To begin at the end of your argument, as John Stuart Mill long ago pointed out, the laws of the universe are laws of sufficiency in terms of minimally sufficient sets of antecedent conditions that make up the laws of nature, instantiation of all of which is necessary for the occurrence of the consequence, but there may well be more than one set of such sufficient conditions on any particular occasion, or to put it another way, any one of those sets may be more than completely instantiated by duplicative conditions, as in your dog-stabbing example.
Contrary to your further statement that lawyers don't have anything interesting to say to philosophers, Hart and Honore's recovery and elaboration of Mill's insight was of great interest to philosophers, including Mackie, who cribbed it without attribution in his INUS formulation before (as is not noticed by philosophers) wierdly retreating to the but-for test for singular instances of causation. Mill's insight continues to be of great interest to philosophers. I have been invited to Germany to speak to several different groups of philosophers about the idea.
To return to the beginning of your argument, you begin by apparently stating that once you have wronged someone, you supposedly are responsible for whatever happens to them, regardless of whether or not your wrong contributed to the later occurrences. If this is your position, you have departed wholesale from any attempt to describe realistically our conceptions of morality or law.
On the other hand, you proceed to argue that there is causation (or something else, unnamed and unelaborated?), using Mackie's (and others') attempts to salvage the but-for test by applying it in the aggregate. I recall your having rejected Mackie's aggregate approach. Perhaps I misread (I'd have to go back and check). In any event, if you then or now accept Mackie's argument, you are in really big trouble, since, as many have pointed out, the aggregate but-for test is wildly inaccurate and over-inclusive, in addition to being illogical (A did not contribute to the death, nor did B, but in the aggregate they did.) Putting aside logic and turning to application, I can add an infinite number of irrelevant conditions into the aggregate along with one or more but-for conditions, and the aggregate will still be a but-for cause. If, as Jane Stapleton attempts with her 'targeted but-for' test, which unfortunately is incorporated in the blackletter of the Restatement Third (a huge error which is mitigated by reliance on the NESS causal sets idea in the comments), you limit the aggregate to those conditions that prevent other conditions from being but-for causes, you stilll have the problem of not being able to distinguish duplicative causation from preemptive causation. Eg, assume the dog is dead from the first stab before the second stab occurs, or the decedent drank a fatal dose of poison for which there is no antidote an instant before his head was chopped off. You seem to argue below ('stuffed dog has value') that the non-causer should still be liable for wrong of causing the death and all consequential damages. The law is clearly to the contrary. There would be no tort liability. There might be criminal liability, for attempted murder (attempted dogicide?), but that is a different wrong, in terms of specification and nature (public versus private).
I agree that in your mortar and brick example, which involves the most difficult causation scenario, overdetermined negative causation, both omissions to supply were causes. The NESS analysis explains this. As you admit, the but-for analysis does not, unless you apply the bankrupt aggregate-but-for test.
We may indeed have legally recognized wrongs that do not involve any 'consequential loss' in the sense of physical or economic or even emotional harm. Assault, battery and false imprisonment are examples, as is a breach of contract. In those instances, nevertheless, there is not 'wronging in the air'. The conduct must cause some defined legally cognizable injury, to honor, dignity, autonomy, reputation, etc. Other torts, most notably negligence and the strict liability torts, do require some consequential loss. Assume the independently sufficient stab wounds were the result of negligence rather than intent. Did each defendant commit a wrong regardless of any consequential loss? If you think so, I think you have departed from the actual law. If you say no, you seem unable to hold either liable for the resulting death.
From: Robert Stevens [mailto:robert.stevens@ucl.ac.uk]
Sent: Thu 3/10/2011 1:07 PM
To: Wright, Richard
Cc: obligations@uwo.ca
Subject: RE: UK Supreme Court Decision on Causation
If a loss would not have been suffered but for the commission of one or more wrongs, each wrongdoer is jointly liable for the loss, even if in respect of each wrong individually the loss would have been suffered anyway because of the wrong of someone else. This is a rule of law. It is nothing to do with what causation means.
So, if a builder contracts with A that he will supply him with bricks and with B that he will supply him with mortar, if both in breach of contract
fail to supply the builder, he can recover against both for the loss caused by not being able to build even though he would not have been able
to build without either the bricks or the mortar alone. That he cannot say that either failure to supply individually caused him this loss, because
'but for' cannot be satisfied in relation to either A or B's individual breach is true, but irrelevant.
So, in the two dog stabber case the result is clear, even if you think (incorrectly in my view) that the only damages recoverable in the law of
torts are for consequential loss. No departure from 'but for' is necessary to explain the result. (I confess that in my original example I assumed
that a dog which has been stabbed and doomed to die is now valueless. You could have it stuffed I suppose.)
We are agreed, I think that both dog-stabbers have committed a tort. The tort is fully constituted by stabbing the dog, just as the breach of
contract was fully constituted by not supplying the material. That we cannot say as against any one dog-stabber "but for your stabbing, the dog
would not have died" is true, but completely irrelevant as a matter of law. It just isn't the relevant legal question.
Further that we cannot say against either of them individually that their stabbing resulted in the dog now being dead does not lead to the logical
absurdity of the dog's death being uncaused. The dog would not have died but for both dog-stabbers, just as the builder's loss would not have been
suffered but for the failure to supply bricks and mortar. I don't reject Mackie's aggregative move. I think he is plainly right. I just don't think
it is particularly helpful in understanding the law. I also don't think lawyers have anything interesting to say to philosophers of causation.
Put another way, those who have focused solely on the defendant's blameworthy conduct, and the current state of affairs which the plaintiff
now complains about, are ignoring the crucial legal question of whether the defendant has wronged the plaintiff at some point in the past.
X caused Y entails "X was necessary in the circumstances for Y". My view is that if you think your understanding of the law requires that that rule
of the universe has to be departed from, your understanding of the law is wrong. Not the universe.
--
Robert Stevens
Professor of Commercial Law
University College London