Moody's overall argument, as presented above, is obviously formally valid,
since it's a simple modus tollens. And, as indicated above,
I think his sub-argument for (1), though vague, can be victoriously specified via
routes I myself have mapped [4].
The problem is the extraordinarily vague premise (2). The
first source of obscurity concerns Moody's underlying modal notions.
In order to analyze these notions, let us use g as
a variable ranging over groups of zombies visiting us from another
world.
And let us deploy the
traditional symbol `
' from modal logic to say that some proposition
or state of affairs p is logically possible (and 
to say that some
proposition p is logically necessary, where
iff 
). Finally, use the predicate P
to denote ``penetrability," where the idea is that if zombies can be found
out, if they can be revealed as zombies, they are said to be penetrable.
One possible contrual of (2), given this (humble) machinery, and the fact
that Moody's gedankenexperiments (as is common) are designed to show that
certain propositions are coherent, and hence logically possible, is
)
) must be the negation of (1)'s
consequent, and that consequent is surely not the proposition that it's not
logically possible that a group of visiting zombies is penetrated. It seems
clear that Moody needs (and wants) to make a claim about the penetrability of
all visiting groups of zombies. What good would it do him to claim that some
groups are penetrable? Or, worse, would good
would it do Moody to claim that it's logically possible that some
groups are penetrable? Such claims are truisms.
After all, some groups of visiting zombies might carry signs
with them declaring that they are zombies. Or, if God is logically possible,
then some groups could show up and be thunderously deemed by the Almighty to
be empty-headed. A much better bet for a more plausible sharpening of (2) is
)
to
indicate `humanly penetrable'):
)
)
Or have we? Isn't it perfectly easy to imagine that a group---call
it the `A Team,' or simply, later, `a'---of zombie visitors
perform always and only those behaviors in keeping with the presence of
consciousness? After all, there is certainly some set of utterances, and
some set of bodily movements, which preclude any inferences to the proposition
that the bearers of these utterances and movements are not conscious. Whatever clever questions humans
pose to the A Team, they come back with the ``right answers." For example,
with respect to the inverted spectrum problem, there is certainly
some set of sets of responses that precludes any suspicion that those giving these
responses are zombies. (That there is such a set of sets follows immediately from
the undeniable fact that we don't suspect, when hearing about ISP from
human philosophers, that they may be zombies.) Surely it is conceivable that
the A Team invariably responds with a set of behaviors from this set.
But now the A Team thought-experiment gives us
follows by existential quantifier
introduction. From this, by quantifier shift, we have
,
which is equivalent to 
, that is, Moody's (2
) is
false---and so his argument fails. It's as easy as that.