Tuesday, February 5, 2013

Problem of free will and God's omniscience

To: Francis M.
From: Geoffrey Klempner
Subject: Problem of free will and God's omniscience
Date: 24th March 2009 13:16

Dear Frank,

Thank you for your email of 16 March, with your essay for the University of London Philosophy of Religion module, in response to the question, 'If God is omniscient, can human beings be free?'

It is interesting that in your answer you avoid the 'compatibilist' route -- arguing that you take your exams 'freely' if and only if you are not put under external constraint, are of sound mind and therefore take the exams 'of your own free will'. All this could be granted, but this would not satisfy someone who (appealing to your argument regarding 'E=the event of me taking two exams on May 14th 2009') argues that if E 'must occur' then your doing E 'freely' is not 'genuine' freedom, but a mere ersatz notion which human beings have constructed in order to justify praise, blame, reward and punishment etc.

Although you dismiss '99%' freedom at the beginning of your essay, it would be perfectly reasonable -- within a compatibilist framework -- to point out that a hardened and unrepentant criminal recently released from jail whom we just *know* is going to commit another crime, nevertheless does so of his own free will.

Or if we take the case of the persistent backslider who makes one resolution after another but always gives in to the craving for tobacco, we *know* that he will break his latest resolution even if he sincerely 'believes' that he won't. This is pathological weakness of will, but still free will according to the compatibilist definition.

What then does 100% add to 99%?

The basis for knowledge is certainly different. If we take causality out of the equation (as you argue) and make this just a kind of 'seeing' (whether from 'outside time', or from a 'fourth dimension') then there is nothing to interfere with the exercise of our freedom. Yet, once again, this is still only compatibilist freedom. In effect, the hypothesis of a God who 'sees' by some non-physical means is no different from the hypothesis of fatalism.

If that is right, then I don't see what sense can be made of a universe where an omniscient God exists but where the future is 'open'. If God knows that E will occur then there is an answer to the question whether E will occur, whether you know this or not. To say that the future is open is to say that there is no answer to the question whether E will occur (e.g. as in Aristotle, there is no answer to the question whether there will be a sea battle tomorrow).

Your thought experiment about time travel appears to be your contribution to the debate. It is always good to try this.

As if conceiving of a God outside time wasn't difficult enough, there is however a serious additional problem with time travel with the assumption that someone has travelled forwards in time to witness E. Let's say your taking the exam (or eating the hamburger) is witnessed by precisely 20 people. That's a fact. After the time travel episode, the event is witnessed by 21 people. In other words, the event is witnessed by precisely 20 people and also by precisely 21 people, which is a contradiction.

This is just a particular case of the general problem with describing time travel episodes: see my Afterword to David Gerrold's 'The Man Who Folded Himself' http://klempner.freeshell.org/articles/afterword.html.

Let's try taking the thought experiment a little further, and imagine not time travel but something more like an observation window which accesses the future. Through this observation window I see you eating the hamburger. The observation window was always there (just hidden in some dark place), and the event was always observable. Problem is, if the event is observable, then you could visit that dark place and observe it too. Now we are in a fix. Because if you look through the window and see yourself eating the burger, then that gives you a reason which you did not have before for not eating the burger. Of course, we could be in an Oedipus scenario; but the problem with Oedipus scenarios is that they require a great deal of stage setting to work. What happens sometimes could not happen always.

(Time travel stories have this too: You visit the future and observe yourself doing X. Do you have to do X just because you saw yourself doing it? The recent independent movie 'Primer' has a brilliant take on this.)

I think that the real conclusion from this is the one which you mention early on in your essay: we are taking a supposition which is, in fact, impossible and then asking what if (counternecessaryfactually) the supposition were true. The proper response should be: If what is necessarily not the case were the case, then anything can be the case. Say what you like.

All the best,

Geoffrey