Paul v Registrar, Domestic Animals Act 2000 (Appeal)
[2019] ACAT 38
•8 April 2019
ACT CIVIL & ADMINISTRATIVE TRIBUNAL
PAUL v REGISTRAR, DOMESTIC ANIMALS ACT 2000 (Appeal) [2019] ACAT 38
AA 51/2018 (AT 59/2018)
AA 4/2019 (AT 72/2018)
Catchwords: APPEAL – administrative review – attack by dog resulting in death of animal – Registrar’s decision to euthanise dog confirmed by original tribunal – original tribunal not reasonably satisfied dog unlikely to be a danger to the public or another animal – circumstantial evidence of dog’s involvement in attack – civil standard of proof applies
Legislation cited: ACT Civil and Administrative Tribunal Act 2008 s 82
Domestic Animals Act 2000 ss 53B, 59, 62, 63, 64, 65, 66, 67, 68, 68A
Subordinate
Legislation cited: ACT Civil and Administrative Tribunal Procedure Rules 2009 (No 2) s 21
Animal Welfare (Welfare of Dogs in the ACT) Code of Practice 2010
Cases cited:August v Commissioner of Taxation [2013] FCAFC 85
Bradshaw v McEwans Pty Ltd (1951) 217 ALR 1
CDJ v VAJ [1998] HCA 67
Chen v State of NSW (No 2) [2016] NSWCA 292Chong & Neale v CC Containers Pty Limited & Ors [2015] VSCA 137
Fox v Percy [2003] HCA 22
Hillier v R [2008] ACTCA 3
Jovanovic v R [2015] ACTCA 29Lau v Registrar, Domestic Animals Act 2000 [2018] ACAT 119
Leighton v The Queen [2017] ACTCA 55
Mansour v Dangar [2017] ACAT 49
Mulholland v Mitchell [1971] AC 666
Phillips v Inspector-General in Bankruptcy [2012] AATA 788
Paul v Registrar, Domestic Animals Act 2000 [2018] ACAT 105
Seltsam Pty Limited v McGuiness; James Hardie & Coy Pty Limited v McGuiness [2000] NSWCA 29Warren v Coombes [1979] HCA 9
Tribunal:Presidential Member MT Daniel
Date of Orders: 8 April 2019
Date of Reasons for Decision: 8 April 2019
AUSTRALIAN CAPITAL TERRITORY )
CIVIL & ADMINISTRATIVE TRIBUNAL )
AA 51/2018
AA 4/2019
BETWEEN:
SAMUEL PAUL
Appellant
AND:
REGISTRAR, DOMESTIC ANIMALS ACT 2000
Respondent
APPEAL TRIBUNAL: Presidential Member MT Daniel
DATE:8 April 2019
ORDER
The Tribunal orders that:
1.The orders of 31 October 2018 are confirmed.
………………………………..
Presidential Member MT Daniel
REASONS FOR DECISION
Background
1.Around 7:00pm on the evening of 18 January 2018, Khan, an adult German Shepherd dog, was found dead in the backyard of the suburban Canberra home where he was kept. Two other dogs, Kogan and Boo, were present in the backyard. Khan’s keeper, who discovered the body, took photographs of Kogan and Boo. He rang Access Canberra and reported the incident.
2.Kogan and Boo were usually kept at the house next door. Their owner (the appellant) had been away for a couple of days for work and returned on the evening of 18 January 2018. When he realised Kogan and Boo were not in the backyard, he called them. Kogan immediately jumped over the Colorbond fence between the two backyards. Boo tried once to jump over the fence and failed. The appellant then helped her over on her second attempt. The appellant observed blood on Kogan, but did not observe any blood on Boo. The appellant observed no fresh injuries on either Kogan or Boo.
3.Rangers from Domestic Animal Services attended at around 8:15pm and inspected the scene. They took possession of Khan’s body for the purpose of a necropsy. They took photographs of the backyard, noting tufts of dog hair and blood at various locations, and inspected the fences which they recorded as being in good repair. They spoke with the appellant and with Khan’s keeper. The rangers ultimately decided to seize Kogan and Boo on the basis that there was a complaint that each dog had been involved in an attack on an animal.
4.On 21 June 2018 the Registrar for the Domestic Animals Act 2000 (DAA) decided to destroy Kogan and Boo. The Registrar made the decision under section 53B of the DAA. That section provides:
53B Dealing with attacking dogs—death or serious injury to person or death of animal
(1) This section applies if the registrar is reasonably satisfied, because of a complaint or otherwise, that—
(a)a dog attacked a person or an animal; and
(b)the attack caused—
(i)the death of the person; or
(ii)serious injury to the person; or
(iii)the death of the animal.
(2) The registrar must destroy the dog.
(3) However, subsection (2) does not apply if, and only if, the registrar is reasonably satisfied the dog is not likely to be a danger to the public or another animal.
(4) For subsection (3), the registrar may consider—
(a)the circumstances of the attack including whether—
(i)the person or animal provoked the dog; or
(ii)the person or animal was attacked because the dog came to the aid of a person or animal the dog could be expected to protect; or
(iii)if the attack was on premises occupied by the keeper of the dog—the person or animal was on the premises without lawful excuse; and
(b)whether reasonable steps can be taken to reduce the risk of the dog endangering the public and other animals; and
(c)any other relevant matter.
(5) The registrar may destroy the dog if—
(a)the registrar gives the dog’s keeper written notice of the decision to destroy the dog; and
(b)the dog’s keeper—
(i)does not, within 7 days after the day the notice is given (the application period), apply to the ACAT under section 120 for review of the decision; or
(ii)applies to the ACAT under section 120 for review of the decision within the application period and the registrar’s decision to destroy the dog is confirmed.
(6) If the registrar decides not to destroy the dog, the registrar—
(a)may issue a control order for the dog to the dog’s keeper or carer; and
(b)if the registrar issues a control order to a carer of the dog—must give a copy of the order to the dog’s keeper.
Note 1The registrar may declare a dog to be a dangerous dog if the dog attacked a person or animal—see s 22 (2).
Note 2The registrar must give a reviewable decision notice for s (2) and s (6) to the keeper or carer and must also take reasonable steps to give a reviewable decision notice to any other person whose interests are affected by the decision (see s 119 and ACT Civil and Administrative Tribunal Act 2008, s 67A).
5.The appellant applied to the tribunal for review of the Registrar’s decision. That application was heard on 2 October 2018.
6.The appellant represented himself at that hearing. He relied on the documents that had been before the Registrar (which included witness statements from the appellant and others), and supplemented these documents with a report from a dog behaviourist (Mr Sean Ehlers), a report from a psychologist (Ms Vanessa Inglis), and a number of letters in the nature of ‘references’ from persons who had observed both Kogan and Boo with the appellant. The appellant submitted to the tribunal at this hearing (the Original Tribunal) that only Kogan had been involved in the attack, and that both dogs could be kept by him in a manner which would ensure the safety of the public and other animals in the future. He provided a quote for the construction of secure cages or runs which he undertook to construct and use if the dogs were returned to him. He said he was willing to have the dogs declared ‘dangerous’ and apply for a dangerous dog licence in relation to each dog.
7.The Registrar was legally represented at the hearing on 2 October 2018. The Registrar did not rely on any additional material, and submitted that the decision to destroy each dog should be confirmed.
8.The evidence was not tested at the hearing. No witnesses gave evidence on oath or affirmation or were cross-examined. The parties made oral submissions and the decision was reserved. On 31 October 2018 the Original Tribunal made orders confirming the Registrar’s decision to destroy each dog. The Original Tribunal published written reasons for its decision (the original decision).
9.The Original Tribunal was satisfied that both dogs were involved in the attack resulting in Khan’s death, thus engaging the provisions of section 53B of the DAA:
In this case the applicant’s dogs jumped a two metre-high fence and attacked Khan, the German shepherd next door. The attack resulted in this animal’s death. The part each dog played in Khan’s assault cannot be determined, as no witnesses to the attack have been found. The investigation report detailed that the German shepherd was larger and weighed more than the applicant’s dogs. It would seem, due to the blood found on Kogan’s nose, that he was definitely involved with the wounds caused to Khan.
While the applicant submitted that there is no evidence that Boo attacked Khan, there is also no evidence that she did not. The applicant gave evidence that both Kogan and Boo are hunting dogs used to hunting in a pack. The fact that tufts of black hair and blood were spread around the concrete ground, underneath a camper trailer and on parts of the internal cement driveway, suggests that Khan was pursued in this area before his death.
Pursuant to section 53B, the dogs must be destroyed unless the Tribunal is reasonably satisfied that the dogs are not likely to be a danger to the public or another animal. In considering this likelihood of danger, the Act details that the Tribunal may consider the circumstances of the attack, including whether the dogs were provoked or came to the aid of another they were protecting.[1]
[1] Paul v Registrar, Domestic Animals Act 2000 [2018] ACAT 105 (the original decision) at [24]-[26]
10.The Original Tribunal then considered the matters set out in subsection 53B(4) of the DAA and concluded that it was not reasonably satisfied that either of the dogs was not likely to be a danger to the public or another animal:
There is no evidence here that Khan provoked the applicant’s dogs. Khan’s owner gave evidence to the investigator that he was on anti-inflammatory tablets due to hip issues, which also prevented him from jumping. The sad situation is that he was in his own backyard separated from the applicant’s property by a high wall when he was attacked.
Based on the ferocity and severity of this attack, the Tribunal is not reasonably satisfied that either dog is unlikely to be a danger to the public or another animal. The respondent submitted that their training as hunting dogs coupled with the killing of Khan, demonstrated a likelihood that this behaviour may be repeated. The Tribunal’s attention was also drawn to the fact that Khan was attacked at the site of his head and neck which, if repeated, would be likely to result in serious injury or death to any future victim.
In Robertson v Domestic Animal Services (Robertson), General President Crebbin reasoned:
When making a decision about something that is likely to happen in the future, ideally we’d have crystal balls, but we don’t. What we need to do is look at what has happened in the past. That criterion requires an examination of the circumstances of the incidents that led to us being here today. Thus ACAT must examine and must take into account the past conduct and behavior of the dogs as a major factor … in predicting the likelihood of harm being caused to any member of the public or an animal. [footnote omitted]
At the hearing, the applicant submitted that the dogs’ pig hunting did not result in the pigs being maimed in the same manner as Khan and so this hunting background should not be taken into account by the Tribunal. The Tribunal considers that the combination of being hunting dogs with the attack of Khan leads to them being considered as a future danger to the public or another animal.
Section 53B(4)(b) of the Act also provides that the decision maker may consider whether any reasonable steps can be taken to reduce the risk of the dog endangering the public and other animals. The applicant has clearly given thought to this consideration, as seen by the tendering of a quotation to modify his backyard. He also gave assurances of how he would restrain the dogs in the future.
In considering this evidence, the Tribunal notes that the applicant told the investigators that he knew his dogs could jump the fences between him and neighbouring properties. Despite this, they were not restrained in his backyard. The investigators found that the applicant had an established dog run with a mesh roof in his backyard but it was not used.[2]
[2] Original decision at [27]-[32]
11.After considering other cases in which the tribunal had declined to accept undertakings as to how dogs would be kept, the Original Tribunal concluded at paragraph 36 of the original decision that “there are not reasonable steps which can be taken to reduce the risk of the dogs endangering the public and other animals.”
12.The Original Tribunal made orders confirming the Registrar’s decision to destroy Kogan and Boo.
The appeal
13.The appellant has appealed the Original Tribunal’s decision to the Appeal Tribunal. The appellant’s grounds of appeal effectively fell under two headings:
(a)the Original Tribunal erred in finding that Boo was involved in the attack on Khan; and
(b)the Original Tribunal erred in failing to be reasonably satisfied that Kogan and/or Boo is not likely to be a danger to the public or another animal.
14.The appeal was conducted as a review under section 82(b) of the ACT Civil and Administrative Tribunal Act 2008 (ACAT Act). The appellant was represented by an advocate, the Registrar was represented by Counsel.
15.The appellant submitted that, if the Appeal Tribunal was satisfied that the Original Tribunal had erred in confirming the Registrar’s decision, the appropriate outcome would be to set aside the Original Tribunal’s orders and instead order that the decision of the Registrar be set aside and the matter remitted to the Registrar to consider, on the basis of the most up to date information, whether the dogs should be returned to the appellant under sections 62 to 68A of the DAA.
The fresh evidence on appeal
16.The appellant sought to rely on fresh evidence at the hearing of the appeal. This consisted of:
(a)photographs of dog cages now under construction at his premises pursuant to the quote considered by the Original Tribunal;
(b)a collection of media articles about dog attacks; and
(c)general information about the temperament and characteristics of the Bull Arab breed of dogs.
17.Section 21(c) of the ACT Civil and Administrative Tribunal Procedure Rules 2009 (No 2) (the Rules) provides that the Appeal Tribunal “may receive further evidence about questions of fact, either orally in a hearing, by written statement or in another way”. The Rules do not prescribe how the power is to be exercised. The approach of the Appeal Tribunal is to apply the principles established in relation to similar statutory powers in other jurisdictions.[3]
[3] See Leighton v The Queen [2017] ACTCA 55 (Leighton) for a summary of these principles
18.The overarching principle is that the power is “remedial”[4] and “exists to serve the demands of justice.”[5]
[4] August v Commissioner of Taxation [2013] FCAFC 85 (August) at [119]
[5] CDJ v VAJ [1998] HCA 67 at [111]
19.At a systemic level, it is not usually in the interests of justice to allow new evidence on appeal because this is inconsistent with the appellate nature of the proceedings,[6] and the public interest in finality of litigation.[7]
[6] “For the further evidence to be admitted in the present case would be inconsistent with the nature of appellate review. A trial is not a rehearsal for an appeal.” — Leighton at [29]
[7] Mulholland v Mitchell [1971] AC 666
20.However, in an individual case it may be in the interests of justice for new evidence to be admitted on appeal if it is established that:
(a)the evidence would, or was likely to, have produced a different result;[8] and
(b)there is an acceptable reason for the evidence not being put before the original tribunal.
[8] August at [119]
21.For the second factor, the Appeal Tribunal will consider whether the new evidence could have been procured with reasonable diligence for the original hearing, and if it could have been, why it was not.[9] It has been said that a forensic decision not to call evidence at the original hearing “will weigh heavily against its reception on appeal”[10] but evidence may more readily be admitted on appeal where an irregularity in the proceedings prevented a party from putting his or her case effectively.[11]
[9] Jovanovic v R [2015] ACTCA 29 at [22]
[10] Hillier v R [2008] ACTCA 3 (Hillier) at [164]
[11] Hillier [161], quoting CDJ v VAJ [1998] HCA 67 at [57] per Gaudron J
22.In relation to the photographs of the dog cages under construction at his residence, the appellant submitted that these photographs showed that his undertaking as to how the dogs would be kept could be relied upon — contrary to the concerns of the Original Tribunal noted at paragraphs 31 to 36 of the original decision. This evidence could not have been provided to the Original Tribunal — the cages did not exist at the time of the original hearing. In this sense, the photographs are further evidence of subsequent events. However, the photographs were relied upon as going to the issues addressed by the Original Tribunal at paragraphs 31 to 36 of the original decision. A finding as to the unreliability of the appellant was integral to the Original Tribunal’s conclusion at paragraph 36. The Registrar did not oppose the Appeal Tribunal having regard to the photographs. Accordingly, the Appeal Tribunal has had regard to these photographs for the appeal.
23.The appellant also sought to put before the Appeal Tribunal a selection of media articles about dog attacks. The Registrar opposed these articles being considered by the Appeal Tribunal. The appellant submitted that the articles were relevant in understanding the political climate within which the Registrar’s and the Original Tribunal’s decisions to destroy Kogan and Boo were made. The appellant submitted:
It is an inescapable fact that the highly charged political environment surrounding dog attacks in the ACT over the last 18 months must have played a role in the decision … It is clear that persistent political & public criticism of DAS and the elected Government has resulted in a response to ‘toughen up’, with Mr Paul, Boo & Kogan being the collateral damage.[12]
[12] Appellant’s submissions dated 25 January 2019 at page 2
However, the appellant did not specify any errors of fact or law by the Original Tribunal deriving from these propositions. When pressed, the appellant’s advocate clarified that he did not suggest that the Original Tribunal had been affected by bias or duress or unable to impartially perform its merits review function, but rather thought that the general political climate was relevant context.
24.The Appeal Tribunal declined to consider the media articles. There was no explanation for why these articles had not been provided to the Original Tribunal and I was not satisfied that they would or could have led to a different outcome before the Original Tribunal. The articles were also of no relevance to the issues raised on appeal.[13]
[13] It is worth noting that because many of the articles were from local news organisations, both the Original Tribunal and Appeal Tribunal might well have read them when originally published
25.The appellant also sought to put before the Appeal Tribunal general information about the temperament of the Bull Arab breed. This was opposed by the Registrar. Although the temperament of Kogan and Boo was a factor in the decision of the Original Tribunal,[14] I declined to consider this fresh evidence on appeal. There was no satisfactory explanation as to why this information was not put before the Original Tribunal. Also, I was not satisfied that this information would or could have led to different result before the Original Tribunal. This general information would have been of very little probative value given that the Original Tribunal had a specific assessment of each dog by an expert, and specific references in relation to each dog.
[14] Original decision at [14], [18] and [43]
26.At a number of times during the hearing of the appeal, and in his written submissions,[15] the appellant referred to scratches or damage to his side of the fence. Neither party was able to point me to such evidence before the Original Tribunal. No application was made for evidence of scratches on the fence to be admitted on the appeal. I have disregarded these submissions. However I note that even if such evidence had been before the Appeal Tribunal it would only reinforce the conclusion that Boo struggled to get over the fence, which has already been drawn.
[15] Appellant’s submissions dated 25 January 2019 page 2
27.I turn next to consider the errors asserted by the appellant.
The Original Tribunal erred in finding that Boo was involved in the attack on Khan – section 53B(1) DAA
28.The appellant submitted that the Original Tribunal erred in finding that Boo was involved in the attack on Khan, that is, that section 53B(1) was satisfied in relation to Boo. There were a number of limbs to this argument.
29.First, the appellant submitted that, at paragraph 25 of the original decision, the Original Tribunal misdirected itself as to the required standard of proof. I note that the Original Tribunal did not expressly state the standard of proof which it applied. This omission does not of itself mean that the Original Tribunal applied the wrong standard: the application of the civil standard to administrative proceedings[16] is so well accepted nowadays as to be a matter of little comment in most written decisions. The requirement that the Registrar be ‘reasonably satisfied’ — contained in section 53B(1) of the DAA, which the Original Tribunal quoted in full — is itself a reference to the civil standard of proof. Significantly, an examination of paragraph 25 of the original decision shows that the Original Tribunal was not there commenting on the applicable standard, but was simply explaining, in layman’s terms, that there was no direct evidence of Boo’s involvement in the attack. The evidence was purely circumstantial, and the fact of Boo’s involvement, or not, had to be inferred from other established facts. I am not satisfied in this respect that the Original Tribunal erred in applying an incorrect standard of proof.
[16] Phillips v Inspector-General in Bankruptcy [2012] AATA 788 (Phillips) at [224]; Seltsam Pty Limited v McGuiness; James Hardie & Coy Pty Limited v McGuiness [2000] NSWCA 29
30.Secondly, the appellant submitted that the onus to prove Boo’s involvement fell upon the Registrar, and that onus was not met. The appellant submitted that, given the evidence was purely circumstantial, the possibility of Boo not being involved in the attack could not be excluded, and consequently the Original Tribunal could not be satisfied to the required standard that she had been involved in the attack.
31.I do not accept these submissions to be a correct statement of the applicable law. First, it is well established that, unless the legislation specifically imposes it, there is no ‘onus’ on one or another party to proceedings for review of an administrative decision.[17] Further, the requirement to exclude reasonably available hypotheses consistent with innocence references the criminal, not civil, standard of proof. In Chen v State of NSW (No 2) it was said “[i]t is not necessary, in order to make a finding on the civil standard, to exclude other possible explanations.”[18]
[17] Phillips at [221]-[223]
[18] [2016] NSWCA 292 at [34]
32.The appellant’s final and overarching submission in relation to section 53B(1) of the DAA was that the established facts supported the inference that Boo was not involved in the attack on Khan, and the Original Tribunal erred in not drawing that inference.
33.In conducting an appeal by way of review, the Appeal Tribunal does not lightly interfere with facts found by the Original Tribunal after conducting a hearing.[19] Usually the Original Tribunal has had the benefit of hearing and observing witnesses, which provides a fact-finding advantage to the Original Tribunal. However, where the fact in issue is to be inferred from a number of established facts, the Appeal Tribunal may be in as good a position as the Original Tribunal to draw the necessary inferences and, if so, it should not shrink from doing so.[20] The circumstances of the original hearing, in which the evidence was not tested and witnesses were not called, also makes it appropriate to take that approach. Accordingly, the parties addressed the Appeal Tribunal on the established facts[21] and inferences that should be drawn from them.
[19] Mansour v Dangar [2017] ACAT 49 at [22]
[20] Warren v Coombes [1979] HCA 9; Fox v Percy [2003] HCA 22
[21] As is discussed below in paragraphs 54 to 59, the Original Tribunal’s reasons for decision included some factual errors contained in the T-documents. In considering the established facts, I have relied upon the facts agreed by the parties in the appeal: for example, that both Kogan and Boo weighed more than Khan
34.The appellant submitted that the most likely scenario is that Kogan went over the fence first, possibly looking for food, and Khan responded defensively which provoked Kogan into attacking. Hearing the commotion, Boo dug out of the dog run and scrambled over the fence, only to arrive after the attack was over and Khan was dead. One problem with this hypothesis is that the evidence before the Original Tribunal established Boo was not in the dog run on the morning in question.[22] Nonetheless, the conclusion that Boo arrived after the attack was over is available to be drawn from the established facts.
[22] T-documents at page 69 (email dated 15 May 2018)
35.The respondent submitted that weight must be given to the fact that Kogan and Boo are trained to hunt together, and both were found in the yard with Khan’s body, the most likely inference from those facts being that they both participated in the attack on Khan.
36.The question is whether the established facts give rise to a reasonable and definite inference that Boo was involved in the attack, rather than ‘conflicting inferences of equal degrees of probability’. As the High Court has unanimously stated:
Of course as far as logical consistency goes many hypotheses may be put which the evidence does not exclude positively. But this is a civil and not a criminal case. We are concerned with probabilities, not with possibilities. The difference between the criminal standard of proof in its application to circumstantial evidence and the civil is that in the former the facts must be such as to exclude reasonable hypotheses consistent with innocence, while [in] the latter you need only circumstances raising a more probable inference in favour of what is alleged. In questions of this sort, where direct proof is not available it is enough … [if] the circumstances appearing in the evidence give rise to a reasonable and definite inference: they must do more than give rise to conflicting inferences of equal degrees of probability so that the choice between them is mere matter of conjecture (see per Lord Robson, Richard Evans & Co Ltd v Astley). But if circumstances are proved in which it is reasonable to find a balance of probabilities in favour of the conclusion sought then though the conclusion may fall short of certainty it is not to be regarded as a mere conjecture or surmise.[23]
[23] Bradshaw v McEwans Pty Ltd (1951) 217 ALR 1 at [6]
37.In considering whether the inference that Boo was involved in the attack is one which, on the balance of probabilities, arises from the established facts, I am also conscious that the serious consequences of such a finding affects the degree of satisfaction required.[24]
[24] Chong & Neale v CC Containers Pty Limited & Ors [2015] VSCA 137 at [44], quoting Rejfek v McElroy [1965] HCA 46
38.I am satisfied that if Boo was in Khan’s backyard when the attack was underway, she would have participated. I am also satisfied that if Boo heard the attack underway when she was not in Khan’s backyard, she would have frantically found a way to get there and become involved.
39.It is a fact that Boo was not observed by the appellant to have blood on her, while Kogan did. However, visible blood is not an inevitable consequence of involvement in an attack. A dog can be involved in a joint attack on an animal in ways that will not lead to the dog having blood on it, for example by chasing the animal, biting at its tail or limbs, or flushing it out from a place of safety. The absence of blood on Boo does not exclude her from involvement in the attack, but it does make it less likely that she was involved in the attack in a way which would result in blood covering her. I do not consider the absence of blood on Boo to be definitive as to her lack of involvement, given the varied ways in which a dog may participate in an attack on an animal.
40.The hair and blood throughout the yard was said to be indicative of a prolonged attack. The appellant submitted that this supported an inference that Boo was not involved in the attack. He said that if she had been involved, it would have been over quickly because she “goes in hard”.[25] The appellant also pointed to Boo’s observed difficulty in jumping over the fence as lending support to the inference that she arrived in Khan’s backyard only after the attack was completed.
[25] Oral submissions by the appellant on the appeal
41.I consider that Boo’s difficulty in scaling the fence and the evidence that the attack was protracted lends support to the inference that Boo was not involved in the attack, but also supports an inference that her involvement came while the attack was still underway.
42.I raised with the parties during the hearing of the appeal that the lack of injury to either Kogan or Boo might support an inference that both dogs were involved in the attack, thus limiting Khan’s ability to inflict injury while defending himself. The appellant said that the involvement of more than one dog in an attack can have the effect of the attacking dogs injuring each other accidentally. While this may be correct on rare occasions, or with dogs not trained to hunt together, I do not accept the appellant’s reasoning applies in the circumstances of this case. I note also the evidence of Khan’s owner was that he would, in her view, be unlikely to defend himself. Nonetheless, given the length of the attack, I consider it likely that Khan tried inadequately to defend himself. The multiple injuries to Khan and the lack of injury to Kogan and Boo supports the inference that both dogs were involved in the attack.
43.In the end, Boo’s presence in the yard with Khan’s dead body, her training and experience hunting jointly with Kogan, her freedom to roam in her backyard on the morning of the attack, her difficulty getting over the fence, the prolonged nature of the attack, and the multiple injuries to Khan and the lack of injury to either Kogan or Boo, are facts which lead me to be reasonably satisfied that Boo was involved in the attack on Khan, most likely after the attack had commenced. The possibility of Boo arriving in Khan’s backyard after the attack was entirely over cannot be excluded, but seems to me an unlikely scenario.
44.It follows that I do not consider that the Original Tribunal was in error in finding that Boo was involved in the attack causing Khan’s death, with the consequence that section 53B of the DAA applied.
45.For completeness, I note that even if section 53B did not apply to Boo, the test for return of the dog under section 64 of the DAA[26] requires the Registrar to be satisfied that release of the dog, potentially under a control order, does not involve unacceptable risk to the safety of the public and other animals. For the reasons set out below, I would not have been so satisfied if required to consider that issue on the appeal.[27]
The Original Tribunal erred in failing to be reasonably satisfied that either Kogan or Boo were “not likely to be a danger to the public or another animal” – section 53B(3) DAA
[26] Section 64 of the DAA applies where a dog has been seized under section 59, as it appears Kogan and Boo were, based on the T-documents (at pages 30-33)
[27] I note that the Original Tribunal, at paragraph 41 of the original decision, expressed a similar view
46.The appellant submitted that the Original Tribunal erred in failing to be reasonably satisfied that each dog is not likely to be a danger to the public or another animal, as specified in subsection 53B(3) of the DAA. Subsection 53B(4) of the DAA sets out matters the Registrar may consider in reaching a conclusion for subsection 53B(3), including ‘any other relevant matter’. The appellant submitted that the Original Tribunal had reached the wrong conclusion on a number of those matters, and that there were other errors which affected the conclusion that 53B(3) was not made out. The respondent conceded that there were some factual errors in the Original Tribunal’s decision, however submitted that these were minor and not material to the decision.
47.It is convenient to set out the areas of error asserted by the appellant under individual headings. Again, because of the nature of the question and the way the original hearing was conducted, the Appeal Tribunal is in as good a position as the Original Tribunal to consider whether section 53B(3) is satisfied. That task is best undertaken after consideration of the errors asserted by the appellant.
Failing to find that Kogan and Boo were provoked – section 53B(4)(a)(i)
48.At paragraph 27 of the original decision the Original Tribunal found that there was no evidence Khan provoked the attack.
49.The appellant’s submission about provocation was connected to his submission about the applicable standard of proof. He argued that if the Original Tribunal could be satisfied, without direct evidence, that Boo was involved in the attack, then it could also be satisfied that Khan provoked the attack. While there was some evidence that Khan had previously run up and down the fence,[28] the appellant said he did not rely on that as the provocation. Rather he asked the Original Tribunal, and now the Appeal Tribunal, to infer that Khan reacted aggressively or defensively when Kogan was in Khan’s backyard, thus provoking Kogan to attack.
[28] T-documents at page 65, [8]
50.The respondent submitted that the Original Tribunal was correct in stating that there was no evidence of provocation by Khan, and that it was correct in not finding that the attack was provoked.
51.Counsel for the Registrar pointed out that Mr Ehlers reported observing Boo and Kogan to display elevated behaviour simply at the sight of a German Shepherd dog on the other side of a fence, and without provocation. Counsel said this evidence supported an inference that similar behaviour occurred on the day of the attack. I have placed little weight upon this evidence: it is too tenuous to infer from this observation that Khan did not provoke Kogan or Boo.
52.The inference of provocation that the appellant says should be drawn is inconsistent with other established facts. Khan’s owner wrote in her witness statement “I believe that Khan would have been terrified and tried to get away from the attack as he is not an aggressive dog in any way and has always been a big teddy bear to our family.”[29] While in general it might be thought that a dog would act aggressively or territorially in its own backyard, the evidence in this case does not support this inference and I do not draw it.
[29] T-documents at page 63 (witness statement of Kathleen Fisher)
53.I do not think the Original Tribunal erred in failing to find that the attack was provoked.
Multiple factual errors
54.The respondent submitted that the Original Tribunal erred in finding that Kogan and Boo were Bull Arabs, when they were only partly Bull Arab, and in finding they were not registered in the Australian Capital Territory (ACT) when they were registered. However, an examination of the original decision shows that the Original Tribunal made no such findings. At paragraphs 1 and 8 of the original decision the Tribunal referred to Kogan and Boo as Bull Arab cross dogs. There was also no finding that the dogs were not registered in the ACT, and the transcript shows that this assertion, contained at a number of points in the T‑documents, was corrected at the hearing.[30] I do not find that the Original Tribunal made any factual errors in these two respects.
[30] Transcript of proceedings 2 October 2018 at page 16, line 25; page 17, line 17
55.At paragraphs 1, 7 and 24 of the original decision the Original Tribunal refers to the fence being two metres high,[31] and in paragraph 27 it is referred to as a ‘high wall’. However, on the appeal the appellant submitted, and the Registrar conceded, that the fence was a 1.6 metre high Colorbond fence.
[31] This information was replicated at a number of points in the T-documents, and was not corrected at the original hearing
56.At paragraph 24 of the original decision the Original Tribunal noted the investigation report contained in the T-documents detailed that Khan was larger and weighed more than Kogan and Boo. At paragraph 43 of the original decision the Original Tribunal “notes Khan’s size compared to the applicant’s dogs” in reinforcing the conclusion that Boo was involved in the attack on Khan. However, it was agreed by both parties on the appeal that the investigation report was incorrect, and both Boo and Kogan weighed more than Khan.
57.The appellant submitted that the effect of these two errors of fact on the Original Tribunal’s ultimate conclusion was unknown, and thus the entire decision should be set aside. For example, the appellant submitted that the findings that the fence was higher than it was, and that Khan was bigger than Kogan and Boo, might have led the Original Tribunal to conclude that Kogan and Boo were more ferocious or driven to attack than they are, and consequently led the Original Tribunal to consider they posed a greater risk to safety of other animals than they do.
58.The Registrar submitted that the two factual errors were not material and their correction would not have led to a different outcome. Consequently, the original decision should stand.
59.It is a difficult task to isolate the effect of these two factual errors on the Original Tribunal’s conclusion under 53B(3) of the DAA. However, they appear integral to the Original Tribunal’s reasoning. For this reason, in addition to the reasons as set out in paragraph 47 of this decision, it is appropriate in the appeal for the Appeal Tribunal to reach its own conclusion on whether section 53B(3) is satisfied, or not.
Insufficient weight given to the expert report of Mr Ehlers
60.The appellant submitted that in reaching its conclusion, the Original Tribunal gave insufficient weight to Mr Ehlers’ report. The appellant acknowledged that the assessment of Kogan and Boo had some shortcomings: it was conducted some seven months after the dogs had been seized, did not involve the dogs being tested in their usual environment, and did not include the appellant. However, the appellant submitted that, even with those shortcomings, Mr Ehlers determined that neither Boo nor Kogan was aggressive, posed a risk, or could not be returned with adequate safeguards in relation to their enclosure.
61.I do not consider this is a correct summary of Mr Ehlers’ report, although the appellant may have interpreted Mr Ehlers’ report in this way.
62.Mr Ehlers acknowledged that his assessment was not complete, for the reasons listed above. He assessed the dogs only once, on 1 August 2018, and wrote that “neither dog demonstrated overt signs of aggression” during his observation. He observed them both to be dominant, both appeared amicable and comfortable with people, and neither dog showed any real level of obedience. Mr Ehlers opined from their reaction to the German Shepherd dog that they had been socialised with other dogs. Mr Ehlers did not express a view whether Boo or Kogan is an aggressive dog, or not, or poses a risk to the public or other animals, or not.
63.Mr Ehlers recommended Kogan and Boo have training with their owner and an experienced dog behaviourist, be de-sexed, and be kept in an escape-proof yard. However, nowhere in his report does Mr Ehlers state that in his view these steps are adequate safeguards to permit Kogan or Boo to be returned to their owner.
64.The Original Tribunal referred to Mr Ehlers’ report at paragraphs 18 and 43 of the original decision. It is clear that the Original Tribunal was aware of Mr Ehlers’ report. The appellant also submits the Tribunal erred in giving insufficient weight to the appellant’s undertaking to do those things recommended by Mr Ehlers. However, the Original Tribunal referred to the appellant’s willingness to do those things in paragraphs 5, 15, 19, and 31 of the original decision. The original decision demonstrates that the Original Tribunal seriously considered this evidence in reaching its conclusion.
65.I do not consider that the Original Tribunal was in error in the way it dealt with Mr Ehlers’ report or the weight given to the appellant’s assurance that he would implement Mr Ehlers’ recommendations.
Failing to find reasonable steps could be taken to reduce the risk of the dog endangering the public and other animals – section 53B(4)(b) DAA
66.At paragraphs 31 and 32 of the original decision the Original Tribunal noted the appellant had given thought as to how he would keep the dogs in the future, but also noted that in the past an available dog run had not been used, and the dogs had been left unrestrained in the backyard when the appellant knew they could jump the fence. After referring to a number of decisions in which other tribunals explained why they had declined to accept undertakings as to how dogs would be kept in the future, the Original Tribunal concluded that there were no reasonable steps that could be taken to reduce the risk of the dogs endangering the public and other animals.
67.The appellant submitted that the Original Tribunal erred in reaching this conclusion, by giving insufficient weight to his assurances as to how the dogs would be kept, de-sexed and trained, and in relying on irrelevant precedents. I have noted, above, my view that the Original Tribunal did not give insufficient weight to the appellant’s assurances. It does however appear from the Original Tribunal’s reasons at paragraphs 33 to 36 of the original decision that the Original Tribunal interpreted the appellant’s submissions as being that the dogs should be returned to him simply upon him giving an undertaking as to how they would be kept. The legislation provides that a dog may be returned under a control order or dangerous dog licence. Indeed, the Original Tribunal later proceeded to consider whether it had the power to make a dangerous dog declaration, or grant a dangerous dog licence. The Original Tribunal’s focus at paragraphs 33 to 36 of the original decision on steps being taken outside the legislative framework may be indicative of an error in reaching the conclusion that there were no reasonable steps that could be taken to reduce the risk of the dog endangering the public and other animals. It is not necessary to reach a concluded view on this point, because it is appropriate for the Appeal Tribunal to form its own view, from the established facts, whether reasonable steps can be taken to reduce the risk of Kogan or Boo endangering the public or other animals.
68.There is no suggestion that either Kogan or Boo poses a risk to the public.
69.From the evidence I am satisfied that if Kogan and Boo were kept in the cages currently under construction, perhaps under a control order or pursuant to a dangerous dog licence, the risk of Kogan and Boo endangering other animals in their vicinity would be reduced.[32]
[32] Whether that risk will be reduced to an acceptable level is a different question to be dealt with under section 53B(3)
70.There is always a question as to whether such steps can be taken.[33] In Lau v Registrar Domestic Animals Act 2000 I noted that the tribunal would not contemplate a dog being kept in a confined manner, even pursuant to a control order, without expert evidence that this was appropriate for the particular dog.[34] This is especially so in a case where there is evidence that such confinement might be contrary to the dog’s welfare. In this case there is evidence that each dog may be discontent being confined — Kogan had to be untethered because of his barking and then he jumped out of his yard,[35] and Boo previously dug her way out of the run she was put in.
[33] While the Animal Welfare (Welfare of Dogs in the ACT) Code of Practice 2010 contemplates dogs may be kept in such structures, this is contingent upon arrangements for exercise and enrichment, and the needs of the individual animal
[34] [2018] ACAT 119 at [70]-[71]
[35] Kogan’s reluctance to remain in the yard must have been known, hence being the reason why he was tethered in the first place
71.I am satisfied that if Kogan and Boo were kept confined in cages the risk they pose to other animals in their neighbourhood will be reduced, but I am not satisfied that those steps can be taken in this case. Thus the Appeal Tribunal reaches the same conclusion on section 53B(4)(b) as the Original Tribunal, albeit for different reasons.
All things considered, did the Original Tribunal err in not being reasonably satisfied that each dog is not likely to be a danger to the public or another animal, as specified in subsection 53B(3) DAA?
72.Because of the nature of the question posed by section 53B(3) and the way in which the original hearing proceeded, it is appropriate that the Appeal Tribunal, in conducting the appeal by way of review, itself consider whether either Kogan or Boo is not likely to be a danger to the public or another animal. This approach must also be taken because the effect of the two factual errors on the Original Tribunal’s conclusion for section 53B(3) cannot be isolated. In undertaking the task, I have considered the factors listed in section 53B(4) and other relevant matters below.
73.The starting point is to understand what is meant by the phrase “the dog is not likely to be a danger to … another animal.”[36] Section 53B of the DAA uses this expression, while other provisions in Division 2.6 and 2.7 of the DAA rely on the concept of “unacceptable risk to the safety of … other animals”.
[36] There being no suggestion Kogan or Boo pose a risk to the public, I have focussed on the risk to animals in the following paragraphs
74.The ordinary English meaning of the word ‘danger’ as a noun is:
liability or exposure to harm or injury; risk; peril[37]
the possibility of suffering harm or injury[38]
[37] Macquarie Dictionary (Pan McMillan, 7th edn, 2017); Pollard, E and Liebeck, H (eds), The Oxford Paperback Dictionary (Oxford University Press, 4th edn, 1994)
[38] Oxford Living Dictionaries, viewed 8 April 2019 <en.exforddictionaries.com>
75.These definitions, when applied in the phrase “not likely to be a danger to … another animal” impose a different test than the phrase “unacceptable risk to the safety of … other animals”. Applying the definition within section 53B(3): the Registrar must be satisfied that the dog is not likely to be a possibility of harm or injury to … another animal. In effect, the ‘danger’ test specifies when the level of risk is unacceptable, setting it at a low level of possibility, whereas the ‘unacceptable risk’ test leaves assessment of the acceptable level of risk to the judgment of the Registrar. This may be intentional, because the ‘danger’ test is being applied where the Registrar is satisfied the dog has already killed an animal, or killed or seriously injured a person.
76.It is not clear from the DAA, or from the explanatory material, whether the test under section 53B is intended to have a different scope of operation to the more frequently used expression. It is not necessary to determine whether the tests are the same or not because, for the reasons set out below, I am not satisfied that Kogan or Boo would meet either test.
77.The appellant proposes that the dogs be returned to him, and it can be assumed they will be used for pig hunting. There was no assurance given to the Registrar or Original Tribunal, or to the Appeal Tribunal, that they would not be so used. In that respect, both Kogan and Boo are very likely, in my view almost certain, to pose a risk of harm or injury to an animal in the future.[39]
[39] This is because even if the hunt proceeds in a textbook fashion, the actions of the dogs holding the pig with their teeth are likely to cause the pig to suffer injuries such as torn or punctured ears or bite marks
78.By reason of their proposed future use in pig hunting alone, the Appeal Tribunal cannot be reasonably satisfied that either Boo or Kogan is not likely to be a danger to another animal.
79.For completeness, I have also considered the likelihood of Kogan or Boo posing a possibility of harm or injury to other animals, apart from their future involvement in pig hunting.
80.The appellant says he is willing to keep the dogs in the newly constructed cages, and continue to transport them in the cages on the back of his vehicle. He will have them de-sexed and, if required, undertake training with a dog behaviour expert. Mr Ehlers opined that the dogs appeared to have been socialised, and the multiple references attest to the dogs being well behaved and sociable with other dogs when with their owner. When seized by the rangers they walked quietly, off leash, to the van and jumped in when told.
81.The Registrar submitted that the dogs have already shown their willingness to go over a 1.6 metre fence and attack and kill another animal. Their response to the German Shepherd dog, when assessed by Mr Ehlers, was dominant: their hackles were raised. They did not show any real obedience when assessed by Mr Ehlers some seven months after being seized. The respondent submitted that the appellant might well move residence, leaving the dog cages behind. It was submitted that the appellant has been demonstrably deficient in the past in leaving the dogs in the care of a person who did not know they had to be kept confined or that they could jump over the fence, and may not have fed them properly.
82.Given that the dogs have attacked and killed an animal, and they have hunting experience, I consider that they are more likely to attack again, especially if similar circumstances arise. Although Kogan most likely was the instigator, the appellant’s submission that Boo ‘goes in hard’ leads the Appeal Tribunal to have grave concerns as to what Boo might do if she was ever to be roaming free and encountered an animal she regarded as prey or felt threatened by.
83.I am not satisfied that this attack was provoked, or that there were any other exculpatory circumstances as listed in section 53B(4).
84.I am satisfied that if Kogan or Boo gets loose again they are capable of entering most suburban backyards due to their ability to scale a fence or dig.[40]
[40] The basic urban fence is only 1.5 metres high and made of palings
85.I am satisfied that if Kogan or Boo gets loose again each is likely to attack and kill any animal they identify as prey or a threat. I have had regard to the references provided in relation to their past behaviour in the presence of their owner,[41] however it is not when the dogs are under the owner’s direct supervision that my concerns arise. There is no evidence that Boo or Kogan, either individually or together, will interact passively with any of the backyard animals[42] they can be expected to encounter if they were to get loose in the suburbs again. Their response to the German Shepherd dog leads me to have little confidence that if they get out, Boo or Kogan will avoid altercations.
[41] It was noted that they had not acted aggressively to other working dogs or livestock
[42] For example, Kogan and Boo were not assessed interacting freely with cats, chickens, rabbits or other typical backyard animals
86.Because of their likelihood to attack, and their ability to get out of their yard, if not kept under some sort of confinement or constraint, Kogan and Boo each pose a risk to other animals.
87.While the appellant proposes to keep the dogs confined, I am not satisfied that either Kogan or Boo can be kept in that way long term. But assuming they were to be kept in the cages, it remains possible that the dogs will get out, particularly if the appellant has to rely on a third party to care for them from time to time. Given the dogs appear to have years left to live, I expect escape from the cages into the backyard will occur on one or more occasions. If this happens, both Kogan and Boo have the ability to get out of the yard.
88.On the facts of this matter, the intention of the appellant was to leave the dogs for a number of days with Kogan tethered and Boo in a run. No arrangements for exercise or play were made, and on his return the appellant was uncertain as to whether the dogs had been properly fed in his absence. These facts demonstrate both the difficulty of ensuring that dogs kept in a confined manner on a long term basis will have their needs met, and the risks of escape where a third party has to be relied upon to keep them confined. I am not satisfied that Kogan or Boo either can, or will reliably, be kept confined if returned to the appellant.
89.The appellant submitted that it is not necessary for the Registrar to be satisfied that it is impossible for the dogs to ever cause injury or harm again. I agree. However the legislation requires that the Registrar must be reasonably satisfied that the dog is not likely to be a danger, that is, it is not likely to pose a possibility of harm or injury, to an animal. Having considered the matters set out above, I am not reasonably satisfied that either Boo or Kogan is not likely to be a danger to another animal.
90.Overall then, the Appeal Tribunal reaches the same view on section 53B(3) as the Original Tribunal. I do not consider the Original Tribunal was in error in failing to be satisfied that Kogan or Boo was not likely to be a danger to the public or another animal.
91.This means that the decision of the Original Tribunal, confirming the decision of the Registrar in relation to each dog, must be confirmed.
…………………………………..
Presidential Member MT Daniel
HEARING DETAILS
FILE NUMBER:
AA 54/2018 and AA 4/2019
PARTIES, APPELLANT:
Samuel Paul
PARTIES, RESPONDENT:
Registrar, Domestic Animals Act 2000
COUNSEL APPEARING, APPELLANT
N/A
COUNSEL APPEARING, RESPONDENT
Dr Jarvis
SOLICITORS FOR APPELLANT
N/A
SOLICITORS FOR RESPONDENT
ACT Government Solicitor
TRIBUNAL MEMBERS:
Presidential Member MT Daniel
DATES OF HEARING:
12 February 2019
16
0