My review of Elaine Howard Ecklund's Science vs. Religion is online and will be in the print edition of your Washington Post this Sunday. I'm unaccustomed to reviewing books in 300-400 words, so there's a bunch I'd have liked to say but couldn't, and I felt like I should wait to blog the book until the Post review was out.

The very short version of the review is that the book is good. It's written mostly as guidance for scientists trying to sort out to deal with science and religion in their own lives, but there's valuable insight for nonscientists as well. I rather like the opening:

Americans are almost evenly divided between those who feel science conflicts with religion and those who don't. Both sides have scientific backers. Biologist Richard Dawkins rallies atheists by arguing that science renders religious faith unnecessary and irrational. Geneticist Francis S. Collins (before becoming NIH director) organized evangelical scientists to offer a vision of science and faith reinforcing each other.

Rice University sociologist Elaine Ecklund offers a fresh perspective on this debate in "Science vs. Religion." Rather than offering another polemic, she builds on a detailed survey of almost 1,700 scientists at elite American research universities -- the most comprehensive such study to date. These surveys and 275 lengthy follow-up interviews reveal that scientists often practice a closeted faith. They worry how their peers would react to learning about their religious views.

Ecklund makes no bones about thinking scientists can and should do more to reach out to religious America, but her book's goal is not polemic. She lets the scientists speak for themselves about the value they do or don't see in outreach to religious communities, and about the difficulties they've found in such outreach.

Most remarkable are the moments when scientists pause while talking about their own religious views, literally lacking the vocabulary to express how they think about these issues. Academic culture, especially in the sciences, places little weight if any on discussion of religious issues, leaving scientists literally at a loss for words when such issues arise. And they do arise, whether scientists are addressing creationist attacks (a common theme in her interviews), or controversies over stem cells, global warming, bioethical implications of synthetic biology, or a host of other contentious social issues rooted in science. When scientists cannot address these issues, or can only address them by saying "God is dead, get over it," it doesn't help.

Don't believe me that it doesn't help? Geoffrey Munro of Towson University reports a study in the Journal of Applied Social Psychology that bears on the matter. In "The Scientific Impotence Excuse: Discounting Belief-Threatening Scientific Abstracts," he finds that people were more likely to reject science outright when confronted with a scientific claim that conflicted with their own deeply held beliefs. In the study, people were presented with scientific evidence that would either confirm or disconfirm their belief in a link between homosexuality and mental illness. They were then asked to evaluate whether science was capable of addressing such a link, and those whose beliefs were disconfirmed tended to be less confident in science's ability. Then the researchers asked whether science could address a range of unrelated questions, from clairvoyance to the death penalty, and again, people's confidence in science's ability to address those unrelated issues declined when they were simply confronted with a disconfirmation of their beliefs.

Does this mean no one should ever debunk anything? Of course not. But it does mean that such debunkings should take account of where the audience is. In the creationism/evolution debate, the belief that evolution and religion are incompatible is a major stumbling block, and efforts to educate people about evolution that specifically do not challenge people's deeply held religious views might be less likely to cause people to reject evolution and all of science than a more forthright approach. But if, as Ecklund suggests, scientists lack the experience with religious believers and with religious vocabulary to understand which beliefs are deeply meaningful to religious people, and how to address issues that might butt up against those beliefs, their outreach efforts could backfire and undermine not only the immediate issue of scientific literacy at hand, but acceptance of science as a way of learning about the world more generally.

For what it's worth, about half of scientists identify with a religious tradition, and half (not quite the same half) attend religious services periodically. Those religious scientists are not all theists (only a third of scientists are theists in any sense, the remaining two thirds are evenly split between agnostics and atheists), and they often feel that their religious beliefs isolate them in professional settings. They fear that mentioning their religious beliefs or practices would cause colleagues to treat their research with less respect, even though half of their colleagues, on average, also have religious beliefs or practices. This self-ignorance of academics about their colleagues' religious practices contributes to the depauperate religious conversation among scientists, and the incomplete engagement of scientists with the broader religious public.

Ecklund's advice to scientists is simple and sensible. To quote the Washington Post piece again: "the bottom line is recognizing and tolerating religious diversity, honestly discussing science's scope and limits, and openly exploring the disputed borders between scientific skepticism and religious faith."