need the paucity of fine-tuned universes simpliciter, but simply in our local area. Besides,
we do have good reason to think that if we look at all possible universes, it is not the case
that the majority of them can support life.
Finally, the many universes anthropic principle (MUAP) can be brought in. This principle
states that there exist infinitely many universes, either sequentially or simultaneously, and
thus it is not improbable that some of them would contain observers, while evidently we
can observe only a universe that can contain observers. The MUAP claims in general that
we have no right to be surprised to observe a feature of the universe necessary for the
production of intelligent life, since it is likely that at least one of the infinitely many
universes would contain that feature, and we cannot observe any other. Thus, perhaps,
there are infinitely
end p.131
many universes, in which case we would expect that at least one would exhibit the kind
of fine-tuning that makes life possible, and obviously we couldn't observe any other.
There are two forms the MUAP takes. First, it might be that, necessarily, all logically
possible universes concretely exist, as in David Lewis's (1986) extreme modal realism.
Unfortunately, Lewis's theory runs into a multitude of paradoxes. To give just the
simplest, note that Lewis's theory undercuts inductive reasoning. Suppose God phoned
you and, after having assured you with sufficiently impressive miracles that he is God,
told you that he created at least as many universes with the same past as yours in which
gravity fails to hold tomorrow as ones in which gravity continues tomorrow, but
neglected to tell you which kind of universe he put you in. By standard canons of
reasoning, you would be rationally required to assign at least as great epistemic
probability to the claim that the law of gravitation will not hold tomorrow as to the claim
that it will. Therefore, your inductive inference that tomorrow gravity will hold as it has
always held would be undercut. But Lewis's theory is just like this call from God: Lewis
tells us that all logically possible universes exist, and certainly then there will be at least
as many worlds that have the same past as this world in which gravity will fail to hold
tomorrow as ones where gravity will continue as before. Thus, Lewis's theory gives data
undercutting induction, and hence we should reject Lewis's theory.
Alternatively, it could be that all or infinitely many universes exist satisfying the same
basic laws of nature, albeit with different constants in them. It does not matter here
whether these universes exist simultaneously or sequentially. This version of MUAP,
however, fails to block the question of why these basic laws of nature hold rather than
others. It might, after all, be that the vast majority of possible sets of laws of nature could
not support intelligent enmattered life because the vast majority would involve massive
irregularity. For instance, intuitively, there are a lot more possible laws of gravitation that
involve many discontinuities and irregularities in the formula for the force as a function
of the distance than there are highly regular laws, and it might be that life could exist only
in what is intuitively only a small fraction of the universes governed by such irregular
laws, though making these intuitions more precise would be a nontrivial task.
It is worth noting parenthetically that a multiple-universe theory has also been used to
neutralize the argument against theism from evil. Donald Turner (2003) proposes that a
perfectly good God would create all universes that are sufficiently good, that is, which it