GrahamH wrote:..
zoon wrote: I'm making the same point from a different angle: I think the useful point is whether punishment, or the threat of punishment, will influence the agent so that they are less likely to repeat that action in the future. This is where the question of whether or not an action was free-willed influences decision-making. For example, if someone barges into me, I will probably be annoyed with them, but if it turns out that someone else pushed them into me, my annoyance is transferred to the person who now looks like the originating agent. Being angry with the person who actually bumped into me is useless, a waste of effort and social capital, the originator is the one who can usefully be targeted. Perhaps this is where our evolved intuitions (in this case, that the originator's action was free-willed, but the other person's bumping into me was not) do in fact track useful distinctions, at any rate most of the time.
It might make more sense to dispense with a concept of "originating agent" or "original cause" and instead look at points of control. It doesn't matter where the origins of the bump might be, whether inside or outside this or that body. You can often identify points where you can interact to exert some control. Speaking to the person who bumped into you could influence how alert that person is to those around them, say. If you direct your attention to the other person you may discover they are unsteady on their feet because they are unwell or partially sighted and assisting them may have greatest control on future collisions.
Someone assuming free will is perhaps more likely to respond more aggressively and with less effect.
Yes, the examples you give are of cases where it would be an error to punish or reward the person, because, as you say, assistance would give better outcomes. But this is already a part of the system of rewards and punishment and moral responsibility, that people need to avoid using it in such cases: those are the cases where the person was not held to be exercising their free will. Group sanctions are sometimes (as in those cases) inappropriate, but it does not follow that they are always inappropriate.
Free will and moral responsibility involve the control of individuals by the group as a whole, using the ongoing threat of group sanctions, from disapproval upwards, for antisocial behaviour. This may be called “aggression” in the scientific sense, it’s clearly related to the way other species of primate control each other’s behaviour and sometimes gang up on each other. However, the word “aggression” is in more ordinary language loaded with value, it’s generally applied to attacks or threats which are
not agreed on by the community, and if I read you correctly, that was how you were using it in your post which I quote above. You said that if I were to berate someone who indirectly caused someone else to barge into me, without realising that they were unsteady on their feet, then I would be behaving aggressively, I would be deserving of censure. You were in fact appealing to the concept of moral responsibility, and saying, correctly, that reasonable people would disapprove of my behaviour. It’s this kind of use of loaded language, implying threats from the community, which is pervasive in human social life. While we do not know the details of brain mechanisms, it seems to me that we need this continual sense of holding each other responsible, it’s vital to human societies.
GrahamH wrote:zoon wrote:The problem with Frankfurt cases (implants in the brain of which the subject is unaware) is that they presuppose a level of knowledge about how brains work, and control over the detailed mechanisms, which so far we don't have. My line throughout is that if/when neuroscience reaches that level of understanding, there would probably be a major rethink of free will, and it is likely eventually to become an entirely redundant concept. As you have said, there is likely to be a long period of overlap.
It's not really a problem in thought experiments to suppose technology that doesn't exist. The brain tumour example makes much the same point, we still don't understand exactly how, but there actual cases.
Certainly, I agree with you that, on current evidence, we are ultimately determinate. It may well be that in the future it will become common to control human behaviour in detail by altering the mechanisms of brains. I think that if/when this happens, human social life will eventually change in ways that we cannot at the moment predict; we don’t know the forms the control will take, or how gradually or suddenly changes may happen. I think it’s likely that the concept of free will would become irrelevant. So far, however, curing criminal behaviour with brain surgery is very rare, and Frankfurt cases are still firmly in the realm of science fiction. For the time being, the only effective methods of detailed control of other people are less direct, and involve creating consequences which the person may like or dislike. If this method is to work, the behaviour being sanctioned needs to have been chosen by that person, i.e. to have been “free willed”.
by 
