Is there really much difference?

[Angrboda]

This is related to a question which frequently comes up in debates, namely, how much of the population is atheist. It turns out that while there is a substantial segment of society that is non-religious, only a small portion of that is self-consciously atheist. (And determining that number is itself complicated; see Zuckerman.) I would suggest that on the other side of the line, there are those who identify as religious, who are actually better described as non-religious. I would suggest that the differences are real, but gradual, and not necessarily aligning with the standard guideposts of belief or non-belief (e.g. ethnic Jews who still participate in Jewish religious traditions, secular Japanese who observe Buddhist and Shinto traditions, and Asian Indians who still maintain consciousness of varna and jati in determining whom to marry and associate with). Moreover, the official categories don't always correlate with the anthropological ones (as noted, some Christians believe in reincarnation).

I suppose I would make two points. First, despite its incompatibility with folk psychological notions of belief, it is possible to believe mutually inconsistent things, and even be aware of the inconsistency, without having to give priority to one or the other or otherwise reconcile the two. Therefore, it wouldn't be seen as unusual to me for a Christian to believe in reincarnation, yet also believe that people will burn in hell for not embracing Christ. It doesn't make sense according to most people's understanding of "normal psychology," but it is my contention that "normal psychology" doesn't accurately describe the psychology of the species.

Another point that comes to mind is that religious beliefs and behaviors have multiple levels and dimensions. So religion may be similar to politics: some people might be intensely focused on social issues, another on economic ones, another not especially political, and another, intensely involved politically only if their specific ideological faction is involved in the issue.

So I think trying to create a set of ready-made boxes of limited number to describe religious belief and behavior is going to fail to capture important aspects of the phenomena. In the biological sciences, scientists are generally divided into two groups: the lumpers, who want to see different examples as just variations around a central locus — they want to lump all examples into one broad category; then there are the splitters, who attempt to give meaning and significance to any variation, no matter how small, so things end up split into discrete but meaningful categories. Needless to say there are disadvantages and advantages to both approaches. Perhaps what may be more important is to recognize the utility of both approaches, and to understand how one's approach — lumping or splitting — is going to distort and create artifacts of interpretation as a consequence of the approach you've adopted, and, not to be misled into believing that those artifacts are themselves present in the data.