An agent has free will iff it engages in action<->reflection cycles.
Also:
An agent A "could have taken" action X vs Y: ActX(Env|CharA) vs ActY(Env|CharB) in the sense that there is a graph "of some sort" linking their CharA to the CharB which could realise ActY.
Eg., Bob can either steal the money or return it. Bob chooses to steal it. Bob is morally responsible for this act because there is a realisable "action<->reflection" path from bob's actual character to the one he should have had. If bob had had this character, he would not have stolen.
This is a very interesting definition of "free will" and not one I would expect most philosophers to agree with, but more measurable and useful than most would propose. In this case "free will" reduces to self-modifying action with randomness injected by the environment, which feels more like a stochastic model than genuine self-determination. That said, the existence of a "reflection" function is still the primary determinant of any moral responsibility.
It's really just a restatement of compatibilism, which has been the major position in philosophy for a long time. And I think, in practice, what most people actually think.
Only somewhat recently have philosophers intuitions changed a little more to incompatibilism.
Action: Act(Environment|Character)
Reflection: Act(Character|Environment)
An agent has free will iff it engages in action<->reflection cycles.
Also:
An agent A "could have taken" action X vs Y: ActX(Env|CharA) vs ActY(Env|CharB) in the sense that there is a graph "of some sort" linking their CharA to the CharB which could realise ActY.
Eg., Bob can either steal the money or return it. Bob chooses to steal it. Bob is morally responsible for this act because there is a realisable "action<->reflection" path from bob's actual character to the one he should have had. If bob had had this character, he would not have stolen.