Date: Wed, 3 May 2006 14:23:50 +0100

From: Robert Stevens

Subject: Childs v. Desormeaux

I don't think this scenario is a non-feasance case at all. The defendant has failed to prevent the driver from drunkenly injuring another, but that doesn't make it a case of non-feasance.

If the defendant had not acted by having a party, the car crash would not have occurred. If the party had been carefully organised, so that drivers were not supplied with excess alcohol, the accident would not have occurred.

A next door neighbour who knows the wild party is going on and hears the drunken revellers getting into their cars but does nothing to prevent them driving off, has not by his actions made any subsequent victim worse off. By contrast, the person who throws the party cannot say he has done nothing, and cannot claim that he has merely failed to protect the victim from a harm which would otherwise have occurred. That is like a car driver saying that his failure to apply the brakes was merely non-feasance.

If I load a gun and leave it in my lap and a child picks it up and shoots you, I have by my actions made you worse off. Your complaint is not (simply) that I failed to stop the child from shooting but either (1) that I was careless in loading the gun and leaving it where the child could get it or (2) by enabling by my actions the child to obtain the gun I came under a positive duty to act to protect others.

There are easy precautions to take to prevent guests leaving your parties drunk. If you don't want to take them, don't throw parties.

Robert Stevens
Barrister
University of Oxford

"Goldberg, John" writes:

Hi Steve:

Thanks, your comments are helpful, as always.

Although there may be no breach in Scenario 2, I don't think the example hinges on that conclusion, or needs to hinge on it, because there is a prior question of duty and a subsequent question of proximate/superseding cause. Thus, if we imagine, as you do, various precautions H might have taken but unreasonably failed to take with respect to the risk of his adult guest's drunk driving, I think that there is still a strong argument that H should not be held liable precisely because H was not under a duty to take reasonable affirmative steps to protect the likes of P against drunk driving by the likes of G. Contrary to your assertion, nonfeasance has everything to do with this analysis -- it hinges on the fact that P is complaining of a failure on H's part to control the conduct of another competent and responsible adult, as opposed to H's having played the active role imagined in scenario 1. Some no duty cases really are no-duty cases, rather than no-breach cases.

Of course, one can question (as I did in my e-mail, briefly) whether there is some reason to think that a scenario 2 social host case warrants an exception from the general rule of no duty that governs nonfeasance. (More radically, one could question the rule itself, though I think its rejection would prove unpalatable for a number of reasons.) But notice that nonfeasance has already done a lot of work by placing the burden on the plaintiff to argue for the applicability of an exception. And the analysis of that argument can (and I think should) look quite different from the policy inquiry that you suggest. For one thing it is an inquiry that, at least in the first instance, should be about law, not policy per se. That is, lawyers and judges ought to look to recognized doctrinal exceptions to the no-duty rule, such as Tarasoff, and attempt to analogize or build on those. Second, the inquiry is an inquiry into obligation, not an inquiry into social welfare, and, at a minimum, your framing of the issue as one of policy runs the risk of conflating these two. The question is not, whether, all things considered, it would be better from society's perspective to hold social hosts liable to drivers injured by drunk guests (here I essentially agree with Robert Stevens' doubts about the propriety or wisdom of asking judges to answer these sorts of questions, although I am less moved by his apparent view that morality demands that social hosts engaged in ordinary, non-criminal activity should routinely be held responsible for the wrongful acts of others who participate in such activity). It is rather the question of whether, given tort law's hesitancy to recognize affirmative duties, there is something compelling about this sort of scenario that warrants or counsels against creating a binding norm -- violations of which will generate potentially significant liability to others -- specifying that social hosts must from now on take steps to keep track of and control their guests' behavior. (Yes, yes, one can argue that this latter question 'really' just 'boils down' to the all-things-considered social welfare question, but the possibility of being a reductionist is hardly an argument for the propriety of being a reductionist.)

Finally, I would suggest that, even if one is looking at tort law as a social engineering mechanism, before one throws it out as a too-cumbersome and too-expensive mechanism for deterrence and compensation, one would have to take account of the intangible (hard-to-measure) deterrence value provided by a body of law that tracks and reinforces norms of responsible conduct. That is, one can't just assess the efficacy of tort law on the assumption that society is a collection of Holmesian bad men, one also has to take into account the effect on norm-followers' behavior of a system of norms that is shored up, in part, by tort law as one among many social practices concerned with holding persons responsible to moderate their conduct in light of the interests of others.

<<<< Previous Message  ~  Index  ~  Next Message >>>>>