By “metaphysically possible” I mean actually possible in the real world. There are some things that are logically possible (such as, “The President of the United States is a Corgi”) but not metaphysically possible (such as, “The President of the United States is, indeed, a Corgi!”).
Well, I am glad I asked. What do you mean that in real life it is impossible for a “corgi” to become president? It is simply not legally allowed at this time, but laws can be changed in a heartbeat. And people could exercise their right to “write-in” and then that dog would become president - just like Caligula’s horse (Incitatus) became a senator.
As things are today, even an extremely smart, 12 years old genius could not be elected, since he is not of the legal age, but, laws can be changed.
Or, let’s consdier, that due to the pervasive hostility toward atheists, it is “impossible” toady that an atheist would elected to be president, even though it would be perfectly legal. Again, customs can change.
The only obstacle against a “possible world” is a logical or physical impossibility. If we cannot agree on this point, it is useless to go on. The concept of “possible world” is a very useful tool, and the ONLY requirement is that the state of affairs would not be logically contradictory. The idea of “metaphysically impossible” is simply nonsense.
But this is precisely what I mean. Any isolated individual can logically freely choose the morally good in any circumstance, and not do something evil. It’s not just a matter of “adding” people, because to add isolated individuals is basically just to multiply the original situation. What I’m suggesting is that we are not dealing with isolated individuals but those in community who compete over certain limited resources. In these kinds of situations, it is less doubtful that individuals will freely choose to never commit evil actions.
And this is where you are wrong. Free will cannot be contingent upon the decisons of others, since it would not be free any more. Let me illustrate with a simple example. Take two people, called A and B. They both face a dilemma. There are four possible worlds:
- both A and B choose the moral option.
- A chooses the moral option, and B chooses the immoral option.
- A chooses the immoral option, and B chooses the moral option.
- both choose the immoral option.
Now the two dilemmas can be independent, or they can be related, it simply does not matter. If option #1 cannot be actualized (for whatever reason), then we need to consider option #2. Person A chose the moral option, and by virtue of his choice person “B” cannot choose the moral option - as such he has no significant free will. Before you ask, the concept of free will presupposes that in a morally significant scenario the agent has the option to choose either the moral or the immoral choice. (Not that the person is free to choose a blue tie or a red tie).
So it is exactly the “free will” which makes option #1 possible - and thus there is a possible world, where both agents (everyone!) chooses the moral option.
The only objection against this proof is that “two” is a small number, and “maybe” with larger numbers the same logic does not hold. Of course that objection is incorrect. Nowhere was the number “2” exploited.
Suppose that we already have a possible world, with “n” moral dilemmas, and all of them are solved in a morally proper manner. Let’s add one one dilemma. There are only two options:
- either there is a possible world, where the “n + 1”-st dilemma is also solved morally, or
- there is no such world - meaning that the “n +1”-st dilemma can only be solved immorally - but that denies free will!
The employed method is called “mathematical induction”, where we prove that a certain property inherits from any arbitrary “n” to the next “n +1”-st case. The same could be done with the binomial theorem, of course.
First, the number of moral agents is not relevant. The same person may face none, one or more dilemmas.
- A possible world is one where there are no moral agents at all. Naturally, in this world there are no moral dilemmas, so this world cannot contain “evil”.
- Another possible world could be with unlimited number of moral agents, but without moral dilemmas. Example could be the “garden of eden”, without the prohibition against the “tree”. If there are no “rules” to obey, or disobey, there are no moral problems, as such no evil.
- Since these are possible worlds, the problem is solved. Of course there are people who insist that such “trivial” solutions “do not count”. So we can accommodate them.
- Let’s consider a world with exactly “n” moral dilemmas. Each dilemma can be decided either morally or immorally. The number of possible worlds where exactly “k” of the dilemmas is solved morally is (n over k), which is n! / (k! * (n-k)!), where k = 0, 1, 2, … ,n - 1, n. The number of possible worlds is exactly 2[sup]n[/sup] - see the binomial theorem. If “k” equals “n”, we have a world, where each dilemma was solved morally (no evil). If “k” is less than “n” then we do have at least one immoral decision and so there is “evil”.
But there is no reason to assert that “k = n” is somehow “impossible”.
So, no matter how we view it, there are zillions of worlds, with unspecified number of moral argents, and unspecified number of moral dilemmas, where all the dilemmas are solved morally. God, of course can “preview” all the worlds, and choose any one of those.