Thy Will Be Done

Blind studies and unanswered prayers

Illustration by Sean Kelly

"Okay. Now sit up. That's great. And close your eyes and relax all your muscles, starting with your feet, your calves, your thighs. Loosen up your belt. Shrug your shoulders around. Roll your head and neck around—just loosen up."

Herbert Benson, M.D., the president of the Mind/Body Medical Institute and an associate professor at Harvard Medical School, spoke in a voice of implacable serenity, and I did what I was told. Benson is the author of The Relaxation Response, which has sold some four million copies since it was published, in 1975, and he was demonstrating some of the techniques he uses to help people achieve a more beneficial state of mind. People pay considerable sums to attend seminars at which Benson introduces those techniques, and from my own experience I must admit that there is much to recommend them.

However, I had come to Benson's consulting room to learn not about relaxation but about his most recent research project, which has to do with prayer and healing. Specifically, it involves what is known as "intercessory prayer": prayer offered by one person for another's welfare—in this case, another person's recovery from illness. Benson has mounted a controversial scientific effort to determine what may be indeterminable: whether intercessory prayer "works."

Benson came to this question gradually. His medical training was in cardiology, which led him to an interest in hypertension, and ultimately to the idea that a reduction in mental stress could have positive physical consequences: the "relaxation response." The notion that there is a connection, however hard to define, between state of mind and state of body is by now so widely accepted that it's hard to remember how much resistance ideas like Benson's initially met. Meditation and similar techniques were one way of producing a state of mental well-being, Benson maintained, and he came to believe that meditation's close kin, repetitive personal prayer, might be another. His own religious background is Judaism, and though he is not observant, he is a believer in God. He has speculated in his writings that human beings are "wired for God": for evolutionary purposes, believing in God and an afterlife might have conferred a survival advantage. In 1995 Benson established the first course on spirituality and healing at Harvard Medical School (though interest in the subject there goes back at least to William James). Around the same time he began to think about taking his research on prayer into another dimension.

By now a significant body of research backs up the notion that personal prayer can be a healthful activity, if only because of the placebo effect. Remote intercessory prayer, in behalf of patients who don't know they are being prayed for, is quite another matter. Here the benefits from what may merely be psychic self-delusion are missing, and prayer has to speak for itself.

Herbert Benson is not alone in his inquiries—studies of intercessory prayer have been under way at Duke University and at Temple University. Such investigations have a long history. In the nineteenth century Sir Francis Galton tried by various means to evaluate the efficacy of intercessory prayer. For instance, he looked at the lifespan of members of royal houses, assuming that royalty would be the most prayed-for people in any kingdom, but found, as he reported in The Fortnightly Review in 1872, that royal personages were "literally the shortest-lived of all who have the advantage of affluence," even when deaths by accident or violence were excluded. Galton also examined maritime records to determine whether there was any difference in the rate of misadventure at sea between vessels carrying missionaries to distant lands and vessels plying the same routes but engaged in trade, including the slave trade. Insurance companies, he discovered, "absolutely ignore the slightest difference between them."

The use of double-blind clinical trials to investigate intercessory prayer was pioneered in a 1965 study conducted by C.R.B. Joyce and R.M.C. Welldon at the London Hospital Medical College. They assigned patients afflicted with "stationary or progressively deteriorating psychological or rheumatic disease" to two groups: one was prayed for by volunteers, often in teams, and the other was not. The patients were not told they were participating in a clinical trial, and the examining physicians did not know to which group the patients were assigned. After many months the researchers assessed the conditions of the "treated" and "control" patients and discerned no clear pattern.

In 1988 Randolph C. Byrd, of the San Francisco General Medical Center, reported on a study he had conducted of 393 patients in a coronary-care unit, roughly half of whom unknowingly "received IP" from born-again Christians who prayed for them by first name outside the hospital. Byrd determined that the control patients "required ventilatory assistance, antibiotics, and diuretics more frequently than patients in the IP group," and concluded that "intercessory prayer to the Judeo-Christian God has a beneficial therapeutic effect."

In 1999 the cardiologist William S. Harris and several colleagues published a study similar to Byrd's in the Archives of Internal Medicine. The Harris study involved 990 patients admitted to the coronary-care unit at the Mid America Heart Institute, in Kansas City, Missouri. Once again patients were randomly assigned without their knowledge to a treatment or a control group, with the treatment group being prayed for by a team of outside intercessors. The study found that the treatment group registered better outcomes on a specially devised coronary-health scale. "This result," Harris and his colleagues concluded, "suggests that prayer may be an effective adjunct to standard medical care." The authors skirted the question of the precise causal mechanism, noting that "when James Lind, by clinical trial, determined that lemons and limes cured scurvy aboard the HMS Salisbury in 1753, he not only did not know about ascorbic acid, he did not even understand the concept of a 'nutrient.'"

Not surprisingly, the small world of investigators attempting to document the power of intercessory prayer has given rise to a parallel and contrary world of investigators attempting to show that such claims are nonsense. Many of these researchers are grouped around the Council for Secular Humanism and the Committee for the Scientific Investigation of Claims of the Paranormal, both based in Amherst, New York, and around the magazines Free Inquiry and Skeptical Inquirer.

The anti-IP argument has several parts. One of them takes issue with how studies have been conducted. For instance, the "blindedness" of the Byrd study was not as complete as it should have been, and the definition of what constitutes a "good" outcome in the Harris study was far too fluid. Also, how does one interpret the fact that many people in the IP groups didn't show any improvement at all? (Thirteen of those in the Byrd study died.) How does one even begin to get a handle on IP's dose-response effect, which would be crucial if IP were a drug? Is it possible to receive an overdose?

A broader line of attack questions whether a scientific evaluation of the efficacy of prayer is even theoretically feasible. How does one know that the people who are supposed to be doing the praying are really doing it? Must one not also consider the impact of malicious supplication? The scientific distinction between prayed-for groups and not-prayed-for groups is probably impossible to maintain: people who don't receive formal IP treatment may be getting it in other ways—from relatives, for instance, or as spillover from general prayers by the devout for the sick. This is a problem that skeptics refer to as "background prayer."

And then there are the many questions posed by theology. "If you are specifically assigning any efficacy to the Judeo-Christian God or the biblical God, you are in big trouble," says Hector Avalos, a professor of religion at Iowa State University. Avalos was a Pentecostal faith healer as a child and is now an atheist and a prominent skeptic. "There are circumstances in which the biblical God will not answer prayer. How does a scientific study ever establish that subjects are meeting God's conditions? There are passages in the Bible, for example, which say that if a whole nation is not doing well, he's not going to answer its people's prayers. There's a verse in Jeremiah which says don't even bother to pray for evil people. It's beyond me how scientists will ever establish what God's mood is any particular day."

This is the context in which Herbert Benson and his colleagues have organized their study of intercessory prayer. Benson won't discuss details of the research design, except to say that he believes he has avoided the flaws of previous studies and has addressed the issue of background prayer. The data were gathered in widely separated hospitals from more than a thousand patients divided into three groups. One group knew it was being prayed for. The other two groups didn't know they were in a study; one of them was prayed for and the other wasn't. The research has been supported by a large grant from the John Templeton Foundation.

Benson is careful to emphasize that he's investigating the matter dispassionately—just looking for correlations. Although his data have been collected, he and his team remain blinded and do not yet know the results. A full study will be published early in 2002. The research effort has already caused "some slight discomfort," as doctors say, in both the scientific and religious realms. Among scientists, reaction to undertakings like Benson's falls heavily toward the contemptuous end of the shrug-to-sneer spectrum. (The National Institutes of Health, however, is accepting grant proposals for research on "therapeutic prayer," broadly defined.) Among people in religion there are of course many who have no doubt that a personal God answers individual prayers and many who detect only vanity in that notion; but both sorts of people may deem it prideful, or foolish, to put God to the test. Some, like those in charge of the Templeton Foundation, believe that science and religion can be—should be—reconciled. But far more people in both science and religion have accepted that their two worlds were rent asunder centuries ago, and see every reason to preserve the breach.

Such will be the contours of the discussion when Benson reveals his results, and that discussion will probably be a heated one. I'm no doctor, but my advice to interested spectators comes from Benson's earlier research: Shrug your shoulders around. Roll your head and neck around—just loosen up.