Dangerous Dogs in Canadian Law
- Jacquelyn Shaw
- Animal Legal & Historical Center
- Publish Date: August 2009
- Place of Publication: Michigan State University College of Law
The partnership between dogs and people is undoubtedly ancient, having existed for anywhere between 10, 000 -100,000 years,  depending on the author and method of study. It is a relationship which has benefitted both parties. As a result, dogs successfully spread to nearly every culture on earth with early humans. People have in turn benefitted in many ways from this relationship with dogs, from herding food animals, to acting as guides, or as companions that improve their owners’ health and longevity. 
In modern North America, most of the approximately 80 million dogs are companion animals.  Dogs first evolved from wolf-like ancestors, and through selective breeding,  they became genetically much tamer than wolves, retaining many puppy-stage characteristics. Yet dogs still possess some wolf-like characteristics, which can be valuable to people, when territorially protecting their family pack and den. Dogs usually see their ‘pack’ as their human owners and their ‘territory’ as their home and the area around it. 
Unfortunately, these same valuable wolf-like traits can lead to unwanted aggressive behaviour in dogs at times. Each year, 4.5 million dog-bites occur in the United States  (500,000 /year in Canada ), suggesting that 1 dog in 16, or roughly6.2% of dogs, may bite humans. People may also suffer fall-related injuries, or even psychological harms, which may result in a fear of dogs strong enough to restrict certain employment. Dogs of all sizes and breeds may bite, although larger dogs may cause more harm. In 1999, direct medical care costs alone exceeded USD$235 million.  While only 3.5% of bites needed surgery,  roughly 1.6 fatal dog bites occur yearly, per population units of the same size, in both the United States and Canada.  In relative terms, serious dog-bites are rare and declining infrequency;  more people may be injured by bicycles.  Nonetheless, dog-bites are a concern, especially because children are the main victims. 
This article explores Canadian legal approaches to dog-related injuries, but will omit discussing ‘Breed-Specific Legislation,’ which prohibits certain dog breeds, such as pit bull terriers. In Canada, two main legal approaches address to dog-bites or other dog-related injuries: the traditional (judge-made) common law and the more recent (government-enacted) statutes. Both approaches use the threat of victim compensation and possible prison sentences to encourage responsible dog control. In severe cases, criminal charges may be laid under a statute not specific to dog-related injuries-- the Criminal Code of Canada-- which employs the threat of imprisonment and public condemnation to encourage responsible dog ownership. 
The first part of this article discusses the traditional common law approaches of ‘scienter’ and ‘negligence,’ and their challenges for dog owners and victims. Next, Canada’s statutory liability approaches and their shortcomings are discussed with respect to the dangerous dog statutes of Canada’s provinces and territories and sections of the federal Criminal Code of Canada. Finally, the article concludes with an analysis of how the statutory schemes have successfully ameliorated much past unfairness in the compensation of dog-bite victims, thereby encouraging responsible owner behaviour. This change comes at a price, however. Some statutes may go too far, potentially causing unfairness even to responsible owners and their dogs, by misdirecting attention away from an owner’s conduct to that of the dog. Ultimately, a statutory or common law approach based on a negligence standard may be the fairest compromise that considers the interests of both responsible owners and innocent human victims.
II. Differences Between Canada and the United States
Canada and the United States share many similarities in dog-bite patterns, but there are also some differences based on geographic and cultural distinctions. For instance, possibly because sled-dog packs are used frequently for transport in Canada’s less accessible and remote north, more fatal bites are reported from such rural areas, compared with the United States.  Canada also has a higher rate of dog-pack over single-dog fatal attacks on people.  It is not clear how often neglect or starvation are involved in attacks on humans. During the warmer months, some sled-dog teams may be turned out and left relatively unattended --other than periodic feeding-- in unpopulated wilderness areas, such as uninhabited islands.  There, dog teams may learn to hunt cooperatively like a wolf-pack,  for food or from boredom. These communal pack hunting experiences may pave the way to predatory behaviour towards humans,  if unwary people encounter the unattended pack while camping, fishing or picking berries. Wolf-dog hybrids with less tame and predictable temperaments may also be more common in remote areas due to interbreeding, accidental or otherwise. 
Furthermore, Canada’s land-mass is largely non-urbanized, with vast areas of undeveloped wilderness, ‘First Nations’ (aboriginal) reserves, and rural farmland, surrounding the half-dozen major urban centres of the south.Researchers observe that, for unknown reasons, dog-bites are more frequent in areas of lower socio-economic status.  Ironically, there may be greater or more vocal public concern over dog-bites in more affluent, densely populated urban areas. Unfortunately, further study of both rural and urban Canadian dog-bite dynamics may be impeded by the fact that Canada --unlike its southern neighbour, the United States-- does not keep consistent, annual records of dog-related injuries. This presents an obstacle to better understanding the issue from a Canadian perspective.
III. Traditional Common Law Approaches to Dog-Bites or Other Dog-Related Injuries
The challenge in assessing Canada’s dog-bite problem is compounded by some variability in application of the law among the provinces and territories. As a result of Canada’s patchwork of differing statutes, in some regions dog-bite cases must be brought under the two common law categories of ‘negligence’ and ‘scienter’, while others employ a combination of statutory and negligence law. The choice of approach is mainly dictated by whether a given province possesses a dog-bite statute or not. In regions with no statute, actions are typically brought under both scienter and negligence. In contrast, where a statute exists, this typically bars the option of bringing a common law action under scienter, but not negligence. This section will describe the application of scienter and negligence to dog-related injuries. These common law approaches, derived originally from English law, are equivalent to civil lawsuits in the United States.
A. Negligence Approaches to Dog-Related Injuries
Negligence means that an owner failed to reasonably consider those who might be at risk of injury from a potentially dangerous dog, by taking proper precautions to prevent injuries to those foreseeably at risk (e.g., by muzzling a dog or keeping it in a fenced yard). Foreseeability of potential harm to others by a dog was an issue in the case of Bates (next friend of) v Horkoff. Therean owner admitted that his German Shepherd was stranger-aggressive and played roughly with his own children. He also knew that the fence separating the dog from the neighbouring daycare had large gaps that let the dog`s head and shoulders through, but he neither fixed the fence nor used the secure dog run on his property. When the dog bit a daycare child, the court found that it was foreseeable that the neighbouring daycare children might be put at risk by the dog, and that the dog`s owner had failed to take reasonable precautions to minimize risks to the local children. He was found liable  in negligence.
In some negligence cases, if a victim was partly to blame for his own injuries, courts have found victims ‘contributorily liable’ (partly responsible) for their own care costs. This may because a bite victim was careless about his own safety, perhaps approaching a dog despite a ‘Beware of Guard Dog’ sign or ignoring other notices of danger, or if the victim deliberately provoked the dog’s aggression. For instance, in Bacon (Litigation Guardian of) v. Ryan, a mother brought her child to visit a family with a Pit Bull that had recently bitten the owner`s son. She knew about the bite incident, but let her child play on the floor near the dog anyway. After her child was also bitten, the mother was held one-third contributorily liable. When a dog in another case, Whelen v. Barlow, bit a drunken man who was shouting at its owner’s wife, the man was found two-thirds contributorily liable for his own injuries. These examples illustrate that under the common law of negligence, a victim’s actions as well as the dog’s and owner’s acts are considered in the legal assignment of blame for dog-related injuries.
B. Scienter Approaches to Dog-Related Injuries
The other common law approach, scienter, is an ancient ‘strict liability’ concept, in which an owner is said to keep a dangerous animal or thing (such as dynamite or a wild animal) ‘at one’s peril’. If a thing known by the owner to be dangerous is kept despite the danger and hurts someone, the owner must compensate the victim for any harm done. The owner cannot use the defence that he made all reasonable efforts (termed ‘due diligence’) to prevent the harm: he is strictly liable. No matter how careful he was, or what precautions he took, he is still be liable for a victim`s injuries if certain conditions (noted below) are met. The British Columbia court in Prasad v Wepruk discussed the application of scienter. There, an elderly newspaper deliverer was mauled by a guard-dog that was usually chained up at its owner`s scrap-yard, but which escaped into the street.  Local residents gave evidence of the chained dog`s aggression towards passersby. Seemingly, the owner knew his dog was aggressive and not only kept it anyway ‘at his peril’ but kept it because it was vicious, to frighten away scrap-yard thieves. The fact that he kept the dog chained to protect innocent passersby was seemingly no defence. He was liable in scienter. 
Scienter has attracted the criticism that it makes it too difficult for victims to obtain compens-ation, even if they sustain severe, life-long injuries. Victims must prove that: a) the dog behaved viciously in the past and b) the owner knew of this. Both may be difficult to prove in court sincea biting dog may have never bitten or acted ‘viciously’ before,  and if it did, it may not have done so in its owner’s presence. In Janota-Bzowska v. Lewis, a guest fell and hurt her hand when her host`s tethered, usually placid Labrador lunged at her. Because the dog had never acted aggressively before, the Court of Appeal overturned the trial court’s finding and concluded that the owner was not liable in scienter.
For a scienter claim to succeed, a dog may not need to have shown a past propensity to do the specific thing complained of presently by a plaintiff, such as biting children or killing livestock. Instead it need only exhibit general aggressive or uncontrolled behaviour. Yet as to the type of aggression needed--viciousness or simply dangerousness of any kind—there remains confusion, which may leave some victims uncompensated. Shelvey v Bicknell involved an appeal where a large Rottweiler knocked a woman down by bumping her legs while it was playing, causing her to hit her head and lose conscious-ness. The dog was friendly, but big and clumsy; it had accidentally knocked down its owner and a child previously. On appeal, scienter was not argued, perhaps because the dog showed no ‘vicious’ history. Yet in another scienter case, Fisher v Liptak, the court stressed that scienter is based on the idea that “anyone who maintains an animal that is known to be dangerous to humans... in any other way does so at his peril” (emphasis added). This suggests that the key trait is dangerousness of any kind, not viciousness. Thus in Shelvey, a scienter argument might have succeeded, if argued, due to the dog’s history of clumsily bumping into people and knocking them down.
In scienter, what counts as ‘proof the owner knew’ of their dog’s dangerousness may confuse victim and owner alike, especially if the dog was gentle with its owners. For example, in Prasad, evidence of owner knowledge was inferred from the owner `s purpose in keeping the aggressive dog to deter intruders; the dog was seen to be affectionate and calm with its owner. Another case, Morsillo v Migliani, involveda German Shepherd biting a local boy; much of the evidence claimed to prove the owners’ knowledge of its viciousness  revolved around their actions of responsibly keeping it in a secure yard and using a choke chain and leash when taking it outside.
In scienter, owners may completely escape liability if it is shown that a victim ‘provoked’ the dog, thereby voluntarily assuming all risk. But what if the provocation is unclear? In some cases, the provocation is obvious and deliberate. For example, an attacking dog may have been tormented by a child victim throwing stones at it.  In other situations, provocation may not have been intended as such by the victim. For example, a stranger startling a sleeping dog by petting it or accidentally stepping on it, the smell of strange dogs on an examining animal control officer, an injured dog’s reaction to those trying to help it, a toddler’s inappropriate play, or perceived threats to an owner or puppies may all provoke a dog to bite. Curious young children, used to harmless cartoons, stuffed toys or petting-zoo babies, may unwittingly provoke or confuse a dog through inappropriate play, or by their small size, high-pitched voices and quick movements. An older case, Richard v. Hoban,sadly still relevant today, involved a young girl who was bitten when she innocently put her arm around the neck of a chained German Shepherd to play, evidently not having been taught appropriate behaviour with strange dogs.
Unfortunately for both dogs and owners, many scienter cases take a human --rather than dog’s eye --view of provocation.  The defendants in Morsillo unsuccessfully argued that the child victim provoked their dog by firing his clicking toy gun at it. Yet the fact the boy was knowingly playing on an allegedly vicious and supposedly ‘terrifying’ dog’s front lawn, a few feet from where it was being walked was not argued as provocation.Dogs do not always share the human legal view of who is an intruder, what seems to be a threat or where their territory ends.  Animal experts can introduce a dog`s perspective to correct presumptions of viciousness, but this adds cost and delay to a trial, and so may be omitted. 
Overall, as in negligence, under the common law of scienter, the actions of the dog-bite victim are considered by courts in calculating legal responsibility. Yet in scienter, the stakes for owner and victim are “all or nothing’ because a victim’s blameworthy, provoking acts can completely cancel the owner’s strict liability for knowingly keeping a dangerous animal, rather than leave the owner only partly liable.
IV. The Statutory Approaches to Dog-Related Injuries
Due to scienter’s problems and the perception of past injustice to bite victims, many Canadian provinces have enacted statutes to make victim compensation more accessible. All threenorthern territories and nine of Canada’s ten provinces have a statute, statutory sections or regulations devoted to dog-related injuries. British Columbia alone has no statute, its Animals Act having been struck down and not replaced.Existing dog-bite statutes vary in their features. Ontario’s Act incorporates a ban on pit bulls, while the three northern territories and Newfoundland/Labrador expressly include wolf-dog hybrids among ‘dogs’. Some place restrictions on the use of Husky or sled-dogs near settled areas, and Saskatchewan’s restrains the keeping of dogs for fighting purposes. All the statutes impose penalties when human beings and domestic livestock (but possibly not always domestic pets) are at risk; in some northern areas, attacks on inanimate ‘property,’ such as food caches also attract penalties. In addition to the provincial and territorial statutes, there is also a federal statute, the Criminal Code of Canada, under which owners may be charged when their dogs attack people.
A. Penalties under the Provincial/ Territorial Dangerous Dog Statutes
All of these statutes and regulations allow a dog that bites (or tries to) to be seized and destroyed, either with or without a judge’s order, possibly at the scene of the incident.Statutory fines for infractions such as bites or ‘running at large’  are imposed in some provinces, in addition to victim compensation, but vary greatly by region. In Nova Scotia, Prince Edward Island and Quebec, and in British Columbia, which has no statute, no additional fines are imposed beyond victim compensation. In several areas, statutory fines exist but are almost trivial: in Alberta $5/day, in New Brunswick a lump sum of between $10-50, in Nunavut and the North-West Territories a maximum of $25 and in Newfoundland /Labrador, up to $200. Yet in other regions, fines may be much more substantial: up to $10,000 in Ontario and Saskatchewan, and in Manitoba, $5,000 for a first offence and $10,000 for subsequent incidents.
Prison terms are an alternative in some regions; longer terms seem to be available where substantial fines exist. The length of such terms can be up to 30 days in Nunavut or the North-West Territories, up to three months in Newfoundland/Labrador, up to six months in Ontario, Saskatchewan, and Manitoba (for a first offence, or for subsequent re-offences, up to one year) and an unspecified length in New Brunswick.
It is unclear why such large discrepancies exist in the penalties for dog-related infractions in different parts of Canada. There may be important regional differences both in victim characteristics and in attitudes towards dogs, especially comparing more urbanized and densely populated regions, such as the province of Ontario, with less affluent, more rural or northerly regions, such as Nunavut or the Yukon. In more urban and prosperous Ontario, there may be a more vocal, better informed and powerful population of potential (or actual) victims able to lobby effectively for stricter laws, following high-profile dog attacks. Less urbanized but strongly agricultural regions such as Saskatchewan and Manitoba, which may also possess powerful lobby groups concerned about potential financial losses from livestock attacks, may have followed Ontario’s example. Another possibility is that in urban Ontario, where people are somewhat disconnected from the natural world including animals, there may be a sense that dogs represent unnecessary health and safety risks to be discouraged, rather than indulged.
By contrast, in the less populous north, where more traditional Aboriginal subsistence lifestyles and a frontier mentality persist, bite victims may be impoverished, dispersed on reserves away from mainstream media, and not well-organized enough to lobby for legislative change. Northern attitudes to dogs may differ from those in the south, and may have changed little from pre-colonial times. In a recent Yukon case, a neglected German Shepherd-Rottweiler cross that bit several people was championed by the local pound and defended legally, being described as ‘a good skookum dog’, a Chinook Aboriginal term meaning a dog valued for its strength and bravery.  In the north, dogs and their associated risks may be seen as a ‘necessary evil’, like motor vehicles elsewhere, given their importance for transport in less accessible regions, for hunting game or for protection of vital winter food caches against marauding wildlife. For such tasks, a dog’s survival instincts, stamina and some aggressiveness may be looked on more with approval and admiration than condemnation in these areas.
B. Dangerous Dog Statutes: Absolute Liability as the ‘New’ Strict Liability Approach
Canada’s statutes seem to have studiously tried to correct the old scienter-based injustices to victims. Yet in the process, some may have gone to the opposite extreme, arguably causing unfairness to responsible owners and their dogs. As in scienter, unreasonable or ‘humanizing’ standards may be required of a dog, under some statutes. Of the 11 existing statutes, only two Acts —those of Prince Edward Island and Ontario-- systematically include the dog’s (as well as the human) perspective  on the circumstances of a bite, for instance, factoring into a judgment details such as whether a dog was injured, teased or protecting its puppies. In Nunavut and the North-West Territories, the contribution of neglect or starvation to a dog’s becoming a safety risk is noted, but this may not excuse an attack. In Saskatchewan, the defence of provocation can, in theory, introduce a dog’s perspective, but the defence is weakened by accepting only deliberate human acts that were intended to provoke the dog. Thus unintentional acts such as tripping over and falling on a dog, or deliberate acts not intended to provoke a dog, such as hugging or petting an unfamiliar dog, will not qualify as ‘provocation’ in Saskatchewan, although they might well seem provocative to some dogs. While in Saskatchewan provocation could, if proven, bar all liability, in Manitoba and Ontario, proof of provocation only reduces the compensation payable to the victim.
Of the 11 Canadian Acts, nine  expressly remove the traditional scienter-based need for victims to prove a dog’s dangerous propensity and owner knowledge of that propensity. Yet three of the nine statutes-- Ontario, Manitoba and Newfoundland/ Labrador-- also remove the need for victims to prove owner fault or negligence, such as a lack of owner due diligence in confining or supervising the dog, making an owner’s disproof of fault irrelevant in those provinces. Thus in these three provinces, an owner may not be able to escape liability even if he undertook all reasonably foreseeable public safety precautions. Of course, to have bitten, a dog must have managed to circumvent such precautions, but this could have been due to a freak, unforeseen event such as a natural disaster, or an intervening act by another person. For instance, pranksters may release a dog, or an illiterate or ill-intentioned person may ignore a ‘Beware: Guard Dog on Duty’ sign , such as when a child climbs over a fence to retrieve a lost ball. However, such possibilities may be irrelevant to whether penalties attach in these provinces, although they probably influence the size of penalties. This risks some unfairness to responsible owners and dogs.
Thus there may be some question as to whether, in Ontario, Manitoba and Newfoundland/ Labrador, the statutes represent a true ‘strict liability’ regime. In Morsillo, the court described the local statute, the Ontario Dog Owner Liability Act, as a ‘strict liability’ approach to dog-bites. The possibility of a prison term may also imply that such statutes are intended to represent only a strict liability approach.  Yet some argue that by neither requiring proof of propensity and owner knowledge, nor owner negligence, these statutes make it nearly impossible for owners to escape liability, regardless of how careful they were. Thus the old scienter-based ‘strict liability’ approach may have been replaced in these three provinces with something closer to ‘absolute liability’.  Usually, absolute liability denotes an activity seen as inherently harmful or risky to society as a whole, which is to be deterred. This does not seem in keeping with dog ownership, which offers society many benefits, including dogs’ companionship, guardianship and other services, and improved owner health and happiness, at relatively low costs.
C. Federal Criminal Liability for Dog Attacks
In very severe attacks, especially those that have attracted significant media attention and public outrage, owners are sometimes charged with ‘criminal negligence’ to symbolize the desire for future deterrance. Despite the similarity in name to ‘negligence,’ criminal negligence actually represents a different legal approach based on Canada’s federal criminal law, rather than being related to the common law of negligence. The statutory definition of criminal negligence involves doing any act or omitting to do a legal duty that shows ‘wanton or reckless disregard for the lives or safety of others.’ Use of the words ‘others’ suggests that only attacks on human beings, rather than pets, livestock or inanimate property, can trigger charges under this statute. In alleged criminal cases, it is the State  rather than the attack victim who lays the charges. In such cases, for some reason, it does not appear to be usual for civil charges to be pursued simultaneously by parties, in addition to the criminal charges laid against the owner.
In this less common  approach to severe dog-attacks, an owner may be charged with contravening either s.221 (‘causing bodily harm by criminal negligence’) or s.220 (‘causing death by criminal negligence’) under the federal Criminal Code of Canada (R.S.C. 1985, c. C-46, s.219 - 221). The criminal law approach may be used to denote public condemnation of an owner suspected of highly irresponsible behaviour and may allow stiffer penalties, in order to deter future incidents. For example, in contrast with some of the lighter penalties existing under the provincial dangerous dog statutes, the maximum penalty for a criminal conviction under s. 221 is a ten-year prison sentence, while those violating s.220 may be sentenced to life imprisonment. A criminal record will follow a convicted owner even after the completion of a prison term, adding to the seriousness of the penalty and its associated social stigma. However, there are much greater obstacles to proving and obtaining a criminal conviction than exist in the non-criminal approaches discussed earlier.
The case of R. v. Baird illustrates the greater difficulty in obtaining a criminal conviction than in finding non-criminal liability for dog-bites. In Baird, the dogs’ owner was charged with criminal negligence when his two Eskimo huskies escaped and severely bit a young boy. One of the dogs had nipped the owner previously while he held a puppy and they had bitten a boy who climbed over the fence surrounding their yard. These incidents suggested aggressive tendencies. Yet Baird was not convicted since he took proper safety precautions (a fenced yard, and chained collars with special locks), and because someone’s deliberate act in releasing the dogs while he was away was considered a significant factor in causing the incident. To be convicted would have required proof of his negligence ‘beyond a reasonable doubt’, a much stricter standard of proof than required in common law cases, where only proof ‘on a balance of probabilities’ is needed. As a result of these greater evidentiary challenges, more commonly, non-criminal negligence law is used to encourage owners to protect the public from their dogs.
Overall, Canada’s non-criminal statutory schemes seem to misplace the focus on a dog’s conduct,  rather than its owner’s, while often ignoring the dog`s perspective on the circumstances of a bite. Experts have long claimed that “there are no bad dogs, only bad owners” or that “focussing on what is at the end of the leash”--the individual dog or breed—is less effective in reducing dog-bites than addressing “what is holding the leash.”  A sensible approach to canine aggression should focus more on owner conduct, which is more amenable to correction by law,  education or other means. This may result in better accountability for dogs’ actions than an approach that simply results in the dog’s destruction, since a bad owner may simply obtain a new dog and treat it with similar irresponsibility in the future.
The trend towards increasing statutory liability in dog-injury cases may have begun with the realization that bite victims were unfairly denied compensation under the old scienter scheme, even for serious injuries. Yet it now seems that, especially in more urban and populous areas, the pendulum may have swung too far in the opposite direction. In more populated regions—unlike in Canada’s north, where there is still a greater connection to and respect for nature--there may also be a subconscious belief that dogs should conform to the same consumer safety guarantees as other inanimate products that people casually acquire and treat as disposable. This is unrealistic and ignores dogs’ needs as living things and the importance of respecting their non-human differences.
Future statutory amendments could include reforms based on a negligence model. Some authors suggest that a negligence standard offers the best compromise between the extremes of victim-unfriendly scienter and owner-unfriendly statutory (strict or absolute) liability.  A negligence-based statutory model would require owners to take reasonable precautions, tailored to their particular dog, such as confining, training or muzzling their dog to minimize foreseeable risks to the public, and would punish owners who did not display reasonable foresight in taking such steps. This resembles both parental responsibility for a child’s conduct and occupiers’ responsibility for the safety of inanimate property to others. 
Other statutory reforms might be modelled on motor vehicle safety in high-risk age-groups, such as periodic refresher courses and testing (for owner and dog alike) before licensing, a demerit point system  to track offending owner, non- trivial penalties or ownership bans, consistent in size among provinces, and earmarking of fine monies for the safety education of children or other potential victims.  As noted, some provinces have drawn a connection between the humane treatment of dogs (providing adequate food, water, shelter, supervision and medical care) and reduced dog-attacks.  Thus public complaints of cruelty, neglect or dog-fighting may be useful ‘sentinel events’ that should be given higher enforcement priority than in the past. 
Ultimately, dogs are comparable to neither dynamite nor plush toys: they are living things that warrant respect for their own sake as well as for human safety. While rare, canine aggression is a serious issue for victims, owners and dogs alike. Future legal changes should aim to strike a balance between protecting public safety and preserving the ancient bond between people and dogs.
 A. Miklosi, Dog: Behaviour, Evolution, and Cognition, (Toronto: Oxford University Press, 2007) at 56; K. Stafford, The Welfare of Dogs (Springer: 2006) at 1, 4; X. Wang & RH Tedford, in P. Jensen ed., The Behavioural Biology of Dogs (CABI International, Oxfordshire: 2007) at 3, 33.
 The US` National Canine Research Council (NCRC) reported 72 million dogs in the United States in 2007, suggesting an estimated 7 million in Canada, with one-tenth the US population; `Media reporting of canine aggression: fact vs fiction`, National Canine Research Center, online at: www.nationalcanineresearchcouncil.com (hereafter NRCR report)
 (US) Centers for Disease Control & Prevention, `Dog bite prevention,` online at: www.cdc.gov/HomeandRecreationalSafety/Dog-Bites/biteprevention.html (hereafter `CDC Report`).
 Thus in total, there are roughly 15 -20 US deaths per year, and 1-2 Canadian deaths per year. S. Avis, Dog pack attack: hunting humans, (1999) 20 Am J. Forensic Med. Pathol. 243 at 243, M. Raghavan, Fatal dog attacks in Canada, (2008) 49 Can. Vet. J. 577 at 577.
 From 1994-2007, dog-bite hospitalization fell 30.3 %, NRCR report; M.E. Lang and T. Klassen, Dog bites in Canadian children: a five-year review of severity and emergency department management (2005) 7 Can. J. Emerg. Med 309 at 313.
 L.A. Epstein, There are no bad dogs only bad owners: replacing strict liability with a negligence standard in dog bite cases, (2006) 13 Animal L. 129 at 129 (hereafter Epstein).
 Children represent over half of dog bites treated and 70% of fatal attacks. American Veterinary Medical Association Task Force on Canine Aggression and Human-Canine Interactions: a community approach to dog-bite prevention. (2001) 218 JAVMA 1732 at 1741. [Hereafter ‘AVMA report’].Children up to 14 years were 3.4 times more likely to bebitten than all other age groups combined, Canadian Hospital Injury Reporting and Prevention Program, Injuries associated with dog bites and dog attacks, 1996, Public Health Agency of Canada, online: www.phac-aspc.gc.ca. More severe bites occur to the head and neck of younger children, who may play inappropriately, hurting dogs, touching their food, acting like prey or simply having their heads closer to a dog’s muzzle; I.L. Reisner, F.S. Shofer & M.L. Nance, Behavioural assessment of child-directed canine aggression, (2007) 13 Injury Prevention 348 at 348. Boys may be 50% more likely to be bitten than girls; J.R. Matthews & K.A. Lattal, A behavioural analysis of dog-bites to children, (1994) 15 Behav. Devel. Pediatrics 44 at 47.
 This occurred in a fatal pack attack on a human family by eight Labrador husky sled-dogs in Newfoundland/ Labrador. S. Avis, Dog pack attack: hunting humans, (1999) 20 Am. J. Forensic Med. Pathol. 243 at 243
 Deliberate hybridization of wolves and sled-dogs may be uncommon because it can lead to counter-productive dominance traits in a dog-team. Yet inherited paw traits unique to wolves may allow hybrid sled-dogs’ feet to avoid gathering ‘snowballs’, permitting longer pain-free running. R. Coppola & R Schneider, Ch. 2 ‘Evolution of working dogs’ in J. Serpell (ed.), The domestic dog: its evolution, behaviour, and interactions with people, (Cambridge University Press, Cambridge: 1995) at 25, 26.
 In one study, 66% of dogs that bit a child had never bitten a child previous and 19% had never bitten any human previously; I.L. Reisner, F.S. Shofer & M.L. Nance, Behavioural assessment of child-directed canine aggression, (2007) 13 Injury Prevention 348 at 348.
While this was resolved under statutory strict liability, the court discussed the evidence initially as if in a scienter case, before concluding that the statute had replaced scienter in that particular province.
 Epstein at 4 of 21 online (Lexis)
This is not to say that the dog’s view should necessarily prevail: if a dog persistently views children as prey, this may not be tolerable in many contexts, especially in densely-populated areas, where many children may be put at risk.
 Epstein at 4 of 21 online (Lexis)
 Only in Ontario are owners not liable if the person bitten was committing a criminal act, but even then, only if keeping the dog was ‘not unreasonable.’
 This is because the possibility of a jail term for an owner (for instance if he cannot pay a fine) requires giving him a chance to free himself by showing he fulfilled all due diligence. If no such option exists, the offence is one of absolute liability. In Canada, imposing prison for absolute liability offences is impermissible: it unconstitutionally violates the right not to be deprived of liberty except according to the principles of fundamental justice, under s.7 of the Canadian Charter of Rights and Freedoms. It seems possible that the confusion over whether, in effect, a strict liability or absolute liability approach exists in these three provinces’ dog-bite statutes, all of which make prison terms an option, might conceivably support a future Charter constitutional challenge to these provinces’ statutes that could result in their being struck down.
 Epstein 3 at 21 online (Lexis). The court in McAllister v. Wiegand also noted that the Ontario statute could be seen as resulting in absolute liability for dog owners.
 Epstein at 2 of 21 online (Lexis).
 T. Follett, ‘Welcome to Dog-Trax,’ online at: http://www.preventdogbites.ca/press.aspx
 Based on a negligence standard rather than strict liability; Epstein at 6-7 of 21 online (Lexis).
 Epstein at 7 of 21 online (Lexis).
 Epstein at 6 of 21 online (Lexis).
 The Centers for Disease Control recommends that children not approach unfamiliar dogs (especially without an adult) or bother them near their food or puppies; if confronted by a menacing dog, do not scream or run (like injured prey) but avoid eye contact, be still like a tree, and if knocked down, roll into a ball and be still like a log. Dog Bite Prevention, Centers for Disease Control, online at: www.cdc.gov/HomeandRecreationalSafety/Dog-Bites/biteprevention.html