I don't necessarily disagree with Adam about the need for caution in relation to appeal or review, but I think the problem is a bit deeper.
If you insist on dividing questions into "law" and "fact", and making that distinction decisive for the standard of review, you nearly always end up with anomalies. The meaning of a contract in a commonly used form (whether or not strictly speaking a "standard form" in the sense of being a contract of adhesion or a contract issued by an institution or a trade association) may be something that deserves close appellate scrutiny; the meaning of a contract which is a "one-off" may not. How much turns on the case for the parties may surely also matter. But if your only tool for deciding whether an appeal court should look at the case is an abstract binary classification as "law" or "fact", it's quite difficult to explain in principled terms why a misunderstanding of the very same words in one contract is "law" and in another is "fact".
Of course, strictly speaking the (English) Court of Appeal *doesn't* depend on such a distinction: you can have appeals on the facts as well as the law, and although lip-service is paid to the idea that the court is reluctant to interfere, the reality is messier, and *how* reluctantly it will interfere with even a definitely factual finding depends a great deal on the particular finding in question, and the reasons for it. In arbitration there is that distinction, but there the categorisation of something as law isn't (in England at least) determinative because there are additional requirements which mean that many points of law are not appealable in practice, which is probably as it should be, and in any event seen as necessary to allow England to compete as an "arbitration venue". Since the parties can anyway completely exclude appeal if they want to, and often do, it probably doesn't matter.
Basically, making appeal standards critically dependent on classification as law or fact is troublesome. And we end up with Sattva which effectively turns the thing on its head and avowedly categorises the question in order to induce what the court thinks is just the right amount of intrusion. Whenever you have a rule which says "You can X if Y" and you find a court saying "in order to decide if Y I am going to decide if I think it would be a good idea to X" you should smell a rat, because the rule has effectively become "You can call this Y if you think you should X". That is more or less where we have (always) got to with the law/fact distinction: a smokescreen or fig-leaf for instrumental reasoning.
If you adopt a Sattva sort of approach, you just end up with the same sort of uncertainty: a mixed question of fact and law which gives judges free rein to "extract" a legal question if they want to (and really, do we actually want *more* navel gazing over "the rules of legal interpretation", whose constant downpour from the Rainy Sky has filled page after page of law reports to precious little benefit in terms of predictability or clarification of principle for these many years?) or refuse to if they don't. It just creates a sort of intellectually dishonest satellite argument, where people pretend to be arguing about which of two incomprehensible categories a putative error should be put into, when the real discussion ought to be about whether it is worthwhile to have an appeal court consider the issue at all.
I'm not against the idea that there are plenty of cases where it might make good sense for an appeal court to say, "it shouldn't matter whether we agree with the judge or not: it's a close case which is important only for the parties and depended on a good deal of detailed evidence, and the judge's answer seems reasonable, so we'll let it stand even though we might have decided it differently ourselves". But insisting that *every* time a judge has decided "what a contract means" is one of those cases, or not, seems silly.
Grant or restriction of rights of appeal depends on too many things that can't be neatly captured in a rather fruitless attempt to sort questions (or errors) into two boxes. To me, that makes Sattva a rather dull case, not because its result is silly, but because it's dealing with a silly question (and in any event, not a question of *contract law* at all, but one of civil procedure).
Paul
--
Paul Stanley QC
Essex Court Chambers
24 Lincoln’s Inn Fields
London WC2A 3EG
+44 20 7813 8000
This email has been scanned for email related threats and delivered safely by Mimecast.
For more information please visit http://www.mimecast.com