On 17 December 2011 a car driven by the claimant, Mr Martyn Williams, collided with a Charolais steer on a dual carriageway section of the A465 road at Glynneath Bank near Port Talbot. He suffered significant injuries, although fortunately not life-threatening. The steer itself was killed.

In due course he commenced proceedings against the estate of Mr Derfyl Llewellyn Hawkes, a local farmer who had been the owner and keeper of the steer. The claim was framed both in negligence and under the Animals Act 1971 (“the 1971 Act”). Damages claimed were limited to £50,000.

After a trial on liability in the Cardiff County Court, the judge, Mr Recorder Lloyd Williams QC, dismissed the claim in negligence. But he found in favour of the claimant and against the defendant under the 1971 Act. The decision was given on the 23 March 2016. Directions were given at that stage for a further hearing before a District Judge on quantum.

The defendant now appeals against that decision on liability with leave granted by Treacy LJ. The appeal involves, among other things, close consideration of s.2 of the 1971 Act set in the context of the facts of the case. That section has received much judicial and other criticism over the years, which does not need further exposition here: suffice it to say that the wording of the section is marked by both linguistic and conceptual obscurity. In the event, the particular focus of the argument before us on this appeal has been on causation. The appeal itself was heard by us sitting in Cardiff.

The facts are set out with exemplary thoroughness and lucidity by the judge. They can be summarised as follows.

At around 6 pm on the evening of 17 December 2011, in the hours of darkness, Mr Williams was driving on the A465 in the direction of Merthyr Tydfil. There was no lighting on that section of the road and he was using his headlights. He was driving a Ford Mondeo belonging to a Mr Clement. There was no suggestion pursued at trial that Mr Williams was driving too fast or that the car was in anything other than proper condition or anything like that. No defence of contributory negligence was pursued at trial.

By Glynneath Bank the car came into contact with a steer which was on the inner (left) lane of the dual carriageway. Mr Williams himself suffered amnesia as a result of the accident, as well as other injuries, and had no recollection of events.

At the time a car driven by a Mr Bryant was driving about four car lengths behind Mr Williams. He estimated that both were travelling at about 45mph. He then noted the car ahead undergoing sudden and heavy braking. Mr Bryant also braked hard and swerved to the right. As he did so he heard a loud bang and saw what he described as a cow being thrown up from the front of the car ahead. The animal appeared to land on the roof. The Ford Mondeo then came into contact with the bank at the side of the road some yards further up the road. The car was very extensively damaged.

The steer was killed by the impact. It was subsequently found lying at an angle on the inner lane of the carriage way. A witness was later to note that it had “bits of fencing and wood tangled in its coat.”

As it happened, the police had already received reports of a cow loose on the road. Mrs Maria Richards had been travelling on that road with her husband and young daughter. She explained in an agreed witness statement that they had had to swerve to avoid a cow (as she described it) on the inner lane of the carriageway. They then pulled in and commendably contacted the police on their mobile telephone. The police arrived at the scene of the accident (they already being on their way as a result of that and other calls received) at 6.13pm. Mr Bryant had himself by now reported the accident.

The steer had belonged to Mr Derfyl Hawkes. He lived and farmed at a small farm called Pencomin Farm and had done so for many years. The farm itself was, in direct terms, about a mile and a half from this section of the A465.

Mr Hawkes had bought the steer, and two others, at a cattle market at Llandovery the previous day (16 December 2011). His brother had been there with his son, Mr Jeffrey Hawkes (Mr Hawkes Junior), to purchase them on his behalf. The steer was a Charolais, as was one of the others (the third was a Belgian Blue). They were large and heavy animals, each weighing between 370kgs and 380kgs. The undisputed evidence was that Mr Hawkes Senior was an experienced and knowledgeable beef cattle dealer. The judge accepted evidence to the effect that he was experienced and reliable in the handling and keeping of cattle (including Charolais). Mr Hawkes Junior was also knowledgeable and experienced in handling cattle.

Mr Hawkes Junior was to say that they would always try and ensure that cattle they purchased were docile and that that was the case here. When they returned from Llandovery the three steers were taken to Pencomin Farm. The farm included a field or paddock of about one acre with some cattle sheds in it. In accordance with good practice the steers were fed and watered and placed in one of the sheds overnight. The following day they were let out into the field: the door to the shed was left open so that they could access hay and water. The judge found as a fact that Mr Hawkes Senior (who lived on the farm) had been checking on the well-being and behaviour of the steers throughout the day.

Mr Hawkes Junior had been absent for much of the day, having gone Christmas shopping in Swansea. On his return at some time between 3pm and 4pm (when it was still light) he noticed, from his house which adjoined the field, that one of the steers appeared to be missing: one of the Charolais. He spoke to his father who said that he had been checking on them. Mr Hawkes Junior went to investigate. He noticed, in the furthest corner of the field, skid marks leading up to the fence and damage to the fence and an adjoining gate, indicative of the steer having jumped the fence (which was some 6 foot high). In the light of the agreed expert evidence, the judge found – although he had considered it somewhat surprising – that the steer had succeeded in jumping the fence. He also found that the fencing was “entirely appropriate”. It is sufficient here to say that to the extent that any claim in negligence was pursued at trial it was rejected by the judge; and there is no appeal from that conclusion.

The judge further found that the steer must have jumped or forced its way through a number of hedges or fences before it reached the road. Quite what route it had followed was not, however, established.

At all events, Mr Hawkes Junior and his daughter went to look for the steer but could not find it. They also telephoned around, unsuccessfully. This was the first experience of an escape of an animal from the farm. Later that evening, however, he received a report of a dead animal on the A465. He drove there and saw a steer lying dead at the roadside. It was the Charolais steer which had escaped from the field at Pencomin Farm.

Mr Hawkes Senior died in May 2012, at the age of 89.

The claimant called Professor Eddison as an expert witness. The defence had instructed Mr Greetham. Prior to the trial, and after exchange of reports, the experts had met. There was in truth relatively little between them and they submitted a detailed agreed joint memorandum before trial, dated 13 January 2016. Professor Eddison gave oral evidence at trial and was extensively cross-examined. Mr Greetham was not called to give oral evidence. The judge was impressed by Professor Eddison as a witness.

All were agreed that something had startled, or “spooked”, the steer causing it to leap the fence and escape. Quite what that was was never identified.

It was said in the written report of Professor Eddison that Charolais steers could be “flighty” – that being explained as connoting a rapid response to an unexpected event and the taking of avoiding action to avoid what was described as an “averse stimulus”. Such stimuli could include, for instance, being in a strange place, a loud noise, a sudden light in darkness, disturbance by an animal such as a dog and so on.

Professor Eddison said that once the steer had jumped the fence it would have been very frightened, being alone in a strange environment. That fright, and accompanying unpredictable behaviour, would have continued as it made its way over or through other fences or hedges. He was much pressed with the proposition that as the road was, in direct terms, no more than about a mile and a half from the farm the effect of the original averse stimulus (whatever it was) would have been superseded by the time it got onto the road some two hours later. Professor Eddison rejected that. He said that, in his opinion, its frightened and unpredictable behaviour would have continued throughout. Indeed, when on the road the problem would have been exacerbated by the lights and noise of passing cars, which would have been inexplicable to the steer. All these could be further averse stimuli resulting in wholly unpredictable reactive behaviour on the part of the steer. Professor Eddison also rejected the proposition that the steer would, by the time it reached the road, have been physically exhausted. It was put to him that the steer would in any case have been “mentally exhausted”. His response was that steers do not have a considered thought process but would simply react to an averse stimulus in a wholly unpredictable way.

The judge accepted all this evidence. Indeed in the agreed joint memorandum the experts both had said this at paragraph 16:

It was of course the case that there was little direct evidence of precisely what had happened at the time of the actual collision (Mr Williams having no memory and Mr Bryant only observing from some distance back). But the judge was entitled to have regard to the evidence of Professor Eddison, as he did, in deciding where the probabilities lay in what occurred that night.

The judge set out the background facts very fully. He carefully assessed the expert evidence. He went through the relevant provisions of the 1971 Act and referred to certain legal authorities (to which I will myself come). He said that “the question of liability and in particular causation requires careful analysis of what took place.” The core of his decision is to be found in the concluding three paragraphs of his judgment.

Clearly there are competing policy considerations in deciding the extent to which keepers of animals should be liable for the actions of their animals which cause damage. It is those considerations which are to be taken as addressed by Parliament in the 1971 Act. As stated by Lord Nicholls in the leading case of Mirvahedy v Henley[2003] UKHL 16, [2003] 2 AC 491 at paragraph 9:

Thus the general theme of the 1971 Act is of strict liability but with certain exceptions or modifications provided. Central to a case such as the present is s.2 (a section described by Lord Nicholls as “opaque”). That provides as follows:

It is to be noted that s.5 provides for various exceptions from liability. These include, among others, cases where damage is suffered by a person who has voluntarily accepted the risk. Section 8 provides for a duty to take reasonable care to see that damage is not caused by animals straying onto a highway.

One problem arising under s.2 (2)(b) came to be identified. It was this: was a keeper of an animal, not in itself being an inherently dangerous animal, strictly liable for damage caused by the animal when the animal’s behaviour in the circumstances was in no way abnormal for an animal of that species in those circumstances?

The views of the courts had differed on this. Some thought that that was what the wording of the section mandated. Others, however, thought that it could never have been the intention of Parliament to impose strict liability on keepers in circumstances where the animal in question was behaving normally for its species in those circumstances.

It was that point that had to be settled by the House of Lords in Mirvahedy (cited above). Mirvahedy itself was a case involving horses. Three horses had panicked as a result of an unidentified event and escaped from their field. They eventually got onto a dual carriageway road. One horse crashed into the side of the car of Mr Mirvahedy: the horse being killed, the car being severely damaged and Mr Mirvahedy suffering serious personal injuries. By a majority of 3-2, the House of Lords decided in favour of the claimant. In doing so, it rejected the latter approach set out above. It answered the question posed (as set out in paragraph 28 above) in the affirmative. It decided that such a conclusion reflected the language used by Parliament. It was acknowledged by Lord Nicholls (at paragraph 43) that that interpretation of s.2 (2)(b)

Given the facts as found by the judge, given the judge’s acceptance of the evidence of Professor Eddison and given the interpretation of s.2 (2)(b) as authoritatively decided by the majority of the House of Lords in Mirvahedy, it is at first sight rather difficult to see how the defendant in this case could escape liability under s.2 (2) of the 1971 Act.

However, what is said by Mr O’Leary is that the claim should have failed on the issue of causation. He submitted that there was no basis for concluding that the collision was caused by characteristics of the steer which are not normally found in animals of the same species or are not normally so found except at particular times or in particular circumstances. He asserted that the judge had failed sufficiently to address the question of whether, but for the section 2 (2)(b) characteristics, the claimant would have suffered damage.

In my view, in approaching this ground of appeal it is essential to follow – as certainly the judge below had endeavoured to do – the structure of the sub-section.

The first point, then, is to consider whether damage was caused by an animal which did not belong to a dangerous species (as defined). Here, the answer is obvious. It was.

The second point is to consider the requirement of s.2 (2)(a). Here, too, the position was agreed at trial. The damage was of such a kind. It was of such a kind just because of the size and weight of the Charolais steer. Both the experts were agreed on that, stating that any injury such a steer might cause was likely to be severe on account of their size. The judge rightly accepted that. The position corresponds with that posited by Lord Nicholls in Mirvahedy in paragraph 46 of his speech (cited above).

Moving, next, to the fourth point- the question of knowledge under s.2(2)(c)- again this was an agreed matter at trial. It was accepted that Mr Hawkes Senior knew of the relevant characteristics.

That, then, brings one back to s. 2(2)(b). No one suggested that the likelihood of the damage was due to the characteristics of the animal not normally found in animals of the same species (that is, the first limb of s.2(2)(b)). But, on the expert evidence, the likelihood of damage was due to characteristics not normally so found except at particular times or in particular circumstances. The damage was not simply attributable to the size and weight of the cow (that satisfied s.2(2)(a) alone). The damage, as the judge found, applying Mirvahedy, was also attributable to the steer behaving in this dangerous way in the particular circumstance of it having been spooked by the various averse stimuli, as steers are wont to do in such circumstances.

Mr O’Leary, however, drew our attention to the facts of Mirvahedy. In that case, the horse in question had careered into the side of the unfortunate Mr Mirvahedy’s car. In the present case, the steer had been in front of Mr Williams’ car. The trial judge apparently had thought that Mr O’Leary’s argument had primarily rested on the proposition that the steer was static at the time Mr Williams’ car came into contact with it: that is, that it was critical, on Mr O’Leary’s argument, that Mr Williams’ car had collided with the steer rather then the steer colliding with the car. Indeed that seemed to be one focus of Mr O’Leary’s written argument on this appeal (he described as a “crucial element” of Mirvahedy that the horse ran into the car and not the other way round). I would very much query if this point – that is, whether or not the steer was in effect static at the time of the collision – could of itself in this particular case necessarily avoid strict liability under the 1971 Act. Could it, for example, really have made a difference if the steer, acting under the influence of the averse stimuli, had careered down the road, with cars avoiding it, but then been struck by a car at a time when it had temporarily come to a halt? I doubt it. Further, as Mrs Collins observed, it surely cannot make any difference in principle in this case if Mr Williams had crashed his car having swerved so as (successfully) to avoid coming into contact with the steer. But in any event the point is met by the judge’s finding of fact that the steer was running in a panic on the carriageway at the time of collision, at a time when it was subject to the averse stimuli.

It was maintained, nevertheless, that s.2(2)(a) and s.2(2)(b) were linked; and in particular that there must be a causal link between the characteristic in question and the (likelihood of the) damage caused: see McKenny v Foster [2008] EWCA Civ 173 at paragraph 33 of the judgment of May LJ, (a case itself involving a cow on a highway, albeit it was a case which was ultimately decided on the lack of knowledge for the purpose of s.2(2)(c): see paragraph 39 of the judgment).

The high water-mark of Mr O’Leary’s argument, perhaps, was the ex tempore decision of a two judge Court of Appeal in the unreported case of Jaundrill v Gillett (16 January 1996). That decision is binding on us, sitting as we were in a two judge court in the instant case. But it is only binding on us for what it actually decides. And it is distinguishable on its facts, as Mr O’Leary fairly and rightly accepted.

In Jaundrill, some horses were let out of a field by some malicious intruder or intruders. Shortly thereafter, they got onto a road, this being in the hours of darkness, and were picked up by car headlights galloping down the road. The plaintiff’s car collided with two of the horses: his car was damaged and he was injured. The Court of Appeal, reversing the trial judge, decided that there was no liability under the 1971 Act.

The trial judge had found on the evidence that the horses had been driven in a group onto the road by the intruders and “in the dark, being removed from their accustomed environment, it was likely that they would behave unpredictably and tend to panic and gallop as they did”: there being expert evidence, accepted by the trial judge, to the effect that horses do not gallop unless they are stimulated to do so either by a rider or by fright.

Notwithstanding those findings of fact, the Court of Appeal felt able to interfere. In fact, a reservation was expressed by Russell LJ (with whom Singer J agreed) as to whether a horse that gallops on a highway at night and is driven to panic can be said to be displaying a characteristic within the ambit of s.2. Be that as it may, Russell LJ went on to say (at page 6 of the transcript of judgment) that “there has to be a causal link between the characteristic in question and the damage suffered.” He then went on to hold that the “only sensible approach” to the question of causation was that “the galloping and/or panicking of these horses was not causative of the accident…in any real sense.” He further went on to say this:

With respect, I have some difficulty with this decision. For one thing, the case was decided before Mirvahedy; and aspects of Russell LJ’s remarks (for instance, the reservation as to a relevant characteristic being present) are not altogether easy to align with the approach taken in Mirvahedy. For another, it departed from the trial judge’s findings of fact and did so on a basis not entirely easy to follow. Moreover, the bald statement that there was no evidence that the galloping of horses aggravated the situation seems, to me at least, somewhat out of line with the “sensible approach” which Russell LJ himself was advocating. Indeed, the suggestion that it was the presence of the horses on the highway that was causative perhaps comes close to saying that this particular collision would have happened anyway, whether the horses were galloping or not: again, a proposition I find difficult.

However the present case is, as I say, distinguishable. It involved a steer with no natural tendency to “gallop” and which, as found, was throughout acting both under the influence of the initial averse stimulus causing it to jump the six foot fence and escape and then under the further averse stimuli of travelling through hedges and fences in a strange environment and then the lights and noise of the road traffic. The present case, on its facts, thus cannot be shoehorned into the ambit of Jaundrill. Nevertheless, I would for myself say that I think that the case of Jaundrill is to be treated with great caution and certainly it is to be treated as a decision confined to its own particular facts and circumstances.

This is also, I think, borne out by the outcome in Mirvahedy itself. For that case had involved not only the question of interpretation of s.2 (2)(b) discussed above but also an issue of causation. In the county court, the trial judge, following Jaundrill, had found against the claimant on the ground that the characteristics in question had not caused the damage. Counsel for the defendants had argued before the Court of Appeal, as he had below, that causation was not established because it was the presence of the horses on the highway and their size, rather than any characteristics which the horses exhibited, that was causative of the damage sustained: see [2001] EWCA Civ 1749, [2002] QB 769 at p772C. (This closely reflects, I note, the argument which Mr O’Leary was advancing before us). However, in giving her reserved judgment (with which the other two members of the court agreed), Hale LJ, after referring to Jaundrill, said this at paragraph 16:

This approach as to causation was endorsed by the majority in the House of Lords. Thus Lord Nicholls said this at paragraph 48:

It is true that neither the House of Lords nor the Court of Appeal purported expressly to overrule or not follow the decision in Jaundrill. They did not need to. It was distinguishable on its facts. But certainly their decisions lend no endorsement whatsoever of some more general approach to be extracted from Jaundrill. To the contrary: the decision of the county court judge, who had purported to follow Jaundrill, was reversed.

All that said, I return to the facts of the present case. It is true that (unlike the horse in Mirvahedy) the steer was not careering, as such, across the highway; nor did it crash into the side of Mr Williams’ car. But those are not material differences. It is futile, in the circumstances of this particular case, to ask whether the steer collided with the car or the car collided with the steer. The steer was, as found, running on the A465 in a panic, acting under the original (but ongoing) averse stimulus which had given rise to the escape from the field in the first place and then exacerbated by subsequent averse stimuli such as the lights and noise of the cars. The linked requirements of s.2 (2)(a) and (b) thus were satisfied on the facts of this case. That was the conclusion of the judge at paragraphs 38 to 40 of his judgment: and it was a conclusion wholly open to him on the evidence.

There is no basis for interfering with the findings or conclusion of the trial judge. I would in fact pay tribute to the care and thoroughness evident in his judgment. I would dismiss this appeal.

For the reasons given by my Lord, Davis LJ, I would also dismiss this appeal.

I would only add that I too find Jaundrill difficult, particularly after the exposition of the law in Mirvahedy; and agree that Jaundrill is properly construed as a decision confined to its own peculiar facts as the court found them to be. As such, in my respectful view, it provides no assistance as to the relevant legal principles; and it is very unlikely to be of any assistance to a court in applying those principles. Therefore, whilst Mr O’Leary was in my view right to draw Jaundrill to our attention, in the future considerable caution should be given before it is cited as an authority.