Fiddes v Channel Four Television Corporation & Ors

These are the court’s reasons, to which all members have contributed, for a decision given on an application for permission to appeal which we heard on Thursday 10 June. The application was brought on for hearing as a matter of urgency, with the appeal to follow if permission was granted, by direction of Elias LJ. It was an interlocutory appeal, and the urgency arose from the fact that the trial of the libel action to which it relates was set to begin on Monday 14 June, that is to say four days from the hearing of the appeal (although the hearing date was put back a week and the case has now settled). Having heard argument on the morning of the 10 June, we concluded that we should grant permission to appeal, but dismiss the appeal. Because of the urgency of the matter, we gave our decision straight away, and these are our reasons.

The order in issue was Tugendhat J’s direction on 28 May that the trial of the action be heard by judge alone. It was made, unusually, on the defendants’ late application following the making of a consent order for trial by judge and jury. But no suggestion is made that the matter was res judicata. The question is whether Tugendhat J was entitled to make the order on the merits.

The proceedings started in December 2008, when the claimant, Matthew Fiddes, a martial arts expert, who owns a chain of martial arts schools in the West Country, issued a claim form alleging that he had been libelled in a television programme, transmitted on 27 November 2008 as part of the “Cutting Edge” series on Channel Four. The programme, entitled “The Jacksons are Coming”, had been made for Channel Four Television Corporation, the first defendant, (“Channel Four”), by the second defendant, Studio Lambert Limited (“Studio Lambert”), who had employed the third defendant, Jane Preston, as the producer, director and narrator of the programme.

The whole of the programme is said by Mr Fiddes to be libellous. In very summary terms, it is a documentary about the proposed move to Devon of members of the Jackson family, the famous musicians. As the defendants accept in their Defence, the programme suggests that Mr Fiddes betrayed the trust of the Jackson family, and “hypocritically deceived them by using his position as an apparent friend by seeking to exploit their fame for his own personal benefit, and was a manipulative and dishonest individual” (although Mr Fiddes puts the implication rather more strongly). The Defence raises defences of justification and of fair comment; in his Reply, Mr Fiddes not only comprehensively seeks to rebut those defences, but alleges malice. In particular, it is contended that, through insidious narration and manipulation of film footage, Ms Preston (and Studio Lambert, through Stephen Lambert) suggested that Mr Fiddes had deliberately leaked stories about the Jacksons’ trip to Devon, although he knew that they wished to avoid publicity.

As Tugendhat J said, “each side is setting out to prove about the other allegations of the greatest gravity”. The time estimate (for what it is worth in a heavily contested libel action) is 20 days, although Mr Thwaites QC, counsel for Mr Fiddes, suggested that that was on the generous side, at least if the trial was conducted with a jury, as Mr Fiddes would then be curtailing his case compared with what it would be if the trial was by judge alone. The problems for the Judge, and indeed for us, were made no easier by the fact that, among other things, the parties were still locked in combat about the contents of the trial bundles less than five days before the trial is due to commence.

When the Judge gave directions on 7 October 2009, he ordered, effectively on the basis of both parties agreeing, that the action be tried with a jury. The defendants had a change of heart, and issued an application on 12 May for an order that the trial be by Judge alone. That application came on for hearing on 27 May, and, after about six hours of argument, having said that his mind had wavered during the course of the hearing, and that he had “not found this case an easy one to decide”, the Judge ordered that the hearing be by Judge alone. It is against that decision that Mr Fiddes seeks to appeal.

We can take the relevant basic principles from the opening paragraph of the judgment of Lord Bingham of Cornhill LCJ in Aitken v Preston [1997] EMLR 415, 418, as it is precisely in point for present purposes:

“The decision as to mode of trial is governed by section 69 of the Supreme Court Act 1981 which so far as relevant reads:

‘(1) Where, on the application of any party to an action to be tried in the Queen's Bench Division, the court is satisfied that there is in issue --

....

(b)

a claim in respect of libel....

the action shall be tried with a jury, unless the court is of opinion that the trial requires any prolonged examination of documents .... which cannot conveniently be made with a jury.

....

(3)

An action to be tried in the Queen's Bench Division which does not by virtue of subsection (1) fall to be tried with a jury shall be tried without a jury unless the court in its discretion orders it to be tried with a jury.’

The judge correctly identified the three issues which he had to decide as follows:

(i)

Will there be a prolonged examination of documents?

(ii)

If so, can it conveniently be made with a jury?

(iii)

If not, should the court nonetheless exercise its discretion to order trial with a jury?

The judge answered the first question in the affirmative and the second in the negative. On the third, having regard to a number of considerations, he decided that he should exercise his discretion in favour of trial by judge alone.

All three of these conclusions are attacked by the defendants, and plainly if the judge was wrong in his answer to either of the first two questions then the defendants would indeed be entitled as of right to trial by jury. If, however, he was right in his answer to the first two questions, then, as is well settled and not disputed, this court may properly interfere with his exercise of discretion only if he was plainly wrong.”

As May LJ said in Viscount de Lisle v Times Newspapers Ltd [1988] 1 WLR 49, 57H, the three questions which a Judge has to decide under section 69 “requires a value judgment, based on what he is told by counsel, and his experience at the Bar and on the Bench.”

Jury trial in defamation cases is a constitutional right, whose importance was well expressed by Nourse LJ in these terms: “whether someone’s reputation has or has not been falsely discredited ought to be tried by other ordinary men and women and, as Lord Camden said, it is the jury who are the people of England” in Goldsmith v Pressdram Ltd [1988] 1 WLR 64, 74. As a result of section 69, it is no longer an absolute right, but its constitutional significance is perhaps emphasised by the fact that there is a right to a jury trial unless both the first and second section 69 questions are satisfied, and the fact that, even in a case where the court is satisfied that those two questions are satisfied, there will still be a jury trial unless the judge, in his reasonable discretion, otherwise decides.

Although the first two section 69 questions do not involve an issue of discretion, an appellate court should be slow before it interferes with a first instance assessment of such a nature. It is significant that Lord Bingham in Aitken [1997] EMLR 415, 426 referred to the judge as being “fully entitled to form the view that the trial would require prolonged examination of documents and that that examination could not conveniently be made with a jury” (emphasis added). The point is even clearer in the next sentence, where Lord Bingham said that it was “right that this court should respect his judgment on these questions, given his close familiarity with the case and his record as an immensely experienced and respected judge”.

In this case, Tugendhat J had the benefit of having case managed the case for nearly a year: the parties had been before him on contested interlocutory matters, requiring him to give reasoned judgments on 3 July 2009, 6 and 7 October 2009 and 28 January 2010. Furthermore, he was due to be the trial judge. Thus, like the Judge in Aitken [1997] EMLR 415, he plainly had “close familiarity” with the case; additionally, bearing in mind his “experience at the Bar and on the Bench” in the defamation field, he is certainly relevantly as well as “immensely experienced and respected” in this very field.

Nonetheless, this does not mean that this court should rubber stamp his decision. That would be fundamentally inconsistent with the existence of a right to appeal his decision (subject of course to permission being given for an appeal). Given that it is Mr Thwaites’s contention on behalf of Mr Fiddes that, although the Judge asked himself the right questions, he went wrong in principle and in details in each of his answers, it is right for us to examine his reasons for concluding that the three section 69 questions should be answered in such a way as to negative a jury trial.

Before turning to the specific arguments, we should mention one other point. In paragraph 23 of his judgment, the Judge referred to the “vast costs in this case”, a fair description on our understanding of the figures. The approach to the litigation is exemplified by the fact that the parties, or at least one of them, have, or has, seen fit to obtain full transcripts of the six-hour hearing of the instant application before the Judge. It is only in the most exceptional circumstances that it can ever be justifiable even to consider obtaining a transcript of an interlocutory hearing for the purposes of an appeal – e.g. to show that an argument which is said to have been taken was not raised, to establish that some concession was or was not made, or to show some misconduct or misunderstanding on the part of somebody at the hearing. Even in such cases, one would hope that the relevant facts could be agreed, or that only a small part of the hearing need be transcribed.

On this appeal, the transcript of the hearing below was referred to only to support the contention that the Judge took a view on certain points which did not feature in his full reasoned judgment. This was inappropriate. During a hearing, a judge will often advance a view to counsel. It may be the judge’s view at the time it is advanced, but the purpose of oral argument is to enable the judge to refine and, where appropriate, to change his preliminary views. Often, however, the view may embody a point which the judge is not currently inclined to accept, but which, in his opinion at the time, should at least be discussed. It would be highly regrettable, both in the interests of justice and in terms of fairness to judges, if a point raised in argument in order to develop the judge’s thinking, to advance the debate and, as is often the case, to assist the advocates, should be cited on appeal, when the point formed no part of the eventual reasoning and was not mentioned in the judgment.

In this case, when considering the three section 69 questions, the Judge took the applicable principles from the judgment of Lord Bingham LCJ in Aitken [1997] EMLR 415, 421-422, where he said this:

“(i)

The basic criterion, viz that the trial requires a prolonged examination of documents, must be strictly satisfied, and it is not enough merely to show that the trial will be long and complicated (Rothermere v Times Newspapers Ltd [1973] 1 WLR 448). However, the word ‘examination’ has a wide connotation, is not limited to the documents which contain the actual evidence in the case and includes, for example, documents which are likely to be introduced in cross-examination (Goldsmith v Pressdram Ltd [1988] 1 WLR 64).

(ii)

‘Conveniently’ means without substantial difficulty in comparison with carrying out the same process with a judge alone. This may involve consideration of several factors, for example:

(a)

the additional length of a jury trial as compared with a trial by judge alone;

(b)

the additional cost of a jury trial taking into account not only the length of the trial but also the cost of, for example, additional copies of documents;

(c)

any practical difficulties which a trial by jury would entail, such as the handling of particularly bulky or inconvenient files, the need to examine documents alongside each other, and the degree of minute scrutiny of individual documents which will be required;

(d)

any special difficulties or complexities in the documents themselves (Beta Construction Ltd v Channel Four Television Co Ltd [1990] 1 WLR 1042 especially per Stuart Smith LJ at page 1047C-D and per Neill LJ at page 1055H, referred to and applied in the recent case of Taylor v Anderton [1995] 1 WLR 447).

(iii)

The ultimate exercise of discretion will in each case depend substantially on the circumstances of each individual case, and it would be idle to attempt to enumerate all the factors which might arise.

There are, however, four factors which have been identified in the earlier cases, which have some general application and which are presently relevant, as the judge recognised:

(1)

The emphasis now is against trial by juries, and this should be taken into account by the court when exercising its discretion (Goldsmith v Pressdram (supra) at page 68 per Lawton LJ with whom Slade LJ expressly agreed). This conclusion is based on section 69(3), which was a new section appearing for the first time in the 1981 Act to replace section 6(1) of the Administration of Justice (Miscellaneous Provisions) Act 1933, the provision in force at the date when Rothermere v Times Newspapers was decided.

(2)

An important consideration in favour of a jury arises where, as here, the case involves prominent figures in public life and questions of great national interest (Rothermere v Times (supra)).

(3)

The fact that the case involves issues of credibility, and that a party's honour and integrity are under attack is a factor which should properly be taken into account but is not an overriding factor in favour of trial by jury (Goldsmith v Pressdram (supra) at page 71H per Lawton LJ).

(4)

The advantage of a reasoned judgment is a factor properly to be taken into account (Beta Construction v Channel Four Television (supra)).

It was suggested on behalf of Mr Fiddes that these principles were not entirely consistent with earlier authorities, but we do not accept that. Inevitably, there are some dicta in other judgments which put some of these points slightly differently, but there is no inconsistency between Lord Bingham’s illuminating summary of the applicable principles when approaching the section 69 questions and other authoritative observations from this court. Lord Bingham went on to point out the value of a reasoned judgment (which would not be available in a jury trial), particularly to the successful party.

Having said that, there are six points we think it right to make about Lord Bingham’s analysis of the applicable principles, in the light of the arguments advanced to us.

First, we would like to emphasise the need for caution when invoking the additional length, and (even more) the additional cost, of a jury trial as factors to be taken into account on the second, convenience, section 69 question. Jury trial will almost always take longer, and cost more, than trial by judge alone. The extra time taken, and the extra costs involved, in a jury trial may often be a useful sort of quantitative cross-check of what might otherwise be a purely qualitative assessment of the extra inconvenience of a jury trial (as was done in Beta Construction [1991] 1 WLR 1042). However, it would be dangerous if those two factors were given much independent weight, as it would risk undermining the important right to a jury trial which section 69(2) gives – to defendants as well as to claimants – in libel actions.

Secondly, the number of documents is not the issue when it comes to the first and second section 69 questions. As Slade LJ said in a passage cited by the Judge, “[t]here may be many cases where numerous documents will be required to be looked at, but no substantial practical difficulties are likely to arise in their examination being made with a jury”, and, by contrast, there can be cases where “relatively few documents will require examination, but nevertheless long and minute examination of them is likely to be required”.

Thirdly, it is important to appreciate that the inconvenience to be considered in the second section 69 question is that arising from “the prolonged examination of documents”: the court should not, at that stage, look at any other inconvenience which may arise as a result of a jury trial, although it could well be relevant when considering the third question. Fourthly, the fact that one party is a public figure may often be a reason for favouring a jury trial, but that does not mean that the fact that neither party is a public figure is a reason against a jury trial.

Fifthly, it is fair to say that the constitutional importance of the right to trial by jury was not mentioned in Aitken [1997] EMLR 415, but that aspect was clearly in the Judge’s mind in this case, as he cited Nourse LJ’s observation in Goldsmith [1988] 1 WLR 65, 74, referred to above. That is undoubtedly a factor which has to be borne in mind on the issue of convenience as well as of discretion.

Sixthly, as the Judge pointed out in this case, the fact that juries in criminal trials (especially those trials involving allegations of complex financial fraud and the like) sometimes have to consider complex documentation does not really bear on the three section 69 questions. It may well be that, in some such criminal trials, the section 69 questions would result in the conclusion that the trial should be by judge alone, but the questions do not arise in the criminal field even in relation to such cases: there is an absolute right to a jury trial, save in circumstances which are very different from those covered by section 69.

Turning to the facts of this case, as the Judge explained, there are five principal issues. First, there is the issue of the meaning of the words and pictures used: this will depend on watching the programme. Secondly and thirdly, there are the issues of justification and fair comment. On the defendants’ case, this depends on incidents (i) before the genesis of the programme in March 2008, (ii) just after the Jacksons arrived in this country in May 2008, at Heathrow Airport, on the road to Devon, and in Devon, (iii) during the next two weeks or so, while the Jacksons were in this country, and (iv) during the four months after the Jacksons returned to the United States, in all of which Mr Fiddes is said by the defendants to have publicly exploited his connections with the Jackson family. Mr Fiddes relies on many incidents mostly after January 2007, but some as early as 1997. Fourthly, there is the malice issue. At least before a jury, this would involve consideration of (i) some ten incidents, all of which are based on the proposition that examination of the original filming, the so-called rushes, establishes that the footage included in the programme was deliberately manipulated by Ms Preston so as to create a false and misleading impression, adverse to Mr Fiddes’s reputation and honesty, and (ii) a number of documents and events which are said to show that Studio Lambert knew the programme was false and misleading. The fifth issue is republication of the alleged libel, which is, as the Judge said, “relatively short”.

As the Judge went on to explain, pursuant to previous orders which he had made, he had the benefit of witness statements and proposed bundles of documents (as well as the identification of the relevant video footage), which, having been prepared in accordance with CPR 32.4 and paragraph 7.10.4(2) of the Queen’s Bench Guide, enabled him better to assess “what and how many documents are likely to be referred to at the trial, and how long that is likely to take.”

He then turned to the first section 69 question, namely whether the issues raised in this case would require prolonged examination of documents. He said that (i) Miss Page QC, for the defendants, had put forward her contentions on “a worst case basis”, which he would try and ensure was avoided by exercising his case management powers at trial, (ii) “the central issue in this case is who is believed”, which was likely to turn on “relatively few of the total number of issues that are canvassed in the pleadings and witness statements”, and (iii) “only ... a relatively small number of issues ... will require a prolonged examination of at least some documents and footage.”

However, he decided that there would be a sufficient number of documents (including film footage on video) which would require sufficiently detailed examination at trial to satisfy the first section 69 question. He gave examples, which he emphasised, and which Mr Thwaites very fairly accepted, were only examples, of what he had in mind in this connection.

Nonetheless, Mr Thwaites submitted that the Judge exaggerated the extent to which the documents he identified would involve prolonged or difficult examination, namely the footage showing the Jacksons’ arrival and interview at Heathrow, the footage showing an incident in the garden of the house they stayed at in Devon, and Ms Preston’s amended electronic diary (as well as some email chains, which he referred to later in his judgment as often being “in quite small print and ... quite confusing”). He argued that the footage which had to be looked at was really pretty short and was anyway “merely cross-examination material”, and that “watching television is something which jurors are particularly well-suited to do”. He further took us to the electronic diary to establish that it was not very long or complex, and that there was only one set of amendments which were easy to follow. He also relied on the Judge’s powers of case management, which, he said, made the pre-CPR cases unreliable precedents.

We are not convinced that the footage will only be relevant for cross-examination: much of it may be used to establish primary facts. In any event, the fact that much of the video material will be used for cross-examination does not advance Mr Fiddes’s case on the issue. On any view, the footage to be relied on by Mr Fiddes is highly relevant, and, together with any footage relied on by the defendants, it will very likely amount, in extent, to substantially more than Mr Thwaites suggested. The passages he will seek to rely on may be short, but, like provisions in documents, the defendants will require those passages to be assessed in their context, and that will require substantially more footage to be viewed and assessed.

In her submission to this court, Miss Page took us briefly through one of the ten or so filmed incidents relied on, which related to the filming of the arrival of the Jacksons at Heathrow, and of an interview recorded by Mr Fiddes with a television news channel. She pointed out that the footage would have to be looked at on a large number of different occasions at trial - during at least two speeches, as well as during examination in chief and cross-examination of a number of different witnesses who were likely to be called in connection with this incident. Not only that, she said, but the viewing would have to be carried out from a number of perspectives: it would have to encompass not merely what the filming showed in general terms, but (i) what it showed in detail – who was present and who was not present, and for what purpose they were present, (ii) how the filming was carried out – e.g. whether it was effected to present a misleading impression, and whether it was set up professionally in advance or done on an ad hoc basis, and (iii) what would or should have been appreciated when viewed by Stephen Lambert - in order to consider the case against Studio Lambert.

As to the diary, Miss Page adopted the Judge’s expressed view that its examination at trial would be prolonged. Even if it could not, on its own, satisfy the first section 69 question, when taken together with the other material, she contended that the Judge was entitled, indeed bound, to take it into account. Miss Page also said that the Judge rightly referred to the emails, describing Mr Fiddes as having disclosed a “substantial” and “rich mine of email traffic” filling a substantial part of three lever arch files. She added that, having examined these emails, there could be difficulties in following chains of emails to various different people, especially where the chains had different branches relating to different groups.

It is clear from his judgment that the Judge was well aware of the facts that (i) the central issue was who would be believed, (ii) the defendants’ representatives would (within the bounds of propriety) suggest that the examination of documents at trial would be as long and complex as possible, (iii) a prolonged examination of documents would be required for a relatively small number of issues, and (iv) that he would, as the trial judge, have case management powers which would enable him to control the hearing from getting out of hand.

Despite these points, and despite his scepticism about the full extent of the defendants’ case on this issue, he reached the conclusion that the first section 69 question was satisfied. Of course, it may turn out that the Judge took too pessimistic a view of the amount of minute examination of footage and documents which would be required, but he took great care to reach an informed and balanced view, which cannot be said to have been based on any misunderstanding of fact or law. The Judge’s careful and considered analysis of the issue, supported by the arguments advanced by Miss Page, satisfy us that there are no grounds for interfering with his conclusion on the first section 69 issue.

On the second issue, that of convenience, the Judge made the obvious, but nonetheless significant, point that, in the light of the need for prolonged examination of footage, the diary and emails, it will take up substantially more time, and therefore substantially more cost, if the trial is with a jury. It takes longer to go through documents with a jury, misunderstandings are much less easy to resolve, every aspect of reading and viewing have to be carried out in court time. Although he said that there were “no special difficulties or complexities in the footage or documents themselves”, he concluded that he had “little doubt that it would be substantially more difficult (and less convenient) to [try the case] with a jury than with a judge alone.”

Mr Thwaites made much of the point that the only inconvenience which can be relied on for the purpose of the second section 69 issue is inconvenience attributable to the need to conduct a prolonged examination of documents. We agree, but the Judge’s reasoning did not fall foul of this principle.

Mr Thwaites also made the point that the Judge should not have given so much weight to the increase in time and cost which a jury would involve. We have already made the point that great care must be taken not to give much independent weight to such increased time and cost, as it would otherwise risk undermining the right to a jury trial: a trial with a jury will almost always cost more money and take longer than the same trial before a judge alone. However, where, as here, the judge relies on the extra time and cost as some sort of quantitative guide to the extra inconvenience, there can, in our view, be no objection to taking them into account.

Mr Thwaites further contended that the Judge exaggerated the extent to which the absence of a jury would enable time to be saved, but we consider that there is nothing in that. He suggested that the convenience which section 69 had in mind was that of the jury. He sought to cite extracts from Hansard, but this is emphatically not a case where that was permissible. The Judge also made the point that expense was of particular concern in this case in the light of the large sums already expended on costs. We see nothing wrong with that.

It was also suggested that the Judge was over-influenced by the fact that the first instance decision in Aitken [1997] EMLR 415 to dispense with a jury was upheld by this court. We can see no support for such a suggestion in the judgment or in the decision of Tugendhat J.

There was a suggestion that Mr Thwaites offered the Judge an assurance that Mr Fiddes’s case would be curtailed in terms of allegations if the trial was held with a jury. It did not appear that any firm commitment to that effect was given to the Judge, so the point has no force in fact. However, we would add this. It seems to us that it would be inappropriate for one party to be able to bargain with the judge for a jury trial on the basis of offering to simplify or curtail his case if, but only if, the trial is with a jury. Having said that, the court can with counsel’s assistance, and in an appropriate case should, gauge whether, if the case proceeds before a jury, the case presented by either or both of the parties may be simpler, shorter and less document-ridden than it would be if the trial was conducted before a judge sitting alone.

In our opinion, the Judge’s decision that the second section 69 question was satisfied was unassailable.

As for the third section 69 question, Mr Thwaites submitted that Tugendhat J based his conclusion that the trial should be by judge alone in part on the proposition that trial by jury “would have a ‘chilling effect’ on the defendants’ freedom of expression”. In his judgment, the Judge made reference to the point that the level of costs in libel proceedings could in some cases have a possible chilling effect on freedom of speech, but he accepted that the costs of this case, even if lost by the defendants after a jury trial, would not put Channel Four out of business, and he also accepted that a very large amount of costs had already been incurred. In the end, his conclusion was that the argument in this case had “some, although not the greatest amount of, weight”.

The Judge also referred to two other factors. First, the desirability of a reasoned judgment on “what is and is not acceptable editorial practice in television broadcasts which are presented as factual”. Secondly, he gave some weight to the desirability of a reasoned judgment, which he thought relevant in the light of the Human Rights Act 1998, and article 6 of the Convention.

In our view, these were all perfectly proper factors for the Judge to have taken into account in this case. He would have erred if he had treated a reasoned judgment as inherently preferable to a jury’s verdict, but he did not do that. It was suggested that Mr Fiddes’s article 8 rights supported his case for a jury trial, and that the Judge failed to take those rights into account. We do not understand that point: if anything, with the benefit of a reasoned judgment, it can be said that a successful claimant in a libel action is better off than he would have been with a jury trial.

The Judge was also criticised for not having referred to the fact that there were substantial allegations of dishonesty against three of the four parties. The Judge was plainly aware of this as he had referred to it when summarising the issues in the case earlier in his judgment. It is fanciful to think that he did not have the point in mind when he was considering the issue of discretion. This was an ex tempore judgment in a hard fought and difficult case management decision which required an immediate judgment (as the Judge appreciated that there would be an attempt to appeal by the party who lost). It would be both unreasonable and unfair to expect a judge, particularly in those circumstances, to repeat any point he had already made when summarising the facts and issues or when dealing with the first two section 69 questions, which bore on the third, residual, question of discretion. It is not as if he failed to refer to any points in favour of exercising his discretion to order a jury trial: thus, he did accept the submission “as far as it goes” that jury members were likely to be more conversant with watching television than a judge.

In conclusion, this was a full, careful and considered judgment of a judge, with very wide experience of defamation cases, and with detailed knowledge of the case and of the issues involved, who, after summarising the facts and issues in the case and directing himself correctly as to the applicable legal principles, considered the three section 69 questions by reference to a careful assessment of the way in which the trial was likely to proceed, and, while accepting that the ultimate decision was not easy, concluded that the trial should not proceed with a jury. We do not consider that he erred in law in reaching this conclusion.