First is the view that talk about God involves the use of “models,” an idea fully
developed in Barbour (1974). A model in science, such as the familiar billiard-ball model
of a gas, is “an imagined mechanism or process, postulated by analogy with familiar
mechanisms or processes and used to construct a theory to correlate a set of
observations” (30). It is not a “literal picture” of reality, but it can used to suggest a
variety of features of the reality under investigation. Religious models have a similar
structure and status. They are based on analogies; they too are not literal pictures of
reality, though they can serve to suggest and point to important features of God (50).
They also serve to express attitudes and direct action. But unlike the situation in science,
where once a theory has been suggested by a model, it can eventually be developed so
that the model that gave birth to it can be left behind (though still useful for an
imaginative grasp), in religion models are the closest we can come to a cognition of God.
Barbour does not make fully explicit why he thinks that we cannot adequately grasp
truths about God directly. But he seems to think that God is so radically different from
any creature that no creaturely terms portray God as he is in himself. Even the most
conceptually elaborated theology is dealing with a model by which we can get enough of
a grasp of God and of divine-human relations to inform our religious thoughts, feelings,
attitudes, and practices. We can never zero in on just where these models fall short of an
adequate grasp of God himself.
My final example of these “analogy without a completely explicit specification of the
limits thereof” views is taken from two essays by I. M. Crombie (1955, 1957). Crombie
too feels that even our best efforts fall short of portraying God and his activities just as
they are. He, like Barbour, is not very specific as to what he thinks keeps us from going
further, but again there is the general sense that God is too infinite, too radically different
from creatures, to allow terms taken from talk of creatures, however modified, to be true
of him as he is.
Going back to reference for the moment (Crombie is one of the very few who realize that
the subject and predicate of statements about God present different problems), Crombie
makes the interesting suggestion that reference to God is achieved by directing one's
attention out of the natural world “in a certain direction.” The direction is given by, for
example, reflecting on the contingency of the world and looking toward a contrasting
necessary being (something Crombie thinks one cannot properly conceive), or by
reflecting on our imperfections and thinking of an absolutely perfect being that would be
wholly without such flaws.
To return to our present concern with predicates, Crombie holds that “when we speak
about God, the words we use are intended in their ordinary sense (for we cannot make a
transfer, failing familiarity with both ends of it), although we do not suppose that in their
ordinary interpretation they can be strictly true of him. We do not even know how much
of them applies” (1955, 122). The beginning of this quote implies a literal, indeed
univocal use of the predicates, and the end of it rules out analyzing that literal meaning
into a part that strictly applies to God and a part that does not. And so, like Aquinas and
Barbour, Crombie leaves us wondering how we can suppose we are saying anything
reasonably determinate about God. His originality consists in the answer he gives to this
challenge: “The things we say about God are said on the authority of the words and acts