Causation

The Language Guy has made another extremely interesting post, this time about how people speak about the causes of things.

As a first year law student (last year) I spent months discussing the meaning of "causation" in Torts class. The legal profession generally dissects that word into two parts:

1. Cause-in-fact
2. Proximate cause

"Cause-in-fact" is generally easy to establish, such as: if X did not run the red light, he would not have collided with Y. But it's too easy to establish: if X would have been driving 25 mph instead of 28 mph, he would not have been in the intersection at that time and would not have collided with Y.

Cause-in-fact goes too far. If X's parents had not had sex, then X would not have been born and could not have murdered Y. If you only used cause-in-fact, then X's parents (and grandparents, etc.) would be liable for Y's death.

So, that's the reason we move on to proximate cause, which is also sometimes referred to as legal cause. It's basically a rather squishy concept that's used to decide whether or not someone should be held liable for causing a problem.

A great example of how this is used today (the "scope of the risk" test) is this:
X leaves a gallon can of nitro glycerine on the edge of the kitchen table. This is negligent because it could get knocked off and explode. Y, a child, knocks the can off the table, but miraculously it does not explode. Instead, it breaks Y's foot.
Should X be liable? Probably, no. The reason it was negligent for X to put the can there was the risk of explosion, not the risk of breaking someone's foot. It was not within the scope of the risk. For purposes of this analysis, you should imagine that the can was full of water. If it would have been negligent to put a can of water there, X should be liable. If that is not negligent, then X should not be liable.

Under this analysis, there can be more than one proximate cause of an injury. Assume X and Z both crashed their cars into Y simultaneously. Under the simplest analysis, Y could recover all of his money damages from either X or Z.

An entirely different can of worms is opened when you go into the products liability area of torts. Here, you still generally speak of the proximate cause of the injury, but the analysis gets more and more squishy. This is especially so in the case of fungible materials. For example, assume you can prove that lead paint was the cause of the injury. Assume also that you have no idea who the manufacturer of that paint was, but you know that it was purchased between 1920 and 1925. Courts will then look at the market share of all the manufacturers of that time. A, B, and C each sold 20% of the lead paint during that time, D sold 35%, and E sold only 5%. If that's as far as your analysis goes, then D should pay 35% of the damages, E 5%, and the other three should each pay 20%. It's different, however, when you know the paint was sold in Nebraska, for example, and E had 99% of the market share in Nebraska.
As may be obvious, this is not scientific. But if you assume that everyone injured by lead paint sues all the manufacturers all the time, and every court uses the same analysis and the same facts, then the result should be correct. This probably won't happen.

Well, at least that's a brief overview of what the law says about causation. It's not scientific, but usually it feels right to people that hear the result.