Written evidence submitted by Professor Claire Parkinson (DDL0200)

Executive summary

• The current approach to protecting the public from dog attacks is ineffective.
• Increases in the number of fatalities and injuries requiring hospital admission due to dog bites attests to the failures of current legislation.
• The costs of these failures to the NHS and police forces are substantial as are the impacts on dog welfare and to those families who lose or are separated from their canine companions.
• Media reporting impacts on public understanding of dog risk and has contributed to the demonization of certain dogs and dog owners.
• Subjective visual assessment of the physical characteristics of a dog is fallible and inconsistent. There is no reliable evidence that links outward appearance to aggression.
• Section 1 of the Act should be repealed alongside a programme of education, a national public information scheme and local community actions.
• Overwhelming evidence from multiple international studies make it clear that breed specific legislation does not protect the public from dog attacks.

I am a Professor and Co-Director of the Centre for Human Animal Studies at Edge Hill University. This submission is based on research I have undertaken and published on Breed Specific Legislation (BSL) in the UK, the Dangerous Dogs Act 1991, my wider academic work which examines the socio-cultural contexts of human-animal relationships and a review of relevant academic literature. In addition to my academic expertise in human-animal studies, I have over 35 years’ experience of training and living with large breeds and mixed-type rescue dogs.

How effective is the Government’s current approach to protecting the public from dangerous dog attacks?

1. Deaths and injuries: The current UK approach has not been effective in protecting the public from dog attacks. Despite the introduction of the Dangerous Dogs Act 1991, reported deaths in England and Wales where the underlying cause was a dog bite have increased in successive decades. According to ONS data released in 2016, between 1981 – 1990 there were 11 reported deaths, in the ten years from 1991- 2000 there were 17 deaths and from 2001 to 2010 there were 32 reported deaths.[i] In the years following the introduction of the Act up to 2015, 63 deaths where the underlying cause was a dog bite were recorded.[ii]

2. The figures for dog bite injuries requiring hospital admission showed an increase of 6.5% to 7,227 in 2014/15 compared with 6,783 recorded in the previous 12 months.[iii] In the ten years from 2005 to 2015 hospital admissions due to dog bites increased 76%.[iv] Children in the 0-9 age range accounted for the highest rates of admission in a twelve months period from March 2014 to February 2015.[v] Children in this age group were more likely to suffer injuries to the head and require oral or maxillofacial surgery treatment.[vi] Geographic variations showed that rates of admissions for dog bites in the period 2014 to 2015 were 2 to 3 times higher in deprived areas.[vii]

3. Costs: Estimates of the annual costs to the NHS for the treatment of injuries resulting from dog bites vary considerably from £3 and up to £10 million.[viii] Hospital Episode Statistics (HES) data reported in 2012 stated that 90% of hospital admissions due to dog bites or strikes were emergencies and patients remained in hospital for an average (mean) time of 2.0 days.[ix] Taking this into account, in addition to the costs of individual patient treatment, dog bites represent a significant burden to the NHS in terms of number of bed days.[x]

4. The costs to police forces for kennelling dogs seized under the Dangerous Dogs Act over a five-year period were reported as £5 million in 2015.[xi] In the financial year 2016-2017 Metropolitan Police records show that 1031 dogs were seized.[xii] The costs to MPS for seizure, kennelling, transportation and veterinary treatment for all dogs seized in 2016-2017 was £1,310,830.[xiii] The ‘heavy costs’ of holding dogs in kennels and the impact on animal welfare are acknowledged in Defra guidance.[xiv]

5. Animal welfare and impacts on dog owners: In addition to substantial costs to the NHS and police forces, the costs in terms of animal welfare, and for the families and owners of seized dogs is considerable. Seized dogs will inevitably spend time in kennels, an environment that is a known psychogenic stressor even for short periods.[xv] Continued exposure to limited social interaction, restricted space and other factors linked to kennelling can result in chronic stress for a dog, compromised welfare, and can lead to behavioural problems.[xvi] A close emotional bond between humans and the dogs they live with is commonplace. Although in legal terms they are property, in many UK households, academic and market research consistently finds that dogs are considered members of the family.[xvii][xviii] Removing a dog from their family can thus have significant welfare and wellbeing consequences for both humans and canines.

6. Influence of media reporting: While there have been amendments to the Dangerous Dogs Act 1991, as the above points make clear, legislation has not been effective in protecting the public from dog attacks. It is important to consider why current legislation cannot adequately deal with dog risk and to consider the original socio-cultural context for the legislative intervention. The Dangerous Dogs Act 1991 was a legislative response to widespread public anxieties in the UK about risks posed by dogs. A key driver of the debate at the time was UK media reporting on dog attacks.[xix] The pressure generated by the mainstream media was acknowledged during the House of Commons Dangerous Dogs debate in May 1991. Although dog attacks were reported across the UK newspapers, popular (tabloid) press articles accounted for a higher proportion of the coverage, the reporting style of which was notably sensationalistic.[xx] With a lack of reliable statistical information on dog bite injuries at the time, media reports were thus used to inform and shape substantially public understanding and the political debate on dog risks.[xxi][xxii]

7. Media stereotypes of pit bulls and their owners were established in the public consciousness. Stereotypes simplify and reduce a social group to a few memorable characteristics that can be used to reinforce forms of social exclusion and oppression. In the case of pit bulls and their owners, the stereotype linked a type of dog and the humans associated with them to violence and aggression, social deviance, dog fighting and drug culture.[xxiii][xxiv] Without an institutional authority (such as the UK Kennel Club) to provide a counter against the media reporting on dog attacks, the demonization of certain types of dog and the stereotypes deployed by the press dominated the public debate. Those pit bull owners whose dogs were well behaved family pets did not have any opportunity to challenge the stereotypes and were given little or no public voice.[xxv]

8. Media involvement in the identification of risk is complex however it is safe to say that the media do play a key role in defining public risks and in the constitution of public knowledge and understanding of risk. In the case of dangerous dogs, the media created a moral panic that focused primarily and erroneously on pit bulls and their owners. It magnified the dangers of pit bulls and their owners to the public. In doing so, the public attention shifted away from statistical or scientifically informed measures of dog risk. One consequence of this situation has been a lack of public understanding about the risks posed by dogs generally.

9. Banned dogs: As my research and that of others has demonstrated, there was a class bias in the media reporting on dangerous dogs which inevitably impacted on the 1991 legislation.[xxvi] The situation was illustrated by the Home Secretary responsible for the introduction of the 1991 Act when he wrote in 1994: ‘There was a danger of over-reaction, with demands to have all dogs muzzled and to put Rottweilers, Dobermans and Alsations in the same category as pit bulls. This would have infuriated the ‘green welly’ brigade. However, the ‘pit bull lobby’ came to my aid by appearing in front of TV cameras with owners usually sporting tattoos and earrings while extolling the allegedly gentle nature of their dogs, whose names were invariably Tyson, Gripper, Killer or Sykes’.[xxvii] The media and the 1991 Act were especially concerned with what were referred to as ‘dogs bred for fighting’ and specifically the Pit Bull Terrier (PBT).[xxviii] Although the legislation covers three other specific breeds or types of dog (Japenese Tosa, Dogo Argetinos and Fila Brazilieros) the focus continues to be on dogs identified as PBTs. For example, the Defra guidance for enforcers of dangerous dogs law includes information on identifying PBT types only.[xxix] When the legislation was written there was no reliable estimate of the number of pit bulls in the UK, no evidence that as a ‘type’ PBTs constituted a specific risk to public safety nor that PBTs in the UK were bred predominantly as fighting dogs. It was primarily media reporting on dog fighting, drug culture and dog attacks, that made pit bulls synonymous with the stereotype of the fighting dog that was thought to pose a special risk to the public.[xxx]

10. Pit bulls were and are not recognised by the UK Kennel Club as a breed. As a result, they do not have a recognised breed standard (a document that is authored by the Kennel Club in conjunction with specialists who have close and well-developed working knowledge of a breed). The definition of a pit bull has therefore relied on two main sources: the unofficial media stereotype and the official adoption of the American Dog Breeders Association standard of confirmation published in 1977. Defra advice states that ‘the law does not require a suspected PBT to fit the description perfectly, it does require there to be a substantial number of characteristics present so that it can be considered ‘more’ PBT than any other type of dog.[xxxi] The lack of quantification (‘substantial’ is not quantified) in relation to number of characteristics and that a dog should be ‘more PBT than other type’ is highly ambiguous and open to subjective interpretation. Moreover, despite noting that ‘there are no photographs provided to assist’ with identification, the guidance does in fact include a photograph on the front cover.[xxxii] This conforms to the stereotypical image used in the popular media which I have referred in published work as ‘the performance of dog aggression’.[xxxiii] It is common for media reports of dog attacks to use stock images and these images in addition to that used on the front cover of the Defra guidance are, albeit inadvertently, a key visual reference for the assessment of PBTs and ‘dog risk’. It is thus important to note that the conventions of media reporting and of illustrating guidance and information materials do have any impact on the public understanding of what constitutes a dangerous dog. This contributes to both misidentification and to the public misunderstanding of ‘dog risk’.[xxxiv]

11. The evaluation of dog risk: There is no reliable evidence that breed or physical conformation predisposes a dog to be aggressive or a risk to the public. Despite this, the identification of a dog under existing legislation relies on the visual assessment of an arbitrary number of ambiguous characteristics. Dog Legislation Officers (DLO) usually undertake these assessments. Dog aggression is complex and is not reducible to visual appearance. Instead, dog aggression is multifactorial and as one study suggests there may be over 100 potential risk factors for human-directed dog aggression.[xxxv] Dog aggression has a communicative function for dogs and therefore it is reasonable to assume that any dog will have the potential for aggression although its expression is undesirable for humans.[xxxvi] As many studies make apparent breed or type is not a major risk factor for the expression of aggression.[xxxvii] Instead, lack of puppy training and socialisation, use of positive punishment or negative reinforcement-based training methods, and context for example are heavily implicated as major risk factors in human-directed dog aggression.[xxxviii] To evaluate a dog as a public risk, or as having the potential to harm or be dangerous based on appearance alone has no evidential basis or rational justification. It is instead ethically unsound, misleading in terms of public safety, has serious welfare implications and its consequences are unnecessarily costly for the tax payer.

What changes, if any, should be made to the current approach and legislation?

12. The current approach needs to be changed to address the flaws and inconsistencies noted above. Breed or type should not be the primary consideration in the assessment of a dangerous dog. Instead, the behaviour of individual dogs should be the primary mechanism by which risk to the public is assessed. Broadly speaking, this should involve assessments by behaviour experts and assessment of the dog’s home environment, training and socialisation. The ability of the human(s) responsible for the dog should also be taken into consideration and a judgement made as to their capacity to keep, care for and be accountable for a dog, with regard to both the dog’s welfare and human safety. Dog’s placed into rescue centres such as those controlled by the RSPCA should be assessed for rehoming suitability by the rescue centre. This would ensure that dogs removed to centres as cruelty cases or as unclaimed strays that would otherwise be destroyed as prohibited types, could be rehomed where such an outcome is possible. In short, Section 1 of the Dangerous Dogs Act should be repealed. Legislation should focus instead on the actions of humans, in particular dog owners and breeders.

How can local authorities and police forces be best supported in reducing the number of dangerous dog-related incidents?

13. Education is key to reducing the number of dog bites and dangerous dog-related incidents. Many risk factors associated with dog aggression are preventable. Early socialisation and puppy training, positive reinforcement/ reward-based training and force-free training are factors that can reduce the potential for aggression. Information for adults on how best to approach and handle dogs is required nationally. A national public information campaign that communicates key messages about dogs and human-dog interaction would be cost effective. It is important to acknowledge that knowledge, awareness and access to information about dogs is subject to geographic variability and socio-economic context. It is vital therefore that need is addressed through community and localised approaches. Co-ordination of community response should draw on collective knowledge and support from local authorities and police forces working in conjunction with puppy socialisation groups, ring craft classes, Kennel Club Good Citizen Dog Schemes and local branches of dog welfare charities and organisations. Education of children at all age ranges but particularly under 10 (when they are most vulnerable to dog attacks that can result in severe trauma, head injuries or death) should be instituted as part of the school curriculum. 

What lessons could the UK learn from other countries dealing with similar issues?

14. Evaluation of the Dutch breed specific legislation found that breed was not a useful indicator of bite risk. Instead the study recommended the preventative measures should focus on better understanding of how to handle dogs.[xxxix] As a result of the study, BSL was abolished in the Netherlands.[xl] A 10-year retrospective study of fatal dog bites in Spain found that the implementation of breed-specific legislation in Spain (1999 and 2002) did not reduce dog bite-related fatalities.[xli] The results of a study on temperament testing in Lower Saxony showed ‘no indication of dangerousness in specific breeds’ and concluded that there was no justification for specific breed lists in the legislation.[xlii] As a consequence of scientific studies, ‘legislation in Lower Saxony was changed, and breed lists were withdrawn’.[xliii] A study of dog breed identification amongst shelter staff in the US found a lack of visual consistency in the identification of pit bull type dogs and concluded that visual identification was unreliable.[xliv] A study of dog attacks in Australia found that there was no evidence to uphold the view that the pit bull terrier is uniquely dangerous. It concluded that breed-specific laws in Australia were not justified.[xlv] A study of breed specific legislation in Canada noted that ambiguity around identification of pit bulls likely results in inaccurate reporting and overrepresentation of pit bulls in bite statistics and that non-restricted dogs are mistakenly identified as banned breeds.[xlvi]

June 2018

5