Cart, R (on the application of) v The Upper Tribunal & Ors

The edifice of administrative and adjudicative tribunals created by the Tribunals, Courts and Enforcement Act 2007 (TCEA) is a landmark in the development of the United Kingdom’s organic constitution. For the first time, a single structure has been created within which a huge variety of existing tribunals is gathered.

TCEA Section 3 provides:

The First-tier Tribunal and the Upper Tribunal

(1)

There is to be a tribunal, known as the First-tier Tribunal, for the purpose of exercising the functions conferred on it under or by virtue of this Act or any other Act.

(2)

There is to be a tribunal, known as the Upper Tribunal, for the purpose of exercising the functions conferred on it under or by virtue of this Act or an other Act.

(3)

Each of the First-tier Tribunal, and the Upper Tribunal, is to consist of its judges and other members.

(4)

The Senior President of Tribunals is to preside over both of the First-tier Tribunal and the Upper Tribunal.

(5)

The Upper Tribunal is to be a superior court of record.

Not only does s.3(5) declare the Upper Tribunal to be a superior court of record; s.25 gives the UT in the discharge of its adjudicative functions “the same powers, rights and privileges and authority as the High Court”; and by s.1 the guarantee of judicial independence contained in s.3 of the Constitutional Reform Act 2005 is extended to all tribunal members. It is accepted on all hands that tribunals must act judicially, whether adjudicating between citizen and state or between persons, and must abide by their statutory remit.

The question in this appeal is whether, despite its status, the Upper Tribunal is subject to the jurisdiction of the High Court by way of judicial review and, if it is, on what grounds. The Divisional Court (Laws LJ and Owen J) [2009] EWHC 3052 (Admin) concluded that it was amenable to judicial review, but only on grounds of pre-Anisminic jurisdictional error (which we will call outright excess of jurisdiction) or a denial of the right to a fair hearing (which we will call denial of procedural justice). Neither of these described what had happened to Mr Cart before the Upper Tribunal. The court accordingly gave him permission to apply but dismissed his application.

We have reached the view, albeit for reasons which differ in part from those of the Divisional Court, that its conclusion of law was correct. If it was, Mr Drabble concedes that his appeal must fail.

In setting out our reasons for so deciding, we are able gratefully to rely, without reproducing it, upon the exposition of the nature and content of the new tribunal structure and on much else besides contained in the judgment of Laws LJ. The judgment covers not only the present case but a second and distinct pair of applications concerning the Special Immigration Appeals Commission: these resulted in a conclusion, now unchallenged, that full judicial review was available against SIAC. In what follows we shall take the material parts of the judgment as read.

TCEA is a statute of nationwide application – necessarily, since many chambers of the FTT and UT have absorbed tribunals common to all four parts of the United Kingdom. The Outer House (Lord Glennie) on 31 March 2010 decided a petition in the case of Eba raising questions similar to those in the present proceedings. Lord Glennie had regard to the decision of the Divisional Court in the present case and came to a similar conclusion without adopting the entirety of its reasoning. The Inner House (Lord Hamilton, Lord Kingarth and Lord Brodie) heard Eba on appeal a few days before we sat to hear the appeal in Cart. Since their decision is not yet known, it has seemed appropriate to counsel and to this court, without intending the least discourtesy to Lord Glennie’s cogently reasoned decision, not to embark on analysis of it but to leave this to the Inner House. It may well be that the two cases will meet in another place, but for the present they are running on parallel but separate rails.

Before this court, in addition to hearing Richard Drabble QC on Mr Cart’s behalf, we have heard James Eadie QC on behalf of the Child Maintenance and Enforcement Commission as respondent and of the Secretary of State for Justice as an interested party. The Upper Tribunal itself, properly, has taken no part. We have also read and been much assisted by the submission prepared pro bono by Michael Fordham QC and Tim Buley for the Public Law Project, which has intervened by permission of the Master of the Rolls.

Mr Eadie’s argument has been that the Divisional Court was wrong to hold that judicial review of any kind lay to the UT, but that, if it was right, the power of review is as restricted as the Divisional Court held it to be. Mr Drabble, here as below, has argued principally for a continuing power of judicial review in relation to statutory functions, including adjudications on child maintenance, which in the pre-Leggatt system were judicially reviewable. It has therefore been of considerable assistance to have the Public Law Project’s submissions in support of the larger proposition, which Mr Drabble supports, that the entirety of the UT’s functions is reviewable by the High Court either on ordinary principles or on the principles governing second appeals.

Mr Eadie’s foundational argument is that, if Laws LJ was right to describe the UT as an alter ego of the High Court, but equally if that was an overstatement of the UT’s status, he was wrong to hold that the UT was in any measure subject to the supervisory jurisdiction of the High Court. Mr Fordham’s answer is that, far from being an alter ego of the High Court, the UT is simply an orderly restructuring of a multiplicity of inferior tribunals which remain subject to judicial review as before.

Mr Eadie, founding principally on s.3 and s.25, as well as on its appellate and judicial functions submits that these make the UT a body of equal power and standing to the High Court, placing it logically and necessarily beyond the latter’s supervisory jurisdiction. Mr Fordham submits that they do the opposite: to be a superior court of record is, on authority, not – or not necessarily – to cease to be a tribunal inferior to the High Court; and to be explicitly given a number of powers and privileges derived from those of the High Court is, he submits, to underscore that the recipient would otherwise not possess them.

But it does not follow from this that everything which is justiciable by way of judicial review of other bodies is justiciable in relation to the UT. Mr Drabble, perfectly properly, has confined his argument to the proposition that, whatever the law may be in relation to other chambers of the UT, a jurisdiction which has historically been open, with beneficial results, to judicial review ought not to lose that characteristic simply because it has been transposed into a new structure. If Mr Cart can establish this, he is not concerned about other jurisdictions within the FTT and UT structure. But, with respect, we are, because any such decision will have indeterminate effects on a considerable range of tribunal functions. For this reason we are as much, if not more, concerned with Mr Eadie’s argument that judicial review of the UT as a whole can run no wider than the Divisional Court held it to do, and with Mr Fordham’s counter-argument that, once reviewability is established, there is no principled basis for restricting its ambit.

In addition to its constitutionally salient features described earlier in this judgment, the UT is in certain classes of case given a jurisdiction which replicates the judicial review jurisdiction of the High Court. Laws LJ describes these powers of the UT in detail at §14-15. It is sufficient for us to say that, in our judgment, if the structural aspects of the UT’s status which are relied on by Mr Eadie cannot place it beyond the judicial review power of the High Court, this adjectival aspect cannot help. Far from investing the UT with a status equivalent to that of the High Court, it recognises that the UT lacks it and that ad hoc statutory authority is needed if it is to exercise any such powers. The same has been true for a number of years of the county court, which by s.204 of the Housing Act 1996 is given a jurisdiction which is “in substance the same as that of the High Court in judicial review” (per Lord Bingham in Runa Begum v Tower Hamlets LBC [2003] 2 AC 430, §7) without its ever having been suggested that the status of the county court is thereby altered.

More relevantly, the UT is itself an appellate body: see Laws LJ §11-13. We consider this to have an important bearing on the ambit of judicial review of it, to which we shall come; but we do not accept that an appellate function in itself, given the historical structure of many tribunals, can be an index of complete immunity from judicial review.

It is to be noted, too, that Sir Andrew Leggatt in the report Tribunals for Users (2001) on which the TCEA is built dealt at §6.31 ff with the possibility of judicial review of the new structure. He suggested (§6.33) that making the tribunals a superior court of record would avoid judicial review (a view which we deal with below) but thought it artificial and likely to “blur the clear distinction which we wish to achieve between the courts and the Tribunal System”. The other option, he said (§6.44), was to exclude it by express statutory provision as had been done for limited purposes in respect of the Crown Court by s.29(3) of the Supreme Court Act 1981. This course was not taken in TCEA.

There remain the important indicia to which we referred earlier and on which Mr Eadie relies. The problem with s.25 is that it is equally explicable as a badge of status and as a recognition that, but for the express provision it makes, the UT would lack the inherent powers enjoyed by the High Court. The problem with s.3(5) is that the designation “superior court of record”, while sometimes believed in England to connote immunity from supervisory jurisdiction (see for example R v Regional Office of Employment Tribunals (London North), ex parte Toyin Sojirin unreported, 21.2.2000, per Sedley LJ at §8), is – we understand – a concept unknown to Scottish law and in the law of England and Wales is in truth a concept of uncertain import.

This aspect of the case is the subject of a detailed and scholarly treatment in the judgment of Laws LJ. He concludes first (§28-42) that it would violate the rule of law to permit the long-established power of the High Court to control excesses of jurisdiction by statutory tribunals to be stifled by an implication, even if ouster is what the designation “superior court of record” implies. But secondly, following a voyage (§43-75) through legal history and literature which will command the admiration of any reader, he concludes that the designation “is not a reliable guide, let alone a definiens” of courts which are immune to judicial review. We need say no more than that we agree with him.

In the course of his discussion of the UT’s status, however, Laws LJ remarked (§39):

“To offer the same guarantee of properly mediated law, any alternative source must amount to an alter ego of the High Court.”

Later, after separately considering the common law’s own doctrine and policy in relation to the supervision of other courts and tribunals (§76-81), he turned to the correct characterisation of the UT in relation to the High Court (§87 ff) and reached this conclusion (§94):

“In my judgment UT is, for relevant purposes, an alter ego of the High Court.”

Laws LJ went on to explain exactly what he meant by this. Mr Eadie for his part felt able to found upon it in his skeleton argument: how, if the one was the equal of the other, could it be subject to the other’s supervisory control? The answer, as it seems to us, is that the UT is not an avatar of the High Court at all: far from standing in the High Court’s shoes, as Mr Fordham puts it in his written submission, the shoes the UT stands in are those of the tribunals it has replaced. And in oral argument Mr Eadie, with a realism for which the court is grateful, accepted that he could not sustain the proposition that the UT was the alter ego of the High Court. His case is that Parliament, for the reasons we have been considering, has taken a clear policy decision to place the UT wholly beyond the reach of judicial review.

For the reasons which we have given, we do not agree. In our judgment, as in that of the Divisional Court, the supervisory jurisdiction of the High Court, well known to Parliament as one of the great historic artefacts of the common law, runs to statutory tribunals both in their old and in their new incarnation unless ousted by the plainest possible statutory language. There is no such language in TCEA. The statute invests with standing and powers akin to those of the High Court a body which would otherwise not possess them precisely because it and the High Court are not, and are not meant to be, courts of co-ordinate jurisdiction.

While this conclusion, if right, establishes the possibility of control by the High Court of the legality of the UT’s acts, we recognise that further questions about the scope of the UT’s conferred or inherent powers may arise. These lie beyond the remit of the present proceedings and are questions for another day.

Mr Drabble, supported by the Public Law Project’s submission, contends that if judicial review runs to the UT, that is an end of the case: there is no warrant for cutting down a priori the range of grounds for intervention painstakingly built up by public law since the breakthrough in Anisminic. Of course the grant of permission to apply and of relief remains discretionary, so that it is not every grievance about the UT which will secure judicial review. But if the UT has made a substantive or procedural error of law, albeit within jurisdiction, it is the role of the High Court to say so and to correct it.

This is a powerful claim. It confronts the court with its own jurisprudence of containment of executive and subordinate judicial power within lawful limits, a jurisprudence which it has in part rediscovered and in part developed over the last half century, and challenges the court to have the courage of its convictions. In particular it is able to rely on what this court said in R(Sivasubramaniam) v Wandsworth County Court [2003] 1 WLR 475, §54: that the mechanism of control in judicial review lies in discretion, not law.

But what the claim overlooks is that the reach of judicial review is itself an artefact of the common law. Until 1969 it was tied up in knots of growing complication which the decision of the House of Lords in Anisminic severed at a blow. Since then, any error of law, whether it related to the limits of the tribunal’s powers or lay within them, and any denial of procedural justice has been justiciable. This has applied equally to lower courts of law, but always subject to the principle that judicial review is a remedy of last resort, so that errors of law within jurisdiction have to be corrected, if they can be, by appeal.

For many years it was unnecessary to decide whether the principle that judicial review is a remedy of last resort is a principle of law or a guide to the use of discretion, since in either case the result was going to be the same. It was in Sivasubramaniam that the attempt to cross the divide was rejected by this court. The claim, it has to be said, was a highly eccentric claim, manifestly doomed to failure: the question was when and how to give it its quietus. This court accepted even so that judicial review went in principle to judges of the county court, but held – on established authority - that permission to apply for it would not be granted where satisfactory alternative recourse existed, whether or not it had been exhausted. That was so in Mr Sivasubramaniam’s case. In exceptional cases it might not be so. One such exception, this court held (§52), was asylum cases, because of their unique subject matter. Two others (§56) might well be a pre-Anisminic excess of jurisdiction and procedural irregularity which denied the right to a fair hearing - what we characterised at the start of this judgment as outright excess of jurisdiction and denial of procedural justice.

The rightness of this reservation is not in question. To permit a lower court or tribunal to embark on an adjudication which Parliament had given it no authority to embark on would be a dereliction of the High Court’s duty. So would standing by in the face of evidence that it had denied a party the fair hearing to which everyone is entitled, whether because of actual or ostensible bias or because of some fundamentally unfair procedural step: forms of challenge to which judicial review is in fact peculiarly well suited since, unlike appeal, it is able to investigate and establish the relevant facts.

What was not affirmatively decided in Sivasubramaniam is whether this restriction of the grounds of judicial review was one of legal policy or of judicial discretion. The closest the court came to it was to say in §54 that where applications for judicial review were no more than collateral attacks on appellate decisions taken within an adequate system, judges of the Administrative Court “should dismiss them summarily in the exercise of their discretion”. No more needed to be said there. Here, however, more is needed.

In our judgment the scope of judicial review available in relation to any amenable decision-making body is necessarily a matter of law. As Lord Diplock said in R v IRC, ex parte National Federation of the Self-Employed [1982] AC 617, 639-40, the rules of standing in judicial review

“were made by judges; by judges they can be changed; and so they have been over the years to meet the need to preserve the integrity of the rule of law despite changes in the social structure, methods of government and the extent to which the activities of private citizens are controlled by governmental authorities….”

What is true of the rules of standing is equally true of the substantive principles of judicial review.

The complete reordering of administrative justice brought about by the legislative enactment of Sir Andrew Leggatt’s report Tribunals for Users (2001) (Footnote: ¹) is in our judgment such a change. It calls for a reconsideration of the principles of law by which judicial review of the new tribunals is to be governed. That the High Court (and this court on appeal from it) is empowered to do this derives from the very characteristic that distinguishes the High Court from the tribunals: its status as a court of unlimited jurisdiction makes it “the sole arbiter as to what matters fall within its jurisdiction”: de Smith JudicialReview (6th ed) §19-020.

It is in this context, rather than in excluding judicial review a priori, that the factors highlighted by Mr Eadie are entitled to real weight. The tribunal system is designed to be so far as possible a self-sufficient structure, dealing internally with errors of law made at first instance and resorting to higher appellate authority only where a legal issue of difficulty or of principle requires it. By this means serious questions of law are channelled into the legal system without the need of post-Anisminic judicial review.

What then of Mr Drabble’s contention that social security is a special class of case which has historically, and beneficially, been subject to such review notwithstanding the high legal expertise of the Commissioners, and for that reason should continue to be so? We accept the premise but not the conclusion. One of the principal purposes of TCEA is to unify the procedures of the disparate tribunals gathered into its structure. It contains no space for historical exemptions of the kind Mr Drabble seeks to establish. We put it this way because we are mindful that in Sivasubramaniam this court acknowledged an exception, based on the uniqueness of the subject-matter, in relation to judicial review of asylum decisions. Whether that will continue to be so in the light of what is decided in the present case both here and in Scotland is not for us to predict.

But this leaves open the larger question: why should the full ambit of judicial review not remain available as before across the board?

Mr Drabble makes an attractive case that such review in, for example, the social security and the immigration and asylum fields, has been a salutary fallback where conventional appeal mechanisms have proved inadequate, and has enabled justice to be done where otherwise it would not have been. If cases have occasionally gone on to the highest appellate level, that is a mark of the value of review, not an argument against it. Mr Eadie counters that neither justice nor its resources are infinite; that a comprehensive appeal to a judicial tribunal is how Parliament has provided for the doing of administrative justice; and that judicial policy can and should recognise this by drawing a line which, if it is to be penetrable at all, should allow review only on grounds of radical transgression.

We see much force in both these approaches. Mr Eadie is right to say that finality is not only inevitable but necessary; but we bear in mind what Lord Atkin said about it in Ras Behari Lal v King-Emperor (1933) 60 IA 354, 361:

“Finality is a good thing; but justice is a better.”

Judicial policy in this context can and should correspond with legal principle. It seems to us that there are two principles which need to be reconciled in order to arrive at a proper judicial policy. One is the relative autonomy with which Parliament has invested the tribunals as a whole and the UT in particular. The other is the constitutional role of the High Court as the guardian of standards of legality and due process from which the UT, for reasons we have given, is not exempt. Although central government has opposed this appeal, its interest in ensuring that departures from legality and due process do not occur is at least as great as that of private individuals.

It seems to us that there is a true jurisprudential difference between an error of law made in the course of an adjudication which a tribunal is authorised to conduct and the conducting of an adjudication without lawful authority. Both are justiciable before the UT if committed by the FTT, but if committed by the UT will go uncorrected unless judicial review lies. The same of course is true of errors of law within jurisdiction; but these, in our judgment, reside within the principle that a system of law, while it can guarantee to be fair, cannot guarantee to be infallible. Outright excess of jurisdiction by the UT and denial by it of fundamental justice, should they ever occur, are in a different class: they represent the doing by the UT of something that Parliament cannot possibly have authorised it to do.

Thus if for some reason the UT made an order giving a money judgment which it had no power to give, with the possibility of enforcement under its s.25 powers, it would be inimical to the rule of law if the High Court could not step in, should the appellate system for some reason not do so. Similarly if a member of the UT were to sit when ineligible or disqualified by a pecuniary interest, or if the UT conducted a hearing so unfairly as to render its decision a nullity, the High Court ought to be able to quash the determination. We do not mean this list to be exhaustive but to be illustrative of the kind of error, rare as it will be, which would take the UT outside the range of its decision-making authority. Such a division is, we consider, one of legal principle which can properly form the basis of judicial policy. It applies only to the UT, since it is the role of the UT itself to correct errors of every kind, including outright excesses of jurisdiction and fundamental denials of justice, in the FTT.

But it is not the only possible division short of unrestricted judicial review. In Wiles v Social Security Commissioner [2010] EWCA CIv 258 this court (Sedley, Dyson and Longmore LJJ) considered one of the last applications for permission to seek judicial review of a Social Security Commissioner’s determination before the transfer of the jurisdiction to the Upper Tribunal. In it Mr Eadie, for the Commissioner, submitted that, following both Sivasubramaniam and the Divisional Court’s decision in the present case, judicial review should be restricted to outright excess of jurisdiction and fundamental denials of procedural justice. Alternatively, following this court’s decision in R (Sinclair Gardens Investments Ltd) v Lands Tribunal [2005] EWCA Civ 1305, he submitted that it should be restricted to difficult and unresolved issues of law of general significance. For the claimant, Mr Drabble contended that any material error of law was and ought to be justiciable by way of judicial review.

Dyson LJ (as he then was), giving the leading judgment, concluded:

45.

If the scope of the jurisdiction to grant judicial review in respect of a refusal by a commissioner to grant leave to appeal had not been established for almost 30 years, I would have been inclined to adopt a position somewhere between those contended for Mr Drabble and Mr Eadie. I would reject Mr Eadie’s primary position. I can see no good reason why the court should not have power to grant judicial review of a refusal of leave to appeal in a case which involves a difficult point of law of general importance. It is clearly in the public interest that the court should be able to decide such issues. Accordingly, if exceptional circumstances were the correct test, I would be inclined to include in the category of exceptional circumstances those cases which raise a point of law of general importance (not necessarily circumscribed in the way suggested by Neuberger L.J. in Sinclair Gardens at [57]).

46.

But in my judgment, there is considerable force in the submission that the categories of case in which judicial review should in principle lie in respect of a refusal of leave to appeal by a commissioner should not be limited to exceptional circumstances. In Sivasubramaniam it was accepted by the court that the practice of entertaining applications for permission to apply for judicial review of refusals of leave to appeal by the now defunct immigration appeal tribunal (“IAT”) was justified. The “special factors” justifying this practice were identified at [52]. I accept the submission of Mr Drabble that the nature and functions of the social security commissioners are closer to those of the IAT than to either the county court or the Lands Tribunal. They are an administrative tribunal, frequently called upon to adjudicate on significant legal issues which have far-reaching consequences well beyond the individual case, including important issues of human rights and EU law. I accept that issues such as the right to life and the right not to be tortured are unlikely to arise in a social security case. But a social security case may well involve the right of a claimant to subsistence income and so directly affect their access to the most fundamental necessities of life.

47.

It seems to me that there is much to be said for opening the door somewhat wider than Mr Eadie would allow to reflect the fact that (i) issues that arise in social security cases may affect the lives not only of the individual claimant, but of many others who are in the same position, some of whom are among the most vulnerable members of our society; and (ii) the issues may be of fundamental importance to them, sometimes making the difference between a reasonable life and a life of destitution.

48.

For these reasons, if the matter were free from previous authority, I would have been inclined to hold that the door to judicial review should be opened wider than Mr Eadie has submitted, even on his alternative argument. How much wider? In my judgment, there is much to be said for the criteria which the court applies in deciding whether to give permission to appeal for a second appeal. Section 55(1) of the Access to Justice Act 1999 provides that no appeal may be made unless it is considered that “(a) the appeal would raise an important point of principle or practice; or (b) there is some other compelling reason for the Court of Appeal to hear it.” It seems to me that this formula would strike a fair balance between the competing considerations which arise where a commissioner refuses leave to appeal.

49.

But I do not find it necessary to reach a concluded view on this, since I am persuaded by Mr Drabble that we should not depart from the approach (most clearly and fully articulated in Connolly) that has been established and applied by the courts for more than 25 years.

Longmore LJ added:

79.

I agree with Dyson LJ that the comparatively long line of authority permitting the court to grant judicial review on orthodox grounds of a decision by a Social Security Commissioner to refuse to give permission to appeal to himself from a decision of the SSAT should not be disturbed at this late stage in its existence. Now that the Commissioners have become part of the Upper Tribunal, no doubt the forthcoming decision of this court in Cart will be applicable in future and there may be a shift in the judicial review perspective. If there is, I would warmly endorse Dyson LJ’s view that it might be appropriate to adopt a similar test to that imposed by statute on the Court of Appeal in respect of second appeals.

Sedley LJ expressed no view on this aspect of the case. But what made sense only a few months ago to the other two members of the court, Mr Drabble submits, still makes sense notwithstanding the transfer. The same experienced judicial officers are interpreting and applying the same law. All that has changed is their name. Why should judicial review of their decisions, which has worked well for decades, suddenly retreat?

The answer, in our judgment, is that the new tribunal structure, while not an analogue of the High Court, is something greater than the sum of its parts. It represents a newly coherent and comprehensive edifice designed, among other things, to complete the long process of divorcing administrative justice from departmental policy, to ensure the application across the board of proper standards of adjudication, and to provide for the correction of legal error within rather than outside the system (Footnote: ²), with recourse on second-appeal criteria to the higher appellate courts. The account of the system given by its first Senior President, Carnwath LJ, ‘Tribunal justice – a new start’ [2009] PL 48 bears out not only these characteristics but its potential to develop a legal culture which is not in all respects one of lawyers’ law – a system, in other words, of administrative law. While it could no doubt cope with any of the models of judicial review which we have described, the one which seems to us to implement Parliament’s intent in enacting the Leggatt reforms is one which secures the boundaries of the system but does not invade it. That is the Sivasubramaniam model.

Mr Cart’s case illustrates how the division of labour on which we have settled plays out in practice. The injustice of which he justifiably complained was that the initial decision to vary child maintenance in his wife’s favour was made without notice to him from the Secretary of State. He appealed to the FTT on a number of grounds including this one. The FTT dismissed his appeal on its merits, and Commissioner Jacobs (as he still was) refused him permission to appeal on the present issue. On renewal the UT (Carnwath LJ and, as he had now become, Judge Jacobs) maintained the refusal, and at the hearing of an appeal on other points which they had given permission to argue they declined to reopen the point.

As can be seen from the UT’s decision [2009] UKUT 62 (AAC), §26-30, the reason was that the FTT had given full and fresh consideration to the need for a variation, so that the effect of the unfairness had been entirely cured. Public law has taken different positions at different times and in different situations on the legal effect of denial of an initial fair hearing when followed by a fair hearing on appeal. The UT in §27 picks this up and sets itself on a course on which, with all due respect, it is not likely to be assisted by the intervention of the High Court:

“The effect of procedural failings no longer depends on whether the legislation is analysed as mandatory or directory. The focus is on the impact of the failure and on presumed statutory intention (see, for example, Lord Steyn, in R v Soneji [2006] 1 AC 340 paragraph 23).”

We recognise that this will not always be the case, and that from time to time there will arguably be serious errors of law or process which escape both the prescribed appellate procedure and the ring-fence which the Divisional Court and this court have now constructed. Experience may lead the common law, which remains the sole source of judicial review, to rethink its position; but for the present, for the reasons we have given, we agree with the conclusion of the Divisional Court and dismiss this appeal.