Atheism and Theism 145between on the one hand the notion of objective probabilities rooted in the
natures of physical systems, and on the other the idea of behavioural tenden-
cies issuing from habitual rational agency. Physical events and human actions
may both admit of a high degree of predictability without either resulting
from deterministic causes. In the case of the former, reliable prediction is
based on natural propensities, in the case of the latter upon rational inclina-
tions and responses.
As I argued earlier, the relation between an agent’s reasons and his actions
is not in general a causal one, at least as causation is typically understood.
To explain what someone is doing it is not necessary to identify something
‘lying behind’ his movements – in a more or less literal interpretation of
those words. Action is the exercise of rational and appetitive powers. To
understand how an agent may act freely on a given occasion one needs to ask
how it is possible that a human being should act at all. Stones are moved
by external forces but, as the scholastics say, agents are moved ‘from within’
(ab intrinseco). What this means is something very different from the
neuropsychological events envisaged by present-day causal theorists. An
adequate theory of intentional behaviour needs to combine the idea of
non-random indeterminacy with that of intelligent sources of action.
We are rational animals; living things whose principles of organization and
functioning are ordered towards a form of life that is responsive to reason.
Voluntary action is a capacity of rational agents expressed in intrinsically
intelligible behaviour. When a human being acts there need be no event in
the agent prior to the action and which is its immediate cause. The only
required ‘source’ is the very agent whose powers are exercised thereby. In
a mature human being these powers are possessed continuously even when he
or she is not doing anything ‘in particular’. Thus most action calls for no
explanation, for if one knows that one is dealing with a rational animal then
there is no need to say why it is doing things, for animals are active by nature
(even sitting quietly and sleeping are activities). Activity is the norm, and
most activity is normal, i.e. it is what would be expected of a reasonable
human being in familiar circumstances. The first point is a general one true of
all agents, rational and otherwise; but the second derives from the fact that if
we say a piece of behaviour is an action then we are committed to the claim
that in doing it the agent was aiming at some end (even if this was just the
performance of an action of that sort). Action differs from mere movement in
being purposeful, in aiming to advance an interest of the agent. This thought
is what lies behind the scholastic doctrine that all action is performed under
the species of the good (sub specie boni).
An obvious question to ask is whether the claim is that every action is
necessarily directed towards a real good or merely to what is believed by the
agent to be a good. Clearly the second interpretation is weaker and may seem