Concept

Religion and Violence

Intro

The objection is one of the most widely repeated claims in popular atheism: religion is the leading cause of violence in human history, more wars have been fought in the name of God than for any other reason, and removing religion would remove the principal engine of human bloodshed.

It is worth taking seriously, because the underlying moral concern is legitimate and because the historical record contains real Christian atrocities that should not be minimised. But the claim as stated is false, and it fails in three independent ways.

It fails on the numbers. Charles Phillips and Alan Axelrod's Encyclopedia of Wars (2004) catalogues 1,763 conflicts across recorded history and classifies roughly 121 as religious, about seven percent. Even the broadest classifications keep the figure well under a fifth. Whatever the leading cause of war has been, it has not been religion.

It fails on the twentieth century. The bloodiest century on record was dominated by regimes that were explicitly secular and in several cases officially atheist. This does not show that atheism causes mass killing, and the Christian should not make that argument, because it is the mirror image of the fallacy being answered. What it does show is that removing religion does not remove the phenomenon, which is precisely what the objection predicts it would.

It fails on definitions, and this is the deepest problem. The claim requires a clean separation between religious motives and political, economic, ethnic and national ones. That separation does not survive contact with the cases. The Thirty Years War is the standard example of a war of religion, and it ended with Catholic France funding Protestant Sweden against Catholic Habsburgs, because dynastic interest outranked confession. Northern Ireland's conflict was territorial and ethno-national, with confessional labels marking the communities rather than supplying the grievance.

Quick reply line: "Take the largest catalogue of wars we have. Of 1,763 conflicts, about 121 are classified as religious. That is seven percent. Then look at the deadliest century in history, which was run by explicitly secular states. I am not saying atheism causes atrocity, because that would be the same bad argument in reverse. I am saying the variable you are pointing at is not the one doing the work."

◎ See this page's graph and discover linked topics →

In full

The objection appears in Christopher Hitchens (god is not Great: How Religion Poisons Everything, 2007), Richard Dawkins (The God Delusion, 2006, chapter 8), and Sam Harris (The End of Faith, 2004), and circulates in popular form as a settled historical fact.

The academic literature is considerably less confident. William Cavanaugh's The Myth of Religious Violence (Oxford, 2009) is the most important single treatment and argues that the category of religious violence is incoherent as usually deployed. Karen Armstrong's Fields of Blood (2014) reaches a similar conclusion from a different direction.

The definitional problem

Cavanaugh's argument is the one worth understanding properly, because it reframes the dispute rather than trading atrocity lists.

His case runs as follows. The claim that religion causes violence requires "religion" to name a distinct category of human activity, separable from politics, economics and culture, and possessing properties, absolutism, divisiveness, irrationality, that the secular does not. No workable definition delivers this. Definitions narrow enough to exclude nationalism and Marxism also exclude Confucianism and some forms of Buddhism. Definitions broad enough to include those include nationalism, Marxism and consumerism, at which point the thesis becomes the trivial claim that strongly held commitments can motivate violence.

He then argues that the religious and secular distinction is itself a product of early modern Europe, constructed as the state consolidated power against the church. Its function was to relocate ultimate loyalty from the church to the nation-state. On this reading, the myth of religious violence performs a political job: it marks certain devotions as dangerous and irrational while exempting the devotion that has actually commanded the most killing in the modern period, loyalty to the nation.

The evidence for this is the asymmetry in how deaths are counted. Nobody describes conscription and mass mobilisation for a national cause as violence caused by patriotism, though the body count of twentieth-century nationalism dwarfs that of the wars of religion.

The historical record, honestly stated

Christian apologetics is not served by minimising, and the numbers are better than the popular impression in any case.

The Crusades. Two centuries of campaigns beginning in 1095, with real atrocities, notably the sack of Jerusalem in 1099 and the sack of Constantinople in 1204, the latter against fellow Christians and condemned by the Pope who called the crusade. Recent scholarship, principally Jonathan Riley-Smith and Thomas Madden, establishes two correctives: the campaigns were a response to four centuries of Islamic conquest of formerly Christian territory including Syria, Egypt, North Africa and Spain, and crusading was ruinously expensive for participants, which undercuts the older account of them as colonial land-grabs.

The Inquisition. The number is the corrective here. The Spanish Inquisition operated for roughly 350 years, and the systematic archival work led by Gustav Henningsen and Jaime Contreras, examining approximately 44,000 cases from 1540 to 1700, indicates executions in the range of one to two percent of those tried, with modern estimates of total executions across the entire period in the low thousands. The historical Inquisition was also, in several respects, a more procedurally careful court than the secular courts of the same period, which is why defendants occasionally blasphemed deliberately to have their cases transferred to it. See Inquisition.

The wars of religion. Real, devastating, and considerably more political than the label implies, as the French subsidy of Protestant Sweden illustrates.

The witch trials. Roughly 40,000 to 60,000 executions across Europe over three centuries, mostly by secular courts, and mostly in the early modern period rather than the medieval one. The Spanish Inquisition was notably skeptical of witchcraft accusations, and its 1614 instructions effectively ended witch prosecution in Spain.

None of this is offered as exoneration. It is offered because inflated figures are the objection's main rhetorical asset, and correcting them removes it.

The reciprocal concession

The productive structure here grants the objector's real point and collects a larger one in return.

Grant fully: Christians have committed atrocities, have used Scripture to justify them, and have done so in the name of Christ. The Church has blessed wars it should have opposed. This is not a matter of a few bad actors, and pretending otherwise is dishonest.

Collect in return: the standard by which those acts are condemned is the standard Christianity supplies. The Crusader who killed civilians violated the Sermon on the Mount; he did not follow it. The slaveholder quoting Scripture was answered by abolitionists quoting Scripture, and the abolitionists had the better reading. There is no comparable internal standard by which a materialist condemns the strong exploiting the weak, other than one borrowed from the tradition being criticised. See Stealing from God Argument and Divine Command Theory.

This is why the objection has the shape it does. It is an accusation of hypocrisy, and hypocrisy is a charge that can only be brought against someone who has a standard. The atheist arguing that Christians failed to love their enemies is arguing that they should have obeyed Christ more closely, not less.

What Christianity itself says

• The founder was executed, not an executioner, and rebuked the disciple who drew a sword in his defence (Matthew 26:52).
• The command is explicit: love your enemies, pray for those who persecute you (Matthew 5:44); do not repay evil for evil (Romans 12:17); the weapons of this warfare are not of the flesh (2 Corinthians 10:4).
• The early church was largely non-violent for its first three centuries, while it was the persecuted party rather than the powerful one.
• The tradition developed constraints on violence, not licences for it. Just-war theory in Augustine and Aquinas exists to restrict when force may be used and how; see Just War Theory.

The historical pattern is not that Christian teaching produced the violence but that political power, once acquired, found uses for Christian language. That is a serious indictment of Christians in power, and it is a different claim from the one the objection makes.

The mirror-image error to avoid

The tempting counter-move is to say that atheism killed a hundred million people in the twentieth century. It should be resisted, for two reasons. It is the same fallacy as the objection, treating a belief about God's existence as the operative cause of complex political events. And it is strategically weak, since the interlocutor will correctly note that Stalin did not kill in the name of atheism.

The correct and stronger form is the one stated in the introduction. Secular regimes produced mass killing on an unprecedented scale, which falsifies the objection's prediction that removing religion removes the phenomenon. The variable being blamed is not the one driving the outcome. See Atheist Regime Body Count.

See also

• Religion Causes Violence Objection Defeater, the debate-prep engagement with this objection
• Atheist Regime Body Count, the twentieth-century data
• Inquisition, the numbers and procedure
• Just War Theory, the tradition's constraint on force
• Atheism, the master hub
• Stealing from God Argument, the borrowed-standard diagnosis
• Divine Command Theory, where the condemning standard is grounded
• Morality, the definitional anchor