162 J.J.C. Smart
of the universe, the fact that it exists at all. This is surely not enough for
theism in any sense in which it need be distinguished from atheism. I think
that Haldane’s and Aquinas’s point is that God, as they think he is proved to
exist, is something only very abstractly described, for example as simple and
the cause of the world. As Haldane points out, the proofs are not claimed to
prove a thick ‘whatness’, i.e. God as conceived by some particular religion.
But Haldane rightly points out that there must be some ‘whatness’ in the
conclusion. I also agree with Haldane that any worthwhile concept of God
must describe God as eternal in the sense of being outside space and time, a
changeless cause of change. I would add that changelessness here would be a
matter not of staying the same through time but of being like the number 7,
say, neither changing nor staying the same. I concede that Haldane gives
a subtle and attractive form of the cosmological argument. Nevertheless I am
not persuaded, for the usual reasons as adumbrated in chapter 1. I do not see
how God’sthatnessandwhatnesscan be the same reality. To say this would
surely be to treat ‘exists’ as though it were a predicate.
The arguments that I used against the cosmological argument do not,
however, depend on any extreme empiricism or positivism about meaning,
which would deny any meaning to talk of the transcendental. Indeed, I think
that this ascent to the transcendental can happen in science when meaning is
transferred upward through the hypothetico-deductive method, and further
through considerations of simplicity when the empirical evidence is indeci-
sive. Hence I do go a long way to agree with the remarks about meaning and
the transcendental (see p. 134). My objection to the hypothesis of theism is
the unclarity of the notion of necessity that would be required. On p. 135
Haldane perhaps rightly objects that I give insufficient attention to the way
in which the notion of necessity arises in the argument from contingency.
He says that ‘what we are led to is the existence of something which exists
eternally, which does not owe its being to anything else and which cannot not
exist’. The nub is in the last clause. Following Quine, my notion of modality
is highly contextual. Except for mere logical necessity, where the background
assumptions are null, the notion of ‘can’ is relative to these background assump-
tions. ‘It cannot be the case that p’ can be said when mutually agreed back-
ground assumptions imply (by first order logic) that not-p. For example, ‘you
cannot live without oxygen’ can be said because ‘you do not live without
oxygen’ follows from agreed assumptions about human physiology. Perhaps
the background assumptions could be assumptions of theological theory, or
‘necessary’ here be a primitive of that theory. This, however, would make
theology question-begging and ready to be sliced off by Ockham’s razor.
As I pointed out in chapter 1, the universe could fill the bill of something
that does not require anything else for its existence. According to the atheist
there is nothing beyond the universe and so it is not dependent on anything