Someone recently asked me what my thoughts are on free will. Specifically, they wanted to know if free will is something that could set us apart from robots. If we imagine that there is a very complicated robot capable of sensing and responding to inputs in much the same way that a human does, on what grounds can we say that we are different from that robot? On an intuitive level, one might be inclined to argue that humans are qualitatively different from such a robot because humans have free will, whereas the robot does not.
Before I discuss this scenario I want to stop and ask why we even care about free will. Why do so many of us think that free will is important and that if we found out that we didn’t have free will it would be a tragedy? I think there are several categories of reasons. First we have experiential reasons. We experience the world as if we had free will. We believe that we have options about what actions we are going to choose and we are aware of ourselves making conscious decisions to act one way or another. We have the experience of willing actions that we don’t feel compelled to will. We even will things that go against what our natural inclinations are. While these aren’t necessarily arguments for free will, they do demonstrate why we think free will might be important. If it turns out that we don’t have free will, then reality is very different from how we experience it.
The second category of reasons is moral. Many people think free will is necessary for there to be moral goodness, moral badness and moral responsibility. If people are not free to choose their actions, then how can we appropriately praise or blame them for anything they do? Or how can we call any of their actions morally good? We don’t praise the moral attributes of washing machines, and it seems reasonable to say that one reason for this is because washing machines don’t have free will. Their actions are determined. They either work or they don’t work, but we never hold them morally responsible for their actions either way. And we don’t say that they are morally good when they do work and morally bad when they don’t. Intuitively, it would seem that if humans do not have free will, then humans could not be held morally responsible either, and human actions could never be judged to be morally good or morally bad.
The third category of reasons is theological. These are probably the first reasons that come to my mind when I think about free will. I think a lot of this can be attributed to the explicitly Arminian Christian culture that I grew up in. Free will was an essential part of my understanding of Christianity in my adolescence. It provided an alternative to the unpalatable doctrines of Calvinism. I especially wanted to avoid the doctrine of limited atonement (the doctrine that Christ died only for the elect). Believing in free will also gave me a well established theodicy to draw on. Why is there suffering in the world? Because humans have free will and they make bad choices. Why did God give us free will? Because God didn’t want robots; God wanted people who could make morally substantive choices and who could choose to love God back. It all fits together nicely.
Given these intuitive reasons for believing in and caring about free will, one might wonder why anyone would doubt that humans have free will. Nevertheless, many people do doubt it. There are two major theses in particular that, if accepted, combine to make a powerful case for either the non-existence of free will or the incoherence of the idea of free will. The first thesis is causal determinism and the second is incompatibilism.
The Internet Encyclopedia of Philosophy article on Free Will defines causal determinism as “the thesis that the course of the future is entirely determined by the conjunction of the past and the laws of nature.” On this view, given a set of starting conditions, all future events can be comprehensively explained as the necessary consequences of natural laws being played out. Think of the game Angry Birds. In the world of Angry Birds there are “laws of nature,” the rules of the game that dictate how everything in the world of Angry Birds interacts. Someone sat down and wrote out equations that determine how much force is necessary to kill pigs of different sizes, or to break through different types of barriers, etc. These rules mimic laws that we find in nature. For each level of Angry Birds you are also given a set of initial conditions (i.e. the arrangement of pigs and other objects on the level). The only initial conditions that you have control over (ignoring some of the mid-flight capabilities for the moment) are the angle and velocity with which you release your birds from the slingshot. All the events that occur after you release a bird are completely determined by the combination of the initial conditions and the “laws of nature” that govern the game. Nothing else comes into play. Analogously, given a set of initial conditions in the real world, causal determinism claims that the laws of nature will dictate all future events.
It doesn’t follow from causal determinism that free will does not or cannot exist. First one has to accept an intermediate thesis known as incompatibilism. Incompatibilism simply claims that free will is not compatible with causal determinism. That may seem trivial, but it is a point of a lot of debate. Some philosophers think that even if causal determinism is true that free will would still be possible. These people are called compatibilists. I’m not going to present arguments for compatibilism versus incompatibilism. I’m only going to point out that there is not a general consensus on the matter within the philosophical community. For the purposes of this blog post I’m going to assume that most of my (three) readers are incompatibilists (i.e. they believe that determinism is incompatible with free will) and move on to a discussion of the laws of nature as it relates to determinism.
For a long time it looked to me like causal determinism might be an unavoidable conclusion (pardon my bad sense of humor). For someone who takes mathematical, physical, chemical and biological laws seriously, it’s hard to see where there is any space for free will. For example, where does free will show up in the chemical processes that govern my brain? I know that there is a continuous chain of chemical events that takes place in my brain every time that I perform an action. I also know that if someone were to analyze these chains of events they would be able to explain them completely in terms of the laws of nature acting on a set of initial conditions. They wouldn’t need to introduce the idea of a non-physical will in order to explain what was going on. It would be entirely superfluous and undesirable.
This perplexed me for many years until I discovered a new way of thinking about the laws of nature. There are two major philosophical camps when it comes to conceptions of the laws of nature. One camp claims that the laws of nature are physically necessary in some way, and that they are prescriptive or “pushy,” as the Stanford Encyclopedia of Philosophy article on Causal Determinism puts it. In other words, the laws of nature don’t simply describe what we observe in the physical world, they actually make the physical world behave in the way that it does; they cause events. This is the camp of the necessitarians. Those in the alternative camp are called regularists.
According to the regularity position, the laws of nature are not pushy explanations; they are merely correct descriptions of generalized regularities that we observe in nature. The laws of nature do not govern the universe, dictating the path that every particle must follow and imposing order on the physical world. They do not cause events, as if abstract, non-physical laws were even capable of that. Rather, events are caused by things in the world and the laws of nature are summaries of the regularities that we have observed in these things.
This latter view is the one that I take and it turns out that under this view, the problem of free will versus determinism basically disappears. If the laws of nature are not pushy, if they are not physically necessary, if they are not the cause of events, then causal determinism loses a lot of its plausibility. Moreover, if the laws of nature are simply descriptions of the regularities in the world, then there is plenty of space for free will. The laws of nature do not dictate our actions. Instead, the laws of nature depend on what we choose.
Now I realized that last statement seems like it ignores certain solid and unchosen characteristics of the physical world. For example, our choices do not affect the law of gravity. It remains the same whether we like it or not. But I think this can be accounted for if we distinguish between the causal powers of animate and inanimate objects. Electrons, for example, have causal powers of attraction and repulsion of other electrically charged particles. Rocks have gravitational powers and properties like hardness. And all physical objects have various properties that cause them to interact with other objects in predictable ways. As a result of these causal powers, we notice patterns of interaction and regularities. It’s my view that if the world were limited to inanimate objects that it would be entirely deterministic (excepting quantum-level phenomena which I won’t discuss here), because a rock cannot chose to act in one way rather than another. Its behavior is determined by its non-volitional causal powers and properties interacting with the objects around it.
Animals and humans, on the other hand, have wills. They are volitional beings, able to choose one action over another. If all beings were volitional, then the laws of nature could end up being very chaotic. For example, if every electron could choose what its electrical charge would be, who knows what would happen to the fabric of the universe. Higher-order beings probably could not exist in such an environment. So the stability provided by lower-order, non-volitional beings create a context in which higher-order, volitional beings can move and interact. And while deterministic laws of nature may arise from non-volitional beings, that does not mean that the universe is governed by deterministic laws. The laws of nature that arise from higher-order beings are anything but deterministic. Just look at the hard physical sciences versus the biological or social sciences.
Let me try to bring this back around to the opening question. Is free will something that could separate us from robots? My first instinct is to say “yes” because I think humans are capable of true volition whereas robots are only capable of mimicking volitional behavior. However, that’s a meaty metaphysical claim and I would probably have to get into discussions of consciousness, philosophy of language, and phenomenology in order to address it. I can’t do that here, but I’ll just hint at some trains of thought that might be worth further investigation. The first thought is that if one believes in some form of mind/body dualism, then one could argue that humans are different from robots because humans have minds. The second thought is that humans are language users and, as such, are capable of reflective thought. That is something unique to humans that seems to be qualitatively different from animal communication. One could argue that this ability also sets us apart from robots. The third thought (which may be related to the first two) is that it might be incoherent to talk about a robot making a choice between different courses of action because robots do not have a thought world. One could argue on phenomenological grounds that it is appropriate to say that humans make choices about courses of actions, whereas all we can appropriately say about robots is that they execute the software written for them.
-David
No comments:
Post a Comment