As I've mentioned before, I actually think philosophy has serious implications for the "real world". (The fact that the puzzles are also fun is just a bonus.) And I've come to think that one of the most important puzzles philosophers discuss is that of free will. I don't think there's any free will (of the controversial kind - we'll see what I mean by that in a while). And I think that's okay. And I think that's important.
What is "free will" in the first place?
What is free will? Well, this question itself is up for debate. To some philosophers, your act was free basically if it came from your own desires. Very roughly, if you wanted to do it and then got to do it, then you did it freely; if you did something you didn't want to do, you were not acting unfreely. Gandhi freely went without food in order to free India (he wanted to go without food, all things considered, and then did it). Someone starving to death on a desert island, by contrast, is not freely going without food in this sense.
Well everyone agrees that sometimes people get to do what they want to do. But not everyone thinks that counts as free will. Why not? Well suppose that what you wanted was completely caused by things totally out of your control - maybe someone kidnapped you and then brainwashed you to want it, for example. Or suppose causal determinism is true, and everything is caused by what came before - so anything we ever do was utterly determined by events back 1,000 years ago and more. So for example, already in the year 1016, it was set in stone that you would eat what you did for breakfast. Well, many people feel like even if you ate what you wanted to eat, if that want was not ultimately under your control, then you weren't really free. If freedom just means doing what you want, even when your want was caused by things totally out of your control, then freedom is compatible with your being caused by forces outside your control to do everything you do. That's why we call the philosophers who think you're free when you do what you want, whether or not that want was caused, compatibilists.
Another way to put this: suppose robots get to the point where they genuinely want things, but suppose also that they are following a deterministic program to decide what they want at any given time. Do such robots have free will? A compabilist would say yes. But if you say no, you are an incompatibilist. You think that if your act was "fully" caused by events in motion before you were born, then your act was not truly free.
The incompatibilist thinks that to be truly free, your future at least sometimes needs to be "open" - there needs to be a moment where given the exact same circumstances, you could have either done the act or not done the act. If you believe in that kind of free will, we call you a libertarian. (This is just a label for a philosophical view on free will that only happens to share the name of the political view. But free will has important implications for the politics, as we'll discuss some in this post, and as I plan to discuss in a future post dedicated to the political view.)
Why we don't have the controversial kind of free will
If compatibilists are right, of course there is free will, because of course sometimes some people get to do what they want. But if incompatibilists are right about what it is to have free will, then it's still not clear whether we have free will or not. That's why I call it the "controversial" kind of free will.
I am pretty torn on lots of philosophical positions, having heard opposing arguments from lots of smart people. But I have not yet heard an argument to shake my confidence in this view: we do not have free will in this sense. Libertarianism is wrong. (I'm inclined to think compatibilism is wrong too, which makes me a "hard determinist" - there's no free will of any sort. But I don't think the stakes are nearly as high between those two views, as I'll explain in part 2.)
Why is libertarianism wrong? Well, it says the act must be at some point uncaused by past events. But here's the kicker: part of being uncaused means it was uncaused by anything about you. If some act was uncaused, you didn't do it; it just sort of happened. It might as well have been from a nerve spasm.
Imagine for example it's the end of a date, and you're deciding whether to go for the kiss. Freeze time at the crucial moment. According to determinism, the combination in that moment of your fear / excitement / bravery / attraction / haunting past memories or rejection / hormonal responses / dopamine levels and so on - all the possible relevant factors - will either cause you to lean in for the kiss, or cause you not to. The future is already determined. According to the libertarian, on the other hand, to be free, none of these things can cause you one way or the other; it must still be open in that moment whether you go for the kiss or not. According to them, if you rewind time to that exact moment - with all that same mix of emotions and chemicals - something different may happen. So nothing about you influenced the result.
I'm not into swamping a post with images, but Kiss the Girl is a sentimental favorite for me.
- You might be tempted to say that whether you lean in for the kiss or not is caused by the decision, and the decision is free. But just run the same argument above with the decision: freeze to just before the decision is made, with all the same mix of emotions and chemicals in place. Was the resulting decision caused or not?
- You might say that the combination of factors makes for a high chance that you go for the kiss. But if having a higher chance from those factors makes you go for it freely, then having the highest chance - that is, fully caused - makes you the most free, right? You are really a compatibilist. The random part at any rate is not part of your freedom, right?
In other words, this kind of free will faces a "dilemma argument" - whichever way you go on causal determinism, you can't recover free will.
- If an act is caused by what came before, it's not free.
- If an act is not caused by what came before, it's not free.
- Every act is either caused by what came before, or not.
- Therefore, no act is free.
There are responses to this argument, of course; the most popular of them posit what we call agent causation in order to fix it. Basically, this is a special kind of causation in the world - a kind of causation that science has never seen before, and that can make neurons fire in your brain, and that just happens to (somehow) give you free will. In other words, they posit magic to fix the problem. And from what I can tell that's the most plausible way to get libertarian free will.
That is the summary of why I think we don't have the "controversial" (incompatibilist) kind of free will. In my next post I'll explain why that's not such bad news, and also explain why that's really important for things like our judicial system and just redistribution of wealth.
You say "agent causation [is]... a kind of causation that science has never seen before... magic". Am I correct in assuming you are a materialist and that you believe consciousness to be an emergent property of matter?
Good question! In fact I am a materialist - but this problem isn't really about that. As Peter van Inwagen puts it in this article:
That article and this NYT article by Galen Strawson gives you a sense why. So to be a libertarian you don't just need immaterialism, you need to posit a specific, non-natural type of causation explicitly to make sense of free will. But this looks suspiciously ad hoc - no? It might be the right move if (as van Inwagen thinks) we have to have free will and know we have free will. But as I'll argue in part 2, we can get by just fine without libertarian free will.
Thank you for that article link. I would take issue with some of the premises but I would agree with Strawson's reasoning from his premises.
To your comment: So an Idealist (such as I am at present) would need to posit a specific, immaterial (is that what you mean by "non-natural?) type of causation explicitly to make sense of free will?
It seems that way, yeah. Presumably angels are immaterial (on the standard reading), but as van Inwagen argues, even they face this problem. Very briefly: do such angels act based on how they now are, or not? If the act is from how they are now, we ask how they became that way (whether the answer is from standard materialist physics or from whatever account you give of the lawful relations of ideas), and whether they are ultimately responsible for that. If the act is not a direct result of how they are now, it does not seem to be really their act.
As Strawson says, you need the miraculous causa sui to make sense of libertarian free will - you make yourself as you are, from scratch. But this (as he quotes Nietzsche saying) is a "rape and perversion of logic" - how can anything self-create?! Wouldn't you already have to be some way in order to make yourself another way?! This isn't just non-scientific, it's difficult to comprehend how even magic could do it.
Yes, I think I can see the point being made by Peter and Galen. They suppose that beings must act based on how they are now... and I guess I just do not see why it could not be that "how they are now" presents a set of live options from which they may freely choose. I could see how no being would possess limitless options at any single point in time but not why they must have none at all.
Also, it is true that every person would acknowledge that he at least has the illusion of free will (which is strange enough if we do not). If it were true that we do not freely make choices nothing we think or do would be of any significance whatsoever. Even deciding to believe that determinism is true would be meaningless! Therefore, since free will is a necessary condition of meaningfulness in life, it seems best to assume we have it. If we do not then life is meaningless and who cares?
Makes you think