Reading Time: 4 minutes

People who are ignorant of science sometimes speak as if the scientific method was some esoteric, arcane method of problem-solving, applicable only to a few highly specialized areas of inquiry and having no relevance to everyday life. But nothing could be further from the truth. In reality, the scientific method is just a more sophisticated, more careful way of asking and answering questions about what is true, with extra safeguards built in to counteract the ways that human beings often fool or mislead ourselves. In principle, science can answer any question whose answer is a matter of empirical fact and not just a matter of opinion or subjective judgment.

This fact has implications for a broad range of religious claims, especially about the efficacy of prayer. Large, well-designed scientific studies have repeatedly failed to find any evidence that sick people who are being prayed for recover faster or more completely than people who aren’t. In response, many apologists have retreated to claiming that prayer’s effectiveness can’t be tested scientifically, such as this one:

Luckily for everyone, scientific attempts to prove or disprove God are all doomed to failure. We live in exactly the world the thoughtful Christian would expect to find. For those who believe, hints of God are everywhere. But none are convincing. Faith remains a requirement…

But this claim probably says more than its originator intended. When theists say that prayer is untestable, what they’re really saying, whether they realize it or not, is that prayer has no measurable effect on the world. If it did have a measurable, repeatable effect, we could easily design an experiment that would show it. But since believers say that this can’t be done, they must mean that prayer has no benefits that can be proven by any test. Consider some of the consequences that necessarily follow from this claim:

Sick theists who pray for healing are no more likely to recover than sick atheists. If people who were prayed for recovered more quickly or more fully than people receiving no prayer, we could easily show this with a test. That was the point of the MANTRA study I linked to above. But if prayer is untestable, then that must mean that prayer has no measurable effect on a person’s recovery, regardless of how many people are offering prayers for them or how fervent those people are in their faith.

Theists who pray for success and prosperity are no more likely to receive it than atheists. Prosperity-gospel churches often teach that the more money a believer tithes, the more God will reward them. Again, a longitudinal study tracking the amount of people’s donations and comparing it to their subsequent financial success could easily show this to be so. If prayer is untestable, however, this must mean that the amount of money you give to your church has no effect on the odds of your subsequently becoming rich.

More committed, more faithful believers have their prayers answered at the same rate as more casual, less committed believers. Even if you start with the assumption that God only grants prayers that agree with his will, it seems like a reasonable guess that more devoted, more committed believers would have at least a slightly greater understanding of God’s will than casual, apathetic churchgoers, and hence their prayers would be more likely to come true. But if prayer is untestable, there must be no such measurable effect, which means that one’s level of commitment means nothing to the effectiveness of one’s prayers.

The number of people praying for some outcome makes no difference to its probability. Even if the level of one’s devotion makes no difference, you might guess that the number of people praying for some outcome would be correlated with how likely that outcome is. But if prayer is untestable, then it must make no difference whether one, a hundred, or a million people pray for something – it would be just as likely, or rather unlikely, to come true.

The specific beliefs of the people praying for some outcome makes no difference to its probability. If there’s one true religion, it seems likely that God would only answer the prayers of believers in that religion, or at least would answer their prayers more frequently than the prayers of heretics. But that would also be an easily testable effect. If prayer is untestable, there must be no such effect, and this means that people of all religions – Christian, Muslim, Mormon, Hare Krishna, Jain, Zoroastrian, Shinto – would see their prayers come true with roughly the same frequency.

People who pray daily are no more happier, no more virtuous, and no more trustworthy than people who rarely or never pray. Some people claim that prayer doesn’t produce miraculous effects in the world, but is intended to strengthen the faith and improve the character of the believer. But even this can’t be true if prayer is untestable. If people who are otherwise alike in social standing are measurably different in any positive psychological trait, depending on whether or how often they pray, this would be a testable effect. We could measure it with the same kind of epidemiological surveys that measure the beneficial health effects of diet or exercise. If this kind of test wouldn’t work, then it must be the case that prayer produces no detectable change in the character of the believer.

Nations populated by people who pray frequently are no more socially healthy than irreligious nations. Building on the last point, if prayer has no measurable effect, this must apply to nations as well as people. This means that nations of fervent believers who pray frequently are no different from godless, atheist nations in every measure of social health: divorce rates, crime rates, number and severity of natural disasters, overall happiness of the populace, and so on.

Avatar photo

DAYLIGHT ATHEISM Adam Lee is an atheist author and speaker from New York City. His previously published books include "Daylight Atheism," "Meta: On God, the Big Questions, and the Just City," and most...