In my view, the best way to describe Bonnington is not as a but-for causation case (which I agree with Sandy is unhelpful), but as a mitigation of the strict application of the but for test.

Richard

Dr Richard Goldberg
Reader in Law
School of Law
Taylor Building
King's College
University of Aberdeen
Old Aberdeen
AB24 3UB
Tel: 012224 272745
________________________________________
From: Sandy Steel [as730@hermes.cam.ac.uk] On Behalf Of Sandy Steel [as730@cam.ac.uk]
Sent: 18 December 2011 19:16
To: Robert Stevens
Cc: Wright, Richard; Neil Foster; David Cheifetz; obligations@uwo.ca
Subject: RE: ODG: Amaca v Booth- HCA on causation in asbestos cases

To say Bonnington is a but-for causation case is in itself unhelpful.

It is necessary (as ODG-ers have pointed out) to define the damage in
respect of which but-for causation was established.

The happiest interpretation would then be Rob's: D's breach of duty caused
some segment of the overall damage, because the breach made the disease
worse in extent. But, since apportionment wasn't before the HL, C succeeded
in full.

But the judgments are just ambiguous with regard to that view of the causal
mechanism. This theory that "the more you are exposed, the worse the extent
of your injury" was stated by counsel in the Court of Session, but it's not
clearly accepted by the judges in the HL (even if in 2011 we know that this
is the mechanism).

Lord Keith says that the disease probably wouldn't have happened when it
did but for D's breach. If the above causal mechanism was assumed, then
this would be an odd thing to say (even if logically consistent with such a
view of the mechanism). It also makes the defender's argument that the
disease would have happened anyway look pretty much incomprehensible if
that was the disease mechanism assumed.

My nit-picking point is that Rob is right, but in the HL this was not for
the reason that the disease is made cumulatively worse by exposure. Rather
the disease was treated as indivisible, but its occurrence was accelerated
by D's breach. (This is a pretty fine point). As Bailey in LS 2010 points
out, it should then have been the case that C recovers only for the loss up
to the point where the indivisible injury would have occurred anyway. That
question was not addressed, and so C recovered in full.

Even if you could take from Bonnington that D was a cause of the over-all
disease even if D's breach did not accelerate the contraction of the
disease (and this becomes a logically possible interpretation if the
disease was treated as indivisible), it still remains to ask whether the
disease would have happened anyway (whether you treat this as a causal
(Rob) or non-causal (Richard, David) issue. And it's for that reason,
namely, that, on whatever view, you need to ask a further question to
determine the extent of liability, and that question was not asked in
Bonnington, that the case should not be cited for anything like what it was
in Bailey v MoD.

Sandy

On Dec 18 2011, Robert Stevens wrote:

>
> However, there was no finding that it
>> was necessary for the disease, per se, rather than only for(theoretically
>> separable) increments in severity, under Rob's theory, in which case the
>> defendant should only be liable, under the but-for test, for those
>> increments?
>>
>
>Yes. The question of quantum of liability was not before the court.The HL
>in Bonnington Castings made no apportionment because the only quiestion
>argued before them was one of liability, not quantification. When the
>question of quantification has been argued in equivalent cases of
>increments in severity, an apportionment has, rightly, been made (eg
>Holtby v Brigham [2000] 3 All ER 421).
>
>Nobody would still be citing Bonnington Castings if it had not been cited
>for a proposition for which it did not stand in McGhee.
>Rob
>
>
>


The University of Aberdeen is a charity registered in Scotland, No SC013683.