Sunday, April 18, 2010

Balaguer on Free Will, Part Two

There are two steps in Mark Balaguer’s argument (contained in chapters 3 and 4 of his book, respectively).  In the first step, the subject of this post, he argues that the question of freedom reduces to a question of indeterminism in decision-making.  In the second step (to be discussed in the next post) he argues that the presence or absence of the relevant indeterminism is an open scientific problem.


Appropriate non-randomness


Balaguer says that a person is free “IFF she makes at least some decisions that are such that a) they are both undetermined and appropriately non-random, and b) the indeterminism is relevant to the appropriate non-randomness in the sense that it generates the non-randomness, or procures it, or enhances it, or something along these lines. (sec.3.1, p.65-6).”

What is “appropriate non-randomness”?  For a decision to be appropriately non-random it needs to have been “authored and controlled by the agent in question, that is, it has to have been her decision and she has to have controlled which option was chosen. (p.66)”

Now, some think libertarian freedom (L-freedom) is untenable because they think indeterminism can’t help.  An undetermined event is just accidental, it “just happens”, or it’s “just random”.  Or putting it the other way, for a decision to be appropriately non-random, it needs to be authored and controlled by us; this would mean we “determine” the choice for certain reasons.  So they can’t be undetermined.    An event can’t be both undetermined and appropriately non-random.

Or, maybe the critic can concede that there is some indeterminism in our decision-making process which doesn’t completely destroy non-randomness.    But nonetheless, it’s hard to see how indeterminism in the process could generate or help or increase appropriate non-randomness.  Inserting an undetermined event into the process adds randomness, not non-randomness!

Balaguer says these sorts of reasoning are wrong.  If certain decisions are undetermined in the right way, then they are appropriately non-random, and the indeterminacy in question does increase or procure the appropriate non-randomness (note in this discussion we are assuming materialism and event causation).

Torn decisions

To make the case, Balaguer focuses on particular decisions he calls “torn decisions”, where we have reasons for 2 or more options, no belief as to which is best, and we just choose without resolving which is best.

He says:  If a torn decision is undetermined at the moment of choice, then it is L-free.  The final decision in the torn decision scenario, which is undetermined, is conscious, intentional, and purposeful.  These are decision-events which settle which option is chosen in a conscious purposeful way.  So they are indeed authored and controlled by the agent in question, but are undetermined.

Furthermore, the indeterminism here increases or procures the authorship and control:  in fact the undetermined case maximizes authorship and control compared to the case of a causally determined decision.  Remember, it is a torn decision:  making an undetermined choice in this case is making it under maximum authorship and control.  In cases where some determining event (external or subconscious internal) tips the scales, there is less authorship and control.  And, this result is important, he says, because we make a lot of torn decisions.

(Note:  if people still have the intuition that internal indeterminism can’t suffice for freedom, it’s probably a hangover from thinking of freedom as necessitating old-fashioned interactive dualism.  But if the indeterminism is in me, then it is mine and it makes my choice free in a manner we can’t improve on and remain naturalists).

Are these decisions rational?  Yes, because the choice is made for good reasons, even though the reasons weren’t’ enough to tip the scales vs. another option for which we also had good reasons.   There is “plural” authorship, control and rationality, since at least one unchosen option, if it had been chosen, would have met the criteria also.

What about untorn decisions? Well, other types of decisions (e.g. where we have reasons which cause a “lean” toward one option over others) are probably L-free, too, if there is indeterminism at the point of decision; it’s just the torn case brings this out most clearly.  Also, even determined decisions could be said to be L-free is they are made by an ”L-free person”, that is, one who makes a significant number of undetermined L-free decisions also.

So, the question of freedom reduces to the question of indeterminism at the point of decision.

No comments: