Sunday, March 11, 2018

Thoughts on Causation and Responsibility

The next section on my Metaphysics and Epistemology course is "causation".

As I have reported earlier, I have studied value theory - without much emphasis on metaphysics and epistemology. However, on the subject of causation, there is an area of overlap.

That area concerns moral responsibility. To say that Agent A is responsible for an accident is to say that, in a sense, he caused it. So, it seems, having a theory of responsibility requires having a theory of causation.

Before getting into this study of causation, there are two preliminary points worthy of consideration.

(1) Free Will

In discussing moral responsibility, one of the first topics to come up is that of free will. We cannot hold a person responsible for an action unless, in some sense, he "could have done otherwise". This is often taken to assume that the agent has some strange contra-causal power that allows him to break the laws of physics and cause matter to go one way or another - to perform one action instead of another - regardless of what would happen under the determined laws of physics.

Desirism is unique, I think, among moral theories in that it not only denies the existence of free will, it denies that the free will hypothesis has anything to do with morality. Morality was invented in a determined universe and it was invented so that it works in a determined universe. Those thinkers who postulated some sort of free will were wrong from the start.

Typically, philosophers distinguish positions regarding free will having to do with different positions on two different questions:

Question 1: Do humans have free will?

Question 2: Is free will required for moral responsibility?

Libertarianism (not to be confused with the political view, with which it has no necessary connection): The position known as "libertarianism" says that the answers are "yes" (humans have free will) and "yes" (free will is required for moral responsibility). This is considered the standard or default view of morality as it is practiced. We hold people morally responsible for actions when they could have done otherwise and, because humans have free will, they could have done otherwise - unless compelled to act by some outside force.

Hard Determinism: The position known as "Hard Determinism" says that the answers are "no" (humans do not have free will) and "yes" (free will is required for moral responsibility), which means that there is no such thing as moral responsibility. This whole practice of holding people morally responsible for their actions is a mistake. We are all only doing that which prior historical events cause us to do and to say that we are to be condemned because we could have done otherwise is a mistake.

Soft Determinism/Compatibilism: The position known as "soft determinism", also often known as "compatibilism," effectively argues that the answers are "no" (humans do not have free will) and "no" (free will is not required for moral responsibility). What is required for moral responsibility is that one's actions are caused by one's character or the type of person one is - by one's own beliefs and desires. When the action is caused by the type of person you are - with who you are as a person - then you are responsible for that action even though you are the type of person you are as a result of the determined laws of nature.

Desirism is a particular kind of soft determinism. One of the problems with compatibilism is that it says that A is compatible with B - that the two can exist side by side in moral harmony. It allows that a type of "free will" is compatible with determinism because it defines "free will" in terms of acting on one's own beliefs and desires - with auctions being caused by the type of person one is. Desirism is a particularly harsh form of compatibilism because it dismisses free will entirely. It does not want to live in peace and harmony with "free will". It wants to dump "free will" entirely - kick it out of the house and be rid of it. It was a bad idea at the start and it remains a bad idea.

Instead, desirism holds that the actions that one is responsible for are those that come from one's character, and that character itself is under the influence of praise and condemnation (reward and punishment). We hold people responsible for their actions because, in a determined world, holding people responsible (reward and punishment; praise and condemnation) has effects on people's character, and we have reason to value those effects.

(2) Practical Responsibility

So, fine, a person is morally responsible for an act when it comes from a part of his or her character that is susceptible to change through praise/reward and condemnation/punishment.

This still does not give us a theory of causation.

On this matter, I have tended to adopt a pragmatic view of cause. To say "A causes B" is to say, "Hey, as a matter of practicality, if we want to control B, then the best way to do this is to find some way to control A."

So, a house burns down. The fire marshal is interested in what caused the fire. It wasn't caused by the presence of oxygen in the atmosphere - we certainly are not going to start building oxygen-free houses. It was caused by faulty wiring, since we can change wiring codes and install safety measures (e.g., circuit breakers) that reduce the chance of fires due to wiring.

And, so, a person is responsible for an action when the action is caused by a part of his character that we can influence using reward/praise and condemnation/punishment. We hold people morally responsible for actions - and apply reward/praise and condemnation/punishment to the action, precisely because it was caused by a type of character trait vulnerable to reward/raise and condemnation/punishment.

Cause by Absence

I think one of the merits of this type of account of causation is that it makes sense of the possibility of the absence of something being a cause.

What caused the fatal accident?

Answer: A callous disregard for the interests of other people on the part of the machine operator.

Most competing theories of causation require that, to make sense of A causing B, there must be an A that causes B. They have no place for the possibility that the absence of A can cause B.

However, on this practical cause model, we can make sense. If the absence of A is something that we can influence - something that we can change - in order to prevent further occurrences of B, then "the absence of A" is a perfectly reasonable cause of B. It simply means that we have the power, in the future, to make it more likely that there will be an A. In the example above, we need to promote a regard for the well-being of others, which we can do (in part) by registering condemnation or punishing those who display a callous disregard for the interests of others.


Doug S. said...

Are you at all familiar with the book "Causality" by Judea Pearl? It's a formal mathematical treatment of the topic.

Alonzo Fyfe said...

No, I am not. I did not study causation formally, until now.