When should a judge not be a judge?

Few people in this country, I would guess, reading this headnote to the official report of a recent decision of the US Supreme Court, would regard it as a difficult case:

After a West Virginia jury found respondents, a coal company and its affiliates (hereinafter Massey), liable for fraudulent misrepresentation, concealment, and tortious interference with existing contractual relations and awarded petitioners (hereinafter Caperton) $50 million in damages, West Virginia held its 2004 judicial elections. Knowing the State Supreme Court of Appeals would consider the appeal, Don Blankenship, Massey’s chairman and principal officer, supported Brent Benjamin rather than the incumbent justice seeking re-election. His $3 million in contributions exceeded the total amount spent by all other Benjamin supporters and by Benjamin’s own committee. Benjamin won by fewer than 50,000 votes. Before Massey filed its appeal, Caperton moved to disqualify now Justice Benjamin under the Due Process Clause and the State’s Code of Judicial Conduct, based on the conflict caused by Blankenship’s campaign involvement. Justice Benjamin denied the motion, indicating that he found nothing showing bias for or against any litigant. The court then reversed the $50 million verdict.

To all appearances, a key member of the appeal court which absolved the appellant company of liability had owed a debt of gratitude to that company. He should not have sat, and the decision in which he participated should be regarded as a nullity. What not everyone in this country will know is that in the 39 states of the US where judges are elected, it is routine for candidates’ campaigns to be sponsored by the law firms, and sometimes by the litigants, who will be appearing before them. Recusal in such states is going to require more than the occasional dollar in the judicial collecting tin.*

Would it matter if the same judge had in other cases found against the company that had sponsored his campaign? The risk of bias is not limited to favouritism. There is an equal and opposite risk that a judge, in endeavouring not to show favour, will bend over backwards and lose his balance that way. The only safe route is the exit.

The West Virginia court of appeals was by no means insensitive to the issue of apparent bias. When the petitioners applied for a rehearing of the appeal, there were motions for disqualification of three of the five original justices. Photos had been produced of one of them, Justice Maynard, vacationing with Don Blankenship on the French Riviera while the case was pending; Maynard correctly stood down. Justice Starcher too agreed to stand down because he had said publicly that ‘Blankenship’s bestowal of his personal wealth, political tactics and “friendship” have created a cancer in the affairs of this court.’ That his remark might have been entirely justified made no difference. But Justice Benjamin, now in the role of acting chief justice, refused to stand down.

What Europeans may find remarkable is not only that it was by a single vote that the US Supreme Court eventually struck down the West Virginia decision for apparent bias, but that it was the four judicial conservatives, Roberts, Scalia, Thomas and Alito, who voted to uphold the judgment. They may also now know that, thanks to a subsequent Supreme Court ruling, Massey need no longer channel its subventions through its chairman: it can donate to judicial re-election campaigns corporately and without limit. What is still uncertain is whether the Caperton decision means that corporations are likely to be wasting their money because the bare fact of their donations will disqualify the beneficiaries from sitting on their cases, or whether (as the majority judgment suggested) disqualification on this ground is going to be confined to facts as extreme as those in Caperton.

The seeds of Supreme Court abstentionism were sown long ago in the US. Justice Kennedy, writing for the Caperton majority, put it this way:

It is axiomatic that ‘[a] fair trial in a fair tribunal is a basic requirement of due process.’ As the court has recognised, however, ‘most matters relating to judicial disqualification [do] not rise to a constitutional level.’ The early and leading case on the subject is Tumey v. Ohio. There, the court stated that ‘matters of kinship, personal bias, state policy, remoteness of interest, would seem generally to be matters merely of legislative discretion’ … As new problems have emerged that were not discussed at common law, however, the court has identified additional instances which, as an objective matter, require recusal. These are circumstances ‘in which experience teaches that the probability of actual bias on the part of the judge or decision maker is too high to be constitutionally tolerable.’

The Tumey case was, at least to our eyes, another case in which there was a manifest risk of bias: it concerned a mayor with judicial powers whose municipal funds were swelled by any fines he imposed. It echoed the early decision of Chief Justice Coke in Dr Bonham’s Case. Justice Kennedy explained:

The Tumey court concluded that the Due Process Clause [of the Constitution] incorporated the common-law rule that a judge must recuse himself when he has ‘a direct, personal, substantial, pecuniary interest’ in a case. This rule reflects the maxim that ‘no man is allowed to be a judge in his own cause; because his interest would certainly bias his judgment, and, not improbably, corrupt his integrity.’

Those last words are Madison’s, writing in the Federalist in 1787, and they put in crystalline form what is still today the essential logic on both sides of the Atlantic. It is their application which is fraught with difficulty. The first main difficulty is to decide what makes someone else’s cause the judge’s own. The second, which is far from being merely technical, is who decides and how.

The dissenting judgment in Caperton sets out a list of 40 questions allegedly begged by the majority opinion, a list too long and in places too contrived to reproduce here. But all the questions drive at a single object: to demonstrate that once the test for recusal is set at any but the highest constitutional level, the grounds for removing judges from the court, and with them the spate of motions for recusal, will be unending. That is a serious consideration, and one which the British courts, which set the bar much lower, are already having to face up to. But serious as it is, it cannot be decisive. As Chief Justice Holt said three centuries ago, when the same argument was advanced in the great election corruption case of Ashby v. White, if wrongs are multiplied, remedies must be multiplied.

Even so, as Justice Scalia said in a pointed dissent:

The court’s opinion will … [add] to the vast arsenal of lawyerly gambits what will come to be known as the Caperton claim. The facts relevant to adjudicating it will have to be litigated – and likewise the law governing it, which will be indeterminate for years to come, if not for ever. Many billable hours will be spent in poring through volumes of campaign finance reports, and many more in contesting nonrecusal decisions through every available means.

Britain, like other countries in the Commonwealth, has already embarked on this road. But, at least in countries where judges are appointed and not elected, the road has not turned out to be a highway to hell; neither, however, is it a yellow brick road to contentment. It is a tortuous and sometimes stony road which is worth following even if its destination is uncertain.

When does somebody else’s cause become the judge’s own? When the judge has a private interest in that person’s prosperity, plainly. The case which set the tone in Britain, in 1848, involved the then lord chancellor, Lord Cottenham. Cottenham turned out to hold shares in the canal company in whose favour he had decided a case brought by a litigious solicitor named Dimes, who had bought a piece of land in order to hold the canal company to ransom for crossing it. Dimes had litigated without much success for more than ten years when luck delivered Cottenham into his hands, and he finished by pocketing the £700 which the canal company finally had to pay him to go away. The turning point, however, was not as neat or sharp as my account so far may have suggested.

Cottenham, on appeal, had affirmed an order made against Dimes by the vice-chancellor, the senior judge of the chancery division of the High Court, restraining him from interfering with the canal. When Dimes finally petitioned to have Cottenham’s order set aside, Cottenham correctly recused himself and handed the decision to the master of the rolls, Lord Langdale. Langdale was able to duck the issue on the ground that an appeal from the vice-chancellor lay to the lord chancellor and nobody else, so that disqualifying him was impossible. It was only when Dimes resorted to sabotaging the canal and its towpath, and Cottenham had to issue injunctions against him, that Dimes was able once again to challenge Cottenham’s jurisdiction. This time the House of Lords took the case. Dimes got nothing out of it, since the striking down of the lord chancellor’s decision left the vice-chancellor’s decision standing. But, although Lord Campbell went out of his way to say that ‘no one can suppose that Lord Cottenham could be in the remotest degree influenced by the interest that he had in this concern,’ the Lords held that he should not have sat.

It now seems that Lord Cottenham’s stake in the Grand Junction Canal Company was quite substantial. But the principle of recusal which Dimes established had nothing to do with the extent of the judge’s interest: it had to do solely with the fact of it, and the Lords took trouble to emphasise that the principle applied from the highest to the lowest tribunal in the land.

Pecuniary interest, precisely because it is completely unrefined, is therefore the simplest and most radical ground of recusal. It appears to operate independently of the will of the parties, rendering a decision void even if a party raises the matter only after finding it has lost. As a result, judges who recognise that a party to a case in their list is a company in which they hold shares make sure routinely that the case is tried by somebody else without waiting for an objection. But it has not been unknown in modern times for a judge, typically in the chancery division, to mention to the parties that he or she holds shares in one of them and to have the objection summarily waived, as often as not with a polite expression of hope that ‘your Lordship’s shares will prosper whatever the outcome.’ Trouble may be in store there.

Fast forward 150 years from Lord Cottenham to Lord Hoffmann in 1998, sitting as a member of the appellate committee of the House of Lords on the prosecutors’ appeal against the High Court’s decision to quash the Spanish extradition warrants against General Pinochet on the ground of sovereign immunity. A group of individuals and organisations including Amnesty International were allowed to appear by counsel as interveners. By three to two the law lords held that Pinochet could be extradited to Spain. One of the three composing the majority, Lord Hoffmann, was chairman of a trust which conducted Amnesty International’s charitable work in the UK, and his wife was an employee of the organisation itself.

Pinochet’s lawyers applied to the House of Lords to set aside its own judgment on the ground of apparent bias, and a differently composed appellate committee, in a still controversial judgment, acceded to the application. The law lords for the most part stressed that simply supporting a cause was not a disqualification: it was Lord Hoffmann’s active role as a trustee of an organisation that was siding with the prosecutor which made the difference. But Lord Goff of Chieveley pushed the door wider open: there could be cases, he said, where ‘the interest of the judge in the subject matter of the proceedings arising from his strong commitment to some cause or belief … could shake public confidence in the administration of justice.’

This seems to some observers a step too far. If the editors and public moralists whose views pass for public opinion are to be believed, the trouble with British judges is that (unlike editors and public moralists) they live lives remote from the real world. If every judge who supports a good cause is to be regarded as a partisan and disqualified from sitting on cases which touch on its subject matter, it will be hard to find a competent tribunal to try a controversial case.

What rings a clearer bell, and in fact became the law of the early Scottish courts, is the bright-line rule of the early canon law against adjudication by a judge linked to a party by consanguinity, affinity, friendship, enmity, or previous retainer as a party’s advocate. Like other early church laws, it was much honoured in the breach. When Urbain Grandier, the curate of Loudun, was accused in 1634 of bewitching and seducing the nuns in the local Ursuline convent, Cardinal Richelieu, whom he had displeased, made sure that a member of the court that retried and convicted him after an initial acquittal was a relative of the mother superior. You can see poor Grandier’s ashes to this day in the local museum. When a juror was challenged in 1572 because he was related in the ninth degree to a party, the court was driven to say: ‘All the inhabitants of the earth are descended from Adam and Eve and so are cousins of one another [but] the further removed blood is, the more cool it is.’

The principle is nevertheless a sound one even if in practice it is frequently unnecessary or occasionally unenforceable. In an appeal I argued at the bar against Shell and BP in 1980 about the lead additives in their petrol, the registrar of civil appeals was not able to assemble a court of judges who had no shares in either defendant. We had to make do with two judges who had shareholdings and one who did not.

The common law world today, however, casts the recusal net much wider. In one form or another it asks whether a sensible observer, knowing what the case was about, who the parties were and what connection the judge had with any of them or with the issues in the case, would think that the judge might be influenced by these things. The trouble with any such simple formula is that it immediately demands fuller detail. How cautious or suspicious is a sensible person? What kind of link will she think matters? How robust will she expect the judge to be?

The minority in the Caperton case set store by the last of these questions. The hypothetical observer needs to be aware that the judge on taking office has sworn a public oath to do right by all manner of people without fear or favour, affection or ill-will. How is the observer to gauge the risk that such a judge will break his or her oath? Indeed why, the Caperton dissenters asked, should the sensible onlooker expect it at all?

The first answer is that in the case of a pecuniary interest, the law itself presumes it. The second is that judges are human. The third is that, whatever the reality, appearances matter, and nowhere more so than in the administration of justice. The problem of freemasonry illustrates the second point. Lord Irvine, when he became lord chancellor, let it be known that he was minded to require judges to disclose whether they were freemasons. It was pointed out by objectors that this could well be an invasion of the right of association guaranteed by Article 11 of the European Convention on Human Rights. But what was more to the point, though never so far as I know acknowledged, was that it matters not at all if one person in court is a freemason. The problem arises if two freemasons are present and one is the judge, because masons are bound to help one another. So seen, the problem is not difficult to solve: any judge to whom a litigant or witness signals membership of the freemasons is obliged by his judicial oath to stop the trial. He is also, I would think, entitled to make a party who has sought favour in this way pay the costs of the consequent adjournment, and is probably also obliged to report the attempt to pervert the course of justice to the DPP.

Such judicial robustness is not always needed. The judge whose court is disrupted by abusive or violent conduct may be compelled to use his or her contempt powers on the spot. But in the absence of a pressing need to act, courts on both sides of the Atlantic recognise that the right course is for the offended judge to back off and to send the contemnor for trial by another judge. As Chief Justice Roberts said in his dissent in the Caperton case:

It may also violate due process when a judge presides over a criminal contempt case that resulted from the defendant’s hostility towards the judge. In Mayberry v. Pennsylvania, 400 US 455 (1971), the defendant directed a steady stream of expletives and ad hominem attacks at the judge throughout the trial. When that defendant was subsequently charged with criminal contempt, we concluded that he ‘should be given a public trial before a judge other than the one reviled by the contemnor’.

It may be that at least in this regard some of the judges in this country have the edge. One circuit judge some years ago was called a four-letter word by a defendant on whom he had just passed sentence. The judge had him brought back into the dock. ‘In a short while,’ he said, ‘I am going to go home to a good dinner. I shall spend the evening reading in front of the fire before retiring to a comfortable bed with my wife. You, thanks to your own greed and stupidity, are going to be taken in a van to a prison where you will have to slop out every day, eat prison food and stare at the same blank walls for the next two years. Now which of us would you say was the … ?’

So far I have mentioned only professional judges. Clearly, however, all this applies to the lay magistrates who try more than 95 per cent of our criminal cases. But they are not the only lay judges in our system. In all serious criminal cases, and in some civil cases and coroners’ inquests, the judges of fact are a jury of 12 ordinary people. How do these principles apply to them?

One of the principal justifications for the jury is its randomness. The abolition of the right of peremptory challenge in 1988 was a recognition of this. Unless some good reason for objection to an individual juror is apparent, both sides face judgment by 12 effectively anonymous individuals. We have never had, as the Americans have, a right to cross-question jurors-in-waiting in order to see whether there is cause for recusing them. But our random system brings its own problems. A juror may know the accused or a witness, or may have heard things about them from neighbours, or may work for a company involved in the case. He or she may have fixed opinions about something on which the jury is going to need an open mind. Yet without questioning there is no way of finding out.

The way most trial judges deal with this is to bring all the jurors in waiting into court, let them know who is to be tried, tell them what the case is about and get the court clerk to read them the list of witnesses. If any of the names or events are known to a potential juror, that juror is stood down. But this leaves gaps. I recall one juror, who had rightly said nothing when asked if she worked for the company whose property the defendant was accused of damaging, at the last minute raising a hand and asking whether it mattered that she and her husband had shares in the company.

But this kind of direct personal or economic connection is only the fringe of the problem. The real difficulty is what you might call institutional prejudice. Until 1973, jury lists included the jurors’ occupations; but in the ‘flying pickets’ trials which followed a prolonged and bitter building workers’ strike, the defence lawyers found that by using the seven peremptory challenges to which each of the accused were entitled they could get a jury composed wholly or mainly of workers from the local steel mills. This produced a series of acquittals until the Lord Chancellor, Lord Hailsham, directed that jurors’ occupations should be deleted from the printed lists. But, while the Crown’s right to stand jurors by was rarely arbitrarily exercised, the defence went on using peremptory challenges to get rid of jurors wearing suits, jurors carrying the Daily Telegraph or the Guardian, young jurors, elderly jurors, female jurors – whatever happened to bump up against some idiosyncratic notion entertained by the accused or their lawyers – until that too was first reduced and then stopped by legislation. The result is that, names apart, you genuinely and designedly do not know now who is sitting on a jury. And since the implementation of the Auld Report almost all the exclusions from jury service (other than for convicted criminals) have gone. Judges, police officers, prison officers, lawyers who defend, lawyers who prosecute can all sit as jurors.

In consequence the criminal division of the Court of Appeal (CACD) is currently considering whether it should quash a verdict of guilty brought in by a jury which included two police officers and a prosecuting lawyer. It has already had to consider what, if anything, can be done about the near certainty that one or more jurors will go home after the first day of a trial and google the defendant and anything or anyone else whose name has come up. Research carried out by Cheryl Thomas of UCL has suggested that something approaching one juror in eight does this. The CACD has made it clear that if any really prejudicial material finds its way by this means into the jury room, it is likely to vitiate the verdict. But the equally important principle of the sanctity of the jury room means that it is only by chance that the introduction of such material is likely to be revealed.

Nevertheless, judges do from time to time take what precautions they can with a jury. In 1969, after the Kray twins had been convicted, with massive publicity, of murdering two rival gangsters, George Cornell and Jack ‘the Hat’ McVitie, they had to stand trial again for murdering Frank Mitchell, the Mad Axeman. The trial judge, Mr Justice Lawton, agreed that the adverse publicity was not going to do much for a fair trial this time round, and to everyone’s surprise allowed the defence to question the prospective jurors about what they knew of the Krays. Within half an hour they had a jury of 12 citizens who apparently never read a newspaper or watched the television news.

For the rest, however, the courts have to rely on the jurors’ oath to return a true verdict according to the evidence, reminding them why it would be thoroughly unfair to both sides to bring into account factual material they have obtained from an outside source without subjecting it to open scrutiny. This gives force, in turn, to the question why the courts should not rely similarly on the judicial oath to answer, or at least to begin to answer, many concerns about apparent judicial partiality.

The European Convention on Human Rights, like the EU Charter, guarantees the right of every litigant or defendant to an independent and impartial tribunal. The two adjectives, and the difference between them, matter. Independence is a state of being. It signifies the separation of the judicial power from that of both the legislature, whose enactments it is for the courts alone to construe and apply, and the executive, whose acts the courts will respect so long as they stay within the law. It is independence which is undermined by fear – fear of official displeasure, for instance – and by favour, for example promotion, or the hope of it, for pleasing the executive. Impartiality, by contrast, is a state of mind, and it is the other pair of vices that the judicial oath abjures – affection and ill-will – which jeopardise it. It is to this above all that recusal is directed, whether the affection or ill-will is real or, as is most often alleged, only apparent – for nobody doubts that justice must be seen to be done as well as actually done.

But the question remains: why should the judicial oath not be regarded, save in the sharpest and clearest cases, as an answer to the appearance of partiality? Lord Hailsham, looking back in his autobiography, recalled that within his lifetime half the Queen’s Bench had consisted of former MPs, and regretted that convention had stopped him appointing more by the time he himself mounted the Woolsack. Impartiality,’ he wrote, ‘does not consist in having no controversial opinions or even prejudices … Impartiality consists in the capacity to be aware of one’s subjective opinions … to weigh evidence and argument and to withhold concluded judgment until the case is over.’

Whatever the tests for recusal, who decides whether they are met, and how do they go about deciding?

Judges as a rule get little notice of the cases they are going to hear, and when they do see the list of cases they will be taking the following week (or sometimes the following day) it is not always apparent from the name of the case who is going to be involved or what the case is going to be about. The result is that the parties, if they are lucky, may get notice of a possible conflict of interest a week or so before the trial; if unlucky, a day or so before; and possibly not until the outset of the hearing.

What the public will not know is that, without waiting for objections, judges from time to time ask to be replaced on cases in which they recognise that they have a possibly conflicting interest. In these instances the problem is anticipated and resolved administratively without a fuss, though a judge cannot do this for insufficient reasons. The difficulties occur in cases in which, while recognising that there is a connection which might give rise to an objection, the judge him or herself does not consider it sufficient to require recusal. At the same time, by disclosing it and inviting submissions about it the judge has implicitly recognised that others might take a different view. And that is where the problems start.

There are inevitably cases in which one party’s lawyers may find such a disclosure a welcome chance to get rid of a judge they would prefer not to have trying their case. But the temptation to take the line of least resistance and abandon the case is one that has to be resisted. It is a principle of prime importance that no party may choose its tribunal, whether by insisting on a particular judge or by objecting to one without sufficient cause. And there is no reason why the other party should lose its place in the trial list and probably incur a great deal of extra expense simply to accommodate an opportunistic objection by its antagonist. So no judge should stand down simply because one side has raised, or has taken advantage of the judge’s disclosure of, a possible disqualifying interest. Someone has to decide whether the objection taken on the basis of it is well founded. But who? And how?

In almost all cases the first practical step is to invite the judge to stand down voluntarily. The bar of England and Wales is good at saying awkward things without giving offence, and there is no reason why an application for recusal on grounds of ostensible bias should not be made directly to the judge concerned. It may be less easy to put a case of actual bias directly to that judge. But it need be no secret that judges do from time to time ask not to be put on an upcoming case because they have firm personal views or allegiances which may make it hard for them to be dispassionate about the case. That seems to me to be exactly what the judicial oath requires. The difficulty arises where one of the parties believes the judge holds committed views and the judge does not agree.

The first thing that needs to be stressed is that it is professionally improper for any lawyer to advance such a contention without sound evidence to support it. The second is that not every strongly held opinion disqualifies a judge. A judge is entitled to be prejudiced against dishonesty or violence. But if there is a foundation for believing a judge to have idiosyncratic views which will jeopardise the fairness of the coming trial, it is counsel’s duty to raise the matter, and if it is raised, it is undesirable that the judge him or herself should decide it.

The problem that arises at such a point is acute because it is not merely logistical: it affects the substance of justice. The judge has either to take on the invidious role of sitting as judge in his own cause – in effect having to analyse his own motivation – or to stand the application over for determination by another court, with serious consequences for one or both of the parties who were ready to have their case tried. What is more, if he does the first of these, decides that he is neither actually nor apparently biased and goes ahead with the hearing, an eventual decision by a higher court that he was mistaken will bring the entire edifice of judgment down and, possibly years down the line, send the parties back to square one.

All of this points towards the desirability of any serious challenge being decided swiftly by another judge or court. I say any serious challenge because in the post-Pinochet era the courts of this country have had to deal with a good many opportunistic or plainly ill-founded applications for recusal which it would be a waste of judicial resources to involve another court in. But where the judge whose status is seriously challenged not only does not accept that the challenge is well-founded but is anxious not to let it throw a last-minute spanner into the machinery of justice, there needs to be some system for deciding the issue swiftly, so that a sound challenge can result in the immediate replacement of the judge and an unsound one be put out of the way. By an unsound challenge I mean not only one which is intrinsically fanciful but one which the party making it has kept up its sleeve until it seemed useful. In the latter kind of case, it will be unusual for the court not to hold that the objection (unless it relates to an automatic disqualification) has been waived by delay or acquiescence.

In a court of three or five judges, this can often be accomplished by letting the other members of the court advise the judge on the challenge. It also, and importantly, enables the judge to state his or her own position, which may not have been fairly or accurately represented in the objection. In a jury trial, there is a judge on hand to decide the issue, and an appeal court if things nevertheless go wrong. But what of the judge who is about to sit alone to try a major piece of litigation and is confronted with an objection that could not arise until it was known which judge’s list the case stood in?

There may be a need for some kind of fire brigade protocol in such a situation. If humanly possible, the objection ought to be renewed before a different judge or court and be determined in time for the trial to go ahead with the same judge if recusal is not required or with another judge if it is. In deciding it, the second judge or court may need to consider whether the objector has forfeited the objection by even minor delay in advancing it, since its impact on the administration of justice may well be serious. But the important thing is that the system should not compound one paradox – a judge who is unbiased but might reasonably be thought not to be – with a further paradox: a judge who, in order to decide whether he will be sitting as judge in his own cause, has to sit as judge in his own cause.

The easy way out of a recusal application, no doubt, is to accede to it every time in the interests of a quiet life. But to do this is not only to enable one side to go some way towards handpicking its court: it is to sacrifice one of the pillars of judicial impartiality by implicitly accepting that one qualified judge is not as good as another. This aspect of the problem was recently highlighted in a much publicised case. The counselling service Relate, which subscribes to a code of ethics forbidding discrimination on grounds of sexual orientation, employed a counsellor who believed that as a Christian he ought not, in the course of counselling same-sex couples, to endorse physical activity which the Bible taught was sinful. The conflict became intractable and he was dismissed. An employment tribunal, upheld in due course by the Employment Appeal Tribunal, held the dismissal to have been fair and not unlawfully discriminatory. He then applied to the Court of Appeal for permission to appeal further. So far so unexceptional.

Then, while the application was pending, a witness statement prepared by or on behalf of a former archbishop of Canterbury, Lord Carey, was either released to the press or found its way into their hands. The statement took issue with the view that manifestation of Christian belief could amount to discrimination and expressed concern at two other recent cases which in Lord Carey’s view were insensitive to Christianity. These were opinions to which Lord Carey was fully entitled, though their admissibility as evidence was not obvious. But the thrust of the statement was to support an application on the counsellor’s behalf ‘for his case to be heard before the lord chief justice and a specially constituted Court of Appeal of five lords justices who have a proven sensitivity to religious issues’ and for the setting up of ‘a specialist panel of judges designated to hear cases engaging religious rights’. The judges in the cases to which he had objected in his statement, Lord Carey wrote, ‘should recuse themselves from further adjudication on such matters as they have made clear their lack of knowledge about the Christian faith’. I will not set out the reportage which this document prompted in sections of the media, but it included a headline without quotation marks or attribution: ‘Christians are being crucified by our judges.’

It is central to my theme that what a judge privately thinks or believes neither qualifies nor disqualifies him in the ordinary way for deciding sensitive and controversial cases. But the law of unintended consequences is never far away, and the judge of the Court of Appeal before whom the application happened to come was, according to the press, a communicant Anglican. This, among other things, is what he said:

Lord Carey’s observations are misplaced. The judges have never, so far as I know, sought to equate the condemnation by some Christians of homosexuality on religious grounds with homophobia, or to regard that position as disreputable, nor have they likened Christians to bigots. They administer the law in accordance with the judicial oath, without fear or favour, affection or ill-will.

The judge went on to consider Lord Carey’s view that there was a want of judicial understanding of or sensitivity to religious belief. He wrote:

These concerns are formulated at such a level of generality that it is hard to know precisely what Lord Carey has in mind. Broadly, however, the argument must be that the courts ought to be more sympathetic to the substance of the Christian beliefs referred to than appears to be the case and should be readier than they are to uphold and defend them.

In a free constitution such as ours there is an important distinction to be drawn between the law’s protection of the right to hold and express a belief and the law’s protection of that belief’s substance or content. The common law and ECHR Article 9 offer vigorous protection of the Christian’s right and every other person’s right to hold and express his or her beliefs, and so they should. By contrast, they do not, and should not, offer any protection whatever of the substance or content of those beliefs on the ground only that they are based on religious precepts. These are twin conditions of a free society.

As I have shown, Lord Carey’s statement also contains a plea for a special court. I am sorry that he finds it possible to suggest a procedure that would, in my judgment, be deeply inimical to the public interest.