Je ne suis pas venu apporter la paix, mais l’épée. Car je suis venu mettre la division entre l’homme et son père, entre la fille et sa mère, entre la belle-fille et sa belle-mère; et l’homme aura pour ennemis les gens de sa maison. Jesus (Matthieu 10: 34-36)
Il n’y a plus ni Juif ni Grec, il n’y a plus ni esclave ni libre, il n’y a plus ni homme ni femme; car tous vous êtes un en Jésus-Christ. Shaoul dit Paul
Croire en un Dieu cruel rend l’homme cruel. Thomas Paine
Au nom de Dieu, amen. Nous soussignés, loyaux sujets de notre respecté souverain Jacques, par la grâce de Dieu Roi de Grande-Bretagne, de France et d’Irlande, défenseur de la foi, etc . Ayant entrepris, pour la gloire de Dieu, pour la propagation de la foi chrétienne, et l’honneur de notre roi et de notre pays, un voyage pour implanter la Première Colonie dans les régions septentrionales de Virginie, par la présente, nous convenons solennellement ensemble, devant Dieu et devant chacun d’entre nous, de nous constituer en un corps politique civil, pour notre administration et sauvegarde et par delà, aux fins susdites ; et en vertu de cela de nous conformer, de décider et de concevoir à l’occasion des lois, ordonnances, actes, décrets et obligations, aussi justes et équitables qu’il semblera à propos et convenable d’adopter pour le bien public de la Colonie, et auxquelles nous promettons toute la soumission et l’obéissance requises. En témoignage de quoi nous avons ci-dessous apposés nos noms à Cape Cod, ce 11 novembre de la quatrième année du règne de notre souverain seigneur Jacques, dix-huitième roi d’Angleterre, de France et d’Irlande, et cinquante-quatrième roi d’Écosse. Anno Domini 1620. Mayflower Compact
Les hommes sont inégaux par nature: voilà le point d’où il faut partir. Clémence Royer
Les athées ne commettent pas d’abominations au nom de l’athéisme. Richard Dawkins
High religiosity is not universal to human populations, and it is actually inversely related to a wide range of socio-economic indicators representing the health of modern democracies.
Once a nation’s population becomes prosperous and secure, for example through economic security and universal health care, much of the population loses interest in seeking the aid and protection of supernatural entities. This effect appears to be so consistent that it may prevent nations from being highly religious while enjoying good internal socioeconomic conditions. Gregory Paul
Many Americans agree that their churchgoing nation is an exceptional, God-blessed, shining city on the hill that stands as an impressive example for an increasingly sceptical world. In general, higher rates of belief in and worship of a creator correlate with higher rates of homicide, juvenile and early adult mortality, STD infection rates, teen pregnancy and abortion in the prosperous democracies. The United States is almost always the most dysfunctional of the developing democracies, sometimes spectacularly so. (…) Europeans are increasingly repelled by the poor societal performance of the Christian states (…) The non-religious, proevolution democracies contradict the dictum that a society cannot enjoy good conditions unless most citizens ardently believe in a moral creator. The widely held fear that a Godless citizenry must experience societal disaster is therefore refuted. Gregory Paul
Conservative religious ideology is a probable contributing causal factor of societal dysfunction. Specifically, the U.S. scores the most dysfunctional in homicide, incarceration, juvenile mortality, gonorrhea and syphilis infections, abortions, adolescent pregnancies, marriage duration, income disparity, poverty, (and) work hours. Gregory Paul
Prior research has emphasized the good consequences of religion and Paul emphasizes the negative aspects, with the results often depending on what factors are picked to base a judgment on. God-versus-Satan religious cosmologies encourage homicide, while nations with belief in a benevolent God have low rates of homicide. Both the ‘religion good’ and ‘religion bad’ perspectives are great oversimplifications.Gary Jensen (sociologue, Vanderbilt University)
Attention: la liberté peut être dangereuse pour votre santé !
Criminalité, incarcération, maladies véneriennes, grossesses précoces, avortements adolescents, alcoolisme, divorces, disparités économiques, pauvreté, mortalité infantile, espérance de vie …
En ce 390e anniversaire du Mayflower compact …
Y aurait-il (mis à part étrangement les cambriolages et le suicide ou l’alcoolisme ou la cyrrhose du foie?) des maux que ne posséderait pas au plus haut point la societé, comme le rappellent nos chercheurs les plus pointus à l’instar du paléontologue Gregory Paul, la plus mortifere du monde civilisé ?
Mais faut-il s’en étonner quand on sait l’origine chrétienne explicitement revendiquée de son peuplement ?
Le premier acte gouvernemental de la colonie de Plymouth du cap de la Morue et célèbre précédent de la démocratie en Amérique imposait-il en effet autre chose en ce fatidique 11 novembre de l’an de grace 1620 (calendrier julien) …
Que la servitude (certes à durée limitée) dont, sous prétexte du dépaysement accidentel de leur lieu de contrat, tentaient de se dégager une minorité de 18 colons ?
Et l’imposait-il autrement, comme en témoigne la débauche de terminologie chrétienne qu’il contient (« Au nom de Dieu, amen », « Ayant entrepris, pour la gloire de Dieu, pour la propagation de la foi chrétienne », ‘Anno Domini 1620″), qu’au nom d’une religion qui avait déja provoqué comme on le sait la Chute de l’Empire romain ?
September 27, 2005
RELIGIOUS belief can cause damage to a society, contributing towards high murder rates, abortion, sexual promiscuity and suicide, according to research published today.
According to the study, belief in and worship of God are not only unnecessary for a healthy society but may actually contribute to social problems.
The study counters the view of believers that religion is necessary to provide the moral and ethical foundations of a healthy society.
It compares the social performance of relatively secular countries, such as Britain, with the US, where the majority believes in a creator rather than the theory of evolution. Many conservative evangelicals in the US consider Darwinism to be a social evil, believing that it inspires atheism and amorality.
Many liberal Christians and believers of other faiths hold that religious belief is socially beneficial, believing that it helps to lower rates of violent crime, murder, suicide, sexual promiscuity and abortion. The benefits of religious belief to a society have been described as its “spiritual capital”. But the study claims that the devotion of many in the US may actually contribute to its ills.
The paper, published in the Journal of Religion and Society, a US academic journal, reports: “Many Americans agree that their churchgoing nation is an exceptional, God-blessed, shining city on the hill that stands as an impressive example for an increasingly sceptical world.
“In general, higher rates of belief in and worship of a creator correlate with higher rates of homicide, juvenile and early adult mortality, STD infection rates, teen pregnancy and abortion in the prosperous democracies.
“The United States is almost always the most dysfunctional of the developing democracies, sometimes spectacularly so.”
Gregory Paul, the author of the study and a social scientist, used data from the International Social Survey Programme, Gallup and other research bodies to reach his conclusions.
He compared social indicators such as murder rates, abortion, suicide and teenage pregnancy.
The study concluded that the US was the world’s only prosperous democracy where murder rates were still high, and that the least devout nations were the least dysfunctional. Mr Paul said that rates of gonorrhoea in adolescents in the US were up to 300 times higher than in less devout democratic countries. The US also suffered from “ uniquely high” adolescent and adult syphilis infection rates, and adolescent abortion rates, the study suggested.
Mr Paul said: “The study shows that England, despite the social ills it has, is actually performing a good deal better than the USA in most indicators, even though it is now a much less religious nation than America.”
He said that the disparity was even greater when the US was compared with other countries, including France, Japan and the Scandinavian countries. These nations had been the most successful in reducing murder rates, early mortality, sexually transmitted diseases and abortion, he added.
Mr Paul delayed releasing the study until now because of Hurricane Katrina. He said that the evidence accumulated by a number of different studies suggested that religion might actually contribute to social ills. “I suspect that Europeans are increasingly repelled by the poor societal performance of the Christian states,” he added.
He said that most Western nations would become more religious only if the theory of evolution could be overturned and the existence of God scientifically proven. Likewise, the theory of evolution would not enjoy majority support in the US unless there was a marked decline in religious belief, Mr Paul said.
“The non-religious, proevolution democracies contradict the dictum that a society cannot enjoy good conditions unless most citizens ardently believe in a moral creator.
“The widely held fear that a Godless citizenry must experience societal disaster is therefore refuted.”
Can we all get along? Maybe not when it comes to science and religion.
Just ask scientist Francis Collins, installed last month as head of the National Institutes of Health.
It wasn’t just newspaper editorials — scientists such as Harvard’s Steven Pinker, who called Collins an « advocate of profoundly anti-scientific beliefs, » criticized placing an outspoken evangelical Christian in the post.
On his first day on the job, Collins stepped down from the BioLogos foundation he founded to foster a rapprochement between the spiritual and the scientific worlds, after such complaints. « I want to reassure everyone I am here to lead the NIH as best I can, as a scientist, » Collins said at an August briefing.
Amid the noise over books like Richard Dawkins’s The God Delusion; the 2005 Kitzmiller court case that found « intelligent design » a form of creationism; and the Bush administration’s 2001 announcement of a (since-overturned) religiously motivated policy limiting human stem cell research funding to a few cell families, it’s easy to overlook what science says about religion itself. But it might explain why the religious-minded sometimes similarly view science with suspicion.
In the June Journal of Religion, Wendy Cadge of Brandeis University, for example, looks at the history of medical studies examining the « power of prayer » to heal the sick. Since 1965, about 18 studies have looked at the effect, to much controversy, with a 2006 re-analysis of past studies finding no health benefits to stranger’s prayers.
But the studies, positive or negative, likely say more about the scientists than about prayers themselves, Cadge concludes. The first studies looked only at Protestant prayers, while later ones added Jewish, Buddhist and other prayers, reflecting growing cultural awareness among researchers of religious variety. Studies used double-blind clinical trial methods to assess prayers, which Cadge in a statement called « fraught with problems. » Scientists couldn’t know for sure if people in their « no prayers » control groups were truly not being prayed for, for example. And they couldn’t agree on the right « dosage » of prayer.
« Scientists tried their best to study something that may be beyond their best tools, » said Cadge. « And reflects more about them and their assumptions than about whether prayer ‘works.’ «
This kind of criticism doesn’t keep scientists from studying religion. In the journal Evolutionary Psychology in July, for example, independent scholar Gregory Paul constructed a « successful society index » to try and analyze the societal effects of religious belief. « What theological, social and economic arrangement produces the best possible societal conditions? » Paul asks in the study.
Scholars have argued this question at least since 1776, when Edward Gibbon’s first volume of The History of the Decline and Fall of the Roman Empire blamed Christianity for helping to sap the empire’s resolve against barbarian invasions. More recently, the German economist Max Weber argued in 1904’s The Protestant Ethic and the Spirit of Capitalism that Protestantism, broadly speaking, led to the development of modern economies.
In his bid to answer the question, Paul looked at 25 indicators of economic well-being, crimes such as murder, corruption, alcoholism and life expectancy in Australia, New Zealand, Japan, Europe and the United States, and then compared them to surveys of religious devotion in each nation.
« Conservative religious ideology is a probable contributing causal factor of societal dysfunction, » he concludes, bad news for the USA, the most religious country in his survey. « Specifically, the U.S. scores the most dysfunctional in homicide, incarceration, juvenile mortality, gonorrhea and syphilis infections, abortions, adolescent pregnancies, marriage duration, income disparity, poverty, (and) work hours, » says the study.
Looks bad for religion, then? Not so fast, says sociologist Gary Jensen of Vanderbilt University. « Prior research has emphasized the good consequences of religion and Paul emphasizes the negative aspects, » he says, with the results often depending on what factors are picked to base a judgment on.
Jensen’s own surveys paint a more mixed picture of religion’s impact on society. « God-versus-Satan religious cosmologies encourage homicide, while nations with belief in a benevolent God have low rates of homicide, » he says, based on a 2006 Journal of Religion & Society journal study. « Both the ‘religion good’ and ‘religion bad’ perspectives are great oversimplifications, » he says
In the name of God, Amen. We whose names are underwritten, the loyal subjects of our dread Sovereign Lord King James, by the Grace of God of Great Britain, France and Ireland, King, Defender of the Faith, etc.
Having undertaken, for the Glory of God and advancement of the Christian Faith and Honour of our King and Country, a Voyage to plant the First Colony in the Northern Parts of Virginia, do by these presents solemnly and mutually in the presence of God and one of another, Covenant and Combine ourselves together into a Civil Body Politic, for our better ordering and preservation and furtherance of the ends aforesaid; and by virtue hereof to enact, constitute and frame such just and equal Laws, Ordinances, Acts, Constitutions and Offices, from time to time, as shall be thought most meet and convenient for the general good of the Colony, unto which we promise all due submission and obedience.
In witness whereof we have hereunder subscribed our names at Cape Cod, the 11th of November, in the year of the reign of our Sovereign Lord King James, of England, France and Ireland the eighteenth, and of Scotland the fifty-fourth. Anno Domini 1620.
1. John Carver
2. William Bradford
3. Edward Winslow
4. William Brewster
5. Isaac Allerton
6. Miles Standish
7. John Alden
8. Samuel Fuller
9. Christopher Martin
10. William Mullins
11. William White
12. Richard Warren
13. John Howland
14. Stephen Hopkins
15. Edward Tilly
16. John Tilly
17. Francis Cooke (sic)
18. Thomas Rogers
19. Thomas Tinker
20. John Ridgdale
21. Edward Fuller
22. John Turner
23. Francis Eaton
24. James Chilton
25. John Craxton (sic)
26. John Billington
27. Joses Fletcher (sic)
28. John Goodman
29. Digery Priest (sic)
30. Thomas Williams
31. Gilbert Winslow
32. Edmund Margeson
33. Peter Brown
34. Richard Bitteridge (sic)
35. George Soule
36. Richard Clark (sic):37. Richard Gardiner
38. John Allerton
39. Thomas English
40. Edward Doten (sic)
41. Edward Leister
Traduction francaise :
Au nom de Dieu, amen. Nous soussignés, loyaux sujets de notre respecté souverain Jacques, par la grâce de Dieu Roi de Grande-Bretagne, de France et d’Irlande, défenseur de la foi, etc ».
« Ayant entrepris, pour la gloire de Dieu, pour la propagation de la foi chrétienne, et l’honneur de notre roi et de notre pays, un voyage pour implanter la Première Colonie dans les régions septentrionales de Virginie, par la présente, nous convenons solennellement ensemble, devant Dieu et devant chacun d’entre nous, de nous constituer en un corps politique civil, pour notre administration et sauvegarde et par delà, aux fins susdites ; et en vertu de cela de nous conformer, de décider et de concevoir à l’occasion des lois, ordonnances, actes, décrets et obligations, aussi justes et équitables qu’il semblera à propos et convenable d’adopter pour le bien public de la Colonie, et auxquelles nous promettons toute la soumission et l’obéissance requises. En témoignage de quoi nous avons ci-dessous apposés nos noms à Cape Cod, ce 11 novembre de la quatrième année du règne de notre souverain seigneur Jacques, dix-huitième roi d’Angleterre, de France et d’Irlande, et cinquante-quatrième roi d’Écosse. Anno Domini 1620. »
Religious Cosmologies and Homicide Rates among Nations: A Closer Look1
Gary F. Jensen
Department of Sociology
Nashville TN 37235
Although religion has been viewed as playing an important role in the maintenance of moral order, the most recent analysis of variation in homicide rates among nations argues that homicide is facilitated by high levels of religiosity (Paul 2005). That analysis, however, was based on scatterplots for eighteen “prosperous nations” and focused primarily on the United States compared to asecular nations. Because there are numerous dimensions to religiosity and a variety of alternative explanations of homicide rates, a more complex analysis is required before more definitive conclusions can be reached. This study attempts such an analysis for a much larger sample of nations and tests Durkheim’s hypothesis that religious passion as a variable characteristic of nations is a positive correlate of homicide rates. A multiple regression analysis reveals a complex relationship with some dimensions of religiosity encouraging homicide and other dimensions discouraging it. The relationships found not only survive controls for variables proposed in prior research, but suggest major modifications to theories focusing on economic variables.
Religion and Homicide: Conflicting Themes
Any thorough attempt to assess the role of religion in relation to violence will encounter two conflicting themes. On the one hand, religious institutions, beliefs, and practices are commonly depicted as discouraging crime, either directly or indirectly through links with other social forces. The two sociologists who have written the most about religion and deviance, Rodney Stark and William Sims Bainbridge argue that church attendance contributes to moral integration and, thus, inhibits a wide range of forms of deviance (1996). In fact, Bainbridge and Stark propose that the preventative effect of religiosity is one of the few general propositions in the sociology of deviance that applies at both the individual and the collective level. Indeed, a meta-analysis of research on religion and crime (Baier and Wright 2001: 3) concluded that “religious behaviors and beliefs exert a moderate deterrent effect on individuals’ criminal behavior.” Stark and Bainbridge propose the same conclusion, but argue that it applies at both the individual and ecological level.
In contrast to the preventative argument, some dimensions of religiosity have been proposed as sources of violence in one form or another. Religious zeal has been cited as a source of national and international problems, ranging from homicide, hate crime and terrorism to genocide and ethnic cleansing. Reflecting upon When Religion Becomes Evil (2003), Charles Kimball proposes that religious belief systems tend to become destructive when they are characterized by absolute truth claims, notions of a cosmic struggle between God and the Devil (cosmic dualism), and rigid dichotomies between good and evil. In short, when certain types of religious attitudes predominate in a nation, Kimball’s argument suggests that they can contribute to an inclination towards interpersonal violence.
The view that religious variables can have negative consequences was highlighted most recently in an article by Gregory Paul (2005). In an analysis of eighteen “prosperous nations,” Paul reports positive relationships between a variety of measures of religiosity and homicide rates as well as other social problems. He concludes that secular nations have lower homicide rates and less serious social problems than found in the United States. His conclusions were based on an examination of scatter-plots for a small set of nations with no attempt to consider alternative explanations nor to encompass the research in the larger body of sociological theory and research on the topic.
Prior Theory and Research
The view that religion might play a role in encouraging homicide has a long history, beginning with Durkheim=s classic study of suicide over a century ago (1897). Although Durkheim is best known for his argument that religious integration inhibits suicide, he proposed a contrary argument for homicide. Depicting homicide as “a violent act inseparable from passion,” he proposed that passionate attachment to religious group life encouraged homicide (1951 edition: 340). The fact that he emphasized “passion” in his explanation suggests that his argument might not apply to all forms of religiosity.
A few clues about religious passion, dualist cosmologies and violence can be found in more recent theory and empirical research. For example, in an attempt to make sense of high rates of homicide and low rates of suicide in the American South, Luper, Hopkinson and Kelly (1985) suggest that southern Protestant fundamentalism ascribes intentionality to people’s actions, prompting people to react to a wide variety of situations as intentional attacks requiring a personal counter attack. Grasmick, Davenport, Chamlin and Bursik (1992) have developed similar arguments to explain southern support for punitive sanctions, and Unnithan, Huff-Corzine, Corzine and Whitt (1994: 149) have extended this line of argument by proposing that adherence to a fundamentalist doctrine would increase the chances of attributing the causes of one’s failures to the malevolent acts of others, thus resulting in aggression being directed outward rather than inward.
Recent research on homicide among cities in the United States reports findings quite compatible with the religious passion argument. Among southern cities, Ellison, Burr and McCall (2003) found the percent of evangelical Protestants to be a positive correlate of homicide rates when other relevant variables were controlled. As a type of social bond, a type of conventional activity, religion may inhibit a variety of types crime, but measures of religiosity that tap into religious passion, evangelical dualism, or belief in malevolent forces may have the opposite effect on homicide.
Much of the literature suggests that certain forms of religiosity are likely to contribute to high rates of homicide, and a far more complex assessment of that issue should be part of that future research. It seems reasonable to expand on Kimball=s perspective to propose that when the moral and religious universe encompassing individuals involves cosmic struggles between benevolent and malevolent forces, moral struggles between “good guys” and “bad-guys,” and dichotomous choices between good or evil, then there is little or no inclination to consider any middle ground, negotiation, or flexibility in dealing with lesser conflicts and struggles in everyday life. It may be that a religious cosmology with moral “wars” and “dueling deities” sets the stage for culture wars (Hunter 1991), facilitates interpersonal wars, and encourages people in conflict to think in terms of dueling contenders for righteousness. When moral boundaries are rigid, it may be easier to offend or “dis”others and harder to assume a personal responsibility for generating conflict. When there is only good and evil and there has to be a clear moral winner or immoral loser, then the options for controlling violent outcomes may be greatly restricted. In summary, there are precedents for proposing that different types of religiosity, differentially structured religious belief systems (religious cosmologies), and related dualistic world views can affect the structure of lethal violence among societies.
Such an elaboration may make sense of the disparate perspectives on religion reflected in Stark=s recent work as compared to Kimball,s and Paul,s arguments. In For the Glory of God (2003: 376), Stark rejects the notion that “religion is all about ritual” and proposes that “Gods are the fundamental features of religions.” However, the focus of that work is on “monotheistic” conceptions of God and his final argument is that such beliefs contribute to moral order. However, the dominant religious cosmology in many nations, including the United States, incorporates both benevolent and malevolent elements and passionate dualisms that cannot be encompassed under the simple notion of monotheism. Contrary to Stark’s exclusive emphasis on the positive consequences of features of Gods, this paper suggests that certain religious cosmologies characterized by contending Gods are associated with dualisms that facilitate high rates of homicide. The Gods do matter, but in a far more complex fashion than proposed.
A Multiple Regression Analysis
There has been very little research on religion and homicide among nations and Paul’s recent analysis is limited to eighteen nations with the primary focus on the United States compared to a few “secular” nations. Paul calls for further research and debate, but expresses a very negative view of the methodology necessary for progress in understanding the relation between religion and homicide–multivariate analysis.
Among reasons given for precluding a multivariate analysis is the argument that such analyses “risk manipulating the data to produce errant or desired results (2005: 12).” However, his presentation of simple scatter-plots focusing heavily on the United States runs the same risk. Without some form of multivariate analysis, the findings may be “errant” (e.g. spurious) and the bivariate analysis may be preferred because it produces desired results.
The key issues suggested in the theoretical and research literature cannot be addressed without a multivariate approach. The literature suggests that some forms or dimensions of religiosity are more conducive to homicide than others, an idea that requires an assessment of separable, independent associations between measures of religiosity and homicide. There are competing theories of variation among nations (e.g. institutional anomie) emphasizing economic characteristics that cannot be controlled fully by limiting the analysis to scatter-plots for a small set of prosperous nations. In short, multivariate analysis is mandatory for adjudicating among alternative perspectives and is a necessary tool for furthering research and debate. This study uses multiple regression techniques to isolate religious covariates of homicide among World Value Survey nations.
The World Values Surveys (WVS) were conducted between 1990 and 1993 and between 1995 and 1997.2 The surveys were limited to persons 18 years of age and older, randomly selected from randomly selected locations. Samples from as many as 54 nations provided data on several dimensions of religiosity in one or both of the surveys. In both waves, respondents were asked questions about the importance of God and religion in their lives, beliefs in the Devil, Heaven and Hell, belonging to a religious faith and attendance at religious services.
Measures of intensity are particularly important to tap the passion invoked in Durkheim’s argument, and other items are relevant to arguments about dichotomous conceptions of the religious cosmos with malevolent and benevolent dimensions. If arguments about the direction of violence are correct, measures of religiosity that tap intensity, dualism, or malevolence should be positive correlates of homicide and negative correlates of suicide. In contrast, measures of religiosity that tap mere acknowledgment of a belief in God’s existence, belief in Heaven, belonging to a religious faith, and attendance at services, should be less relevant to the structure of lethal violence.
Data on average homicide and suicide rates among nations are derived from reports posted on the Internet by the World Health Organization in their compilation of causes of death (World Health Statistics Annual 1997-1999, Online Edition, 2000). The homicide and suicide rates are based on the three most recent reports between 1992-1998 when possible and the estimate for one or two years when data for three years was missing. Interpol estimates were used to fill in missing cases when available and earlier WHO data were used as a last resort. The same procedure was followed for suicide rates.
Although some patterns for both homicide and suicide rates will be presented, the focus in this paper will be on homicide rates for several reasons. For one, there is a sizeable body of research on suicide while no definitive relationship has been established between any dimension of religiosity and rates of homicide (See Lester 1987). Hence, a thorough analysis relevant to the effects of religion on homicide is the first priority at this point in time. Second, Paul’s study has received national attention, despite its weaknesses, and a more elaborate assessment needs to be presented as soon as possible. Third, the analysis will contrast the impact of religious cosmologies with variables central to theories focusing on dimensions of economic culture specifically proposed to explain homicide rates (See Messner and Rosenfeld 1997). This set of goals is quite ambitious and a detailed analysis of the stream analogy and the direction of violence will have to be considered in future analyses.
Table 1 summarizes the bivariate correlations between logged homicide and suicide rates for eight WVS items tapping distinct features of religion. The findings are remarkably consistent with Durkheim’s argument in that fifteen of sixteen correlations are statistically significant and are patterned as predicted. Moreover, measures that tap intensity and belief in malevolent religious or cosmic forces are more strongly correlated with homicide rates than the measures tapping belonging, attendance, belief in God, and belief in Heaven.
The average correlation with homicide rates for the intensity and malevolence items is +.512 as compared to +.220 for the measures that tap institutional belonging, social involvement, and belief in God and Heaven. Consistent with Stark’s argument,
features of the Gods and the religious cosmos are stronger correlates than the measures of ritual (attendance and belonging).
(Table 1 about here) Because the measures of religiosity are positively correlated with one another, an analysis based solely on bivariate relationships could yield quite errant conclusions. Among nations, the percentage who believe in God is very highly correlated with the percentage who believe in the Devil. Belief in Heaven tends to be correlated with belief in Hell. In fact, the alpha index for all eight religiosity items in Table 1 exceeds .90 which would justify creating a single index of general religiosity. However, this analysis is guided by theoretical precedents that imply a structure to the relationships with homicide–a structure that may reflect Stark’s view that religion is far more than ritual and Durkheim’s emphasis on passion. Measures of religiosity may be highly correlated with one another, but still enter into quite distinct relationships with homicide.
Table 2 summarizes the results of regressing logged homicide rates on summed z-scores for 1) the two items relevant to a malevolent religious cosmos (the Devil and Hell), 2)the two items tapping intensity or passion (the importance of God and religion in respondents’ lives), 3) two items tapping relatively benevolent beliefs (God and Heaven), and 4) two items tapping more ritualistic dimensions of religion (belonging and participation).
The pattern of findings is remarkably consistent with both Kimball’s argument about the properties of religious cosmologies that facilitate violence and the implications of Stark’s arguments about the contribution of belief in God for moral order. When the different dimensions of the religious cosmos are introduced, measures of passion and malevolence are strong positive correlates of homicide rates among nations while the more benevolent beliefs are strong negative correlates. The measure of ritual is not related to homicide rates when included in the same analysis with the other measures.
Moreover, the pattern found is not a product of the effect of high levels of collinearity among independent variables on the coefficients isolating the impact of different sets of items. Not only do ridge regression tests uniformly indicate that multi-collinearity is not a significant problem for the analysis in Table 2, but ridge regression analysis correcting for any such problem do not alter the results. Although highly correlated with one another, the different measures appear to be tapping different features of religion that covary with homicide rates in different ways. (Table 2 about here)
In an attempt to assess Paul’s argument about the benevolent consequences of secularism and the implication of Kimball’s argument that religious dualism encourages homicide, nations were classified based on the percentages of respondents that believe in God and the percentage who believe in the Devil. The scatter-plot for these two items is reported in Figure 1 with the United States located in the upper right corner. The United States can be considered dualist in that 96 percent believe in God and 76 percent believe in the Devil.
South Africa, the Philippines, and the Dominican Republic fall in the upper right corner as well. In contrast, the nations in the bottom left corner exhibit relatively low levels of belief in either God or the Devil. For example, 56 percent of the respondents from Sweden believe in God and 18 percent believe in the Devil. The nations in this quadrant tend to be viewed as secular. But, it should be noted that the sizeable percentages of respondents in these nations report they believe in God.
Hence, the nations designated as secular should be viewed as relatively more secular than other nations. (Figure 1 about here) The respondents in some nations exhibit a low level of belief in the Devil, but a relatively high belief in God. These nations might be considered as the most “monotheistic” in that respondents do not recognize the Devil as a significant actor in their religious cosmos. For example, 85 percent of respondents from Iceland indicate belief in God, but only 19 percent acknowledge a belief in the Devil. Spain (91 vs 32), Switzerland (84 vs 32), and Austria (87 vs 23) exhibit similar disparities. These nations are neither secular nor dualist and, for the sake of simplicity, will be categorized as “God-Only.”
There are several nations that might be considered dualist because of the combination of belief in both God and the Devil, but do not exhibit particularly high levels of belief in God. For example, 80 percent of Australians express belief in God and 47 percent express belief in the Devil. Australia is certainly more dualist than Iceland, but does not fall in the top third in terms of belief in God and belief in the Devil. Seventy-seven percent of Ukrainians express belief in God and 47 percent express belief in the Devil. Again, there is greater dualism than for the God Only nations. Hence, two categories of dualism were created to encompass the nations that did not fall towards the two ends of the continuum. That middle set was differentiated into those where the ratio of God to the Devil exceeded the median (Hi Dual) and those below the median (Dual). (Table 3 about here)
As summarized in Table 3, the high dualist nations have the highest homicide score followed by the lesser dualist nations with God-Only and secular nations exhibiting lower scores. However, the statistically significant contrast is between the dualist nations and the God-Only and secular nations. Relative to dualist nations, nations with a sizeable percentage believing in God (but not the Devil) have a significantly lower score. The most secular nations exhibit a significantly lower score than dualist nations as well. But, contrary to Paul’s emphasis on secular versus religious nations, there is no difference between the non-dualist, God believing nations, and the relatively more secular nations. These patterns are quite consistent with the multivariate analysis reported above. Some features of religious belief systems are negative correlates of homicide and some features of religion are positive correlates.
A legitimate response to the findings involving religiosity is to propose that there is a reasonable probability that the relationships reported are spurious; that is, the relationships may be attributable to other variables that have not been taken into account. This criticism is legitimate and some attempt should be made to eliminate plausible alternatives. One approach to the issue is to consider the relationships involving religious variables together with other variables that have been proposed as crucial to the explanation of international variation in homicide rates and the high rate in the United States. If the relationships endure when incorporated into such models, then the patterns observed gain credibility.
A particular brand of criminological theory called “institutional anomie theory” has dominated analysis of homicide rates among nations in recent years and should provide a strong contender for eliminating or explaining the patterns reported. Messner and Rosenfeld (1997, 2001) attribute variation in homicide among nations to variations in the strength of national investments that free citizens from market forces. Such investments are referred to as “decommodification,” and their analysis supports the argument that there is a negative relationship between such investments and homicide rates. In Crime and the American Dream, Messner and Rosenfeld apply their argument to the high rate of homicide in the United States. From this perspective, the United States has an exceptionally high rate as a product of stresses and strains stemming from “economic dominance,” a form of “institutional imbalance” that generates high rates of crime.
Two economic variables central to institutional anomie theory, welfare funding and income inequality, are significant bivariate correlates of logged homicide rates in this analysis. Consistent with Messner and Rosenfeld’s perspective, the greater the national investment in decommodification (based on welfare expenditures relative to GDP) the lower the homicide rate ( r= -0.351). Moreover, the greater the income inequality (based on a gini score) of a nation, the higher the homicide score ( r = +.546). Greater income equality and national investments in decommodification are associated with lower homicide scores.
Table 4 summarizes the results of a multiple regression analysis incorporating a score designated “passionate dualism” (PASSDUAL) combining z-scores for intensity, dualism (aGod/Devil ratio, and malevolence), the benevolence score, and the two economic variables. The findings continue to support the relevance of religious variables to homicide. The two religious variables persist as strong correlates of logged homicide rates, while the coefficients for the two economic variables are not statistically significant. The economic variables may have an indirect effect with religious variables acting as mediating mechanisms, or their association with homicide is spurious due to shared connections with religious variables. Regardless of the interpretation, the relationships involving the religious variables cannot be attributed to spurious shared associations with economic variables. (Table 4 about here)
Table 5 summarizes the results for a model regressing logged homicide rates on the two religious scores together with four other variables (Latin American nation, civil war, newly established government, and cultural diversity) found to correlate with variation in homicide rates in prior research (Jensen and Akers 2003). Each of the six variables is a significant independent covariate of homicide. The measure of passionate dualism is a strong positive correlate while the benevolence score is a strong negative correlate.
Categorization as a Latin American nation, the presence of civil war in a nation, a relatively new form of government, and cultural diversity are all positive correlates of homicide, explaining nearly 75 percent of the variation in logged homicide rates among 41 nations. In short, the relationships involving religious variables persist and are not spuriously attributable to other variables examined in prior literature. (Table 5 about here)
The United States
Paul focuses primarily on the high homicide rate and other selected ills characterizing the United States in a set of eighteen prosperous nations, attributing that unique position to a high level of religiosity. This approach can be badly misleading and a similar approach could be taken to highlight problems in more secular nations. For example, the Netherlands, Belgium, Denmark, Sweden, Germany and seven other nations have higher burglary rates than the United States (based on Interpol and United Nations data). The United States ranks ninth in cirrhosis death rates with at least four of the secular nations, including Japan, Denmark, France, and Germany exhibiting higher rates. The United States ranks thirteenth in suicide rates, seventh in estimates of daily consumption of narcotic drugs (Interpol estimates), and fourteenth in estimates of net annual alcohol consumption (Interpol estimates). In short, Paul’s analysis generates the “desired results” by selectively choosing the set of social problems to include to highlight the negative consequences of religion.
The more complex analysis in this study applies to the set of nations studied and does not allow a conclusive determination of the fundamental source of the high rate of homicide in the United States. The model reported in Table 4 persists when the United States is dropped from the nations in the analysis. The patterns found apply to variations among nations included in the World Values Surveys where data on homicide and other data are available. Those nations with the highest homicide rates tend to be those that can be characterized as high in religious passion or passionate dualism and the United States falls in that set. Moreover, that religious cosmology appears to be the most prominent shared characteristic of nations with high homicide rates.
However, the United States also has a high rate of income inequality, a relatively under-funded welfare safety net, and other characteristics that could explain the high homicide rate. Indeed, religious passion and dualism could be intimately intertwined with other characteristics of the United States. High levels of inequality and economic insecurity may encourage religious cosmologies and passions that facilitate homicide.
Conclusions and Discussion
There are obvious limitations to this analysis, and just as sociologists should not over-generalize about the positive effects of religion based on prior research, negative consequences have not been established conclusively. The data are cross-sectional and the temporal links among the variables studied have not been established. Moreover, the analysis is a secondary analysis of data that have already been collected which limits their applicability to more complex arguments.
Yet, this analysis is the first step towards a more meaningful specification of the complex links between religiosity and homicide rates at the ecological level using nations as units of analysis. Some of the results are remarkably consistent with Durkheim’s passion hypothesis about religion and homicide and are contrary to over-generalizations about religion as a barrier to crime. On the other hand, relatively secular nations do not have lower homicide rates than nations where people accept God and Heaven, but do not embrace their malevolent counterparts, the Devil and Hell.
Collective beliefs suggesting a relatively benevolent religious cosmos are negatively correlated with homicide when included in a regression analysis with more malevolent, dualist dimensions of the religious cosmos. These patterns were supported using nations as units of analysis, and should not be used to reach conclusions about the characteristics of individuals and their involvement in violence. Yet, the findings certainly have implications for reasonable speculation about homicide at other levels of analysis. The findings are consistent with Ellison, Burr, and McCall’s (2003) analysis of the strength of “Evangelical Protestantism” and city homicide rates. It seems quite reasonable to hypothesize that the evangelical movement encourages high levels of passion and moral and/or religious dualisms. It is plausible to propose that religious and moral dualisms may coincide with other forms of dualism at the individual level. As Luckenbill and Doyle (1989) argue, homicide is one outcome of situated transactions where honor is at stake with a narrow range of options for responding and heightened sensitivity to what might appear to be minor affronts. Whether called a “culture of violence” or a “code of the street” (Anderson 1999), disputes are easily triggered and there is little flexibility in acceptable responses. In short, other cultural or sub-cultural dualisms may help explain variation in behavior at the individual level. If a youth grows up in a world where there are rigid boundaries for attaining honor, a wide range of situations that are interpreted as disrespect, and limited cultural means for reestablishing honor, the range of situations generating interpersonal violence are enhanced.
Similarly, if there is only one appropriate way for a spouse to behave with dualist “macho” conceptions of the male as the lord of the household, then challenges to such authority may elicit passionate attempts to reestablish “moral order.” Although it may seem to be a huge step from dualist cosmologies to forms of interpersonal violence, there may be a great deal of asymmetry between dualisms that can operate to generate violence in the family, on the street, and elsewhere.
At a minimum, this research suggests that far more attention needs to be paid to moral and religious cosmologies in criminology, including their etiology and consequences at ecological and individual levels of analysis. The findings are significant for current controversies in sociological criminology as well. For example, institutional anomie theorists argue that it is a disparity between widely shared cultural aspirations for pecuniary success and realities of limited opportunities that generate high levels of homicide. In short, it is economic culture that contributes to a high rate of homicide unless ameliorated by state decommodification policies. Not only is there little or no support for the decommodification argument, but recent attempts to assess such economic values as a source of variation in homicide have failed to support that argument, whether applied to international variations or to the unusual rate of homicide in the United States (See Jensen 2003).
A more reasonable explanation for the high homicide rates would focus on religious and moral cosmologies. Indeed, it is reasonable to propose that variables such as inequality may have significant, but indirect, consequences for homicide by reinforcing dualistic moral cosmologies. High levels of inequality may be associated with high levels of us-versusthem views of the moral cosmos and tendencies to blame external forces for interpersonal problems. This paper does not establish the sources of such dualist cosmologies, but does point to promising avenues for further research.
It is interesting to consider the possibility that Messner and Rosenfeld may have located the United States in the wrong category in terms of their conceptions of “institutional imbalance.” They argue that within a social system “dominance by family and religion generate crimes in defense of the immediate, parochial social order such as vigilanteeism, hate crimes and violations of human rights.” Such offenses allegedly reflect a climate of “extreme moral vigilance” and “a strong sense of interpersonal obligations… restricted to those with whom they share particular social statuses or identities (2001: 156).” The extra-ordinary primacy accorded both the family and religion by U.S. respondents to World Value Survey items could justify categorizing the United States as parochially organized and dominated by strong in-group-out-group distinctions as well as extreme moral vigilance. Such characteristics are likely to coincide with familial, religious, and moral dualisms.
Moreover, these dualisms should be far more relevant to homicide than to instrumental offenses, an argument supported by the fact that the United States does not have an unusual rate of the very forms of instrumental crime that would reflect a dominance of economic values and institutions. As Zimring and Hawkins note “Crime is not the problem (1997).” Rather, it is lethal violence that distinguishes the United States from other industrial nations.
Finally, this analysis supports Stark=s proposal for a reconsideration of the ways in which religion affects moral order. He chides Durkheim for putting too much emphasis on ritual rather than “features of the Gods.” However, when Durkheim introduces the notion of “passion,” he appears to go beyond simple ritual. If the dimension of the religious cosmos captured by belief in God and Heaven is a form of public embracement of monotheism, then Stark=s argument that such features of the Gods encourage morality and discourage deviance is consistent with the findings reported above. On the other hand, findings that other dimensions of the religious cosmos are positive correlates of homicide are consistent with arguments that religion can “become evil.” Paul’s “first look” at popular religion as a positive correlate of homicide is partially correct. But, a “closer look” supports a more complex, but quite meaningful, set of findings that accord some validity to each perspective.
1. The author is grateful for support and encouragement for this project from the Center for Religion and Culture at
Vanderbilt University, Volney Gay and Douglas Knight, Directors.
2. The World Values Surveys were administered a third time in 2002, but the homicide data and other measures are from the
mid- to-late 1990s. Hence, this analysis is limited to the first two waves of surveys. The survey data can be found in the MicroCase data Archive available through Thomson Learning.
Anderson, E. 1999. Code of the Street: Decency, Violence, and the Moral Life of the Inner City. New York: W. W.
Norton and Company.
Baier, Colin J. and Bradley R. E. Wright. 2001. “’If You Love Me, Keep My Commandments’: A Meta-Analysis of the Effect
of Religion on Crime.” Journal of Research on Crime and Delinquency 38: 3-21.
Durkheim, E. 1951.. Suicide. Translated by J. A. Spaulding and G. Simpson. New York: The Free Press.
Ellison C.G., J. A. Burr, and P. McCall. 2003. “The Enduring Puzzle of Southern Homicide: Is Regional Religious Culture the Missing Piece?” A Homicide Studies 2003; 7: 326-352
Grasmick, Harold G.,Elizabeth Davenport, Mitchell B. Chamblin and Robert J.Bursik, Jr. 1992. AProtestant Fundamentalism
and the Retributive Doctrine of Punishment. Criminology 30: 21-45.
Gold, Martin. 1958. “Suicide, Homicide, and the Socialization
of Aggression.” American Journal of Sociology 63: 651-61.
Henry, Andrew and James Short. 1954. Suicide and Homicide: Some Economic, Huff-Corzine, L., J. Corzine and D.C. Moore. 1991. Deadly Connections: Culture, Poverty, and the Direction of Lethal Violence. Social Forces 69: 715-732.
Hunter, James Davison. 1991. Culture Wars. The Struggle to Define America. New York: Basic Books.
Jensen, Gary F. 2003. “Institutional Anomie and Societal Variations in Crime: A Critical Appraisal.”
International Journal of Sociology and Social Policy. Special Issue, Rebecca Block Editor.
_____ and R. L. Akers. 2003. “Taking Social Learning Global.”
Pp. 9-37 in R.L. Akers and G.F. Jensen (Eds.), Social
Learning and the Explanation of Crime. New Brunswick, New Jersey: Transaction Publishers.
Kimball, Charles. 2003. When Religion Becomes Evil. San Francisco: HarperCollins Publishers.
Lester, David. 1987. “Cross-national Correlations among Religion, Suicide and Homicide.” Sociology and Social
Research 71: 103-4.
Luckenbill,D.F. and D. P. Doyle. 1989. “Structural Position and Violence: Developing a Cultural Explanation.”
Criminology 27: 419-436.
Luper, M., P. J. Hopkinson and P. Kelly. 1988. “An Exploration of the Attributional Styles of Christian Fundamentalists and Authoritarians.” Journal of the Scientific Study of Religion 27: 389-98.
Messner S. F. and and R. Rosenfeld. 1997. “Political Restraint of the Market and Levels of Criminal Homicide:
A Cross-national Application of Institutional Anomie Theory.” Social Forces 75: 1393-1416.
Paul, G. S. 2005. “Cross-National Correlations of Quantifiable Societal Health with Popular Religiosity and Secularism
in the Prosperous Democracies, A First Look.” Journal of Religion and Society. Volume 7: Online.
Stark, Rodney. 2003. For the Glory of God. Princeton, New
Jersey: Princeton University Press.
Stark, Rodney and William Sims Bainbridge. 1996. Religion, Deviance, and Social Control. New York: Routledge.
Unnithan, N. Prabha, Lin Huff-Corzine, Jay Corzine and Hugh Whitt. 1994. The Currents of Lethal Violence: An
Integrated Model of Suicide and Homicide. Albany, New York: State University of New York Press.
World Health Organization. 2000. World Health Statistics
1997-1999, Online Edition.
TABLE 1: HOMICIDE AND SUICIDE RATES (NATURAL LOGARITHMS)
BY MEASURES OF RELIGIOSITY (WORLD VALUE SURVEY NATIONS)