... which in turn suggests that, if there is a principled basis for combining the restitution and the contract doctrines, it isn't to be found in the mistake/misprediction distinction - unless of course there is a principled reason why one particular type of misprediction is treated differently from all the others.

If we ask rather about the basis of the transaction (or "the parties' common assumptions", if you prefer), the similarities are more straightforward. It is enough that the parties' agreement embodied an assumption, which turned out to be inaccurate. Whether the mistake could in principle have been discovered before the payment or not is beside the point. This leaves us with the problem of saying which types of assumption are fundamental enough to matter, but I think we have that problem anyway.

Of course, most of the cases will in practice involve mistakes - the argument is that it is not convenient or analytically correct to base them on that. Indeed, as we know, the same principles have been applied where one party was not mistaken at all, but nonetheless their transaction embodies a false assumption - as where money is demanded and paid on the basis that it is due, the payor already realising that it is not in fact due (Woolwich, the last 5 overpayments in Nurdin). The question is not, ultimately, what the parties expected or predicted, but what sort of circumstances their agreement provides for.

All of this, of course, suggests a very narrow ambit for the doctrine of frustration, as well as "mistake".

Steve Hedley

=============================================
FACULTY OF LAW, UNIVERSITY OF CAMBRIDGE

ansaphone : +44 1223 334931
www.stevehedley.com
fax : +44 1223 334967

Christ's College Cambridge CB2 3BU
=============================================