Lord Nimmo Smith

Lady Dorrian

Sir David Edward, Q.C.,

[2009?2009] CSIH NO.60



delivered by SIR DAVID EDWARD, Q.C.,

in the Reclaiming Motion

in the cause



Pursuer and Reclaimer;



Defender and Respondent:


Pursuer: MacAulay, Q.C.; Love; Digby Brown, SSC

Defender: Di Rollo, Q.C.; Davie; Ledingham Chalmers LLP

[Date of Issue]9 July 2009


[1] The pursuer lives in Wellbank near Dundee. On 14 March 2005, she went for a walk with her golden retriever, Cava, in an open field beside the village. Also in the field was the defender accompanied by his four year-old daughter and their black Labrador bitch, Ebony. Both dogs were off the lead and were running around together. At a certain moment, the pursuer stopped, turned round and called "Cava come". Both dogs ran towards her. Ebony collided with the inside of her right knee and knocked her over. Because of the way in which she was standing, with her feet at right angles to the dogs' direction of travel, the pursuer suffered severe injury to her right knee, for which she claims damages from the defender. The Lord Ordinary found, and it is not disputed, that Ebony weighed about 25 kg. and that "it is plain from the severity of her injuries that Ebony must have struck the pursuer with considerable force".

[2] In support of her claim for damages, the pursuer pleaded a statutory case of strict liability based on section 1 of the Animals (Scotland) Act 1987 ("the 1987 Act"), and a common law case based on the alleged negligence of the defender. The quantum of damages (with interest to 22 January 2008) was agreed, in the event of liability being established, at £160,000. The Lord Ordinary found against the pursuer on both cases and assoilzied the defender.

[3] In her reclaiming motion, the pursuer does not challenge the Lord Ordinary's decision on the common law case. So the only question argued before us was whether the Lord Ordinary erred in law in finding against the pursuer on the statutory case of strict liability.

The Animals (Scotland) Act 1987

[4] The passing of the 1987 Act followed a period of gestation of almost a quarter of a century, including recommendations in 1963 of the Law Reform Committee for Scotland (Cmnd.2185); the passing of the Animals Act 1971 (applicable to England and Wales); a recommendation in 1978 by the Royal Commission on Civil Liability and Compensation for Personal Injury (Cmnd.7054) that the 1971 Act be applied to Scotland; a consultative memorandum (No.55) in 1982 by the Scottish Law Commission; and finally a report in 1985 by that Commission (Scot. Law Com. No.97).

[5] In all substantive respects, the layout and terms of the Act are the same as those of the draft Bill set out in Appendix A to the Commission report. Counsel for the pursuer drew our attention to the relevant passages of the report as a guide to interpretation. We do not find it necessary to decide whether it is permissible to refer to the report as a guide to interpretation since, in our view, the terms of the Act are clear once its structure is understood. Indeed, it is a tribute to the skill of the Commission draughtsmen that the Act requires no guide to its interpretation.

[new paragraph here please] [6] The relevant provision for the purposes of the present case is section 1. This is entitled "New provisions as to strict liability for injury and damage caused by animals". Section 2 provides for certain exceptions from liability under section 1. Section 1(8) provides that sections 1 and 2 together

"replace -

(a) any rule of law which imposes liability, without proof of a negligent act or omission, on the owner or possessor of an animal for injury or damage caused by that animal on the ground that the animal is ferae naturae or is otherwise known to be dangerous or harmful;

(b) the Winter Herding Act 1686,

(c) section 1(1) and (2) of the Dogs Act 1906 (injury to cattle or poultry)."


[new paragraph here please]The Act does not affect common law liability for injury or damage caused by animals where a negligent act or omission (fault) is proved, but otherwise sweeps away all the prior law on liability for animals, including the so-called scienter rule. Knowledge that a particular animal is dangerous or harmful may be relevant to proof of fault, but not otherwise.

[68] Section 1 creates strict but not absolute liability in the sense that, once the three criteria set out in subsection (1) are satisfied, liability will follow subject only to the exceptions set out in subsections (4) and (5) and section 2. For present purposes the relevant provisions are subsections (1), (2) and (3).

[79] Section 1(1) sets out the three criteria that must be satisfied in order to establish strict liability:

"a person shall be liable for any injury or damage caused by an animal if -

(a) at the time of the injury or damage complained of, he was a keeper of the animal;

(b) the animal belongs to a species whose members generally are by virtue of their physical attributes or habits likely (unless controlled or restrained) to injure severely or kill persons or animals, or damage property to a material extent; and

(c) the injury or damage complained of is directly referable to such physical attributes or habits."


[new paragraph here please]SSection 5 defines the expression 'keeper' for the purposes of the Act. In the present case, there is no dispute that, at the time of the injury suffered by the pursuer, the defender was "keeper" of the dog Ebony. The requirement of subsection (1)(a) is therefore satisfied and this provision need not be further considered here.

[811] As regards paragraph (b) of subsection (1), subsection (2) provides that the expression "species" includes -

"(a) a form or variety of the species or a sub-division of the species, or the form or variety, identifiable by age, sex or other such criteria as are relevant to the behaviour of animals; ..."

Subsection (3) further provides that

"For the purposes of subsection (1)(b) above -

(a) dogs .... shall be deemed to be likely (unless controlled or restrained) to injure severely or kill persons or animals by biting or otherwise savaging, attacking or harrying;..."

[912] In order to satisfy the requirements of subsection (1)(b), it is necessary, first, to identify the species of animal under consideration, bearing in mind the extended definition of the word 'species' in subsection (2). This definition goes considerably beyond Linnaean classification since the criteria for determining the "species" include any that are "relevant to the behaviour of animals".

[1013] Having identified the species, the next question is whether its "members generally" are "likely" to do certain things, and whether they are likely to do so "by virtue of their physical attributes or habits". So, while some characteristics of the animal concerned (e.g. age or sex) may be relevant for the purpose of defining the "species" under consideration, the characteristics of that particular animal are no longer relevant once the species has been identified.

[1114] The next task is to identify the relevant "physical attributes or habits" that are shared by members of the species generally. Having done so, the next question in the normal case will be whether the members of the species are "likely", by virtue of those physical attributes or habits, unless controlled or restrained, to injure or kill persons or animals, or damage property, noting that the injury must be "severe", or the damage to property "material".

[1215] Having satisfied subsection (1)(b), the last step is to satisfy the test of causation set out in subsection (1)(c). This requires that the injury or damage complained of must be "directly referable" to the physical attributes or habits considered under subsection (1)(b). This requirement is different from the test of causation normally applied in cases of delict and is important in at least two respects.

[1316] First, the process of analysis called for by subsection (1)(b) may be short-circuited by subsection (3), in terms of which dogs are deemed to be likely to injure or kill by biting, savaging, attacking or harrying. So, if the injury complained of is "directly referable" to biting, savaging, attacking or harrying by a dog, then it will not be necessary to go further into the physical attributes or habits of that species of dog.

[new paragraph please][17] Second, if it were to be shown that members of a particular species of animal are likely to cause damage to property, but not to injure or kill persons or animals, this proof of the propensity to cause damage to property would not be sufficient to establish strict liability in a case where, on an isolated occasion, an animal of that species injured or killed a person or animal. That does not, however, as indicated above, exclude the possibility of establishing common law liability for negligence.

[1418] Finally, as regards interpretation of the statute, there was some discussion in the debate before us as to the meaning of the word "likely". Counsel for the defender, in particular, referred us to three cases in which the word was considered. In Bailey v Rolls Royce [1984] ICR 688, the Court of Appeal in England held that "likely" in section 72 of the Factories Act 1961 should be construed as "probable" or "more probable than not". In North Uist Fisheries Ltd v Secretary of State for Scotland 1992 SC 33, the Extra Division held that "likely" falls to be understood as referring to what is probable. Most relevantly for present purposes, in Hunt v Wallis [1994] PIQR P128, Pill, J., construing the word "likely" in section 2(2) of the Animals Act 1971, referred to various definitions proposed by judges of the Court of Appeal ("foreseeable", "such as might happen", "such as might well happen", "where there is a material risk that it will happen" and "more probable than not" - see page P137). In the event Pill, J., gave the word "the broadest meaning I properly can".

[1519] The word "likely" is one that is in ordinary usage and has a range of connotations, depending on the speaker and the context. For our part, we do not think it "likely" that we will assist others in applying the 1987 Act by proposing a definition of the word which the statutory draughtsman has not thought it necessary or appropriate to provide.

Applying the Act to the present case

[1620] As we have noted (paragraph [note change of paragraph numbers - previously 7, now 9, I think][910] above), it is not in dispute that the defender was Ebony's "keeper" for the purposes of section 1 (1)(a). It is not suggested that Ebony "bit, savaged, attacked or harried" the pursuer, so as to bring the deeming provision of subsection (3) into play. Consequently, we are thrown back to the generality of subsection (1)(b).

[1721] While there might be some dispute as to the extent of detail that should be brought into definition of the "species", we know that Ebony was a fully grown black Labrador. We also know that she was female, but there was no suggestion before the Lord Ordinary that her sex was of any relevance. Bearing in mind that subsection 1(2) refers expressly to the age (and, where relevant, sex) of the animal, we might therefore formulate the question raised by subsection (1)(b) as being this:

"Are fully grown black Labradors, by virtue of their physical attributes or habits, likely (unless controlled or restrained) to injure severely or kill persons or animals?"


[new paragraph here please]An example of the way in which this question might have been addressed is to be found in Hunt v Wallis, supra. The issue in that case concerned the characteristics of Border collies. There was some lay evidence, and expert evidence was led, on one side, from a consultant in animal behaviour and, on the other, from a retired veterinary surgeon in a rural practice. We do not suggest that expert evidence will always be necessary. But proof of strict liability under section 1 requires that a finding be made as to the physical attributes and habits of the "members generally" of the species in question. The evidence must therefore address the statutory criteria.


[ new paragraph here please]In the present case, the evidence led for the pursuer as narrated by the Lord Ordinary at paragraphs [2] and [3] of his Opinion (the notes of evidence were not extended) was almost entirely concerned with the characteristics and behaviour of the dog Ebony : that she weighed 25 kg; that she was large, lively and boisterous; that she was friendly towards people and other dogs; that she did not always respond to verbal commands to return to her owner; that although excitable, she was neither aggressive nor vicious; and that, on one previous occasion, ignoring her master's call, while wagging her tail, she had run up to another lady and jumped up placing her paws on that lady's chest.

[1824] The pursuer led an expert in dogs and dog handling, Mr O'Meara. His evidence was that Labradors are "biddable" and that they enjoy completing tasks and following orders, hence their popularity for use as guide and police dogs. Beyond that, his evidence as summarised was that

"when training a dog of whatever size or breed, ... it [is] essential to instil obedience to stop or return commands. ... A dog in a public place can be seen as a threat and can scare people, It can run off and get into danger. ... If not 'recall-proof', a dog should be kept on a lead when in a public place. If his dog was running towards someone, he would command it to stop and return to him".

[1925] Counsel for the pursuer sought valiantly to persuade us that this evidence was sufficient to satisfy the requirements of section 1(1)(b). In our opinion, it fell far short of what the statute requires for proof of strict liability. It went no distance at all towards demonstrating that black Labradors (or any sub-division of the breed, whether by reference to age, sex or any other criterion) are, by virtue of their physical attributes or habits, likely to injure severely or kill persons or animals.


[2026] That is sufficient to dispose of this reclaiming motion. We shall therefore refuse it and adhere to the interlocutor of the Lord Ordinary.