Bad arguments about religion: faith and evidence

There’s an atheist bad argument which runs something like this: “Faith is believing stuff without evidence, believing stuff without evidence is always bad, therefore faith is bad”.

This seems reasonable at first, but sooner or later you meet a William Lane Craig or similar apologist type, as Jerry Coyne did recently:

Craig argues that science itself is permeated with assumptions about the world that cannot be scientifically justified, but are based on faith. One of these is the validity of inductive reasoning: “Just because A has always been followed by B every time in the past is no proof at all that A will be followed by B tomorrow.” To suppose the latter requires faith.

According to Coyne, as well as the problem of induction, Craig mentions last-Thursday-ism and the idea that we’re all in the Matrix as beliefs that we reject on faith. Some of commenters on Coyne’s blog react as if Craig is advocating these ideas that we all reject, that is, as if he really thinks that the Sun might not rise tomorrow or that we’re in the clutches of a cartesiandaemon. But that’s not Craig’s point. Nor is Craig being inconsistent if he gets on an aeroplane assuming that the laws of physics will carry on working as they always have to keep it flying. After all, he’s not the one claiming that it’s always wrong to believe things without evidence.

The problem here, which makes the atheist’s argument a bad one, is that the atheist has cast their net too broadly. Craig is right to say that there are things that atheists (and everyone else) believe “on faith”. To say that these beliefs are always unwarranted leaves the atheist open to Craig’s counter-argument that, to be consistent, the atheist should then discard those beliefs or admit that it’s not always wrong to believe things without evidence.

Doing better

Nevertheless, something has gone wrong with Craig’s argument if it’s supposed to be a defence of religious faith (as all Craig’s arguments ultimately are). Religious faith is different from belief in induction or the existence of an external world. The atheist should abandon the claim that unevidenced beliefs are always bad, and concentrate on the distinction between religious beliefs and, say, the belief that the external world is real.

One way of doing that would be to turn Craig’s allegation of inconsistency back on him. As Chris Hallquist puts it

belief in the Christian God isn’t very much at all like most of the common-sense beliefs commonly cited as threatened by Descartes & Hume-style skepticism (like belief in the reliability of our senses), but is an awful lot like beliefs most Christians wouldn’t accept without evidence – namely, the beliefs of other religions.

The atheist’s discomfort is now the apologist’s: either he must accept that, say, Muslims or Scientologists are right to take things on faith (in which case, why not join up with them instead?); or further distinguish his religion from theirs (probably by making arguments about the resurrection of Jesus). The atheist’s acceptance of the real world doesn’t come into it.

Hume’s own solution to radical scepticism was to note that he couldn’t entertain that sort of thing for long. Creatures like us soon fall unavoidably back on treating other people as if they were conscious, the world as if it were real, and so on. The great man tells us:

Most fortunately it happens, that since reason is incapable of dispelling these clouds, nature herself suffices to that purpose, and cures me of this philosophical melancholy and delirium, either by relaxing this bent of mind, or by some avocation, and lively impression of my senses, which obliterate all these chimeras. I dine, I play a game of back-gammon, I converse, and am merry with my friends; and when after three or four hours’ amusement, I wou’d return to these speculations, they appear so cold, and strain’d, and ridiculous, that I cannot find in my heart to enter into them any farther.

Among educated folk, thoughts of gods rely on meetings with other believers to keep them going: believers are chronic sinus sufferers. They rarely anticipate the world being any different from a godless one, and those who act as if God is real are called crazy even by their fellow believers. To be sure, that doesn’t mean their avowed beliefs are false. But again, they are not like the commonplace beliefs that everyone takes on faith. In my experience, they fly forgotten, as the dream dies with the dawning day. How about a nice game of backgammon?

Edit: gjm11 suggests another reasonable response in this comment: admit that believing stuff without evidence is bad, and try to minimise it, and say that the problem with religious faith (in so far as that means holding unevidenced beliefs) is that it means having way more unevidenced beliefs than necessary.

Edit again: I’ve also commented with a shorter version of this on Coyne’s original posting, so there’s some discussion there too.

See also

8 Comments on "Bad arguments about religion: faith and evidence"

  1. Another option is to maintain the position that unevidenced beliefs are always bad; to admit that we all have some (e.g., that the world behaves in a somewhat-consistent manner); to say that that’s a necessary evil (since the alternative is complete epistemic collapse); and to say that the problem with religious faith (in so far as that means holding unevidenced beliefs) is that it means having *way more* unevidenced beliefs than necessary.

    On this view of things, Craig’s argument is a bit like saying “So, you say I shouldn’t have stolen your car? But don’t you admit that it’s OK to steal bread when the alternative is for you and your family to starve to death? Aha, gotcha: you endorse theft too, so how dare you object when I do it?”.

    (I think I think the best approach is a combination of this with the sort of position expressed in, e.g., these articles on Less Wrong.)


  2. The thing about Science is that it is _not_ belief. It does not prove anything to be absolutely right, or anything to be absolutely wrong. Instead it provides models with predictive ability.

    If any of the parts of it ceased to provide predictive models (because cause and effect ceased to function, for instance) then we would cease to use them.

    If people say “God doesn’t exist, there is no scientific evidence” what they are basically saying is “The models of the universe which have the most predictive power do not involve God”.

    There are models of the universe that involved God creating it in 4004BC (or thereabouts). However, these did not fit the facts that we had, and the models created based on it did not match with the evidence around us (see this for instance).

    Anyone who has studied Philosophy of Science knows that science isn’t perfect, doesn’t provide 100% reliability, etc. That’s not what science does.


      1. I think there’s a qualitative difference here. In that most belief systems hand you a bunch of stuff that’s The Truth, and Science doesn’t do that. It hands you a bunch of models that have been tested to a variety of different degrees, that you might find useful for dealing with life.


  3. The passage from Jerry Coyne that you quote doesn’t distinguish between two philosophical concepts: inductive fallibility (inductive arguments can never be completely certain) and inductive skepticism (an inductive argument provides no reason to believe its conclusion).

    I think the modern consensus is that inductive fallibility is right and inductive skepticism is wrong: that is, it’s rational to believe the conclusion of an inductive argument, even though the argument is not logically valid, because we do have evidence, of a statistical or probabilistic nature. (Probabilistic induction is of course fallible too: we can’t ever know for certain that our sample is not an outlier, or even that nature admits of a statistical description. But again, we don’t take these things on faith, we have evidence for them.)

    So I am not persuaded that induction is a good example of something that people believe in without evidence. Which is not to say that there isn’t such a belief, just that induction is not it. It’s a good try, though.


    1. I think the modern consensus is that inductive fallibility is right and inductive skepticism is wrong.

      I’m not sure how that works, unless you’re allowed some sort of circularity (possibly of the “lens that sees its flaws” kind that Yudkowsky talks about in the articles gjm11 linked). Probabilistic reasoning about the future surely rests on induction, so you’re looking at inductive justifications of induction.


      1. That’s right: the argument has a kind of circular flavour. This would be fatal if we expected induction to be logically valid, but we don’t, we only expect it to be reasonable.


Leave a Reply to gareth_rees Cancel reply

Your email address will not be published.