350 RELIGION, MORALITY, FAITH, AND REASONlimited itself in this way. Nor is there any reason why they should, and
there is excellent reason why they do not.
Robust foundationalism
Robust foundationalism has a high standard for proper belief. According to
it, one properly believes a proposition P only if it is logically impossible
that one believe that P and P is false – only where co-presence of belief and
falsehood cannot arise. This limits objects of proper belief to necessary
truths and belief-entailed propositions. Even if those monotheists who
joined Anselm in thinking that God exists is a logically necessary truth are
right, few if any religious beliefs can be properly believed on robust
foundationalist standards. All or most religious beliefs, in that respect, join
There are trees, Texas covers more territory than Rhode Island, There are
pigeons in New York City, and London, England is not located in Delhi,
India, as well as almost everything else anyone believes. This is not a
source of proper concern, since It is proper to believe only what robust
foundationalism says it is proper to believe is itself not something even a
robust foundationalist could properly believe.^8
Religious claims
When we deal with such questions as whether seven is greater than five,
what twenty-nine times forty-eight is, whether the argument form If P
then Q, Q or R, not-Q, hence R or not-P is valid, and whether Necessarily,
P entails Necessarily (Necessarily, P) we deal with things to be thought
through without essential appeal to sensory experience. Reports of sensory
experiences will not tell us what the answers to these questions are; only
abstract reasoning will do that. When we deal with questions of whether
there is a tree in the yard, how apples taste, whether crows will be kept
from crops by playing rock music in the fields, and what black bears eat, we
deal with things to be answered by reference to reports of sensory
experiences; abstract reasoning will not tell us the answers. Sensory
evidence and abstract reasoning is each precious relative to their
knowledge-potential, though they are different; it is foolish not to value
both highly if one values knowledge.
If one looks at the sorts of claim that are central to religious traditions –
the ones presupposed or entailed by the diagnoses and cures that such
traditions offer – it is obvious that they are not merely sensory reports and
they are not merely reports of the result of abstract reasoning. Like