Abstract

The respondent was a police officer. He cared for two police dogs at his home and was paid an additional allowance for doing so. The care provided included feeding the dogs, and exercising and training them. The dogs were being exercised in a field when one of them attacked and injured a runner. In respect of the incident, the respondent was charged with an aggravated offence contrary to s. 3 of the Dangerous Dogs Act 1991, that is, of being the owner or a person in charge of a dog which was dangerously out of control. Before the Crown Court, he relied successfully on the exemption set down in s. 10(3) of the 1991 Act, the material part of which provides that for the purposes of the Act: references to a dog injuring a person or an assistance dog or there being grounds for reasonable apprehension that it will do so do not include references to any case in which the dog is being used for a lawful purpose by a constable or a person in the service of the Crown. (emphasis added)
The appeal heard by the Court of Appeal (Criminal Division) was concerned with two issues. The first related to a jurisdictional matter, namely whether email notification satisfied the requirements of s. 58 of the 2003 Act, or whether this could only be achieved in open court. The second issue related to the correct interpretation of s. 10(3) of the 1991 Act.
Commentary
The Dangerous Dogs Act 1991 is a classic example of how laws may be passed very quickly where there is a consensus within Parliament that a pressing social problem needs to be addressed by legislation. It is also, of course, an Act that has sparked controversy for several reasons, not least of which was because it originally afforded magistrates little option but to order that a dangerous dog be destroyed. This feature of the legislation was described by one commentator as ‘a draconian attempt to ensure even enforcement’: see Hood, ‘Assessing the Dangerous Dogs Act: where does a regulatory law fail?’ [2000] PL 282 at 292. Its contentious nature stemmed in part from the fact that, as noted by another commentator, ‘there are very few mandatory sentences or ancillary orders in English law’: see Tucker, ‘Dangerous Dogs Act 1991: Law bites dog’ [1996] Crim LR 568 at 570.
Various amendments have sought to deal with the flaws in the Act as originally passed, as well as extending the scope of its protection in important ways. Thus an assistance dog can now be a victim of an attack for the purposes of the Act: see ss 3(1) and 10(2). Also, the Act now applies to dogs dangerously out of control in either the private or public domain. Furthermore, the maximum penalty for a dog attack resulting in a human fatality has been increased to 14 years’ imprisonment: see s. 3(4A)(a).
Despite the brevity of the 1991 Act, it has generated much litigation before the courts and has been the subject of a number of Home Office Circulars, for example, Circular 67/1991. Thus, for example, a body of case law evolved on the meaning of ‘public place’ for the purposes of the s. 3 offence when the location of a dog attack was particularly significant: see DPP v Fellowes (1993) 157 JP 936, Bates v DPP (1993) 157 JP 1004, Cummings v DPP, The Times, March 26, 1999 and R v Bogdal [2008] EWCA Crim 1. The present appeal also involved a matter of interpretation; the meaning of the words in s. 10(3) which provide for an exemption to the s. 3 offence where a police dog is ‘being used for a lawful purpose by a constable’. As far as the present commentator is aware, this is the first occasion on which the appeal courts have been asked to consider the scope of the exemption.
There is an inescapable logic to the Court of Appeal’s conclusion that police dogs and their handlers do not enjoy a blanket immunity from the reach of s. 3 by virtue of the exemption. In the judgment of the Lord Chief Justice, Lord Burnett, the exemption ‘imports four concepts’: identifying the dog whose behaviour is under scrutiny; whether the dog was being used at the time; whether there was a lawful purpose; and whether the use was by a police constable (or other Crown servant). These concepts needed to be interpreted in the context of the statutory purpose of s. 3, that is, to protect the public from dogs which are dangerously out of control. Of the four concepts, it was whether the dog was being ‘used’ which was central to the operation of the exemption. Since it was an ordinary word, regard was had to the Shorter Oxford English Dictionary where it was defined as: ‘The act of using a thing for any (especially a profitable) purpose; utilisation or employment for or with some aim or purpose’. As Lord Burnett explained: ‘being used’ in s. 10(3) ‘suggests the active engagement of the dog in a directed task or in support of the person concerned for an identifiable purpose’ (at [41]). To illustrate this meaning, he referred to other instances when dogs can be said to be ‘used’. Thus, for example, the shepherd ‘would be using his dog when working with sheep or taking it to and from the fields’, as would the gamekeeper whose gundog would be used for ‘flushing out and retrieving game’. Similarly analogous was the case of ‘sniffer dogs used to detect drugs, explosives or human remains’ (at [41]). However, in Lord Burnett’s judgment, it was a ‘misuse of language’ to suggest that a dog being taken for a walk was being used, whether or not it was a police dog.
Given the public interest in the police being able to use dogs to carry out various operations, such as policing football matches or apprehending fleeing suspects, and in recognition of the fact that the dogs will have been trained to attack on command, it would have been short-sighted of Parliament if such behaviour was caught by s. 3 of the 1991 Act. Had this been the case, the use of dogs by the police would have been confined to activities where there was little or no prospect that the dog would attack a person. Instead, as the Court of Appeal’s decision makes clear, while there are strong public policy reasons for the s. 10(3) exemption, there are equally strong reasons why it ought to be limited to those instances where a dog (whether a police dog or not) is involved in a lawful policing purpose or activity. On the facts of R v PY, exercising a police dog clearly did not involve a policing activity. Rather, it was something which millions of dog owners do every day.
In the light of the above finding, it was unnecessary for the Court of Appeal to consider a more complex matter; when a policing purpose or activity may cease to be lawful. However, in an obiter remark, Lord Burnett suggested that examples of a police dog being used for an unlawful purpose may include ‘using a dog to fell a suspect to arrest him, or unreasonable force to prevent crime’ (at [43]). In the case of the former, much would depend upon the particular circumstances. Since the police are entitled to use reasonable force when making an arrest (see s. 117 of the Police and Criminal Evidence Act 1984), it might include the use of a police dog in an appropriate situation: see Dhesi v Chief Constable of the West Midlands [2000] Po. L.R. 120.
Jurisdiction
The jurisdictional matter raised by the appeal was of no little practical importance. It was notable that the submission that the appellant had not complied with the procedural requirements in s. 58 of the 2003 Act was made not by counsel for the respondent but by the advocate to the court. Put another way, in a joint submission, the parties to the appeal were in agreement that the court had jurisdiction to hear the case. It was the advocate to the court who argued that in accordance with the proper construction of s. 58, and the authorities, notice of an intention to appeal and ‘the acquittal undertaking’ both had to be given in open court.
Rather more of the Court of Appeal’s judgment was given over to this issue than to the interpretation of s. 10(3) of the Dangerous Dogs Act 1991. Significantly, the recent authorities indicated a shift in judicial thinking over a relatively short period of time. Thus as Lord Burnett noted, remarks made by Hughes LJ (as he then was) in R v Arnold [2008] EWCA Crim 1034 to the effect ‘that the right of appeal must be invoked in open court’ had caused some ‘to question whether an email notification may suffice’ (at [17]). However, some ten years later, in R v The Knightland Foundation Jacob Friedman [2018] EWCA Crim 1860, Hallett LJ VP expressed certain obiter views which Lord Burnett in R v PY described as ‘valuable observations on the nature of the prosecutor’s appeal and electronic notification’ (at [20]). Thus by the time that the present appeal was decided, although it remained the case that complying with the s. 58 requirements in open court was considered to be a good practice, the door had been left ajar in relation to electronic notification.
In the judgment of the Court of Appeal, the underlying purpose of the s. 58 requirements is to ensure that ‘there is no unnecessary delay caused by a prosecution appeal’ (at [26]). Crucially, while the provision is clear as to the steps which need to be taken, it is not prescriptive as to how they are to be accomplished. Thus as Lord Burnett noted, there was ‘no impediment to the prosecution informing the court by email’ (at [27]). Such a conclusion was arrived at through the process of interpretation. It also sat ‘comfortably’ with what was described as ‘the modern trend to use electronic means of communication in connection with proceedings when it is convenient to do so’ (at [29]).
Conclusion
Both aspects of the decision in R v PY are to be welcomed since they reflect a common sense approach to the interpretation of provisions in different statutes. In practical terms, the acceptance that the s. 58 requirements can be met via email rather than in open court does not mean that electronic notification is now likely to become the norm. Rather, the effect of the decision in the present appeal is that on those comparatively rare occasions when the parties are not in court to hear a ruling that can be appealed by the prosecution, it will be appropriate for the prosecution to signify its intentions electronically. On all other occasions, the s. 58 requirements will continue to be satisfied by a statement made in open court.
Turning to the second issue raised by the appeal, had Parliament wanted to completely exclude police dogs from the terms of the 1991 Act it could easily have done so via an appropriately worded provision. The fact that it enacted s. 10(3) reflects its intention to take a more nuanced approach to the police dog exemption, the effect of which has now been clearly explained by the Court of Appeal. Thus although it may be the case that ‘the law relating to dangerous dogs is pock-marked by a multitude of uncertainties’ (see Bleasdale-Hill and Dickinson, ‘“Dangerous Dogs”: different dog, same lamppost?’ [2016] J Crim L 64 at 75), on this point at least, there is now some clarity. Police dog handlers must therefore take care to ensure that their dog is under control when not engaged in a policing activity since if it is not, and it attacks a person, they will be liable to be prosecuted under s. 3 in the same way that the ordinary citizen would if it was their dog which was dangerously out of control.
Allowing the appeal and remitting the matter to the Crown Court for the proceedings to continue will almost inevitably result in a conviction for an aggravated offence given that s. 3 imposes strict liability: see R v Bezzina [1994] 1 WLR 1057. Assuming that this is so, there are various potential consequences for both handler and police dog. A person convicted of an offence under the 1991 Act may be disqualified from future dog ownership (see s. 4(1)(b)) in addition to being fined or imprisoned. For the police dog, it is possible that it may be made the subject of a destruction order (s. 4) or a contingent destruction order (s. 4A). In the circumstances, the latter may be the more likely sentencing option. If this proves to be the case, its days as a police dog are likely to be at an end since the terms of an order, for example, muzzling, being kept on a lead, will severely restrict its ability to be used for policing purposes.
Coincidentally, the decision in R v PY was handed down shortly before the Animal Welfare (Service Animals) Bill, or ‘Finn’s Law’, passed all of its stages in the House of Commons. If it becomes law, as seems likely, it will amend s. 4 of the Animal Welfare Act 2006 so as to provide additional protection to a service animal, for example, a police dog, which is injured by a defendant while it is under the control of a constable and being used by them in the course of their duties in a way that was reasonable. Had the amendment been the law at the time of the events of R v PY and had the runner responded to the attack by injuring the police dog, he would clearly not be guilty of causing unnecessary suffering since the dog was neither under the control of the handler nor being used in the course of the constable’s duties.
