I’m going to give a talk next month on atheists and stereotypes next month. It occurs to me that “atheist” is not a natural kind, and I’m wondering if the following analogy might help to bring this across.

People who don’t believe in a God are like people who don’t drink bottled water. Statistically speaking, the group might have some noticeable characteristics, but you’re still lumping distinct subgroups together. People who refuse bottled water might be, statistically speaking, more likely to be environmentally conscious. But only some people refuse bottled water because it is environmentally a bad idea. There are also others who stay away from bottled water for entirely different reasons: because it’s a waste of money when tap water is just as good, because they can’t afford it, because they just haven’t developed the habit, etc. etc. It doesn’t make a lot of sense to generalize about non-drinkers of bottled water as if this was a coherent, natural grouping of people.

Similarly, people do not believe in God because of all sorts of reasons. Some are skeptical about popular religion, and it spills over into all kinds of God-talk. Some are at home in a scientific subculture where nonbelief is routine. Some think religion is a social evil. Etc., etc. There are differences between people who join local atheist clubs, and people who do not believe but are indifferent to religion. People who go without God in an academic environment are different from people who shed religion as part of a political movement. “Atheists” are not a natural, coherent group.