On Free Will as Categorical and Conditional Freedom

Sammanfattning: This dissertation is about a complex of problems, related to the question: ‘Can we ever act differently from how we in fact act?’In Part I, the meaning of ‘can’ and ‘could’ is discussed. It is argued that when we say that an agent could do something he didn’t do (in a sense of ‘could’ involving control), this means, in what is called ‘Decision-Contexts’, that he was conditionally free to do it, and, in what is called ‘Strong-Autonomy-Contexts’, that he was categorically free to do it. That he was conditionally free to do it means, roughly, that if he had decided to do it he would have done it. That he was categorically free to do it means, roughly, that, given the fixity of everything causally relevant, he had control over whether he did it or not. Therefore, that he was categorically free to do it seems to imply that he could agent-cause it, i.e., that he could cause it in a way, not reducible to event causation. It is argued that categorical freedom is necessary for moral responsibility and strong autonomy, but not for making sense of our praxis of saying that someone ought to do so-and-so.In Part II, Peter van Inwagen’s Consequence Argument for the incom-patibility of determinism and our ability to act otherwise is discussed. Dif-ferent versions of the argument are compared. It is concluded that the argu-ment is sound under a certain interpretation but unsound under some others.In Part III, it is argued that each member of a certain family of arguments (Galen Strawson’s argument for the non-existence of moral responsibility, the Consequence Argument and the Mind argument) has as an implicit prem-ise that only event causation takes place, and that no one can do anything about that only event causation takes place. An argument is given, in the form of a series of thought-experiments, for the plausibility that this premise is false, and by doing so a first step in giving an argument for the existence of categorical freedom is taken.