분위기도 환기시킬 겸, 최근에 쓴 과제를 올려봅니다. 짧아서 읽기 괜찮지 않을까 싶어요. 여담으로 (제가 생각하기에) 헤겔이 자유의지에 갖고 있는 스탠스를 옹호하는 글입니다. 헤겔을 언급하진 않지만요. 그리고 요약 + 반박으로 이루어져있는데, 요약이 조금 아쉬운 면이 있습니다. 근데 고치긴 귀찮기도 하고, 아주 나쁘진 않아서 그냥 올립니다.
A Compatibilist Account of the Agent-Cause Account of Free Will: A Frankfurtian Argument
Agent-causal accounts of free will hold that, if an agent acts with free will, her action is caused by herself, not by any of the prior causes. If I, for example, attend a lecture with free will, champions of the agent-causal accounts will say that my attending the lecture is caused by me and nothing else. Furthermore, the champions hold that, if I act with free will, nothing causes me to cause my action; in attending the lecture, I am the uncaused cause for my action.
These accounts of free will are often subject to an objection called “rationality objection.” The idea is that, if none of the prior events caused my actions, then my having certain reasons for attending the lecture cannot cause myself to attend the lecture. But there certainly seems to be a causal role played by my reasons; if it were not the reasons for going to the lecture (e.g. I will lose my participation grade if I didn’t), then I would not have attended the lecture. But if so, these early agent-causal accounts fail: if prior events cause the action, then I cannot be the cause. For regardless of what I do, I am determined to act in a certain way, in which case, there is no “causal work” left for me.
In his “Toward a Credible Agent-Causal Account of Free Will,” Clarke replies to the rationality objection. To see Clarke’s point, suppose that the world is probabilistic, such that, if an event E occurs, A or B happens, and which of them happens is probabilistic. Note that in this case, E doesn’t causally necessitate either A or B; E just causes there to be possibilities. This lack of causal necessity leaves some causal work for agent causation. For example, suppose that I have a set of reasons to attend a lecture, e.g. I want to get a full participation grade, and also a set of reasons to go for a walk instead, e.g. the weather is nice. But these reasons don’t causally necessitate my action; I might choose to attend the lecture or to go for a walk. What does causally necessitate my action is myself: I prioritize or order sets of reasons in such a way that I attend the lecture, e.g. I decide that my participation grade is more important than the enjoyment I’d experience going for a walk, etc. Then, Clarke has successfully replied to the rationality objection, by ascribing some causal role to previous events and also leaving some causal work for the agent.
Now, throughout the paper, Clarke insists that his account is a libertarian account, which holds that free will is incompatible with determinism. We can speculate why Clarke takes his account to be a libertarian account: (1) if determinism is true, there is only one possibility for the agent caused by a previous event, (2) if there is only one possibility for the agent caused by a previous event, there is no causal work left for the agent, and (3) therefore there will be no causal work left for the agent in a deterministic universe. If, for example, the only action I could take was to attend the lecture, e.g. I can’t prioritize my reasons such that doing so differently would cause a different action, then there is no causal work left for me. Therefore, the agent-causal account seems incompatible with determinism.
But I think this account is compatible with determinism, because (2) is false. To see why, consider the following thought-experiment. Suppose that Jones is about to vote for the Republicans or the Democrats. But further suppose that Black, a person with a mysterious power, wants Jones to vote for the Republicans, such that, if Jones tries to vote for the Democrats, Black will manipulate Jones and cause her to vote for the Republicans, but if Jones tries to vote for the Republicans, Black will not do anything. Now, in the case of Black manipulating Jones, Jones certainly was caused by Black to vote for the Republicans. However, in the case of Jones voting for the Republicans without Black’s manipulation, Jones caused herself to vote for the Republicans. Jones had certain sets of reasons to vote for the Republicans or Democrats, and it was Jones that ordered the reasons in such a way that she voted for the Republicans. So, in voting for the Republicans, there was causal work left for Jones. But could Jones have voted for the Democrats? Certainly not; Black would have manipulated Jones to vote for the Republican. Then, what this thought-experiment shows is that (2) is false: there is only one possibility for Jones, but there still was causal work left for Jones. Therefore, Clarke’s account of free will is compatible with compatibilism