Breaking the Spell: Religion as a Natural Phenomenon

A research topic of particular urgency, but also particular ethical and political sensitivity, is the effect of religious upbringing and education on young children. There is an ocean of research, some good, some bad, on early-childhood development, on language learning and nutrition and parental behavior and the effect of peers and just about every other imaginable variable that can be measured in the first dozen years of a person’s life, but almost all of this—so far as I can determine—carefully sidesteps religion, which is still largely terra incognita. Sometimes there are very good—indeed, unimpeachable—ethical reasons for this. All the carefully erected and protected barriers to injurious medical research with human subjects apply with equal force to any research we might imagine conducting on variations in religious upbringing. We aren’t going to do placebo studies in which group A memorizes one catechism while group B memorizes a different catechism and group C memorizes nonsense syllables. We aren’t going to do cross-fostering studies in which babies of Islamic parents are switched with babies of Catholic parents. These are clearly off limits, and should remain so. But what are the limits? The question is important, because, as we try to design indirect and noninvasive ways of getting at the evidence we seek, we will confront the sort of trade-offs that regularly confront researchers looking for medical cures. Perfectly risk-free research on these topics is probably impossible. What counts as informed consent, and how much risk may even those who consent be permitted to tolerate? And whose consent? The parents’ or the children’s?

All these policy questions lie unexamined in the shadows cast by the first spell, the one that says that religion is out of bounds, period. We should not pretend that this is benign neglect on our part, since we know full well that under the protective umbrellas of personal privacy and religious freedom there are widespread practices in which parents subject their own children to treatments that would send any researcher, clinical or otherwise, to jail. What are the rights of parents in such circumstances, and “where do we draw the line”? This is a political question that can be settled not by discovering “the answer” but by working out an answer that is acceptable to as many informed people as possible.

It will not please everybody, any more than our current laws and practices regarding the consumption of alcoholic beverages please everybody. Prohibition was tried, and by general consensus—far from unanimous—it was determined to be a failure. The current understanding is quite stable; we are unlikely to go back to Prohibition anytime soon. But there are still laws forbidding the sale of alcoholic beverages to minors (with age varying by country). And there are plenty of gray areas: what should we do if we find parents giving alcohol to their children? At the ball game, the parents may get in trouble, but what about in the privacy of their own homes? And there is a difference between a glass of champagne at big sister’s wedding, and a six-pack of beer every evening while trying to do homework. When do the authorities have not just the right but the obligation to step in and prevent abuse? Tough questions, and they don’t get easier when the topic is religion, not alcohol. In the case of alcohol, our political wisdom is importantly informed by what we have learned about the short-term and long-term effects of imbibing it, but in the case of religion we’re still flying blind.

Some people will scoff at the very idea that a religious upbringing could be harmful to a child—until they reflect on some of the more severe religious regimens to be found around the world, and recognize that in the United States we already prohibit religious practices that are widespread in other parts of the world. Richard Dawkins goes further. He has proposed that no child should ever be identified as a Catholic child or a Muslim child (or an atheist child), since this identification in itself prejudges decisions that have yet to be properly considered.

We’d be aghast to be told of a Leninist child or a neo-conservative child or a Hayekian monetarist child. So isn’t it a kind of child abuse to speak of a Catholic child or a Protestant child? Especially in Northern Ireland and Glasgow where such labels, handed down over generations, have divided neighbourhoods for centuries and can even amount to a death warrant? [2003b]