Long ago, in an essay on “The Decline of Religion,” C. S. Lewis made an interesting comment on the “decline” of chapel attendance in the colleges of Oxford and Cambridge:

Now it is quite true that chapels which were quite full in 1900 are empty in 1946. But this change was not gradual. It occurred at the precise moment when chapel ceased to be compulsory. It was not in fact a decline; it was a precipice. . . . The withdrawal of compulsion did not create a new religious situation, but only revealed the situation which had long existed.

I’m recalling this point as I read the new Pew report on the growth of the religiously unaffiliated.

In the last five years alone, the unaffiliated have increased from just over 15% to just under 20% of all U.S. adults. Their ranks now include more than 13 million self-described atheists and agnostics (nearly 6% of the U.S. public), as well as nearly 33 million people who say they have no particular religious affiliation (14%).

The question I would ask is this: Has there been an actual increase in religiously unaffiliated people, or do people who are in fact unaffiliated simply feel more free than they once did to acknowledge that fact? My suspicion is that until quite recently a person born and baptized into the Catholic church who hadn’t attended Mass in fifteen years would still identify as a Catholic; but recently is more likely to accept his or her unaffiliated status. There is less social (and perhaps also psychological) cost in saying “I have no particular religion that I’m connected to” than there once was.

That is, the poll may reflect not a change in behavior but a change in how people think of their behavior — a change that brings their self-descriptions more closely into line with reality. And that wouldn’t at all be a bad thing: there’s always something to be said for the removal of illusions, for “reveal[ing] the situation which had long existed.”