The Natural Environment's Impact upon Religious Ethics: A Cross-Cultural Study Author(s): John Snarey Source: Journal for the Scientific Study of Religion, Vol. 35, No. 2 (Jun., 1996), pp. 85-96 Published by: Wiley on behalf of Society for the Scientific Study of Religion Stable URL: http://www.jstor.org/stable/1387077 Accessed: 29-04-2015 19:28 UTC
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact
[email protected].
Wiley and Society for the Scientific Study of Religion are collaborating with JSTOR to digitize, preserve and extend access to Journal for the Scientific Study of Religion.
http://www.jstor.org
This content downloaded from 170.140.26.180 on Wed, 29 Apr 2015 19:28:23 UTC All use subject to JSTOR Terms and Conditions
The Natural Environment's Impact upon Religious Ethics: A Cross-Cultural Study* JOHN SNAREYt
This study explores the ability of variations in the natural environmentto predict variations in the religious-ethicalviews espoused by different cultural groups. Drawing from the theoretical perspectives of William James and Max Weber,it was hypothesized that societies with environmentalconditionsof potentially life-jeopardizingwater scarcity would be significantly more likely to show an elective affinity for a morallyconcernedSupreme Deity in order to promotethe prosocialuse of natural resourcesand contribute to societal survival. To test this hypothesis, a worldwide survey was conductedby using the 186 cultural groupsin the StandardCross-CulturalSample. After the influence of missionizationwas controlledfor, the findings confirmedthat in societies with a marginally sufficient water supply, a Supreme Deity was significantly more likely to be understoodas concernedwith, and supportiveof, human morality. In contrast, in societies in which water was abundant,the Supreme Deity typically was not concernedwith the morality of human behavior.These findings have implications for developingmore fruitful understandingsof religious environmentalethics.
iNost speculations to date regarding the relationship of religion and nature have suggested that religious-moral orientations function to shape the ways in which peoples of the world relate to the natural environment. For instance, it has been suggested that some theological orientations have promoted or legitimated the domination and abuse of the natural environment (Eckberg and Blocker 1989; Greeley 1993; Toynbee 1972; White 1967). Although this study also brings together concerns with nature and comparative religious ethics, in contrast to prior studies it considers the power of the natural environment to shape religious ethics. The ancient Hebrews, for instance, received ten commandments for moral behavior from Yahweh within the desert environment of the Sinai (Exodus 19:3), and, as farmers and herdsmen, believed that "rain was one of the rewards promised by Yahweh" (Weber 1963:57). Subsequently, "thirsting after righteousness" has functioned as a theological metaphor in the history of both Judaism (Joel 2:23) and Christianity (Matthew 5:6). The thesis of this study suggests that these associations are "natural." THEORETICAL BACKGROUND
Religion functions to maintain the active equilibrium of the social system as a whole and thereby to satisfy the needs of individuals and society at large. This assumption has
This study was originallypresentedas an invited paper at the InternationalConferenceon Eco-EthicalThinkingfrom a Cross-CulturalPerspective,sponsoredby the InternationalAssociationfor Cross-CulturalPsychology,the Ministryof Germany.Special thanks to Science and Culture of the Saarland, and the University of the Saarland, &;aarbrucken, David Bromley,LutzEckensberger,Jean Gearing,Steve Tipton,and anonymousreviewersfor their helpful suggestions. t John Snarey is professorin the School of Theologyand in the Departmentof Psychologyat Emory University,Bishops Hall, Atlanta, GA30322.
?)Journal for the Scientific Study of Religion, 1996, 35 (2): 85-96
This content downloaded from 170.140.26.180 on Wed, 29 Apr 2015 19:28:23 UTC All use subject to JSTOR Terms and Conditions
85
86
JOURNALFORTHE SCIENTIFICSTUDYOF RELIGION
been one of the theoretical mainstays in the psychological and sociological study of religion, including the classic works of William James (1842-1910) and Max Weber (1864-1920). James's psychologicalanalysis of the varieties of firsthand "personal"religious experience highlighted religion's many positive functions in terms of meeting human needs. Religion, for James, was "anessential organ of our life, performinga function which no other portion of our nature can so successfully fulfill" ([1902] 1994:59). Although he was always careful to avoid the reductionism of medical and environmental materialism regarding the value of religious phenomena, James believed that "everyreligious phenomenon has its history and its derivation from natural antecedents" (6). This does not mean, however, that religious experiences are unilaterally conditioned by the environment. Religion, for James, also imaginatively responds to, and dynamically influences, the environment (cf. Browning 1980). James theorized that "the original factor in fixing" the election of different types of gods in different societies was that a society could "use"their God. "Theyrequired him,"for instance, as "a curber" of crimes (361). Morality and religion were also interrelated for James because he believed religion could provide the energy to be moral when a purely "athletic"type of ethics "inevitably"weakens and disintegrates (53). James argued that a primary function of religion was not simply to supply moral prescriptionsthat allowed one to accept the nature of the universe with fear or "stoic resignation," but rather to transcend "moralitypure and simple" and accept the universe with "passionate happiness" (47-48, cf. 86). As James states, "we are in the end absolutely dependent on the universe"- we are finite, bounded, limited - "andinto sacrifices and surrenders of some sort . . . we are drawn and pressed as into our only permanent positions of repose" (59). Religion thus makes attractive and genuine "whatin any case is necessary" (59). Sociologist Max Weber'scitation of William James's work in The ProtestantEthic and the Spirit of Capitalism ([1904] 1930:232-33) is, on balance, a friendly critique of James's judgment regarding the secondary importance of "the rational element" in religion. But James's pragmatic view of ideas was actually more paradoxical than Weber originally realized, making it quite compatible with Weber's own dialectical approach (Myers 1986:471). Similarly, while Weber is primarily known for demonstrating the independent contribution of religious ideas to the rise of economic ethics, he actually stressed their dialectical interplay - religion was one of the determinants of an economic ethic and, in turn, "the religiously determined way of life is itself profoundly influenced by economic and political factors" ([1922-23] 1946:268). After his analysis of calvinistic Puritanism and entrepreneurial capitalism, Weber analyzed the empirical details and meaning of functionally equivalent processes in other societies, including China (Confucianism and Taoism), India (Hinduism and Buddhism), and ancient Israel. His use of comparative evidence may have helped pave the way for the systematic cross-cultural survey research method developed by anthropologists - the method used in this present study. Weber also developed a distinction regarding religious development that shows an intellectual affilnity with James's distinction between spontaneous religious experiences by which individuals originate new insights, and the subsequent process by which some insights, if they "provecontagious enough,"are socially selected for institutionalization as an orthodoxy(James [1902] 1994:34, 369). Weber, more specifically, distinguished the personal and charismatic roots of religious ideas from their subsequent routinization and social impact. It is during the process of routinization, Weber claimed, that society "elects"those religious ideas or elements of ideas with which they have a sympathetic "affinity."Weber'sconcept of "elective affinity"is especially useful for understanding the relations between nature and religion. Weber noted, for example, that peasants are typically inclined towards magic and nature worship because their economic life is directly dependent upon nature and elemental forces. He noted, in contrast, that a practical rationalism tended to characterize the
This content downloaded from 170.140.26.180 on Wed, 29 Apr 2015 19:28:23 UTC All use subject to JSTOR Terms and Conditions
ENVIRONMENTSIMPACTON RELIGIOUSETHICS
87
religion of traders and craftsmen. He argued that their elective affinity for this latter type of religion was "conditionedby the nature of their way of life, which is greatly detached from economic bonds to nature" ([1922-23] 1946:283, 284). Yet these religious views do not function as simple projectionsof economicideas, Weber reminded us. A particular religious view, once established, becomes an important societal power in its own right. Often the "worldimages" that arose from "material and ideal interests" came to function as "switchmen";they "determined the tracks along which action has been pushed by the dynamic of interest" (280). Weber, like James, also contributedto our understanding of religious ethics. The "two primordial methods of influencing supernatural powers," according to Weber, were to "subjectthem to human purposes by means of magic"or "bygratifying their egoistic wishes" ([1922] 1963:37). For instance, "throughoutthe world the magician is in the first instance a rainmaker, for the harvest depends on timely and sufficient rain, though not in excessive quantity"(56). But with the rise of a Supreme Deity, who is understood as the creator of the natural order, "it is naturally postulated that god will protect against injury the order he has created. The intellectual implementation of this postulate . . . stimulated the development of a religious ethic." This development "added obedience to the religious law as the distinctive way to win the god's favor"(37). "Again,"for instance, "rain was one of the rewards"promised by a Supreme Deity to his law-abiding "devotees"(57). Guy Swanson extended streamlined versions of Weber'scomparativeresearch method and his theory that different types of cultural groups tend to select or "give birth" to religious beliefs that functionally reflect the purposes and powers of the larger sociocultural system (1960:vii-viii). Based on a cross-cultural survey of 50 preindustrial societies, Swanson (1960) found that high-god or monotheistic-like religions originated in societies with a parallel political "hierarchyof three or more sovereign groups' (81). Swanson suggested that high-god religions arise when social complexity requires a supreme authority to unify conflicting interests. Subsequent similar research by Ralph Underhill (1975) suggested that both economic and political complexity make independent contributions to the belief in a High God. Neither Swanson nor Underhill, however, found a relationship between social complexity and the likelihood that a High God, once conceived, also would be understood as supporting human morality. Swanson was able to show that supernatural support of human morality was more commonin societies that also had parallel explicit social rules legitimating the morality of "interpersonaldifferences according to wealth" (1960:174). Of course, alternative interpretations of Swanson's and Underhill's findings are possible (for reviews, see Ember and Levinson 1991:98-102; Swanson 1975). In contrast to both James and Weber, Swanson and Underhill seldom moderate the reductionistic logic of cultural materialism. These diverse scholars, nevertheless, hold in commonthe view that religion "doessomething"that generally contributes to a state of equilibrium and that this function is important because human needs are experienced within an environment characterized by a greater or lesser degree of scarcity. This continuity within otherwise diverse streams of theorizing might lead one to expect that all religious conceptions of a Supreme Deity would function to endorse prosocial moral behavior and buttress the interest of the larger society. Yet religions vary in the degree to which they are explicitly concerned with human morality. Swanson (1960) observed, more specifically, that a High God may be understood in several ways: Some High Gods play no active role in the universe - after creation they became idle and of no use. Other High Gods take "an active part in the lives" of people; they may "governthe crops or warfare or the seasons, but seem not to care whether virtue triumphs or the wicked go unpunished" (78). Finally, in some societies, the Supreme Deity is "active in human affairs"and gives "specific support to human morality" (210). Such High Gods exercise control"overthe moral relations between particular individ-
This content downloaded from 170.140.26.180 on Wed, 29 Apr 2015 19:28:23 UTC All use subject to JSTOR Terms and Conditions
88
JOURNALFORTHE SCIENTIFICSTUDYOF RELIGION
uals" and generally "rewardor punish people in accordancewith the way [they] treat their fellows"(159, 78). THEsIs This study addresses the broad thesis that a society's evolutionary survival under environmental conditions of extreme scarcity is promotedby a Supreme Deity's legitimization of moral codes designed to protect natural resources. To test this idea, it had to be operationalized. Under what environmental conditions would this association most likely be measurable?After the need for oxygen, the most essential resource for human existence is water. Besides the human body's physiological need for water, the availability of plants that directly or indirectly provide humans with food is largely determined by precipitation. In this sense, water is the most important feature of any ecological system on our planet (Raikes 1967) and, thus, the primary expression of our finitude. The availability of water is also an especially useful variable in cross-culturalresearch because it is universally crucial for survival but not evenly distributed among diverse populations. When water is scarce, the balance between biological needs and their actual satisfaction can become precarious, and the potential societal costs of violations of prosocial norms governing the use of resources and protecting the well-being of society are quite high. In such a situation, a Supreme Deity may be more likely to endorse prosocialmoral behavior in order to minimize norm violations and ensure the maintenance of the economicbase (e.g., animal herds) necessary for the survival of the population. Ethical deities are not strictly confined to monotheism or monotheisticlike supreme deities, of course, but Weberbelieved that "itis at the level of monotheism that this development has particularly far-reaching consequences" (1968:429). These consequences are uniquely significant because a Supreme Deity's ethical requirements are universalizable, command a distinctly high level of adherence, and cannot be nullified by magic (Weber 1968). To summarize, the study asked the following basic research question: Do variations in the availability of water in the natural environment predict variations in the social conception of a High God's active support of conventional social ethics? To address this question, the study tests the following hypothesis: In societies in which ensuring a sufficient supply of water is difficult, the members of that society will be significantly more likely to conceive of a Supreme Deity who is concerned with, and supportive of, human morality. Societies in which a shortage of water is not experienced, in contrast, will be notably less likely to conceive of a morally concernedSupreme Deity. METHOD
Sample The unit of analysis in this study is the cultural group rather than the individual; more specifically, the study is based upon the 186 societies in the Standard Cross-Cultural Sample (SCCS) (Murdockand White 1969). The SCCS was developedby George P. Murdock and Douglas White to provide a representative worldwide sample of preindustrial societies that were ethnographically well described. The 186 societies are clearly defined in terms of historical time (ethnographic present) and geographical location (longitude and latitude of specific community) and were chosen to represent historically unrelated cultural groups. Together they represent all major cultural areas of the world, including 28 cases from the circum-Mediterranean area, 28 cases from sub-Saharan Africa, 34 from Eastem Europe and Asia, 31 from the Insular Pacific, 33 from North America, and 32 from Central and
This content downloaded from 170.140.26.180 on Wed, 29 Apr 2015 19:28:23 UTC All use subject to JSTOR Terms and Conditions
ENVIRONMENTSIMPACTON RELIGIOUSETHICS
89
South America. In sum, the major strengthg of the sample are that it minimizes the effects of cultural diffusion and adequately represents non-Westem societies. Procedure The ethnographicreports designated by Murdockand White (1969) for each of the 186 societies constitute the actual database for this study, in which the cross-cultural survey method is used. Cross-culturalsurvey research is a method of statistical comparisonthat allows for the systematic investigation of the conditions that predict the occurrenceof various kinds of sociocultural phenomena (Ember and Levinson 1991; Whiting 1968). The method can provide or fail to provide support for a particular hypothesized pattem of association, but, of course, it cannot prove causality (Ember, Ross, Burton, and Bradley 1991; Swanson 1960). The variables under consideration were coded from the ethnographies for these societies. When the appropriate ethnographies were included in the Human Relations Area Files, they were used by the raters. Otherwise, the original ethnographic publications were obtained. Independent raters, blinded to the purpose of the study, examined the ethnographic sources and rated each cultural group in terms of the following factors: High God's supportiveness of human morality; annual precipitation;surface water; and missionization. High God's Morality. Each cultural group's religious beliefs regarding a High God's presence, level of activity, and support for morality were rated by using the classification system originally developed by Swanson (1960). A High God, in Swanson's classification system, refers to a "spiritual being who is believed to have created all reality and/or to be its ultimate governor (1960:210). Swanson's high-god rating system includes the following categories: Absent: a High God is absent or is not reported in substantial descriptions of religious beliefs. Present: a High God is present but inactive and unconcerned with human affairs. Active: a High God is present and active in human affairs but does not offer positive support to human morality. Moral: a High God is present, active, and specifically supportive of human morality. Overall, Swanson's model has the advantage of being more inclusive than rare conceptions of strict monotheism, but also of being more easily rated than categories of diverse types of pantheons and thus more readily applicable as a system for rating the uneven ethnographicliterature on diverse cultural groups. Murdock(1967) reported Swanson ratings for 151 of the 186 societies in the Standard Cross-Cultural Sample. For this study, all 186 societies were also rated by two trained blinded raters. The interrater reliability was .83, using Cohen's Kappa, which corrects for chance agreement (Cohen 1960). When the raters did not agree, both raters' notes and the original ethnographies were carefully restudied, and a final consensus rating for each society was assigned. The final set of ratings was then compared with those provided by Murdock.There was 94%exact agreement between the new ratings and Murdock'sratings;, interrater reliability was .91, using Cohen's Kappa. When the new ratings and Murdock's ratings differed, however, Murdock'sEthnographic Atlas ratings were always used for the sake of consistency with previous studies. Precipitation. Annual precipitation in number of inches for each society was also rated. The data were recorded directly from the ethnographic reports when the information was provided (N = 104). For the remaining cases (N = 79), precipitation was estimated by applying each community'slongitude and latitude to the Klimadiagram weltatlas, a climatic world atlas (Walter and Lieth 1967). To establish the reliability of ratings based on the ethnographic reports, a second blinded rater rerated the records for 50 randomly selected societies. The interrater reliability correlation coefficient was .94. To estimate the reliability of using atlas rather than ethnographic information to rate precipitation, a third blinded rater also rerated 50 addi-
This content downloaded from 170.140.26.180 on Wed, 29 Apr 2015 19:28:23 UTC All use subject to JSTOR Terms and Conditions
90
JOURNALFORTHE SCIENTIFICSTUDYOF RELIGION
tional societies for which the ethnographies provided annual precipitation, using only the climatic atlas. When the atlas ratings and the ethnographic ratings were compared,the reliability coefficient was .84. These figures suggest that atlas ratings are acceptable, although not perfectly equivalent, substitutes for ethnographic ratings. For purposes of contingency table analysis, the average precipitationratings were categorized into three levels by adapting Goode's (1953) customary classification system (Espenshade 1986): high: more than 40 inches per year; moderate:10 to 40 inches of precipitation per year; low: under 10 inches of precipitation per year. Surface Water. Terrain served as an approximate index of available surface water, based on the Whiting and Sodergren ratings (1975). Each society's terrain (continental, insular, coastal, and riverine or lacustrine) was determined by applying the community'slongitude and latitude coordinates to The Time Atlas of the World (Bartholomew et al. 1975). For purposes of contingency table analysis, each community was assigned to one of the following three categories of surface water availability based on terrain: high: a riverine or lacustrine community;moderate:an insular or coastal community that was not also riverine; low: a continental or mainland community that was not also riverine, lacustrine, insular, or coastal. Because interrater reliability was not previously reported,the availability of surface water was rated directly from the ethnographic reports for 50 communities by a blinded rater using the same criteria. The interrater reliability was .72, using Cohen's Kappa. Overall Water Supply. To evaluate the combined ability of precipitation and surface water to predict religious morality, a third variable, overall water supply, was constructed. A low water budget was defined by a natural environment in which neither precipitation nor surface water was abundant (i.e., precipitation and surface water both rated as low or moderate). A high water budget was defined by a natural environment in which either precipitation or surface water was abundant (i.e., precipitation or surface rated as high). Missionization. Christianity and Islam are both "missionary"religions that conceive of a Supreme Deity who is concernedwith the moral behavior of individuals. As Malefijt (1968) notes, such religions generally include a belief that "culturallyapprovedbehavior will be rewarded after death; disapproved behavior will be punished. Such beliefs are relatively rare in the religions of nonliterate peoples, and when they are found, missionary influence can often be discovered"(169). Because missionary activities undertaken in the name of these major religions specifically aim to persuade others to adopt a conceptionof a morally concerned High God, this factor must be controlled for in the analysis. Each community was rated accordingto the following distinctions: Premissionization: the ethnographicpresent is prior to contact with Christian or Islamic religious traditions, or missionization began concurrentlywith the ethnographic present, without sufficient time for missionaries to impact local religion, or contact occurred prior to the ethnographic present, but it was documented to have been completely unsuccessful at the time of the ethnographic present. Postmissionization: the ethnographic present is after missionaries have had an observable influence on the community as indicated by an ethnographic report of conversions, baptisms, the establishment of a native congregation, or other specific evidence that elements of outside beliefs had been assimilated. Because Islam is a relatively recent monotheistic religion, the following additional criteria were also used: If a communitywas part of the original rise of Islam (from the Hijra in 622 to ca. 700 A.D.), it was rated as being indigenously Islamic, but when a community became Islamic after the period of the original establishment of Islam, it was rated as having been missionized. There were also a small number of communities in the sample that had had contact with outside religions that themselves did not satisfy Swanson's criteria for believing in a morally concemed or ethically offendable Supreme Deity (e.g., Buddhism, Hinduism). Such communities were not assigned to the postmissionization category because the presence or absence of a belief in a High God could
This content downloaded from 170.140.26.180 on Wed, 29 Apr 2015 19:28:23 UTC All use subject to JSTOR Terms and Conditions
ENVIRONMENTSIMPACTON RELIGIOUSETHICS
91
not be attributed to the influence of the nonmonotheistic religions with which they may have had contact. Two blinded raters assessed all societies for missionization. The interrater reliability was .79, with Cohen's Kappa. When the two sets of ratings differed,the raters' notes and the original ethnographies were carefully restudied so a final consensus rating could be assigned. RESULTS
The belief in a High God who actively supported human morality characterized 23.1% of the 186 cultural groups. By comparison, a High God was understood as active in human affairs but unconcerned with human morality among 12.4%of the cultural groups, as both inactive and unconcerned, among 27.4%of the cultural groups, and as simply not part of the religious beliefs of 37.1% of the cultural groups. The average annual level of precipitation was 58.3 inches (SD = 42.6) among the 186 cultural groups. The results of a one-way analysis of variance test showed that the environments of the four different categories of cultural groups, based on their conceptions of a High God, differed significantly in the number of inches of precipitation they received each year (F (3,182) = 6.60, p < .00 1). Post hoc analysis (Student-Newman-Keuls procedure)showed that the difference lay between only two subsets - morally oriented versus nonmoral conceptions of a High God (p < .05). Cultural groups in which a High God was understood as supportive of prosocial moral values had an environment with a significantly lower average level of annual precipitation (M = 35.1 inches) than did cultural groups in which a High God was seen as morally unconcerned but active in human affairs (M = 57.7 inches), or inactive in human affairs (M = 62.9 inches), or absent (M = 69.5 inches). Results of the first cross-tabulation reported in the upper section of Table 1 summarizes the relation of the four conceptions of a High God with categorical indices of the cultural groups' natural environments. The average annual precipitation was low for 9.1% of the cases, moderate for 31.2%, and high for 59.7%. The association between three levels of annual precipitation and conceptions of a High God was moderately strong (Cramer's V = .32) and highly significant (p < .001). Societies in which precipitation was low were most often rated as having a morally concernedHigh God, whereas societies with moderate precipitation were more evenly distributed across the four categories, and societies in which precipitation was high were most often rated as having either an inactive or absent High God. The availability of surface water in the natural terrain was low for 56.5%of the cases, moderate for 19.8%,and high for 23.7%.The second cross-tabulation shows that the association between surface water availability and conceptions of a High God was moderately strong (V = .26) and highly significant (p < .00 1). A morally concerned High God was the most frequent belief about the deity among communities with little surface water, whereas it was an infrequent religious belief among communities with abundant surface water. The two indices of water, precipitation and surface, were modestly but significantly associated (V = .19, X2 = 13.54, df = 4, p < .0 1). Results of the third cross-tabulationshown in the upper section of Table 1 indicate that a combination of precipitation and surface water factors was better able to predict religious morality than either variable alone. Overall, 32.6% of the societies were characterized by a low water budget, and 67.7% of the societies were characterized by a high water budget. Water availability had a strong (V = .40) and highly significant (p < .001) association with conceptions of a High God. Societies with a low water budget most frequently conceived of a High God who was supportive of prosocial moral values, whereas societies with a high water budget most frequently had no conception of a morally active High God.
This content downloaded from 170.140.26.180 on Wed, 29 Apr 2015 19:28:23 UTC All use subject to JSTOR Terms and Conditions
JOURNALFORTHE SCIENTIFICSTUDYOF RELIGION
92
TABLE 1 ENVIRONMENTALPREDICTORS OF CONCEPTIONS OF A SUPREME DEITYS MORALITY
Predictor Variables
Absent
Annual Precipitation 4 Low 15 Moderate 50 High SurfaceWater Low Moderate High
31 21 17
Precipitationand Surface Neither High 14 Either High 55 Missionized No Yes
33 36
Coneegtions of a High God Present, Present, Active, Active, Morally Morally Present, Concerned Detached Inactive
23 12 16
15 1 7
36 3 4
10 41
8 15
28 15
V
p
.32
