Earlier today, the Supreme Court gave judgment in Jones (by Caldwell) v First-tier Tribunal  UKSC 19. It contains an interesting discussion of the distinction between errors of fact and of law, and, I think, provides a further insight (following its decision in R (Cart) v Upper Tribunal  UKSC 28) into the Supreme Court’s approach to administrative law.
The sad facts of Jones are relatively straightforward. Barry Hughes jumped into the path of an articulated lorry on a busy dual carriageway and succeeded in his apparent attempt to kill himself. The lorry swerved, hitting a vehicle being driven by Gareth Jones. Jones suffered very severe injuries such that he now requires full time care. Via his mother, Jones sought compensation from the Criminal Injuries Compensation Authority. The relevant part of Jones’s argument before the CICA was that Hughes had inflicted upon Jones grievous bodily harm contrary to section 20 of the Offences Against the Person Act 1861. The CICA declined to award any compensation on the ground that it was not satisfied that the offence had been committed. As such, its power to award compensation—which arises when a “crime of violence” is committed—was not triggered.
The procedural background to the case is rather more complex, but needs to be explained in outline:
- First-tier Tribunal Jones unsuccessfully appealed against the CICA’s decision to the First-tier Tribunal. The FTT declined to overturn the CICA’s decision, holding that no s 20 offence—and hence no crime of violence—had occurred.
- Upper Tribunal Jones then sought judicial review (under s 15 of the Tribunals, Courts and Enforcement Act 2007) of the FTT in the Upper Tribunal. That was also unsuccessful.
- Court of Appeal There then was a further—and this time successful—appeal to the Court of Appeal, which held that the FTT had made an error of law by (the Court of Appeal thought) assuming that Hughes’s actions could not amount to a crime of violence. (In fact, the FTT concluded that his actions did not, on the facts, amount to such a crime.)
- Supreme Court Eventually, the matter reached the Supreme Court. It unanimously overturned the Court’s Appeal’s decision and upheld the original decision of the FTT.
The key issue in the case was whether it had been lawful for the FTT to conclude (in common with the CICA) that no “crime of violence” had been committed. If the Supreme Court had wanted to exercise a high degree of control, then it could have held that both the meaning of the term “crime of violence” and whether Hughes’s acts amounted to such a crime constituted questions of law. (Consider Khawaja v Home Secretary  1 AC 74 in which the court held that it had to be satisfied that the person concerned was an “illegal entrant”.) That would have allowed the Court to insist that the only lawful answers to those questions were the answers it considered to be correct. This, in turn, would have enabled the Court to hold that the FTT had erred when it concluded that no s 20 offence (and hence no crime of violence) had occurred on the ground that Hughes lacked the requisite mens rea.
Unsurprisingly, the Supreme Court chose not to proceed in this way. It was taken as read by Lord Hope, delivering the leading judgment, that a s 20 offence is necessarily a crime of violence within the meaning of the criminal injuries compensation scheme: in that sense, the classification of a s 20 offence for the purpose of that scheme is itself a question of law admitting of only one correct answer. But the logically prior question—whether Hughes’s conduct in the first place amounted to a s 20 offence, and, in particular, whether he had the relevant mens rea—was to be treated as a question of fact.
None of this is really very surprising. It is commonplace for reviewing courts to hold that whether a given set of circumstances satisfies a statutory criterion is to be treated as a question of fact, such that the decision-maker’s conclusion stands absent unreasonableness. The Supreme Court might have been presented with greater difficulty, however, if the FTT had arguably erred in relation to its understanding of what, in the first place, s 20 meant. That might have raised a “pure” question of law, rather than (as actually arose in the case) a question concerning the application of facts to settled law. The Court was not required to confront that issue—but it did hint at how it might have done so. For instance, Lord Hope said:
A pragmatic approach should be taken to the dividing line between law and fact, so that the expertise of tribunals at the first tier and that of the Upper Tribunal can be used to best effect. An appeal court should not venture too readily into this area by classifying issues as issues of law which are really best left for determination by the specialist appellate tribunals.
Meanwhile, Lord Carnwath—which whom the other Justices agreed—quoted extensively from his own article on “Tribunal Justice: A New Start”  PL 48, in which he had said that
the division between law and fact in such classification cases is not purely objective, but must take account of factors of “expediency” or “policy” … Logically, if expediency and the competency of the tribunal are relevant, the dividing line between law and fact may vary at each stage. Reverting to Hale LJ’s comments in [Cooke v Secretary of State for Social Security  3 All ER 279 paras 5-17], an expert appellate tribunal, such as the Social Security Commissioners [now replaced by the Upper Tribunal], is peculiarly fitted to determine, or provide guidance, on categorisation issues within the social security scheme. Accordingly, such a tribunal, even though its jurisdiction is limited to ‘errors of law’, should be permitted to venture more freely into the ‘grey area’ separating fact from law, than an ordinary court. Arguably, ‘issues of law’ in this context should be interpreted as extending to any issues of general principle affecting the specialist jurisdiction. In other words, expediency requires that, where Parliament has established such a specialist appellate tribunal in a particular field, its expertise should be used to best effect, to shape and direct the development of law and practice in that field.
Where does this leave us? It suggests that while the Upper Tribunal should intervene relatively readily in decisions taken by the FTT, the regular courts—whether they are judicially reviewing or determining appeals against decisions of tribunals—should tread more warily. It is already clear from R (Cart) v Upper Tribunal  UKSC 28 that tribunals’ decisions are unlikely in the first place to end up before the regular courts (whether on appeal or review) thanks to the applicability of the second-tier appeals criteria. But what Cart had very little to say about was the nature of the scrutiny that applies to those cases that clear the hurdles erected by those criteria. Viewed thus, Jones may be regarded as the conceptual counterpart to the highly pragmatic decision in Cart, in that Jones suggests that reviewing or appellate courts will be slow to discover an error of law. This is achieved by reliance upon the law/fact distinction. Lord Carnwath, for instance, quoted with approval from Lord Hoffmann’s speech in Moyna v Secretary of State for Work and Pensions  1 WLR 1929, in which he said that
there are two kinds of questions of fact: there are questions of fact; and there are questions of law as to which lawyers have decided that it would be inexpedient for an appellate tribunal to have to form an independent judgment [and which we therefore call questions of fact].
In Jones, then, we have confirmation (if it were needed) that the law/fact distinction has come to eclipse the distinction between jurisdictional and non-jurisdictional errors of law as the principal frontier in this area of judicial review. The latter distinction is largely spent thanks to Page, and so the heavy lifting falls to be undertaken by the former. But if the distinction between jurisdictional and non-jurisdictional errors of law is malleable, then that which distinguishes law from fact appears to be positively liquified. This is nowhere more clearly illustrated than in Lord Carnwath’s endorsement of the intellectual sleight of hand that would enable the Upper Tribunal to characterize as a question of law something that the regular courts would characterize as a question of fact: the characterization of the issue may differ depending on the nature of the body that is doing the characterizing. Like the jurisdictional/non-jurisdictional error of law distinction, the law/fact distinction thus becomes a conclusory device that signals but does not justify judicial (non-)intervention. Plus ça change.
Update For further comment on Jones v FTT, see Paul Daly’s post on his Administrative Law Matters blog and Alison Young’s post on the UK Constitutional Law Group blog.