Friday, September 10, 2010

Responsibility and causality

Previously, I've talked about causality from a scientific perspective, but causality is also very important in ethics.  Specifically, causation tells us something about who we hold morally responsible for various outcomes.

For example, let's say that I borrow a friend's laptop and it breaks.  Should I be held morally responsible for breaking the laptop?  It depends on whether my handling of the laptop caused it to break.

Case 1: My handling of the laptop caused it to break.  That is to say, it would not have broken if I had not mishandled it.  Therefore, since prefer our laptops in working condition, my mishandling should be discouraged.  That is to say, I should be held morally responsible for the mishandling.

Case 2: My handling of the laptop did not cause it to break.  That is to say, it would have broken anyway.  Discouraging my treatment of the laptop is pointless because the results could not have been prevented.  So I shouldn't be held morally responsible.

"I didn't do it, honest!"

In this context, causality just means that the results wouldn't have happened if not for the actions.  Moral responsibility means that the actions should be encouraged or discouraged.  We intuitively understand that causality is a necessary condition for moral responsibility, because there is little point in encouraging or discouraging actions that have no bearing on the results.

On the other hand, causality is not a sufficient condition to establish moral responsibility.  Most obviously, the laws of nature are what cause earthquakes, hurricanes, and other natural disasters, but we don't hold them responsible for anything.  This is because the laws of nature are not human and don't respond to encouragement or discouragement.

But even when all parties are human, this is still true.  Consider a general scenario where Alice performs action X.  Bob, in response, performs action Y.  Y results in negative consequences.  Both actions caused the result, in the sense that if either action had been avoided then the consequences would have been avoided.  But should Alice or Bob be held morally responsible?


This is an outcome table for our scenario.  It's assumed that Alice and Bob have some motivation for acting the way they do, and are inconvenienced if they don't do it.  Depending on the precise situation, this inconvenience could be great or small.

If both parties are being coerced with threats of violence, then the "inconveniences" are quite large, and we couldn't hold them morally responsible.  If both parties are just being lazy, we might hold them both morally responsible.  For example, if Alice is a lab safety supervisor, and Bob is a lab worker, then we expect Alice to inspect the lab conditions and for Bob to be prepared for any hazards.  Safety is more important than laziness.

But sometimes, one party is much more inconvenienced than the other.  Unsurprisingly, the first example that comes to mind is queer-related.  If Alice goes out in public with her girlfriend, or if she is MTF and goes out in public presenting as female, then this might cause Bob to harass her.  But it's a pretty major inconvenience for Alice to never show herself in public, and not a big inconvenience for Bob to stop harassing people.  So Bob should be given all the moral responsibility.  Alice may recognize the chain of causality, and may choose to be more discrete, but she is not morally obligated to do so.

On the flip side, it could be Alice who holds all the responsibility.  For example, Alice could be someone who advertises healing touch therapy with false claims, and Bob could be a customer.  Certainly Bob bears some responsibility for investigating the validity of the services he pays for, but Alice is far more responsible because she is lying.

Come to think of it, in the above scenario, Alice has the greater inconvenience, because she's missing out on a lot of money that she could have conned from people.  Perhaps it's better to measure it by inconvenience to society in general rather than inconvenience to Alice or Bob specifically.  Or maybe there are other factors I'm not considering.  Point to ponder.

In general, this is a very difficult ethical dilemma, finding out who is morally responsible.  For many situations, there are social conventions for who is expected to step aside.  Sometimes, these conventions are neither good nor bad, it's just that society has to pick one or the other (like the convention that all cars drive on the right side of the road).  Other times the conventions are bad, but they're hard to change without some intermediate negative consequences.  Ethics is hard.

(This post came to mind when I read something on The Thinker.  I have no comment on the specific example he uses.)

Other posts in this mini-series:
Colds and Causality
Women and Causality
Responsibility and Causality
Nature/nurture and Causality
Physics and Causality 
Math and Causality 

6 comments:

Larry Hamelin said...

Note that The Thinker's account of the subprime mortgage crisis is incorrect -- or at least highly controversial: the mortgage crisis was probably not caused directly or indirectly by offering "easy credit to low-income earners."

Larry Hamelin said...

As you note, the best way to view moral causality is, first of all, to look at it as what specific ideas and properties of minds we want to actually correct.

Another important point is that, because we can't predict the specific outcome of our actions in very much detail, we have to look at the general case. Would an action in general or usually lead to undesirable consequences? There's no point in correcting beliefs that generally lead to good outcomes. If, for example, you change the oil in your car at the recommended time, but this action happens in one case to lead to a bad result (the new oil filter is perhaps undetectably defective) that doesn't argue you shouldn't change your oil.

Third, we often have to combine the general case with abstract properties of beliefs and actions. Does the class or category of actions generally cause undesirable results?

Basically, an effective strategy when we have a moral disagreement is find the abstract level at which we have substantial agreement about the general desirability.

Secret Squïrrel said...

Good stuff (BB's comments, too). However, we also need to consider intent as well as consequence.

If we were walking along the footpath/sidewalk and I pushed you over, you'd have every right to be upset. But what if someone had dropped an anvil out of a fourth story window (yes, we are in a cartoon) and my pushing you over prevented you from getting a mountainous bruise on your head? You'd probably thank me. Unless I happened to push you in front of a moving vehicle in which case you'd have to perform some "lesser of two evils" calculation.

What if I had meant to push you over and only incidentally saved you from the anvil?

Larry Hamelin said...

However, we also need to consider intent as well as consequence.

Intent is precisely that which we specifically need to correct (if indeed it needs correcting).

miller said...

If an action incidentally prevents me from being anvilled in the head, we can't really encourage that specific action. As Barefoot Bum said, we can only encourage or discourage general classes of actions. In general, pushing people randomly is bad, even if we can think of specific scenarios where it will accidentally give a good result.

Larry Hamelin said...

On the other hand, if I really do know that an action that would generally be bad will almost certainly have an unequivocally good result, that's a good argument for permitting that specific instance. If I actually see the anvil coming for your head, I'm justified in pushing you out of the way.

That's one reason why we actually try each individual case in a court of law, to let human beings sort out the individual complexities of each case.