Consider libel. If I call you an axe murderer in print, I breach my duty
not to do so and commit libel. There is no further requirement.

RWW: Again, the publication of the defamatory statement constitutes the required legal injury, here an injury to reputation, dignity, whatever, that is distinct from the conduct that caused that injury, which I have a duty not engage in.

When you say that, as a matter of definition, we require, in addition,
causation and attibutable responsibility, causation of what, attributable
responsibility for what?

RWW: See above.  Causation of the specified legal injury for the particular tort.

> The issue in the Sikh case does not, as Rob implies, turn on the the
> debate between those, like me and the Restatement Third, who agree with
> Andrews' view of a duty owed to the world at large not to create
> unreasonable foreseeable risks to others (which, by the way, the English
> courts have adopted without acknowledging doing so by treating any
> nonfantastic risk as foreseeable in Wagon Mound 2), rather than Cardozo's
> narrow (and incoherent given the many exceptions made to it) view of a
> duty owed only to specifically foreseeable (classes of) plaintiffs (on
> both points, see my article, The Grounds and Extent of Legal
> Responsibility, in the San Diego Law Review, which is accessible through
> the hyperlink in my 'signature').

I think the result in this case will be the same regardless of the view
one adopts. I was responding to the argument that I was confusing duty,
breach and causation.

RWW: You were the one who mentioned the Cardozo-Andrews divide as if it were the answer to my argument.

 As I stated in my original post, the
> victim in the Sikh case was within the class of foreseeable plaintiffs
> (assuming it is foreseeable that persons with allergies to eggs would or
> might reasonably rely on there being no eggs in the food served at the
> Sikh event), so a duty was owed according to Cardozo's view as well as
> Andrews' view.  Cardozo himself separated the issue of duty from the
> subsequent distinct issues of causation, both actual and 'proximate'.
>

Absolutely

> Rather, what divides Rob from me (and, I believe, many others) is whether
> one has the duty concept swallow up the entire tort, by including within
> the duty analysis the issues of causation and attributable responsibility.
>  Rob believes that a duty is owed to others only not to injure them (in
> foreseeable, nonremote ways) through conduct that creates an unreasonable
> foreseeable risk of such injury.  No matter how unreasonable one's conduct
> is and no matter how foreseeable and significant the risks to others, one
> has no duty not to engage in such conduct unless it subsequently turns out
> to have actually injured others (in a foreseeable and nonremote way).
> For Rob, duty is equivalent to wrong

No. Breach of a duty owed to anothr is a civil wrong. Duty itself does not
mean wrong.

RWW:  I guess I don't understand how you formulate duty distinct from breach of duty.  I thought you believe that the duty is a duty not to injure (by certain types of conduct), which thus (tautologically) is breached if and only if one injures (by certain types of conduct).  If you do have a concept of duty that does not include non-injury, what is it?  In any event, your notion of breach of duty apparently encompasses all the requirements for a civil wrong, including actual causation and attributable responsibility.  I have the same objections to this formulation of breach as I do to what I thought was your formulation of duty.

 and requires ex post actual causation
> and attributable responsibility analysis as well as ex ante foreseeable
> risk analysis.

So, on my view the causal enquiry doesn't disappear, rather whether there
has been a breach will, in part, turn upon whether the defendant's conduct
has caused injury to P. The conduct is not the wrong in itself.

RWW:  Agreed, the conduct is not the wrong in itself.  In the usual view, the tortious conduct is the negligent conduct, the 'wrongful' (in a loose sense) conduct, the conduct that will subject you to liability if it actually and attributably results in the required legal injury for the particular tort.

  He, I assume, defines negligence or 'potentially liability
> creating conduct' not as a breach of duty but rather as creating an
> unreasonable foreseeable risk to others.  But what makes it unreasonable?
> It would seem that, according to Rob, it is not unreasonable if not
> wrongful in a strict sense, that is, if it does not end up actually
> causing injury to another.

Getting drunk and driving round a roundabout is very unreasonable. It just
isn't, without more, a civil wrong. Lots of negligent conduct is not
wrongful. Negligence in the air is still negligence, just not a tort.

  It is not even enough in the Sikh case for the
> unsuspecting guest to have eaten the egg (as Rob states); it must also be
> the case that eating the egg caused an allergic reaction, and that the
> allergic reaction caused physical harm.
>

You need injury (ie a right violation).

RWW:  Agreed, entirely.  Injury is one of the several required elements for a wrong.  But not, as you sometimes seem to state, 'injury' or 'rights-violation' 'in the air'; rather, the specific injury required for the particular tort.  For negligence, generally some physical harm to the plaintiff's person or property.  Without such, there is no 'rights violation' and no 'wrong'.

> Perhaps Rob has some other name for what everyone else calls
> duty/breach/negligence.  It may just be semantics, but I don't find his
> semantics helpful.  Moreover, I don't think his approach to duty, which
> has also been espoused by Ripstein and Zipursky, is merely idiosyncratic
> semantics, I think it also is bad conceptually.  On his view, it seems one
> should not be able to enjoin risky conduct unless it is certain that the
> risky conduct will injure (in a foreseeable, nonremote way) a foreseeable
> victim, and perhaps not even then.  After all, on his view there has been
> no breach of duty until a cognizable legal injury has been suffered by the
> plaintiff.
>

That does not follow. It is unnecessary in order to obtain an injunction
to show that you are the victim of a wrong (injunctions in contract law,
for example, obviously don't require the breach actually to have
occurred). It is also unnecessary to show that the defendant will
certainly commit a wrong. The injunction may be awarded to prevent a
potential wrong from being committed, not just to stop a wrong which is
ongoing. However, do we all have standing to obtain injunctions against
anyone else to restrain them from engaging in unreasonable conduct? No, we
must rely upon threats to our own rights in order to obtain the relief.

I don't think this is just semantics. So, for example, my view is that if
you don't separate out correctly the two causal questions (i) has D's
negligent conduct injured P and (ii) what consequences has the wrong P has
suffered had, that you soon get into difficulties. You may even argue that
X can cause Y, even where Y would have happened regardless of X.

RWW: I agree that there are distinct causal issues.  There is the issue of whether any wrong has been committed at all, that is, whether the plaintiff has suffered the required legal injury for the particular tort as an actual and attributable result of the defendant's tortious conduct.  There may also be a separate issue of whether that legal injury has actually and attributably resulted in further, consequential harm.

RWW: You should not throw such a flame at me, Rob! :>)  As you know, it opens up an entirely different debate.  Of course I believe, as I think any rational person must, that X can and does cause Y even where Y would have happened regardless of X.  It is the old two fires, each independently sufficient, scenario, where they combine to burn down a house (each is a 'duplicative' cause), or the shot which immediately kills a person bound to die due to having drunk a slow-acting but fatal poison for which there is no antidote (shot is a preemptive cause, poison is a preempted, noncausal condition).  To hold otherwise is to believe in mysteriously uncaused events.

best
Rob
--
Robert Stevens
Professor of Commercial Law
University College London

best back,
Richard