Mike Arnautov
Oxford Philosophical Society Forum (members only)
19 February 2017

Cecil [not the real name] assures us that our actions are driven not by causes but by reasons. My initial thought was to argue that his position is undermined by the existence of predicate dualism approaches to the mind/body issues. But then I thought: why not tackle the matter from the other end? Let's do that!

First, though, I must clarify some of the terminology used in my argument below.

• By 'reason' I do not mean one's capacity to reason, but a reason one may have for something – e.g. there being something I want to watch on the TV is a reason for turning on the TV.
• 'Reason' need not be a singular, atomic reason – e.g. taking an umbrella because it might rain and because I do not wish to get wet, can be viewed as a reason, rather than as two separate reasons.
• By 'reason' I do not mean only logical and/or conscious reasons. How I feel about things may also constitute a reason, however unreasonably so. And e.g. being angry is not excluded from my reasons for an action, however much I may wish to deny being angry.
• Finally, by 'action' or 'acting' I do not mean only physical action. E.g. making a decision also counts as an action. Deciding to take an umbrella is an action whether or not I do take an umbrella in the end.

Bearing this in mind, let's consider reasons and actions. When I act, there are two possibilities:

1. I act for no reason, in which case I act arbitrarily.

2. I act for a reason. There are two possibilities:

2.1 I act for a reason necessarily. There are two possibilities:

2.1.1 I have but one available action or

2.1.2 I have a range of actions to choose from. I may

2.1.2.1 Choose one action for a reason – so add that reason to reasons already included in the overall reason and go back to (2).

2.1.2.2 Choose one for no reason – i.e. arbitrarily (which would be at odds with the premise 2.1, of course, but that's humans for you! :-)).

As a finite being, I cannot continue with the loop (2) -> (2.1) -> (2.1.2) -> (2.1.2.1) indefinitely. The loop must terminate either in (2.1.1) or in (2.1.2.2).

2.2 I act for a reason contingently. Again, there are two options.

2.2.1 I act arbitrarily within the constraints imposed by my reason or

2.2.2 there is a further reason involved – so add that reason to reasons already included in the overall reason and go back to (2).

As a finite being I cannot continue with the loop (2) -> (2.2) -> (2.2.2) indefinitely. The loop must terminate either in (2.1) or in (2.2.1).

The upshot is that my actions are either fully determined by my reasons, or I act arbitrarily within constraints imposed by my reasons, or I simply act arbitrarily.

Since my reasons must be a part of my state of mind (I cannot act for a reason if I don't have that reason), this looks entirely like a mixture of determinism and randomness, just translated into the mental sphere. Note that this argument does not rely on any connection between reasons and causes. It merely points out the parallel between determination by reasons and determination by causes.

So here are some questions for Cecil and for anybody agreeing with him... Is determination by reasons any less problematic than determination by causes? If yes – why? If no – why is it so important to rule out any kind of idenity between physical causes and mental reasons?