Campbell Purton's website

HOME      Publications:      BUDDHISM        COUNSELLING   



Published in:  Philosophy Vol. 54 (1979)  pp. 108-109



Agent-determination and Freewill




In his article ‘Who’s in Control Here?’ (Philosophy 51 (1976), 421-430)

Robert V. Hannaford presents an interesting defence of the view that a

human agent’s freedom in determining what he does is compatible with

an underlying physical determination of the movements in which his actions

are embodied. Much of what he says seems to me to be correct; in par-

ticular that there is an important distinction to be made between physical

agency and voluntary agency, between the physical determination of one

event by others, and the determination of a course of action by the agent.

However, I want to argue that Hannaford does not come to terms with a

crucial point in the argument that human freedom is incompatible with

physical determinism. I will illustrate the point by an example and then

consider Hannaford’s reply.


Suppose my car’s brakes fail on a hill, and that at the bottom of the hill

the road forks. There is no way of stopping the car before it reaches the

junction, and if I do nothing the car will follow the left fork because of the

way the road slopes. As we ordinarily understand the situation I can choose

to turn the wheel to the right or I can let the car take its natural course to

the left. To say that I have the choice is to say that the future is still open

for me; that I have not yet determined which way the car will go. Now

Hannaford’s claim is that this opportunity for me to choose is compatible

with its being physically determined which way the car will go. Yet if it is

physically determined (because of the laws of physics, the state of my brain,

etc.) that the car will turn to the left, then it follows that the statement ‘I

will turn the wheel to the right’ is false. For my action of turning the

wheel to the right logically involves its turning to the right. But then the

future is not open as between my turning the wheel to the right and letting

it turn to the left. Note that this argument does not involve, as Hannaford

wrongly supposes, the assumption that actions are just a species of move-

ments or other physical events. All that is required is the unexceptional

point that from a certain physical-event-description (the wheel will turn to

the left) it logically follows that a certain action-description (I will turn the

wheel to the right) cannot be true.


Hannaford briefly takes note of the sort of objection I have just illus-

trated, and his reply is that ‘no doubt purely physical events could take

place in brain states so that we could not avoid some .specific movement

. . . but to the extent that [brain] operations or other physical events make

only one response possible, we would refuse to describe the response as the

person’s choice or as his voluntary action’. But this will not do: the thesis

of physical determinism is precisely that physical events always make only

one physical response possible; and as we have seen there are quite ordinary

cases (not involving such things as brain operations) where we can deduce

from the fact that only one physical outcome is possible the fact that only

one action is possible, or at least that of two seemingly possible .actions

only one can be carried out. Now Hannaford himself admits that where

only one action is possible there is no real choice, no determination by the

agent, yet if ‘determinism is true’ it is almost always going to be the case

that only one action is possible. I say ‘almost’ here merely to exclude

possible cases in which the two actions between which choice is to be

made are such that they will be embodied in qualitatively indistinguishable

physical events  -  if there are such cases they are clearly the exception

rather than the rule.


I conclude that Hannaford’s attempt to reconcile voluntary agency with

physical determinism fails, and that the admittedly important distinction

between physical determination and agent determination has no import-

ant bearing on the question of whether free action is possible in a physically

deterministic universe. Although actions are to be distinguished from the

physical events in which they are embodied, the links between actions’ and

physical events are too tight to allow room,  in a physically deterministic

universe, for that freedom which we cannot coherently doubt that we possess.


University of St Andrews

Return to beginning of paper


HOME      Publications:      BUDDHISM        COUNSELLING       FOCUSING       PHILOSOPHY