Given that there are several types of “free-will” defenses to the problem of evil, each taking on subtle nuances to include or explain different aspects of the environment (e.g. Plantinga's attribution of seemingly gratuitous evils to Satan), they all share the idea that evil is the consequence of free will granted by God to (some of) his creations, and that free will is good enough in itself to justify any evil. It seems to me that this defense only works when assuming incompatibilist (libertarian) notions of free will.
Incompatibilists, in general, hold the following to be a necessary condition of free will:
The Condition of Ultimate Responsibility (UR): “to be ultimately responsible for an action, an agent must be responsible for anything that is a sufficient reason, cause, or motive for the action's occurring,” and an agent that has true free will must be ultimately responsible for her actions (Kane, Robert. "Ultimate Responsibility." A Contemporary Introduction to Free Will. Oxford: Oxford UP, 2005)
The interesting and relevant connections between UR and the free will defense surface when discussing reasons, causes and motives in context of a defense of the Anselmian God.
For our purposes, let's make a few assumptions:
A1: UR is true
A2: Having control of the all of the sufficient reasons, causes, and motives for an agent's action (usually thought of as the laws of physics, events in the past, an agent's genetic makeup, and the agent's own initial/early desires and character) is impossible for any agent that exists and is not God
A3: God has the power to control these things (by virtue of his omnipotence and omniscience)
A4: Determinism is true
If we then postulate a version of free will that is OK with these assumptions and maintains that free will is true, it seems like it will necessarily be the case that any attempt to use this view of free will will be vulnerable to the “it could be better” objection. If God controls all sufficient reasons, causes, and motives for an agent's action's occurring, or really even just one of those (the essential (e.g. dispositional or genetic) makeup of His creation), it follows He could always have set the circumstances for every agent's every action to be good, or minimally unpleasant (if unpleasantness/evil is necessary for good to exist); He could have reined in the magnitude of evils that exist AND preserved free will at the same time. And of course, if these heinous acts are not necessary for free will to exist, which would be the case as postulated, then they cannot be justified by free will's existence; in short, we could have had free will without heinous acts.
So, what is the employer of the free will defense to do? What are the alternatives? She could challenge A2, arguing that we can, in fact, control all the sufficient reasons, causes, and motives for our actions. But, this seems prima facie false, and would require a ton of argumentation to support this claim, not to mention a complete rethinking of other positions in metaphysics, philosophy of religion, and epistemology, among others. The burden of proof for this objection is on the proponent of this objection, and will most likely turn out impossible to prove.
She could challenge A3, arguing that God cannot control all the things that will affect the future actions of an agent. But, in a deterministic system, this seems to conflict with the Anselmian conception of God as omnipotent and omniscient.
She could object to A1, but it seems as though this objection will undermine the free will defense itself for the same reason given above; if an agent does not have to be ultimately responsible for her actions to have free will, then God could make the sufficient reasons, causes, or motives for her actions in such a way that she causes no (or little) evil while also preserving her free will. If God created every human being with an essentially good nature, and this nature did not affect free will (as would be the case with compatibilist, soft determinist theories), it does not seem like the heinous evils follow necessarily from free will (and therefore can't be justified by the existence of free will).
Taking an indeterminist approach (rejecting A4) seems to be the only way to make the free will defense sensible. Indeterminism allows for true randomness, might more easily lend itself to explaining God's ability or inability to control future events, and posits that there are some actions that don't necessarily have sufficient reasons or motives, and that, on account of the random aspects of the view, even if characters and motives could be defined for a person, they would merely “incline without necessitating” the relevant agent to act in a certain way.
What do you all think? Can the free will defense make sense within a soft determinist framework, or should those using the free will defense brush up on their libertarianism?