Greta Christina on converting the religious

Greta Christina has an excellent piece on AlterNet. She argues persuasively against the silly idea that there's something inherently intolerant about attempts by atheists to persuade the religious to become atheists. While this should go without saying, it's useful for her to put the detail of the argument. Basically, it goes like this: religion is one hypothesis - or, rather, a set of numerous hypotheses - about how the world works; and there is nothing wrong with testing such hypotheses, arguing about them, and attempting to reach agreement on their epistemic status.

I totally agree, and I add that there is no great benefit to diversity if it means a diversity of false beliefs. Diversity of art forms, cuisines, literary traditions and so on can be aesthetically pleasing, but it's rather different with beliefs. We should look kindly on diversity of beliefs insofar as it is evidence that no one belief is being imposed by political power, something we all have reason to fear. However, this kind of diversity is not something to be pursued for its own sake. We should prefer true beliefs to a diversity of (mostly false) beliefs.

Where I do have a quibble is that I think Greta goes close to something like a logical positivist position, with all its inherent paradoxes. Once you start saying that hypotheses are unacceptable unless they are empirically falsifiable, the obvious rejoinder is to ask whether what you just said is itself an empirically falsifiable hypothesis - if it's not, what is it? It doesn't, for example, look like an analytically true claim (something that must be correct as a matter of logic). You need an epistemological position that is at least rich enough to accommodate its own claims.

But I don't think she needs to go down a path that involves any paradoxes. All that needs to be said is the following.

The senses of human beings are limited, so there are many truths about the world that we can't observe directly. Indeed, our senses often mislead us. When it comes to events that happened a long time ago, or which are very distant from us, or which take place on a very small scale, it is extremely difficult for us to obtain knowledge. Direct observation won't do the job.

However, for the past 400 years science has become increasingly able to obtain robust knowledge about very ancient, very distant, and very small-scale events. In principle, its methods can be useful in assessing many kinds of hypotheses about events that are hidden from direct observation. The techniques it uses involve, among other things, the rigorous application of an everyday method of reasoning, namely hypothetico-deductive reasoning. By combining this with mathematical modelling, arguments from consilience, and instruments that extend the senses (telescopes, cloud chambers, geiger counters, etc.), science is able to develop powerful lines of convergent evidence to support hypotheses that would otherwise be no more than arbitrary speculations.

Some religious claims - for example those about the Earth's past - can be tested and falsified by these distinctively scientific means. Others, however, are left in the category of arbitrary speculation, since we have no way to tell whether they are true or false. They are not like logical principles (such as modus ponens) or epistemological principles (such as "prima facie, believe what you see with your own eyes under good conditions") that are widely and inter-culturally accepted and can be discussed, clarified, and refined, as happens in philosophy. They are more like speculations that were made about very distant events at a time in history when human beings had no real evidence to ground such speculations and no real way to test them. Religious claims fitting in this category are, indeed, useless, because it is possible to imagine an infinite number of them without having any reason to believe that any particular one is true, or any way at all to choose among them. It would be freakishly unlikely for any particular one of these arbitary claims to be true.

No paradoxical epistemological principles are required here, just the true observation that it is arbitrary to believe in any particular one of the infinite number of claims that could be made about unobservable and unevidenced events.

With that caveat, I'll send you off to read Greta Christina's wonderful essay.