Welsh v Stokes & Anor

On 31 July 2001, the claimant was riding a 9 year old horse called Ivor on a road near Sticker in Cornwall. She was 17 years of age at the time. She fell off the horse and suffered a serious head injury. She has no memory of the accident. She had been working as a trainee at the defendants’ yard since October 2000. The judge found that, although there were some doubts as to her riding abilities, she was competent to ride a “sensible” horse on her own. Ivor was a “sensible” horse with no history of misbehaviour or vice of any kind. The accident occurred at a cross-roads. There were two motorists in the vicinity at the time. Mr Wragg, who gave evidence, was one. The other was an unidentified man, who failed to remain at the scene or leave his name and address. Mr Wragg (and his wife who was travelling with him) did not see the accident. But he said that the other motorist had told him that he had seen the accident. In his second statement, Mr Wragg said that the other man had said that “the horse had reared up, then the girl had fallen off and the horse had fallen onto her.”

The claimant issued these proceedings alleging that the accident was caused by the negligence of the defendants. Further and alternatively, she alleged that the defendants were strictly liable to her pursuant to section 2(2) of the Animals Act 1971 (“the 1971 Act”).

There was a trial on the issues of liability. In a judgment given on 19 October 2006, Judge Tyzack QC dismissed the claim in negligence. But he found that strict liability was established under section 2(2) of the 1971 Act. An essential element of his reasoning was his decision under the Civil Evidence Act 1995 (“the 1995 Act”) to give weight to, and accept the reliability of, the hearsay evidence given by Mr Wragg. Without that evidence, the claimant could not have proved how the accident had occurred and the judge would have been bound to dismiss the claim.

The defendants appeal with the permission of Rix LJ. There are two principal issues. The first is whether the judge was right to give weight to the hearsay evidence. The second is whether the judge correctly applied section 2(2) of the 1971 Act.

So far as material, the 1995 Act provides:

The judge said (para 34) that the hearsay evidence had led to the “greatest difficulty in this case”. One of the difficulties was that there were differences between Mr Wragg’s first and second statements. In his first statement, he said that the unidentified man had said: “Apparently, the horse had stepped backwards and then the young girl had fallen off and the horse had fallen onto her.” The claimant’s solicitors asked for amplification of this statement. One of the questions asked was: “What exactly the untraced driver said to you about the circumstances of the accident. Did he say that the horse reared, or did he say that it stepped back?” In response, Mr Wragg said in his second statement: “Apparently the horse had reared up and then the young girl had fallen off and the horse had fallen onto her.”

At para 36 of his judgment, the judge referred to the “obvious danger and concern” that the questions had influenced Mr Wragg or led him to change his statement. But in his oral evidence Mr Wragg was “adamant that he was told that the horse was being encouraged to go forward, it did not want to, and so it reared and fell back with the claimant.”

At para 42, the judge turned to the provisions of section 4(2) of the 1995 Act and considered each of paras (a) to (f) in turn. He found that it would not have been reasonable or practicable for the claimant to have produced the unidentified maker of the original statement (para (a)). The original statement was made reasonably contemporaneously with what had happened (para (b)). Mr Wragg’s evidence did not involve multiple hearsay (para (c)). There was no evidence that any of the circumstances stated in paragraphs (d),(e) or (f) applied.

The judge said that he had to consider all the facts and determine whether any inferences could reasonably be drawn from them. In deciding which version of the hearsay evidence to accept, the judge also had regard to the evidence of the claimant’s father (paras 46 and 47). Mr Welsh said in his statement that he had spoken to Mr Wragg a few weeks or months after the accident and that Mr Wragg had told him that the unidentified man had said: “the horse stepped back---went backwards. The horse reared and the girl fell off. The horse fell backwards onto her.” As the judge said, this was second-hand hearsay. But it was consistent with the account given by Mr Wragg in his second statement.

There was expert evidence on the question of whether Ivor was likely to “rear”. At para 36 of his judgment, the judge said:

The judge set out his conclusions of fact at para 48:

It can be seen, therefore, that the judge gave weight to the hearsay evidence. It was the essential basis for his crucial finding that the claimant fell from Ivor when the horse reared because it was unwilling, for some reason, to go forward and the claimant was unable to handle the situation.

Miss Rodway QC submits that the claimant’s case depended entirely on the reported statements of an unidentified witness. She (correctly) concedes that the hearsay evidence was admissible. But she submits that, as a matter of law, the uncorroborated and untested hearsay evidence of the unknown motorist carried insufficient weight to allow a court to make specific findings of fact based upon it. The general rule is that any fact that needs to be proved by the evidence of a witness should be proved by oral evidence: see CPR 32.2(1)(a).

The “tick-box” approach to section 4(2) of the 1995 Act adopted by the judge should not have automatically led to the hearsay evidence being accorded sufficient weight to decide the central issues of fact in the case. The claimant was unable to test the evidence. She was unable to explore issues such as how good a view the unidentified motorist had of the accident, precisely what he saw, whether he saw only part of the incident, whether his eye was drawn to the accident before, during or after the accident, whether he was telling the truth and whether he was involved in the accident in some way.

The judge was in error in failing to consider the credibility and reliability of the unknown motorist. He appears simply to have taken his account at face value because it was “credible”. Miss Rodway criticises the judge for placing any weight on the evidence of Mr Mackie that the account given by the unknown motorist was “credible”. She submits that the opinion of Mr Mackie was based on the assumption that the unknown motorist did in fact see what Mr Wragg reported. She relies on the following exchange in the course of her cross-examination of Mr Mackie:

Yes, for sure.

Because anything could have happened to cause the Claimant to fall. She could have simply lost an iron and tipped off herself, could she not?

Certainly. I am basing this on Mr. Wragg’s second statement where the motorist said the horse reared.

We do not know and you do not know and Mr. Wragg was fair enough to say we have no way of knowing whether the motorist is telling the truth or not?

Sure.

And we have no way of testing what the motorist actually saw?

There is no way at all, except that it is consistent in my view.

If Aimee was lying on the ground beside a horse in the middle of the road, that is in itself consistent with her falling for any number of other reasons, is it not, you would agree?

Yes”.

In her skeleton argument, Miss Rodway refers to a number of authorities to support her submission that hearsay evidence should not be relied on to prove the central issues in a case unless supported by other evidence, although in her oral submissions she showed less enthusiasm for them. In The “Ferdinand Retzlaff” [1972] 2 Lloyd’s Rep120, the plaintiff shipowners claimed damages for detention following a collision with the defendants’ ship. There was an issue as to how long the ship repairs would have taken if they had been done at Bremen. The defendants adduced evidence on this point in the form of an exchange of letters. In considering what weight to give to the letters under section 6(3) of the Civil Evidence Act 1968, Brandon J identified a number of circumstances to which he had regard. Apart from these matters, he said (p 127) that it was necessary to take account of the fact that the evidence was unsworn and could not be tested by cross-examination. This was particularly important where the evidence related to hypothetical rather than actual facts. He then made this general observation on which Miss Rodway relies:

In The “Kilmun” [1988] 2 Lloyd’s Rep 1, Leggatt J said at p 6 that, although giving evidence by way of statements under the Civil Evidence Act 1968 was convenient, “it is obvious that it is not a satisfactory way of resolving disputed issues of fact”.

In Polanski v Conde Nast Publications Ltd [2005] UKHL 10, [2005] 1 WLR 637, Baroness Hale said at para 78: “It might be grossly unjust to the other party, even contrary to his right to a fair trial under article 6 of the European Convention on Human Rights to decide a claim principally on the untested evidence of a party who has not been subject to cross-examination of any sort”.

I would dismiss this ground of appeal substantially for the reasons given by Mr Stead. The judge directed himself properly in his consideration of the weight to be accorded to the hearsay evidence. He had regard to each of the factors in section 4(2) of the 1995 Act. He was entitled to do so and Miss Rodway does not criticise his assessment of the factors. His approach to section 4(2) did not “automatically” lead him to giving the hearsay evidence critical weight. He satisfied himself that Mr Wragg’s second statement gave an accurate account of what had been said by the unidentified motorist. He carefully considered whether the second statement had been influenced by the solicitors’ questions and concluded for rational reasons that it had not been. He was able to assess Mr Wragg as a witness and also took into account the evidence of Mr Welsh which was consistent with Mr Wragg’s second statement.

The judge was plainly alive to the difficulty of assessing the reliability of the account given by the unidentified witness. But he did have expert evidence to the effect that the account was “credible”. I take “credible” to mean “plausible”. That was important evidence which the judge was entitled to take into account in deciding to give weight to the evidence. I accept that the defendants were unable to explore the reliability of the account or to raise issues of the kind that Miss Rodway would have wished to canvass in cross-examination. But there was no doubt that the claimant fell off Ivor and as a result sustained a head injury. In these circumstances, there was a limited number of plausible explanations for the accident. Miss Rodway suggested to Mr Mackie in cross-examination that the claimant might simply have lost an iron and tipped off herself. Another possibility was that the horse could have reared causing her to fall off as the judge found. No doubt there are other possibilities too, but Miss Rodway did not suggest any in the course of argument. In these circumstances, the expert evidence that the account attributed by Mr Wragg to the unidentified witness was “credible” assumed particular significance. I do not accept that the opinion of Mr Mackie that the account given by the unknown motorist was credible was in any way dependent on the truth of that account. All that Mr Mackie was saying was that the account given by the motorist provided a plausible explanation of how the claimant came to be injured.

In my judgment, therefore, the judge was entitled to infer that the hearsay evidence was reliable and conclude that he should give it weight. He reached a conclusion that was reasonable and well within the ambit of conclusions that were reasonable for a court to reach.

Even if the hearsay evidence were the only evidence on which the claim was based, I would not accept that this was necessarily a reason for giving it no weight. It would depend on all the circumstances. I accept that there will be cases where it is so unfair to hold a defendant liable solely on the basis of hearsay evidence that a court should place little or no weight on the evidence. Consideration of the factors stated in section 4(2) will point the way, but will not necessarily be determinative. In some cases the defendant may be able to adduce evidence to contradict, or at least cast doubt on, the hearsay evidence. But there will also be cases, like the present, where the defendant is not in that position. Apart from the unidentified motorist and the claimant, there were no witnesses to the accident. In such a case, there may be said to be unfairness to the defendant in having to face hearsay evidence which he cannot directly challenge. On the other hand, there would be unfairness to the claimant to place no weight on the hearsay evidence, since without it her claim would inevitably fail.

The decision what weight (if any) to give to hearsay evidence involves an exercise of judgment. The court has to reach a conclusion as to its reliability as best it can on all the available material. Where a case depends entirely on hearsay evidence, the court will be particularly careful before concluding that it can be given any weight. But there is no rule of law which prohibits a court from giving weight to hearsay evidence merely because it is uncorroborated and cannot be tested or contradicted by the opposing party. I do not consider that the statements in the authorities relied on by Miss Rodway in her skeleton argument support such an extreme proposition.

In The “Ferdinand Retzlaff”, before making the general comment relied on by Miss Rodway referred to at para 16 above, Brandon J had (at p 126-127) given four reasons (specific to the facts of that case) to which regard should be had in judging the weight to be attached to the statements in the letters. He also said that it was particularly important that the unsworn and untestable evidence related to hypothetical rather than actual facts. Nor was there any reason to suppose in that case that the makers of the statements could not have been called to give oral evidence. That is the context in which the observations made by Brandon J fall to be considered. It is important to note that they were general observations: viz “I cannot think that the Civil Evidence Act 1968 was intended, in general, to change the long established system...” It is clear from the passage that he thought that the Act was being “abused” by the defendants in that case.

I do not consider that the brief statement of Leggatt J in The “Kilmun” provides any support for the broad statement of principle for which Miss Rodway contends. The same may be said of the passage in opinion of Baroness Hale in Polanski. Baroness Hale was not considering a case where the maker of the statement cannot be subjected to cross-examination, nor was she addressing the question of how the court should estimate the weight to be given to hearsay evidence under section 4 of the 1995 Act. At para 74, she referred to the 1995 Act and said that the principal safeguard to the opposing party who is unable to cross-examine the witness is the weight to be given to a statement which has not been tested by cross-examination: “The court is to be trusted to give the statement such weight as it is worth in all the circumstances of the case”.

For these reasons, I would dismiss the appeal in so far as it is based on the hearsay point.

So far as material, section 2 provides:

The judge correctly directed himself that all three conditions in section 2(2) had to be proved before strict liability could be established. As regards the first limb of subsection (2)(a), he found at para 51 that the damage was not of a kind which, unless restrained, Ivor was likely to cause. Ivor was a docile animal and the defendants had no reason to believe that he would rear up in the way that he did.

As regards the second limb of subsection (2)(a), he said at para 52:

In relation to subsection (2)(b), the judge said at para 53 that the likelihood of the damage being severe was “possibly” due to characteristics of Ivor which were not normally found in other horses, because horses do not normally rear up and fall on their riders. He continued:

At para 54 he said:

As regards subsection (2)(c), the judge said at para 55 that the “characteristics” that were to be known to “that keeper” were the characteristics proved in (b), namely that Ivor does not normally rear save in particular circumstances. The judge continued:

Miss Rodway does not challenge the judge’s finding that the damage was of a kind which, if caused by Ivor, was likely to be severe, ie that the second limb of subsection (2)(a) was satisfied.

Her challenges to the judge’s conclusions in relation to subsection (2)(b) have fluctuated and I confess that I have not found them altogether easy to follow. In para 11 of her grounds of appeal, she contended that the judge erred in law in finding that subsection (2)(b) was satisfied for the reasons set out in para 9 of the grounds. Para 9 stated that the judge’s findings “of the circumstances of the accident were inconsistent with his other findings in the case and the weight of the evidence”. Particulars of these other findings were then given. In short, they were the judge’s findings that Ivor had never previously “misbehaved”. Yet at paras 3 and 4 of her skeleton argument for the appeal dated 4 December 2006, Miss Rodway said that the issues arising under the 1971 Act were “concerned with the correct approach to the interpretation of sections 2(2)(a) and 2(2)(c) of the Act upon which there is little direct authority”: there is no reference to subsection (2)(b) here. But at para 39 she said that she would seek to persuade the court that the judge was wrong to find that subsection (2)(b) was satisfied because he did not have sufficiently cogent evidence before him to make such a finding.

In the course of her oral argument, she submitted that the judge’s decision in relation to subsection (2)(b) was flawed because he did not direct himself correctly as to the causal link between the kind of damage that Ivor was likely to cause (subsection (2)(a)) and the characteristics found in subsection (2)(b). Save in this limited respect, she made no complaint that the judge had misdirected himself as to the meaning of subsection (2)(b).

During the course of Mr Stead’s submissions, the question was raised by the court as to whether “normally” in subsection (2)(b) means “usually” or “naturally”. We invited counsel to make written submissions on this one point and we received these after the completion of the oral argument. In her written submissions, Miss Rodway argued for the first time that the judge had adopted the wrong approach to subsection (2)(b) because, instead of considering whether rearing was a characteristic normally found in horses in particular circumstances, he considered whether rearing was behaviour of which Ivor was “capable”.

It is unsatisfactory that the appellant’s case on subsection (2)(b) should have developed in this way. Nevertheless, since Mr Stead has been able to respond to all of Miss Rodway’s arguments, I shall attempt to deal with all of them in this judgment.

Miss Rodway also submits that the judge’s decision in relation to section 2(2)(c) was wrong because he considered whether the defendants knew that horses in general had the characteristics he found in Ivor rather than whether they knew that Ivor had those characteristics.

The relevant particular circumstances for the purposes of the second limb of subsection (2)(b) are that, for some reason, the horse did not want to go forward and had a rider on board who was unable to handle him and give him confidence in the situation (para 53). I shall refer to these as “the particular circumstances”. As I understand her argument, Miss Rodway submits that the judge found that the likelihood of severe damage was due to the rider falling on a tarmac road or the horse falling on her and that these are not characteristics of the horse within the meaning of subsection (2)(b).

The basis for this submission is the penultimate sentence of para 52 (see para 28 above). But in that sentence the judge was considering the likelihood of severe damage by reference to the specific facts of this case, ie as a result of Ivor rearing and causing the claimant to fall on a tarmac road and the horse falling on her. In my view, the link between the likelihood of severe damage and the characteristic found by the judge was obvious. If a horse rears in the particular circumstances and the rider falls from the horse, she is likely to suffer severe damage. That may be because she falls on a hard surface (which need not be a tarmac road) or because the horse falls on her or in some other way. I do not understand Miss Rodway to challenge this. It seems to me that, once the judge found that the damage that was caused was likely to be severe and that Ivor had the characteristic of rearing, it was inevitable that he would find that the likelihood of the damage being severe was due to that characteristic. I would, therefore, reject this criticism of the judgment.

The meaning of section 2(2)(b) has been authoritatively explained by the House of Lords in Mirvahedy v Henley [2003] UKHL 16, [2003] 2 AC 491. The claimant suffered injury when the car he was driving collided with the defendants’ horse which had panicked and escaped with others from its field. It was not clear what had frightened the horses. The House held by a majority of three to two that the defendants were liable under section 2(2). To bolt was a characteristic of horses which was normal “in the particular circumstances”, these being some sort of fright or other external stimulus. The main issue concerned the true meaning of the second limb of subsection (2)(b). The majority adopted that favoured by the Court of Appeal in Cummings v Grainger [1997] QB 397. Thus the fact that an animal’s behaviour, although not normal behaviour generally for animals of the species, is nevertheless normal behaviour for the species in the particular circumstances does not take the case outside section 2(2)(b).

At para 43, Lord Nicholls of Birkenhead noted of the Cummings interpretation that:

The question of what is meant by “normally” was not in issue in Mirvahedy (or any other case that has been cited to us). There are passages in the opinions of their Lordships which, it might be said, suggest that they considered that “normally” means “usually”, rather than “conforming to type” or “naturally”. The clearest is at para 3, where Lord Nicholls said that the behaviour of the horse in that case was “usual in horses when sufficiently alarmed by a threat”. At para 23 he also said that the horse “was not behaving any differently from the way any normal horse would have behaved in the circumstances” (emphasis added). But since the meaning of the word “normally” was not in issue, these statements do not provide a secure basis for deciding precisely what it means. In any event, it is plain that, if it is usual for horses to bolt when sufficiently alarmed, it is also natural and conforming to type for horses to bolt in such circumstances. But it does not necessarily follow that, if it is unusual for horses to bolt when sufficiently alarmed, it is abnormal for them to bolt in such circumstances.

The Oxford English Dictionary provides a definition of “normal” as being “according to or squaring with a norm; constituting, conforming to, not deviating from a type or standard; regular, usual 1828”. A “norm” is defined as a “rule or authoritative standard”. “Abnormal” is defined as “deviating from the type; contrary to rule or system; unusual 1835.” Depending on the context, therefore, “normal” can mean “conforming to a type” or “usual”. The latter meaning connotes a greater degree of regularity or frequency of occurrence than the former. But even the former must connote some frequency of occurrence. If a characteristic is rarely found in animals of the same species, it may be difficult to say that the characteristic conforms to the type of animal in question.

In some contexts, it is clear that the word “normally” means “usually”. If I say: “I normally travel to work on the No 18 bus”, I am saying that I usually travel to work on that bus. I may occasionally travel to work by different means, but that is an exception to my usual practice. In other contexts, however, the position is different. It is a proper use of language to say “horses will most often turn and flee when faced with a frightening stimulus, but it also normal for them to rear in such circumstances.” It is normal for horses to rear when frightened in such circumstances, because it is natural for them to do so, although rearing may be a less usual response than turning and fleeing. Another way of making the same point is to say that it is not abnormal (even if it is unusual) for horses to rear when frightened.

It seems to me that the core meaning of “normal” is “conforming to type”. If a characteristic of an animal is usual, then it will certainly be normal. The best evidence that a characteristic conforms to the type of animals of a species is that the characteristic is usually found in those animals.

I can find nothing in the context of subsection (2)(b) to suggest that Parliament did not intend “normally” to bear this core meaning. It is difficult to see why Parliament should have intended to exclude from the ambit of subsection (2)(b) cases where the relevant characteristic is natural, although unusual, in the animal which has caused the damage. There is no need for such a narrow interpretation because a claim will not succeed unless the knowledge requirement in subsection (c) is also satisfied. To adopt the language of Lord Walker of Gestingthorpe in Mirvahedy (para 157), if section 2(2)(b) is interpreted in this way, there is nothing unjust or unreasonable, as between the keeper (who can decide whether “to run the unavoidable risks involved in keeping horses” and whether or not to insure against those risks) and the victim of the horse’s behaviour, in requiring the keeper to bear the loss.

There was evidence directed to the question of whether, and if so in what circumstances, it was normal for horses to rear. I have already referred to Mr Mackie’s written answer quoted by the judge at para 36 of his judgment that rearing in the particular circumstances comes within the parameters of “normal equine behaviour”(see para 10 above). In his oral evidence, Mr Mackie said (p 3) that fear generally causes horses to stop. “A horse will tend to try to run from any fear. That will be its initial reaction. It is when it cannot run that it is likely to rear….. If the rider controls the horse, particularly in a slightly overbearing way, not necessarily, but very commonly, the horse cannot go forwards, cannot run round and run away, and so it will rear.” He said this was “very common”. At p 4 the following exchange took place:

At p 7, Mr Mackie said that a three day test such as that to which Ivor was subjected when first acquired by the defendants would certainly rule out any vices or bad habits, “but of course that does not rule out the possibility of the horse behaving unpredictably in a particular circumstance.” At p 16 he said:

At p 17, Mr Mackie was asked about the passage in his report where he had stated: “Rearing, bucking and kicking are normal behaviour for horses which they do in certain circumstances…”. He explained that he was saying that “rearing is within the range of a horse’s natural behaviour”. A little later, he said:

Finally, in re-examination, Mr Mackie said that horses can react in a number of ways to different circumstances. They “can rear under the sort of circumstances I have already described.”.

Mr Stokes was also asked about the circumstances in which horses may rear. He has worked with horses all his life and has run a business dealing in horses for 18-20 years. He accepted (p 10) that horses may rear, although it was not a “natural move”. He said that horses do not usually tend to rear at the first sign of something frightening them. “They’ll probably just whip away quick”. Or “they will step back from what they’re not liking.” In answer to the question whether, “if they are pushed on, ridden on at that moment, it is a characteristic of some of them to rear”, Mr Stokes said: “I’ve seen some horses rear, yes.” He agreed that it is their way of avoiding what is frightening them. At p 20, he said that he could count on his hand the number of times horses have reared up with him when frightened and not wanting to go forward.

It was on the basis of this evidence that the judge said that Ivor was “capable” of rearing in the particular circumstances. He used the word again in para 56 when dealing with the question of knowledge: like any horse of his kind, Ivor was “capable” of rearing. A little later in the same paragraph, the judge said that horses are “unpredictable” animals which are “quite capable of behaving in an unpredictable way, especially if they are being ridden by someone who lacks the necessary skills to handle a crisis situation.”

Miss Rodway criticises the judge in two principal respects. First, she submits that the sole basis for the judge’s conclusion that subsection (2)(b) was satisfied was that rearing in the particular circumstances was something of which Ivor, like any horse, was “capable”: that is not the statutory test. The second limb of the subsection is directed at behaviour which is normally found in animals of the same species in particular circumstances, not behaviour of which animals of the same species are capable in such circumstances.

Secondly, Miss Rodway submits that a conclusion that rearing was behaviour normally found in horses in the particular circumstances was inconsistent with other central findings made by the judge. In particular, the judge found at para 51 that rearing was not likely or foreseeable and that the defendants had no reason to believe that Ivor would rear up in the way that he did. Furthermore, at para 53 the judge said that “rearing up in the way I find Ivor did was not a normal characteristic for him.”

She also relies on a number of detailed findings by the judge which she submits are inconsistent with a finding that Ivor was capable of rearing in the particular circumstances. Thus at para 12, the judge found that the claimant was “a sufficiently competent, confident and experienced rider to ride different horses on the roads and lanes near the defendants’ yard, both accompanied and unaccompanied” and that Mrs Stokes “thought the claimant was both a competent and confident rider”. As against that, however, the judge also found in the same para that this “needed to be balanced by the undoubted facts that she did fall off Tabitha Twitch about five months before the accident, largely due to inexperience, and did not pass the assessment on 16^th June 2001 in respect of riding a schooled horse.”

Miss Rodway also points to para 14 where the judge described the experiences of Ivor without any sign of a problem before the accident. At para 20, the judge said that the defendants had sold Ivor on and that they had had “good reports and no complaints” from the purchasers. At para 21 he said that there had been no evidence that Ivor had ever misbehaved before the accident by rearing or in any other way and there had been no evidence that he had manifested any vices since the accident. At para 24 the judge said that the experts were agreed that three days of testing “was sufficient to discover whether he had any bad habits such as napping, spooking or rearing” and none had been shown.

I cannot accept any of Miss Rodway’s criticisms of the judgment. As regards her first criticism, it is true that subsection (2)(b) is not concerned with behaviour of which an animal may be capable. The second limb is concerned with characteristics not normally found in animals of the same species except in particular circumstances. But in my view the judge did not apply the wrong test. Para 53 is the critical paragraph (see para 30 above). Having said that rearing generally was not a normal characteristic for Ivor because he did not rear regularly or repetitively, the judge said: “However, I do find that Ivor was capable of rearing up…in [the] particular circumstances…” In my view, it is clear in this context that, by saying that Ivor was capable of rearing up in the particular circumstances, the judge meant that it was a normal characteristic of Ivor to rear up in such circumstances (because it was natural for horses to do so from time to time). The word “however” shows that the judge was explaining why, although it was not normal for Ivor generally to rear up, it was normal for him to do so in the particular circumstances.

There was ample evidence to support such a finding. I refer to paras 48-52 above. Particularly striking was the evidence of Mr Mackie that it was “very common” for horses to rear if ridden by a person who does not allow the horse to run away from a source of fear but insists on riding forwards, and his reference to rearing in the particular circumstances being within the parameters of “normal equine behaviour.” I also refer to his statement that “rearing is within the range of a horse’s natural behaviour”.

The answer to the second criticism is that it is nothing to the point that there was no evidence that Ivor had ever reared in this way before. The relevant question is whether rearing was a characteristic normally found in horses as a species in the particular circumstances. It may be that, so far as the evidence went, Ivor had never been exposed to the particular circumstances, so that his response to them had never been tested. The relevant question was not whether Ivor tended to rear generally, but whether he had the characteristic of rearing in the particular circumstances. I do not accept that the judge made inconsistent findings. When he said that Ivor was not likely to rear up and that rearing up was not a normal characteristic “for him”, he was saying no more than that Ivor had no track record of rearing. He was not saying that it was not normal for horses as a species to rear up in the particular circumstances.

It is implicit in his reasoning that the judge considered that “normally” means “conforming to type” and not “usually”. For the reasons given at paras 41-47 above, in my view he was right to do so. For these reasons, I would reject all of Miss Rodway’s submissions in relation to subsection (2)(b).

Miss Rodway submits that the judge failed to direct himself correctly as to the need for the claimant to establish that the defendants had specific knowledge that Ivor had the characteristic of rearing in the particular circumstances. The knowledge found by the judge in paras 55 and 56 was knowledge of horses in general. Thus, for example, the judge said: “the defendants as experienced keepers of horses would have known that Ivor, like any horse of his kind, was capable of rearing…” (emphasis added).

Miss Rodway submits that what is required is specific knowledge on the part of the keeper of the allegedly dangerous propensity of the specific animal. Knowledge that animals of the species in general have the relevant propensity is insufficient. She relies on the decision of this court in Breeden v Lampard (unreported) 21 March 1985, Court of Appeal (Civil Division) Transcript No 1035 of 1985. She also seeks to rely on the debates in Parliament when the Animals Bill was being discussed. Finally, she seeks to derive support from Professor North’s The Modern Law of Animals (1972).

In Breeden, the plaintiff had been injured by a horse in a riding accident. The judge found that the defendant’s horse was young but not a “kicker”, since it had not kicked before and there was no evidence to suggest that it was anything more than a normal 5-year old horse. The judge held that there was no breach of section 2 of the 1971 Act. The plaintiff appealed. The first judgment was given by Sir George Waller. Although he said at p 4D that subsection (2)(c) had no application to the case, it seems clear that he in fact decided the appeal on the basis of that subsection as well as on the basis that subsection (2)(b) was not satisfied. At p 5B-D, he said that subsection (2)(c) required that the characteristics should be known to the defendant and the judge’s finding of knowledge was correct. Lloyd LJ said at p 7F that the short answer to the claim under section 2(2) was that, even if subsection (2)(b) was satisfied, the judge accepted the defendant’s evidence that she did not know that the horse possessed the particular characteristic.

It will be seen, therefore, that both Sir George Waller and Lloyd LJ decided that subsection (2)(c) was not satisfied because the judge was entitled to find on the evidence that the defendant did not have the necessary knowledge. Neither of them considered what the requirement of knowledge entailed. Oliver LJ said that he agreed that the appeal should be dismissed for the reasons given by the other members of the court. At p 11, however, he added the following:

It is significant that Oliver LJ said that it was not sufficient to establish knowledge to show that the relevant behaviour is sufficiently common in the relevant species to put the keeper “on notice” that the particular animal had the characteristic. I make two points about this. First, it is not clear whether Oliver LJ was saying anything more than that constructive knowledge is insufficient: the phrase “on notice” is perhaps significant. Secondly, these observations were obiter, since the judge accepted the defendant’s evidence that she did not know of the particular characteristic. That must have been a reference to her actual knowledge.

It follows that the decision in Breeden does not resolve the question at issue.

As regards the debate in Parliament, Miss Rodway drew our attention to a number of passages in Hansard which she submits show that, with regard to non-dangerous animals, the intention of Parliament was that knowledge of a specifically dangerous characteristic in the particular animal was required. A keeper is not, therefore, liable for “one-off” circumstances, but is only liable if he knows of the “mischievous propensity” as distinct from the general possibility that the animal may behave in a certain way in particular circumstances. Thus, for example, at the second reading before the House of Lords on 29 October 1970 at column 842 the Lord Chancellor clarified the intention of the Bill. He said that under the common law “where you are keeping for instance, a horse, which is an animal mansuetae naturae…you are not responsible unless you know its particular proclivity… That is the law as it is, and broadly speaking that is what we intend to continue.” When the Bill came before the Second Reading Committee on 27 January 1971, the Attorney-General said:

I do not consider that subsection (2)(c) is ambiguous or obscure. Nor does the interpretation adopted by the judge lead to absurdity. I do not, therefore, consider that any of the conditions for recourse to Hansard prescribed by the House of Lords in Pepper v Hart [1993] AC 593 is present. But even if I am wrong about that, I do not consider that the statements made in the House of Lords cast light on the question at issue. They do not show that it was intended that knowledge of the relevant characteristic could not be acquired by knowing how animals of the same species normally behave in the particular circumstances. None of the passages relied on by Miss Rodway deals with the question of what knowledge is required or how it may be proved.

It is not in dispute that subsection (2)(c) requires it to be shown that the keeper knew that the particular animal which caused the damage had the characteristics found to satisfy subsection (2)(b). The only question is how that knowledge can be proved. Miss Rodway submits that it can only be proved by showing that the keeper knew that the particular animal had previously behaved in that way.

I do not agree. I do not see why a keeper’s knowledge that a horse has the characteristic of normally behaving in a certain way in particular circumstances cannot be established by showing that the keeper knows that horses as a species normally behave in that way in those circumstances. Indeed, Mirvahedy shows that subsection (2)(b) may be satisfied where the characteristic is displayed by the animal in the same particular times or circumstances as by other animals of the same species. It is a general characteristic of horses to bolt in the particular circumstances of the facts of Mirvahedy, or to rear in the particular circumstances of the present case. It makes no sense to require a keeper, if aware of that general characteristic, to have some additional and more particular knowledge.

As regards Professor North’s book, Miss Rodway cites from pages 58-61. Professor North identifies a number of ways in which he says that the requirement of knowledge may be satisfied. These include proof that the animal has, to the knowledge of the keeper, actually caused this kind of damage before, that the animal has attempted to do the harm in question but, so far, without success, and proof that the keeper knows that the animal has a vicious characteristic even though it has not yet attempted to cause any injury.

But Professor North does not consider the particular issues raised by the second limb of subsection (2)(b) and, of course, did not have the benefit of the elucidation provided in Mirvahedy. Nor is there any discussion of the particular issue of whether knowledge of the characteristic found in the particular animal in particular circumstances can be acquired by knowledge that the characteristic is to be found generally in the particular circumstances in animals of the same species.

In my judgment, therefore, the judge was entitled to find on the evidence that the defendants had the knowledge required by subsection (2)(c).

For the reasons that I have given, I would dismiss this appeal.

I agree.

I also agree.