Original post by banterboyI'll try, i find some parts confusing myself. In the book i have by him (The Nature of necessity) it's like a 20 page argument.
Well, basically, there's an atheistic argument that goes roughly like this:
1. God is omnipotent.
2. God is omni-benevolent
3. Evil exists in the possible world A.
4. 3 contradicts one or both of 1 or 2.
5. So God cannot exist in A.
6. A is the actual world.
7. God does not exist in the actual world.
(btw possible worlds are just complete sets of non contradictory propositions which account for each other, so there is a possible world which is exactly the same apart from I have blue not brown hair. There are no possible worlds where I dye my hair blue successfully but have entirely brown hair.).
When the theist points out that evil is required for free will, the atheist tends to say
"God could have made some possible world P where free will exists and no evil occurs, otherwise he wouldn't be omnipotent".
Plantinga therefore says that any morally relevant definition of free will has to include the ability to do evil on every possible world (Call this def.1). Plantinga calls this Transworld Depravity.
So suppose I can choose to kill someone or not to kill someone on world A at time t, and i can't kill someone and not kill someone at t. I will use this as an example, but the point extends to the moral decisions of everyone in a possible world.
Call the world in which i choose not to A1, the one where i do A2.
Suppose God could instantiate possible world A1.
Then by def.1 and def.2 "I do not have free will with respect to the fact I did not kill someone at t in world A1. As By def.1, free will requires my ability to make the morally relevant choices P and not P in the same possible world. So, if God creates a world in which not P must obtain, and therefore P is impossible, I have no free will to choose P or not P. So I am not free on world A1.
Therefore, if God creates a world A1, where it is impossible that i choose to kill, God has on that world removed evil by removing free will.
In A1, therefore, God has not "created a world in which myself and everyone else freely chooses not to do evil", he has simply attained the world in which moral evil does not occur.
Now imagine I am on A2, and that God's will does not enforce my murderous choice on A2. Also suppose that on A2, up until time t when i actually kill the bastard, the proposition I could choose not to kill him, is true. Universalise this to everyone on A2, and we have it that:
For all morally evil choices on world A2, it is possible that they could have been avoided.
As a crucial consequence of that, however unlikely; on a world like A2, it is possible that humans have free will and always choose the morally good option.
Now make one final assumption. God prefers his children to have free will than for them to lack free will but have no evil in their lives. Given an even elementary understanding of Christian theology, this assumption is surely desirable.
Now we have established the premises we need to make the following argument:
1.The best possible world for an omnibenevolent and omnipotent God is one where free will exists and evil does not occur.
2. Morally significant free will requires that on every world one has free will, one can choose between a moral and immoral action.
3. It is impossible for him to actualise a world like A1 where "everyone freely chooses good all the time", because on that world no one can choose evil, and by 2. this world does not contain free will, making it an impossible world.
4. If God actualises a world like A2, then it is possible but not necessary that everyone freely chooses good.
5. Given 1, 2 and 3, this is best possible world even an omnipotent God could create.
6. Therefore, there is no contradiction between evil occurring on a world like A2 and God's omnipotence/omnibenevolence.
7. Our world is like that of A2 in the relevant respects.
8. Therefore, and omnibenevolent and omnipotent God can exist in the actual world.
I hope that explains the core of the argument somewhat.