In his July 6, 2015 decision in R. v. Hartman, Justice Doherty, for a unanimous Ontario Court of Appeal, quashed a conviction and ordered a new trial, confined to the issue of whether the accused should be convicted or found Not Criminally Responsible by Reason of Mental Disorder (“NCR-MD”). The Court admitted fresh evidence that could suggest the accused suffered from “sexsomnia” at the time of committing a sexual assault. The decision is a very thorough review and interesting application of:
- the test for admission of fresh evidence;
- appellate courts’ reviewing the competence of trial counsel; and
- the orders an appellate court can grant when quashing a conviction (e.g., substituting an acquittal or finding of NCR-MD, or ordering a new trial, possibly confined in scope).
The accused had been convicted of sexual assault before the Provincial Court. At trial and on appeal to the Summary Conviction Appeal Court, he denied sexually assaulting the complainant at all. Seeking leave to appeal to the Court of Appeal, he admitted that he had sexually assaulted the complainant, but argued that he was suffering from “sexsomnia” at the time, and should be found NCR-MD. Justice Doherty began his judgment by noting:
 New defences advanced years after the relevant events, only when other defences have been tried and failed and the convicted offender stands at the prison gate, must be viewed with considerable judicial suspicion. However, there will be cases in which the interests of justice require that an appellant be allowed to present a new defence on appeal. For the reasons that follow, this is one of those rare cases. I would grant leave to appeal, admit the fresh evidence, allow the appeal and order a new trial. I would direct that the new trial be limited to a determination of whether the appellant is NCR-MD or guilty.
Before the Court of Appeal, the accused tendered a report from a psychiatrist (“Dr. Gojer”) who opined that the accused was asleep when he sexually assaulted the complainant and therefore was not culpable. He also filed affidavits from himself, his mother and his girlfriend, explaining his history, and that of his family, of sleepwalking. Justice Doherty noted that if this fresh evidence were admitted, the appeal must be allowed. He summarized the applicable principles:
 Evidence is admissible on appeal if it is “in the interests of justice” to receive the evidence […]. When, as in this case, the evidence proffered on appeal challenges a factual finding essential to the verdict reached at trial, the admissibility of the evidence will depend on three inquiries […]:
- Is the proffered evidence admissible under the generally applicable rules of evidence governing criminal proceedings?
- Is the evidence sufficiently cogent to warrant its admission?
- What is the explanation for the failure to tender the evidence at trial, and should that explanation affect the admissibility of the evidence on appeal?
 The first two inquiries address prerequisites to admissibility. It will never be in the “interests of justice” to admit evidence on appeal that is not legally admissible under the normal rules of evidence. Nor will the “interests of justice” ever be served by admitting evidence that lacks sufficient cogency. The third inquiry does not go to a prerequisite to admissibility, but instead considers the explanation for the failure to lead the evidence at trial and whether that explanation compels the exclusion of the evidence even though it would otherwise be admissible on appeal. This inquiry, often referred to as the “due diligence” inquiry, recognizes that the “interests of justice” in s. 683(1) embrace administration of justice concerns such as finality that go beyond the specific interests of the appellant. For example, it may not be in the “interests of justice” to receive evidence on appeal, even though it is legally admissible and could reasonably be expected to have affected the result, if that evidence was available at trial and a tactical decision was made to not lead that evidence […]
 The cogency inquiry addresses three features of the evidence tendered on appeal. The proffered evidence must be relevant, in that it bears upon a potentially decisive issue at trial. The evidence must also be credible, in that it is reasonably capable of belief. Finally, the evidence must be sufficiently probative, in that it could reasonably be expected to have affected the result when taken with the other evidence adduced at trial and on appeal. On this appeal, the Crown and appellant’s disagreement centres on the probative value of Dr. Gojer’s opinion. They otherwise agree that his evidence is legally admissible, relevant and reasonably capable of belief.
“Due Diligence” and Ineffective Assistance of Counsel
In an effort to explain the failure to lead the evidence at trial, the accused argued that his trial lawyer provided ineffective assistance. The trial lawyer admitted she was unaware of the existence of the defence of sexsomnia. Justice Doherty noted that, though the defence is rare, “as a practising criminal lawyer, counsel should have been aware of its existence” (para. 24). However, he held that, on the facts known to her, counsel’s failure to investigate the defence did not amount to incompetence. Nonetheless, he did not feel that this should weigh against the accused’s fresh evidence application. In explaining why this was so, he provided useful guidance for assessing appellate claims of ineffective assistance of counsel, and how this can affect motions to admit fresh evidence:
 […] Trial lawyers, like medical clinicians, are trained to look for horses, not zebras. While some lawyers may have twigged to the possibility of a parasomnia defence, in the circumstances, I cannot conclude that it was incompetent for counsel to fail to recognize it as a potential defence.
 Although I would reject the appellant’s contention that he was the victim of ineffective assistance of counsel at trial, I do not think that the failure to lead the sexsomnia defence at trial should preclude the admissibility of the evidence on appeal. There is no suggestion that the failure to lead the sexsomnia defence was the result of a tactical decision made by trial counsel. Nor is there any suggestion that the appellant deliberately withheld the relevant information from trial counsel for some ulterior reason.
 The appellant did not tell his trial lawyer about the information that is said to be relevant to the sexsomnia defence because he did not appreciate the availability of that defence. Trial counsel did not make the inquiries that might have revealed that information, first, because she did not know that sexsomnia was a defence and, second, because there was nothing in the information given to her that would cause a reasonable lawyer who was aware of the sexsomnia defence to make inquiries into the availability of that defence. In those circumstances, nothing relating to the failure to lead evidence of sexsomnia at trial requires that evidence which would otherwise be admissible on appeal be excluded “in the interests of justice” […]
Admitting the Fresh Evidence on the Facts of this Case
Dr. Gojer found nine factors to suggest that the accused was asleep when he sexually assaulted the complainant. The Crown filed a responding affidavit by a Dr. Pressman, who concluded that the accused was not asleep but drunk when he sexually assaulted the complainant. The Crown vigorously attacked the credibility of Dr. Gojer’s opinion, by suggesting that many of the nine factors that formed the basis of his opinion were flawed if not demonstrably untrue. Moreover, Crown counsel emphasized that Dr. Pressman was, by any fair account, more familiar with sleep disorders than Dr. Gojer.
Justice Doherty held:
 To succeed on the appeal, the appellant does not have to convince the court that Dr. Gojer’s opinion should be accepted and preferred over Dr. Pressman’s. If the appellant satisfies this court that a reasonable trier of fact could, in the context of the entirety of the record, including Dr. Pressman’s opinion, accept Dr. Gojer’s opinion that the appellant was asleep on a balance of probabilities, the appeal must be allowed and the matter returned to the trial court. It would be for the trial court to decide whether Dr. Gojer’s evidence should ultimately be accepted.
 I think the appellant has cleared that hurdle. […]
Justice Doherty agreed that many of the attacks on Dr. Gojer’s opinion had force. However:
 Despite these criticisms of Dr. Gojer’s opinion, I am satisfied the appellant has met his burden on the fresh evidence application. A reasonable trier of fact might well disagree with the criticisms outlined above. Furthermore, apart from those criticisms, there was believable evidence that the appellant had a family history of parasomnia, a personal history of parasomnia and associated sleep disorders, and that he had on other occasions engaged in sexual activity while asleep […]
 R.C.’s statement to the police and her evidence at trial also provided some independent support for the appellant’s sexsomnia claim. […]
 For the sake of completeness, I should add that counsel for the appellant also made a vigorous and effective attack on several aspects of Dr. Pressman’s opinion. […]
 My analysis of the proffered evidence, considered in the context of the trial record, leads to the conclusion that there are valid reasons to accept or reject the opinions of Dr. Gojer and Dr. Pressman. That conclusion compels the finding that the appellant has met his onus of showing that a reasonable trier of fact could accept Dr. Gojer’s opinion and find on the balance of probabilities that the appellant was asleep at the time of the assault. That finding renders the verdict a miscarriage of justice. The evidence should be received under s. 683(1) and the conviction quashed.
When an appellate courts quashes a conviction, it can substitute an acquittal, substitute a finding of NCR-MD, or order a re-trial. It was agreed that it would have been inappropriate for the Court of Appeal to make a finding of NCR-MD in this case given the need to have further findings of fact. Moreover, sexsomnia cannot lead to an outright acquittal. The real question, therefore, was whether the Court of Appeal “could and should” order a re-trial confined to the question of whether the accused should be found guilty or NCR-MD. Justice Doherty explained this area of the law and applied it to these facts:
 When this court allows an appeal and orders a new trial pursuant to s. 686(2) of theCriminal Code, the court may, under s. 686(8), “make any order, in addition, that justice requires.” In Warsing, at paras. 72-74, the Supreme Court held that where an NCR-MD defence was raised for the first time on appeal from conviction, the court could not, in ordering a new trial based on the evidence of mental disorder, rely on s. 686(8) to limit the new trial to a choice between a verdict of NCR-MD and a verdict of guilty. In so holding, the Supreme Court concluded that an order foreclosing the possibility of an acquittal on the new trial would interfere with the accused’s right to control his own defence and the presumption of innocence, and could not be in the interests of justice. In Warsing, the appellant wanted an opportunity at the second trial to secure an acquittal on the merits.
 In Luedecke, this court considered the Warsing analysis of the scope of s. 686(8) in circumstances that have some similarity to the present case. In Luedecke, the Crown appealed from an acquittal based on a sexsomnia defence. The court held that the appeal should succeed as the trial judge had misapplied Stone in entering an acquittal, as opposed to a finding of NCR-MD. The respondent (the accused) argued that the new trial should be limited to a choice between NCR-MD and not guilty. He took the position that the trial judge’s finding that his actions were not voluntary should stand and that he should not be exposed to the risk of a conviction at a new trial. This court accepted this submission holding, at paras. 137-38:
The submissions made on behalf of the respondent have convinced me that s. 686(8) does authorize an order ancillary to an order for a new trial made under s. 686(4) limiting the scope of the new trial. That ancillary order may be made only where it does not interfere with any of the accused’s rights and is otherwise consistent with the demands of justice in the circumstances.
 Through counsel, the appellant has acknowledged the assault on R.C. in this court. He does not seek the opportunity to gain an acquittal at the new trial, but rather seeks an opportunity to show that he was not criminally responsible for the assault on R.C. In light of the position taken on appeal, I do not think an order limiting the new trial to a choice between NCR-MD and guilty would offend the presumption of innocence or interfere with the accused’s right to control his own defence. To the contrary, as in Luedecke, an order limiting the scope of the new trial would reflect the manner in which the appellant wishes to conduct his defence at the new trial.
 Finally, I think an order limiting the new trial would serve the administration of justice in one additional way. R.C. has asserted from the moment it happened that she was sexually assaulted by the appellant. Now, after many years and contrary to the position taken throughout, the appellant acknowledges that he did indeed sexually assault R.C. He seeks an opportunity to show that he was not criminally responsible for that action because he was asleep. For the reasons set out above, I think the appellant should have that opportunity.
 That opportunity is not, however, in any way compromised by an acknowledgement of the validity of R.C.’s complaint. An order limiting the new trial to a determination of whether the appellant was NCR-MD or guilty as charged would provide that acknowledgement to the complainant and remove the possibility, however small, that at the new trial the validity of R.C.’s assertion that she was sexually assaulted would once again be tested, despite the admission made on behalf of the appellant in this court.