Friday, September 26, 2014

Against defining religion

What is a religion?  How do we decide whether something is or is not a religion?

One standard definition is that a religion is a community with ritualistic practices and an overarching supernatural belief system.  But it depends on who you ask.

There are at least a few good things that come out of philosophy.  The philosophical analysis of concepts and categories is one of those good things.  One important lesson is that concepts and categories don't necessarily have definitions.

It's worth learning about the prototype theory of concepts vs the classical theory of concepts.  (I will only introduce these theories, and if you'd like to read more in depth, there are better resources for that.)

The classical theory of concepts says that every concept has a definition.  A bachelor is an unmarried man.  A triangle is a polygon with three sides.  A chair is a piece of furniture intended for one person to sit on.  However, this cannot explain why many concepts have a strong association with particular items.  For example, why, when we think of a bird, are we quicker to think of a robin than an ostrich?  Why, when we think of a fruit, are we quicker to think of an apple than a plum?

The idea behind prototype theory is that most concepts have one or more prototypical examples which serve as focal points.  Other objects are judged by their proximity to the prototypical examples.  Prototype theory comes from psychological research in the 1970s (although it has some roots in Wittgenstein's philosophy in the 1950s).  Note that prototype theory is not necessarily a completely correct description of how we think of concepts, but it is a decent approximation, and a step forward from the classical theory.

When people attempt to explain what religion really is by coming up with a definition, they've gotten it wrong from the beginning.  On a fundamental level, religion does not have a definition, because that's not how we think of most words.  In the US, we tend to think of Christianity, Judaism, and Islam.  And then maybe Buddhism and Hinduism.  These different examples provide prototypes for what "religion" is. Any other objects are judged by their proximity to the prototypes.

Definitions of religion are essentially attempts to interpolate the prototypes, to identify common features so that we can more easily decide whether other things are religions or not.  In some sense, we know that Hinduism is a religion before we reflect on the definition of religion, before we even learn any details about Hinduism.  If your definition does not include Hinduism, then something is wrong with that definition.

But sometimes I see people using more nonstandard definitions of religion.  For example, religion as "any all-encompassing belief system".  This definition clearly is not an attempt to interpolate prototypes (or it's a very inept one), nor is it an attempt to describe how we actually think of the concept of religion.  I am forced to consider that these overly broad definitions are prescriptive rather than descriptive.

It's not that prototypes are above criticism.  For example, we here in the US have a bias towards certain religions.  To us, the prototypical religion has a god or gods, an eschatalogy, a sacred text, and it perpetuates itself through evangelism and inheritance.  But not all these properties apply to Eastern religions.  Maybe there are some good reasons to have such a biased prototype, but in this case we don't have good reasons.  It just comes down to ignorance and unfamiliarity with religions outside our immediate surroundings.  It comes down to western colonialism, because if eastern civilizations had more cultural power, they'd be the ones controlling the concept of religion.  Perhaps a prescriptive definition may help to remedy our cultural bias.

But for most prescriptive definitions of religion, I have to ask why.  Why is it good to consider, say, Marxism or Capitalism a religion?  We cannot draw any novel conclusions from a novel definition.  Although sometimes certain conclusions are made more cognitively accessible.  So what does this definition help us understand?  That's what I want to hear when people propose definitions of religion.