Criminal Injuries Compensation Authority v First-Tier Tribunal (Social Entitlement Chamber) & Anor

On 15^th August 2002 at about 5.15 p.m. TS, then aged 14, was riding his bicycle along the pavement of a quiet residential street near his home when a small dog, which had escaped from its owner’s garden, rushed up to him barking in an aggressive manner. TS instinctively swerved away from the dog on to the road and into the path of a car. He was seriously injured. He spent four months in hospital and is now quite severely disabled. The owner of the dog was well aware that it was aggressive towards strangers (although it had apparently never bitten anyone) and had in the past frightened them by barking at them. She was also aware that it would do its best to escape from the garden where it was normally confined. Proceedings were taken against her under section 3(1) of the Dangerous Dogs Act 1991, but the prosecution was discontinued after she relinquished ownership of the dog and so no conviction ensued.

TS submitted a claim for compensation to the appellant, the Criminal Injuries Compensation Authority (“CICA”), but it was rejected on the grounds that his injuries were not directly attributable to a crime of violence. (Under the scheme governing the payment of compensation for injuries sustained as a result of crime it is necessary for the claimant to establish that his injuries are “directly attributable to a crime of violence.”) An appeal was lodged with the First-tier Tribunal, which exercises the jurisdiction of the former Criminal Injuries Compensation Appeals Panel. The tribunal disagreed with CICA. It found that the dog had a history of being aggressive if it escaped and that it had been aggressive towards TS. It considered on a balance of probabilities that the injuries suffered by TS were directly attributable to a crime of violence and, after a further hearing to consider the medical evidence, on 2^nd December 2010 it awarded him £499,155 by way of compensation.

There being no provision for an appeal from the First-tier Tribunal to the Upper Tribunal on a matter of this kind, on 28^th February 2011 CICA started proceedings for judicial review of the decision on the grounds that the tribunal’s finding contained an error of law or was irrational. In substance, CICA contended that it was not possible to regard the only offence committed by the dog’s owner, namely an offence under section 3(1) of the Dangerous Dogs Act 1991, as a crime of violence. On 25^th August 2011 Dobbs J. transferred the claim to the Upper Tribunal.

In the Upper Tribunal the case came before Upper Tribunal Judge Levenson. On 12^th December 2011 he gave permission for the claim to proceed limited to the question whether the injuries suffered by TS had been attributable to a crime of violence. An oral hearing took place before Judge Levenson on 8^th November 2012. As he himself remarked, it is regrettable that matters took so long to get to that point.

In a decision delivered on 19^th November 2012 Judge Levenson upheld the decision of the First-tier Tribunal. He directed himself that what matters for the purposes of deciding whether there has been a crime of violence is the nature of the crime rather than its likely consequences. Having considered the leading authority on the meaning of a “crime of violence” in this context, R v Criminal Injuries Compensation Board ex parte Webb [1987] Q.B. 74, and the decision of this court in R (Jones) v First-tier Tribunal (Social Entitlement Chamber) [2011] EWCA Civ 400, [2012] Q.B. 345, he upheld the tribunal’s decision. He expressed his conclusion in the following way:

“24.

Applying the law as explained above, the first question is whether an offence under section 3 of the Dangerous Dogs Act 1991 is a crime of violence for the purposes of the 2001 scheme. This is a question of law. I can see nothing in the legislation, the 2001 scheme or the case law that would prevent such an offence being classed as a crime of violence. The offence can only be committed if a dog is “dangerously out of control” (my emphasis). Even the non-aggravated offence carries a possible sentence of six months imprisonment. The concept of the dog being dangerously out of control involves grounds for reasonable apprehension that it will injure any person (my emphasis). The fact that no mens rea (mental attitude that must be established before the offence can be proved to have been committed) is specified does not stop it being a crime of violence for the purposes of the scheme, notwithstanding Mr Johnson’s assertions to the contrary

. . .

26.

I accept that the reasons given by the panel in the present case could have been more detailed and extensive but I reject Mr Johnson’s caricature of the decision as being that “an unidentified offender had committed some unspecified crime of violence”. In my view the First-tier Tribunal did enough to establish that the dog’s owner had committed a crime of violence under the 1971 Act. It referred to the history of the dog being aggressive if it got loose from the back yard and it explicitly accepted evidence which (although the panel did not spell it out) established that the offence had been committed.”

Mr. Collins for CICA submitted that the decision of the First-tier Tribunal (and therefore that of the Upper Tribunal) was wrong in law for two reasons: first, because the owner of the dog had not committed a crime of any kind; and secondly, because, if she had, it was not a crime of violence. Miss Shah for TS submitted that, although not convicted, the owner of the dog had committed an offence under section 3(1) of the Dangerous Dogs Act 1991 and that since violence or the threat of violence was an essential element of the offence, it could properly be regarded as a crime of violence. It was common ground that it was not necessary for there to have been a conviction in order for a crime of violence to have been committed.

The only crime which anyone has suggested might have been committed in this case is an offence under section 3(1) of the Dangerous Dogs Act 1991. It is convenient, therefore, to begin by setting out those parts of the Act that are of particular relevance to this case. Section 3(1) provides as follows:

“If a dog is dangerously out of control in a public place –

(a)

the owner . . . is guilty of an offence or, if the dog while so out of control injures any person, an aggravated offence, under this subsection.”

Section 10(3) defines the meaning of “dangerously out of control” as follows:

“For the purposes of this Act a dog shall be regarded as dangerously out of control on any occasion on which there are grounds for reasonable apprehension that it will injure any person, whether or not it actually does so.”

It follows that in a case such as the present the offence is committed simply by being the owner of a dog which is in a public place in circumstances giving rise to grounds for reasonable apprehension that it will injure someone.

In Jones the court was concerned with a claim for compensation for injuries suffered by a lorry driver in a road traffic accident arising out of a suicide. The deceased had run into the path of another lorry on a busy road; the driver had swerved, unsuccessfully, to avoid him causing his vehicle to collide with that driven by the claimant. The claimant was injured as a result. The questions for decision were whether the deceased had committed an offence under section 20 of the Offences against the Person Act 1861 and, if so, whether that amounted to a crime of violence. The First-tier Tribunal was not persuaded that the deceased had committed an offence, because it was not satisfied on the evidence before it that he had intended to cause harm or had been reckless whether his actions might cause harm to others. In paragraph 24 of his judgment in this court (with which Mummery and Rix L.JJ. agreed) Patten L.J. held that whether a criminal offence had been committed was undoubtedly a question of fact for CICA or the tribunal to decide, a view which was later endorsed by Lord Hope when the matter came before the Supreme Court. It will be necessary to refer again to the Supreme Court’s decision in Jones, but for present purposes it is sufficient to note that, since the question whether a crime has been committed is one of fact, the court cannot set aside the decision of the First-tier Tribunal save on the grounds that it is one to which no rational tribunal could properly have come.

The tribunal did not expressly identify the crime that it was satisfied had been committed, but I think it is clear from the way in which it expressed itself that it must have found that the owner of the dog had committed an offence under section 3(1) of the Dangerous Dogs Act 1991. In the present case the First-tier Tribunal found that the dog had behaved aggressively towards TS (although it did not actually attempt to bite him). Judge Levenson in the Upper Tribunal said that it had been “barking and snapping”. That was not part of the findings of the First-tier Tribunal, but I am prepared to assume that it is broadly what it meant. Mr. Collins submitted that a dog, particularly a small dog, may be aggressive without giving grounds for reasonable apprehension that it will cause injury, but even small dogs are armed with sharp teeth and biting is the only serious method of attack available to them. In the case of a dog, therefore, aggression almost inevitably involves a more or less immediate threat of being bitten. In those circumstances I am unable to accept the submission that no offence was committed in this case and that the tribunal was not entitled to find that it had been. If a dog is behaving aggressively towards a person there will almost always be grounds for reasonable apprehension that it will injure him. The nature and severity of the injury are irrelevant for this purpose.

The more difficult question is whether the First-tier Tribunal was entitled to find that the offence was a crime of violence. The provisions of the scheme governing the payment of compensation contain no definition of a crime of violence, so it is necessary to turn to the authorities for assistance.

In Ex parte Webb the court was concerned with claims for compensation for personal injury in the form of psychological harm suffered by the drivers of trains which had struck and killed people trespassing on the line. In three cases the deaths were suicides, but the fourth involved a senile man who may not have known what he was doing. The claims were made on the basis that each of the deceased had committed an offence under section 34 of the Offences against the Person Act 1861 which provides that:

“Whosoever, by any unlawful act, or any wilful omission or neglect, shall endanger . . . the safety of any person conveyed or being in or upon a railway, . . . shall be guilty of a misdemeanour . . . ”

Lawton L.J., with whom Stephen Brown L.J. and Sir John Megaw agreed, described the history of the compensation scheme and the approach to be taken to the construction of its terms as follows:

“A scheme to compensate the victims of crime was first introduced by the Government in 1964. That scheme provided for compensation to be payable for “personal injuries directly attributable to a crime.” This brought within the scheme persons injured as a result of breaches of regulatory statutes such as the Factories Act 1961 and Food and Drugs Act 1955. In 1969 the scheme was modified by limiting its operation to a “personal injury attributable to a crime of violence.” The words “crime of violence” are not a term of art. The scheme is not a statutory one. The government has made funds available for the payment of compensation without being under a statutory duty to do so. It follows, in my judgment, that the court should not construe the scheme as if it were a statute but as a public announcement of what the government was willing to do. This entails the court deciding what would be a reasonable and literate man’s understanding of the circumstances in which he could under the scheme be paid compensation for personal injury caused by a crime of violence.”

Later, when dealing with the question whether the offences in that case were “crimes of violence” he said at pages 79C-80B:

“Mr. Wright submitted that the correct approach to this problem is to start by construing the words in their grammatical context. The word “crime” by itself covers all unlawful acts or omissions for which the law imposes a penalty. The draftsman of the scheme as amended clearly intended to limit the meaning of the word “crime.” He did so by the use of the qualifying words “of violence.” These words are adjectival and indicate the nature of the crime to which the scheme applies. The nature of a crime is different from its consequences. Injury to a person may be the probable consequence of a failure to fence a dangerous part of a machine, contrary to section 14 of the Factories Act 1961 (which is an offence), but no one would say that such a failure amounted to a crime of violence. If consideration of probable consequences is what makes a crime one of violence, a motorist who leaves his vehicle in a dangerous position contrary to section 24 of the Road Traffic Act 1972 commits a crime of violence. Mr. Wright accepted that the acts of the four deceased caused psychiatric injury to the applicants; but that was because of the consequence of their acts, not their nature.

. . .

In my judgment, Mr. Wright’s submission that what matters is the nature of the crime, not its likely consequences, is well founded. It is for the board to decide whether unlawful conduct, because of its nature, not its consequence, amounts to a crime of violence. As Lord Widgery C.J. pointed out in Clowes’s case, at p. 1364, following what Lord Reid had said in Cozens v. Brutus [1973] A.C. 854 , the meaning of “crime of violence” is “very much a jury point.” Most crimes of violence will involve the infliction or threat of force but some may not. I do not think it prudent to attempt a definition of words of ordinary usage in English which the board, as a fact finding body, have to apply to the case before them. They will recognise a crime of violence when they hear about it, even though as a matter of semantics it may be difficult to produce a definition which is not too narrow or so wide as to produce absurd consequences, . . .”

Those statements of the law were subsequently approved by the Supreme Court in Jones [2013] UKSC 19, [2013] 2 A.C. 48.

It follows from Ex parte Webb that when deciding whether a crime is a “crime of violence” it is necessary to have regard to the nature of the offence rather than its consequences. However, since some conduct may constitute an offence whether or not it is accompanied by violence, it is necessary to have regard to the facts of the offence itself in order to decide whether it amounts to a crime of violence: see R (August) v Criminal Injuries Compensation Appeals Panel, R (Brown) v Criminal Injuries Compensation Appeals Panel [2001] 1 Q.B. 774. That is so in the case of the offence created by section 3(1) of the Dangerous Dogs Act 1991. In R v Bezzina [1994] 1 W.L.R. 1057 this court held that mens rea is not required for the offence, which is one of strict liability. The position was described by Kennedy L.J. at page 1061E as follows:

“ . . . if a dog is in a public place, if the person accused is shown to be the owner of the dog, if the dog is dangerously out of control in the sense that the dog is shown to be acting in a way that gives grounds for reasonable apprehension that it would injure anyone, liability follows.”

The offence may therefore be committed without any violence on the part of the owner or even on the part of the dog. If, however, the owner of a dog known to be vicious were to release it in a public place in order to enable it to attack someone, I can see that the crime might be categorised as one of violence. In those circumstances, however, other, more serious, offences of a violent nature are likely to have been committed.

In Jones the First-tier Tribunal had dismissed the claimant’s appeal on the grounds that the deceased had not intended to cause harm and had not been reckless whether someone might be harmed by his actions. He did not therefore have the mens rea necessary for an offence under section 20 of the Offences against the Person Act 1861 and so no crime had been committed. The Upper Tribunal dismissed a claim for judicial review of that decision, holding that the finding was one which a rational tribunal could properly make. This court, however, held that it was highly improbable that anyone who ran into the path of traffic on a busy road would not foresee the possibility of harm being caused to other road users and that the First-tier Tribunal had therefore erred in law in finding that there was no evidence from which it could infer foresight of some harm on the part of the deceased.

The decision of this court was subsequently overturned by the Supreme Court in a judgment delivered after the decision of the Upper Tribunal in the present case. Lord Hope, giving the leading judgment, pointed out that the first question the First-tier Tribunal had to decide was whether an offence had been committed at all. The only offence suggested was that of unlawfully inflicting grievous bodily harm contrary to section 20 of the Offences against the Person Act 1861. That being so, the tribunal had to decide as a matter of fact whether the deceased had intended to cause harm or had been reckless whether harm might be caused by his actions, in the sense that he had actually foreseen that some physical harm to others might ensue and yet had gone on to take the risk. The tribunal was not satisfied of either of those matters and was therefore unable to find that a criminal offence had been committed. The decision was open to a rational tribunal on the evidence before it and it was therefore not open to challenge. In those circumstances it was unnecessary for the court to consider whether the crime, had one been committed, would have been a crime of violence for the purposes of the scheme. In paragraph 24 of his judgment below Patten L.J. had expressed the view that the question was one of law. I can see nothing to suggest that Lord Hope dissented from that view, although he made it clear that the question whether a section 20 offence is necessarily a crime of violence “admits of only one answer” (paragraph 27).

Both Lord Hope and Lord Carnwath emphasised that the decision of the First-tier Tribunal in a case of this kind is to be read benevolently and that both the Upper Tribunal and the Court of Appeal should be cautious about interfering with its decisions, even when they are less fully expressed than might be desirable. In paragraph 16 of his judgment Lord Hope referred to the acceptance by Buxton L.J. in August of the proposition that whether a crime of violence has taken place is a jury question. He was of the view that it would be more accurate to say that it is for the tribunal which decides the case to consider whether the words “a crime of violence” do or do not apply to the facts which have been proved and he agreed with Lord Carnwath that it is primarily for the tribunals, not the appellate courts, to develop a consistent approach to those issues. Lord Hope was also of the view that “a pragmatic approach should be taken to the dividing line between law and fact, so that the expertise of the tribunals can be used to best effect.” His views were echoed by Lord Carnwath, who suggested that some flexibility may be appropriate in this context in applying the traditional distinction between decisions of law and decisions of fact.

In the light of these observations and of the fact that the proceedings before the Upper Tribunal involved a claim for judicial review, it can be seen that the question for the Upper Tribunal was whether it was open to the First-Tier Tribunal to find that the claimant’s injuries were directly attributable to a crime of violence. If the circumstances of the offence were incapable of amounting to a crime of violence, it was entitled to interfere, but not otherwise. The same principles apply to this court.

For the reasons I have given I have no doubt that the First-tier Tribunal was right to find that an offence under the Dangerous Dogs Act 1991 had been committed, but in any event that was a finding of fact reasonably open to it on the evidence and so not open to review by the Upper Tribunal or this court. Whether the offence was properly to be characterised as a crime of violence for these purposes is, however, more difficult. The tribunal held that it was, without explaining why it reached that decision. The decision may have been instinctive, as Lawton’s L.J.’s comment in Ex parte Webb suggested might be the case, but it is none the worse for that, provided it was one to which a rational tribunal, correctly applying the law, could come.

The critical distinction for these purposes is between the nature of the crime and its consequences. The tribunal did not find, expressly or by implication, that the dog had been deliberately allowed out of the garden; on the contrary, the indications are that the owner was negligent at worst in failing to prevent its escape. I find it difficult to accept that negligently to allow a dog to escape, even a dog known to be aggressive, constitutes a crime of violence, giving that expression its ordinary meaning. It is clear from Ex parte Webb that a crime does not become a crime of violence for these purposes just because it results in injury to the person. The nature of a crime is different from its consequences, even if they might reasonably have been foreseen, as in the case of failing to fence a dangerous part of a machine or leaving a vehicle in a dangerous position on the road, two examples given by Lawton L.J. in that case.

In my view the offence in this case could only be described as a crime of violence if one were to have regard to its consequences rather than its nature. Although he had referred to Ex parte Webb at an earlier stage in his reasons, Judge Levenson did not really deal with the distinction between nature and consequences when he came to make his decision. In my judgment, in common with the First-tier Tribunal he concentrated on the consequences of the crime rather than its essential nature. Whether a crime of violence has been committed will depend on the particular circumstances of the case. The fact that the dog was known to be aggressive clearly weighed with both the First-tier and Upper Tribunals, but I do not think it is sufficient to enable the offence to be characterised as a crime of violence, any more than would be the case of an unfenced machine known to be dangerous.

For these reasons I do not think it was open to the First-tier Tribunal to find that the offence committed by the owner of the dog was a crime of violence and accordingly I would allow the appeal.

I agree.

I also agree.