When it comes to discussing suicide bombers, the controversial topic of religion is never far behind. Scholars and pundits have proposed several theories to explain why people would sacrifice their lives to take those of others, and conjectures about religious views seem easy to defend. After all, anthropologist Scott Atran estimated that since 2000, 70% of suicide attacks have been carried out by religious groups, and Islamic ones in particular.
But for all the speculation, very few people have examined the supposed link between religion and suicide attacks with an objective scientific eye. Enter Jeremy Ginges from the New School for Social Research in New York. He has used four related studies to show that there is indeed a link between religion and support for suicide attacks, but it's a complicated one.
Ginges studied a wide variety of religious people from various cultures and faiths - from Palestinian Muslims to Israeli Jews, and from British Protestants to Indian Hindus. Across the board, Ginges found that a person's stance on martyrdom had little to do with their religious devotion or to any particular religious belief. Instead, it was the collective side of religion that affected their stance - those who frequently took part in religious rituals and services, were most likely to support martyrdom.
Various commentators have suggested that religious devotion makes it easier for people to buy into the ethos of suicide attacks because some religious beliefs denigrate those of other faiths, promise rewards in the afterlife or glorify the notion of martyrdom. According to Ginges, the advocates of this idea, Richard Dawkins among them, tend to bias their attention towards the more violent aspects of religious traditions or texts, in a fairly simplistic way.
An alternative idea says that the social side of religion is the more powerful influence. During religious rituals such as church or mosque services, large groups of people move or speak as one, invoking a powerful sense of shared identity. By strengthening bonds within a group, these rituals can augment a person's loyalty to that community, often to the exclusion of those outside it. Suicide attacks, which sacrifice a person's life for the sake of the collective cause, could be viewed as the extreme dark side of this cliquey behaviour.
Ginges, together with Ian Hansen and Ara Norenzayan, carried out four studies to distinguish between these two theories and they've consistently found support for the latter, across a variety of religions.
In the first, they analysed a survey of 1,151 Palestinian Muslims in 1999, which looked at their religious beliefs and behaviour, and whether they support "martyrdom attacks". Their answer had little to do with their religious devotion, as measured by how often they prayed (Ginges confirmed that people who said their religion was particularly important to them were also those who prayed most frequently). However, their attendance at mosque services did predict their support for suicide attacks. Those who attended daily were more than twice as likely to back martyrdom than those who paid fewer visits.
Seven years later and little had changed. In a more sophisticated survey of 719 Palestinian Muslim students, Ginges found the same trends. This time, each student was asked whether Islam forbids, allows, encourages or requires Muslims to killing themselves in order to kill their enemies. Again, those who attended mosque services daily were over three times more likely to believe that Islam "requires" suicide attacks than less frequent visitors. But again, devotion to Islam, as indicated through prayer, had nothing to do with their support for martyrdom.
What is it about mosque services that influenced the students' views? It's possible that some clerics use services to spread anti-Israeli propaganda, but Ginges saw the same effect whether or not the students empathised with Israelis or had dehumanised them. Likewise, it didn't matter if the students supported Hamas or the PIJ, groups which most Palestinian suicide attackers have belonged to.
Together, these two surveys support the idea that it's the group ritual aspect of religion that affects support for suicide attacks, rather than religious belief by itself. But Ginges wanted more - he craved experimental support for his idea, and he wanted to see if it applied to people from other faiths. To do that, he interviewed 198 Jewish settlers living in the West Bank and Gaza. And this time, he simply put the idea of prayer or synagogues into their heads.
During the interview, he asked them either how often they prayed or how often they visited a synagogue, and later about their support for a famous suicide attack - the Cave of the Patriarchs Massacre (Israeli suicide attacks are relatively rare, so Ginges had to focus on a single incident).
The results were clear. About a quarter (23%) of the interviewees who had synagogues on their minds described the attack as "extremely heroic". However, just 6% of those who were asked about prayer did so, and an in-between 15% agreed when neither thought was planted. So priming people's brains with the very idea of a religious service made them more likely to support suicide attacks than usual, and contemplating prayer actually dulled their fervour.
For his final study, Ginges went global. He analysed surveys conducted by the BBC, looking at the beliefs of representative people from six religious groups - Indonesian Muslims, Mexican Catholics, British Protestants, Russian Orthodox Christians, Israeli Jews, and Indian Hindus. Once again, he confirmed that people who prayed more frequently were most strongly devoted to their respective creeds, and prayer was a better indicator of strength of belief than actual attendance at religious services.
Obviously, suicide attacks are rare or non-existent in many of these faiths, so Ginges looked at whether were "willing to die for [their] God/beliefs" and whether they "blame people of other religions for much of the trouble in this world". As predicted, a person's inclination for prayer had nothing to do with their answers to these questions. But overall, people were twice as likely to agree if they regularly attended a mosque, church, synagogue or temple.
This effect remained even after accounting for the different demographics and economics of the six countries, but it did vary from group to group. It was only statistically significant (unlikely to be a fluke result) for Indian Hindus, Russian Orthodox Christians and Israeli Jews. However, Ginges warns against overinterpreting these differences - obviously the six samples differed in many ways. The important point was that all of them showed a similar trend.
Together, these four studies - three survey analyses and one experiment - contradict the idea that religious belief and devotion in themselves are the driving force behind the suicide bombing mindset. Nor is this mindset exclusive to Islam, as the third and fourth experiments show. Instead, it seems that the link between religion and suicide attacks is more to do with collective rituals. Ginges's theory is that these rituals strengthen an individual's loyalty to a community, but risk hardening their hearts against outsiders. That's certainly a reasonable interpretation but it's worth noting that religious services are complicated affairs, which have many sides to them besides group behaviour. People who worship en masse may get very different perspectives on their faiths that those who pray alone don't, and this study doesn't really take that into account.
Regardless, the results are certainly interesting and consistent; the big danger is in drawing simplistic interpretations from them. The research isn't saying that people who attend religious services will become suicide bombers, or even that they're likely to. It simply implies that they are more prone to supporting the ideals of martyrdom than people who don't.
Likewise, it would be both naive and reactionary to use this study to conclude that religious services are evil and harmful. Ginges is very clear that they have a strong positive side and can build strong communities. Only rarely, and in specific political and cultural environments, does the same group-building ethos turn sour and twists into the foundation for suicide attacks.
And obviously, there are many ways for secular groups to create the same mentality, including youth groups and military parades. The Tamil Tigers of Sri Lanka are a classic example of a non-religious group that uses such methods to recruit support for suicide attacks. Indeed, it would be very interesting to see if similar studies in secular commuities would produce the same results when thinking about ideology, group activities and martyrdom.
And as a final caveat, obviously there are many other factors that affect support for suicide attacks, or intent to carry them out. Some will be religious, as examined through this study, others will be economic or political. Either way, it can only be a good thing to put this most divisive of topics under the scientific limelight.
Reference: Psychological Science in press. 2009. Vol 20, 224-230.
More on belief:
- Lacking control drives false conclusions, conspiracy theories and superstitions
- In conflicts over beliefs and values, symbolic gestures matter more than reason or money
Wow, that's a really interesting study. It makes a lot of sense, though. I mean, think about cults. They use a highly social context to brainwash and control their members. Just believing isn't enough - they have a high level of structure behind it, too. It just goes to show how powerful culture and social context is on how a person thinks and even behaves. Nice article!
How do you get in press articles from PsySci? I'm a PhD. student of ecology from Czech republic and I sometimes write popular science articles to the media. I thing, this is a good topic, but if i want do write about it, I need original manuscript.
So Ginges expects me to believe that attendance to religious services is somehow disassociated with his definition of "religious devotion" and not used as a metric for its measurement while the act of prayer is a form of this type of devotion? How convenient for his theory. Last I heard, people usually went to religious services specifically for the purpose of engaging in the act of PRAYER. weak.
Religious devotion was measured by asking people a variety of questions about whether they had ever doubted their faith, whether they had always been religious and their views on creation myths, divine intervention, the afterlife and so on.
Prayer frequency was more strongly related to "religious devotion" than attendance at religious services was. As noted in the post. From the fourth experiment:
Prayer was again a stronger predictor of the religious-devotion
index than was attendance at collective religious services.
Sorry Ed but hi has a point. In my opinion devotion is shown by actually putting your money where your mouth is. Attending services involves sacrificing your time for your religion where having doubted your faith, views on creation myths etc. are economically "cheap".
In other words anyone can claim unshakeable faith in their religion but it takes someone sincerely devout and dedicated to get his arse off his bed on a cold Sunday morning and trod down to the local temple.
Words are cheap, economists have learned this a long long time ago.
I consider myself a religious person, and I have a huge problem with this study - namely that it seems to lump all religion together. Religion means different things to different people. Saying religion is the problem is like saying government is the problem. There are many different types of both, and many have completely different philosophies. The religious group I belong to emphasizes respect for others and personal growth. We have little in the way of rote ceremony.
It seems to me that the problem here, as mentioned in your post, is the emphasis on ideology, rather than religion. I've met hard core communists that were just as scary as the hard core religious fundamentalists I've encountered. It is true that many religions place an emphasis on an authoritarian ideology, but to say that all religion is therefore bad is like saying that all forms of government are bad because absolute monarchy is an abusive form of government.
Edward - No one's saying that "all religion is therefore bad". The researchers had a valid research question - given that the majority of suicide attacks are carried out by religious groups, what aspects of religion associate with support for suicide attacks? So yes, it may be ideology or group activities in general, but the study only look at these factors within the context of religion. They are very clear about the limitations of their study, and the need to study similar processes in secular groups.
From reading the paper, I get no sense of them criticising religion in general or placing value judgments on the results. They're not saying that religion is bad and I don't think you should read that into either the study or my summary of it. You'll note that I specifically say:
Likewise, it would be both naive and reactionary to use this study to conclude that religious services are evil and harmful.
Miguel - A good point, well made. Although you could equally argue that some people who attend services do so because they feel obliged to, are dragged along, want to be seen to be attending, and so on.
For the record, I am more than happy to hear criticisms of the study - I am, after all, just reporting it. I do however have a natural antipathy against snarky whines especially by anonymous commenters (not you, Miguel - your comment was a clear example of how to debate with both substance and graciousness).
Hm... activation of religious cognitive pathways in a social context increases chances of violent social action from the religious framework.
That actually makes a hair of sense from a broader perspective.
Just thinking out loud here. I wonder if the effect works for taking actions in general, destructive or not -- that is, if people participate in a social group that bonds around certain behaviors and belief systems, are they more likely to act than people who claim to hold similar beliefs but don't belong to a group?
I know people who volunteer at soup kitchens, volunteer for Habitat for Humanity, take in foster kids, and set up medical clinics to treat the homeless. These people all attend religious services regularly, and they see their charitable actions as one way they live out their religious commitments. If suicide bombers are one side of the coin, is founding hospitals and finding housing for homeless people the other?
Very interesting studies. I think it would be illuminating to repeat the study while also asking questions about helping others of different communities. For example, for people who frequently attend services, would they consider it a responsibility to help people whose house had burned down? People of a different religion whose house had burned down? A different religious institution that had burned down? And how would that compare to the other answers. In other words, how does the sense of solidarity work in mustering good works for other communities or does it?
I was surprised that just thinking about going to synagogue had an effect. I wonder if it would for people who had never attended synagogue frequently, or whether it evokes the us/them sense only for people who do.
Is there actually such a word as "Islamist"? I mean, I know you can find it in any given morning's copy of the Metro on the tube, but that doesn't make it a real word.
Thanks for highlighting this study for us, not sure about most of us but certainly some us, myself included, would never have found out about it if people like you weren't blogging it. To me this is the beauty of blogging, we who do not work in the field of science but are curious and interested are given the privilege of being able to read and share our views with the ones who are actively working to expand our knowledge. It kin'a democratises science... a little bit.
You have a very good point on attendance. No doubt, it can also be deceitful as an indicator since so many other factors influence it, peer pressure, family influence, society's rules, etc. I still think it is possibly a good indicator of commitment provided you can stabilise the influencing factors and I don't know if this was done in the studies mentioned.
Having said this, I think the study is still very valid and poses some very interesting questions... but more research is needed, something both the authors and you in your report fully acknowledge. Some very interesting comments too, Anna K I think is very close to something, maybe seeing our beliefs validated by peers is a very strong mechanism for action?
Keep up the good work!
I'd be interested to know if the studies cited included women, and if so, if the results differed for men and women. News reports suggest that women become suicide bombers for different reasons than men do:
I loved Anna's and Lilian's suggestions for further experiments too.
Ian - ahem. I don't know what you're talking about. ;-)
Miguel - I wholeheartedly agree about the benefit of blogging, in putting curious people in direct contact with knowledgeable ones. It's brilliant. You'll see in these comment threads that from time to time, the scientists themselves crop up to answer questions.
DRK - Good question, but the samples in all four studies included a pretty equal split of men and women, and there weren't any significant differences in their behaviour.
Just thought I'd offer another angle for discussion. Since the vast majority of suicide bombing is done by Muslims, I think it's worth examining the cultural perspective. When looking at the motivation behind suicide bombing, it can help to look at its purpose. By and large the Western way of warfare is attritional, meaning that offensive and defensive actions are taken with resources in mind. Troops, cash, weapons, logistics, morale: all resources that can be taken into consideration in attritional calculations. On the other hand, the 'Eastern' (read Persian, Muslim, Arab as relevant over the centuries) is less about attrition, and more about aesthetics. Generally, suicide bombings are about graphic, emotive images, rather than killing as many people as possible. They make statements. In that way, I suppose you could say they're a perverse form of art. Not science perhaps, but it does offer an interesting insight.
Actually, this is fairly consistent with most studies about religious belief vs practice that I know about: religious belief doesn't seem to have many measurable effects other than the assertion of beliefs. Attendance is correlated with various outcomes in physical and mental health, social connections, etc.
... It was only statistically significant (unlikely to be a fluke result) for Indian Hindus, ...
Most Hindu temple visits consist of solitary prayer (and offering, etc.) in front of the idol within. In fact visits to the more popular temples can consist of standing in line waiting your turn for hours, after which you get only a few minutes inside. Visiting temple is normally not a communal or group activity.
That said, I don't know if this weakens their conclusions because they failed to take this into account (did they?) or strengthens it as there have been relatively few religiously motivated Hindu suicide killers.
To address Miguel's previous point, whether or not talk is cheap, the primary point is the separation of constructs. It seems like the devotion-index was just a separate construct made simply as something to distinguish itself from the... well, I don't know what the author called it (perhaps I missed it in the summary), but the martyrdom-index or what-have-you (the construct under question by the series of studies). The validity of the devotion index construct is not so much important as long as there is a solid distinction between the measurable outcomes of prayer behavior and ritual behavior.
It seems to me that if attendance at mosque, synagogue, church etc isn't correlating with religious belief as such it presumably shows something else. It is at least plausible that it is measuring the strength with which people are identifying with the group they are part of. So this may be a case where religion is acting to shore up group identity and, along with this, identification of others as an outgroup. Which makes these others valid targets. My reading of Dawkins suggests that he has been saying this for a long time and this would appear to confirm his views in this regard.
Hi Ed. Interesting article! Obviously, the degree of someones religious devotion will be hard to measure by mere means of asking people questions (I'm thinking some kind of neuroscientific method would be more objective, but that's pretty futuristic I suppose). Needless to say, asking the right kind of questions is most important. However, IMHO the questions that have been asked in the survey do a poor job since they suggest a certain correlation between the degree of devotion on the one hand; and previous doubts, previous lack of belief etc. on the other. It may sound obvious at first, but is it really a safe assumption? I don't see how. If you doubted, and you lost your faith, then you're probably not in the survey at all. If instead you reaffirmed your beliefs, I could think of a couple of psychological reasons why the beliefs might come back even stronger than before. The same goes for people with no religious upbringing, or a different one; where does the assumption come from, that they are less devout than those with a religious upbringing?
Anyhow, even though the idea of looking at "religious devotion" (prayer) vs. "religious community" is interesting, I really fail to see how one can ignore which religious claims & teachings any religious devotion (including the sort demonstrated in community ceremonies) is aimed towards. Of course it matters whether it's a Muslim or a Buddhist ceremony.
Why was 2000 the cut-off point for the 70% statistic? If they go back to the eighties, then they'd have more data from Ireland. With that in mind, the better negative control wouldn't be other religions but rather non-religious communes/cults.
Ed - in response to your response to my response....
In my opinion the way in which the question is asked and the labeling of the variables (e.g. "religious devotion") indicate a bias on the part of the authors of the study. As I indicated in my previous reply - "religion" and "religious" are not well defined terms and mean different things to different people. There are a number of people who equate being "religious" with being a theistic ideologue. It is be my hypothesis, based on personal observation, that it is the ideologue aspect of the equation that leads to extreme actions, not the theistic part. Many religions do, unfortunately, encourage people towards being ideologues. Thus, a positive correlation is expected between being an ideologue and being "religious." Also, it also seems to me that ideologues seek simple answers to questions that have complex answers or no answer, and many religions are all too willing to provide such simple answers. As a result, these religions attract ideologues. As Anna and Lilian point out, sometimes people's ideologies can lead them to do good things as well as bad.
It comes down to correlation vs. causation. I agree that there is a correlation here, but I think the cause is not religion. We don't see many suicide bombers who come from comfortable middle class families in the U.S.A., however religious they are.
It seems obvious to me that religious devotion does not necessarily correlate with church attendance - based on my own experience as well as that of many of the people I know and who I grew up with. People who I knew, including my family, went to church mostly so that others would see them as conforming to community norms, as being good people. And it seemed common in my community.
Religiosity can be a personal rather than a communal experience and I suspect that many truly religious people see it that way and recognize the tribal nature of many religious services, and are probably somewhat repelled. I think most churches realize they must provide that tribal emotional hit in order to fill the collection plate. I think this could be proven (or disproved) by studying the types of services churches provide (along that tribal-emotional vs personal experience dimension) as compared to overall attendance and how much money they bring in. Who could watch an evangelical TV broadcast and not see the tribal nature of it?
Another way to read the study results is that tribal behavior (participating in structured rituals with other tribal members on a regular basis) correlates with the willingness to sacrifice one's personal well-being for the sake of the tribe. i.e. those who seek tribal emotional fulfillment are likely to have tribal personalities and behave in tribal ways. Is that so mysterious?
Perhaps including this non politically correct factor by name as a variable did not occur to the author for some reason. But all the factors he describes can be seen as tribal identity behavior.
The willingness to sacrifice one's personal well-being for the sake of the tribe is probably especially likely when tribal enemies are identified and denigrated during those tribal ceremonies and rituals. Those who then sacrifice their personal well-being for the tribe by attacking those enemies are virtually guaranteed tribal admiration (as martyrs for example). Add a chance to spend eternity with the namesake of one's tribal religion and all your friends and family who have died and you have an open invitation for endless death and destruction - if church leaders wish to lead the tribe in that direction rather than say, missionary work or healing lepers.
Finally Ed, you say, "And as a final caveat, obviously there are many other factors that affect support for suicide attacks, or intent to carry them out. Some will be religious, as examined through this study, others will be economic or political. Either way, it can only be a good thing to put this most divisive of topics under the scientific limelight."
As usual the psychologists ignore the more obvious emotional motivations and search for complex cognitive answers. Why shouldn't they include tribal emotional susceptibility along with religious, economic and political?
now, how do you juxtapose the above with this http://bhascience.blogspot.com/2009/01/how-does-religion-prevent-suicid… . Does religious service reduce suicidal ideation or does it cause suicide bombing ideation?
Very interesting article. Thanks. And a lot of interesting well thought out replies. Many of which would make fascinating future research.
One slight curiosity for me is the definition of Martyrdom. Perhaps due to a Catholic upbringing my predominant image of this is almost pacifist- eaten by lions for refusing to give up your faith. Or "Tale of Two Cities"- sacrificing for another's life. Suicide bombing is a rather different interpretation of Martyrdom.
Not trying to make a point about Catholicism, or criticising the article, just a curious comment about the meaning of the word.
Responding to Sandeep (and perhaps Richard a bit) . .
The article at Epiphenom seems to offer an enigma. If religious Canadians receive less social support in their lives than non-religious Canadians, then why does the effect of controlling for religious support in that cohort double the effective outcome - and in the opposite way one would expect?
Perhaps the social support aspect of religiosity is not a dominant variable. That would still leave open my premise that it is the tribal identity emotions provided by religious practice that has a dominant effect on suicide.
If such an effect were to evolve in human minds I assume it would operate like all such effects to increase the probability that carriers of the genes that expressed this trait would reproduce more successfully.
I would suggest that Richard is right, suicide is not the same thing as risking your life (or giving your life) to kill an enemy. Both behaviors can be selected for and are beneficial in the proliferation of those genes through a population.
Not killing yourself (not ending the chance for the reproduction of your genes) serves the same end as risking or killing one's self in the service of destroying a mortal threat to all those in your tribe whose genes you share to some extent. Both these behaviors would have the net effect of increasing of the survival of those genes and the tribe that carries them.
Further, I'd suggest that spirituality (the recognition or experience of mysterious forces that control one's life) is somewhat different from and even opposite in many ways to tribal emotion - which is more an emotional feedback loop that can emerge from identity membership in a specific group (one's tribe) especially when that tribe is under threat (aren't we all). This can be seen in an extreme example as mob behavior.
I do think spirituality can be molded and harnessed to support and even amplify tribal emotions in many cases though. I think this is especially true in cultures where the religion is intimately tied to the political leadership and are often one and the same.
Canada is probably not the place I would look for good examples of this except perhaps at the hockey rink. ;-)
To clarify above, when I say "Both behaviors can be selected for and are beneficial in the proliferation of those genes through a population."
I meant: Both not killing yourself (avoiding suicide) and risking your life (or giving your life) to kill an enemy - can be selected for and are beneficial in the proliferation of those genes through a population.
I don't know if this will work but here's a cartoon quite on topic:
If it doesn't appear just copy this:
. . into your browser's "go to" field and click ENTER.
It took me a while to get it finished, but I thought this post warranted a proper discussion beyond a mere comment... which can be found here. Not to overly self-promote, but people are welcome to check it out.
Whilst this was an interesting article, based upon your reporting I get the feeling these guys rushed to print prematurely.
As well as the arguments which have been presented here regarding the "religiosity" index potentially being false, there are other issues too. For example, they may really be picking up a simple difference in personality types, with people who are more likely to suicide bomb also being the type of people for whom going to church is also something they like to do - a type I error (false positive). Church going may be a spandrel, associated with rather more explosive behaviours.
Another possibility is that they would have gotten a much better relationship had they used biblical knowledge / adherence rather than church attendance, which may simply be a proxy of biblical knowledge / attendance. People who do not have the negative aspects of religion reinforced in a regular manner may be more likely to be selective about their interpretation of said holy texts (which would correlate with the much higher rates of church attendance in Muslim countries).
B+ - could do better.
I think Ginges is asking the wrong questions. When we come right down to it, all extremist actions taken out by religious fanatics are driven by a political agenda. Why aren't the political motivations of those religious institutions accounted for in Ginges' study?
Christianity is the most violent religion. The most basic goal of the faith is the end of the world. That's pretty disgusting. They are rooting for the end of all humanity.
Many of the hardline Jewish settlers have made themselves believe that Baruch Goldstein massacred the Muslims at the mosque because a pogrom was imminent from the members of that mosque with weapons they had stored there. They point to the fact that no children were (there to be) killed. Others believe a conspiracy theory that says the only people who died there were trampled to death and Goldstein shot some people in the legs. So they've made it easier in their minds to say "heroic."
"We don't see many suicide bombers who come from comfortable middle class families in the U.S.A., however religious they are."
The 7/7 bombers in the UK were from comfortable families - they were not destitute Bangladeshi immigrants or the like, the ringleader was a teacher which is unarguably a middle-class profession - and Osama bin Laden is an independently wealthy member of one of the richest families in the Gulf region. The morons who failed to blow up Glasgow Airport in 2007 and died instead were highly qualified doctors who lived a privileged life in the UK. One of them was a brain surgeon.
If I recall rightly, Mohammed Atta, the 9/11 bomber, wasn't exactly on the breadline either. Ayaan Hirsi Ali isn't a fan of this Marxist interpretation of suicide bombing where it's just another aspect of the 'opium of the people' and the bomber is somehow a victim. Having rescued numerous members of her family from starving to death in a refugee camp full of rapists, she has seen a thing or two, and in her book 'Infidel' she stated that the utterly poor care only about where their next meal is coming from. They are not walking around with the intent to forment complex political plots against 'the west', contrary to popular belief. The evidence from my own country, where the silver-spoon brigade have strapped on semtex shirts, would appear to back her observations up.