The PMC members would be expected to act with discretion as fiduciaries for the deep code. The code is not supposed to produce rigidity and delay adaptation in the organization by simply trying to anticipate and prescribe all behaviors; actions are discretionary but referenced to the code’s rationalization and terminal values. In other words, action is guided but must be supported by reasons. It is critical that the code be set up as a deep code: be publicly explainable, genuinely and strongly supported, reflect those terminal values, be shared widely among its practitioners, and be systematic in construction.

Moreover, we would introduce into this bounded, recognizable community the threats of ostracism and/or of damage to reputation from the Greif model: Retention of the entry ticket, the new MBA degree, would then depend on compliance. This design would go far beyond taking a professional oath or signing a promise to abide by a code of conduct. Precautionary levels of compliance would be expected of group members, and noncompliance would be detected and publicly policed by the group itself in order to maintain the group’s credibility and identity. In this uncertain and dependent world, the fences erected outside other fences would preserve what is highly valued in society, and prevent tragic losses.



Heroic marginality may take the form of production of extreme returns to the principal’s interest at levels above those originally specified as acceptable by the principal (assuming of course that more-is-better), just to avoid any risk that returns might fall below the acceptable level. Or returns may exceed levels inherent in contractual provisions or in common practice or what outside disinterested observers not in the agency role might view as appropriate, given the apparent circumstances of agency.

Heroic marginality may also be observed in situations that appear not to be ones of agency, in that the party performing the marginal behavior is not apparently in the role of acting for another. For example, under Jewish law, halakha, many situations of compliance involve scrupulous concern for the avoidance of prohibited behaviors. The avoidance behaviors may take the form of heroic marginality, as described later in this article. As discussed below, however, compliance to halakha is not a judgmental issue concerning law made by fellow humans that may or may not reflect human frailties or judgmental errors; halakha is a law deduced from what is taken as an original divine basis. There is no choice about compliance.

Because the individual is serving divine intent, an implicit agency relation exists. Note also that in secular common law (American Law Institute, 2006), agency is performed under the actual or assumed directives or orders of the principal; in this article, agency relations are treated in the more general sense used in social theory, as relations of “acting for” (on agency theory, see Mitnick, 2018a). But because the commandments of Jewish law have a divine basis, and so are mandatory, they bear a structural similarity to legal agency, that is, acting under the orders of the principal.


We will use the terms “middleman minorities” and “ethnically homogeneous middleman group,” despite their lack of gender neutrality, because they were used extensively in the older literature. To change to another term now, however appropriate and desirable, would cause confusion in reference to works in that literature. A term sometimes used to refer to similar groups, “ethnic entrepreneurs,” is not a satisfactory substitute: the groups in question were not always distinguished purely on ethnic grounds (this article focuses on distinction based on religious practices), and the groups themselves are not always entrepreneurs (in this article they are merchants acting both as agents and principals); they are not entrepreneurs per se.


There are numerous sources in English on Jewish law, including business law, and on the behavior of Jews in commerce; for examples, see Barzel, 1992; Block, 1990; Cohen, 1991; Dorff, 1997; Elon, 1994; Fand, 1989; Fendel, 1983; Gershfield, 1988; Green, 1997; Hauptman, 1994; Hecht et al., 1996; Hertz, 1960; Jung, 1978; Levine, 1980, 1987; Liebermann, 1981; Miller, 1993; ; Ohrenstein & Gordon, 1992; Passamaneck, 1983; Pava, 1997; Rabinovich, 1993; Schnall, 1993; Siker, Donahue, & Green, 1991; Tamari, 1987, 1990, 1991, 1994, 1997; Wagschal, 1990; Zipperstein, 1983; see also a number of relevant articles in the Encyclopedia Judaica; on religion in organizations, see, e.g., Tracey, 2012). An especially relevant argument is advanced by Botticini and Eckstein (2012), who argue that the absolute requirement of literacy in Judaism in order to be able to study the divinely originated written Torah and, later, the oral Torah as written, and the voluminous codifications and commentaries that appeared, together with external historical shocks, strongly affected the economic behavior of the Jewish people, including the occupations to which they gravitated. For our purposes, this argument also supports making the case that the content of the divine commandments must have shaped behaviors – given the mandate that all men (at least in the historical context) read and study it. Hence, we have reason to expect that observant Maghribi traders would tend to avoid cheating because the Torah tells them to.


There is a bitter dispute in the literature regarding the adequacy with which Avner Greif cited the works of Janet Landa (see Rowley, 2009 and response by Greif 2009). Greif argues that the mechanisms of their explanations of the reliable behaviors of the middleman agents are different, and that he cited Landa appropriately. Landa’s published work preceded Greif by eight years; Greif cites only unpublished work by Landa and does not pose her work as a prime predecessor. Greif is correct in noting that neither he nor Landa (1981) cited the original work by Bonacich (1973) on middleman minorities; he also notes in his response piece (Greif, 2009) work by Sombart (1953) – which Landa does not cite and even Greif does not cite in his major book (2006a). Neither Landa (1981) nor Greif (2006a) cite Banfield (1958), who was among the first to raise the critical question about the identification of trustworthy economic exchange partners. Greif is correct in that the critical mechanisms in each of their approaches are different. But Landa notes in her 1981 article, in explaining why kinship/ethnic status is so effective, that “outsiders may substitute reputation for kinship/ethnic status. But acquisition of reputation is not costless.” (p. 356). Landa also notes the importance of the size and identifiable composition of the trading group, factors that are important in Greif’s approach. What Greif does, in effect, is advance a mechanism that reduces the cost of use of reputation. Moreover, his model absolutely depends on cultural factors that include the creation of boundaries around the Maghribi traders (else why wouldn’t traders from other ethnic groups seek to enter and gain the benefits of the reputational self-enforcement system?). The similarity to considerations of ethnic/cultural identifiers and social distance, arguments used by Landa, is inescapable. Thus, the natural flow of citation would have been for Greif to use the preexisting work by Landa as jump-off point. This observation does not diminish the significance or creativity of Greif’s work, which introduces important claims about the performance of reputation mechanisms, but it does accord Landa a special status in the literature. That is why we have labeled a fundamental problem in the operation of markets, the “Landa Problem,” the problem of identifying a trustworthy economic exchange partner. We believe that her work warrants such recognition.

The original argument that the selection of agents who hold genuine normative preferences, including such social norms as the fiduciary norm, to serve the principal’s interest is a means of economizing in social agency is due to Mitnick (1973, 1974, 1975a, 1975b). Mitnick’s work on agency was known to Landa but was cited by neither Landa nor Greif.


See also the chumrah, a prohibitive behavior undertaken in order to take care in compliance to a halakhic rule. It is based on:

When you build a new house, you shall make a parapet for your roof, so that you do not bring bloodguilt on your house if anyone should fall from it. (Deuteronomy 22:8; Jewish Publication Society, 1985, p. 308)

Thus it may be understood as d’oraita, with consequent prescriptive force.

A construct in halakha that is superficially similar in some respects is lifnim mishurat hadin, or “going beyond the letter of the law” (see, e.g., Pava, 1997, pp. 141–157, who discusses it in the context of corporate social responsibility). The examples that we discuss of “fences outside fences” in prescriptions of precautionary behavior under Jewish law deal with circumstances of taking exceptional, and even extreme care, mostly in order to avoid an unacceptable outcome. Under lifnim mishurat hadin, however, the law is recognized as potentially incomplete in specific provisions for all circumstances and/or in being unable to adequately handle societal variability and change. Where the law (including rabbinic statements) leaves gaps in instructing behavior, the construct of lifnim mishurat hadin instructs the use of ethics to guide behavior. Citing Lichtenstein (1978), Pava (1997, p. 154), argues that “The law, by carving out a domain in which individuals are expected to apply ethical criteria, recognizes and benefits from the advantages of contextual morality.” The resulting lacunae can still generate clear expectations to take special care via ethically guided behaviors of repair or extension. Later, we will point out the fiduciary expectations commonly placed on agents when the principal is highly dependent. But such expectations do not involve a “fence outside a fence” – they simply prescribe the highest level of care. In essence, dilemmas raised by incompleteness in the law’s applications are to be resolved at the margins via ethics. Pava (1997, p. 157), quotes rabbinical authority as asserting that “Jerusalem was destroyed only because they gave judgments therein in accordance with biblical law … and did not go beyond the letter of the law (lifnim mishurat hadin)” (cf. Pava, 2003 on “ethical improvisation,” pp. 113ff).

Note that lifnim mishurat hadin is seen as a means of completing normatively guided action in a way that simply adapts to and takes special care to be morally consistent with reasonably unforeseen contingencies; it is formally not the “fence outside fences” that we have been discussing. The only sense in which it may be considered as such a fence is its strong instruction to guard against not attaining the basic values of the legal system, which it does by making special consultations with accepted morality not encoded in the legal system – hence, in a sense, it tells participants in the legal system to go outside the normal “fences” of the legal system.


There is no explanation for this requirement. It is accepted and has strong prescriptive effect because it is biblical in origin. Remember that the law is primary and does not require rationalization, especially where the original prescription is taken to have divine provenance.


Note that the theory does not require a particular religion; indeed, the literature does not appear to show clear ties between many aspects of religion or religious practice and ethical behavior (see Ford & Richardson, 1994, p. 206). But the model used here, halakha, is an especially suggestive one because of its sensitivity to its fundamental, genuinely-held normative sources, its developed set of instructions for behaviors derived from that normative base, and its empirical display of the kinds of going-above-and-beyond/fences-outside-fences behaviors that demonstrate how defense of outcomes in service to the normative base is achieved.

The point is that the deep codes that instruct ethical behaviors must be anchored in basic, genuinely-held beliefs such as a religion supported in the community. A corporation’s code of conduct, for example, is normally not held as a genuine, deeply-held, terminal set of beliefs or values, even if employees agree with it and it has wide general support. A problem for many leaders is to create or install such deep, genuine beliefs.

It happens that strength of religious belief does show an effect; see McNichols and Zimmerer (1985), cited by Ford and Richardson (1994, p. 206).


Kaufman’s (1960) classic study of the US Forest Service has been criticized by Forrest Fleischman (2017) as ignoring the effects of external political stakeholders and changing policy coalitions and contexts. The Forest Service faces diverse policy demands and interest groups, and is far less homogeneous in composition as well (e.g., Anderson, Hodges, & Anderson, 2013; Fleischman, 2017; Tipple & Wellman, 1991). Kaufman (2017), in a brief response to Fleischman, notes that he did in fact recognize such effects, but focused in his book on the internal mechanisms. Moreover, Iverson (2017) calls attention to Kaufman’s (1994) “paradox of excellence,” which Kaufman labels the “down side of cohesion”:

[…] these influences tended to lock members – especially leaders who had come up through the system – into a prescribed set of ideas and behaviors formed in a particular context, and thus were likely to impede formation of new patterns for which changes in the context might one day call. (p. 3).

Kaufman (1994) refers to this effect as “mental blinders” (p. 3).

But the model of a deep code we have described is intended not as a set of operating procedures that ensure compliance, but as a normative guide to be administered with fiduciary discretion. Still, the possibility of change in even terminal values, or at least in the trade-offs the system implements among such values, must be recognized. We argue, however, that normative guides of the kinds we are describing tend to exhibit relative stability.


Aldrich, H., Cater, J., Jones, T., & McEvoy, D. (1983). From periphery to peripheral: The South Asian petite bourgeoisie in England. In I. H. Simpson & R. L. Simpson (Eds.), Research in the sociology of work (Vol. 2, pp. 132). Greenwich, CT: JAI Press.

Alvesson, M., & Berg, P. O. (1992). Corporate culture and organizational symbolism: An overview. Berlin, Germany: Walter de Gruyter.

Alvesson, M., & Lindkvist, L. (1993). Transaction costs, clans, and corporate culture. Journal of Management Studies, 30(3), 427452.

American Law Institute. (2006). Restatement of the Law Third, Agency (Deborah A. DeMott, Reporter). Philadelphia, PA: American Law Institute.

Anderson, S. E., Hodges, H. E., & Anderson, T. L. (2013). Technical management in an age of openness: The political, public, and environmental forest ranger. Journal of Policy Analysis and Management, 32(3), 554573.

Appel, G. (1975). A philosophy of mizvot: The religious-ethical concepts of Judaism, their roots in biblical law and the oral tradition. New York, NY: KTAV Publishing House.

Arrow, K. J. (1985). The economics of agency. In J. W. Pratt & R. J. Zeckhauser (Eds.), Principals and agents: The structure of business (pp. 3751). Boston, MA: Harvard Business School Press.

Ashforth, B. E., & Anand, V. (2003). The normalization of corruption in organizations. Research in Organizational Behavior, 25, 152.

Banfield, E. C. (1958). The moral basis of a backward society. New York, NY: Free Press of Glencoe.

Barley, S. R., Meyer, G.W, & Gash, D. C. (1988). Cultures of culture: Academics, practitioners and the pragmatics of normative control. Administrative Science Quarterly, 33(1), 2460.

Barzel, Y. (1992). Confiscation by the ruler: The rise and fall of Jewish lending in the Middle Ages. Journal of Law and Economics, 35(1), 113.

Berkovits, E. (1983). Not in heaven: The nature and function of halakha. New York, NY: KTAV Publishing House.

Bernstein, L. (1992). Opting out of the legal system: Extralegal contractual relations in the diamond industry. Journal of Legal Studies, 21(1), 115157.

Bialik, H. N., & Ravnitzky, Y. H. (1992 [1908]). The book of legends (Sefer ha-aggadah): Legends from the Talmud and Midrash. New York, NY: Schocken Books.

Blalock, H. M., Jr. (1967). Toward a theory of minority group relations. New York, NY: John Wiley.

Blau, B. M. (2017). Religiosity and the volatility of stock prices: A cross-country analysis. Journal of Business Ethics, 144, 609621.

Bleich, J. D. (1977). Contemporary halakhic problems. New York, NY: KTAV Publishing House.

Bleich, J. D. (1981). Judaism and healing: Halakhic perspectives. New York, NY: KTAV Publishing House.

Block, W. (1990). Jewish economics in the light of Maimonides. International Journal of Social Economics 17(3), 6068.

Bonacich, E. (1973). A theory of middleman minorities. American Sociological Review, 38(5), 583594.

Botticini, M., & Eckstein, Z. (2012). The chosen few: How education shaped Jewish history, 70–1492. Princeton, NJ: Princeton University Press.

Boyarin, D. (1993). Carnal Israel: Reading sex in Talmudic culture. Berkeley, CA: University of California Press.

Bradach, J. L., & Eccles, R. G. (1989). Price, authority, and trust: From ideal types to plural forms. Annual Review of Sociology, 15, 97118.

Cai, J., & Shi, G. (2017). Do religious norms influence corporate debt financing? Journal of Business Ethics, Published online 3 October 2017.

Calton, J. M., & Lad, L. J. (1995). Social contracting as a trust-building process of network governance. Business Ethics Quarterly, 5(2), 271295.

Carmell, A. (1988). Siyata lig’mara [Aiding Talmud study] (5th ed.). New York, NY: Feldheim.

Carr, J. L., & Landa, J. T. (1983). The economics of symbols, clan names, and religion. Journal of Legal Studies 12(1), 135156.

Chait, B. (1992). The 39 avoth melacha of Shabbath. Spring Valley, NY: Philipp Feldheim.

Chill, A. (1974). The mitzvot: The commandments and their rationale. New York, NY: Bloch Publishing Company.

Chintrakarn, P., Jiraporn, P., Tong, S., & Chatjuthamard, P. (2017). Exploring the effect of religious piety on corporate governance: Evidence from anti-takeover defenses and historical religious identification. Journal of Business Ethics, 141, 469476.

Clark, R. C. (1985). Agency costs versus fiduciary duties. In J. W. Pratt & R. Zeckhauser (Eds.), Principals and agents: The structure of business (pp. 5579). Boston, MA: Harvard Business School Press.

Cohen, A. (1991). An introduction to Jewish civil law. Spring Valley, NY: Feldheim.

Cohen, M. R. (1980). Jewish self-government in medieval Egypt: The origins of the office of the head of the Jews, CA. 1065–1126. Princeton, NJ: Princeton University Press.

Coleman, J. S. (1988). Social capital in the creation of human capital. American Journal of Sociology 94, Supplement, S95S120.

Conroy, S. J., & Emerson, T. N. (2004). Business ethics and religion: Religiosity as a predictor of ethical awareness among students. Journal of Business Ethics, 50, 383396.

Cook, K. S. (Ed.). (2001). Trust in society. New York, NY: Russell Sage Foundation.

Cook, K. S., Hardin, R., & Levi, M. (2005). Cooperation without trust? New York, NY: Russell Sage Foundation.

Cook, K.S., Levi, M., & Hardin, R. (Eds.) (2009). Whom can we trust? New York, NY: Russell Sage Foundation.

Cooter, R., & Freedman, B. J. (1991). The fiduciary relationship: Its economic character and legal consequences. New York University Law Review, 66, 10451075.

Cooter, R., & Landa, J. T. (1984). Personal versus impersonal trade: The size of trading groups and contract law. International Review of Law and Economics, 4(1), 1522.

DiTomaso, N. (1987). Symbolic media and social solidarity: The foundations of corporate culture. In 
S. B. Bacharach (Ed.), N. DiTomaso (Guest Ed.), Research in the sociology of organizations (Vol. 5, pp. 105134). Greenwich, CT: JAI Press.

Donaldson, T., & Dunfee, T. W. (1994). Toward a unified conception of business ethics: Integrative social contracts theory. Academy of Management Review, 19(2), 252284.

Donaldson, T., & Preston, L. E. (1995). The stakeholder theory of the corporation: Concepts, evidence, and implications. Academy of Management Review, 20(1), 6591.

Donin, H. H. (1972). To be a Jew: A guide to Jewish observance in contemporary life. New York, NY: Basic Books.

Dorff, E. N. (1997). Judaism, business, and privacy. Business Ethics Quarterly, 7(2), 3144.

Easterbrook, F. H., & Fischel, D. R. (1993). Contract and fiduciary duty. Journal of Law and Economics, 36(1(Pt.2)), 425446.

Edwards, J., & Ogilvie, S. (2012). Contract enforcement, institutions, and social capital: The Maghribi traders reappraised. Economic History Review, 65(2), 421444.

Eider, S. D. (1980). Halachos of Chanukah. Lakewood, NJ: Rabbi Shimon D. Eider.

Elon, M. (1994). Jewish law: History, sources, principles (B. Auerbach & M. J. Sykes, Trans.). Philadelphia, PA: Jewish Publication Society.

Fand, J. (1989). Business ethics and the Jews. The Public Interest, 97, 148152 (Review of Meir Tamari, 1987).

Feldman, D. M. (1978). The structure of Jewish law. In M. M. Kellner (Ed.), Contemporary Jewish ethics. (pp. 21–37) New York, NY: Sanhedrin Press.

Fendel, Rabbi. (1983). The Halacha and beyond. New York, NY: Hashkafah Publications.

Frankel, T. (2011). Fiduciary law. Oxford and New York: Oxford University Press.

Friedman, M. (1953). The methodology of positive economics. In M. Friedman, Essays in positive economics (pp. 343). Chicago, IL: University of Chicago Press.

Finn, P. D. (1977). Fiduciary obligations. Sydney, Australia: The Law Book Company.

Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention, and behavior. Reading, PA: Addison-Wesley.

Fiske, S. T., & Taylor, S. E. (1991). Social cognition (2d ed.). New York, NY: McGraw-Hill.

Fleischman, F. (2017). Questioning Kaufman: How cross-level political coalitions interact with organizational structure. Public Administration Review, 77(5), 743751.

Ford, R. C., & Richardson, W. D. (1994). Ethical decision making: A review of the empirical literature. Journal of Business Ethics, 13(3), 205221.

Frederick, W. C. (1995). Values, nature, and culture in the American corporation. New York, NY: Oxford University Press.

Frederickson, H. G. (Ed.). (1993). Ethics and public administration. Armonk, NY: M. E. Sharpe.

Ganzfried, S. (1987 [1870]). Kitzur Shulchan Aruch, Vol. 1 of 3 (A. Davis, Trans.). New York, NY: Metsudah.

Gatewood, R. D., & Carroll, A. B. (1991). Assessment of ethical performance of organization members: A conceptual framework. Academy of Management Review, 16(4), 667690.

Gershfield, E. M. (1988). Private property in Talmudic legal tradition. International Journal of Social Economics, 15(8), 4553.

Gil, M. (2003). The Jewish merchants in the light of eleventh-century Geniza documents. Journal of the Economic and Social History of the Orient, 46(3), 273319.

Gil, M. (2008). Shipping in the Mediterranean in the eleventh century A. D. as reflected in documents from the Cairo Geniza. Journal of Near Eastern Studies, 67(4), 247292.

Giorgi, S., Lockwood, C., & Glynn, M. A. (2015). The many faces of culture: Making sense of 30 years of research on culture in organization studies. The Academy of Management Annals, 9(1), 154.

Goitein, S. D. (1967). A Mediterranean society: The Jewish communities of the Arab world as portrayed in the documents of the Cairo Geniza (vols. i–vi). Berkeley, CA: University of California Press.

Goitein, S. D. (1974). Religion in everyday life as reflected in the documents of the Cairo Geniza. In 
S. D. Goitein (Ed.), Religion in a religious age (pp. 3–17). Cambridge, MA: Association for Jewish Studies, KTAV Publishing House.

Green, R. M. (1997). Guiding principles of Jewish business ethics. Business Ethics Quarterly, 7(2), 2130.

Greif, A. (1989). Reputation and coalitions in medieval trade: Evidence on the Maghribi traders. Journal of Economic History, 49(4), 857882.

Greif, A. (1993). Contract enforceability and economic institutions in early trade: The Maghribi traders’ coalition. American Economic Review, 83(3), 525548.

Greif, A. (1994). Cultural beliefs and the organization of society: A historical and theoretical reflection on collectivist and individualist societies. Journal of Political Economy, 102(5), 912950.

Greif, A. (2006a). Institutions and the path to the modern economy: Lessons from medieval trade. New York, NY: Cambridge University Press.

Greif, A. (2006b). The birth of impersonal exchange: The community responsibility system and impartial justice. Journal of Economic Perspectives, 20(2), 221236.

Greif, A. (2009). The curious commentary on the citation practices of Avner Greif. Public Choice, 141(3/4), 273275.

Greif, A. (2012). The Maghribi traders: A reappraisal? Economic History Review, 65(2), 445469.

Greve, H. R., Palmer, D., & Pozner, J.-E. (2010). Organizations gone wild: The causes, processes, and consequences of organizational misconduct. The Academy of Management Annals, 4(1), 53107.

Hakohen, I. M. (1990 [1931]). Sefer hamitzvot hakatzar [The concise book of mitzvoth]. New York, NY: Feldheim.

haLevi, P. (1988 [1523]) . The book of [mitzvah] education (Sefer haHinnuch): For the month of Tishrei (C. Wengrov, Trans.). New York, NY: Feldheim.

Hammer, R. (Trans.). (1986). Sifre: A Tannaitic commentary on the book of Deuteronomy. New Haven, CT: Yale University Press.

Hardin, R. (1992). The street-level epistemology of trust. Analyse & Kritik, 14, 152176.

Hardin, R. (1996). Trustworthiness. Ethics, 107(1), 2642.

Hardin, R. (1997). The economics of religious belief. Journal of Institutional and Theoretical Economics, 153(1), 259278.

Hardin, R. (2002). Trust and trustworthiness. New York, NY: Russell Sage Foundation.

Hardin, R. (Ed.). (2004). Distrust. New York, NY: Russell Sage Foundation.

Hardin, R. (2006). Trust. Malden, MA: Polity Press.

Hauptman, J. (1994). Judaism and a just economy. Tikkun, 9(1), 5556, 95.

Hecht, N. S., Jackson, B. S., Passamaneck, S. M., Piatelli, D., & Rabello, A. M. (Eds.). (1996). An introduction to the history and sources of Jewish law. New York, NY: Oxford University Press.

Hertz, Dr. J. H., C. H. (Ed.). (1960). The Pentateuch and Haftorahs (2nd ed.). London: Soncino Press.

Hosmer, L. T. (1995). Trust: The connecting link between organizational theory and philosophical ethics. Academy of Management Review, 20(2), 379403.

Husted, B. W. (1994). Transaction costs, norms, and social networks: A preliminary study of cooperation in industrial buyer–seller relationships in the United States and Mexico. Business & Society, 33(1), 3057.

Iverson, D. C. (2017). Kaufman’s paradox. Public Administration Review, 77(5), 753.

Jewish Publication Society. (1985). Tanakh: A new translation of the holy scriptures according to the traditional Hebrew text. Philadelphia, PA: Jewish Publication Society.

Jones, T. M. (1991). Ethical decision making by individuals in organizations: An issue-contingent model. Academy of Management Review, 16(2), 366395.

Jones, T. M. (1995). Instrumental stakeholder theory: A synthesis of ethics and economics. Academy of Management Review, 20(2), 404437.

Jung, L. (1978). The ethics of business. In M. M. Kellner (Ed.), Contemporary Jewish ethics (pp. 332343). New York, NY: Sanhedrin Press.

Kahneman, D. (2011). Thinking, fast and slow. New York, NY: Farrar, Straus, and Giroux.

Kaufman, H. (1960). The forest ranger: A study in administrative behavior. Baltimore, MD: Resources for the Future/Johns Hopkins University Press.

Kaufman, H. (1994). The paradox of excellence. Unpublished paper posted on U.S. Forest Service web site now defunct (Link from D. C. Iverson, 2017).

Kaufman, H. (2017). Let’s look again. Public Administration Review, 77(5), 752.

Kramer, R. M. (Ed.). (2006). Organizational trust. New York, NY: Oxford University Press.

Kramer, R. M., & Cook, K. S. (Eds.). (2004). Trust and distrust in organizations: Dilemmas and approaches. New York, NY: Russell Sage Foundation.

Kramer, R. M., & Lewicki, R. J. (2010). Repairing and enhancing trust: Approaches to reducing organizational trust deficits. The Academy of Management Annals, 4(1), 245277.

Kramer, R. M., & Tyler, T. R. (Eds.). (1996). Trust in organizations: Frontiers of theory and research. Thousand Oaks, CA: Sage Publications.

Kreps, D. M. (1990). Corporate culture and economic theory. In J. E. Alt & K. A. Shepsle (Eds.), Perspectives on positive political economy (pp. 90–143). Cambridge: Cambridge University Press.

Krupnik, E. (1981). The gateway to learning: A systematic introduction to the study of Talmud. New York, NY: Feldheim.

Kurland, N. B. (1995a). Ethical intentions and the theories of reasoned action and planned behavior. Journal of Applied Social Psychology, 25(4), 297313.

Kurland, N. B. (1995b). Ethics, incentives, and conflicts of interest: A practical solution. Journal of Business Ethics, 14, 465475.

Kurland, N. B. (1996a). Sales agents and clients: Ethics, incentives, and a modified theory of planned behavior. Human Relations, 49(1), 5174.

Kurland, N. B. (1996b). Trust, accountability, and sales agents’ dueling loyalties. Business Ethics Quarterly, 6(3), 289310.

Laitin, D. D. (1995). Marginality: A microperspective. Rationality and Society, 7(1), 3157.

Lamm, M. (1969). The Jewish way in death and mourning. New York, NY: Jonathan David Publishers.

Lamm, M. (1991). Becoming a Jew. Middle Village, NY: Jonathan David Publishers.

Landa, J. T. (1981). A theory of the ethnically homogeneous middleman group: An institutional alternative to contract law. Journal of Legal Studies, 10(2), 349362.

Landa, J. T. (1994). Trust, ethnicity, and identity. Ann Arbor, MI: University of Michigan Press.

Landa, J. T. (1996). Doing the economics of trust and informal institutions. In S. G. Medema & W. J. Samuels (Eds.), Foundations of research in economics: How do economists do economics? 
(pp. 142162). Cheltenham: Edward Elgar.

Landa, J. T. (2005). Bounded rationality of homo classificus: The law and bioeconomics of social norms as classification. Chicago-Kent Law Review, 80, 11671180.

Landa, J. T. (2016). Economic success of Chinese merchants in Southeast Asia: Identity, ethnic cooperation and conflict. Berlin, Germany: Springer-Verlag.

Lehmann, J. (1988) A guide to shatnez: Forbidden wool and linen combinations. Spring Valley, NY: Feldheim.

Leib, E. J., & Galoob, S. R. (2016). Fiduciary political theory: A critique. Yale Law Journal, 125, 

Levine, A. (1980). Free enterprise and Jewish law: Aspects of Jewish business ethics. New York, NY: KTAV Publishing House and Yeshiva University Press.

Levine, A. (1987). Economics and Jewish law: Halakhic perspectives. Hoboken, NJ and New York, NY: KTAV Publishing House and Yeshiva University Press.

Lichtenstein, A. (1978). Does Jewish tradition recognize an ethic independent of halakha? In M. M. Kellner (Ed.), Contemporary Jewish ethics (pp. 102123). New York, NY: Sanhedrin Press.

Liebermann, Y. (1981). The Coase theorem in Jewish law. Journal of Legal Studies, 10(2), 293303.

Light, I. (1984). Immigrant and ethnic enterprise in North America. Ethnic and Racial Studies, 7(2), 195216.

Light, I., & Bonacich, E. (1988). Immigrant entrepreneurs. Berkeley, CA: University of California Press.

Longenecker, J. G., McKinney, J. A., & Moore, C. W. (2004). Religious intensity, evangelical Christianity, and business ethics: An empirical study. Journal of Business Ethics, 55, 373386.

Margolis, H. (1997). Religion as paradigm. Journal of Institutional and Theoretical Economics, 153(1), 242252.

Mayer, R. C., Davis, J. H., & Schoorman, F. D. (1995). An integrative model of organizational trust. Academy of Management Review, 20(3), 709734.

Mazmanian, D. A., & Sabatier, P. A. (1983). Implementation and public policy. Glenview, IL: Scott, Foresman.

McNichols, C. W., & Zimmerer, T. W. (1985). Situational ethics: An empirical study of differentiators of student attitudes. Journal of Business Ethics, 4, 175180.

Merton, R. K. (1957). Social theory and social structure (Revised ed.). Glencoe, IL: The Free Press.

Meyer, J. W., & Rowan, B. (1977). Institutionalized organizations: Formal structure as myth and ceremony. American Journal of Sociology, 83(2), 340363.

Milgrom, P. R., North, D. C., & Weingast, B. R. (1990). The role of institutions in the revival of trade: The law merchant, private judges, and the champagne fairs. Economics and Politics, 2(1), 123.

Miller, G. P. (1993). Contracts of genesis. Journal of Legal Studies, 22, 1545.

Mitnick, B. M. (1973). Fiduciary rationality and public policy: The theory of agency and some consequences. Paper presented at the 1973 Annual Meeting of the American Political Science Association, New Orleans, LA. Proceedings of the APSA, University Microfilms. Retrieved from https://ssrn.com/abstract=1020859 or http://dx.doi.org/10.2139/ssrn.1020859

Mitnick, B. M. (1974). The theory of agency: The concept of fiduciary rationality and some consequences (Vol. Univ. Microfilms No. 74-22,881). Unpublished Ph.D. dissertation, Department of Political Science, University of Pennsylvania.

Mitnick, B. M. (1975a). The theory of agency: The fiduciary norm. Paper presented at the 1975 annual meeting of the American Sociological Association, San Francisco, CA.

Mitnick, B. M. (1975b). The theory of agency: The policing “paradox” and regulatory behavior. Public Choice, 24(Winter), 2742. Retrieved from https://ssrn.com/abstract=1021143

Mitnick, B. M. (1985). Agents of legitimacy: Pantheonic directorates and the management of organization environments. Paper presented at the Fifth Annual Sunbelt Social Networks Conference, Palm Beach Hilton Hotel, Palm Beach, FL, February 14–17.

Mitnick, B. M. (1995). Systematics and CSR: The concept of normative referencing. Business & Society, 34(1), 533.

Mitnick, B. M. (1996) The hazards of agency (October). Retrieved from https://ssrn.com/abstract=1417412 or http://dx.doi.org/10.2139/ssrn.1417412

Mitnick, B. M. (1999a). Credibility and the theory of testaments. Paper presented at the 1999 annual meeting of the Academy of Management, SIM Division, Chicago, IL, August 8–11. Earlier versions of some of the material were presented at the 1998 annual meeting of the American Political Science Association, Boston, MA, and the 1996 annual meeting of the International Association for Business and Society, Santa Fe, NM.

Mitnick, B. M. (1999b). Credible testaments, property, and the role of government. In N. Mercuro & W. J. Samuels (Eds.), The fundamental interrelationships between government and property. The Economics of Legal Relationships (Vol. 4, pp. 165176). Stamford, CT: JAI Press.

Mitnick, B. M. (2000). Commitment, revelation, and the testaments of belief: The metrics of measurement of corporate social performance. Business & Society, 39(4), 419465.

Mitnick, B. M. (2009). Assurance and reassurance: The role of the board. In R. W. Kolb & D. Schwartz (Eds.), Corporate boards: Managers of risk, sources of risk (pp. 294315). Hoboken, NJ: Wiley-Blackwell.

Mitnick, B. M. (2018a). Agency theory. In R. W. Kolb (Ed.), Sage encyclopedia of business ethics and society (2nd ed., pp. 63–71). Thousand Oaks, CA: Sage Publications.

Mitnick, B. M. (2018b). The Landa Problem. Unpublished working paper.

Mitnick, B. M. (2018c). The theory of testaments. Presented at the AMR conference on the Implications of Uncertainty for Management and Organization Theories Conference, University of Pittsburgh, May 17–19, 2018.

Mitnick, B. M., & Backoff, R. W. (1984). The incentive relation in implementation. In G. C. Edwards, III (Ed.), Public policy implementation (Vol. 3, pp. 59122) of S. Nagel (Ed.), Public policy studies: A multi-volume treatise. Greenwich, CT: JAI Press.

Neusner, J. (1984). Invitation to the Talmud: A teaching book. San Francisco, CA: Harper & Row.

Neusner, J. (1986). The oral Torah: The sacred books of Judaism: An introduction. San Francisco, CA: Harper and Row.

Noreen, E. (1988). The economics of ethics: A new perspective on agency theory. Accounting, Organizations and Society, 13(4), 359369.

North, D. C. (1990). Institutions, institutional change and economic performance. Cambridge: Cambridge University Press.

North, D. C., & Weingast, B. R. (1989). Constitutions and commitment: The evolution of institutions governing public choice in seventeenth-century England. Journal of Economic History, 49(4), 803832.

Ohrenstein, R. A., & Gordon, B. (1992). Economic analysis in Talmudic literature: Rabbinic thought in the light of modern economics. Leiden, The Netherlands: E. J. Brill.

Ouchi, W. G. (1980). Markets, bureaucracies, and clans. Administrative Science Quarterly, 25(1), 120142.

Padawer, E. E. (1990). Piskei Hilchos Shabbos (Vol. 4). New York, NY: CIS Publishers.

Palmer, D. (2012). Normal organizational wrongdoing: A critical analysis of theories of misconduct in and by organizations. New York, NY: Oxford University Press.

Passamaneck, S. M. (1983). A synopsis of the Jewish law of sale. In S. M. Passamaneck (Ed. & Trans.), The traditional Jewish law of sale: Shulchan Arukh Hoshen Mishpat chapters 189–240, 
(pp. 11–68). Cincinnati, OH: Hebrew Union College Press.

Pava, M. L. (1997). Business ethics: A Jewish perspective. New York, NY: KTAV Publishing House/Yeshiva University Press.

Pava, M. L. (2003). Leading with meaning: Using covenantal leadership to build a better organization. New York, NY: Palgrave Macmillan.

Pondy, L. R., Frost, P. J., Morgan, G., & Dandridge, T. C. (Eds.). (1983). Organizational symbolism. Greenwich, CT: JAI Press.

Posner, R. A. (1980). A theory of primitive society, with special reference to law. Journal of Law and Economics, 23(1), 153.

Powell, W. W., & DiMaggio, P. J. (Ed.). (1991). The new institutionalism in organizational analysis. Chicago, IL: University of Chicago Press.

Putnam, R. D., Leonardi, R., & Nanetti, R. Y. (1993). Making democracy work: Civic traditions in modern Italy. Princeton, NJ: Princeton University Press.

Rabinovich, L. (1993). Introduction to secured transactions in Halakha and common law. Tradition, 27(3), 3650.

Reiss, S. (2000). Who am I? The 16 basic desires that motivate our actions and define our personality. New York, NY: Tarcher/Putnam.

Reiss, S. (2004). Multifaceted nature of intrinsic motivation: The theory of sixteen basic desires. Review of General Psychology, 8(3), 179193.

Reiss, S. (2015). The 16 strivings for God. Macon, GA: Mercer University Press.

Reiss, S., & Havercamp, S. M. (1998). Toward a comprehensive assessment of fundamental motivation: Factor structure of the Reiss profiles. Psychological Assessment, 10(2), 97106.

Richman, B. D. (2004). Firms, courts, and reputation mechanisms: Towards a positive theory of private ordering. Columbia Law Review, 104(8), 23282368.

Richman, B. D. (2006). How community institutions create economic advantage: Jewish diamond merchants in New York. Law and Social Inquiry, 31(2), 383420.

Richman, B. D. (2017). An autopsy of cooperation: Diamond dealers and the limits of trust-based exchange. Journal of Legal Analysis, 9(2), 247283.

Ring, P. S., & Van de Ven, A. H. (1994). Developmental processes of cooperative interorganizational relationships. Academy of Management Review, 19(1), 90118.

Roberts, R. N. (1988). White House ethics: The history of the politics of conflict of interest regulation. New York, NY: Greenwood Press.

Roth, J. (1986). The halakhic process: A systemic analysis. New York, NY: The Jewish Theological Seminary of America.

Rowley, C. K. (2009). The curious citation practices of Avner Greif: Janet Landa comes to grief. Public Choice, 140, 275285.

Sarason, R. S. (Trans.). (1982). The Talmud of the land of Israel: A preliminary translation and explanation, Vol. 3: Demai. Chicago, IL: University of Chicago Press.

Schnall, D. J. (1993). Exploratory notes on employee productivity and accountability in classic Jewish sources. Journal of Business Ethics, 12, 485491.

Schwartz, S. H. (1994). Are there universal aspects in the structure and contents of human values? Journal of Social Issues, 50(4), 1945.

Schwartz, S. H. (2012). An overview of the Schwartz theory of basic values. Online Readings in Psychology and Culture, 2(1). Retrieved from https://doi.org/10.9707/2307-0919.1116.

Schwartz, S. H., & Boehnke, K. (2004). Evaluating the structure of human values with confirmatory factor analysis. Journal of Research in Personality, 38, 230255.

Scott, A. W. (1949). The fiduciary principle. California Law Review, 37(4), 539555.

Scott, W. R., & Meyer, J. W. and associates. (1994). Institutional environments and organizations: Structural complexity and individualism. Thousand Oaks, CA: Sage Publications.

Scott, W. R. (1995). Institutions and organizations. Thousand Oaks, CA: Sage Publications.

Shane, S. (1994). The effect of national culture on the choice between licensing and direct foreign investment. Strategic Management Journal, 15(8), 627642.

Siker, L. V. W., Donahue, J. A., & Green, R. M. (1991). Does your religion make a difference in your business ethics? The case of Consolidated Foods. Journal of Business Ethics, 10(11), 819832.

Simon, H. A. (1957). Models of man: Social and rational; mathematical essays on rational human behavior in a social setting. New York, NY: John Wiley & Sons.

Sombart, W. (1953 [1916]). Medieval and modern commercial enterprise. In F. C. Lane & J. C. Riemersma (Eds.), J. Riemersma (Trans.). Enterprise and secular change: Readings in economic history (pp. 25–40). Homewood, IL: Richard D. Irwin .

Steinsaltz, A. (1989). The Talmud: The Steinsaltz edition. A reference guide. New York, NY: Random House.

Stinchcombe, A. L. (1986). Stratification and organization: Selected papers. Cambridge: Cambridge University Press.

Tamari, M. (1987). “With all your possessions”: Jewish ethics and economic life. New York, NY: Free Press and Collier Macmillan.

Tamari, M. (1990). Ethical issues in bankruptcy: A Jewish perspective. Journal of Business Ethics, 9(10), 785789.

Tamari, M. (1991). In the marketplace: Jewish business ethics. Southfield, MI: Targum Press.

Tamari, M. (1994). Wealth, Torah, and morality. Tikkun, 9(1), 5758, 94.

Tamari, M. (1997). The challenge of wealth: “Jewish business ethics.” Business Ethics Quarterly, 7(2), 4556.

Telushkin, J. (1991). Jewish literacy: The most important things to know about the Jewish religion, its people, and its history. New York, NY: William Morrow and Company.

Tenbrunsel, A. E., & Smith-Crowe, K. (2008). Ethical decision making: Where we’ve been and where we’re going. The Academy of Management Annals, 2(1), 545607.

Thornton, P. H., Ocasio, W., & Lounsbury, M. (2012). The institutional logics perspective: A new approach to culture, structure, and process. Oxford: Oxford University Press.

Tipple, T. J., & Wellman, J. D. (1991). Herbert Kaufman’s Forest Ranger thirty years later: From simplicity and homogeneity to complexity and diversity. Public Administration Review, 51(5), 421428.

Tracey, P. (2012). Religion and organizations: A critical review of current trends and future directions. The Academy of Management Annals, 6(1), 87134.

Treviño, L. K., & Weaver, G. R. (2003). Managing ethics in business organizaitons. Stanford, CA: Stanford Business Books/Stanford University Press.

Twersky, I. (1967). The Shulhan ‘Aruk: Enduring code of Jewish law. Judaism: A Quarterly Journal, 16(2), 141–158.

Van Buren, H. J., III. (1999). Acting more generously than the law requires: The issue of employee layoffs in halakah. Journal of Business Ethics, 19, 335343.

Van Maanen, J., & Barley, S. R. (1984). Occupational communities: Culture and control in organizations. In B. R. Staw & L. L. Cummings (Eds.), Research in organizational behavior: An annual series of analytical essays and critical reviews (Vol. 6, pp. 287365). Greenwich, CT: JAI Press.

Vaughn, R. (1979). Conflict-of-interest regulation in the Federal Executive Board. Toronto, Canada: Lexington Books.

Wagschal, Rabbi. (1990). Torah guide for the businessman. Spring Valley, NY: Feldheim.

Waldinger, R., Aldrich, H., & Ward, R. (1990). Ethnic entrepreneurs. Beverly Hills, CA: Sage Publications.

Weber, M. (1958). The Protestant ethic and the spirit of capitalism (T. Parsons, Trans.). New York, NY: Scribner.

Werblowsky, R. J. Zwi. (1977). Joseph Karo: Lawyer and mystic. Philadelphia, PA: Jewish Publication Society.

White, R. (2003). Information, markets, and corruption: Transcontinental railroads in the Gilded Age. Journal of American History, 90(1), 1943.

Wood, D. J. (1991). Toward improving corporate social performance. Business Horizons, 34(4), 6673.

Zipperstein, E. (1983). Business ethics in Jewish law. New York, NY: KTAV Publishing House.


The authors wish to thank Nancy Kurland, Manuel London, Scott Eckers, and several anonymous reviewers for their helpful comments. We are especially grateful to one reviewer who wrote a paper-length review filled with extremely thoughtful, helpful observations, and not a few corrections that we greatly appreciated the opportunity to fix. This reviewer certainly built us a fence outside our fence and enabled us to write a much better paper. We would also like to thank the editors, Howard Harris and Michael Schwartz, for being willing to consider and publish a paper whose content is unusual, though the research question it addresses – the problem of identifying and maintaining fidelity in business relationships – is not unusual at all. Any remaining errors are most certainly ours. Scholarship support from the David Berg Family Fund partially supported one of the authors during this research, and is gratefully acknowledged.