Modal logical proofs of God

We will continue with Anselem's argument which we began with Anselem and Descartes and follow it as it continues into the 20th century. Despite that Russel killed Anselem's argument for good and Hume proved that you can't prove the existence of anything a priori, with new logical tools developing in the 20th century, someone is bound to attempt to try to update Anselem's proof. The man to do that is Alvin Plantinga and the technique he uses is modal logic.

Modal logic is simply just logic dealing with possibility, and Plantinga uses it to prove that since God possibly necessarily exist then God necessarily exist. Philosophers nowadays use necessarily existent as meaning existent in all possible worlds and possibly existent as existent in some possible worlds. Thus possibly necessary should lead to the confusing proposition that in some possible worlds something exists in all possible worlds, which some philosophers have proposed means it exists in all possible worlds. In other words, they propose that possibly necessary = necessary.

What Plantinga does it he basically defines God as maximally excellent and great, which to him means God is omniscient, omnipotent and wholly good and must exist in all possible worlds (in other words, God necessarily exists). He proposes that a God by this definition is possible. Thus, if this God is defined as necessarily existing, then saying this God is possible, means God possibly necessarily exists, which, as we explained above means God necessarily exists. In other words, God exists in all possible worlds, which of course includes this world.

Looking at this argument one should grow a bit skeptical. We made a very quick leap from saying God could exist to saying God must exist. Our first approach is to try and refute it through counter-example. For example, I propose that it is possible that I necessarily have six toes, which means I necessarily have six toes, which means I should look down and count six toes on my foot. I look and I count only five--argument refuted.

But hold on a minute. If I necessarily have six toes then I should have six toes in this world. The fact that I don't proves that I can't necessarily have six toes. Thus we can't define something as possibly necessary unless we can empirically observe it in this world. But then all we have to do is to take as a counter-example something we haven't empirically observed. For example, it's possible that in all possible worlds there is life on more than one planet. With this proposition, I've proved there's life on other planets. But of course, we should notice that to accept that it is possible that life must necessarily exist on more than one planet is immediately to accept there is life on other planets. We can see the same problem in Plantinga's argument. To propose that a God as he defines it could exist is to immediately propose that such a God exists, which is of course exactly what he is trying to prove. In short, Plantinga has merely found a new technique for secretly importing his conclusion into the argument as a premise. The question still unanswered is whether God as he defines it is possible.

This turns out to be the real crux of the argument: if a God that is necessarily existent can possibly exist then that God must exist. Plantinga's definition has put us in the position of saying that God as he defines it is impossible or necessarily exists. There are arguments that argue that God as he defines it (omniscient, omnipotent and wholly good) is irreconcilable with a world which has evil. Hume in his Dialogues on Natural Religion, which I discussed last post, mentions a trilemma which he ascribes to Epicurus, which basically says God is either unable to prevent there from being evil (in which God couldn't be omnipotent and omniscient) or is unwilling to prevent evil (in which case God couldn't be wholly good) or there is no evil in the world (which seems to disagree with experience). Plantinga, to his credit, does try to show that God, as he defines God, is reconcilable with a world which contains evil. Nonetheless the whole argument still rests on this question of whether such a God is possible, which we might be skeptical of. We should of course remember that even if we can show that a God as Plantinga defines it is impossible, it doesn't mean God doesn't exist, merely that God as Plantinga defines God doesn't exist.

On the other hand, we might also attack the whole argument by doubting the proposition that possibly necessary = necessary. Remember, possible necessary means that in some possible world something exists in all possible worlds, which I think most normal people would perceive as complete nonsense, but philosophers have a higher tolerance for nonsense and in fact many philosophers do accept this.

Next Post, proofs God doesn't exist.