Benham Ltd v Kythira Investments Ltd & Anor

This is the claimants’ appeal against the order made by His Honour Judge Geddes sitting as a judge of the High Court on 13 May 2003 dismissing with costs their claim against the respondents. The order was made following argument at the close of the claimants’ evidence on the fourth day of the hearing. In making it the judge was acceding to the respondents’ submission of no case to answer without first putting the respondents to their election. The case illustrates yet again the dangers of adopting such a course. Later in this judgment I shall have to examine some of the cases on this topic. I shall be considering too the question whether the court’s entitlement in certain circumstances to draw adverse inferences from a party’s failure to call evidence ought properly to bear upon the approach to be taken to submissions of no case to answer; no attempt seems hitherto to have been made to inter-relate these two lines of authority.

First, however, it is necessary to indicate the circumstances in which this claim arose and the background to the present appeal. I shall do so comparatively briefly, shortening and simplifying the facts wherever possible.

The appellant, Benham Limited, trading as Benham & Reeves, whom I shall call simply Benhams, are a well-known firm of estate agents based principally in Hampstead. Until February 1999 Benhams’ senior employee (and since 1994 a director) had been Jeremy Gee (“Mr Gee”). He was then summarily dismissed for gross misconduct (unrelated to the present case), on the ground that he took a secret profit.

The respondents are property companies under the control of the Tsentas brothers who acquire and develop properties for investment, mostly in north London. I shall refer to them simply as Kythira.

On a number of occasions from 1993 onwards, Kythira retained Benhams (acting through Mr Gee) as their agents with regard to the purchase of suitable investment properties, usually for a fee of 2% of the purchase price plus VAT. Sometimes Benhams, in addition, were retained for a similar fee as sole agents on the resale of such properties.

In May 1998 Hong Kong International Properties Limited acting through Mr Marcus Cooper (“the vendors”) were selling two contiguous properties in Branch Hill, London NW3, known respectively as Leavesden Cottage (“Leavesden”) and Spedan Cottages (“Spedan”) for a combined purchase price of £1,650,000. Both properties were purchased by Kythira. Contracts were exchanged on 22 May 1998, unconditionally in respect of Leavesden for an apportioned price of £450,000; conditionally by way of an option agreement in respect of Spedan (the freehold of which Kythira had meantime to acquire from Camden London Borough Council) for £1,200,000. The sale of Leavesden was completed on 5 June 1998, that property later being developed and resold for £1,200,000. Kythira exercised their option in respect of Spedan on 18 August 1998, the sale being completed on 18 September 1998. That property was later developed and eventually resold in September 2003 for £3,345,000.

It is Benhams’ claim in these proceedings that they, through Mr Gee, were acting as agents for Kythira in connection with these transactions and that they accordingly became entitled to their commission in respect of them. There being no agency agreement in writing, however, and the circumstances of Mr Gee’s departure from the firm making it impossible for them thereafter to invite his cooperation and call him, their case is largely dependent upon what is essentially circumstantial evidence.

Before turning to the evidence upon which Benhams most particularly rely, it is convenient first to set out the core paragraphs of the respective pleadings. Benhams’ Re-amended Particulars of Claim include the following:

“6.

In or about the spring of 1998, Jeremy Gee of the Claimant had dealing with the First Defendant and/or those connected with it in relation to a property known as Spedan Cottages and Leavesden Cottage Branch Hill London NW3 (‘the property’).

7.

Jeremy Gee was at all material times acting in his capacity as an employee of the Claimant and the First Defendant knew him to be an employee of the Claimant by virtue of its previous dealings with Mr Gee and the Claimant.

8.

The First Defendant subsequently retained the Claimant to act as its agent on the purchase of the property and as its sole agent on any subsequent resale of the property (‘the retainer’).

8A. To the best of the Claimant’s knowledge this retainer was agreed orally, no later than the first half of May 1998, between the Claimant’s agent Jeremy Gee and one or more of the Tsentas family who at all material times controlled and/or directed both the Defendants. it was, however, evidenced in writing by various documents relating to the purchase faxed to the Claimant by the First Defendant’s solicitors in 18 May 1998. No further particulars can be given as Mr Gee was dismissed from the Claimant’s employ for misconduct, but the Claimant will say the agreement was in consideration and/or in consequence of his introduction of the property to them and/or his instrumentality in the Defendants’ acquisition thereof.”

Kythira in their Re-amended Defence respond as follows:

“7

The Defendants make no admissions as to paragraphs 6 and 7 save that:

a)

It is denied that the First Defendant instructed Jeremy [Gee] as alleged or at all.

(b)

It is admitted that the First Defendant approached Jeremy Gee in May 1998 in relation to the sale of the freehold of Spedan Cottages and Leavesden Cottages as Jeremy Gee was believed by the First Defendant to have a business relationship with Marcus Cooper, the vendor, and the First Defendant thought Jeremy Gee might broker a reduction in the purchase price.

c)

It is admitted that Jeremy Gee was, at the material time, an employee of the Claimant.

d)

Jeremy Gee attended a meeting on 14^th May 1998 between the First Defendant and Marcus Cooper to negotiate a sale of the freehold of Leavesden Cottages and Spedan Cottages.

e)

At the meeting on 14^th May 1998 Marcus Cooper was unable to show good title for the freehold of Spedan Cottages and there were no further dealings between the First Defendant and Jeremy Gee in relation to the sale of the freehold.

8.

Paragraph 8 is denied. The property was introduced to the First Defendant by Adrian Levy of The Estate Office. The First Defendant did not retain the Claimant in respect of the property as alleged. The only retainer the first Defendant entered into with the Claimant was to market that part of the property known as Leavesden on a multiple agency basis, particulars whereof are pleaded at paragraph 15.2 of this Defence.”

What, then, at the close of their evidence, were the facts and circumstances upon which Benhams sought to rely to make good their claim that an agency agreement had indeed been made between the parties under which they were entitled to their commission? Principally, they were:

Benhams contend that the most obvious interference to be drawn from all these documents is that their firm, through Mr Gee, were retained by Kythira with regard to this property transaction, just as they had been retained in connection with various other properties in the past. Although Kythira plead in paragraph 7(e) of their Re-amended Defence that, after the meeting on 14 May 1998, they had no further dealings with Mr Gee in relation to this purchase, that seems to be belied both by Mr Gee’s scribbled notes and diary entries and by Kythira’s solicitor’s practice of copying to them a number of obviously relevant documents with regard to the transaction. Above all, of course, Kythira’s pleading is inconsistent with their own solicitor’s express assertion in the Additional Enquiries sent to the vendors on 20 May 1998:

“We … have also spoken to our client’s agents, Benham & Reeves …”

Benhams recognise, of course, that even with regard to certain of the evidence which they themselves put before the court, there are arguments available against them. For example, in The Estate Office’s Sales Information Sheet dated 18 May 1998 the space under the heading “Other Agents” is left blank (although the layout of the form tends to suggest that the only “Other Agents” intended for mention in that space are any co-agents for the vendors). Perhaps a stronger argument against the inference which Benhams submit should be drawn from these various facts and circumstances is that Mr Gee on their behalf seems never to have submitted an invoice claiming payment of commission in respect of either limb of the completion process, ie Leavesden in June 1998 and Spedan in September 1998. Could it be, however, that at some point during Kythira’s acquisition of these properties Mr Gee may have decided not to account to his employers for the commission earned on this transaction? Whatever may be the answer to this question and whatever uncertainties surround it, it is Benhams’ submission before us that the judge below could not properly accede to a submission of no case to answer; still less could he do so without putting Kythira to their election.

It is convenient at this stage, before coming to examine the judge’s reasoning which led him to uphold the respondents’ submission, to look (as indeed, at the outset of his judgment, the judge himself looked) at the recent authorities governing the approach to be taken to submissions of no case to answer. We were referred to four such cases, all of them decisions of this court. Since it appears that none has been reported (at any rate in the official or All England Law Reports), it may be helpful if I cite from each at somewhat greater length than would otherwise be necessary. I take them chronologically.

In Boyce -v- Wyatt Engineering [2001] EWCA Civ 692, a personal injury case brought against three defendants, the judge below allowed a submission of no case to answer made by all three defendants without putting any of them to their election. This court allowed the claimant’s appeal (brought solely against the second and third defendants) and in the result remitted the case for retrial before a different judge. Mance LJ gave the leading judgment in this court:

“4.

The course taken by the judge of deciding the case following the hearing of the claimant's evidence without putting the defendants to their election is one which calls, on any view, for considerable caution. I mention two particular considerations. First, where a defendant is put to his election, that is the end of the matter as regards evidence. The judge will not hear any further evidence which might give cause to reconsider findings made on the basis of the claimant's case alone. The case either fails or succeeds, even on appeal. But, where no such election is called for, the judge is required to make up his mind as to facts on the basis of one side's case, and then, if he is against the defendant, to hear further evidence and to retain and apply an open mind in relation to all the facts at the end of the trial. That is an inherently difficult exercise. Hence the difference in normal practice between criminal cases (where submissions of no case are common but are determined by a different test and by the judge rather than the jury) and civil cases (where the practice has been for the judge to put the defendant to his election).

5.

In this respect, despite the objectives of the new Civil Procedure Code and the broad powers of court management which it contains, there remains force, in my view, in the general observation made in this Court in Alexander -v- Rayson [1936] 1 KB 169 at 178 that it is not right that the judge of fact should be asked to express any opinion upon the evidence until the evidence is completed. There may be some cases, probably rare, in which nothing in the defendant's evidence could affect the view taken about the claimant's evidence or case, but this is not one of them, and care would be required in identifying them.

6.

Secondly, there is another consideration which is independent and general. If no election is extracted, then there is the risk, as here, that if the claim is dismissed, there may be a successful appeal against the judge's view of the merits, and the matter may then have to be remitted, quite likely to a different judge, for a complete retrial. That may waste far more money than might have been saved by hearing the defendant's evidence at the first trial.”

Potter LJ and Sir Martin Nourse agreed, the former adding at paragraph 36:

“If such breaches of statutory duty might procedurally have been established against either or both of the second and third defendants, it was wrong for the judge to come to any conclusion as to whether both or either were liable without hearing the defendants' evidence (if they proposed to call any), unless the evidence of the claimant himself disclosed that he was bound to fail, in particular by failing to establish that the defendants' breaches of duty were causative of the accident.”

Lloyd -v- John Lewis Partnership [2001] EWCA Civ 1529 was another personal injury case in which the judge below allowed the defendant’s submission of no case to answer without putting them to their election and again the claimant’s appeal succeeded, albeit on this occasion with an order that the matter be sent back to the court below for the trial to continue before the same judge. Sir Murray Stuart-Smith gave the leading judgment. He noted that the trial judge had been persuaded that the rule in Alexander -v- Rayson [1936] 1 KB 169 had been altered by the Civil Procedure Rules “and that as a general rule a judge was not required to put defence counsel to his election, although there might be exceptional circumstances in which he would do so”. Sir Murray continued:

“9.

Since the judge determined the point there has been a decision of this court on the point. It is called Boyce -v- Wyatt Engineering and Others . In paragraphs 4 to 6 of the judgment, Mance LJ (with which the other members of the court agreed) said that, on the contrary, the general rule was as stated in Alexander -v- Rayson. There were good reasons for that rule and it should be, as a general rule, adopted unless there are circumstances which indicate the contrary.

10.

It is unnecessary to refer to what Mance LJ said in any more detail, it can be read by those who wish to do so. I have no doubt that if the judge had had the benefit of that authority, rather than the authority to which he was referred, he would have come to the conclusion that he should put the defendants' counsel to his election. What he would have elected to do we do not know. But it seems to me that there is really nothing in this point bearing on the appeal because the judge made his decision. If he was right on the question of causation, then the appeal fails: if he was wrong on that point, then the appeal succeeds and there will have to be a new trial or continuation of this trial by Judge Cooke.”

The third in this series of cases is Bentley -v- Jones Harris & Co [2001] EWCA Civ 1724, an accountants negligence case, in which again the judge below acceded to a submission of no case to answer without putting the defendant to his election, albeit this time the claimant’s appeal was dismissed. I need cite only two short passages from the judgments, first this from Burton J’s leading judgment:

“61.

At the time of this trial it was, it seems, common ground between counsel for both parties and the judge that under the CPR there was no need for a party making a submission of no case to answer to be put to his election. This has now been clarified by this Court in Boyce -v- Wyatt Engineering and Others [2001] EWCA 692 per Mance LJ, so that certain limits are set on that proposition. But it is clear that on the basis of the learned judge's judgment he concluded that in this case, in the light of the evidence given by the claimant, nothing in the defendants' evidence could affect the view he had taken (see per Mance LJ at para 5 and the learned judge's judgment at p.15D-F). In any event, the question of further evidence from the defendants never arose because the learned judge found that there was no case to answer. We have to decide whether that conclusion was right.”

Burton J then recorded counsel’s agreement as to what the “no case to answer” test was, namely “whether the court comes to the conclusion that realistically there is no basis upon which a jury could, properly directed, find in favour of the claimant on the evidence that the claimant has adduced”, and noted the trial judge’s conclusion to have been “that there was no such case to put to himself, wearing his jury hat”.

Latham LJ added:

“75.

As Mance LJ has said in the case of Boyce, to which my Lord has already referred, it will only be in a rare case that the judge should be asked to determine the issues before him before all the evidence has been completed. However, it seems to me that, if a judge concludes at the end of a claimant's evidence, whether on the application of the defendant or of his own motion, that the claimant has no real prospect of success or, in other words, is bound to fail, on his assessment of the evidence before him at that stage, he is in my view entitled to give judgment for the defendant, in the same way as if there had been an application at an earlier stage in the proceedings for summary judgment under CPR Part 24.2. In that way he will be giving effect, in the circumstances of a trial, to the overriding objective and in particular to the need to contain within limits the expenditure of time and costs on the particular case before him.”

The fourth and final case in the series is Miller (t/a Waterloo Plant) -v- Margaret Cawley [2002] EWCA Civ 1100, a very different case from the others on the facts since the trial judge there had put the defendant to her election and in the result she had elected not to give evidence. The oddity was that with the agreement of both counsel the judge had thereupon applied the test of whether “realistically there is no prospect of the claimant proving his case upon the evidence that he has put before the court thus far”, his conclusion upon that being:

“Mr Miller was not the best of witnesses. … Nevertheless it would be a bold judge who would say that realistically there were no prospects of any jury being able to find in his favour, and in my view, I am unable to come to that conclusion.”

He therefore rejected the submission of no case to answer and concluded that the claimant had proved his case.

The defendant’s appeal in Miller succeeded; clearly the judge had adopted the wrong approach. Once again, Mance LJ gave the leading judgment in this court. His was, indeed, the only reasoned judgment: the other members of the court, Latham LJ and myself, simply agreed with it. Mance LJ explained the position where a defendant is put to election and elects to call no evidence in paragraphs 17 and 18 of his judgment:

“17.

Where a defendant is put to his or her election and elects to call no evidence, the position is quite different. As I said in Boyce at para. 4:

‘First, where a defendant is put to his election, that is the end of the matter as regards evidence. The judge will not hear any further evidence which might give cause to reconsider findings made on the basis of the claimant's case alone. The case either fails or succeeds, even on appeal.’

18.

The issue after an election is, in other words, not whether there was any real or reasonable prospect that the claimant's case might be made out or any case fit to go before a jury or judge of fact. It is the straightforward issue, arising in any trial after all the evidence has been called, whether or not the claimant has established his or her case by the evidence called on the balance of probabilities.”

Mance LJ then repeated in paragraph 20 that:

“… once a defendant has elected to call no evidence, … the only issue then becomes whether in the light of the evidence already adduced the claimant has made out his case on the balance of probabilities and that was the test (more favourable to the defendant) which the judge ought to have been invited to apply”.

As to whether or not the judge should put the defendant to his election before entertaining a submission of no case to answer, Mance LJ said this:

“12

The determination by a judge of fact of a submission of no case to answer without putting the defendant to any election to call no evidence has been likened to the determination of a pre-trial application under CPR Part 24.2 on the basis that the claimant has no real prospect of success: cf Bentley -v- Jones Harris & Co. [2001] EWCA Civ 1724, per Latham LJ at para. 75. The differences in context mean that the analogy may not be precise: for example, a judge pre-trial may make allowances for the possibility of development or amendment, whereas, by the close of a claimant's case, that case and the evidence supporting it will have been definitely identified. But it is clear that in some circumstances a submission of no case to answer at the close of a claimants' case can be appropriate and may, in the exercise of the judge's discretion, be entertained without the defendant being put to his or her election – cf both Bentley itself and Boyce -v- Wyatt Engineering [2001] EWCA Civ 692, per Potter LJ at para. 36 (last 31 words). Some flaw of fact or law may, for example, have emerged for the first time, of such a nature as to make it entirely obvious that the claimant's case must fail, and it may save significant costs if a determination is made at that stage.

13

However, as I said in Boyce, considerable caution is necessary before a judge entertains such a submission or undertakes such a determination, without requiring any election. The trial is now in progress, and although the test (no real prospect) differs from that applicable after hearing all possible evidence (balance of probability), caution is dictated for reasons along the lines indicated in paras. 4-6 in my judgment in Boyce . The submission interrupts the ordinary trial process, and it is not desirable that, during that process, the judge of fact should be put in a position where he may find himself having to express first an initial view on the basis of the claimant's evidence alone and then (if he allows the claim to proceed) a further final view after taking into account further evidence, even though he does so by reference to different tests. There may be cases, as I pointed out in Boyce, where this consideration is of less force, because nothing in the defendant's evidence could affect the view taken of the claimant's evidence or case. But there is also the second and very important consideration that, if the judge rules that the claimant's evidence does not show a real prospect of success (whether this is for reasons of fact or law), he may prove wrong on appeal. In that event, the procedure adopted will prove to have caused much unnecessary cost, involving a re-trial (quite likely before a different judge, as was ordered in Boyce). It was considerations like these that led, as explained in Alexander -v- Rayson [1936] 1 KB 169, 178-179, to the general practice of entertaining applications at the close of the claimant's case in a civil trial only on the basis of an election by the defendant to call no evidence (in which event the position becomes as stated in paragraphs 17-18 below).

14.

Where a judge does, however, embark at the close of the claimant's case on a determination whether the claimant's case has no real prospect of success without requiring any election, the judge will, if he determines that the claimant's case has no such prospect, dismiss the claim, and this will, subject to any appeal, be the end of the matter. If, on the other hand, the judge determines that the claimant's case has a real prospect of success, he must go on to hear the defendant's evidence and thereafter to find the factual position on the whole of the evidence and on the balance of probabilities.”

Those, I repeat, are the four recent Court of Appeal authorities in point. It is worth reminding oneself, however, of just what had been said on this question by the Court of Appeal all those years ago in Alexander -v- Rayson [1936] 1 KB 169, 178:

“Where an action is being heard by a jury it is, of course, quite usual and often very convenient at the end of the case for the plaintiff, or of the party having the onus of proof, as the defendant had here, for the opposing party to ask for the ruling of the judge whether there is any case to go to the jury, who are the only judges of fact. It also seems to be not unusual in the King’s Bench Division to ask for a similar ruling in actions tried by a judge alone. We think, however, that this is highly inconvenient. For the judge in such cases is also the judge of fact, and we cannot think it right that the judge of fact should be asked to express any opinion upon the evidence until the evidence is completed. Certainly no one would ever dream of asking a jury at the end of a plaintiff’s case to say what verdict they would be prepared to give if the defendant called no evidence, and we fail to see why a judge should be asked such a question in cases where he and not a jury is the judge that has to determine the facts. In such cases we venture to think that the responsibility for not calling rebutting evidence should be upon the other party’s counsel and upon no one else.”

That final sentence becomes more readily understandable when one recognises this simple fact: if on a submission of no case to answer the judge does put the defendant to his election, then one of two consequences necessarily follows. Either the defendant withdraws his submission - in which case, of course, the problem resolves - or he elects to call no evidence - in which case the position is as set out by Mance LJ in paragraphs 17, 18 and 20 of Miller (see paragraph 20 above), subject only to what I say in paragraph 30 below. In other words, it is only when the judge does not put the defendant to his election that it becomes necessary to consider the difficulties arising from a submission of no case to answer. Although one talks about entertaining such a submission without putting the defendant to his election, it is in fact meaningless to refer to a submission of no case except on the basis that the defendant has not been put to his election. Strictly, therefore, it is tautologous to refer both to entertaining a submission and also to not putting the defendant to his election. When hereafter I refer to entertaining a submission of no case, I am to be taken as referring to the hearing of such a submission without putting the defendant to his election.

Before looking further at that body of authority I wish now to introduce into the debate the separate line of authority to which I referred at the outset of this judgment - the cases establishing the principle that in certain circumstances the court is entitled to draw adverse inferences from a party’s failure to call evidence. Until the present appeal no one seems to have noticed that this principle might have some bearing on entertaining submissions of no case. It is, I suggest, relevant, first, as to whether a defendant seeking to make such a submission should be put to his election (ie to whether the court should entertain a submission) and, secondly, assuming he is not put to his election (ie assuming the court does entertain it), what in those circumstances should be regarded as sufficient evidence to establish a real prospect of success.

Rather than myself having to trawl through the line of cases explaining just when adverse inferences can properly be drawn from a party’s failure to give evidence, I am in the fortunate position of being able to draw on Brooke LJ’s leading judgment in this court in Wisniewski -v- Central Manchester Health Authority ([1987] PIQR P324, [1998] Lloyd’s Rep Med 223) itself unfortunately unreported. Brooke LJ in Wisniewski analysed the various cases and derived from them a number of principles (which, if I may respectfully say so, seem to be both accurately and concisely stated). The cases he examined were McQueen -v- Great Western Railway Company (1875) LR 10 QB 569, Chapman -v- Copeland (1966) 110 SJ 569, Herrington -v- British Railways Board [1972] AC 877, O’Donnell -v- Reichard [1975] VR 916, Hughes -v- Liverpool City Council (Court of Appeal transcript, 11 March 1988) and T C Coombs -v- IRC [1991] 2 AC 283.

The principles Brooke LJ derived from those cases are:

“(1)

In certain circumstances a court may be entitled to draw adverse inferences from the absence or silence of a witness who might be expected to have material evidence to give on an issue in an action.

(2)

If a court is willing to draw such inferences they may go to strengthen the evidence adduced on that issue by the other party or to weaken the evidence, if any, adduced by the party who might reasonably have been expected to call the witness.

(3)

There must, however, have been some evidence, however weak, adduced by the former on the matter in question before the court is entitled to draw the desired inference: in other words, there must be a case to answer on that issue.

(4)

If the reason for the witness's absence or silence satisfies the court then no such adverse inference may be drawn. If, on the other hand, there is some credible explanation given, even if it is not wholly satisfactory, the potentially detrimental effect of his/her absence or silence may be reduced or nullified.”

The fourth of those principles, of course, can have no possible relevance to the consideration of no case submissions. The first three, however, seem to me clearly relevant. Let me explain why.

The test to be applied by the judge if he does entertain a no case submission is whether or not on the evidence adduced by the claimant the claimant has “a real prospect of success”. As the cases make plain, this is a different and lower test than that of a balance of probabilities, the test to be applied once the court has heard all the evidence that is to be called. What, however, is less plainly discernible from the cases is just what in this context is meant by “a real prospect of success”. There is, of course, a good deal of authority as to the nature of this test when it applies under CPR Part 24.2. It seems to me difficult, however, to relate the approach to be taken before trial directly to the situation arising at the conclusion of the claimant’s evidence at trial. Mance LJ made much the same point in paragraph 12 of his judgment in Miller (see paragraph 21 above), commenting on what Latham LJ said in Bentley (see paragraph 18 above).

It is at this stage that the relevance of the principles stated in Wisniewski becomes apparent. The judge entertaining a no case submission should in my opinion clearly recognise and bear in mind the real possibility that the defendant, were his submission to fail, might choose to call no evidence (or, indeed, call evidence which in the event proves helpful to the claimant, something in the experience of all of us) thereby entitling the court to draw adverse inferences which go to strengthen the claimant’s case. Of course such adverse inferences can only be drawn when the claimant’s own evidence itself establishes a case to answer. A case to answer, however, as the third Wisniewski principle indicates, is established by “some evidence, however weak” (“only a scintilla of evidence … to support the [relevant] inference” as May LJ put it in one of the earlier authorities, Hughes -v- Liverpool City Council).

Obviously, the possibility of drawing adverse inferences only arises where the defendants have material evidence to give on the issue in question. But generally that will be the case and manifestly it was so in the instant case. As Mance LJ said in paragraph 5 of his judgment in Boyce (see paragraph 14 above):

“There may be some cases, probably rare, in which nothing in the defendant's evidence could affect the view taken about the claimant's evidence or case, but this is not one of them, and care would be required in identifying them.”

Mance LJ reiterated the point in paragraph 13 of his judgment in Miller (see paragraph 21 above). What, however, neither Boyce nor Miller specifically drew attention to is the possibility of a claimant’s case being strengthened after the conclusion of his evidence, either by the defendant not calling evidence or by that evidence in the event damaging his defence.

The point is worth making too even in those cases where the defendant elects to call no evidence. True, as Mance LJ made plain in Miller (see paragraph 20 above), the only issue then is whether the claimant has established his claim on the balance of probabilities. But it must be recognised that he may have done so by establishing no more than a weak prima facie case which has then been strengthened to the necessary standard of proof by the adverse inferences to be drawn from the defendant’s election. Such adverse inferences can in other words tip the balance of probability in the claimant’s favour.

The linking of the two strains of authority in this way to my mind lends added weight to the need for caution at the half way stage of a trial. The disadvantages of entertaining a submission of no case to answer are plain and obvious and have been spelled out already in the cases. Essentially they are twofold. First, as Mance LJ explained both in Boyce and in Miller, the submission interrupts the trial process and requires the judge to make up his mind as to the facts on the basis of one side’s evidence only and applying the lower test of a prima facie case with the result that, if he rejects the submission, he must then make up his mind afresh in the light of whatever further evidence has been called and on the application of a different test. This, to say the least, is not a very satisfactory procedure. The second disadvantage, as again Mance LJ made plain in Boyce and Miller, is that if the judge both entertains and accedes to a submission of no case, his judgment may be reversed on appeal with all the expense and inconvenience resulting from the need to resume the hearing or, more probably, retry the action.

Let me state my central conclusion as emphatically as I can. Rarely, if ever, should a judge trying a civil action without a jury entertain a submission of no case to answer. That clearly was this court’s conclusion in Alexander -v- Rayson and I see no reason to take a different view today, the CPR notwithstanding. Almost without exception the dangers and difficulties involved will outweigh any supposed advantages. Just conceivably, as Mance LJ suggested at the end of paragraph 12 of his judgment in Miller (see paragraph 21), “some flaw of fact or law may … have emerged for the first time, of such a nature as to make it entirely obvious that the claimant’s case must fail, and it may save significant costs if a determination is made at that stage”. Plainly, however, that was not the case here and hardly ever will it be so. Any temptation to entertain a submission should almost invariably be resisted.

With this conclusion in mind, let me at last turn to consider how the submission came to be dealt with below. The judge in effect took two decisions: first, to entertain the submission of no case, ie not to put the defendants to their election; second, to uphold the submission on the basis that “there really is no realistic possibility that this case could succeed on the evidence which I have heard”. Let me consider each in turn.

Having referred to the relevant authorities the judge accepted that although generally the defendant will be put to his election, the judge has a discretion not to do so “in an exceptional case”. He thought that such an exceptional case could arise when two conditions were satisfied: one, that nothing in the defendant’s evidence (which he took to mean the “defendant’s statements and pleadings”) could affect the view taken of the claimant’s evidence, and secondly that it was entirely obvious that the claimant’s case must fail. In the event he thought it “obvious the claimant’s case must fail and that there is nothing in the defendant’s evidence that will alter that view”.

For my part I think that the judge was wrong to take that view. When Mance LJ in Boyce (paragraph 5 of his judgment) and Miller (paragraph 13 of his judgment) spoke of cases where “nothing in a defendant’s evidence would affect the view taken of the claimant’s evidence or case”, he was not to my mind referring to the “defendant’s statements and pleadings” as the judge below supposed, but rather to whether the defendant’s witnesses “might be expected to have material evidence to give on an issue in [the] action” - see the first of the Wisniewski principles set out above. Secondly, it seems to me that whether or not “it is entirely obvious that the claimant’s case must fail” can hardly be the touchstone whether the judge should entertain the submission (ie, not put the defendant to his election); rather it is the test to be applied if the judge does entertain the submission.

Even, however, assuming I am right to conclude, as I do, that the judge ought not to have entertained the submission here in the first place, it does not necessarily follow that Benhams’ appeal must succeed. It is one thing to point to the disadvantage of entertaining a submission of no case (the possible need to examine the claim at two different stages on two different bodies of evidence according to two different tests) and the danger (if the submission is entertained) of ruling erroneously upon it (with the attendant need for a retrial); quite another to conclude in any particular case that the disadvantage has in fact proved fatal and that the danger has materialised. This, indeed, was the point made by Sir Murray Stuart-Smith in paragraph 10 of his judgment in Lloyds -v- John Lewis Partnership (see paragraph 15 above), and by Burton J at the end of paragraph 61 of his judgment in Bentley (see paragraph 16 above). We too “have to decide whether [the judge’s] conclusion [that there was no case to answer] was right”.

It is accordingly necessary to consider whether in fact the judge here, having chosen to entertain the submission, then ruled correctly upon it.

The judge below referred in his judgment to all of the material I have sought to summarise in paragraph 10 above but concluded that it fell short of constituting a case to answer. His essential approach to the question appears most clearly from the following passages in his judgment:

“10.

There is no direct evidence that there was any agreement between Mr Gee and the defendant that he would do anything other than attend the meeting on 14 May or that Benhams would be paid for his services. I have to ask myself whether that is an inference on balance of probabilities that I can draw from the documents and any other evidence that may be relevant, and in my judgment the answer is no, it is not a reasonable inference that I can draw on balance of probability.

…

12.

In my judgment it is more probable that either Mr Gee attended that meeting in the hope or expectation of receiving a secret commission or that he hoped and expected to receive instructions on the resale. I find those inferences in the circumstances to be more probable than that he entered into an agreement honestly to act on behalf of the defendant for reward. As I said, it is perfectly clear that he was in one respect acting on the defendant’s behalf, that is what he was asked to do when he came to the meeting, but for the reasons I have given it seems to me that it is not a probable inference that he did that for reward for Benhams.

…

16.

I am not prepared, I am afraid, from those entries [in Mr Gee’s diary] to infer that Mr Gee had been employed for reward acting on behalf of Benhams in the negotiation and purchase of these properties. It seems to me that these entries are equally consistent with him keeping abreast of events, partly because he was a man who clearly liked to do that and was very skilful at doing that, but also because he had, it seems to me, an interest in arranging the resale of the properties. In my judgment those diary entries are far too tenuous to permit an inference on balance of probability that he had been instructed in the manner that I have indicated.”

Consistently throughout these passages the judge can be seen to have been asking himself what on the balance of probabilities was the correct inference to draw from the claimants’ evidence in the case. That, to my mind, is not the correct question for the judge to be asking himself on a submission. Rather, as indicated above, the question to be resolved at that stage is whether or not the claim has a reasonable prospect of success, a question which, in a case like this where the defendants’ witnesses clearly have material evidence to give on the critical issue in the action, can be re-formulated variously as follows: have the claimants advanced a prima facie case, a case to answer, a scintilla of evidence to support the inference for which they contend, sufficient evidence to call for an explanation from the defendants? That it may be a weak case and unlikely to succeed unless assisted, rather than contradicted, by the defendant’s evidence, or by adverse inferences to be drawn from the defendants’ not calling any evidence, would not allow it to be dismissed on a no case submission.

Had the judge asked himself the correct question with regard to the evidence adduced, in my judgment he would have been bound to reject the defendants’ no case submission. It is unnecessary - and, indeed, inappropriate given that a new trial must now inevitably follow - to express any view as to just how strong or weak Benhams’ prima facie case is. Suffice it to say that in my judgment it could not properly be characterised as having only a fanciful, rather than a realistic, prospect of success, the touchstone of striking out a claim even at the earlier, Part 24.2 stage. I have no doubt that the case crossed the evidential threshold required to defeat a no claim submission.

For all these reasons, expressed I fear at altogether too great length, I for my part would allow this appeal and remit the case for retrial before a different judge.

I agree.

I agree with the judgment of Simon Brown L.J and with the order that he proposes. It seems to me that the wise words of Romer L.J in Alexander –v- Rayson in 1936 still hold good today. Only in the most exceptional circumstances should a judge entertain a submission to dismiss an action at the close of the claimant’s evidence without putting the defendant to his election. This was not such a case and it is difficult to envisage many situations when such a course would be appropriate.