Decision regarding December 2006 motion
COURT FILE NO.: 05-CV-295948PDI
SUPERIOR COURT OF JUSTICE - ONTARIO
RE: Catherine Cochrane v. Her Majesty The Queen in the Right of Ontario as represented by the Attorney General
BEFORE: Herman J.
COUNSEL: Clayton C. Ruby, Caroline Wawzonek, for the Applicant
Robert E. Charney, Michael T. Doi, S. Zachary Green, for the Respondent
DATE HEARD: December 21, 2006
E N D O R S E M E N T
 The application in this matter was heard on May 15, 16 and 18, 2006. Eleven days after the conclusion of the hearing, Mr. John Robert Martin was killed as a result of an attack by his dog. The dog was identified as a pit bull by an animal control officer and a veterinarian.
 A few weeks later, I was advised by counsel for the Attorney General of their intention to bring a motion for leave to adduce evidence concerning Mr. Martin’s death. The Attorney General proposes to introduce affidavit evidence from the following: two eye-witnesses to the attack; two police constables who responded to the attack; the animal control officer who impounded the dog; and the veterinarian who identified the dog to the police as a “cross breed or mixed breed with dominant pit bull characteristics.” It is the Attorney General’s position that this evidence is directly relevant to two factual matters that are at issue in this application: the number of human fatalities in Ontario attributable to pit bull attacks; and the ability and willingness of veterinarians to identify dogs as pit bulls.
 The applicant opposes the motion on the basis that the evidence in question does not meet the test for allowing the reopening of a case and the admission of further evidence. If, however, I agree to allow the Attorney General to adduce the further evidence, the applicant seeks to produce evidence in reply. The reply evidence is: the report of the animal behaviourist who was commissioned by the Coroner’s office that investigated the death; the report and record of the veterinarian who euthanized the dog; and a copy of the Trempe Inquest Recommendations and Report (a 1999 Coroner’s Inquest) that were part of the Coroner’s file in the investigation into Mr. Martin’s death.
 The Attorney General accepts that the evidence with respect to the identification of the dog is proper reply evidence but objects to the introduction of a copy of the Trempe Inquest Recommendations and references to the Trempe Inquest.
The Test for the Admission of New Evidence
 Until judgment is entered, a judge has the discretion to reopen the case after the conclusion of evidence in order to admit further evidence. The overriding consideration is, “…that a miscarriage of justice is to be avoided” (Gateway Realty Ltd. v. Arton Holdings Ltd.,  N.S.J. No. 338 (N.S. Sup. Ct). at 2 (QL). The discretion to reopen the trial should be exercised, “…‘sparingly and with the greatest care’ so that ‘fraud and abuse of the Court’s processes’ do not result” (671122 Ontario Ltd. v. Sagaz Industries Canada Inc., 2001 SCC 59 (CanLII),  2 S.C.R. 983 (S.C.C.) at 1010).
 In cases where reasons for decision have been delivered but judgment has not yet been taken out, courts will apply the following two-part test:
First, would the evidence, if presented at trial, probably have changed the result?
Second, could the evidence have been obtained before trial by the exercise of reasonable diligence?
(671122 Ontario Ltd. v. Sagaz Industries Canada Inc., 2001 SCC 59 (CanLII),  2 S.C.R. 983 at para. 20; Scott v. Cook,  2 O.R. 769 (H.C.J.).
 The Attorney General submits that this test should be modified in cases where leave to admit further evidence is sought prior to the release of the reasons for decision because it is not possible to determine whether the evidence would “probably have changed the result” before the decision is rendered. In such cases, it proposes that the test be that the moving party must demonstrate that the evidence goes to important matters that may affect the result (Crawford (Litigation guardian of) v. Penney,  O.J. No. 116 (Sup. Ct. J.) at paras. 10 and 18-21).
 The Attorney General further submits that the discretion to admit new evidence is particularly important in constitutional cases. As held by the Supreme Court in Irwin Toy Ltd. v. Quebec (Attorney General), 1989 CanLII 87 (S.C.C.),  1 S.C.R. 927, the government should draw upon the “best evidence currently available” in attempting to establish that a legislative objective is pressing and substantial or that a legislative measure is proportionate to the objective.
 Because the decision on this motion is being released at the same time as the decision on the application, it is possible for me to answer whether the evidence in question would “probably have changed the result”.
 Both parties agree that, regardless of which test is applied, the overriding principle to be applied in the exercise of my discretion is whether a miscarriage of justice would occur if I do not allow the Attorney General to adduce this additional evidence.
(i) Did the Evidence exist before the Hearing?
 There is no question but that the evidence concerning the fatal attack on Mr. Martin did not exist before the hearing. So too, the evidence of the animal control officer and the veterinarian who identified the dog responsible for the attack did not exist before the hearing. At the time of the hearing, the evidence was that there had been one reported death that was attributed to a pit bull. Unfortunately, there are now two.
 To the extent that the evidence is being adduced to address the issue of vagueness, in particular, whether veterinarians are able to identify pit bulls, as defined in the Dog Owners’ Liability Act, it is not, in my opinion, evidence that did not exist before the hearing. While the evidence with respect to the identification of this one particular dog was not previously available, there was ample opportunity for both parties to present evidence with respect to the ability and willingness of veterinarians to identify dogs as pit bulls.
 However, to the extent that the evidence is directed to the issue of constitutional overbreadth, and, in particular, the dangerousness of pit bulls, it is evidence that did not exist before the hearing.
(ii) Would the Evidence have changed the Result?
 I have delivered my reasons on the application at the same time as the reasons on this motion. Evidence that there was an additional death attributable to a pit bull would not have changed the result in my decision, given my conclusion that there was already sufficient evidence to establish a “reasoned apprehension of harm”.
 Even were I to apply the less stringent standard, that is, that the evidence may have changed the result, I would conclude that the new evidence did not have the potential to change the result. The statistics on reported fatal attacks attributable to pit bulls were one piece of evidence amongst many other pieces of evidence concerning the alleged dangerousness of pit bulls. The new evidence would not have made a difference in my overall assessment.
 I am also satisfied that a refusal to admit this new evidence will not result in a miscarriage of justice.
 The Attorney General’s motion for leave to adduce further evidence is therefore dismissed. It is unnecessary to consider the applicant’s motion to adduce evidence in reply.
 If the parties are unable to agree on the disposition of the costs of this motion, they may provide submissions at the same time that they provide submissions on the costs of the application.
DATE: March 23, 2007
Decision regarding Dog Owners' Liability Act
COURT FILE NO.: 05-CV-295948PD1
SUPERIOR COURT OF JUSTICE
B E T W E E N:
- and -
Attorney General of Ontario
Clayton C. Ruby, Breese Davies, for the Applicant
Robert E. Charney, Michael T. Doi, Zachary Green, for the Respondent
HEARD: May 15, 16 and 18, 2006
 The applicant, Catherine Cochrane, seeks a declaration that the provisions of the Dog Owners’ Liability Act, R.S.O. 1990, c. D.16 that deal with pit bull dogs violate ss. 7 and 11 (d) of the Canadian Charter of Rights and Freedoms and are ultra vires the provincial government.
 There is much debate as to the wisdom of enacting legislation that targets a particular breed of dogs. Many argue that it is the dog owner or the dog’s environment, not the dog’s breed, that is the primary determinant of whether a particular dog is dangerous. However, it is not my task to determine whether the legislature made the correct policy choice in restricting the ownership of pit bulls. Rather, the issue is whether the provisions in question are constitutionally valid.
 The basis for the applicant’s claim is three-fold:
(i) the pit bull provisions are unconstitutionally overbroad and vague;
(ii) the provision that allows a document from a veterinarian stating that a dog is a pit bull within the meaning of the Act to stand as proof that the dog is a pit bull results in trial unfairness and offends s. 11 (d) of the Charter; and
(iii) the designation of a particular breed of dogs is ultra vires the provincial government because the designation of breeds is addressed in a federal statute, the Animal Pedigree Act, R.S.C. 1985, c. 8 (4th Supp.).
 This application was heard on May 15, 16 and 18, 2006. I reserved my decision at the conclusion of the hearing. The Attorney General of Ontario subsequently brought a motion seeking leave to introduce new evidence. That motion was heard on December 21, 2006. I have dismissed that motion and have released my reasons for doing so at the same time as the release of this decision. I therefore rely solely on the evidence that was introduced at the original hearing of the application.
 The Dog Owners’ Liability Act (DOLA) establishes a scheme that applies to owners of all types of dogs. It provides that the owner of a dog is liable for damages resulting from a bite or attack by that dog. A dog owner is required to exercise reasonable precautions to prevent the dog from harming or posing a menace to persons or domestic animals. Contravention of the Act or regulations constitutes an offence. The court may order that the dog be destroyed or that the owner take measures for more effective control of the dog.
 In 2005, the Ontario legislature added breed-specific provisions to the Act that targeted pit bulls. This was done in response to several brutal attacks by dogs that had been identified as pit bulls. It is these provisions that are being challenged.
 The Act defines “pit bull” to include: a pit bull terrier; a Staffordshire bull terrier; an American Staffordshire terrier; an American pit bull terrier; and “a dog that has an appearance and physical characteristics that are substantially similar” to these dogs.
 The Act places various restrictions on the ownership and transfer of pit bulls. No person shall own, breed, transfer, abandon, allow to stray, import or train a pit bull for fighting.
 An individual may, however, own a “restricted pit bull”. A restricted pit bull is a pit bull that was owned by an Ontario resident on the day that the amendments came into force or that was born in Ontario within 90 days after the amendments came into force. In such a case, certain restrictions apply, including a requirement that the dog be sterilized and that it be muzzled and leashed in a prescribed manner.
 As with the Act’s general provisions, an individual who contravenes the pit bull-specific provisions is guilty of an offence. The individual is liable on conviction to a fine of not more than $10,000 or a term of imprisonment of not more than six months or both.
 At a hearing to determine whether an individual charged with an offence has breached the Act, the prosecution has the onus to prove its case beyond a reasonable doubt. A document that is purported to be signed by a member of the College of Veterinarians of Ontario stating that the dog in question is a pit bull is receivable in evidence as proof that the dog is a pit bull, in the absence of evidence to the contrary.
The Applicant and Her Dog
 The applicant, Ms. Cochrane, has a dog, Chess. Ms. Cochrane adopted Chess from the Humane Society. The Humane Society’s documents identified Chess as a Staffordshire terrier cross.
 Ms. Cochrane stated in her affidavit that Chess is a playful, affectionate and friendly dog, who is well loved by Ms. Cochrane’s family and friends. According to Ms. Cochrane, Chess has become a “fixture in my neighbourhood”. Ms. Cochrane likes to take Chess to an off-leash area in the park where she plays with other dogs. While Ms. Cochrane is at work, a dog walker takes Chess out for a walk with five to six other dogs of various sizes and breeds.
 Ms. Cochrane and Chess attended private dog training classes for a period of time. Ms. Cochrane reports that Chess has never been involved in any aggressive behaviour. The dog walker has told Ms. Cochrane that Chess is always well behaved.
 Ms. Cochrane is concerned that the provisions of the Act will restrict Chess’ opportunities to socialize in the community and with other dogs. She will not be able to take Chess to an off-leash park and let her play with other dogs. She is also concerned that the provisions will make it difficult for the dog walker to walk Chess with other dogs, since Chess will have to be on a short leash and muzzled. If the dog walker cannot walk Chess, the dog will have to be left in Ms. Cochrane’s apartment for nine to ten hours at a time while Ms. Cochrane is at work.
Injuries Caused by Dogs Identified as Pit Bulls
 There is no doubt that individuals in Ontario have been seriously injured by dogs that were identified as pit bulls. Affidavits filed by the respondent relate instances of extremely savage attacks that resulted in significant and permanent injuries. In several of these cases, the individuals reported that the dog attacked without warning and without provocation.
 C.H., for example, sustained severe life-threatening injuries after being attacked by three pit bulls in Brantford, Ontario. She had to undergo three surgical procedures. She continues to suffer from double vision, foot drop and will likely bear permanent scars on her face, arms and legs. She is unable to walk properly. Portions of her hair are permanently missing as a result of the dogs tearing her scalp. Portions of flesh are missing from her leg. Dr. Carolyn Mary Levis, a plastic reconstructive surgeon, said that C.H.’s injuries were the worst injuries sustained by a dog attack victim she had ever seen.
 D.W., a letter carrier in Dresden, Ontario, reported that she was mauled and seriously injured by two pit bulls that attacked her as she was delivering mail.
 G.C. of Ottawa related that three pit bulls viciously attacked him and his two-year old son as they were walking home. His neighbour, R.M., was also seriously injured when he tried to stop the dogs. The dogs had, apparently, attacked other children a month prior to this incident. The dogs’ owner had been directed to muzzle and control the dogs but had not done so. The owner was charged and convicted of criminal negligence.
 Another parent, L.E. of Toronto, stated that her five-year old child sustained serious injuries when she was mauled by a pit bull. Having been assured by the owner of the dog that it was friendly and could be patted, L.E. let her daughter pat the dog, after which the dog lunged and attacked the child. Her daughter suffered serious injuries to her face, underwent five hours of surgery and received 300 stitches.
 There were several affidavits from police officers who reported having confronted and shot pit bulls in the course of their duties. In some instances, the police officers were responding to reports that a dog was attacking someone. In other instances, individuals suspected of criminal conduct used pit bulls to aggressively threaten police officers.
Are the Provisions Overbroad or Vague?
 The applicant does not challenge the government’s legitimate interest in taking steps to prevent dog attacks on individuals. Rather, the applicant submits that the means the government is using to accomplish that objective are overbroad and vague and therefore contrary to s. 7 of the Charter.
 The parties agree that there is no Charter right to own or breed a dog or to allow a dog to be without a muzzle or off-leash in public. Rather, the Charter, in particular s. 7, is triggered because a term of imprisonment may arise from a conviction under the Act. Section 7 of the Charter states that every person has the right to “life, liberty and security of the person and the right not to be deprived thereof except in accordance with the principles of fundamental justice”. Because an individual may receive a term of imprisonment for a contravention of DOLA, the provisions of that Act must conform to the principles of fundamental justice.
 Courts have traditionally given considerable latitude to a legislature in deciding what means it wishes to employ in accomplishing legitimate objectives. As noted by Gonthier and Binnie JJ. in the reasons of the majority in R. v. Clay, 2003 SCC 75 (CanLII),  3 S.C.R. 735:
The task of the Court in relation to s. 7 of the Charter is not to micromanage Parliament’s creation or continuance of prohibitions backed by penalties. It is to identify the outer boundaries of legislative jurisdiction set out in the Constitution. Within these boundaries, it is for Parliament to act or not to act. …The Court’s concern is not with the wisdom of the prohibition but solely with its constitutionality (at para. 4).
 Similar sentiments were expressed by Dickson C.J. in Reference re ss. 193 and 195 (1)(c) of the Criminal Code (Man.), 1990 CanLII 105 (S.C.C.),  1 S.C.R. 1123:
The issue is not whether the legislative scheme is frustrating or unwise but whether the scheme offends the basic tenets of our legal system (at 1142).
 It is a principle of fundamental justice that a law should be neither vague nor overbroad. A law may be contrary to s. 7 of the Charter if it is so vague that it does not give notice to individuals of the conduct that is permitted or prohibited. A law may also be contrary to s. 7 if it is overbroad in its application, that is, if it uses means that are broader than necessary to accomplish a legitimate objective.
 The applicant’s arguments that the provisions are overbroad and vague relate to two different factual matters. The basis of the overbreadth argument is that the restrictions on all pit bulls are too broad because pit bulls are not inherently dangerous dogs and only a small number of pit bulls are dangerous. The vagueness argument focuses on the difficulty of determining whether a particular dog is a “pit bull” as defined in the Act.
 Overbreadth and vagueness are two concepts that may be applied separately or may be interrelated. Legislative provisions may be clear but overbroad. Alternatively they may be so vague that their result is overbroad (R. v. Zundel 1987 CanLII 121 (ON C.A.), (1987), 31 C.C.C. (3d) 97 (Ont. C.A.) at 125-26).
 Cory J., citing the proposition in Zundel with approval, added further to the relationship between the two concepts in R. v. Heywood, 1994 CanLII 34 (S.C.C.),  3 S.C.R. 761:
Overbreadth and vagueness are related in that both are the result of a lack of sufficient precision by a legislature in the means used to accomplish an objective. In the case of vagueness, the means are not clearly defined. In the case of overbreadth the means are too sweeping in relation to the objective (at 792).
Are the Provisions Overbroad?
 There is no disagreement as to the legitimacy of the objective of this legislation, which is the protection of society from harm, and in particular, the protection of the public from attacks by dogs. The issue, rather, is whether the means the legislature has chosen, that is the regulation of all pit bulls, are disproportionate to that objective.
 Are the provisions disproportionate or too sweeping in their reach? This question can be divided into two interrelated parts:
(i) Is there a rational basis for legislating and restricting the ownership of pit bulls?
(ii) Are the means disproportionate to the objective?
 The Attorney General takes the position that the evidence establishes the pressing nature of the state’s interest in protecting the community from the harm caused by pit bulls. The applicant maintains that there is no rational connection between the legitimate objective of protecting society from harm and banning pit bulls.
Are Pit Bulls Dangerous?
 This is a highly contentious issue, characterized by strongly held opinions on both sides of the debate. It is the applicant’s position that many, if not most, pit bulls will never be involved in an attack; there are other breeds of dogs that are equally, if not more, dangerous; and a dog’s breed is not the primary determinant of dangerousness. The respondent, however, contends that pit bulls pose a particular threat to public safety.
 Dr. Lehr Brisbin has a doctorate in zoology. He has been qualified as an expert in U.S. courts in the area of canine behaviour, genetics, anatomy and training in relation to American pit bull terriers and other breeds of dogs used in hunting and law enforcement. He has also owned and trained three American pit bull terriers. He stated that he is not aware of any scientific study that identifies physical, physiological or behavioural differences between American pit bull terriers and other breeds that would indicate that American pit bull terriers present more of a risk to public safety than other types of dogs.
 Dr. Brisbin referred to statistics published by the American Temperament Test Society that indicate that over 83% of American pit bull terriers and American Staffordshire terriers tested passed the temperament test administered by the organization. The pass rate for American pit bull terriers was higher than the pass rate for border collies, King Charles spaniels and English sheepdogs and almost identical to the rate for golden retrievers.
 Dr. Irene Sommerfeld-Stur has a doctorate degree in veterinary medicine and is an Associate Professor at the Institute of Animal Breeding and Genetics at the University of Veterinary Medicine in Vienna, Austria. She has conducted extensive research on canine genetics and has given her opinion on breed-specific legislation to provincial governments in Austria. She stated that there are no scientifically valid studies that establish that certain breeds of dogs have more bite strength or higher pain tolerance than other breeds. Further, there is no objective scientific evidence that the danger posed by an individual dog is primarily caused by its breed affiliation. Rather, in her opinion, the danger of an individual dog is, in large measure, a product of that dog’s environment.
 Dr. Karen Delisle is a veterinary technician in West Virginia who has conducted extensive research on dog bite incidents. In particular, she has studied fatal dog attacks. She has identified three factors that play a critical role in canine aggression towards humans: function of the dog (e.g. whether the dog is acquired for fighting); owner responsibility (e.g. whether the dog is allowed to roam loose; whether the dog is neglected or abused); and reproductive status of the dog (e.g. unaltered male dogs; bitches with puppies).
 In an article entitled “The ethology and epidemiology of canine aggression”, Dr. Randall Lockwood indicated that the multiplicity of factors involved in dog bites “…makes it difficult and often meaningless to base predictions of a particular animal’s aggressive behavior on a single characteristic, such as breed”. Dr. Lockwood also noted that the dogs of choice for persons who want to own aggressive dogs have changed in the last 20 years, “…from German shepherds, to Dobermans, to pit bulls, to Rottweilers to a current surge in problem wolf-dog hybrids”.
 The authors of a study on fatal human attacks by dogs in the United States identified several interacting factors that affect a dog’s propensity to bite, including “…heredity, sex, early experience, socialization and training, health (medical and behavioral), reproductive status, quality of ownership and supervision, and victim behavior”.
 Dr. Tim Zaharchuk presented evidence on the situation in Canada. Dr. Zaharchuk is an Ontario veterinarian and was the president of the Ontario Veterinary Medical Association (OVMA) in 2004. He referred to several studies regarding dog bites and dog attacks in Canada. A study in 1996 found that pit bull terriers accounted for 4% of reported dog bites in Toronto, ranking ninth on the list of identified breeds. Another study conducted by the Canadian Hospitals Injury Reporting and Prevention Program examined dog bites reported at eight hospitals between August 1995 and August 1996. The most common offenders were German shepherds, cocker spaniels, rottweilers and golden retrievers.
 Dr. Zaharchuk also reported that there have been 23 reported human fatalities in Canada due to dog attacks, involving a total of 55 dogs. Only one of those dogs, an American Staffordshire terrier, would fall within the definition of “pit bull” in the Act.
 Thomas Skeldon is the Chief Dog Warden for Lucas County in Toledo, Ohio. He disagrees with the proposition that pit bulls are not aggressive dogs. Mr. Skeldon manages a dog control department and supervises a staff of twenty-one dog wardens. He has frequently appeared in court in prosecutions for dog-related offences and has testified in approximately 50 cases involving pit bulls. In each case, he was qualified by the court as an expert witness in the area of dog identification, behaviour and handling. Mr. Skeldon stated that he has dealt with thousands of pit bulls. He has observed highly aggressive behaviour that he says is unique to pit bulls and not exhibited by any other type or breed of dog. Pit bulls have posed a high risk of serious injury to law enforcement officers since they have, according to Mr. Skeldon, become the guard dogs of choice for criminal elements.
 Dr. Alan Beck is a professor of animal ecology and director of the Center for the Human-Animal bond in the School of Veterinary Medicine, Purdue University, West Lafayette, Indiana. He too disagrees with those who take the position that pit bulls are not inherently dangerous dogs. It is his opinion that pit bulls are a recognized danger to public safety. While he agrees that there are many contributing factors that account for dog attacks, he disagrees with the applicant’s witnesses that breed plays no role. In particular, he strongly disagrees with the statements of Dr. Brisbin and Dr. Irene Sommerfeld-Stur that there are no studies that indicate that pit bull terriers are more dangerous than any other breed.
 One of the studies that Dr. Beck referred to was a report on dog bite-related fatalities in the United States from 1979 through 1988. The report found that pit bull breeds were involved in 41.6% of 101 deaths where the breed of dog was reported, by far the greatest proportion attributable to a particular breed. However, the authors of that report expressed caution with respect to the reliability of the data: the term “pit bull” has been used to describe a variety of pit bull terrier and bull terrier breeds; it has been suggested that any short-haired, stocky dog may have been identified as a pit bull; and public attention to pit bulls may have led to the overreporting of pit bull incidents relative to other breeds. Another problem with statistics on dog bites is that there are no reliable statistics on the population of particular breeds of dogs. Thus, it is not possible to determine what proportion of dogs of a particular breed are involved in this behaviour. Notwithstanding these concerns, the authors concluded that “…the pit bull excess in deaths is real and increasing”.
 In a subsequent study, it was reported that pit bull-type dogs were involved in approximately one third of dog bite related fatalities in the United States from 1981 through 1992. Pit bull-type dogs and rottweilers accounted for 67% of dog bite related fatalities between 1997 and 1998. Since it is extremely unlikely that these breeds accounted for 67% of dogs in the United States during that period of time, the authors concluded that “…there appears to be a breed-specific problem with fatalities”. 
 Authors of another report on pit bulls concluded that pit bulls account for a disproportionate number of fatal attacks and serious injuries but acknowledged the problems of overreporting where pit bulls were concerned. They noted the tendency to label any stocky short-haired animal involved in an attack as a pit bull. They also expressed the opinion that one of the reasons that pit bulls account for a disproportionate number of attacks may be that, because of their reputation as mean dogs, a disproportionate number of these dogs belong to the type of dog owner who is likely to exercise less responsibility for their care and supervision.
 Dr. Delisle stated that, based on her research, there appeared to be an overreporting of dog attacks when the breed involved was suggested to be a pit bull or pit bull-type dog and that the initial assumption regarding the dog’s breed was often mistaken. There was, in her opinion, a corresponding underreporting of attacks by dogs of other breeds.
 The applicant questions the weight that should be placed on the studies referred to by the respondent since they are all based on the experience with pit bulls in the United States. There was no evidence before me as to why the behaviour of pit bulls in Canada might be different than that of pit bulls in the United States.
Are There Alternative Means?
 The applicant accepts that preventing dog attacks on humans is a legitimate objective. She submits, however, that there are both less restrictive and more effective means to address the problem. The Attorney General contends that there are no alternatives to breed-specific legislation that would meet the legitimate objective of reducing the risk of pit bull attacks.
 Is there a way to target dangerous dogs without regulating an entire breed? Is it possible, for example, to identify and target dangerous animals before they are involved in an attack?
 Although the authors of a report on the breeds of dogs involved in fatal human attacks in the United States between 1979 and 1998 concluded that there was a relationship between breeds and fatalities (pointing, in particular, to rottweilers and pit bulls), they did not recommend breed-specific legislation. Instead they recommended that individual dogs and owners be regulated on the basis of their behaviour, noting that, “…our reading of the fatal bite reports indicates that problem behaviors (of dogs and owners) have preceded attacks in a great many cases and should be sufficient evidence for preemptive action”.
 In his article, Dr. Lockwood recommended that legislation emphasize responsible and humane ownership of genetically sound animals, as well as the responsible supervision of interactions between children and animals.
 The authors of “Dog Bites in Urban Children” recommended that there be efforts to identify a potentially dangerous dog ahead of time. They noted preexisting circumstances such as a prior unprovoked serious injury to a human or animal or being trained for dog fighting. They recommended a “conscientious animal ownership plan” that would include a reporting system to keep track of breed ownership, dog bite injuries and aggressive training or behavior.
 Another report recommended that efforts to address the problem should focus on dog owners. People should know that previous signs of aggression may be predictive of a severe attack; if a dog displays signs of aggression, owners should be encouraged to seek help from specialists; and potential victims should be educated in dog-bite avoidance.
 A problem with these various approaches is that their effectiveness depends to a significant extent on conscientious dog owners. It is questionable how effective such measures would be in view of the evidence that there are owners who choose to own pit bulls precisely because of their reputation as aggressive dogs.
 Another problem with the recommendations is that they depend on the ability to identify a dangerous dog before the dog attacks. There is evidence that pit bulls may attack without warning or provocation. If that is the case, it may not be possible to identify a potentially dangerous dog in advance of an attack. There also may be nothing that a potential victim can do to prevent such an attack.
 In a paper entitled “Dos and Don’ts Concerning Vicious Dogs”, the authors note that pit bulls are “unquestionably the most dangerous and unpredictable”. Thomas Skeldon, the Chief Dog Warden in Toledo, Ohio, stated that there are pit bulls that have been friendly pets and yet pulled children out of their strollers and killed them.
 However, Dr. Brisbin referred to the 83% pass rate of a temperament test by American pit bull terriers and American Staffordshire terriers and indicated that a high passing rate for a breed suggests that the breed is highly predictable in its temperament. In cross-examination, Dr. Zaharchuk stated that there are behavioural characteristics that are easily identified long before anybody gets bitten.
 If a particular breed of dog is, in fact, more dangerous, can that danger be reduced or eliminated through training? While there was evidence that training and socialization can be factors in the dangerousness of dogs, there was no evidence as to whether a pit bull can be trained not to be dangerous, for example, by attending obedience school.
 Breed-specific legislation exists in many jurisdictions in Canada and the U.S., often at the level of a municipal by-law or ordinance. Although there are different opinions as to the wisdom of breed-specific legislation, I was not referred to any studies on the effectiveness of such legislation in reducing either the incidence or the severity of dog attacks. The authors of a report on dog-bite fatalities in the United States, published in 2000, stated that they were unaware of any formal evaluation of the effectiveness of such legislation.
 Given this evidence, is a law that restricts the ownership of all pit bulls overbroad? An overbroad law uses means that are “too sweeping in relation to the objective” (R. v. Heywood at 792). As articulated by Cory J. in Heywood:
In considering whether a legislative provision is overbroad, a court must ask the question: are those means necessary to achieve the State objective? If the State, in pursuing a legitimate objective, uses means which are broader than necessary to accomplish that objective, the principles of fundamental justice will be violated because the individual’s rights will have been limited for no reason. The effect of overbreadth is that in some applications the law is arbitrary or disproportionate (at 792-793).
 In determining whether the means are too sweeping, the court needs to consider whether there are narrower, alternative means that could be used to accomplish the same objective (R. v. Heywood).
 In R. v. Demers, 2004 SCC 46 (CanLII),  2 S.C.R. 489 the court addressed the validity of provisions dealing with persons found unfit to stand trial. Iacobucci and Bastarache JJ., writing for the majority, concluded that the law was overbroad because, “the means chosen are not the least restrictive of the unfit person’s liberty and are not necessary to achieve the state’s objective” (at 515).
 At the same time, the means chosen by the legislature are to be given a measure of deference. A judge should not intervene just because she or he might have chosen a different means.
 In R. v. Clay, the Supreme Court, in considering the prohibition against marihuana, noted the relationship between overbreadth and arbitrariness. There must be a “rational basis” for extending the prohibition to all users of marihuana. The court concluded that the evidence showed that a narrower prohibition would not be effective because the members of at least some of the vulnerable groups and chronic users could not be identified in advance. Furthermore, the effect on users in the acute phase established “…a rational basis for extending the prohibition to all users…” (at para. 40).
 In R. v. Malmo-Levine, 2003 SCC 74 (CanLII),  3 S.C.R. 571 the majority opinion of the Supreme Court articulated the extent of harm necessary to justify a legislative prohibition:
We do not agree…that harm must be shown to the court’s satisfaction to be “serious” and “substantial” before Parliament can impose a prohibition. Once it is demonstrated, as it has been here, that the harm is not de minimis, or…not “insignificant or trivial”, the precise weighing and calculation of the nature and extent of the harm is Parliament’s job. …A “serious and substantial” standard of review would involve the courts in micromanagement of Parliament’s agenda. The relevant constitutional control is not micromanagement but the general principle that the parliamentary response must not be grossly disproportionate to the state interest sought to be protected…(at para. 133).
Application of the Law to the Evidence
 In determining whether legislation that specifically targets pit bulls is overbroad, it must be determined whether the provisions are “grossly disproportionate” to the objective. The parties accept that the objective of the DOLA provisions is legitimate. It is the means that are in question.
 There is substantial evidence that a number of dogs identified as pit bulls have been involved in very serious attacks on a number of Ontario residents. There is also substantial evidence that many pit bulls are warm, affectionate animals and will likely never be involved in a dangerous incident. There is no doubt, as well, that other kinds of dogs have also been involved in serious attacks on individuals.
 There is conflicting evidence as to whether pit bulls are inherently dangerous dogs. While the experts agree that there are many factors that may account for a particular dog’s dangerousness, there is strong disagreement as to whether a dog’s breed is a significant factor in determining its dangerousness.
 How is the court to deal with conflicting opinions and evidence? Do I need to make a finding as to whether breed is a factor in determining a dog’s dangerousness or whether pit bulls are dangerous dogs?
 In Harper v. Canada (A.G.), 2004 SCC 33 (CanLII),  1 S.C.R. 827 the court considered limitations on election spending. Bastarache J. dealt with the problem of inconclusive or competing social science evidence:
The legislature is not required to provide scientific proof based on concrete evidence of the problem it seeks to address in every case. Where the court is faced with inconclusive or competing social science evidence relating the harm to the legislature’s measures, the court may rely on a reasoned apprehension of that harm (at para. 77).
 Similarly, in R. v. Sharpe, 2001 SCC 2 (CanLII),  1 S.C.R. 45, a case involving the prohibition of the possession of child pornography, McLachlin C.J., writing for the majority, stated that the applicable standard was a “reasoned apprehension of harm” (at para. 85).
 In considering whether a prohibition was overbroad, the court in R. v. Malmo-Levine, required a demonstration that the harm not be “insignificant or trivial”.
 As in Harper v. Canada (A.G.), there is inconclusive and competing evidence in the case at hand. However, conclusive evidence is not required before a government can take action. It is also not necessary for the court to resolve the conflicting evidence. There is, in my opinion, sufficient evidence to conclude that the legislature, in enacting these provisions, had a “reasoned apprehension of harm” concerning the dangerousness of pit bulls.
 One of the applicant’s arguments is that there are other dogs that are as dangerous or more dangerous than pit bulls. In R. v. Malmo-Levine, the court considered the legislature’s choice to criminalize the use of marihuana, while failing to criminalize the consumption of tobacco and alcohol, which are also harmful substances. Gonthier and Binnie JJ. noted that Parliament “ …does not lose [its] jurisdiction just because there are other substances whose health and safety effects could arguably justify similar legislative treatment” (at para. 139). Thus, a law that targets one breed of dogs is not overbroad because it has failed to regulate other dog breeds that may be as dangerous or more dangerous than the breed it has regulated.
 The next part of the inquiry is a consideration of whether the means chosen by the legislature are disproportionate to the objective. This inquiry has two aspects: is it legitimate to restrict the ownership of all pit bulls, although only a minority may, at some point, exhibit dangerous behaviour; and, are the restrictions themselves disproportionate to the objective?
 The first question, that is, whether it is legitimate to restrict the ownership of all pit bulls relates, in part, to the question of whether it is possible to identify a dangerous pit bull ahead of time. The applicant accepts that it is legitimate for the legislature to craft a response that does not require the dog to have been involved in an attack before the state can act. Is it possible to identify a dangerous dog before it attacks? Are there less restrictive means that could target fewer dogs? Would it be possible, for example, to permit owners of pit bulls to escape the restrictions in the legislation if they obtain a certificate from a veterinarian that their dog is not dangerous?
 Again, there is conflicting evidence. The experts disagree as to whether a dangerous dog can be identified ahead of time. Some experts have expressed the opinion that pit bulls, in particular, are unpredictable. According to these experts, a pit bull may attack even though it has not been provoked and even though it has previously been known to be a friendly animal. Other experts disagree. In their view, potentially dangerous dogs can be identified before they are involved in a dangerous attack.
 It is my opinion that, in the face of this conflicting evidence, the legislature was entitled to decide that there was a sufficient body of evidence with respect to the inability to identify dangerous pit bulls in advance of an attack so as to justify restrictions that apply to all pit bulls. The recommended alternative approaches to breed-specific legislation largely depend on either a previous dangerous act or a responsible dog owner identifying a dangerous dog and taking appropriate action. Where public safety is concerned, it was open to the legislature to choose the more cautious approach.
 Thus far, the consideration of the means has focused on the broad scope of the provisions, that is, their application to all pit bulls. Another aspect in considering whether the means are disproportionate is to examine the restrictions themselves. What restrictions are placed on owners of pit bulls and others who may fall within the prohibitions in the Act? Are these restrictions disproportionate to the objective?
 In R. v. Clay, the court considered whether the law prohibiting simple possession of marihuana was overbroad. Gonthier and Binnie JJ., writing for the majority, stated that a legislative measure was overbroad where its adverse effect on the individual was grossly disproportionate to the state interest it was seeking to protect.
 What is the adverse effect on individuals in this case? Dog ownership is not a right. The provisions impose restrictions on the ownership and management of pit bull dogs. Those who own pit bulls are required to sterilize them and, when in public, muzzle them and put them on a short leash.
 These restrictions are not, in my opinion, in any way comparable to the restrictions in R. v. Heywood and R. v. Demers, two cases in which the legislative provisions were found to be unconstitutionally overbroad. In Heywood, persons convicted of specified offences were prohibited from loitering in certain public spaces. In Demers, the court considered the continued subjection of a permanently unfit accused person to the criminal process. Both cases involved significant restraints on liberty.
 It is not my task to substitute my opinion for that of the legislature as to how best to protect the public. It is also not necessary for me to resolve the conflicting evidence as to the role that breed plays in determining whether a dog is dangerous and whether pit bulls, as a breed, are dangerous. The legislature, in determining how to accomplish its objective, is not required to have conclusive evidence before it enacts legislation. The evidence with respect to the dangerousness of pit bulls, although conflicting and inconclusive, is sufficient, in my opinion, to constitute a “reasoned apprehension of harm”. In the face of conflicting evidence as to the feasibility of less restrictive means to protect the public, it was open to the legislature to decide to restrict the ownership of all pit bulls.
 I have considered not only the ambit of the restrictions, that is, that they apply to all pit bulls, but also the nature of the restrictions, that is, the restrictions they impose on the owners of pit bulls. Dog ownership is not a right. The impact of these restrictions on individual dog owners is not, in my opinion, disproportionate to the objective of protecting the public.
 As noted by the authors of one report on the pit bull debate, “(i)n the absence of conclusive data, legislators tend to err in the direction of safety”. When one considers the interests at stake, that is, the objective of public safety as against the restrictions on dog owners, it is my opinion that this was a choice that the legislators were entitled to make. I conclude, therefore, that the means the legislature has chosen are not too sweeping in relation to the objective and the provisions are not unconstitutionally overbroad.
Are the Provisions Vague?
 The applicant also submits that the definition of “pit bull” is unconstitutionally vague. In Canadian Foundation for Children, Youth and the Law v. Canada, 2004 SCC 4 (CanLII),  1 S.C.R. 76 McLachlin C.J.C. explained the vagueness principle as follows:
A law must set an intelligible standard both for the citizens it governs and the officials who must enforce it (at para. 16).
 The applicant’s submission that the definition is vague rests on her assertion that it is often not possible to determine whether a particular dog is a “pit bull” within the meaning of the Act. An individual cannot know, therefore, whether the dog that she or he owns falls within the purview of the restrictions.
Definitions of Pit Bulls
 There are several different definitions of “pit bull” that are commonly in use. While this decision concerns the definition in the Ontario legislation, the different definitions should be kept in mind in assessing the evidence.
 The definition in DOLA is as follows:
1 (1) “pit bull” includes,
(a) a pit bull terrier,
(b) a Staffordshire bull terrier,
(c) an American Staffordshire terrier,
(d) an American pit bull terrier,
(e) a dog that has an appearance and physical characteristics that are substantially similar to those of dogs referred to in any of clauses (a) to (d);
1 (2) In determining whether a dog is a pit bull within the meaning of this Act, a court may have regard to the breed standards established for Staffordshire Bull Terriers, American Staffordshire Terriers or American Pit Bull Terriers by the Canadian Kennel Club, the United Kennel Club, the American Kennel Club or the American Dog Breeders Association.
 Pit bulls and pit bull terriers are not breeds that are registered by Canadian or American Kennel Clubs. There are therefore no breed standards. Dr. Brisbin stated that “pit bull” is a generic category that includes registered breeds of dogs but also bull terriers and non-pure-bred dogs related to the breeds. Leslie Steeves of the Canadian Kennel Club stated that “pit bull” is a generic term that is used to describe a wide variety of mixed-breed dogs.
 A review of breed-specific legislation in Canada and the United States reveals a range of definitions. There are three common types of definitions in use: (1) definitions that define “pit bulls” as Staffordshire bull terriers, American Staffordshire terriers and American pit bull terriers and dogs that are physically similar to those three breeds; (2) definitions that do not list these three breeds but merely refer to pit bulls or pit bull terriers; and (3) definitions that do as DOLA does, that is, list the three breeds and add pit bulls or pit bull terriers and dogs that are physically similar to the three breeds or to pit bulls.
 There are several different working definitions of pit bulls. Many studies refer generally to “pit bulls” or “pit bull-type dogs”. Others define “pit bulls” as incorporating specific breeds.
 Thomas Skeldon, the Chief Dog Warden in Toledo, Ohio, stated that he and other dog wardens in his office frequently identify pit bulls “…by virtue of their appearance and similarity to Staffordshire bull terriers, American Staffordshire terriers and American pit bull terriers”.
 The American Veterinary Medical Association, in its booklet, “Dos and Don’ts Concerning Vicious Dogs” refers to illustrations of American pit bull terriers and American Staffordshire terriers, which it says comprise the majority of the pit bull group. The booklet states that the three breeds associated with the term “pit bull” are the Staffordshires or Staffordshire bull terriers, American pit bull terriers and American Staffordshire terriers.
 In a study he conducted, Dr. Brisbin considered American Staffordshire terriers, American pit bull terriers, Staffordshire bull terriers and any animal that dog owners called a pit bull.
 Karen Delisle explained that for the purposes of another study, “pit bull” meant American pit bull terriers, American Staffordshire terriers, bull terriers and Staffordshire bull terriers.
 In a study entitled “Are ‘Pit Bulls’ Different? An Analysis of the Pit Bull Controversy”, the authors noted the definitional problem:
…people commonly use the term pit bull to describe a variety of registered and unregistered dogs, including the American pit bull terrier (registered by United Kennel Club and American Dog Breeders Association), the American Staffordshire terrier, the Staffordshire bull terrier, the bull terrier, and the bulldog (all registered by the American Kennel Club), and many mixtures of these breeds with one another and with other breeds.
 The authors also noted the tendency to label “any stocky short-haired animal” involved in an attack as a pit bull.
 In Toledo v. Tellings, 2006 Ohio 975 (Ohio C.A.) (currently under appeal), Skow J. indicated that the lack of an exact statutory definition of “pit bull” in the Toledo ordinance meant that dogs were sometimes called pit bulls on the basis of whether their jaw was massive enough, their chest was muscular enough or their brow was broad enough.
Can a Pit Bull be Identified?
 The parties agree that a dog’s breed cannot be determined by DNA profiling. It can only be definitively determined if information as to its lineage is available.
 The parties also agree that it is possible to identify a pure-bred Staffordshire bull terrier, American Staffordshire terrier or American pit bull terrier that has been registered with a designated American or Canadian Kennel Club. However, the number of dogs that are registered is a small minority of the dogs that might arguably fall within the definition of “pit bull”. (In 2005, there were two American Staffordshire terriers and 114 Staffordshire bull terriers registered with the Canadian Kennel Club. The Canadian Kennel Club does not recognize the American pit bull terrier.)
 The parties disagree as to whether it is possible to identify other dogs that potentially fall within the definition in the Act. These include a dog whose lineage is unknown; mixed-breed dogs; a “pit bull terrier” or “pit bull”; or a dog that may be substantially similar to a pit bull terrier or to any of the three designated breeds. Although there was no evidence on dog populations, it is undisputed that these dogs constitute a significant portion of the population that could potentially be identified as “pit bulls” under the Act.
 Dr. Zaharchuk is an Ontario veterinarian and past president of the Ontario Veterinary Medical Association (OVMA). He stated that the only way to be sure of the breed of a dog is to know that dog’s lineage. There is, according to Dr. Zaharchuk, no scientific way to affirm or refute whether a mixed-breed dog has any of the specific breeds as part of its lineage. Furthermore, the “pit bull terrier” is not a known breed of dog.
 Dr. Zaharchuk expressed concern that the definition in the Act does not provide any guidance as to what aspects of a dog’s appearance or physical characteristics are to be considered in determining whether a dog is “substantially similar” to a pit bull. In his opinion, it is not possible to conclusively determine whether a dog has the appearance and physical characteristics that are substantially similar to the specified breeds. That is because breed standards contain a wide variety of characteristics and each characteristic that is identified for the pure-breed dogs is also found in the description for other dogs. Dr. Zaharchuk indicated that there are at least 24 pure breeds of dogs that arguably have substantially similar appearances and physical characteristics to the breeds specified in the Act.
 The OVMA has recommended to its members that they not sign the document prescribed by the Act in s. 19 (1). Dr. Zaharchuk stated that the primary reason for this position was that a veterinarian cannot accurately determine whether an individual dog falls within the definition unless it is both pure-bred and registered. However, the communication from the OVMA to its members states that the reason it was recommending that they not sign the certificate was to avoid possible civil liability.
 The OVMA uses the term “pit bull” in its communication and indicates that veterinarians do, indeed, designate pit bulls in maintaining patient records. However, the communication alludes to the identification problem when it notes that if the dog is not a specified pure-breed but may have physical characteristics substantially similar to the dogs specified in the legislation, “…you could inform the client that the dog might be considered a pit bull by law enforcement officers and the courts” [emphasis added].
 Ms. Lee Steeves is on the Board of Directors of the Canadian Kennel Club (CKC). She indicated that there is a 19-step process that the CKC follows before recognizing a new breed of dogs. The CKC recognizes and registers Staffordshire bull terriers and American Staffordshire terriers but does not recognize or register the American pit bull terrier as a pure-breed dog, although it is recognized by the United Kennel Club in the United States. The CKC also does not recognize pit bulls or pit bull terriers.
 Ms. Steeves explained that some standards are well–defined, while others are very general. In the case of the American Staffordshire terrier, the CKC breed standards provide descriptions under the categories of: general appearance, size, coat and colour, head, neck, forequarters, body, hindquarters, tail and gait. The CKC breed standards for the Staffordshire bull terrier provide descriptions of: general appearance, temperament, size, coat and colour, head, neck, forequarters, body, hindquarters and tail. The breed standards established by the United Kennel Club for American pit bull terriers describe the breed under these categories: history, general appearance, characteristics, head, neck, forequarters, body, hindquarters, feet, tail, coat, colour, height and weight and gait.
 In general, the closer a dog is to the breed standards, the easier it is to identify. Ms. Steeves stated that the standards are a “guideline” to assist the public to determine the type of dog. However, according to her, “a lot of it is subjective”. Ms. Steeves indicated that it could, for example, be difficult to differentiate between a poorly bred Labrador retriever and a poorly bred American Staffordshire terrier.
 Dr. Alexandra Soltan is a veterinarian in London, Ontario. She stated that in many cases, it is impossible to know the dog’s breed, since dogs may be a mixture of “one, two, three, four, five, six different breeds”. Dr. Soltan has a dog whose mother was a boxer and whose father was reported to be a Labrador retriever. She indicated that her dog has been mistaken for a pit bull since he was a puppy and she is concerned that this will create a problem under the legislation.
 Many dog owners do not know the lineage of their dog. Like Ms. Cochrane, Michael Martin obtained his dog from the Humane Society. Humane Society staff told him they were unsure of the dog’s breed. His dog has been variously identified by staff at veterinary clinics and Humane Society staff as “bull & terrier”, “pit bull terrier”, “American Staff/Cattle Dog” and “Australian Sheppard”.
 The authors of a study of dogs involved in fatal human attacks in the United States stated that the owners of mixed-breed or unregistered dogs have no way of knowing whether their dog is one of the types identified in breed-specific legislation and, therefore, cannot know whether they are required to comply with the legislation. Attempts to address this problem by including a description of the breed are problematic, in the authors’ opinion, because they rely on subjective visual observation and result in many more dogs than those of the specified breed being subject to the restrictions.
 Counsel for the applicant conducted an experiment and asked four people to identify pit bulls based on photographs. These were individuals who had witnessed violent incidents by dogs that they had identified as pit bulls and had provided affidavits for use by the respondent in this application. They were shown 25 small colour photos of dogs and asked to identify the dogs. The four witnesses provided different answers as to which dogs were pit bulls.
 The respondent objects to the admissibility of the results of this experiment for several reasons. The individuals did not see the actual dogs, rather they only viewed pictures that were very small and were a partial view; the witnesses gave qualified answers to some of the questions; and it is not possible to know definitively what the correct answers were because no one saw the actual dogs or breed papers or spoke to the owners.
 While it is not possible to come to a definitive conclusion as to what the correct answers were to the identification of the 25 dogs in the photos, the fact is that the four witnesses did not agree as to which dogs were pit bulls. The weight to be given to this lack of consistency in identification is tempered by the fact that the witnesses did not see the actual dogs but based their answers on very small pictures that presented only a partial view of the dogs.
 A court in Ohio recently addressed the issue of identifying a pit bull in Toledo v. Tellings. Thomas Skeldon, the Chief Dog Warden in Toledo, was a witness in that case. Skow J. referred to Mr. Skeldon’s testimony:
Dog Warden Skeldon acknowledged that even if a dog was 50 per cent pit bull, if it did not “look like a pit bull,” the owner would not be charged. On the other hand, if a dog did look like a pit bull,” it would be classified as a pit bull and the owner would be subject to the “vicious dog” laws. No definitive description of a “pit bull” was presented. The warden also acknowledged that there is really no way to tell if a dog is or is not a “pit bull” and the determination is made by his or a deputy’s subjective judgment. Regardless of its parentage or behavior, however, if a dog is labeled a pit bull, the owner would be charged under the statutes and city ordinance (at para. 30).
 An example of the difficulty in identifying a pit bull arose in the case of R. v. Jody Kirby (6 April 2006), Sarnia, Ont. (Ont. Ct. J.). Mr. Kirby was charged under DOLA because his dog was identified as a pit bull and was not leashed. A letter from a veterinarian introduced in evidence states that, “(i)n my opinion, this dog does have some pit bull similarities” (at 2). Another veterinarian, Dr. Sandy Taylor, gave evidence on behalf of Mr. Kirby. Dr. Taylor testified that she could not say whether the dog was a pit bull or a Staffordshire terrier without knowing its parentage or lineage. Although it had some tendencies of the bull terrier, it also had similarities to the bulldog. The judge concluded that the dog did not meet the legal definition of “pit bull”.
 A similar situation arose in the prosecution of Noah David Gatzke under a City of Winnipeg by-law. A veterinarian from the Animal Services Agency in Winnipeg examined the dog and concluded that it had the appearance and physical characteristics predominantly conforming to the standards for the breed known as pit bull terrier, Staffordshire bull terrier, American Staffordshire terrier and American pit bull terrier. Mr. Gatzke’s dog was, in fact registered with the National Kennel Club as an American bulldog. He was able to obtain an opinion from another veterinarian that the dog’s appearance and physical characteristics conformed to the standards for the American bulldog, a distinctly different breed than the American Staffordshire terrier. Mr. Gatzke was acquitted on the basis that reasonable doubt had been established regarding the breed of dog.
 The applicant points to these two cases as evidence of the difficulty of identifying pit bulls. The Attorney General, however, submits that these two cases demonstrate that courts are able to adjudicate the issue.
 The Attorney General contends that pit bulls are frequently identified without difficulty. Mr. Skeldon, for example, indicated that dog wardens with the Office of the Dog Warden for Lucas County, Ohio, frequently identify pit bulls. Veterinarians will often indicate pit bull in their patient records and academics will identify dogs as pit bulls in their research. Dr. Delisle and Dr. Brisbin, both witnesses for the applicant, identified dogs as pit bulls for the purpose of their work. The identification of a dog for the purpose of a veterinary practice or for an academic study is, however, a different matter than a determination that a particular dog falls within a legal definition for the purpose of a prosecution.
 Dr. Alan Beck is a professor of animal ecology at Purdue University. In his opinion, breeds are easily identifiable by form and breed specific behaviours. He referred to a booklet published by the American Veterinary Medical Association for veterinarians that gives them information as to how to recognize pit bulls. In his experience, most veterinarians can identify pit bull-type breeds as distinct from other breeds.
 The Attorney General cited numerous cases in Canada and the United States in which pit bull terriers were identified as such by the court. The City of Winnipeg has registered 45 convictions in municipal by-law prosecutions against pit bull owners. This, in the Attorney General’s submission, is evidence that a court is able to identify a dog as a pit bull.
 The evidence on identification may be summarized as follows. There is a small population of dogs covered by the legislation that can readily be identified: those dogs that are pure-bred, their lineage is known and they are registered with a designated Kennel Club. There are differing opinions as to the difficulty of identifying pit bulls that go beyond this small group. Many dog owners do not know their dogs’ lineage. Many dogs are mixtures of two or more breeds. Breed standards are only guidelines. Pit bulls and pit bull terriers are not breeds and there are, therefore, no breed standards. The identification of a dog often has a subjective component. As with any subjective exercise, there may well be a difference of opinion as to whether a particular dog is a “pit bull” as defined in the Act. There is evidence of misidentification and overreporting of dogs as pit bulls.
 The applicant asserts that the definition is unconstitutionally vague in that it is not possible to know whether a particular dog falls within the definition. She points, in particular, to the following problems: there is no known breed of pit bulls or pit bull terriers; the definition is open-ended; many dogs are mixed-breed and their lineage is unknown; and “appearance and physical characteristics that are substantially similar” is difficult to apply. There is, in short, insufficient guidance given to dog owners and to those who have to enforce the legislation as to which dogs fall within its ambit.
 The doctrine of vagueness was reviewed by Gonthier J. in R. v. Nova Scotia Pharmaceutical Society, 1992 CanLII 72 (S.C.C.),  2 S.C.R. 606 and Ontario v. Canadian Pacific, 1995 CanLII 112 (S.C.C.),  2 S.C.R. 1031.
 In Nova Scotia Pharmaceutical Society, Gonthier J. discussed the basis for the doctrine of vagueness. There are two rationales for the doctrine, both of which are included within the broader concept of the “rule of law”. The first rationale is that the individual is entitled to fair notice so that he or she can determine what is lawful conduct. The second rationale is to protect individuals from arbitrary and selective law enforcement. This is sometimes referred to as the “standardless sweep”.
 Language is, by its nature, imprecise. A degree of generality in laws is inevitable. It is not possible to achieve absolute certainty. The court should not require a law “…to achieve a standard of precision to which the subject matter does not lend itself” (at 642). Notwithstanding these inherent limitations, a law should still be intelligible. A law will be unconstitutionally vague where it gives “…insufficient guidance for legal debate” (at 638). Precision is too high a bar. Rather, a law should “…enunciate boundaries, which create an area of risk…Guidance, not direction, of conduct is a more realistic objective” (at 639). The degree of precision that is required will vary depending on the nature and subject matter of the particular legislative provision.
 Gonthier J. considered the doctrine of vagueness further in Ontario v. Canadian Pacific. In determining whether a particular provision is constitutionally vague, the court must consider whether the provision “…provides a sufficient basis for distinguishing between permissible and impermissible conduct, or for ascertaining an ‘area of risk’” (at 1070). The court should consider whether the law “…provides sufficient guidance for legal debate as to the scope of prohibited conduct” (at 1070). The standard of precision required will vary depending on the nature and subject matter (at 1071). The existence of an identifiable “core” of activity that is prohibited is often an indicator that the law provides sufficient guidance for legal debate (at 1047).
 Gonthier J. went on to refer to the “mediating role of the judiciary” in determining whether a legislative provision applied in a particular fact situation. The “mediating role of the judiciary” is of particular importance where it is practically difficult to frame the legislation in precise terms. The court must exhaust its interpretative function before determining whether a particular provision provides sufficient guidance (at 1070-1071).
 Factors to consider in determining whether a law is too vague include:
(a) the need for flexibility and the interpretive role of the courts,
(b) the impossibility of achieving absolute certainty, a standard of intelligibility being more appropriate and
(c) the possibility that many varying judicial interpretations of a given disposition may exist and perhaps coexist (Nova Scotia Pharmaceutical Society at
 McLachlin C.J. summarized the standard for vagueness in Canadian Foundation for Children v. Canada as follows:
A law is unconstitutionally vague if it “does not provide an adequate basis for legal debate” and “analysis”; “does not sufficiently delineate any area of risk”; or “is not intelligible”. The law must offer a “grasp to the judiciary”: R. v. Nova Scotia Pharmaceutical Society, 1992 CanLII 72 (S.C.C.),  2 S.C.R. 606 at pp. 639-40. Certainty is not required (at para. 15).
 While the rationale for the vagueness doctrine addresses the individuals who have to comply with the law and those who have to enforce it, the legal principles developed by the court address both the interests of these individuals, by referring to guidance for conduct and core areas of risk; and the ability of courts to determine the issue, by referring to guidance for legal debate and the mediating role of the judiciary.
 Is the definition of “pit bull” in the Act intelligible? Does it delineate an area of risk? Does it provide an adequate basis for legal debate and analysis?
 The applicant’s principal vagueness argument concerns clause (e) of the definition of “pit bull”, that is, that a “pit bull” includes, “a dog that has an appearance and physical characteristics that are substantially similar to those of dogs referred to in any of clauses (a) to (d)” [emphasis added]. Clauses (a) to (d) are: a pit bull terrier; a Staffordshire bull terrier; an American Staffordshire terrier; and an American pit bull terrier.
 The applicant contends that the legislation gives insufficient guidance to owners of dogs of unknown lineage to determine whether their dog is captured by the legislation. Given the problems with identifying any dog of unknown lineage, particularly mixed-breed dogs, the applicant argues that it is not possible for the court to apply the definition in a manner that will prevent absurd results.
 “Substantially similar” is a commonly used phrase. It means essentially or materially similar, without being identical.
 The Oxford English Dictionary defines “substantially” as:
4. In all essential characters or features; in regard to everything material; in essentials; to all intents and purposes; in the main.
 Black’s Law Dictionary provides the following definition:
SUBSTANTIALLY. Essentially; without material qualification; in the main; in substance; materially; in a substantial manner.
 There are many laws that incorporate the phrase “substantially similar”. There are, as well, numerous cases in which the phrase has been applied. Many of those laws refer to substantially similar facts, circumstances, terms or laws. There are, however examples of laws that deal with substantially similar in the physical sense. The Building Code Act, R.S.O. 1992, c. 23, s. 6 (1) (b), for example, refers to “substantially similar buildings”; the Consumer Protection Act, R.S.O. 2002, c. 30, Sh.A., s. 96 (1) (b) refers to a substantially similar condition of goods.
 The section in which “substantially similar” is located may be a source of assistance in interpretation. In Sansome v. Canada (M.N.R.),  T.C.J. No. 463 (T.C.C.), Cuddihy T.C.J. pointed to the factors set out under the Employment Insurance Act to be considered by the court when deciding whether terms of employment met the “substantially similar test”. The applicant submits that there is no similar list of factors to be considered by the court here.
 The Attorney General, however, asserts that the breed standards referred to in the legislation are factors to guide the court. The Manitoba Court of Appeal held that Winnipeg’s pit bull by-law, which also refers to breed standards, provided “reasonable benchmarks” for determining whether a particular dog was a pit bull (Manitoba Assn. of Dog Owners v. Winnipeg (City),  M.J. No. 661 (Man. Q.B.), aff’d  M.J. No. 643 (C.A.)). The provision in the Winnipeg by-law is arguably more precise than the Ontario provision, in that it mandates the court to consider breed standards when it refers to a dog whose appearance and physical characteristics predominantly conform to those standards. The Ontario provision states, instead, that the court “may have regard” to those standards.
 The applicant contends that the ability to refer to breed standards does not solve the problem. The breed standards include a long list of aspects of appearance and physical characteristics, some of which are well defined, others of which are more general in nature. The definition in the Act provides no guidance as to which aspects of appearance and characteristics are to be considered. Furthermore, there are no breed standards for pit bulls because they are not a breed.
 The definition of “pit bull” “includes” four types of dogs and dogs that are substantially similar to those dogs. The applicant argues that the open-ended nature of the definition that is created by the word “includes” adds to the vagueness of the definition.
 The fact that a definition is inclusive does not mean that the list is open-ended. Rather, it is meant to capture that which falls within the natural meaning of the defined word, plus other specific things that would not normally fall within the natural meaning (see Reynolds v. Commissioner of Income Tax for Trinidad and Tobago,  1 A.C. 7 (P.C.) at 10-11; M. & F. Frawley, Ltd. v. Ve-Ri-Best Co. Ltd.,  1 Q.B. 318 at 323, cited in P. St. J. Langan, Maxwell on The Interpretation of Statutes, 12th ed. (London: Sweet and Maxwell Ltd., 1969) at 271). An inclusive definition does not prevent the word from receiving its ordinary meaning (see Robinson v. Barton-Eccles Local Board (1883), 8 A.C. 798 (H.L.) at 801).
 Whether the word “includes” renders the definition of “pit bull” vague depends on whether the term “pit bull” itself has an ordinary and intelligible meaning. If “pit bull” is intelligible, the fact that it includes specified types of dogs does not render it open-ended or vague. If, however, “pit bull” does not have a meaning that is capable of interpretation, the fact that it is an inclusive definition may well be problematic.
 The respondent points to the fact that in the two Canadian decisions in which the court addressed the issue of the vagueness of the definition, one in Winnipeg and one in Lachine, the law was upheld. It should be noted, however, that both those cases dealt with by-laws. The test for challenging municipal by-laws is whether the by-law represents an abuse of power by the municipality. That is a very different test than the test for finding a law constitutionally vague.
 In Manitoba Assn. of Dog Owners Inc. v. Winnipeg, the Manitoba Court of Queen’s Bench considered a challenge to the by-law on the grounds that it was vague and uncertain; unfair, arbitrary and unreasonable; and discriminatory. With respect to the issue of vagueness, DeGraves J. stated that, “(t)he focus of the pit bull by-law is clear and precise” and that, “(t)he scope and effect of the pit bull by-law are clear and understandable by a reasonable person” (at paras. 11 and 12). The Manitoba Court of Appeal dismissed an appeal from the decision, stating that, “…it is not merely the dog’s appearance but its physical characteristics and other standards set out in the by-law which provide reasonable benchmarks for the determination of the breed by a licensed veterinarian” ( M.J. No. 643 at para. 3).
 The definition in the Winnipeg by-law is different from the Ontario definition in that the Ontario definition is inclusive (“pit bull” includes), while the Manitoba definition is not (“Pit bull dog” means). Furthermore, instead of including dogs that are substantially similar to the mentioned dogs, the Manitoba by-law arguably provides more precision, referring to dogs whose appearance and physical characteristics predominantly conform to established breed standards. The Manitoba provision mandates reference to breed standards; the Ontario law provides that the court “may” have regard to those standards.
 A challenge was also brought to a by-law in Lachine, Quebec (Madronero v. Lachine (Ville),  Q.J. No. 307 (Sup. Ct.)). In addition to prohibiting dangerous dogs, that by-law specifically prohibits:
1.03 Tout chien de race bull-terrier, Staffordshire bull-terrier, American bull-terrier ou American Staffordshire terrier;
1.04 Tout chien hybride issue d’un chien de la race mentionée au paragraphe 1.03 de cet article et d’un chien d’une autre race;
1.05 Tout chien de race croisée qui possède des caractéristiques substantielles d’un chien de la race mentionée au paragraph 1.03 de cet article.
 With respect to the ability to identify a dog falling within the definition, the court referred to the affidavit of Dr. Gilles Demers in which he stated:
Si les trois races de chiens pit bull mentionnées au Tableau I sont reconnues par des associations nationales comme l’AKC, l’UKA et le CKC, c’est parce que chacun des chiens d’une même race possède des caractéristiques physiques qui sont spécifiques à la race et qui ont été transmises par hérédité depuis plusieurs générations. Il est possible, à partir des caractéristiques physiques d’un animal, d’identifier sa race avec une certaine rigueur scientifique et de donner une description de ses attitudes comportementales. D’ailleurs au niveau des associations regroupant les éleveurs de chacune des races de chiens, il existe des normes physiques auxquels chacun des chiens de cette race doit correspondre (cf. ANNEXE 2). Ces normes pourrant être utilisées dans les cas où une identification de race sera nécessaire (at para. 23).
 The court noted at paragraph 24 that there were those who disagreed with Dr. Demers’ opinion regarding the ease of identifying pit bulls. However, the court was of the view that Dr. Demers’ opinion sufficiently relieved the factual problem of imprecision.
 Two Canadian cases referred to by the parties turned on the question of whether the dog in question was a pit bull: the prosecutions of Jody Kirby in Ontario and Noah Gatzke in Manitoba. In both cases, there was conflicting evidence from veterinarians concerning the identification of the dog, and in both cases the defendant was acquitted. This, the respondent maintains, demonstrates that a court is able to apply the definition to the facts before it and reach a decision as to whether a particular dog fits within that definition. The applicant, on the other hand, contends that these cases illustrate the difficulty of identifying pit bulls.
 A recent Nova Scotia case has been brought to my attention by counsel: R. v. Marilyn and Willard Cameron (11 December 2006), Nova Scotia (N.S. Prov. Ct.). The Camerons’ dog had been identified as a pit bull. Defence witnesses testified that pit bulls and pit bull terriers were not recognized breeds and that the dog in question did not have the characteristics of any of the breeds specified in the by-law (American pit bull terrier, Staffordshire bull terrier and American Staffordshire terrier). Stroud J. concluded that, based on the evidence from the defence witnesses, there was a reasonable doubt as to whether the dog fell within the Act. He commented on the by-law as follows:
…it is vague and overreaching and is based upon fiction as opposed to objective or scientific standard… (at 6)
 In an Addendum, dated January 12, 2007, Stroud J. expressed the opinion that the reference to a breed of dogs was unconstitutionally vague in the absence of a clear legal standard to determine the existence of or characteristics of the breed.
 The respondent cited numerous cases from the U.S. in which courts had rejected vagueness challenges. The applicant maintains that those cases are of limited value because the tests for unconstitutionality in the two jurisdictions are significantly different; they deal with different definitions of “pit bulls”; and they proceed on a different evidentiary basis than the one in the case at bar. While I agree with the applicant that the U.S. cases should be referred to with caution, the courts’ considerations of the intelligibility of various definitions are, nonetheless, of assistance.
 In Hearn v. City of Overland Park, 772 P.2d 758 (Kan. 1989), a case cited by the Manitoba court in Manitoba Assn. of Dog Owners Inc. v. Winnipeg (City), the Supreme Court of Kansas considered a constitutional challenge to a municipal ordinance that regulated the ownership of pit bulls in the city. The Supreme Court stated that the existence of marginal cases in which it is difficult to determine which side of the line a situation falls does not render a provision unconstitutionally vague. It applied the following test to determine whether a criminal statute is unconstitutionally vague:
…whether its language conveys a sufficiently definite warning as to the conduct proscribed when measured by common understanding and practice. A statute which either requires or forbids the doing of an act in terms so vague that persons of common intelligence must necessarily guess at its meaning and differ as to its application is violative of due process (at 642).
 The ordinance under consideration in Hearn identified three specific breeds of dogs as well as dogs that, “have the appearance and characteristics of being predominantly” of the three breeds, a definition that is arguably narrower than the DOLA definition. The municipal police department had adopted rules for the application of the ordinance and, in particular had defined “predominantly” as meaning that “…the officer has knowledge through identification procedures, admission by owner, keeper, or harborer, or otherwise that the animal is more than fifty percent pit bull. Predominantly shall further mean that the animal exhibits the physical characteristics of a pit bull more than that of any other breed of dog” (at 643). The court concluded that whether a particular dog qualified as a pit bull, as defined in the ordinance, was a question of fact to be determined by evidence.
 A similar conclusion was reached in Florida v. Peters, 534 So.2d 760 (Fla. Dist. Ct. App. 1988), in which the Court of Appeal of Florida considered an ordinance regulating the ownership of pit bulls. In that case, a dog was a pit bull if it substantially conformed to the kennel club standards for Staffordshire terriers, Staffordshire bull terriers or American pit bull terriers. The court stated that an individual need only look at those standards to determine whether the dog was described by any of them. Like the court in Hearn, the court concluded that whether a dog fell within the definition was a matter of evidence, not constitutional law.
 In Ohio v. Anderson, 57 Ohio St.3d 168 (Sup. Ct. Ohio 1991) the Supreme Court of Ohio considered a provision that defined “vicious dog” to include “(b)elongs to a breed that is commonly known as a pit bull dog.” The court acknowledged that it was difficult to precisely define “pit bull”. However, it noted that “pit bull dogs are distinctive enough that the ordinary dog owner knows or can discover with reasonable effort whether he or she owns such a dog” (at 171). While the physical description of a pit bull was neither absolute nor all encompassing, the traits were common enough among pit bulls such that the vast majority of dog owners would know whether they owned a pit bull. The court concluded that the physical and behavioural traits of pit bulls together with the available knowledge of dog breeds acquired by potential dog owners or possessed by veterinarians and breeders was sufficient to inform an individual whether their dog was a pit bull.
 The respondent cited numerous cases by various American courts that came to similar conclusions that various definitions of “pit bull” dogs were not unconstitutionally vague. I would note that not one definition was identical to that used in Ontario. In several cases, the definitions were arguably more precise or more limited, as for example, when the definition was limited to three specified breeds and dogs that predominantly conformed to specified breed standards. However, in Ohio v. Anderson, the definition of “a breed commonly known as a pit bull dog” was arguably less precise than the Ontario definition.
 There is one American case, however, in which a court has upheld a challenge to the legislation. That is the case of Toledo v. Tellings. The City of Toledo’s ordinance regulated the ownership or vicious dogs. Vicious dogs were defined to include a dog that “(b)elongs to a breed that is commonly known as a pit bull dog”. The court distinguished previous cases on the basis that they relied on “…now outdated information which perpetuated a stereotypical image of pit bulls” (at para. 63).
 With respect to the vagueness argument, the court stated:
We are troubled by the lack of an exact statutory definition of “pit bull”, the evidence presented that more than ten non-pit bull breeds look very much like pit bulls, and the highly subjective nature of the identification process. Particularly troublesome is the fact that, depending on the zealousness and bias of the local agency, criminal charges have likely been brought based on purely individual and speculative decisions on whether the jaw of a dog is “massive” enough or the chest muscular enough or the brow is broad enough to be designated as a “pit bull”, rather than some other similar breed, such as bulldog, boxer-mix, or bull mastiff. Although the Anderson court indicated that persons could easily discern that they owned a pit bull, we respectfully suggest that, some fifteen to twenty years later, with the greater number of a variety of breeds and mixed-breeds, this no longer holds true (at para. 73).
 The court concluded that the provision was unconstitutionally vague on the basis that “…the subjective identification of pit bulls may often include both non-pit bull dogs or dogs which are not vicious, to the extent that an ordinary citizen would not understand that he was breaking the law and which would result in the occurrence of arbitrary arrests and criminal charges” (at para. 76).
Application of the Law to the Evidence
 Given the evidence - a small number of dogs can easily be identified; breed standards are only guidelines and they vary in the degree of their specificity; pit bulls and pit bull terriers are not breeds; the identification of many dogs involves a subjective element; and there may well be differing opinions as to the identification of a particular dog – is the definition unconstitutionally vague?
 The doctrine of vagueness addresses the need to provide guidance to those who have to comply with the law and to those who have to enforce it. The standards that the courts have developed to determine whether a law is unconstitutionally vague suggest that the law should be assessed from two perspectives: (1) whether the law provides guidance to those who have to comply with or enforce the law by defining an “area of risk” or “core of activity”; and (2) whether the law provides guidance to courts such that it provides an adequate basis for legal debate and the courts are able to interpret and apply the law.
 Does the definition delineate a “core of activity” or an “area of risk”? Is there a basis for legal deliberation and debate?
 While a legislative provision does not have to be precise, it should provide guidance. The degree of precision required will depend on the context. As is apparent from the evidence, it is not possible to be precise in defining the large number of dogs who are neither registered, nor pure-bred, nor whose lineage is unknown.
 It is helpful to parse the definition from its most precise to its least precise elements.
“Staffordshire bull terrier, American Staffordshire terrier and American pit bull terrier”
 If a dog is registered with one of the recognized Kennel Clubs as one of these three breeds and is pure-bred, it can be identified. The reference to these three breeds is not vague.
 The DOLA definition of “pit bull” delineates a core in so far as it provides for animals of these three specified breeds. However, in order to withstand Charter scrutiny, the “core” should, in my opinion, be adequate to give sufficient guidance to individuals so that they know whether they fall within an area of risk. Where, as in this case, the majority of dogs will fall outside of the core, the specification of those three breeds may be insufficient, in itself, to define an area of risk that can be usefully applied to the majority of the dogs that are potentially covered by the definition.
“Dog that has an appearance and physical characteristics that are substantially similar to Staffordshire bull terrier, American Staffordshire bull terrier and American pit bull terrier”
 The definition provides the following guidance: a dog is a “pit bull” if it looks substantially like one of the three breeds. Is this sufficient guidance to determine whether a dog falls within the “area of risk”?
 A determination of whether a dog is “substantially similar” includes a subjective element. However, courts often make determinations on issues that involve an element of subjectivity. Courts are also able to determine whether one thing is “substantially similar” to another. The fact that a term requires interpretation does not render it vague (R. v. Morales, 1992 CanLII 53 (S.C.C.),  3 S.C.R. 711 at 729).
 The potential for two veterinarians to disagree on whether a dog is substantially similar to one of the designated breeds does not mean that a court is unable to determine the issue. The fact that a veterinarian may not be able to conclusively determine that a dog falls within the definition is also not fatal. Both the rules of evidence and the onus of proof assist a court when faced with inconclusive or conflicting evidence.
 The three breeds are known breeds. The definition further provides that the court may have regard to the breed standards. Dr. Zaharchuk expressed the opinion that the breed standards were not sufficient for the purpose of determining whether a dog is “substantially similar”; the standards contain a wide variety of characteristics and each characteristic can be found in the description of other dogs. Ms. Steeves indicated that the breed standards vary in their degree of specificity. This may make the standards more difficult to apply but, in my opinion, they nonetheless provide a basis for legal debate and deliberation.
 The breed standards also provide guidance to dog owners and others to assist them in determining whether a particular dog falls within the definition. Individual dog owners, veterinarians and others may not be able to definitively conclude that the dog is a “pit bull” under the Act, but the breed standards will assist them in deciding whether a dog falls within the area of risk, that is, whether the dog might fall within the Act. The OVMA recommended that veterinarians take this approach, that is, that they advise the client that “…the dog might be considered a pit bull”.
 The Ontario provision states that the court “may have regard” to the breed standards. Reference to breed standards is not mandated as is the case with the Winnipeg by-law. The court, however, is required to exhaust its interpretative function before it can be said that a law is vague. While a mandated reference to the breed standards might have been preferable, the provision that the court may have regard to those standards provides an interpretive guide and is sufficient, in my opinion, to provide the necessary guidance or benchmarks.
 The legal standard that the definition has to meet in order to be constitutional is not precision. The court, dog owner or enforcer does not have to be able to conclusively determine that a particular dog falls within the definition. Rather, the definition has to provide a basis for deliberation by the court, standards for enforcers so as to prevent a “standardless sweep” and guidance for dog owners so that they can determine whether there is a real risk that their dog may fall within the definition. In my opinion, the reference to the three breeds and to the breed standards provides sufficient guidance to identify dogs that are substantially similar to the three breeds.
“Pit bull” and “ pit bull terrier”
 Does the DOLA definition include dogs that are neither one of the three specified breeds, nor substantially similar to any of those breeds?
 The evidence is that there are many definitions of “pit bull” and “pit bull terrier” in use. Some define pit bulls as only the three specified breeds and those dogs that are similar to those three breeds; others define “pit bulls” more broadly, referring to pit bulls, pit bull terriers or pit bull-type dogs. In some cases, the difference is a distinction without a difference, that is, the users of the term “pit bull” are, in fact, referring only to dogs that are one of the three specified terms. In other cases, the term is used more broadly to include other dogs, such as the bull terrier and bulldog or, indeed, to apply it to any stocky short-haired animal involved in an aggressive act.
 The Attorney General submits that “pit bulls” and “pit bull terriers” are generic terms that refer to dogs of one of the three specified breeds or dogs that are hybrids or mixes of these three breeds. However, that is not what the DOLA definition says. It states that “pit bull includes…” and it also adds “pit bull terriers” and dogs that are substantially similar to pit bull terriers to the list.
 It should be presumed that the legislature avoids meaningless words and does not pointlessly repeat itself. A meaning should, where possible, be given to every word in a statute (see Ruth Sullivan, Sullivan and Driedger on the Construction of Statutes, 4th ed. (Vancouver: Butterworths, 2002) at 158). I therefore assume that in providing that “pit bull includes” and specifying “pit bull terrier”, it was contemplated that the definition encompasses dogs in addition to Staffordshire bull terriers, American Staffordshire terriers and American pit bull terriers, and dogs substantially similar to these three breeds.
 There is a significant variation in how “pit bulls” and “pit bull terriers” are defined. There is evidence that the misidentification and overreporting of pit bulls is common. Pit bulls and pit bull terriers are not breeds; they do not have breed standards. Opinions as to whether pit bulls can be identified vary, from Dr. Zaharchuk, who stated that there are at least 24 breeds of dogs that are substantially similar in appearance to the specified breeds, to Dr. Beck, who takes the position that pit bulls are easily identifiable.
 While s. 1 (2) provides that the court may have reference to breed standards in determining whether a dog is a pit bull, these standards will be of limited assistance where the claim is that the dog is of a type or is substantially similar to a type other than one of the three specified breeds.
 Given these problems of definition and identification, it is my opinion that the phrases “pit bull includes” and “pit bull terriers” are problematic in so far as they appear to include an undefined number of dogs that fall beyond the three specified breeds and dogs substantially similar to those three breeds. In so far as these terms go beyond the three breeds, they do not, in my opinion, provide sufficient guidance to courts or to those who have to enforce the legislation. They also do not define an “area of risk” for dog owners.
 I conclude that insofar as the legislature has used the phrases “pit bull includes” and “pit bull terriers” to define “pit bull”, there is insufficient guidance provided to individuals who have to comply with the law, to those who have to enforce it and to courts that have to interpret it. These provisions are, therefore, unconstitutionally vague.
 Can a provision that is unconstitutionally vague be justified as a reasonable limit within s. 1 of the Charter?
 It has been suggested that if a law is unconstitutionally vague it cannot be saved by s. l. In R. v. Heywood, Cory J. stated:
This court has expressed doubt about whether a violation of the right to life, liberty or security of the person, which is not in accordance with the principles of fundamental justice, can ever be justified, except perhaps in times of war and national emergencies (at 802).
 In Osborne v. Canada (Treasury Board), 1991 CanLII 60 (S.C.C.),  2 S.C.R. 69, Sopinka J. similarly indicated that where a law is so imprecise that it cannot be interpreted, there is no “limit prescribed by law” and no s. 1 analysis is therefore necessary (at 94-95).
 However, in New Brunswick (Minister of Health and Community Services) v. G. (J.) 1999 CanLII 653 (S.C.C.), (1999), 26 C.R. (5th) 203 (S.C.C.), Lamer C.J.C. suggested that s. 7 violations might be justified under s. 1. Such a justification would, however, be rare for two reasons: the rights protected are very significant and would not ordinarily be overruled by competing social interests; and violations of principles of fundamental justice, specifically the right to a fair hearing, would rarely be upheld (at para. 99).
 The Attorney General submitted that if the DOLA provisions infringed a Charter right, that infringement was justified under s. 1. However, no specific justification was provided for a vague definition. The general s. 1 argument that the Attorney General provided – the legislation is in support of a pressing and substantial objective; there is a rational connection between the provisions and the objective; the provisions minimally impair the rights of pit bull owners; and the provisions are not disproportionate – does not speak to whether a vague definition of pit bull can be justified.
 There was no evidence as to why the legislative provisions had to extend beyond the three specified breeds in order to accomplish the objective of the legislation. Indeed, many legislative provisions in other jurisdictions limit the definition to the three breeds and to dogs that are substantially similar or predominantly conform to these breeds.
 In the absence of any justification, I conclude that the vagueness cannot be justified under s. 1.
Do the Provisions Contravene the Right to Trial Fairness?
 The applicant contends that s. 19 of the Act, which provides for the receipt in evidence of a veterinary certificate, violates both s. 7 and s. 11 (d) of the Charter.
 Section 19 of the Act provides as follows:
(1) A document purporting to be signed by a member of the College of Veterinarians of Ontario stating that a dog is a pit bull within the meaning of this Act is receivable in evidence in a prosecution for an offence under this Act as proof, in the absence of evidence to the contrary, that the dog is a pit bull for the purposes of this Act, without proof of the signature and without proof that the signatory is a member of the college.
(2) No action or other proceeding may be instituted against a member of the College of Veterinarians of Ontario for providing, in good faith, a document described in subsection (1).
(3) For greater certainly, this section does not remove the onus on the prosecution to prove its case beyond a reasonable doubt.
 It has already been noted that s. 7 of the Charter applies by virtue of the potential for imprisonment. Section 11 (d) provides for trial fairness where a person has been charged with an offence:
Any person charged with an offence has the right to be presumed innocent until proven guilty according to law in a fair and public hearing by an independent and impartial tribunal.
 These rights apply to all offences for which there is a possibility of imprisonment (R. v. Wigglesworth 1987 CanLII 41 (S.C.C.), (1987), 37 C.C.C. (3d) 385 (S.C.C.) at 399-403).
 The applicant’s submission is two-fold:
(i) The lack of a provision for cross-examination in respect of the document from the veterinarian contravenes the right to a fair trial; and
(ii) The fact that the document is proof that the dog is a pit bull, in absence of evidence to the contrary, infringes the defendant’s right to be presumed innocent.
 The starting point of both of these arguments is that the determination of the identity of the dog as a “pit bull” within the definition in the Act is an essential element of the offence. Indeed, in many cases, the only issue will be whether the dog in question is a pit bull.
 Another important factor in considering the issues of cross-examination and proof is that the identification of the dog as a pit bull is not a straightforward matter. Individuals, both experts and non-experts, may well disagree as to whether a particular dog fits within the definition in the Act. If the dog is mixed or a cross-breed, as many dogs are, the identification of that dog is further complicated and will depend on a determination as to whether the dog is “substantially similar” to any of the specified breeds. This is a determination that, by its very nature, includes a subjective element.
 Section 19 provides that a document from a veterinarian stating that a dog is a pit bull will be proof, in the absence of evidence to the contrary, that the dog is a pit bull. The section does not expressly provide an opportunity for the defendant to cross-examine the veterinarian. The applicant contends that this failure to provide for cross-examination of the veterinarian who signs the certificate stating that the dog is a pit bull contravenes the defendant’s right to trial fairness.
 The right to cross-examine is a fundamental element of trial fairness. As Cory J. explained in R. v. Osolin, 1993 CanLII 54 (S.C.C.),  4 S.C.R. 595:
There can be no question of the importance of cross-examination. It is of essential importance in determining whether a witness is credible. Even with the most honest witness cross-examination can provide the means to explore the frailties of the testimony…Its importance cannot be denied. It is the ultimate means of demonstrating truth and of testing veracity. Cross-examination must be permitted so that an accused can make full answer and defence. The opportunity to cross-examine witnesses is fundamental to providing a fair trial to an accused. This is an old and well established principle that is closely linked to the presumption of innocence (at 663).
 That is not to say that the right to cross-examination is unlimited. Evidence must, for example, be relevant in order to be admissible. The scope for cross-examination of a complainant has been limited in sexual assault cases.
 The fact that it is less expensive for the Crown to present a document from a veterinarian than to have the veterinarian present at the trial is not a reason to preclude the right to cross-examination. Administrative convenience does not override Charter rights and the right to trial fairness (see Singh v. Canada (Minister of Employment and Immigration), 1985 CanLII 65 (S.C.C.),  1 S.C.R. 177 at 218-219).
 In Waterloo (Regional Municipality) v. Yan,  O.J. No. 4012 (C.A.), the Ontario Court of Appeal considered a provision of the Highway Traffic Act that expressly limited examination of a provincial offences officer on information obtained from a red light camera system. The officer was not required to give evidence unless a summons was issued. In order to issue a summons, the judge had to be satisfied that the defendant would not have a fair trial if the officer were not required to give oral evidence. The court upheld the restriction on oral evidence:
The use of certificate evidence is designed to save the delay and expense of calling the certifier of the evidence, where prima facie, there is no basis to doubt the accuracy and reliability of the certificate. The situation is not analogous to those in which the liberty of the subject may be at stake. Here, the only sanction that may be imposed is a fine; there is no exposure to imprisonment, probation, loss of licence or loss of demerit points. The limits placed on the examination of the provincial offences officer in red light traffic camera situations reflects a proper balance of the various competing interests in my view (at para. 19).
 Thus, the court drew a distinction between a case in which the only sanction was a fine and a case such as the situation at hand, in which the potential consequences are more serious, as for example, where there is the possibility of imprisonment.
 A further distinction may be drawn between the instant situation and the situation in Waterloo v. Yan. The veterinarian’s document is not, in my opinion, akin to the certificate that the court dealt with in Waterloo v. Yan, where the court was satisfied that “…prima facie, there is no basis to doubt the accuracy and reliability of the certificate”.
 The document from a veterinarian is more akin to an expert opinion. As with any expert opinion, it is essential that the defendant have the opportunity to cross-examine the expert who has signed the opinion. The defendant must be able to challenge the veterinarian’s experience, her or his familiarity with pit bulls and the reasons for the statement that the dog is a pit bull. The evidence is that there may well be a difference of opinion as to whether a particular dog is a pit bull or is substantially similar to a pit bull. Indeed, in the Canadian cases referred to above in which there were acquittals, this is exactly what occurred.
 Given the sanction that may be imposed and the nature of the evidence in question, it is my opinion that the opportunity of the defendant to cross-examine the veterinarian is critical to ensuring trial fairness. Section 19 is silent as to whether there is a right to cross-examine. While it does not expressly provide for such a right, neither does it expressly preclude or limit it, as was the case in Waterloo v. Yan. The issue, then, is whether the absence of an explicit provision for cross-examination means that the defendant does not have a right to cross-examine.
 The applicant maintains that the statute should expressly provide for a right to cross-examine. She points by way of comparison to certificates under the Controlled Drugs and Substances Act, S.C. 1996, c. 19 that are admissible in evidence as proof of the facts contained therein. That Act provides that the defence may, with leave of the court, require that the person who issued the certificate either produce an affidavit or solemn declaration attesting to any facts deemed to be proved or appear for examination or cross-examination (ss. 50(1) and (2)). This express provision for cross-examination is missing from DOLA.
 Section 19 should be read in conjunction with the provisions of the Courts of Justice Act, R.S.O. 1990, c. C.43 and the Provincial Offences Act, R.S.O. 1990, c. P.33. The Courts of Justice Act provides that the jurisdiction conferred on the Ontario Court of Justice “shall, in the absence of express provision for procedures for its exercise in any Act, regulation or rule, be exercised in any manner consistent with the due administration of justice” (s. 146).
 The relevant sections of the Provincial Offences Act provide for the issuance of a summons to a witness and the examination and cross-examination of witnesses:
39. (1) Where a justice is satisfied that a person is able to give material evidence in a proceeding under this Act, the justice may issue a summons requiring the person to attend to give evidence and bring with him or her any writings or things referred to in the summons.
46. (2) The defendant is entitled to make full answer and defence.
46. (3) The prosecutor or defendant, as the case may be, may examine and cross-examine witnesses.
 While typically, cross-examination arises after the other side has examined the witness in chief, there is nothing to prevent a judge from allowing the defendant to call a witness and cross-examine that witness.
 Where legislation is imprecise or open to more than one interpretation, it should be interpreted in a way that is consistent with the Charter. As stated in Slaight Communications Inc. v. Davidson, 1989 CanLII 92 (S.C.C.),  1 S.C.R. 1038, “(l)egislation conferring an imprecise discretion must therefore be interpreted as not allowing the Charter rights to be infringed” (at 1078).
 Thus, in determining whether DOLA offends the Charter because it does not expressly confer a right of cross-examination, it should be assumed that the discretion of the court to give leave to a defendant to cross-examine will be exercised in a way that is consistent with Charter rights and, in particular, with the right to trial fairness and the due administration of justice.
 I therefore conclude that the absence of an express right to cross-examine the maker of the document does not contravene the right to trial fairness.
Presumption of Innocence
Mandatory or Permissive Presumption?
 The applicant contends that s. 19 improperly relieves the Crown of the burden of proof by creating a mandatory presumption in its favour. It does so by providing that a document from a veterinarian is proof that the dog is a pit bull. It places the burden on the defendant to provide evidence to the contrary, that is, that the dog is not a pit bull and thereby infringes the defendant’s right to be presumed innocent under s. 11 (d) of the Charter.
 The Attorney General argues that the provision does not offend the presumption of innocence because s. 19 (3) expressly provides that the section does not remove the onus on the prosecution to prove its case beyond a reasonable doubt. Furthermore, the provision is permissive, not mandatory, in that the document is “receivable” in evidence. If it is permissive, it is constitutionally acceptable.
 The document from a veterinarian “…is receivable in evidence…as proof, in the absence of evidence to the contrary, that the dog is a pit bull…without proof of the signature and without proof that the signatory is a member of the College”. Is this a mandatory or a permissive presumption?
 The applicant argues that s. 19 creates a mandatory presumption such that, once a document from a veterinarian is presented, the trier of fact must accept it and must conclude that the dog is, in fact, a pit bull, in the absence of evidence to the contrary. The Attorney General, on the other hand, argues that the word “receivable” means that the judge may choose whether or not to accept the document as proof; the inference is therefore permissive.
 Does “receivable” mean that the judge “may” receive the evidence, in the sense that the judge has a choice; or does “receivable” mean that evidence that would otherwise be inadmissible is rendered admissible?
 The word “receivable” should be interpreted within the context of the provision. The purpose of the provision is, in my opinion, to allow the Crown to introduce a document from a veterinarian instead of having to call the veterinarian as a witness. This is made clear by the words, “without proof of the signature and without proof that the signatory is a member of the College.” Absent such a provision, the Crown would be unable to introduce the document without calling the maker of that document, that is, the veterinarian. The purpose of s. 19 (1) is therefore to provide a mechanism whereby issues of evidence can be dealt with expeditiously and economically. As recognized in Waterloo v. Yan, “(t)he use of certified evidence is designed to save the delay and expense of calling the certifier…” (at para. 19).
 Based on this contextual reading, it is my opinion that “receivable” means that the Crown may introduce into evidence a document that it would otherwise not be able to introduce in the absence of the testimony of the maker of that document. However, once the document is introduced, it becomes proof of the fact that the dog is a pit bull as defined in the Act, absent evidence to the contrary. I therefore conclude that this is a mandatory presumption. Once the document is introduced, the trier of fact is required to draw the conclusion in the absence of evidence to the contrary.
 The Supreme Court considered presumptions in the context of an alleged infringement of s. 11 (d) of the Charter in R. v. Downey 1992 CanLII 109 (S.C.C.), (1992), 72 C.C.C. (3d) 1 (S.C.C.). Cory J. referred to an analysis of presumptions by T.A. Cromwell in “Proving guilt: The Presumption of Innocence and the Canadian Charter of Rights and Freedoms” in W.H. Charles, T.A. Cromwell & K.B. Jobson, eds., Evidence and the Canadian Charter of Rights and Freedoms (Toronto: Butterworths, 1989) 125 at 130. He categorized presumptions as follows:
(1) Presumptions which operate without the requirement of proof of any basic facts.
(2) Presumptions which require proof of a basic fact.
(a) Permissive Inferences: Where the trier of fact is entitled to infer a presumed fact from the proof of the basic fact, but is not obliged to do so. This results in a tactical burden whereby the accused may wish to call evidence in rebuttal, but is not required to do so.
(b) Evidential Burdens: Where the trier of fact is required to draw the conclusion from proof of the basic fact in the absence of evidence to the contrary. This mandatory conclusion results in an evidential burden whereby the accused will need to call evidence, unless there is already evidence to the contrary in the Crown’s case.
(c) Legal Burdens: Similar to the burden in (b) except that the presumed fact must be disproved on a balance of probabilities instead of by the mere raising of evidence to the contrary. These are also referred to as “reverse onus clauses”.
 The provision in s. 19 (1) is, in my opinion, an evidential burden. Upon the introduction in evidence of a document from a veterinarian, the trier of fact is required to conclude, in the absence of evidence to the contrary, that the dog is a pit bull.
Does an Evidential Burden Contravene the Charter?
 Before deciding whether s. 19 (1) contravenes the Charter, as claimed by the applicant, it is necessary to comment on the extent to which an evidential burden may violate the Charter. In R. v. Oakes, 1986 CanLII 46 (S.C.C.),  1 S.C.R. 103 the court considered a burden that was placed on the accused to disprove the presumption on the balance of probabilities. Does a burden that merely requires an accused to raise a reasonable doubt contravene s. 11 (d)?
 In R. v. Osolin, the Supreme Court of Canada concluded that a mere evidential burden did not contravene the Charter. In that case, there was a burden on the accused to point to evidence or adduce sufficient evidence to raise the common law defence of mistake of fact. The Crown, however, still had to establish all the essential elements of the offence.
 In R. v. Downey, the Supreme Court of Canada found that an evidential burden did contravene the Charter (subject to s. 1). At issue was a Criminal Code provision that required the trier of fact to conclude that the accused lived off the avails of prostitution from the fact that the accused lived with or was habitually in the company of prostitutes. The accused could rebut this presumption by adducing evidence that he or she was not living off the avails. Cory J. stated that a mandatory statutory presumption would violate s. 11 (d) if it required the trier of fact to convict in spite of a reasonable doubt. On the other hand, it would be valid “…if the proof of the substituted fact leads inexorably to the proof of the other” (at 13). The Court held that the fact that a person lives with prostitutes does not lead inexorably to the conclusion that the person is living off the avails of prostitution. Therefore, the presumption violated s. 11 (d) because it mandated a conviction despite the existence of a reasonable doubt as to the presumed fact.
 Like the situation in R. v. Downey, the case at hand deals with a fact that is to be presumed, absent evidence to the contrary. The identification of the dog as a pit bull is a part of the case that the Crown would otherwise have to establish; it is not a defence, as was the case in R. v. Osolin. In my opinion, the principles in R. v. Downey apply. In order to determine whether s. 19 (1) of DOLA violates s. 11 (d), I must therefore determine whether proof of the substituted fact leads inexorably to proof of the presumed fact.
Does the Proof of the Substituted Fact Lead Inexorably to the Proof of the Presumed Fact?
 As noted by Cory J. in R. v. Downey, “a statutory presumption will be valid if the proof of the substituted fact leads inexorably to the proof of the other”. A similar approach was taken in R. v. Vaillancourt 1987 CanLII 2 (S.C.C.), (1987), 39 C.C.C. (3d) 118 (S.C.C), in which Lamer J. stated that substitution of one element for another is constitutional only “…if upon proof beyond reasonable doubt of the substituted element, it would be unreasonable for the trier of fact not to be satisfied beyond reasonable doubt of the existence of the essential element” (at 136).
 If the Crown introduces a document from a veterinarian stating that a particular dog is a pit bull, would it be unreasonable for the judge not to be satisfied beyond a reasonable doubt that the dog was, in fact, a pit bull?
 There is ample evidence that the identification of a particular dog as a pit bull is a matter upon which veterinarians and others may well differ. The statement that a dog is a pit bull is akin to an expert opinion; it is a conclusion or inference drawn from the veterinarian’s observations and guided by her or his experience and expertise. The expertise and experience of veterinarians with pit bulls will vary, as will their opinions. This is particularly the case when one is dealing with a dog that is a mixed or cross-breed and the issue is whether the dog is “substantially similar”. As was indicated by Ms. Lee Steeves, a member of the Board of Directors of the Canadian Kennel Club, the application of breed standards is a subjective exercise. This is very different from the situation in Waterloo v. Yan, in which there was no basis to doubt the accuracy and reliability of the certificate from the red light camera system or the situation in R. v. Phillips 1988 CanLII 198 (ON C.A.), (1988), 42 C.C.C. (3d) 150 (Ont. C.A.), where it would only be in relatively rare cases that the results of a breathalyzer test would be inaccurate.
 The difficulties in the identification of pit bulls mean that it will often not be possible for the judge to be satisfied beyond a reasonable doubt that a particular dog was a “pit bull” as defined in the Act solely on the basis of a document from a veterinarian. I conclude, therefore, that s. 19 establishes a mandatory presumption that is contrary to the Charter unless it can be justified under s. 1.
 Is the provision a reasonable limit under s. 1? In order to answer this question, it is necessary to apply the two-part test set out in R. v. Oakes: the legislation must be in support of a pressing and substantial objective; and the means must be reasonable and demonstrably justified. The second part of the test is referred to as the “proportionality test”. It has three components: there must be a rational connection between the limit on the right and the objective; the limit should impair the right as little as possible; and there should be proportionality between the effects of the measure and the objective.
(i) Pressing and Substantial Objective
 The first branch of the test is easily met. The legislation is in support of a pressing and substantial objective, namely, the reduction and elimination of the risk of attacks by dogs. The applicant does not challenge this objective.
(ii) Proportionality Test
 The first part of the proportionality test is a consideration of whether there is a rational connection between the limit on the right and the objective. In the case at hand, the question is whether there is a rational connection between proving that a dog is a pit bull through the introduction of a document from a veterinarian and the objective of the legislation, that is, preventing attacks by dogs.
 I have already indicated that the legislature was entitled to choose to restrict the ownership of pit bulls in order to meet its objective. It follows that the Crown will need to be able to prove that a particular dog is a pit bull. There must be a rational connection between its method of proof, that is, the document from the veterinarian stating that the dog is a pit bull and the finding that the dog is, indeed, a pit bull.
 In my opinion, there is a rational connection between the document that is filed and the finding that the dog is a pit bull in that it is a licensed veterinarian who is attesting to the fact that the dog is a pit bull. The use of a document from a veterinarian to establish that a dog is a pit bull is not “arbitrary, unfair or based on irrational considerations” (R. v. Oakes at 139).
 Is the impairment as minimal as possible? In R. v. Downey, the court considered “…whether Parliament could reasonably have chosen an alternative means which would have achieved the identified objective as effectively” (at 19). Are there alternative means by which the Crown could prove that the dog is a pit bull that would have a less adverse impact on the defendant?
 In R. v. Phillips, the Ontario Court of Appeal considered the constitutionality of Criminal Code provisions that state that the results of a breathalyzer test, in the absence of evidence to the contrary, are proof of the concentration of alcohol in the blood of the accused at the time of the alleged offence. In upholding the provision, the court considered the following factors: the breathalyzer reading is based upon conclusions reached by qualified scientists derived from research and experiments; the presumption could be easily rebutted with information known to the accused; and there were various safeguards in place (the test could only be demanded in certain circumstances; it could only be performed by a certified operator using an approved machine; and two tests were required at specified times). The court concluded that the legislative provision did not place an onerous burden on the accused because the accused only needed to create a reasonable doubt. The accuracy of the breathalyzer machines and the safeguards in place were such that a burden would be cast on only a very small group of late drinkers. Furthermore, it was a burden that would be within their power to meet.
 Similar to the situation in Phillips, the defendant under s. 19 (1) of DOLA need only raise a reasonable doubt. However, in contrast to Phillips, the circumstances in this case are such that there could be a significant number of people affected by the burden. Unlike a breathalyzer reading, the statement by a veterinarian that a dog is a pit bull may have a significant subjective element, particularly in cases where the veterinarian will need to determine whether a dog is “substantially similar” to a specified breed or to a pit bull terrier. It is, as noted, more in the nature of an expert opinion. The Crown does not need to establish the experience and particular qualifications of the veterinarian who is certifying that the dog is a pit bull. The veterinarian may, in fact, have little or no experience with pit bulls. The identification of a dog is, as noted, a matter about which veterinarians may disagree. It will often be difficult for a defendant to rebut the presumption on his or her own; rather, the defendant will have to call another veterinarian or expert to testify in order to provide evidence to the contrary.
 There is an obvious means available to the Crown to prove that the dog is a pit bull: the Crown could call a witness to testify as to the identification of the dog. The legislative provision in question relieves the Crown of the inconvenience and expense but, in effect, places the inconvenience, expense and burden on the accused. Administrative convenience is not a reason to contravene the right to trial fairness (Singh v. Canada (Minister of Employment and Immigration) at 218-219).
Proportionality between the Effect and the Objective
 The final part of the proportionality test is to consider whether there is proportionality between the effects of the measure and the objective. In considering this branch of the test in Downey, the court concluded that there was a relatively minor infringement, that is, the defendant only had to point to evidence capable of raising a reasonable doubt. This was considered minimal as compared to the important objective involved.
 In this case, as noted, while the defendant need only establish a reasonable doubt, the defendant would have difficulty rebutting the presumption on his or her own. It could well be necessary to call another veterinarian or other expert. While the objective of the legislation is an important one, the Crown can easily accomplish that objective by calling its own evidence to establish the identification of the dog. It is not, in my opinion, justifiable to place that burden on the defendant.
(iii) Section 1 - Conclusion
 While the legislation is in support of an important objective, the limitation cannot, in my opinion, be justified. There is a rational connection between the document from a veterinarian and a finding that the dog is a pit bull. However, it is problematic insofar as the identification of the dog may be both debatable and subjective in a significant number of cases. The provision has the potential to place a burden on a significant number of defendants. In many cases, they will only be able to discharge the burden by calling a veterinarian or other expert. There is an alternative means to accomplish the objective: the Crown could present a witness to testify to the identification of the dog instead of being able to rely on a document. While this may have implications in terms of efficiency and cost, the practical effect of placing the burden on the defendant to present evidence to the contrary is to transfer the burden of calling evidence from the Crown to the defendant. This, in my opinion, cannot be justified.
Onus on the Prosecution to Prove its Case
 The final argument of the Attorney General is that the provision cannot contravene the right to be presumed innocent because s. 19 (3) provides that the section “…does not remove the onus on the prosecution to prove its case beyond a reasonable doubt”.
 The addition of this provision cannot, in my opinion, “Charter-proof” a section that is otherwise problematic. As I have noted above, the effect of s. 19 (1) is that if the Crown introduces a document that certifies that a particular dog is a pit bull within the Act, that document is proof that the dog is a pit bull, unless there is other evidence to the contrary. In most cases, it will be up to the defendant to produce the evidence to the contrary. While this does not remove the ultimate burden on the prosecution to prove its case beyond a reasonable doubt, it does place an evidential burden on the defendant and therefore contravenes a Charter right.
 Section 19 (1) creates a mandatory presumption: where the Crown introduces a document from a veterinarian stating that the dog is a pit bull, that document is proof that the dog is a pit bull in the absence of evidence to the contrary. However, the document from a veterinarian that the dog is a pit bull does not lead inexorably to the conclusion that the dog is, indeed, a pit bull as defined in the Act. The provision creates a mandatory presumption which, given the nature of the statement from the veterinarian and the burden that the presumption places on the defendant, cannot be justified under s. 1.
Is the Legislation Ultra Vires the Province?
 The applicant submits that the definition of “pit bull” in the Act conflicts with and frustrates the purpose of a federal statute, the Animal Pedigree Act, and is therefore ultra vires the provincial government.
 The applicant acknowledges that the Ontario legislature has the authority to enact legislation that restricts the ownership and handling of dogs within the province under the “property and civil rights” power granted to it by s. 92 (13) of the Constitution Act, 1867 (U.K.), 30 & 31 Vict., c. 3, reprinted in R.S.C. 1985, App. II, No. 5. However, she submits that the legislature has no authority to define breeds of dogs as it has done in s. 1 of DOLA, nor can it authorize veterinarians to declare whether a particular dog is a pit bull.
 The applicant maintains that, pursuant to the doctrine of paramountcy, the provincial legislation is invalid. She contends that ss. 1 and 6 to 10 of DOLA define and regulate the breeding and sale of pit bulls in Ontario. She submits that these provisions serve the same purpose as the Animal Pedigree Act, which contains a complete scheme for the identification of dogs.
 The doctrine of federal paramountcy provides that where there is an inconsistency between provincial and federal legislation, the provincial legislation is inoperative to the extent of that inconsistency. In order to determine whether there is an inconsistency, two questions need to be answered. Firstly, can a person simultaneously comply with the provisions of both Acts? Secondly, do the provisions of the provincial legislation frustrate the purpose of the federal legislation? (Rothmans, Benson & Hedges Inc. v. Saskatchewan, 2005 SCC 13 (CanLII),  1 S.C.R. 188 at paras. 11 and 15).
 I see no difficulty with an individual complying with both legislative enactments. The fact that an association established under the Animal Pedigree Act may identify and register a dog does not prevent an individual from complying with the provisions of DOLA concerning pit bulls.
 As to whether the provincial legislation frustrates the purpose of the federal legislation, the two enactments have entirely different purposes. The purpose of the Animal Pedigree Act is to enhance the stock of animals that have a commercial purpose (R. v. Neuman 1998 ABCA 261 (CanLII), (1998), 127 C.C.C. (3d) 440 (Alta C.A.) at 443). In order to accomplish this purpose, it provides for the creation of animal pedigree associations to identify and register valuable animals. The purpose of the DOLA provisions, however, is to reduce and eliminate the perceived danger of pit bulls, as defined in that legislation. These provisions do not frustrate the purpose of the Animal Pedigree Act.
 I therefore conclude that an individual can comply with both the federal and provincial legislation. Furthermore, the DOLA provisions do not frustrate the purpose of the Animal Pedigree Act. They are therefore not inconsistent with the federal legislation and are not inoperative.
 This decision is not a judgment as to the wisdom of the legislature in choosing to restrict the ownership of pit bulls. Rather, it is solely concerned with the constitutional validity of the legislative provisions.
 While the evidence with respect to the dangerousness of pit bulls is inconclusive and conflicting, the legislature had a “reasoned apprehension of harm”. In the face of conflicting evidence, it was open to the legislature to choose to target all pit bulls. The means it chose are not disproportionate to the objective. The legislation is therefore not unconstitutionally overbroad.
 The legislation provides sufficient guidance for the determination of whether a dog is a Staffordshire bull terrier, American Staffordshire terrier and American pit bull terrier or a dog that is substantially similar in appearance and physical characteristics to any of these three breeds. However, by providing that “pit bull includes” and by designating “pit bull terriers”, the definition captures dogs beyond these three breeds and beyond those that are substantially similar in appearance and physical characteristics to these three breeds. To that extent, it is unconstitutionally vague and is not saved by s. 1 of the Charter.
 With respect to s. 19 (1) of the Act, it is my opinion that this provision does not offend the right to trial fairness when it fails to expressly grant the defendant a right to cross-examination. However, by providing that the document from a veterinarian is receivable in evidence as proof that the dog is a pit bull, the legislature has created a mandatory presumption that offends trial fairness and the right to be presumed innocent, contrary to s. 11 (d) of the Charter. Such a breach cannot be justified under s. 1 of the Charter.
 The legislative provisions do not conflict with the Animal Pedigree Act and are therefore not inoperative on this basis.
 The parties agreed that, in the event that I concluded that there was a breach of the Charter, they would make subsequent submissions on the appropriate remedy. I would invite them to make arrangements to see me in order to make those submissions. Submissions on costs may be made at the same time.
Released: March 23, 2007
Friday, March 30, 2007
Decision regarding December 2006 motion