Uncertainty, Religion, and Law

Elliott Ash

1. Introduction

The relationship between religion and law in human society has been the subject of untold inquiry. The Platonist philosophers famously believed, for example, that the enlightened philosopher-king should use religion to persuade his subjects to obey the law (Plato, 1901). In the Middle Ages, the religious and legal authorities were equally obsessed with the practice of witchcraft (Russell, 1972). Sisk et al. (2004) presented evidence that judges' religious beliefs have a significant effect on their decisions in religious freedom decisions. Fort (1987) considers law and religion to be ontologically equivalent, with the former constituting secular ethics and the latter constituting moral ethics (pp. 1-2 ). Religious concepts, according to Fort, give laws their authority and convince people to follow them (pg. 17). This argument is consistent with the Constitutional worship documented, for example, by Levinson (1989).

There have been few authors, however, that have directly attempted to conceptually pair law and religion as responses to uncertainty. Hofstede (2001) is one of the few that has made this relationship explicit:

Extreme uncertainty creates intolerable anxiety, and human society has developed ways to cope with the inherent uncertainty of living on the brink of an uncertain future. These ways belong to the domains of technology, law, and religion. I use these terms in their broad senses: Technology includes all human artifacts; law, all formal and informal rules that guide social behavior; religion, all revealed knowledge of the unknown. Technology has helped us to defend ourselves against the uncertainties caused by nature; law, to defend against uncertainties in the behavior of others; religion, to accept the uncertainties we cannot defend against (pg. 146).

These definitions are admittedly vague, but they point toward the topic of this essay: the role of religion and law in managing uncertainty. Specifically, I examine the proposition that religion and law serve as uncertainty-management mechanisms. The relationship between law, religion, and uncertainty is evident in many avenues, but perhaps the central mechanism entails the establishment of a logical narrative of which we have absolute knowledge. We might analogize to the standards for a good plot in soap operas and other narratives: “Loose ends” are seen as great weaknesses in drama, but in real life they are rampant. These aesthetic attitudes can also be explained by a desire for absolute knowledge of a social system.

Religion and law often serve as rationalizations for decisions, actions, or events that are of other origin. The causes of these decisions and events are not self-evident, sometimes mysterious, and other times corrupt or hypocritical. Law and religion not only legitimize such decisions, but also integrate them into a coherent cognitive framework. Telling ourselves that our life decisions result from religious epiphanies—just as telling ourselves our legal conclusions result from constitutional principles—gives us certitude in the rightness of our decisions. In these cases, religion and law serve to pacify intolerable emotions associated with uncertainty. For example, when the Supreme Court decided to give abortion constitutional protection, it was seen as a difficult and uncertain act; but once the opinion was written, it could be cited as law, given full respect as a declaration of our normative fiber. Similarly, an expectant mother’s decision whether to have an abortion is extraordinarily difficult and uncertain; but once the decision is made to keep the baby, the raw religious power of faith provides firm basis to defend that decision.

The problem with these modes of thinking is that neither the framers nor the prophets have much in the way of good advice for living our lives or running our governments. Both the framers and the prophets lived in a pre-scientific landscape where belief in supernatural forces remained strong, and they knew even less than we do today about how the world and how society works. Nonetheless, faith in religion and faith in the law are both comforting mindsets, and they both serve to reduce uncertainty about the world and about our place in it. This paper examines that relationship in some detail. I infer from the subsequent discussion that religion and law play important roles in managing uncertainty in human societies, and in fact some of the most important aspects of religion and law can be explained in this light. I conclude with an appeal for greater empirical research in this area, and with an example of how an experiment to explore these ideas could be designed.

2. Uncertainty

Marris (1996) asserts that “uncertainty is a fundamental condition of human life” (pg. 1). The world is dynamic and complex, while our perceptual and cognitive faculties are limited. Our language is ambiguous, while the behavior of other people is unpredictable. These information asymmetries lead to uncertainty in our decisions, relationships, and existence. It has long been recognized, moreover, that uncertainty is not just an information state but a discrete emotion, like happiness and sadness. Shakespeare illustrates the emotional valence of uncertainty as well as anyone in the following quote from Shakespeare’s Coriolanus, in which the emperor condemns the plebeians who have banished him:

And here remain with your uncertainty!
Let every feeble rumor shake your hearts!
Your enemies, with nodding of their plumes,
Fan you into despair!

Uncertainty does shake our hearts, and it can even lead to existential despair. We dedicate much of our lives to managing uncertainties, whether by gathering information through inquiry or alleviating associated emotional distress. This section describes the different roles that uncertainty plays in human life and the many strategies humans have developed to cope with it.

First, some discussion of terminology. The word “uncertainty” is itself subject to uncertainty because it admits a handful of meanings, connotations, and associated concepts. The sense of uncertainty used in the previous sentence is that of ambiguity, which, following most legal scholarship, I use in the narrow sense to describe language or wording that admits multiple reasonable meanings. While ambiguity is of extraordinary importance in a word-focused enterprise such as the law, it is not the primary focus of the present essay. Another sense of uncertainty that will not be the focus here is the concept of risk, often called “uncertainty” by economists, which refers to probabilistic decision-making. Under risk, decision-makers do not know the outcomes, but they do know the probabilities that each of the range of outcomes will occur. While most humans share a tendency to be uncomfortable with risk (e.g., Holt & Laury, 2002), economists have developed an effective toolkit for making optimal risky decisions.

Risk is therefore a trivial problem compared to uncertainty as I use it: to describe the decision environments in which one does not know which outcomes are probable, and might not even know which outcomes are possible. Decisions under this kind of uncertainty are a far more difficult problem, and formal decision theory is powerless in these cases (Ellsberg, 1961). When specificity is necessary to avoid ambiguity, I will refer to environmental uncertainty as the quality of environments that proscribe the collection and interpretation of sufficient decision-relevant information. Cognitive uncertainty, on the other hand, refers to the mental state of the decision-maker—specifically, his lack of sufficient decision-relevant information. Cognitive uncertainty might not be registered by the decision-maker, however; one can have a belief about the probability of an event occurring even if one lacks the information on which to sensibly determine that probability (Posner, 2009).

A special subset of environmental uncertainty, which I call social uncertainty, is unpredictability in social environments. Depending on the range of social norms and belief systems followed in a particular community, we may have more or less capacity to predict the behavior of other members of that community. This concept is important in the present essay because one of the functions of both law and religion is to motivate convergence to a particular set of social norms.

Cognitive and social uncertainty, as I use them, are memory-related constructs unrelated to affect. Recognition of uncertainty, however, does generally activate an emotional response. Emotional uncertainty, as used here, is the negative emotional response that most people feel in situations of recognized cognitive uncertainty (Greco & Roger, 2001). In addition to emotional uncertainty, social psychologists have a more nuanced concept called personal uncertainty (van den Bos, 2009), which refers to the cognitive and emotional responses associated with being uncertain about oneself, including one’s self-image, decisions, relationships, and worldview.

While most people are uncomfortable with uncertainty, the tendency to experience uncertainty as aversive varies across individuals and across populations. Different people and different cultures have different predispositions for (in)tolerance of uncertainty, and those differences have important social consequences. In the subsequent paragraphs, I will refer to this disposition or attitude toward uncertainty as tolerance (or intolerance) for uncertainty, or uncertainty aversion. I conceptualize uncertainty aversion as a scalar psychological characteristic that, while modifiable, has durable effects on personality and behavior. Experimental economists have observed significant aversion to decisions characterized by environmental uncertainty (see, e.g., De Mauro & Maffioletti, 2004). Moreover, evolutionary scientists have given us good reason to believe that uncertainty aversion was fitness-enhancing in the ancestral environment and therefore an adaptive decision strategy (Rode et al., 1999).

Differences in the attitudinal disposition toward uncertainty result in different behavioral responses to uncertainty. As will be discussed below, humans have found many creative ways to deal with uncertain environments. Hofstede (2001) uses the term “uncertainty avoidance” to refer to what I call uncertainty aversion, but I use uncertainty avoidance to refer to behaviors undertaken prospectively to avoid potentially uncertain situations and events. Uncertainty reduction, intuitively enough, refers to efforts to reduce existing uncertainty. Both of these strategies are subsumed under a broader term, uncertainty management, which presents the central theme of the essay.

With the important terms defined, I now turn to a deeper discussion of uncertainty. Most social psychologists believe that humans have a fundamental desire to feel certain about the world and their place in it (van den Bos, 2009). Under conditions of emotional uncertainty, we lack confidence in our decisions because we cannot predict their outcomes effectively. Emotional uncertainty is a stressful experience for most people, and it is associated with many social and personal factors, including trustworthiness of authorities and institutions, identity issues of inclusion and status, and personal well-being (Feygina & Tyler, 2009).

The feeling of emotional uncertainty is a stressful experience for most people. This emotional distress appears to be an intrinsic component of human psychology. Using controlled psychological experiments, Murray et al. (2002, 2005) found that personal uncertainty activates the same “emotional alarm system” that activates in response to learning that close personal relationships are in jeopardy. Eisenberger & Lieberman (2004) argue that this alarm system is evolutionarily adaptive because it increases awareness for social cues, the response to which will prevent further deterioration of one’s social status and perhaps allow reparation of destabilized relationships. These authors do not mention that the emotional distress imposed by the alarm system might itself have an evolutionary function: That is, the stress of social uncertainty prospectively incentivizes prosocial behavior that would reduce the probability of such socially destabilizing events.

A separate research paradigm has sought to quantify and measure uncertainty aversion. These efforts are reviewed in a lengthy chapter of Hofstede’s Culture’s Consequences (2001). Hofstede’s measurement of uncertainty aversion is based on survey questioning, which has elicited a strong correlation between 1) concern for rule compliance, 2) worry about job stability, and 3) frequency of stress and anxiety. Hofstede interprets the correlation in answers among these questions as evidence of a common cause: uncertainty aversion. Hofstede’s measure is admittedly vague and in need of more precise definition (e.g., Ailon, 2008), but some later work has substantiated its key implications (see Huang, 2008).

Statistical studies show that high uncertainty aversion is associated with higher age, a lower ambition for occupational promotion, a preference for specialist over management positions, a preference for large over small organizations, a higher value in company loyalty, a preference for investing in precious metal and gems rather than stocks, the avoidance of conflict, a preference for consensus decision-making, an aversion to working for foreigners, and a preference for neighbors of the same race (Hofstede, 2001). Low uncertainty aversion is associated with greater use of the internet, newspapers, and books (Hofstede, 2001: pg. 170). These findings indicate that individuals with relatively high aversion to uncertainty express greater discomfort with uncertain situations.

Uncertainty aversion is often discussed in relation to (and sometimes confused with) the authoritarian personality (Adorno, et al., 1964). The distinction, emphasized by Hofstede (2001), is that uncertainty aversion is associated with an emotional need for the authority of rules, while authoritarianism is associated with an emotional need for the authority of persons (see also Duckitt & Sibley, 2009). In the United States, uncertainty aversion is correlated with authoritarianism (Jost et al., 2007); that correlation is not observed in other cultures, however, indicating that the attitudes are separable (Hofstede, 2001).

Hofstede (2001) is most interested in the culture-wide effects of uncertainty aversion, which he defines as “the extent to which the members of a culture feel threatened by uncertain and unknown situations” (pg. 161). For example, uncertainty aversion is associated with cultures that value structure in organizations, institutions, and relationships, with the end being clarity and predictability. Countries with high uncertainty-aversion scores are more conscious of death. Test subjects from countries with high uncertainty aversion report longer duration in feelings of embarrassment in social situations. These measures indicate that individual differences in uncertainty aversion have effects on the broader society. The social forces leading to intercultural variation in uncertainty aversion are currently unknown, but there is some evidence that the orientation of the country’s legal system—whether rule-based or case-based—plays an important role (see Section 4 below).

The social effects of uncertainty aversion are especially pronounced in the socialization process. In countries with high uncertainty aversion, children are taught distinctions between clean/dirty and safe/dangerous at a younger age, and these categories are also more likely to be applied to ethnic groups (Hofstede, 2001: pg. 161-62). There are stronger rule and norm systems, and people are more likely to feel guilty and sinful. Children are taught that the world is hostile and that unknown situations are dangerous. In contrast, countries with low uncertainty aversion are more comfortable with unfamiliar situations, people, and ideas. Moral norms are more flexible, as are social norms for appearance and speech. The one moral norm that low-uncertainty-aversion cultures are more likely to follow is a norm against discrimination based on gender, race, and appearance.

In the schools of high-uncertainty-aversion societies, students and teachers prefer structured learning environments with “precise objectives, detailed assignments, and strict timetables” (pg. 162). Accuracy is the preeminent value. Instructors are expected to be experts in their field, and the use of academic jargon to signal erudition is common and acceptable. Students in uncertainty-averse countries “will not, as a rule, voice intellectual disagreement with their teachers. . . Intellectual disagreement in academic matters is interpreted as personal disloyalty” (pg. 162). In low-uncertainty-aversion countries, in contrast, schools are less structured, and a higher value is placed on originality. It is acceptable for a teacher not to know an answer, but it is unacceptable to use academic jargon. Intellectual disagreement is not taboo. An illustrative example of the educational differences between high-uncertainty- aversion and low-uncertainty-aversion cultures is the methodological approach in the social sciences:

In low-UAI [an index for uncertainty aversion] countries the scientific logic favors induction — that is, the development of general principles from empirical facts. In high-UAI countries deduction —that is, reasoning from general principles to specific situations—is more popular. . . . Up to the present day, empirical studies in the social sciences in such high-UAI countries as Germany and France are rare; in a society with a strong uncertainty [aversion] norm, scholars fear the risk of exposing their truths to experiments with unpredictable outcomes. On the other hand, in lower-UAI countries like the United States and Great Britain, empirical studies dominate. The orthodox methodological justification of such studies is that the progress of scientific knowledge passes through the falsification of hypotheses in testing them on reality; actually attempting to falsify one’s hypotheses requires that one have a large tolerance for uncertainty (pg. 178).

These broad differences in educational structure can be interpreted as reflecting different levels of tolerance for uncertainty.

Many economists are interested in the effect of cultural uncertainty aversion on a country’s productivity. Huang (2007) found that uncertainty aversion was associated with less trade with geographically distant countries, which reduced economic growth. Huang (2008) found that entrepreneurs exhibit less uncertainty aversion than the average person, and that this had positive economic consequences (see also Licht, 2007). Cultures with less uncertainty aversion are more successful in industries requiring innovation in the face of significant environmental and technological uncertainties. Others are seeking to include notions of uncertainty in models of economic behavior. Easley & O’Hara (2009), for example, show that adding uncertainty aversion to a standard investment model creates an efficiency-enhancing role for legal regulation that did not exist in the standard model.

High uncertainty aversion has darker consequences than slower economic growth, however. Hofstede (2001) believes that the predisposition of high-uncertainty-aversion individuals for religious ideology extends to all kinds of ideology: “Dogmatic, intolerant ideological positions are more likely in countries with higher [uncertainty aversion]” (pg. 177). Correspondingly, low-uncertainty-aversion countries should be more pragmatic. Hofstede points to Milgram’s (1974) simulated-electric-shock experiments as evidence of the harmful side of the rule-following behavior associated with uncertainty aversion. “The rules,” comments Hofstede, “protected against the uncertainty of independent judgment” (pg. 147). In a similar vein, Hogg (2005) observes that in cases of extreme personal uncertainty, people tend to choose dogmatic belief systems related to orthodoxy, hierarchy, and extremism.

Many of the effects of uncertainty aversion, for good or for ill, result from efforts to reduce or otherwise manage uncertainty. Because emotional uncertainty is an aversive experience, people are prospectively motivated to avoid situations, events, or activities that might result in feelings of uncertainty. When these avoidance measures fail, the negative feelings associated with personal uncertainty motivate behavior to reduce it (van den Bos, 2009). Posner (2009) writes that

People respond to uncertainty aversion in a variety of ways—trying to transform it into calculable risk when they can do so, as by improving analytical techniques or gathering additional information, and, when they cannot do so, substituting other methods of decision making for cost-benefit analysis, such as simple extrapolation from the past, decision according to rules of thumb, imitation of other people, minimizing sunk costs, flipping a coin, seeking guidance in prayer, adopting a “safety-first” policy, and building relations of trust (often within the family) in order to create a form of insurance that does not rely on the calculation of premiums.

As this quotation implies, conscious effort can result in the elimination of uncertainty, but oftentimes people must instead engage in psychological coping mechanisms that render uncertainty tolerable. For example, Hafer & Begue (2005) show that in uncertain or uncontrollable circumstances, people tend to construe new information in a way that minimizes threat to feelings of controllability. Wagenaar (1988) showed that gamblers use cognitive heuristics to avoid feelings of uncertainty about their gambling habit. According to Hofstede (2001), there is some evidence that gambling is more popular in countries with lower uncertainty aversion (pg. 179).

One way that people avoid uncertainty is through strict adherence to cultural norms and values—and more generally, worldviews (van den Bos et al., 2005). Social psychologists believe that people construct worldviews to systematize the world and one’s life under a comprehensible mental framework (van den Bos, 2009). Worldview construction makes the world seem more stable and predictable, serving as a preliminary defense against life’s vicissitudes. Similarly, Marris (1996) observes that people insulate themselves from the discomfort of emotional uncertainty by cultivating well-defined behavioral routines and rule systems. Social groups and community values help reduce uncertainty by providing self-esteem resources in the form of encouragement and comforting, and they provide knowledge and arguments for constructing and defending a worldview (van den Bos, 2009).

One way that people maintain worldviews is by filtering information. In social-psychology experiments, test subjects are more likely to believe information that supports their worldview and disbelieve information that destabilizes their worldview. This information-capture bias is reinforced by an emotional defense mechanism, in which worldview-supporting experiences are associated with positive emotions and worldview-destabilizing experiences are associated with negative emotions (van den Bos et al., 2005). Moreover, those individuals who have preexisting or experimentally primed uncertainties about their worldview will have more intense (both positive and negative) emotional reactions to both worldview-supporting and worldview-destabilizing experiences (van den Bos, 2001). For example, subjects primed with emotional uncertainty are more likely to react negatively to individuals who criticized their home country and their university (van den Bos et al, 2007; Dechesne et al., 2000). People who believe uncertainty to be “a relatively emotionally threatening experience” demonstrate greater aversion in interactions with homeless persons (van den Bos et al., 2007).

Part of Hofstede’s (2001) study of uncertainty aversion involved empirical work in organizations. Hofstede found that, just as societies use technology, law, and religion to manage uncertainty, so do organizations use technology, rules, and rituals. For example, in one cross-national study of MBA students, test subjects from high-uncertainty-aversion countries were far more likely to recommend that a hypothetical management problem be resolved through massive investment in computer technology (Schneider and De Meyer, 1991). According to Marris (1996), organizations try to manage uncertainty by constructing bureaucracies controlled by well-defined institutionalized incentives. Entrenched bureaucracies function well in static environments, but they cannot adapt to dynamic environments (Marris, 1996: pg. 91).

Rituals alleviate emotional uncertainty by allowing people to control the future, or at least to believe they are controlling the future. Rituals, in religions or otherwise, “are functional because they allow the members of that society to continue their lives in the face of otherwise intolerable uncertainty” (Hofstede, 2001: pg. 148). Rituals are not just religious, however. They include traditions such as birthday celebrations, state ceremonies, business meetings, and worker training programs. All of these rituals have symbolic, non-instrumental content, with sacred texts, sacred language, and sacralized acts. In businesses, rituals that serve to simulate conviction in the face of uncertainty include the ceaseless composition of memos, PowerPoint? presentations, computer modeling, and hiring of expert consultants—who bill themselves as, and are believed to be, “beyond uncertainty” (Hofstede, 2001: pg. 148).

The scholarship on uncertainty management suggests that many of our efforts to reduce emotional uncertainty interfere with our efforts to reduce cognitive uncertainty. In situations of environmental uncertainty, it is preferable to seek objective information rather than to select it based on a preexisting worldview. Flexibility is the best strategic framework for dealing with low-information environments. Feygina & Tyler (2009) summarize this point: “The uncertainty management approach suggests that the motivation to reduce uncertainty is best served by developing an accurate understanding of the world, rather than seeking reassurance of its legitimacy and security.” Similarly, Posner (2009) writes that “during times of economic downturns, the aversion to uncertainty of consumers and business people rises. . . . [G]overnment stimulus is necessary in such periods, like the current one, to reduce the fear of consumers and business to spend and invest.”

3. Religion

The role of religion in managing uncertainty has received a lot of attention in historical and scientific studies of religion. A 2009 Gallup poll, meanwhile, shows that religion remains an important part of people's lives all over the world. This section surveys some of the theoretical literature, as well as recent empirical studies. The existing scholarship is consistent with the notion that religion often serves as an uncertainty-management device.

Religion, writes Marris (1996), “remains the richest and most varied cultural resource in which to find both consolation for the burdens of uncertainty and strength to bear them” (pg. 115). Religion serves to reduce a variety of sources of uncertainty, including uncertainty about one's identity, one's group identification, the nature of existence, rules of morality, and the rules of living (Hogg et al., 2009). Recalling the earlier discussion, religion assists in the construction and defense of worldviews. According to Boyer (2001),

There are places like Melanesia where people perform an extraordinary number of rituals to protect themselves from witchcraft. Indeed, people think they live under a permanent threat from these invisible enemies. So we might think that in such societies magical rituals, prescriptions and precautions are essentially comforting devices, giving people some imaginary control over these processes (pp. 22-23).

But, Boyer continues, it is not comfort in particular that defines religion:

A religious world is often far more terrifying than a world without a supernatural presence. . . [C]onsider the widespread beliefs about witches, ghouls, ghosts and evil spirits allegedly responsible for illness and misfortune. . . [The Fang, a primitive society in Africa,] see evidence of this all the time, in crops that fail, cars that crash and people who die unexpectedly (pg. 23).

These anecdotes support the claim that comfort per se is not the defining aspect of religion. Instead, the common theme underlying these anecdotes is uncertainty reduction. While witches are terrifying and truly difficult to deal with, explaining unfortunate events as the machinations of witches reduces uncertainty about the causal mechanisms underlying those events. I propose that it is usually more important to have a worldview that offers some explanation of mysterious phenomena rather than to have a worldview that is intrinsically comforting.

Other scholars have discussed religion’s uncertainty-management function in terms of a response to environmental complexity. Burkert (1998) observes that religion provides “orientation within a meaningful cosmos for those who feel helpless vis-à-vis infinite complexity” (pg. 26). That function persists today: Just as religion served as a basis for comprehending the incomprehensible in the natural environment in pre-scientific societies, so does it now serve as a basis for comprehending the incomprehensible in the social environment of modern societies. We have precious little scientific understanding of the mechanisms underlying social processes, so religion can be used to validate past harms, existing power structures, and future decisions (Benavides, 2009).

But religion’s uncertainty-managing role might be even more fundamental than what has already been described. According to Guthrie (1993), what we call religion is anthropomorphosis—that is, the conceptualization of and interaction with non-human entities as if they were human. That people frequently and automatically engage in anthropomorphosis is not controversial; everyone develops some kind of relationship with non-humans, including animals, inanimate objects, and even supernatural spirits. More importantly, we perceive that these entities are sending us social messages. Guthrie argues that, in light of the environment's fundamental uncertainty, this biased perceptual framework is fitness-enhancing. Interpreting these messages as human is a good gamble because recognition of human agency would be the most important from a survival standpoint. Guthrie summarizes his view in this way:

The strategy is to interpret the world’s ambiguities first as the possibilities that matter most. Such possibilities usually include living things and especially humans. Although the strategy leads to mistakes, it also leads to vital discoveries that outweigh them. We see shadows in alleys as persons and hear sounds as signals because if these interpretations are right they are invaluable, and if not, they are relatively harmless.

For example, the most dangerous predators faced by ancestral hominids were other hominids, so assuming an approaching threat to be human contributed to survival. There is also plenty of evidence that, in primitive societies, success in social spheres is highly correlated with reproductive success. One must know when humans are around to avoid being ostracized for the violation of social norms, such as engaging in covert mating opportunities. Being overly biased toward detecting human beings facilitated behavior that protected our ancestors’ social status.

Ritual is one important way that religions manage uncertainty; Niklas Luhmann proposes that ritual itself is an uncertainty-avoidance strategy (cited in Beyer, 2009). The distinguishing quality of rituals, according to Luhmann, is that they make no distinction between information and utterance. Ritual is a form of communication, but it has no content. The utterance is the message; ritual is, essentially, communication without communication. Luhmann argues that this aspect of ritual makes it the only form of communicative act that can proceed with zero uncertainty. Because of the fundamental complexity of the world and the fundamental ambiguity of language, we cannot say anything substantial without uncertainty in its truth. From a social perspective, we cannot say anything substantial without fearing that others will express disagreement. We instinctually construe disagreement as a social challenge, which imposes its own emotional discomfort and undermines group cohesiveness. At the same time, interpersonal communication is the foundation of durable human relationships, so barriers to communication are socially harmful. Ritual elides these difficulties by allowing communication to proceed without uncertainty. Ritual thereby facilitates the cultivation of durable social relationships through interpersonal communication. Through rituals, religious communities can foster trust and commitment.

An important component of religious rituals in literate societies is the repetition and veneration of sacred texts. The repetition of words in sacred texts allows the powerless to reclaim some illusion of power (Marris, 1996: pg. 115). Reflecting the value of sacred texts in religious ritual, these texts remain important fixtures in modern life. A 1993 poll found that 92 percent of American households own a Bible. According to a 2000 poll, 59 percent of Americans profess to reading the Bible at least occasionally, with 37 percent reading it at least once a week. Consistent with Marris’s view that sacred texts appeal to those with less power in society, Women are more likely to have read the Bible within the last week, 42 percent compared to 32 percent.

Another interesting piece of data comes from a 1997 poll, which found that 83 percent of Americans prefer the King James Bible to the New International Bible. The King James, being written in archaic English, is more difficult to understand for modern readers. This makes sense in light of the previous discussion about the value of incomprehensibility in religious ritual. Indeed, the incomprehensibility of religious texts is a prerequisite for them to be the basis of religious ritual. If religious texts were comprehensible, then they would have a truth value, which would foster uncertainty and disagreement. Because liturgical verses are incomprehensible, they can be repeated, mindlessly, as communication for the sake of communication. This serves the uncertainty-related goals described in the previous paragraph. Because squeezing substance from these texts undermines their ritualistic value, it might also be hypothesized that religious people who profess belief in sacred texts will nonetheless avoid internalizing the information contained therein. Accordingly, only 14 percent of Americans claim to belong to Bible study groups. Moreover, only half of American adults can name one of the four Christian gospels, and only 37 percent could name all four. These findings are consistent with the view that the value of religious ritual consists in its communication without substance.

A more destructive manner in which religion mediates uncertainty is the waging of holy war (Marris, 1996: pg. 32), which has occurred innumerable times in history up to and including the present day. The incompatible belief systems cultivated by different religious ideologies are not subject to reconciliation by persuasion. At the same time, the presence of opposing belief systems generates uncertainty about one’s own religious worldview. Holy war serves to eliminate one of the opposing belief systems by force. Lest one sit in bewilderment at the idea that people would go to war over the incompatibility of religious beliefs, one should recall the intensely negative emotions that uncertain people feel when their worldview is destabilized (van den Bos, 2001).

Religion’s role in personal uncertainty reduction can have significant effects on the level of society. For example, games of chance (which require “mastery of the supernatural”) are more popular in cultures where gods are seen as benevolent and less popular in cultures where gods are seen as malevolent (Roberts et al., 1959). In a more serious vein, Marris (1996) argues that religion serves to generate consensus about the moral rightness of the social order (pg. 67). There is a tradeoff, Marris argues, between the social need for a legitimized system of control that reduces social uncertainty, and the individual need for autonomy, which alleviates personal uncertainty. At the level of the individual, religious devotion can alleviate the motivation to reduce personal uncertainty by replacing secular progress with spiritual transcendence as the foremost goal in life. These strategies reduce uncertainty but undermine efforts by disadvantaged classes to challenge the system’s established unfairness (pg. 116). People that view the current hierarchy as the will of God, moreover, will view reform efforts as threats to their system-justifying worldview.

A potentially positive effect of religious certitude is its motivational role in social-reform movements. Abolitionism, for example, as well as the civil rights movement, both had strong religious components. Religious conviction in the rightness of one's ideals enables social reform advocates to withstand “great uncertainty and relentless repression” (Marris, 1996: pg. 115). These examples suggest potential social benefits for religious uncertainty reduction.

The theory of religion as uncertainty-management strategy has plenty of anecdotal evidence in the history literature. Travis-Henikoff (2008) recounts religion’s gruesome role in relieving moral uncertainty about cannibalism during famines, an effect so powerful that many societies continued to engage in cannibalism even after the famine subsided. A more familiar example is that of the Puritans; Ericson (2005) emphasizes the role of their religious devoutness in maintaining a sense of certitude:

The hard logic of their creed required the Puritans always to doubt the evidence of their own senses but never to doubt the fundamental precepts of their religion. . . . If a persuasive argument should jar a Puritan's certitude or a clever line of reasoning confuse him, he had every right to suspect that some devilish mischief was afoot (pp. 51-52).

The Salem witch trials, especially, were a destructive example of uncertainty management.

Another possible example is the uncertainty-reducing role of religion among blacks in the slave era. Presumably, worship services strengthened community bonds and provided emotional support for enduring the hardships and uncertainties of slave life. A contemporaneous discourse among theologians, centered at Princeton, was aimed at theological justification of slavery, using legalistic analysis of biblical texts to allay moral uncertainties about slave ownership. The propriety of slavery was a foundational pillar of the Protestant worldview, and theological arguments were wielded in its defense. Slaves were taught this same system-justifying doctrine—that it was God’s will that they be enslaved—and thereby the moral uncertainties of both master and slave were alleviated.

Empirical evidence from the behavioral sciences is also consistent with the view that religion provides an outlet for avoidance of uncertainty. Willer (2009) ran an experiment in which he asked test subjects to write short essays, half writing about watching television and the other half writing about death. After the essay, subjects answered a series of survey questions about their religious beliefs. Subjects in the death treatment reported significantly higher belief in the afterlife and significantly higher belief in God. Willer interprets this as evidence of motivated reasoning in religiosity, and specifically motivation by fear of death. We might also interpret the result as evidence of emotional uncertainty associated with death, however. Supporting this latter interpretation is the study reported in vas den Bos et al. (2006), which found that low tolerance for uncertainty was associated with angrier responses to antireligious arguments by religious test subjects. Being primed for uncertainty and being more religious were associated with more intensely emotional responses.

Also supportive is the meta-analysis of 34 studies by Hackney & Sanders (2003) finding that, generally speaking, religiosity has a small but significant salutary relationship with mental health. Importantly, this correlation is sensitive to the meaning we give to “religiosity.” If by religiosity we mean only involvement in religious organizations, the salutary effect is negligible. If by religiosity we mean a measure of ideological attachment to a religious belief system, we find a significantly stronger association with mental health. Emotional devotion to a supernatural entity, meanwhile, has the strongest positive correlation with mental health. The authors comment that these findings are consistent with a theory of religion as uncertainty-management device—specifically, that religion “shields the individual from existential anxiety and enables the individual to achieve self-esteem and (presumably) life satisfaction through the knowledge that one is a valuable member of a meaningful universe” (pg. 51). Merely engaging in religious ritual (involvement in church) is not sufficient to reduce uncertainty—that requires sincere internalization of the emotional and/or cognitive commitments associated with religion. The fact that emotional commitment is stronger than cognitive commitment in improving mental health is consistent with van den Bos’s (2009) view that emotional uncertainty plays a more important role than cognitive uncertainty in worldview defense and system justification.

The mental-health studies prove nothing about causation, however: Mental health might facilitate religious devotion, rather than vice versa. Correspondingly, people with low mental health might be more likely to seek involvement in religious organizations, thereby bringing the mental-health average down for churchgoers relative to non-churchgoers (even if their mental health would be even worse without religious involvement). This last point is important in the context of understanding religion as an uncertainty-reduction strategy, rather than as an uncertainty-aversion indicator. Religion provides many other benefits besides uncertainty reduction, so we can’t determine with confidence whether religious people are doing it for uncertainty-reduction or other reasons. Moreover, uncertainty-reduction is not the conscious goal of most religious adherents: It’s not the case that emotionally uncertain people actively seek religion; people attend church for other reasons, but it makes emotionally uncertain people feel better about themselves. So they keep attending. These individuals don’t recognize this positive feeling as resulting from reduction in uncertainty, so we can’t measure the phenomenon by survey questioning. We must infer it from measurements of religiosity and emotional uncertainty, which would require, at minimum, a longitudinal study of the effects of religious activities on emotional uncertainty across extended time periods (see also Francis, 2001; Kroll et al., 2007).

Another line of scholarship, as yet unmentioned, makes a distinction between denominations in relation to uncertainty. Hofstede (2001) predicts that in countries with high uncertainty aversion, religions would be more successful if they stressed absolute certainties and manifested intolerance toward other religions (pg. 176). Catholicism—with an infallible Pope, detailed descriptions of heaven, and Godly incorporation in visible physical objects—should be more popular for people averse to uncertainty. Protestantism, with predestination, a rejection of otherworldly comfort strategies, and an invisible God, demands a greater tolerance of uncertainty. Posner (2003) identifies this distinction when he writes that “Protestants are encouraged to read and ponder the Bible for themselves… and Catholics are told to defer to their specialized ‘representatives.’” Consistent with these comments, Catholicism is more popular in countries with high uncertainty aversion, and Protestantism is more popular in countries with low uncertainty aversion (Guiso et al., 2003).

Posner (2003) proposes that Protestantism’s “encouraging people to think for themselves may well have promoted economic progress at the same time that resources were being shifted from the religious to the commercial sector” (pg. 173). This comment is an allusion to Weber's classic work, The Protestant Ethic and the Spirit of Capitalism (1930). Weber’s argument had many working parts, but what is important for the present essay is whether certain forms of religious belief—like colonial Protestantism—spurred economic growth by alleviating uncertainty aversion. As mentioned above, low uncertainty-aversion is the defining quality of the entrepreneur (Licht, 2007). Huang’s (2008) aforementioned study, which related low uncertainty aversion with success in innovative industries, also found that the country’s dominant religion (Protestant or Catholic) explained a significant chunk of the variance in economic productivity. This evidence suggests that, apart from alleviating feelings of emotional uncertainty, religious beliefs might actually be able to mitigate aversion to uncertainty.

But using the Calvinists as an example of this process requires oversimplification of historical events. Calvinists had a peculiar stance toward uncertainty and uncertainty management. The doctrine of predestination was adamant: One's good works in life had no bearing on one's salvation. Unlike most other religions, Calvinism explicitly precluded its adherents from reducing uncertainty about the afterlife by engaging in a quid pro quo with God. So notwithstanding the points noted in the previous paragraph, it is nonetheless true that belief in Calvinism was defined by uncertainty. The Protestant ethic might have some relationship to success in markets, but the role of uncertainty aversion in that process is not understood.

4. Law

A topic that has received much theoretical investigation but less empirical investigation is the relationship of law to uncertainty management. This section surveys some of the existing literature on that relationship, which analyzes three important ways that law relates to uncertainty. First, as indicated by Hofstede (2001), law serves to reduce uncertainty in social environments by constraining the behavior of other agents. Second, the effectiveness of law in constraining behavior can be undermined by widespread uncertainty in its legitimacy. This legitimacy rests, in part, on the ability of courts to appear impartial. But judges themselves, as emphasized by Posner (2008), face “radical uncertainty” when making decisions, both in terms of cognitive uncertainty about factual evidence and emotional uncertainty about the prudence of their substantive judgments. The efforts by judges and other legal actors to manage these uncertainties explains much of the otherwise puzzling qualities of the modern legal system.

As noted, law serves the purpose of reducing uncertainty about the behaviors of other members of society by motivating convergence to a collection of officiated social norms. North (2006), for example, argues that “institutional innovation intended to reduce uncertainty” is the defining quality of sociopolitical progress in human societies (pg. 2). An example of the legal system reducing uncertainty is found in the requirement that governments compensate individuals for the appropriation of property, since “such takings are uncertain events” (Posner, 2009). Another way the law reduces uncertainty is by establishing and enforcing social norms (Licht, 2007), since social norms play a critical role in reducing uncertainty about one’s own decisions (Feldman & Maccoun, 2005), as well as reducing social uncertainty. In previous eras, societies lacked a peaceful way to reconcile disagreements between different systems of social norms and ideological beliefs; the unfortunate solution was to wage holy war (Marris, 1996: pg. 32). Marris argues that the definitive benefit of the law—and in particular, private litigation—is its replacing holy war as the mediator of incompatible ideologies. The law’s peaceful dispute-resolution process paves the way for convergence to one set of social norms, increasing the predictability of social behavior and reducing uncertainty about the social environment. This salutary effect is observed, for example, in the gradual (and substantial) reduction in violence observed over all of human history (Pinker, 2007).

If a plaintiff feels wronged, and a defendant feels otherwise, there is no amount of moral theorizing that can cause them to reconcile. At the same time, legal interventions are often the sort of interventions that would be strictly forbidden, such as the deprivation of property, restriction of agency, and infliction of physical injury. The members of society will submit to the law’s mediation of such disagreements only if the legal system is seen as legitimate (Tyler, 2006). Chase (2005) summarizes this universal social problem:

Americans. . . therefore confront the dilemma faced by all societies in which a neutral is empowered to judge the disputes of others: how to legitimate that person’s decisions so that compliance is likely, so that the constituting authority is not (by unacceptable decisions) delegitimized, and so that the decision maker does not become the object of revenge by the losing party (pg. 32).

A concern for the fairness of legal procedure, moreover, is an almost universal pillar of peoples’ moral worldview (van den Bos, 2009). Empirical support of this view is provided in van den Bos (2001), finding that perceptions of procedural fairness activate positive emotions and perceptions of procedural unfairness activate negative emotions. Moreover, inducing emotional uncertainty in test subjects amplifies the intensity of both emotional responses. Flashing warning lights—presumably associated with the uncertainties about emergencies and crime—have the same effect (van den Bos et al., 2008). According to Feygina & Tyler (2009), “research indicates that procedural experiences provide people with information relevant to the trustworthiness and reasonableness of the system, which reduces uncertainty and insecurity about social environments” (pg. 352). Uncertainties about the law’s legitimacy can therefore undermine social stability.

At the same time, judges make decisions under substantial cognitive and emotional uncertainty (Posner, 2008). They must make confident factual findings; they must order momentous transfers of money; and they must condemn other human beings to lives in captivity—or even death. They must render these decisions under severe time pressure, with insufficient evidence, and with ambiguous legal source materials. They lack the expertise and apparatus needed to predict the consequences of their decisions. And yet, in spite all of these grave problems—these profound sources of uncertainty—judges remain confident in their rulings. Indeed, it is a prerequisite for the perceived legitimacy of a judgment that the judge have complete certitude in its rightness (Chase, 2005). A judge’s uncertainty would pose a threat to social stability. The uncertainty felt by the deciding judge, as well as society’s uncertainty in the legitimacy of the legal system, are the focus of the rest of this essay.

Jerome Frank’s Law and the Modern Mind (1935) was the first attempt to comprehensively examine the law’s response to the problem of uncertainty. Frank’s main goal was to critique the reigning approach in the academy and judiciary, what I call “legalism” but what Frank called “legal fundamentalism.” Legal fundamentalists believe that law must meet exacting standards of purity, the most important being temporal persistence and logical coherence. As an example of this pathology, Frank points to the doctrine of punitive damages in tort, which the Michigan Supreme Court had allowed for a time, then banned for a few years, and then allowed again (pg. 49). The legalists argued that these judicial decisions did not constitute law because law is eternal—it cannot change back and forth every few years. Moreover, according to Frank, the majority of 1930s American lawyers would have agreed with the legalists, believing that “the principles of law are absolute, eternal, and of universal validity” (pg. 54). This phenomenon, which Christenson (1985) called the lawyer’s “quest for certainty,” relates to the earlier discussion of worldviews. As noted, people construct worldviews as a preliminary defense against emotional uncertainty (van den Bos, 2009). Cultivation by judges and law professors of legalist beliefs can be seen as a similar process. The widely-held legalist worldview served to reduce uncertainty about the fundamental problems with existing legal concepts and practices. Lawyers shared a subliminal uncertainty with the law, and the legalist concept of a pure, perfect law served to allay that uncertainty.

Frank (1935) had this to say in response to the legalist mindset:

From the point of view of the ordinary human being, that kind of Absolutist law is meaningless. For the ordinary human being is interested, legitimately, in what happens in court… Principles, rules, conceptions, standards, and the like, may be law for lawyers, regardless of whether such law ever comes into contact with the affairs of life. But not for the rest of humanity. To mere humans, law means what the courts have decided and will decide, and not vague, “pure” generalizations.

For many, Frank’s pragmatic view of the law was (and is) seen as contaminating of its logical perfection. This defensive attitude is to be expected, given that the legalist paradigm was founded on uncertainty aversion. Threats to the legalist worldview were fended off by emotional defense mechanisms.

The intensity of legal uncertainty aversion was so extraordinary in Frank’s (1935) time that some scholars even argued that law should transcend the written word. These scholars honestly believed that “developing a true science of law. . . has been retarded because lawyers have not risen above the word level” (pg. 57). Other academics advocated that courts stop writing opinions, as the writing of opinions forces judges to move from the general to the particular, where pure abstractions are contaminated by facts (pg. 156). These legal philosophers wanted to abandon written law, believing that lawyers could reason on a higher level of abstraction, with semantic and factual analysis being replaced by an interaction of pure Platonic forms (pg. 58).

This admittedly ludicrous approach to law soon died out, but the legalists’ belief in the purity of written law persisted. According to Frank (1935), it is language (and in particular, its emotiveness) that makes legalism possible (pg. 62). He summarizes their method thusly (pp. 61-62). First, state a conclusion. Then, defend it by reference to a fictional premise chosen such that the desired conclusion appears to follow inevitably from the premise. Most importantly, make sure that the premise is of sufficient generality and indeterminacy as to conceal its arbitrariness. The emotive generalities upon which the argument rests, what Frank pans as “solving words,” serve to preclude uncertainty in the validity of the argument. These words, Frank comments, “create a hallucinatory satisfaction. They become substitutes for action” (pg. 62). By settling for emotive verbiage, the legalist sacrifices what real opportunity he had to shape the law to his case. Reflecting on the legalist approach, Frank writes that

We can see now why there is so much talk about the certainty of law. That talk is not descriptive of the facts; it is made up of magical phrases. Law being so largely lacking in the certainty which is desired, resort is made to those twangings of the vocal chords which will yield compensatory satisfaction (pg. 63).

Recitation of these “magical phrases”—as with the recitation of liturgical passages—served to reduce uncertainty in problematic legal concepts.

The philosophical brand of legalism that Frank (1935) so effectively criticizes was most popular in academia. The strategies of uncertainty management used by judges are both more mundane and more subtle. For example, because their caseloads are so demanding, judges are motivated to use legalist shortcuts rather than engage in a more uncertain consequentialist analysis. This process might be compared to the use of cognitive heuristics among gamblers to avoid feelings of uncertainty about their gambling habit (Wagenaar, 1988).

One ubiquitous way that judges reduce popular uncertainty in the legitimacy of their rulings is by displacing responsibility for their rulings onto another (often nonexistent) entity. For example, courts often shift responsibility to the legislative branch, even if the statute in question is ambiguous (Posner, 2008). Chase (2005) offers the following explanation for such behavior:

Particularly in societies in which revenging wrong is expected, the losing party, convinced of the rightness of his cause, will see the judge as an ally of the enemy, and be inclined to seek retribution. If the judge’s decision is seen to be compelled by some neutral force, this problem is attenuated (pg. 33).

Displacing responsibility for a decision onto external forces therefore functions to reduce the society’s uncertainty in the legitimacy of his decision. Other examples of this displacement strategy are observed when judges say that “the Constitution permits” or “the Constitution does not permit” or “the law permits” or “the law does not permit,” all of which are disingenuous statements given that they effectively mean “I permit” or “I do not permit.” Judges often refer to past decisions of the court through such constructions as “we held” or “we decided,” which are suggestive of a unitary body with unitary agency, notwithstanding that courts are composed of many judges with frequent turnover (Chase, 2005). This language is even used when referring to cases in which one or more judges filed a dissent. “Textualism”—law’s equivalent to theology’s literalism—is also popular among judges because it displaces responsibility from the judge onto an inanimate object: a written document. Aside from the responsibility-displacing effect, the textualist mindset also serves to reduce a judge’s personal uncertainty about their rulings. Repeating legal texts worshipfully—as if the words themselves have intrinsic value—reduces uncertainty by relieving judges of the obligation to think carefully about the real-world effects of their decisions (see also Phillipse, 2007).

Perhaps the subtlest example of displacing responsibility for a ruling onto a “neutral force” is the use of syllogistic reasoning in judicial opinions (Frank, 1935). Consider the following stylized example. A young man confesses to his priest that he is about to murder his wife. The priest advises against the murder, but otherwise does nothing, and the wife is dead the next morning. The wife’s parents sue the priest for wrongful death, but the priest claims that the sanctity of the confessional booth relieved him of any duty to warn. Suppose that the court decides to rule in favor of the parents, justifying its ruling based on a fundamental principle: the duty to warn. The published opinion then employs impeccable formal logic. The major premise is the fundamental principle, that all citizens have a duty to warn potential victims if they learn that serious harm is imminent. The minor premise is that the priest learned that serious harm to the wife was imminent. Applying the major premise to the minor premise, the court is compelled to conclude that the priest breached his duty and must pay damages.

The problem with the opinion is the “fundamental principle,” which the court could justify as an inference from previous cases or as the command of fundamental notions of justice (Frank, 1935). The problem is that the same court could have found an equally (il)legitimate “fundamental principle” that would have turned the case in the defendant’s favor. For example, the First Amendment’s Free Exercise Clause presents a tempting opportunity for justifying syllogistic reasoning with a “fundamental principle.” In this case, the major premise becomes the principle that the confidentiality of the confessional booth is sanctified by the First Amendment, while the minor premise becomes the fact that the murder tip was relayed in the confessional booth. The inevitable conclusion emerges: No breach of duty occurred, case dismissed. Both conclusions seem inevitable given the premises—and therefore serve to reduce the judge’s and the parties’ uncertainty about the legitimacy of the outcome—but the premises are fundamentally illegitimate. Frank (1935) writes:

The weakness of the use of formal logic is now exposed. The court can decide one way or the other and in either case can make its reasoning appear equally flawless. Formal logic is what its name indicates; it deals with form and not with substance. The syllogism will not supply either the major premise or the minor premise. . . These are difficult tasks, full of hazards and uncertainties, but the hazards and uncertainties are ordinarily concealed by the glib use of formal logic (pg. 66).

The use of syllogistic reasoning in judicial opinions therefore serves to conceal judges’ uncertainty in the rightness of their judgments by displacing responsibility for that judgment onto a logical fiction. This trick allows judges to preserve the illusion of certitude and retain some modicum of dispute-resolving legitimacy.

The absurdity of such arguments can be deterred, in some cases, if a smart and good-faith judge is also sitting on the panel, ready to dissent. The filing of dissents, however, destabilizes the legalist worldview by demonstrating that law is subject to reasonable disagreement. Many legalists in Frank’s (1935) time therefore wanted dissenting opinions to be prohibited (pg. 156). A pragmatic argument against dissents is that they increase uncertainty in the legitimacy of the legal-reasoning tools that judges use to make decisions. This probably accounts for what Posner (2008) calls “dissent aversion,” the widely followed social norm among judges requiring them not to dissent except in cases of intense disagreement. A potentially beneficial effect of dissents, however, is the signaling of uncertainty in a judicial body about the rightness of a particular ruling. This signaling can have socially beneficial consequences over time by weakening the precedential effect of decisions in which the courts are not confident. That way, those decisions in which the judges are confident—which, lacking a dissent, are more likely to be socially efficient—will have a comparatively stronger influence on later rulings. Through this process, dissents facilitate the law’s long-term goal of motivating convergence to welfare-maximizing equilibria.

Another legal-uncertainty-reduction strategy is codification of the law into statutes. Germans were the first to attempt systematic codification of the law; code advocates were consciously interested in reducing uncertainty (Frank, 1935). That Germany was the first country to undertake this task is consistent with the hypothesis offered by Hofstede (2001) that high-uncertainty-aversion countries should feel a more intense need for legislation than countries with lower uncertainty aversion. Germany has high uncertainty aversion and a labyrinthine statutory regime, while Great Britain has low uncertainty aversion and minimal codification. In Germany’s first attempt at codification, spearheading monarch Frederick the Great envisioned that “all contingencies should be provided for with such careful minuteness that no possible doubt could arise at any future time” (pg. 186). In particular, judicial discretion was explicitly disavowed.

The plan failed miserably (Frank, 1935). The complexities of German society chafed against the idealizations in the code. Without some interpretive discretion given to courts, the law was indeterminate in case after case, and decisive rulings were elusive. The Germans learned the hard way that a naïve belief in the completeness of codified law—and the corollary normative judgment that courts should apply only the text of codes—leads to the opposite effect sought by the legalist. Rather than reducing uncertainty, a code-oriented legal system amplifies it. The Germans under Frederick believed that the code, once written, would be eternal and would require only interpretation. They found—as Americans have found with the Constitution—that if one looks to an unchanging document as authoritative in changing times, a gradually widening gap will emerge between the law as written in the code and the law as practiced in the courts. The legal schizophrenia that results increases the uncertainty of judges in the rightness of their decisions as well as the uncertainty of prospective litigants ex ante.

Frank (1935) advocated that the judicial realities making legal certainty impossible should be given full air and acceptance. He thought that this good-faith acceptance of reality would facilitate effective reform:

If we relinquish the assumption that law can be made mathematically certain, if we honestly recognize the judicial process as involving unceasing adjustment and individualization, we may be able to reduce the uncertainty which characterizes much of our present judicial output to the extent that such uncertainty is undesirable (pg. 157).

Frank’s argument is in the end a pragmatic one: Let’s stop managing uncertainty by defending a false worldview; instead, let’s reform our legal institutions to maximize the beneficial effects of legal transparency and predictability.

The previous paragraphs discussed those legal uncertainty-management strategies involving how to structure rulings to avoid uncertainty among judges and the society. There is a whole other collection of uncertainty-management strategies that have no substantial implications for the content of rulings; we call them rituals. As demonstrated by Chase (2005), ritual is a defining aspect of our legal system (pp. 119- 122). In criminal cases, for example, judges often order two or more sentences to be served concurrently; the extra sentences are purely symbolic maneuvers. Allen (2007) argues that ritual in law serves to facilitate belief in the impartiality of the judge. Like magic, the law enables judges to create truth, just by proclaiming it while invoking the requisite legal incantations. It is law without law: Just as religious ritual helps build trust in the church, so does legal ritual build trust in the courts. Moreover, both religious and legal rituals contribute to trust in other members of the community by cultivating an expectation that they will comply with religious and legal norms.

According to Chase (2005), the rituals of any particular legal system are drawn from symbols that the broader culture associates with fairness, truth, and respect. These symbolic associations subconsciously contribute to the public’s support for the legal system. For example, in the United States, black robes are strictly associated with somber occasions, such as religious gatherings and graduation ceremonies. By wearing black robes, judges capture that symbolic sobriety; observers unconsciously construct a powerful preconception of the judge’s wisdom and competence. Classical architecture evokes reverence among most Americans, an association of which most American courthouses take advantage. Archaic Latin phrases retain an air of veneration, not to mention mystery; judges are the only non-classicists that still use them on a regular basis.

The judiciary’s cultural associations are exemplary of a broader social process whereby legal conventions contribute to judicial legitimacy (Chase, 2005: pp. 119-122). Ceremonies and rituals function to depersonalize and separate the judge from the rest of the society, thereby contributing to perceptions of his impartiality and superiority. Induction ceremonies serve to symbolically separate the judge’s new life as arbiter from his past life as philistine. Judges of all stripes swear a sacralized oath, even private arbitrators (Chase, 2005: pg. 34). In the courtroom, observers rise as the judge enters and as he leaves, commanded to do so by the bailiff. A special appellation, “your honor,” is reserved for judges, which emphasizes the profession’s sacredness and separation. All of these institutions serve to sacralize judges and separate them from society, thus contributing to the legitimacy of their rulings.

One side effect of all of these consecrating rituals might be that judges gain an overly optimistic view of their competence as decision-makers. Support for this view is provided by the poll of 400 state court judges reported in Gatowski et al. (2001). The judges, asked about standards for admissibility of scientific evidence, “overwhelmingly” supported the broad judicial discretion decreed by Daubert. The judges wanted this power, however, while failing to meet minimal standards of scientific literacy as quizzed in the survey. According to the study, “Judges had the most difficulty operationalizing falsifiability and error rate, with only 5% of the respondents demonstrating a clear understanding of falsifiability and only 4% demonstrating a clear understanding of error rate.”

Notwithstanding such evidence of incompetence, there is good evidence that the legal system has managed to convince most Americans of the legitimacy of the legal system. As Schauer & Keckhauser (2009) write, for example, Supreme Court nomination hearings are a “collusive charade. . . designed simultaneously to reassure and mislead the public.” A January 2009 Rasmussen poll found broad support for legalism: 64 percent of Americans believe that “Supreme Court decisions should be based on what is written in the Constitution,” whereas only 27 percent think the decisions “should be guided by fairness and justice.” Another poll, this one from June 2009, found that Americans generally believe that judges follow legalist principles without the influence of their background: 66 percent believe that “well -qualified male and female judges will reach the same decisions most of the time,” with 67 percent believing the same for white and Hispanic judges.

These opinion polls don’t say much about the broader societal effects of law and uncertainty. It might be the case that different legal institutions are associated with higher or lower levels of personal uncertainty and uncertainty aversion. Huang (2008), for example, shows that common-law countries have lower scores for uncertainty aversion than civil-law countries, while the results reported in Botero et al. (2004) indicate that countries with high uncertainty aversion had relatively strict regulations of the labor market. Both of these findings support the idea that high uncertainty aversion is associated with a more codified, rule-based legal system, while lower uncertainty aversion is associated with a more discretionary, case-based legal system. This relationship, if confirmed, would be consistent with one of two views. One possibility is that societies with high uncertainty aversion are motivated to reduce uncertainty, and they attempt to do so by establishing a codified rule regime to deal with any contingency; that codification, however, does not reduce uncertainty aversion significantly. A second possibility is that legal regimes are determined more-or- less randomly, but rule-based regimes (relative to case-based regimes) tend to cultivate high uncertainty aversion in their citizens. The distinction is one of causal direction: Does uncertainty aversion provoke legal reform, or do legal institutions cultivate uncertainty aversion? Because the rule-based/case-based distinction between the common-law and civil-law countries is ancient and has persisted with great inertia, my intuition is that the latter is more likely to be true. As discussed previously, the frequent indeterminacy of statutes makes it difficult to make prudent rulings based solely on those statutes. Because rule-based systems preclude the slow accretion of policymaking knowledge from the observed consequences of decentralized rulings, they cannot reduce social uncertainty as effectively as the case-based systems. The rule-based legal regimes, being less effective regulators than the case-based regimes, increase uncertainty aversion among members of the society.

We can flesh out this view with reference to the finding that countries with high uncertainty aversion have relatively more formality in judicial procedures (Djankov et al., 2007). As mentioned, the rule-based systems, having severe restrictions on the establishment of precedent, lack a mechanism for the gradual, systematic reduction of social uncertainty. What they do have, instead, are mechanisms for reducing emotional uncertainty with the courts’ legitimacy: formality. Dressing up the state machinery that announces these legal rulings with formal accoutrement lowers emotional uncertainty about that legitimacy. Similarly, the higher proportion of civil servants with law degrees in uncertainty-averse countries (Hofstede, 2001) also serves a legitimacy-proving function. Not coincidentally, countries with higher uncertainty aversion also register higher disapproval of the legal system (Hofstede 2001: pg. 171).

5. Conclusion

The cursory inquiry undertaken in the previous sections suffices to show an important psychological and social relation between uncertainty, religion, and law. Granted, it is premature to attempt a broader theoretical subsumption of these ostensibly disparate ideas, but I believe that the present discussion has suggested future directions. Further inquiry might be guided by the perspective that religion and law are not independent social forces, but rather subsets of a broader range of human activities. All human activities, including religion and law, are characterized to various degrees by uncertainty management. On this view, we cannot dichotomize human works into those that manage uncertainty and those that do not; rather, we can view all modes of understanding the world, and all human rhetorical devices, as means to reduce uncertainty in some way. It is no surprise, then, that religion and law involve uncertainty management, as all human works involve uncertainty management. Hofstede's (2001) observation that technology is used to reduce uncertainty accords with this perspective. What has been shown by the present essay's focus is that religion and law are especially powerful institutions in human society, and therefore the uncertainty-management strategies associated with religion and law are also some of the most important. Correspondingly, many of our metaphors for managing uncertainty are drawn from those domains.

Accordingly, I believe that the mode of inquiry taken in this essay could be fruitfully applied to other social domains. The physician-patient relationship, for example, has a significant uncertainty-managing component. Diagnosis of an ailment, for example, serves to reduce uncertainty in both doctor and patient about that ailment. In cases of uncertain etiology, that reduction involves a feeling of control over the ailment. In other cases, it allays feelings of guilt, as when an ADHD diagnosis relieves parents of responsibility for their child's misbehavior.

More generally, the previous sections illustrate the methodological rift between science and history. History's role is contingency; science's role is consistency. Scientists seek ever to simplify through reductionism; historians doggedly refuse to reduce complexity. An example of this is the social-scientific concept of ritual used earlier, which is not only underspecified but also oversimplified. These simplified models are useful, and they can be reliable. But we lose much of the character of ritual when we treat it as a mutually exclusive property, with some human activities being ritual and all others being not-ritual. Rituality has multiple describable components that deserve scrutiny in all of their complexity; this is the kind of inquiry that scientists reject but historians embrace. Whereas the scientist must begin with simple building blocks far removed from the complexity of human experience, historians take what is known about the past and what is known about the science of causality to construct a plausible narrative of historical events. Notwithstanding their different philosophies and methods, both science and history are themselves methods of uncertainty management. They both seek to give plausible explanations of a fundamentally uncertain world. Ritual, for example, is poorly understood, and actually rather mysterious. What is ritual? What does ritual do? Those questions might reveal important truths about human communication, and they can be productively investigated by both scientists and historians.

Field study will be an important component of both scientific and historical research projects into religion and law. We can say with confidence that uncertainty management is lurking in the churches and the courts, but casual theorizing based on current historical and scientific knowledge will only take us so far. We must go into the churches, go into the courts, observe the priests and judges, speak with the adherents and clients. The present essay might be seen as a call for an anthropology of uncertainty management. The churches and the courts will be prominent objects of this research enterprise.

More modestly, we can begin empirical investigation through laboratory experimentation with human subjects. In the subsequent paragraphs, I describe a simple economics experiment that takes a small step toward reducing the scientific uncertainty in this subject area. The experiment is founded in the observation that the existing empirical work on uncertainty and uncertainty management tests the effects of priming emotional uncertainty on subjects who perform a task and then fill out a questionnaire (e.g., van den Bos, 2009). The experimental data consist of the answers to the questionnaires. What hasn’t been done—and which moves us into the area of experimental economics—is to test the effects of salient social priming on uncertainty aversion, as measured in an Ellsberg task (Ellsberg, 1961). My two- fold hypothesis, motivated by the arguments made in the previous sections, is that religious and legalist priming can reduce observed uncertainty aversion in the task.

The Ellsberg task presents the test subject with two jars full of blue and green marbles. One jar, clearly labeled, has five blue marbles and five green marbles. The other jar also has ten marbles in it, but the proportion of blue and green marbles is unknown to the subject. The subject chooses a jar and picks out one marble. If blue, he wins $10; if green, $0. Under standard decision theory, the two jars have the same expected value and the test subject should be indifferent between them. Many experiments have shown, however, that subjects choose the jar with known color proportions about 75 percent of the time.

My proposal is to run this same task, but to presage it with treatment questionnaires designed to prime either a religious or legalist mindset. In the Religious Treatment, the questionnaire will include the following four statements, with agreement and/or disagreement elicited through a 5-point Likert scale:

• “God ensures that good things happen to good people.”
• “God ensures that bad things happen to bad people.”
• “I am a good person.”
• “I am a bad person.”

In the Legalist Treatment, the questionnaire will include the following four statements:

• “Judges ensure that good things happen to good people.”
• “Judges ensure that bad things happen to bad people.”
• “I am a good person.”
• “I am a bad person.”

A Control Treatment will include the following four statements:

• “Hollywood directors ensure that good things happen to good characters.”
• “Hollywood directors ensure that bad things happen to bad characters.”
• “I am a good person.”
• “I am a bad person.”

In this way, I hope to reveal the psychological interaction between legal/religious reasoning and responses to uncertainty. A bias in favor of the known jar relative to controls would suggest that legal/religious thought-processes place individuals into a rule-based cognitive frame that promotes an aversion to uncertainty. A bias in favor of the unknown jar, on the other hand, would suggest that a legal/religious cognitive frame pacifies feelings of uncertainty and increases comfort with the unknown.

As religion and law are widely regarded as powerful social forces, controlled laboratory experiments like this one are essential to elucidate the social-psychological mechanisms that underlie them.