Monday, April 9, 2012

The Dilemma of Libertarianism

I have encountered the question of freedom recently in three related areas: moral responsibility, the problem of evil, and God's knowledge of the future.  These topics have been raised in presentations at conferences, informal discussions with others in philosophy, and readings by noted philosophers of religion.  In each case "freedom" was taken to mean "libertarian freedom," which says that a person is free only if that person could have done otherwise (was not determined).

William James wrote of this in his essay "The Dilemma of Determinism" in which he describes a murder, and argues that determinism makes this act a necessity and something about which we cannot feel regret, whereas open possibilities leaves room for regret and moral outrage.

By way of contrast, I'd like to suggest a dilemma for libertarian freedom.  The dilemma is that if we accept the definition of libertarian freedom we either admit of uncaused events (under various names, like agent causation, reasons are not causes, indeterminent causation, underdetermined cause) and all the consequences of this or we must give up freedom as the possibility to do otherwise.

The alternative definition of freedom is that a person's choice is free if the person did what he wanted to do.  This is compatible with determinism because what a person wants to do is caused by various factors such as the person's beliefs and values.  Furthermore, a person's beliefs and value are themselves part of who the human is, and if the human is created by God then God has predetermined what the person will want to do.

Philosophers like Immanuel Kant and William James found the idea of a compatibilist view of freedom repugnant to moral responsibility.  In its place, Kant made room for uncaused events in the noumenal realm where freedom resides, and James made room for uncaused events in the open possibilities of the future.  Do uncaused events save moral responsibility?

If a choice is uncaused, then it is not clear in what sense it is a choice of the person in question.  If the choice is caused by the person but the person is uncaused then this person has in some sense been brought about by "nothing."  So the moral problem quickly becomes an ontological problem and then a logical problem.  If an event is uncaused, then this means whatever proceeded it was not sufficient to bring it about, and there is nothing else that is sufficient (or else it is determined), therefore it had no sufficient cause.

If some events can be brought about by nothing then why not others, or all events?  Perhaps all being can come into existence from non-being through an uncaused event.  I argue that once an uncaused event is admitted others cannot be consistently denied elsewhere.

Furthermore, the difference between being and non-being is so basic that to confuse them on this point is a significant mistake that reverberates throughout all the rest of one's thought.  To believe that being (or an event) can come either from non-being (no cause) or from being is to suggest that on this point being and non-being are not different.  And yet they are different in every respect.  The theist who relies on uncaused events for free will solutions to the problem of evil or providence cannot consistently reject them as the source of all being.  One's thoughts might be uncaused.  Any appearance of causation might be mistaken, such as the appearance of causation between thoughts, or between words and thoughts, or between premises and conclusions.

In future posts I'd like to apply this to popular solutions such as the free will defense to the problem of evil, and to molinism (middle knowledge) and soteriology.


  1. I think this is a well written and straight forward post. One question though, quite a few are unashamed of their position implying uncaused events. Why do they not see it as obviously incoherent and worthy of derision?

  2. Great question. I suspect they don't see the treat of uncaused events. Do they think these can be limited to the will? If we allow one uncaused event, why not others? Why have causation at all? I suspect their background assumption is a kind of empiricism, and empiricism cannot support causation. What explanations have you heard?