Long ago, I linked Plantinga’s definition. All I’ve been doing is following his definition, since the OP is about his argument, not yours.
He answers your question with: “Possible worlds themselves are typically ‘taken as primitive’, as the saying goes: but by way of informal explanation it may be said that a possible world is
a way things could have been-a
total way.” -
andrewmbailey.com/ap/Actualism_Possible_Worlds.pdf
Those are his italics btw. He seems very clear “a possible world is a way things could have been”. What is it you don’t understand?
So what is it that is required to make a possible world a “could have been” as opposed to a couldn’t have been?
Obviously Plantinga isn’t here to clarify that distinction, and yet, that distinction does seem to be one that needs to be made.
My contention is that what “could have been,” logically speaking, depends entirely upon what is. We have no other resort because we cannot produce from whole cloth what is or is not “logically possible.”
Our entire concept of logically possible depends entirely upon its consistency with that which we know to exist – some actualizing principle or other. In this case, what we know to exist leads us to speculate about other things which, therefore, might possibly exist.
However, we have nada to say with respect to pure “possibility” without some reference or other to actuality.
As I said, the only reason we know this world we currently reside in is, indeed, possible is because it exists. Absent that we have no grounds to make claims about what is or what is not possible.
You can refute that claim quite easily. Provide some logical grounds for what is or is not possible without any reference whatsoever to what exists.
Now you might claim any world that is logically coherent is one that is possible.
Well, okay, then it is up to you to give an example of logical coherency that makes no reference to and obtains no warrant from things that actually do exist or some actualizing principle found therein.
My second line of argument is that any and all existent things – which form all the grounds we have for determining what is possible or not – positively require the following actualizing principle: that which has existence itself as its essence, i.e., Aquinas’ Ipsum Esse Subsistens or that which exists a se.
Without that, no world or entity is possible whatsoever because contingent things – even an infinite chain of them cannot bootstrap themselves into existence. Necessary being (that which has existence as its very essence) is a requirement for the existence of any possible being – that is what I refer to as the “actualizing principle” behind every possible world if it is indeed to be possible.
That is also the argument to be made from the modal version, as I discussed in post #50 which neither of you have properly addressed.
Let’s go back to this counter argument to Plantinga’s, shall we?
I don’t think Plantinga need accept your 2. precisely because he would claim that the very definition of MGB precludes that possibility.
His point would be that for a MGB to exist in any possible world, it would necessarily have the trait of being necessary in every possible world. That means for a MGB to be maximally great in any possible world, it would, by definition, have to be maximally great in every possible. One of the conditions for being “maximally great” would be to have necessary existence and for necessary existence to logically be ‘necessary’ it must imply, in a logical sense – even in modal logic – that it be not merely possible in every possible world, but, indeed, NECESSARY in every possible world, otherwise you aren’t logically speaking of the MGB, but something else entirely.
Again, the point being that the MGB either must exist in every possible world or be logically self-contradictory and impossible in every possible world.