Further Reflections on Atheism 203This is a clever argument. The supposition that God is maximally great
implies that God exists in all possible worlds. If so, he exists in that possible
world which is the actual world. But what about the premise that if God is
maximally great, he exists in all possible worlds? Unlike Plantinga I follow
Quine in parsing proper names as predicates so that for example ‘Socrates’
becomes ‘the x such that x socratises’ which denotes nothing if there is no
socratiser. (You can think of the predicate ‘Socrates’ as (say) ‘has a snub nose
and fought in war and taught the author of the Republic’ or things of this
sort.) Unlike Kripke and Plantinga I don’t treat proper names as so-called
‘rigid designators’, but not much turns on this for the present argument.
However, it is not clear that anything is maximally great in my world, since
it is logically possible that in all worlds any degree of greatness could be
exceeded by a greater. That is, one might be sceptical about the premise that
a maximally great being is possible.
At the propositional level it is provable in C.I. Lewis’s system S5 (see
Prior,Formal Logic^7 ) that if it is necessarily the case that if possibly p then p,
then it is the case that if possibly p it is the case that if possibly p then
necessarilyp. We cannot deduce the consequent of the main hypothetical
from its antecedent in the general case because we do not have ‘if possibly p
thenp’ but Plantinga’s moves with possible worlds suggests that in the special
case of ‘a maximally great being exists’ we may say ‘if possibly p then p’.
Equally, however, as Plantinga concedes, there could be a counter-
argument. It seems possible that in no possible world is there a maximally
great being because however near to maximality a being is, there is another
possible world which has a still greater greatness.
Let us look at the matter more simply and omitting steps. Let p abbreviate
‘a maximally excellent being exists’. Plantinga has argued in effect for the
soundness of the deduction from ‘possibly there is a maximally excellent
being’ to ‘a maximally excellent being exists’, i.e. in the notation of propositional
logic from ‘Mp’ to ‘p’, where ‘M’ abbreviates ‘possibly’ and ‘p’ abbreviates ‘a
maximally excellent being exists’. This is of course not a valid deduction in
pure propositional logic because Plantinga has gone through quantified modal
logic and also has used some definitional statements about maximality and
excellence. These could be queried and an argument could equally be con-
structed from the atheist ‘notp’ to ‘not possibly p’. I have my Quinean doubts
about quantified modal logic cum possible world semantics. But allowing the
logic, the atheist could question one or other of Plantinga’s assumptions. We
might question whether a maximally excellent being is possible. For any
degree of excellence there might always be a greater one.
My conclusion is that Plantinga’s argument is of considerable subtlety
(and I have skated over some moves that he makes explicitly) but I con-
clude that it cannot be used by the theist to convince an atheist (the Fool as
Anselm calls him).