Freedom, responsibility, and higher-level laws

In this post, I present an argument for the following thesis: When determinism holds, a person’s action is free and the person is responsible for the action if determinism does not hold on the relevant higher level on which the action is being described.

Let’s start by unpacking what that means:

  • “When determinism holds.” This means that the world is such that every future state is completely determined by any past state of the world. So, in principle, a godlike intellect staying outside the universe but observing it, who knew everything about the universe’s state at some given point, could also infer its state exactly at any later point.
  • “A person’s action is free.” I can’t fully unpack this here, so I’ll just characterise it. What I mean by “free” is “what we should mean by freedom of the will or of action so that it both conforms to our intuitions as well as it makes sense to, and corresponds to what we want and need.” So in a sense, “real” freedom. Obviously this is a compatibilist position, ie. free will is seen as not contradicting determinism. I’m also assuming that freedom means something like being able to make considered choices that take everything into account.
  • “And the person is responsible for the action.” This mainly means that the person may be punished and blamed for what the did. It will become evident why this is justified when I present the argument proper, but the idea here is that punishment is justified only by its deterrent effect; because it will keep people from doing bad things, not because you’re just supposed to punish people who do wrong just because.
  • “Determinism does not hold on the relevant higher level on which the action is being described.” This will need to be unpacked in several parts.
    • “the higher level on which the action is being described”. This means that you leave out some detail from the lower level in your description. So while the godlike being who can predict what happens in the universe would know about every elementary particle or whatever the universe is made of, we could also describe the universe in a lot more coarse way that focuses on some aspect of it and leaves details out. In the case of an action, the description might be something like “When Alice had taken one drink, she went on drinking all night.” or “When Bob saw the television in the shop window, he broke the window and stole it.” (The “when” structure will be necessary later.)
    • “determinism does not hold on the higher level”. This will be the crucial point. It means that our godlike being could not predict what will happen if it is only given the higher-level description. So it cannot predict that if Alice has one drink, she will go on drinking all night, or that if Bob sees a television on the window, he will steal it. But really this is putting things a bit strongly. Who steals every television? Is there no recovering alcoholic (of the sort who knows that they can’t have even one drink) who would sometimes be able to stop at one drink if they tried it? So let’s just say that there doesn’t need to be an absolute law, only something like it, and that we will need to be really careful about how we formulate the “laws” in the higher-level description.
    • “relevant level”. …Which is why this is very important and carries a lot of burden for the things that are not explained properly here. We need to choose the right level in every case to make this work.

So, the claim can be restated as follows: When determinism holds, a person’s action is free and the person is responsible for it when, when we describe the kind of situation the person is in the right way, determinism doesn’t hold for even roughly all situations that fit that general description.

Okay… so what does that mean? The more evident example is that about Alice as an alcoholic. Let’s say she “always” drinks too much in certain general conditions. So she’s an addict whose freedom is compromised.

What about Bob as a thief? Here’s what someone opposed to determinism/compatibilism might say: If determinism is true, then when Bob steals a television, by the definition of determinism, he couldn’t have done otherwise in those exact circumstances. So he can’t be held responsible.

But, of course, if the purpose of punishment is deterrence, then it makes sense to punish people anyway, as we can expect that (perfectly in harmony with determinism) fewer people will steal televisions or whatever if they know you’re liable to get caught and punished.

What the current argument aims to do is to show the difference between the kind of determinism that threatens freedom and the kind that doesn’t — and why that is relevant for responsibility. If someone is determined to do something only in the exact circumstances in which they did that thing, that means that they and others could do differently in circumstances that can be described in the same higher-level terms. Then we can hope to deter people by punishments, and also educate them beforehand in other ways. If, on the other hand, there is a higher-level generalisation to be made about a person, or people in general, that they will “always” (or often or whatever) act in a certain way in the certain kind of circumstance, not just other things being equal but even if we add deterrent and/or education — then there is no sense in treating people as responsible, because that will change nothing. The choice in such a situation are also not free in the other sense that the person is not able to consider things they should be taking into account. They will do the thing even if they will regret it. This is why we are right to make allowances for pathological addicts, mentally disturbed people, etc.

So, the thesis at the beginning follows given compatibilism with a certain kind of view of freedom, and given a utilitarian view of punishment and responsibility — applies to freedom because it describes circumstances under which one can make considered choices, and applies to responsibility because it describes circumstances under which the concept of responsibility makes utilitarian sense.

Now of course we could question compatibilism, or utilitarianism about responsibility. I have or will have arguments for those elsewhere. We could also question determinism, and I’m only assuming its truth here to simplify things. All I’m trying to do in this post is show that something follows given those assumptions.

Obviously this will need more work, but I hope this post gets the general idea across.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s