Thus far, in speaking about the stranger cases I've talked about the relationship between strict liability, negligence and intentional harms. But there's always another element lurking - Did your particular actions cause the harm in question? The modern view on this subject is to invoke a two part test. The first is called but-for causation, the second is proximate cause. Under the first part somebody asked the question is but-for the negligence of a particular defendant or but-for the action of a particular defendant, would the harm to the plaintiff happen? And if the answer to that question is no, then the defendant turns out to be out. Turns out this test is simply too capacious because there could be all sorts of events. So if you start with a but-for system, you then you always develop a system of proximate cause, a limitation as to what particular injuries that you do cause you're not responsible for and you develop either a foresight or a directness test. If, in fact, you do what was said to be in the Roman system, causation corpore corpori, which is fancy Latin for by the body, to the body; causation begins with a punch to the face. Or a kick in the leg. And the connections here are very tight. And everybody therefore understands what's going on and there's no dispute about it. And so what you then try to do is to go backwards from the immediate cause to see if there's something else in the cycle which you want to take into account. But everybody says that to the extent that you imposed coercion upon an intermediate actor, you should be treated as though you're responsible for the harm, as if you did it yourself. And so what you say is actions done under coercion by a third party mean that the actions of the third party are not quote, unquote too remote. Note that, at this particular point in time, the definition of proximate cause it's oddly contradictory. It started with meaning the closest cause and now the question is to figure out how remote causes can be held responsible.