Skip to main content
On the Unscientific Basis of the Crusade of Some Scientists against Religion
Jul 1, 2008

The voices as well as popularity of certain anti-religion scientists have increased lately. Such figures as evolutionary biologist Richard Dawkins1 and neuroscience researcher Samuel Harris2 enjoy fame and credit that their predecessors could only have dreamed of, and they owe most of their fame to the unyielding crusade they are waging against religion. The tone and severity of this anti-religious crusade is troubling to many, including to some atheists. What is also troubling is that the very methods these scientists use to understand and criticize religions are hardly scientific. Samuel Harris himself makes a distinction between scientific and religious thought as follows: “Pretending to know things you do not know is a great liability in science, and yet, it is the sine qua non of faith-based religion.” 3 Ironically, the new crusade against faith-based religions is based more on “faith” than on fact or reason. This is because the attitude of these scientists toward religions and their adherents is exactly this: pretending to know religion and religious people that they do not know. Thus, the crusaders of science fall prey to the very behavior they claim to be waging war on: dogmatic thinking. In this review, I particularly highlight four problems with the methodology of anti-religious scientists in studying religion: literalism, crimes against empirics, literature-free analysis, and unfair treatment of religion.

Fallacy one: Literalism

A critical fallacy in the arguments of “scientific crusaders” is that their understanding of religions relies on the most primitive way of approaching a text: literalism. Any serious and “scientific” inquiry into a text requires asking certain questions such as “Who wrote this text?” “To whom was it written?” “In which context was it written?” and “Why was it written?” Yet we rarely see the aforementioned scientists using this method when they approach religious texts. Samuel Harris, for example, simply takes certain verses in the Qur’an about killing certain non-Muslims at face value and pays no attention either to the context in which these verses were revealed or to the gigantic Islamic literature on the exegesis of these verses. As such, he simply turns a blind eye to all developments in hermeneutics and uses the most primitive method of text-reading.

Samuel Harris might not be interested in finding out what Muslim scholars have thought about killing innocent people; however, he has to be aware of the fact that it is not the individuals’ own interpretations but mostly scholars’ interpretations that have shaped Muslims’ opinions on any issue. An interesting point is that there is no precedent whatsoever for suicidal terrorism in traditional Islam before the late twentieth century. There has been a solid consensus among Muslim scholars for fourteen centuries on the sanctity of human life, because the Qur’an declares unequivocally that killing an innocent person unjustly is like killing all of mankind (Maidah 5:32). Thus, the critical question becomes: if something (such as terrorism) is “inherent” in a religion (such as Islam), why did that religion not encourage that thing for fourteen centuries but suddenly started to encourage it later? Unfortunately, scientific crusaders do not seem to be very interested in this kind of question.

Fallacy two: Crime against empirics

A second fallacy of scientist crusaders is that their understanding of religion’s influence on society is highly impressionistic and lacks any credible scientific inquiry or understanding. Samuel Harris’ argument about the violent nature of European Muslim youth and Richard Dawkins’ argument regarding oppression of women in Islamic societies are purely impressionistic inferences because they totally ignore other potential factors that can also influence juvenile violence and gender discrimination. The fact that Muslim youth in Europe are also disproportionately poorer and less educated than many other groups does not seem to undermine Harris’ sense of cognitive comfort. Yet any scientist with a basic knowledge of quantitative methods knows that we cannot analyze the influence of one factor on a dependent variable independent of other related factors. Statistics tells us that we will have nothing but biased results in such control-free analysis. Indeed, Scott Atran reports a study that reveals that “religious education is a negative predictor of Muslims entering European prisons.” 4 If a Muslim’s becoming more religious decreases the likelihood of his or her involvement in violence, then religion is a peaceful element. If this is the case, then the root causes of violence may be a combination of the very things which Harris avoids studying.

Dawkins’ argument about oppression of women in Muslim societies suffers from the same problem. He commits the same methodological error as Harris by excluding other potential explanatory factors from his analysis. Thus, his study cannot tell us whether it is Islam or something else (such as tradition, poverty, education, etc.) that distinguishes Muslim societies from the Western Christian societies regarding the oppression of women. Fortunately, we have some other (and more scientific) studies to examine the issue. A report by UNICEF in 2000 revealed that average domestic violence is essentially no different in Christian and Muslim societies. Below is the table from this report that gives the details about domestic violence rates in different Christian and Muslim countries. According to these survey results, average domestic violence against women was 25.5% in Western Christian societies (Canada, New Zealand, United Kingdom, and the United States); 30% in Muslim societies (Egypt, Israeli Arabs, and Tajikistan); and 38.3% in African Christian societies (Kenya, Uganda, Zimbabwe). The survey results indicate that domestic violence rates in Muslim societies were higher than they were in certain Christian societies but lower than they were in other Christian societies. As such, these figures solidly refute Dawkins’ impressionistic argument and rather suggest that something other than Islam is the root cause of domestic violence in Muslim societies.

Here, Dawkins can respond, “These data do not refute my arguments entirely, only the argument that Islam encourages more domestic violence than Christianity. But the argument that both of these religions encourage domestic violence remains valid and sound.” Yet, is this really the case? If one accepts Dawkins’ argument, one must also argue that increasing religiosity of a married Muslim or Christian man should increase domestic violence in his home. It is easy to check this. We can simply ask a random sample of large numbers of married Muslim and Christian women first to grade the religiosity of their husbands on a scale of 0–10, and then ask them how often their husbands behave violently at home. Although we do not have such a study on Muslims yet, to our surprise (and to Dawkins’ shame), we already have one study on Christians, which used exactly this method. In a study of nearly 1,000 New Zealand families, Fergusson et al. (1986) found that the self-reported religious attendance of parents (both fathers and mothers) is inversely related to their self-reported perpetration of domestic assaults. They reported a graded pattern: men and women who attend services once a month or more are least likely to report being involved in domestic violence, followed by those who attend less than monthly. On the other hand, men and women who never attend religious services are much more likely than their more religious counterparts to be involved in domestic violence.5The findings of this earlier study were later supported by another one, which employed a more objective measure of religiosity, on American couples as well. 6 (Although we have yet to prove it, this relationship between religiosity and domestic violence is very likely to hold for Muslim men as well. After all, the Prophet Muhammad, peace be upon him, said, “The best of you is the one who treats his wife in the best manner.” 7) Once again, contrary to Dawkins and Harris, we find that religion is an element of peace, not violence.

It is also worth mentioning that much less crime has been reported in religious schools (imam-hatip) than in secular schools in Turkey. Naci Akay, former Istanbul Head of National Education, said there have never been any activities reported in these schools that pose a threat to public safety. Likewise, Ibrahim Solmaz and Ahmet Gundogdu, chairmen of ONDER and Egitim-Bir-Sen respectively (two civil educational organizations) said parents prefer these schools mainly because they do not have high crime rates and students are much less likely to develop bad habits like drug abuse, alcoholism, or gambling. 8

The holy month of Ramadan in which Muslims observe day-long fast and additional devotion, is also a time of security for the entire Muslim world: crime rates plummet in this month, and sometimes fall to zero.

Fallacy three: Avoiding the literature

Another major problem for the crusaders of science is that they completely ignore the literature on the root causes of conflict and violence. None of the major theories on domestic or international conflict emphasize religion as a central factor, let alone singling it out as the root cause of violence. The major theories explain violence as resulting from peoples’ ambition for power, from liberalism with its inevitable conflict of interests and peoples’ inability to solve them in cooperative ways, and from class conflicts. To the subscribers to any of these theories, the argument that we could minimize or eliminate violence in this world if we could substitute scientific inquiry for religious faith is simply laughable. For them, religion is at most a secondary factor and therefore its elimination does not help to address the root causes of violence. As such, scientific crusaders’ emphasis on religion as a cause of violence is misleading and unhelpful because it moves our attention away from other more influential factors which could be more successfully addressed. In this respect, the anti-religion arguments of Harris, Dawkins, and the like resemble the wishful thinking of the nineteenth-century positivists. Ironically, ignoring the real forces at work neither makes the world more peaceful nor serves these scientists’ very goal of converting people to “scientism”. The new rise of religious faith since the second half of the twentieth century indicates that science and reason alone have failed to provide sufficient answers to peoples’ metaphysical questions or solutions to their moral problems. Hence, the return to religion!

Fallacy four: Imbalanced sheet

Last but not least, we have been taught since elementary school that we should never come to a judgment on any issue without first evaluating its pros and cons. What we see in the books of Dawkins and Harris, however, are only questionably negative aspects of religions and religious people. We see no appreciation whatsoever of the virtues of religions and their positive influences on their adherents. To start with one of the most obvious, is it not intriguing enough a question for Darwinian Dawkins to ask what prevented the Muslim Moors of Spain from following the “survival of the fittest” rule and simply annihilating the Jews and Christians of the Iberian Peninsula? How rational or scientific is it to simply avoid studies that have found that “teens from intact families with frequent religious attendance were the least likely to have ever shoplifted,” 9 “adolescents’ religiosity was related to a decreased likelihood of delinquent behavior,” 10 “increasing individual religiosity decreased tax fraud acceptability,” 11 and “regular religious attendance was inversely associated with the perpetration of domestic violence,” 12 to name but a few? Interestingly, the books of scientific crusaders have no place for such virtuous consequences of believing in God. It seems not only religious people have dogmas!

To conclude, certain scientists’ arguments against religions in general and Islam in particular lack scientific basis and thus do not merit much scholarly attention. There might be some valid grounds to criticize religions and their adherents as they are, but, believers or not, we all deserve more holistic, more scientific, and less hostile evaluations of religions. If atheist scientists want to wage a crusade against dogma, their own would be a good place to start.

Kaan Kerem has a PhD in political science. He is a freelance writer on philosophy and scientific thought.


1. Author of The God Delusion, Houghton Mifflin, 2006.

2. Author of The End of Faith, W.W. Norton & Company, 2004.



5. Ferguson, David M., L. John Horwood, Kathryn L. Kershaw and Frederick T. Shannon (1986). “Factors Associated with Reports of Wife Assault in New Zealand.” Journal of Marriage and the Family, 46: 663–70.

6. Ellison, Christopher G. and Kristin L. Anderson (2001). “Religious Involvement and Domestic Violence among U.S. Couples.” Journal for the Scientific Study of Religion, 40(2):269–86.

7. Ibn Maja, Section: Marriage, Hadith # 50.


9. Fagan, Patrick (2006). A Portrait of Family and Religion in America: Key Outcomes for the Common Good. Washington, D.C.: The Heritage Foundation.

10. Regnerus, Mark D. (2003). “Linked Lives, Faith and Behavior: Intergenerational Religious Influence on Adolescent Delinquency.” Journal for the Scientific Study of Religion, 42(2): 189–203.

11. Steven Stack and Augustine Kposowa (2006). “The Effect of Religiosity on Tax Fraud Acceptability: A Cross-National Analysis.” Journal for the Scientific Study of Religion, 45(3):325–51.

12. See note 7.