I don't, without more, understand how my definition fails. The class of
users who have potentially suffered a loss as a result of the failure to
warn are those who would have read and acted on the warning. The failure
of the cosmetic manufacturer to warn this class of users of its
carcinogenic properties, has increased the risk of cancer for all members
of that class of person. If the claimant was a member of this class of
users, the failure to warn him increased the risk for him.

Similarly in Chester v Afshar the risks from the failure to warn were only
increased for the class of persons who would have acted on the failure to
warn by not having the operation. This plaintiff did not fall into this
class because she would have had an operation on a different day running
precisely the same risk anyway. The majority thought they were departing
from orthodoxy for policy reasons. I agree with that, but disagree as to
the adequacy of those reasons (and so am a Hoffmann-ian/Hoffmann-iac in
dissent).

Contrast the plaintiff in Chappel v Hart who if warned would have had a
lower risk operation with another surgeon. The risk for her was increased
and so the claim rightly succeeded. The division in the High Court of
Australia was between those who thought that evidence supported the
finding that the alternative operation would have been a lower risk one,
and those who did not. None took the view that this didn't matter.

I don't consider the Shavell definition to be orthodox, as can be
demonstrated by considering our leading decision (until Chester v Afshar)
on coincidence: Carlsogie v Royal Norwegian Govt [1952] AC 292, HL).

The plaintiff's ship was wrongfully damaged by the defendant. After
temporary repairs rendering the ship seaworthy she set sail on a voyage to
the United States. On that journey she suffered heavy weather damage,
which she would have done in that kind of storm even if completely
undamaged.

If the initial wrong had not occurred she would have been in another place
at another time and would not have been caught in the storm. However,
looking back from the time the event occurred, the risk of this loss was
not increased by the wrong for this plaintiff.

The Shavell definition without more (and I haven't checked if it is
fleshed out further) fails because looking forward from the time of the
wrong, one ship smashing into another does increase the risk of one ship
suffering further damage in a storm. Ships with gaping holes in their side
are less good in bad weather, and this was the "type" of loss suffered.
However, as things turned out, this loss was coincidental for this
plaintiff. As in Chester.

We could I suppose define "type" very narrowly, but my view is that that
approach will collapse into mine.

Rob



> We should all remind ourselves that Rob uses the term “coincidental”
> in
> a very strange and unorthodox way: “A loss is coincidental if, as things
> turn out, its risk was not increased FOR THIS CLAIMANT by the wrong
> which occurred.” (Torts and Rights 165)
>
> If we used Rob’s definition: when the manufacturer of a cosmetic fails
> to warn potential users that it causes cancer, the later cancer in a
> user is “coincidental” because “the risk of this loss was not
> increased
> by the failure to advise”.
>
> We should remind ourselves of the orthodox meaning of “coincidental”
> consequence. A consequence of breach is only coincidental where it
> “does not increase the probability of the TYPE of accident” (Steven
> Shavell, Economic Analysis of Law, Harvard Univ. Press 1987 p.111). If
> a warning would have reduced the number of consumers using the cosmetic
> (and therefore the number of cosmetic-caused cancers), the cancer
> contracted by an individual user is not a coincidental consequence of
> the manufacturer’s breach and he will be liable for it. That the failure
> to warn did not affect the risks to the individual user inherent in the
> cosmetic is irrelevant.
>
>
>
>
>>>> Robert Stevens <robert.stevens@ucl.ac.uk> 03/25/11 12:31 AM >>>
> I am quite happy to be called a "Hoffmann-ian".
>
>
>
> However, although there may have been a public law wrong in this case,
> in
> private law a tort is not constituted by having an unlawful policy as a
> matter of public law. Detaining someone against their will without
> lawful
> authority is the wrong.
>
>
>
> If I trespass on your land it doesn't cease to be trespass if you would
> have
> given me permission if I had asked nicely. The wrong is constituted by
> walking on the land without permission. the wrong is not constituted by
> the
> failure to ask nicely. Similarly that they could have detained the
> claimants
> lawfully doesn't alter the fact that here they did not.
>
>
>
> If my actions with respect to you are wrongful they are wrongful,
> regardless
> of whether I could have done the same thing non-wrongfully.
>
>
>
> I agree that in Chester v Afshar the wrong was constituted by failing to
> advise. It may have been appropriate to award a sum of money to
> vindicate
> this wrong (Hoffmann's 'modest solatium'). The decision of the majority
> in
> that case to allow her to recover the full loss she suffered as a result
> of
> the surgery, even though this loss was coincidental to the wrong (ie the
> risk of this loss was not increased in this case by the failure to
> advise)
> is unjustifiable in my view.
>
> Rob
>
>
>


--
Robert Stevens
Professor of Commercial Law
University College London