What do you think of this argument for God?

[Nonpareil]

Let's grant that this Necessary Being exists in some possible world, in the same way as unicorns. Why should that be our world? How can you know?

This is where we get into modal logic, which is one of those places where theistic philosophy gets most notoriously stupid.

In theory, there's nothing wrong with modal logic. It's a simple extension from statements that are straightforwardly true or false - it allows for the concept of possibility, such as "it is possible that it will be cold today". The two main concepts of it, then, are possibility and necessity - what could be true and what must be true. From there, of course, it gets much more complicated; it's a very deep field, with a lot of theoretical applications that I won't get into here.

But, as it applies to this discussion, we don't need to get deep. It's enough to deal with the concepts of possibility and necessity.

You see, to model statements made in modal logic, we are asked to consider "possible worlds" - ways that the world might have been, but isn't. In essence, if a coherent statement can be made, then it can be expressed in terms of a possible world (note the use of the word coherent; statements like "this mouse is smaller than itself" and "one times one equals two billion" are not coherent, while statements like "the United States never existed" are). Most reasonable students of modal logic accept that these "possible worlds" are simply abstractions, but there are a handful - most notably Alvin Plantinga, the apologist who formulated the modal ontological argument, which is basically what we're currently dealing with - who maintain that these possible worlds are worlds that actually exist, a la alternate dimensions in science fiction.

The other thing to understand is that, when using modal logic, you need to set down a set of axioms that define how the operators within it work. One set of axioms that sees common use is referred to as "S5". Again, in theory, there's nothing wrong with S5, but it has the misfortune of being the set of axioms that Plantinga chooses to abuse for his modal argument.

The thing about S5 that makes it useful to Plantinga is that, under S5, strings of modal operators can be cut down to the last one, no matter what came beforehand - that is, if something is "necessarily necessarily possibly necessarily possibly possibly true", then, under S5, you can cut that down to say that it is simply "possibly true". The specific thing that this enables that Plantinga cares about is that, if something is possibly necessary, then it is simply necessary under S5.

Now, again, S5 is valid. It isn't the only axiomatic system that modal logic can make use of (other, more refined modal logic systems exist that can handle much more complex statements), but it's useful in its own way. The problem does not lie with S5, but with the argument Plantinga tries to make with it.

Plantinga's modal argument runs thusly:

1. A being has maximal excellence in a given possible world W if and only if it is omnipotent, omniscient and wholly good in W.
   
2. A being has maximal greatness if it has maximal excellence in every possible world.
   
3. It is possible that there is a being that has maximal greatness.
   
4. Therefore, possibly, it is necessarily true that an omniscient, omnipotent, and perfectly good being exists.
   
5. Therefore, it is necessarily true that an omniscient, omnipotent and perfectly good being exists.
   
6. Therefore, an omniscient, omnipotent and perfectly good being exists.

The first two points are definitions, and I do not care to dispute them (though a lot of people certainly like to, as the various formulations of the problem of evil demonstrate).

The third point is the ipse dixit/bare assertion point that I raised in my last serious analysis. It is still the point at which the argument collapses on its face, and no other grounds for rejection is necessary. Even Plantinga concedes this, as I have pointed out previously. For the sake of clarification, however, I will continue.

The fourth point is simply an extension of the definitions given above. Plantinga defines a "maximally great" being as one which possesses "maximal excellence" in all possible worlds - that is, a "maximally excellent" entity that is also necessary. Since the previous premise is that this being is also possible, we can now define a "maximally great entity" as a possibly necessary maximally excellent entity.

From there, S5 allows us to discard all but the last modal, and leaves us with the asserted maximally great entity being simply necessary.

And so we're back at the original issue that I pointed out in my last post. This version of the ontological argument, as with all other versions, simply boils down to defining "God" as "something that, if it is possible, must exist", and hoping that the assertion of possibility will be accepted without justification.

The assertion is not accepted, and the argument is discarded.