R. v. Chaulk, [1990] 3 S.C.R. 1303

Robert Matthew Chaulk and

Francis Darren Morrissette Appellants

v.

Her Majesty The Queen Respondent

and

The Attorney General of Canada,

the Attorney General for Ontario,

the Attorney General of Quebec,

the Attorney General for New Brunswick

and the Attorney General for Alberta Interveners

indexed as: r. v. chaulk

File Nos.: 21012 and 21035.

1990: May 29, 30; 1990: December 20.

Present: Dickson C.J.* and Lamer C.J.** and Wilson, La Forest, L'Heureux‑Dubé, Sopinka, Gonthier, Cory and McLachlin JJ.

Constitutional law -- Charter of Rights -- Presumption of innocence -- Accused presumed sane until contrary is proved -- Insanity to be proved by accused on balance of probabilities -- Whether s. 16(4) of Criminal Code infringes s. 11(d) of Canadian Charter of Rights and Freedoms -- If so, whether s. 16(4) justifiable under s. 1 of Charter.

Criminal law -- Defences -- Insanity -- Person insane under s. 16(2) of Criminal Code if suffering from disease of the mind rendering him incapable of knowing act is wrong -- Meaning of word "wrong" in s. 16(2) of Code.

Criminal law -- Defences -- Insanity -- Delusions -- Whether s. 16(3) of Criminal Code constitutes independent insanity defence -- Whether trial judge's instructions to jury adequate.

Evidence -- Rebuttal evidence -- Crown adducing evidence of accused's sanity in rebuttal -- Whether evidence of accused's sanity should have been adduced by Crown as part of its case-in-chief.

The accused were convicted of first degree murder. The only defence raised at trial was insanity, but this defence was rejected by the jury. The Court of Appeal upheld the conviction. This appeal is to determine (1) whether s. 16(4) of the Criminal Code, which provides that "Every one shall, until the contrary is proved, be presumed to be and to have been sane", infringes the presumption of innocence guaranteed in s. 11(d) of the Canadian Charter of Rights and Freedoms; and, if so, whether s. 16(4) is justifiable under s. 1 of the Charter; (2) whether the meaning of the word "wrong" in s. 16(2) of the Code should be restricted to "legally wrong"; (3) whether s. 16(3) of the Code provides an alternative defence if the conditions of s. 16(2) were not met; and (4) whether the trial judge erred in permitting the Crown to split its case by presenting its evidence with respect to the sanity of the accused in rebuttal.

Held (L'Heureux-Dubé, Sopinka and McLachlin JJ. dissenting): The appeal should be allowed and a new trial ordered.

(1) Presumption of Sanity/Presumption of Innocence

Per Dickson C.J. and Lamer C.J. and La Forest, Sopinka and Cory JJ.: The insanity defence under s. 16 of the Code should be characterized as an exemption from criminal liability which is based on an incapacity for criminal intent. This claim for an exemption will usually be manifested under s. 16 either as a denial of mens rea in the particular case or as an excuse for what would otherwise be a criminal offence.

Section 16(4) of the Code infringes the presumption of innocence guaranteed in s. 11(d) of the Charter. The real concern under s. 11(d) is not whether the accused must disprove an element or prove an excuse, but whether an accused may be convicted while a reasonable doubt exists. When that possibility exists, there is a breach of the presumption of innocence. Accordingly, it is the final effect of the impugned provision on the verdict that is decisive. Whether the claim of insanity is characterized as a denial of mens rea, an excusing defence or, more generally, an exemption based on criminal incapacity, s. 16(4) allows a factor which is essential for guilt to be presumed, rather than proved by the Crown beyond a reasonable doubt. Moreover, the section requires an accused to disprove sanity (or prove insanity) on a balance of probabilities. Section 16(4) therefore violates the presumption of innocence because it permits a conviction in spite of a reasonable doubt in the mind of the trier of fact as to the guilt of the accused. Finally, while the verdict under s. 16 is "not guilty by reason of insanity", the accused raising the s. 16 defence is seeking a "true acquittal" vis‑à‑vis the attachment of criminal culpability and is entitled to the presumption of innocence.

Section 16(4) of the Code constitutes a reasonable limit on the presumption of innocence. The objective of s. 16(4), which is a purely evidentiary section, is to avoid placing on the Crown the impossibly onerous burden of disproving insanity and to thereby secure the conviction of the guilty. This objective is of sufficient importance to warrant limiting a constitutionally protected right. The means chosen by the government are proportional to the objective. First, the presumption of sanity and the reverse onus embodied in s. 16(4) are rationally connected to the objective. Second, s. 16(4) impairs s. 11(d) as little as possible. To reduce the burden on the accused to a mere evidentiary burden would not achieve the objective as effectively. While s. 16 is seldom raised, given the substantial constraint on liberty which follows a successful insanity plea, if insanity were easier for an accused to establish, the defence would be successfully invoked more often. Parliament may not have chosen the absolutely least intrusive means of meeting the objective, but it has chosen from a range of means which impair s. 11(d) as little as is reasonably possible. It is not the role of this Court to second‑guess the wisdom of policy choices made by Parliament. Third, there is proportionality between the effects of the measure and the objective. The burden on the accused is not the full criminal burden; rather, the accused is required to prove his insanity on a balance of probabilities. Section 16(4) represents a compromise of three important societal interests: avoiding a virtually impossible burden on the Crown; convicting the guilty; and acquitting those who truly lack the capacity for criminal intent. The alternatives to this compromise raise their own Charter problems and give no guarantee as to whether they will achieve the objective.

Per L'Heureux-Dubé, Gonthier and McLachlin JJ.: The insanity provisions of the Code relate to the basic preconditions for criminal responsibility rather than to essential elements or defences to criminal offences. The latter approach ignores the historical and philosophical origins of the fundamental precept of the criminal law system that the attribution of criminal responsibility and punishment is justifiable only for those who have the capacity to reason and choose between right and wrong. The accused must be sane before any consideration of the essential elements of the offence or exculpatory defences becomes relevant. This approach also violates the language of s. 16 of the Code, which refers to capacity for criminal responsibility rather than actual states of mind. It is at odds with the fact that insanity in s. 16 can be raised by the Crown in circumstances where neither the elements of the offence nor a defence are at issue. Further, it confuses true acquittal, the result of the absence of an essential element of an offence or the presence of a defence to it, with formal acquittal coupled with alternative coercive measures because mental impairment renders the imposition of true penal responsibility inappropriate.

The presumption of sanity in s. 16(4) of the Code, viewed as the fundamental precondition of criminal responsibility, does not offend the presumption of innocence embodied in s. 11(d) of the Charter, nor does it offend the fundamental notion of procedural fairness which underlies the procedural guarantees of the Charter. The presumption of sanity merely relieves the Crown from establishing that the accused has the capacity for choice which makes attribution of criminal responsibility and punishment justifiable. The Crown must still prove the accused's guilt -- i.e., the actus reus, the mens rea, and the absence of exculpatory defences raised on the evidence -- beyond a reasonable doubt. The presumption of innocence reflects the fundamental precepts upon which our legal system and our Charter are based. There is no suggestion that the present system results in injustice. Accused persons appear to have no difficulty establishing insanity on a balance of probabilities where it exists.

Per Wilson J.: Section 16(4) of the Code infringes s. 11(d) of the Charter. The presumption of sanity requires the accused to establish his insanity on a balance of probabilities. However one conceives the plea of insanity, whether as an exemption, a defence, a justification or an excuse, the persuasive burden imposed on the accused by s. 16(4) permits him to be convicted of a crime notwithstanding a reasonable doubt as to his guilt. Under s. 11(d), it is the net effect of a reverse onus provision on the final guilt or innocence of the accused rather than the precise nature of the provision that must be examined.

Section 16(4) of the Code does not constitute a reasonable and demonstrably justified limit on the presumption of innocence under s. 1 of the Charter. The objective of s. 16(4) is to prevent perfectly sane persons who have committed crimes to escape criminal liability on tenuous insanity pleas. But nothing indicates that successfully fabricated insanity pleas have given rise to an existing pressing and substantial concern. While the legislature may not necessarily wait until such a concern has arisen, the Crown has not succeeded in establishing even a likelihood of its arising. The American experience does not support the contention that a lower standard of proof would result in more people being acquitted by reason of insanity. Further, several reports from Canada and other countries propose that the burden of proving insanity should be made an evidential one. This burden on the accused is seen as a sufficiently high threshold to prevent insanity pleas in cases where there is only tenuous support for such a plea.

In any event, s. 16(4) does not meet the proportionality test. It is rationally connected to the legislative objective but does not impair the accused's right to be presumed innocent as little as is reasonably possible. It is inappropriate in this case to apply the deferential standard of review on the "minimal impairment" issue. The government does not mediate between different groups but acts as the singular antagonist of a very basic legal right of the accused. The government's objective in this case could be met by imposing a purely evidentiary burden on the accused. The possibility of a successful feigned insanity plea diminishes as scientific knowledge about mental illness increases. The burden on the prosecution, while it would not be an easy one, would nevertheless not be an "impossible onerous burden". In the absence of s. 16(4) the onus would still be on the accused to adduce evidence that makes insanity a live issue fit and proper to be left to the jury. The burden on the Crown would then be to remove any doubt in the jury's mind as to the presence of any of the elements of insanity as set out in ss. 16(2) and 16(3). There is no hard evidence that putting a lower burden on the accused would result in more guilty people escaping criminal liability on tenuous insanity pleas. Section 16(4) is accordingly not saved by s. 1 of the Charter. It follows that the common law principle which is reflected in s. 16(4) also infringes s. 11(d) and is not saved by s. 1.

(2) Meaning of "Wrong"

Per Dickson C.J. and Lamer C.J. and Wilson, La Forest, Gonthier and Cory JJ.: The word "wrong" in s. 16(2) should be interpreted to mean "morally wrong" and not "legally wrong". This Court's judgment in Schwartz v. The Queen, 1976 165 (SCC), [1977] 1 S.C.R. 673, is overruled. Under s. 16(2), the courts must determine whether the accused, because of a disease of the mind, was rendered incapable of knowing that the act committed was something that he ought not to have done. To do so, the inquiry cannot terminate with the discovery that the accused knew that the act was contrary to the formal law. A person may well be aware that an act is contrary to law but, by reason of disease of the mind, is at the same time incapable of knowing that the act is morally wrong in the circumstances according to the moral standards of society. The trial judge, therefore, erred in directing the jury that the insanity defence was not available under s. 16(2) of the Code if the accused knew, at the time of committing the offence, that the act was contrary to the laws of Canada. In view of the trial judge's error, a new trial should be ordered.

Per L'Heureux-Dubé, Sopinka and McLachlin JJ. (dissenting): The question under s. 16(2) of the Code is not whether the accused thought an act was morally or legally wrong but whether he was incapable of knowing that he ought not to do it. If the accused is capable of knowing that the act was in some sense "wrong", then it is neither unfair nor unjust to submit the accused to criminal responsibility and penal sanction. The word "wrong" in s. 16(2) should therefore be construed to mean simply that which one "ought not to do", for whatever reasons, legal or moral. The wording of s. 16(2), the history and the purpose of the insanity provisions, as well as the difficulty of determining in every situation what is morally wrong, support that conclusion.

(3) Specific Delusions

Per Dickson C.J. and Lamer C.J. and La Forest, Gonthier and Cory JJ.: As a result of this Court's reconsideration of the meaning of the word "wrong" in s. 16(2), any successful attempt to invoke the insanity defence under s. 16(3) of the Code would also succeed under s. 16(2). Furthermore, if an accused fails to satisfy the conditions set out in s. 16(2), he will not be able to benefit from s. 16(3). It would not, therefore, assist an accused in any way if s. 16(3) was held to constitute a separate and independent defence.

Per L'Heureux-Dubé, Sopinka and McLachlin JJ.: Assuming that s. 16(3) of the Code relating to specific delusions can arise where the conditions of s. 16(2) are not met, the trial judge left open this possibility to the jury and his instructions were adequate.

Per Wilson J.: While most cases of specific delusions under s. 16(3) of the Code will now be caught under the second branch of s. 16(2), there may still be "specific delusions" which are not necessarily attributable to a "disease of the mind". Since legislative provisions should not readily be held to be redundant, particularly ones which are favourable to the accused, it is preferable to leave s. 16(3) open to an accused who might not qualify for the insanity defence under the second branch of s. 16(2).

(4) Splitting the Case

Per Dickson C.J. and Lamer C.J. and Wilson, La Forest, Gonthier and Cory JJ.: The trial judge correctly permitted the Crown to adduce evidence with respect to sanity in rebuttal. While the Crown must tender, as part of its case, evidence that will establish the existence of all elements of the offence with which the accused is charged, the Crown need not adduce evidence in chief to challenge a defence that an accused might possibly raise. This principle is not altered by the fact that the accused may warn the Crown that it intends to raise a particular defence. Further, a requirement that the Crown adduce evidence in chief to establish the sanity of the accused would defeat the presumption contained in s. 16(4). Finally, since the accused were given the opportunity for surrebuttal in this case, they did not suffer prejudice as a result of the fact that the Crown's evidence was adduced in rebuttal rather than in chief.

Per L'Heureux-Dubé, Sopinka and McLachlin JJ.: The Crown was entitled to rely on the presumption of sanity and was under no obligation to present evidence on this issue as part of its case against the accused. The evidence on insanity did not relate to an essential element of the offence.

Cases Cited

By Lamer C.J.

Overruled: Schwartz v. The Queen, 1976 165 (SCC), [1977] 1 S.C.R. 673; applied: R. v. Whyte, 1988 47 (SCC), [1988] 2 S.C.R. 3; R. v. Oakes, 1986 46 (SCC), [1986] 1 S.C.R. 103; distinguished: R. v. Schwartz, 1988 11 (SCC), [1988] 2 S.C.R. 443; referred to: R. v. Godfrey (1984), 1984 3052 (MB CA), 11 C.C.C. (3d) 233; Smythe v. The King, 1940 384 (SCC), [1941] S.C.R. 17; R. v. Simpson (1977), 1977 1142 (ON CA), 35 C.C.C. (2d) 337; Rabey v. The Queen, 1980 44 (SCC), [1980] 2 S.C.R. 513; R. v. Abbey 1982 25 (SCC), [1982] 2 S.C.R. 24; R. v. Holmes, 1988 84 (SCC), [1988] 1 S.C.R. 914; Woolmington v. Director of Public Prosecutions, [1935] A.C. 462; R. v. Edwards Books and Art Ltd., 1986 12 (SCC), [1986] 2 S.C.R. 713; Irwin Toy Ltd. v. Quebec (Attorney General), 1989 87 (SCC), [1989] 1 S.C.R. 927; Reference re ss. 193 and 195.1(1)(c) of the Criminal Code (Man.), 1990 105 (SCC), [1990] 1 S.C.R. 1123; M'Naghten's Case (1843), 10 Cl. & Fin. 200, 8 E.R. 718; R. v. Codere (1916), 12 Cr. App. R. 21; R. v. Windle, [1952] 2 Q.B. 826; Stapleton v. The Queen (1952), 86 C.L.R. 358; R. v. Bernard, 1988 22 (SCC), [1988] 2 S.C.R. 833; R. v. Budic (No. 3) (1978), 1978 2404 (AB SCAD), 43 C.C.C. (2d) 419; R. v. Bruno (1975), 1975 1240 (ON CA), 27 C.C.C. (2d) 318.

By Wilson J.

Applied: R. v. Oakes, 1986 46 (SCC), [1986] 1 S.C.R. 103; R. v. Whyte, 1988 47 (SCC), [1988] 2 S.C.R. 3; distinguished: R. v. Schwartz, 1988 11 (SCC), [1988] 2 S.C.R. 443; Irwin Toy Ltd. v. Quebec (Attorney General), 1989 87 (SCC), [1989] 1 S.C.R. 927; referred to: Clark v. The King (1921), 1921 603 (SCC), 61 S.C.R. 608; M'Naghten's Case (1843), 10 Cl. & Fin. 200, 8 E.R. 718; Woolmington v. Director of Public Prosecutions, [1935] A.C. 462; R. v. Vaillancourt, 1987 2 (SCC), [1987] 2 S.C.R. 636; R. v. Holmes, 1988 84 (SCC), [1988] 1 S.C.R. 914; Law Society of Upper Canada v. Skapinker, 1984 3 (SCC), [1984] 1 S.C.R. 357; Davis v. United States, 160 U.S. 469 (1895); Ortwein v. Commonwealth, 76 Pa. 414 (1874); In Re Winship, 397 U.S. 358 (1970); Mullaney v. Wilbur, 421 U.S. 684 (1975); Jackson v. Virginia, 443 U.S. 307 (1979); Leland v. Oregon, 343 U.S. 790 (1952): Rivera v. Delaware, 429 U.S. 877 (1976), aff'g 351 A.2d 561 (1976); United States v. Pasarell, 727 F.2d 13 (1984), certiorari denied, 105 S. Ct. 107 (1984); United States v. Voice, 627 F.2d 138 (1980); United States v. Samuels, 801 F.2d 1052 (1986); People v. Stockwell, 242 N.W.2d 559 (1976); United States of America v. Cotroni, 1989 106 (SCC), [1989] 1 S.C.R. 1469; R. v. Gill (1963), 47 Cr. App. R. 166; R. v. Abbey, 1982 25 (SCC), [1982] 2 S.C.R. 24; People v. Krugman, 141 N.W.2d 33 (1966).

By McLachlin J. (dissenting)

Rabey v. The Queen, 1980 44 (SCC), [1980] 2 S.C.R. 513; R. v. Simpson (1977), 1977 1142 (ON CA), 35 C.C.C. (2d) 337; R. v. Saxell (1980), 1980 1623 (ON CA), 59 C.C.C. (2d) 176; R. v. Oakes, 1986 46 (SCC), [1986] 1 S.C.R. 103; R. v. Whyte, 1988 47 (SCC), [1988] 2 S.C.R. 3; Schwartz v. The Queen, 1976 165 (SCC), [1977] 1 S.C.R. 673; M'Naghten's Case (1843), 10 Cl. & Fin. 200, 8 E.R. 718; R. v. Codere (1916), 12 Cr. App. R. 21; R. v. Windle, [1952] 2 Q.B. 826; Stapleton v. The Queen (1952), 86 C.L.R. 358.

Statutes and Regulations Cited

Canadian Charter of Rights and Freedoms, ss. 1, 7, 8, 11(d).

Criminal Code, 1892, S.C. 1892, c. 29, s. 11.

Criminal Code, R.S.C., 1985, c. C-46, ss. 13, 16, 17, 614(2).

Young Offenders Act, R.S.C., 1985, c. Y-1.

Authors Cited

American Bar Association. Standing Committee on Association Standards for Criminal Justice. First Tentative Draft: Criminal Justice Mental Health Standards. Washington: American Bar Association, 1983.

Boisvert, Anne-Marie. "Psychanalyse d'une défense: réflexions sur l'aliénation mentale" (1990), 69 Can. Bar Rev. 46.

Canada. Law Reform Commission. Report 31. Recodifying Criminal Law. Ottawa: The Commission, 1987.

Canada. Law Reform Commission. The Presumption of Innocence in the Draft Code of Substantive Criminal Law. Unpublished paper prepared by Patrick Healy, 1986.

Canada. Law Reform Commission. Working Paper 29. Criminal Law: The General Part -- Liability and Defences. Ottawa: Minister of Supply and Services Canada, 1982.

Canada. Royal Commission on the Law of Insanity as a Defence in Criminal Cases. Report of the Royal Commission on the Law of Insanity as a Defence in Criminal Cases. Hull: Queen's Printer, 1956.

Charles, William H., Thomas A. Cromwell and Keith B. Jobson. Evidence and the Charter of Rights and Freedoms. Toronto: Butterworths, 1989.

Colvin, Eric. "Exculpatory Defences in Criminal Law" (1990), 10 Oxford J. Legal Stud. 381.

Colvin, Eric. Principles of Criminal Law. Toronto: Carswells, 1986.

Ferguson, G. "A Critique of Proposals to Reform the Insanity Defence" (1989), 14 Queen's L. J. 135.

Fortin, Jacques et Louise Viau. Traité de droit pénal général. Montréal: Thémis, 1982.

Great Britain. Criminal Law Revision Committee, Eleventh Report. Evidence (General), Cmnd. 4991. London: H.M.S.O., 1972.

Hodgins, Sheilagh et al. Annual Report, Year 1 Canadian Data Base: Patients Held on Lieutenant Governors' Warrants, 1989.

Keilitz, Ingo and Junius P. Fulton. The Insanity Defense and its Alternatives: A Guide for Policy makers. National Center for States Courts, 1984.

Kenny's Outlines of Criminal Law, 19th ed. By J. W. Cecil Turner. Cambridge: Cambridge University Press, 1966.

Klinck, Dennis R. ""Specific Delusions" in the Insanity Defence" (1983), 25 Crim. L.Q. 458.

Mahoney, Richard. "The Presumption of Innocence: A New Era" (1988), 67 Can. Bar Rev. 1.

Matthews Jr., Arthur R. Mental Disability and the Criminal Law: A Field Study. Chicago: American Bar Foundation, 1970.

McWilliams, Peter K. Canadian Criminal Evidence, 3rd ed. Aurora, Ont.: Canada Law Book Ltd., 1990.

Mewett, Alan W. "Insanity, Criminal Law and the Charter" (1989), 31 Crim. L.Q. 241.

Mewett, Alan W. "Section 16 and `Wrong'" (1976), 18 Crim. L.Q. 413.

Mewett, Alan W. and Morris Manning. Criminal Law, 2nd ed. Toronto: Butterworths, 1985.

Perkins, Rollin M. and Ronald N. Boyce. Criminal Law, 3rd. ed. Mineola, N.Y.: Foundation Press, Inc., 1982.

Quen, Jacques M. "Anglo-American Concepts of Criminal Responsibility: A Brief History". In Stephen J. Hucker, Christopher D. Webster and Mark H. Ben-Aron, eds., Mental Disorder and Criminal Responsibility. Toronto: Butterworths, 1981.

Rogers, Richard and R. E. Turner, "Understanding of Insanity: A National Survey of Forensic Psychiatrists and Psychologists" (1987), 7 Health L. Can. 71.

Schiffer, Marc. Mental Disorder and the Criminal Trial Process. Toronto: Butterworths, 1978.

Simon Rita J. and David E. Aaronson. The Insanity Defense: A Critical Assessment of Law and Policy in the Post-Hinckley Era. New York: Praeger, 1988.

Stuart, Don. Canadian Criminal Law, 2nd ed. Toronto: Carswells, 1987.

Taschereau, Henri Elzéar. The Criminal Code. Toronto: Carswells, 1893.

Weihofen, Henry. Mental Disorder as a Criminal Defense. Buffalo: Dennis & Co., 1954.

Ziskin, Jay and David Faust. Coping with Psychiatric and Psychological Testimony, vol. I, 4th ed. Marina del Rey, Calif.: Law and Psychology Press, 1988.

APPEAL from a judgment of the Manitoba Court of Appeal (1988), 1988 84 (SCC), 4 W.C.B. (2d) 218, [1988] Man. D. 5400-03, dismissing the appellants' appeal from their conviction on a charge of first degree murder. Appeal allowed and new trial ordered, L'Heureux-Dubé, Sopinka and McLachlin JJ. dissenting.

John Scurfield, Q.C., for the appellant Chaulk.

G. G. Brodsky, Q.C., for the appellant Morrissette.

George Dangerfield, Q.C., for the respondent.

S. R. Fainstein, Q.C., for the intervener the Attorney General of Canada.

R. Libman, for the intervener the Attorney General for Ontario.

Jacques Gauvin, for the intervener the Attorney General of Quebec.

Bruce Judah, for the intervener the Attorney General for New Brunswick.

Michael Watson, for the intervener the Attorney General for Alberta.

The judgment of Dickson C.J. and Lamer C.J. and La Forest and Cory JJ. was delivered by

Lamer C.J. -‑ This case involves a constitutional challenge, under s. 11(d) of the Canadian Charter of Rights and Freedoms, to the reverse onus clause contained in s. 16(4) of the insanity provisions contained in the Criminal Code, R.S.C., 1985, c. C-46. This Court has also been asked to revisit its interpretation of the meaning of the word "wrong" found in s. 16(2). There are other issues specific to the appeal which are set out further on in these reasons.

Facts

On September 3, 1985, the appellants Chaulk and Morrissette entered a home in Winnipeg, plundered it for valuables and then stabbed and bludgeoned its sole occupant to death. A week later they turned themselves in, making full confessions.

After a transfer proceeding in the Youth Court (Chaulk and Morrissette were 15 and 16 years of age, respectively), the appellants were tried and convicted of first degree murder by a jury in the Manitoba Court of Queen's Bench. The only defence raised was insanity within the meaning of s. 16 of the Code. Expert evidence was given at trial that the appellants suffered from a paranoid psychosis which made them believe that they had the power to rule the world and that the killing was a necessary means to that end. They knew the laws of Canada existed, but believed that they were above the ordinary law; they thought the law was irrelevant to them. They thought they had a right to kill the victim because he was "a loser".

An appeal to the Manitoba Court of Appeal was unanimously dismissed on May 13, 1988.

Relevant Statutory Provisions

Criminal Code

16. (1) No person shall be convicted of an offence in respect of an act or omission on his part while that person was insane.

(2) For the purposes of this section, a person is insane when the person is in a state of natural imbecility or has disease of the mind to an extent that renders the person incapable of appreciating the nature and quality of an act or omission or of knowing that an act or omission is wrong.

(3) A person who has specific delusions, but is in other respects sane, shall not be acquitted on the ground of insanity unless the delusions caused that person to believe in the existence of a state of things that, if it existed, would have justified or excused the act or omission of that person.

(4) Every one shall, until the contrary is proved, be presumed to be and to have been sane.

Canadian Charter of Rights and Freedoms

1. The Canadian Charter of Rights and Freedoms guarantees the rights and freedoms set out in it subject only to such reasonable limits prescribed by law as can be demonstrably justified in a free and democratic society.

2. Any person charged with an offence has the right

(d) to be presumed innocent until proven guilty according to law in a fair and public hearing by an independent and impartial tribunal

Lower Court Judgments

Manitoba Court of Queen's Bench

The appellants were tried before Ferg J., sitting with a jury, in the Court of Queen's Bench. They were convicted of first degree murder. The only defence raised at trial was insanity and this defence was rejected by the jury. Chaulk and Morrissette were sentenced to life imprisonment without eligibility for parole for a period of twenty‑five years.

Manitoba Court of Appeal (O'Sullivan J.A. for the Court)

The appellants appealed their conviction to the Manitoba Court of Appeal on the grounds that s. 16(4) violated the Charter; that the trial judge had erred in his charge to the jury on s. 16(3) and on the effect of mental impairment on the ability to plan and deliberate; that the trial judge had erred in refusing to answer specific questions of the jury regarding ss. 16(2) and 16(3); that the trial judge erred in not ordering separate trials for the two accused and in admitting certain statements by the accused Chaulk to a counsellor without a voir dire; and that the trial judge erred in refusing a defence motion for particulars.

The Court of Appeal was of the view that Ferg J.'s charge to the jury and his answers to their queries "fairly set out the issues for the jury's consideration". The court held that any reasonable jury would, if properly instructed, reject the defence of insanity in this case. O'Sullivan J.A. stated:

In my opinion, the evidence shows that the accused were suffering from megalomania, but it is clear that they knew and fully appreciated the nature and consequences of their acts and they knew that what they were doing was legally wrong (see R. v. Abbey (1982), 1982 25 (SCC), 29 C.R. (3d) 193). There was no evidence of any specific delusions.

In my opinion, medically insane people can be made responsible for their acts if the insanity does not fall within the provisions of s. 16 of the Criminal Code.

The court rejected the argument that the onus provisions of s. 16(4) of the Code contravened s. 11(d) of the Charter. Relying on its judgment in R. v. Godfrey (1984), 1984 3052 (MB CA), 11 C.C.C. (3d) 233, the court held that s. 16(4) was not inconsistent with the Charter. O'Sullivan J.A. was of the opinion that the argument against the presumption of sanity was potentially threatening to liberty, given that insanity may be raised by the Crown or by the judge. He stated that if there were no presumption of sanity, "it would be open to a jury to condemn a prisoner to incarceration as an insane person if they had a reasonable doubt as to his sanity".

As regards the admission of Chaulk's statement in the absence of a voir dire, the Court of Appeal held that the statement had not been made to a person in authority and found no reversible error in Ferg J.'s decision to admit the statement without a voir dire.

Issues

The following constitutional questions were stated by Dickson C.J. on July 13, 1989:

1.Is s. 16(4) of the Criminal Code of Canada inconsistent with s. 11(d) of the Canadian Charter of Rights and Freedoms?

2.If the answer to question 1 is yes, is s. 16(4) justified by s. 1 of the Canadian Charter of Rights and Freedoms and therefore not inconsistent with the Constitution Act, 1982?

The appellants raise three further issues in this Court:

3.What is the interpretation of the word "wrong" which is found in s. 16(2) of the Criminal Code of Canada? Is its meaning restricted to "legally wrong" or may it also be interpreted more broadly to include "legally or morally wrong"?

4.What interpretation should be given to s. 16(3) of the Criminal Code of Canada? Should the section have been left to the jury in the case at bar, and if so, what are the necessary jury instructions?

5.[Did] the learned trial judge err in permitting the Crown to split its case and call the bulk of its case in purported rebuttal?

Analysis

Is s. 16(4) of the Code inconsistent with s. 11(d) of the Charter?

Section 16(4) of the Code sets out a presumption of sanity. This presumption can be rebutted if "the contrary is proved". In Smythe v. The King, 1940 384 (SCC), [1941] S.C.R. 17, this Court held that when insanity is raised by the defence, the accused must prove that he or she was insane, at the time of the offence, on a balance of probabilities. In R. v. Simpson (1977), 1977 1142 (ON CA), 35 C.C.C. (2d) 337 (Ont. C.A.), Martin J.A., citing Smythe at p. 363, stated:

It is, of course, well established that where the accused alleges that he was insane at the time of the commission of the act the onus on the accused to prove insanity is discharged by proof of insanity on the balance of probabilities: ...

The appellant Morrissette argues that the words "until the contrary is proved" in s. 16(4) should be interpreted, based on common law, so as to require an accused to merely raise a reasonable doubt as to his insanity. He assumes that if the words were given this interpretation, s. 16(4) would not violate the presumption of innocence and it would not be necessary to address the Charter arguments in this case.

In my view, the words "until the contrary is proved" cannot be interpreted as requiring an accused merely to discharge an evidentiary burden (i.e., raise a reasonable doubt as to insanity); the words in s. 16(4) clearly impose a persuasive burden on the accused. In R. v. Oakes, 1986 46 (SCC), [1986] 1 S.C.R. 103, Dickson C.J. discussed the different types of presumptions which exist in criminal law in the context of deciding whether s. 8 of the Narcotic Control Act placed an evidentiary or a persuasive burden on the accused. The provision in question required an accused to "establish" that he was not in possession of a narcotic for the purpose of trafficking. Dickson C.J. stated that the phrase "to establish" is equivalent to the phrase "to prove", and that the use of the word "establish" implied that the accused had to prove, on a balance of probabilities, that he or she was not in possession of a narcotic for the purpose of trafficking (pp. 114‑18). It is clear from the reasoning in Oakes that a provision which presumes the existence of some fact (in this case, sanity) "until the contrary is proved", places a burden on an accused wishing to rebut this fact to prove the contrary (insanity) on a balance of probabilities.

Accordingly, it is necessary, in this case, to measure s. 16(4) against s. 11(d) of the Charter.

The appellants argue that the requirement that an accused person prove his or her insanity on a balance of probabilities is contrary to the presumption of innocence, guaranteed by s. 11(d) of the Charter. In order to resolve this question one must consider the nature of the insanity provisions in our criminal law.

The Nature of the Insanity Provisions

There is controversy in academic circles as to whether insanity operates to negate a "condition precedent" to legal responsibility or whether it operates as a defence in the sense that it negates mens rea. Let me begin by saying that I will use the term "defence" to describe a claim of insanity under s. 16, but only in the broad, general sense of the term which means "any answer which defeats a criminal charge" (Law Reform Commission of Canada, Working Paper 29, Criminal Law: The General Part -- Liability and Defences (1982), at p. 35), or "any claim which, if accepted, would necessitate an acquittal" (Colvin, Principles of Criminal Law (1986), at p. 163).

It is true that the exact nature of the insanity defence is very difficult to ascertain. The wording of s. 16 itself is not particularly helpful in this regard. It tells us that a claim of insanity, if proved, will preclude a conviction, but it does not tell us whether a claim of insanity negates mens rea, provides an excuse or justification, or whether it simply exempts an accused from a criminal conviction on policy grounds. Other Code provisions are more explicit; for example, s. 17 tells us that a person who commits an offence under compulsion of certain threats will be excused for committing the offence in certain circumstances.

It is helpful to note, however, that s. 16(1) is worded in a very similar fashion to s. 13, which provides:

13. No person shall be convicted of an offence in respect of an act or omission on his part while that person was under the age of twelve years.

This is a variation on the common law, which provided that a child under the age of seven was presumed irrebuttably to be incapable of possessing criminal intent and that a child between the ages of seven and fourteen was likewise presumed incapable unless there was clear proof of "precocity" so as to establish a real appreciation of the wrong done (see Perkins and Boyce, Criminal Law (3rd ed. 1982), c. 8). Once an accused reached the age of fourteen, the common law presumption of criminal incapacity disappeared and was replaced by a rebuttable presumption of capacity for criminal intent. This accords with the presumption of sanity which existed at common law (enunciated via the M'Naghten Rules), was adopted into Canadian criminal law in the first criminal code (see The Criminal Code, 1892, S.C. 1892, c. 29, s. 11, and accompanying commentary by Taschereau J. (The Criminal Code (1893, at pp. 8-9)), and continues in the current s. 16(4).

In other words, the nature of the insanity defence is revealed if one views the changing presumptions regarding criminal capacity as a continuum. At common law, this continuum began with an irrebuttable presumption that a child under the age of seven could not have the capacity for criminal intent. Our current Code s. 13 provides for an irrebuttable presumption that a child under the age of twelve has no criminal capacity. At common law, the continuum provided for a rebuttable presumption of incapacity for children between the ages of seven and fourteen. Perkins and Boyce state "[t]his presumption is extremely strong at the age of seven and diminishes gradually until it disappears entirely at the age of fourteen" (p. 936). The current Criminal Code cuts off the presumption at age twelve; after a person reaches the age of twelve the presumption of sanity in s. 16(4) comes into play. Thus, at this end of the continuum, individuals are presumed to have criminal capacity until such presumption is rebutted on a balance of probabilities (of course, the Young Offenders Act, R.S.C., 1985, c. Y-1, which incorporates a concept of diminished responsibility, applies to young people between the ages of twelve and eighteen).

While the state of insanity and the state of childhood cannot be equated, the connection between these two situations for the purpose of criminal law is apparent. What these two situations have in common is that they both indicate that the individual in question does not accord with some basic assumptions of our criminal law model: that the accused is a rational autonomous being who is capable of appreciating the nature and quality of an act and of knowing right from wrong. With respect to the state of childhood, these basic assumptions are brought into question because of the immaturity of the individual -‑ he or she has not yet developed the basic capacity which justice and fairness require be present in a person who is being measured against the standards of criminal law. With the state of insanity, these basic assumptions are brought into question because the accused is suffering from some disease of the mind or from some delusions which cause him or her to have a frame of reference which is significantly different than that which most people share. This mental condition means that the accused is largely incapable of criminal intent and should not, therefore, generally be subject to criminal liability in the same way that sane people are. (I note here that s. 16 does not exempt all people with a disease of the mind from criminal liability. The insanity defence is defined in a particular way and only if an accused meets those criteria will his or her mental condition preclude a finding of guilt.)

The foregoing discussion indicates, in my view, that the insanity provisions operate, at the most fundamental level, as an exemption from criminal liability which is predicated on an incapacity for criminal intent. However, in particular cases, this basic incapacity may manifest itself in a number of different ways depending on the claims put forward by the accused. A claim of insanity, with its underlying claim of criminal incapacity, could give rise to a denial of the actus reus or of the mens rea in a particular case. For example, an accused could claim that his or her mental condition is such that when the alleged crime took place, he or she was not acting consciously. This is akin to a claim of insane automatism which denies the essential element of voluntary actus reus on the basis of an internal cause -‑ the accused's disease of the mind (Rabey v. The Queen, 1980 44 (SCC), [1980] 2 S.C.R. 513). An accused could also raise the argument that his or her mental condition was such that while he or she was acting consciously and voluntarily, he or she did not have the requisite mens rea. For example, a person charged with murder could claim that while he consciously and voluntarily did the act of chopping, he thought that he was chopping a loaf of bread in half, when, in fact, he was chopping off the victim's head (see Kenny's Outlines of Criminal Law (19th ed. 1966), at p. 83, n. 1). In such a case, the insanity claim is manifested as a denial of mens rea. The accused had no intention to bring about the consequence of death. In yet another case, an accused, charged with murder, could argue that while she consciously and voluntarily did the act of killing and while she desired to bring about the death of the victim, she did so because her mental condition was such that she honestly believed that the victim was evil incarnate and would destroy the earth if the accused did not kill him. In such a case, the insanity claim is manifested not as a denial of actus reus or mens rea, but rather as a defence in the nature of an excuse or a justification based on the fact that the accused's mental condition rendered her incapable of knowing that the act was wrong. Professor Eric Colvin makes reference to the different roles played by the insanity defence in "Exculpatory Defences in Criminal Law" (1990), 10 Oxford J. Legal Stud. 381. He states (at pp. 394 and 401):

The traditional approach to the insanity defence has been to conceive of it as negativing all culpability. Insanity can, of course, operate either as a special way of denying a mental element of the offence or as an exculpatory defence. . . .

Deficient mental capacity may provide the basis for defences of several different kinds. Impairment may be used to deny definitional elements of an offence, or to claim an exculpatory defence which is specifically geared to the problem of impairment, or to make a special claim to one of the exculpatory defences which are based on contextual permission.

All three types of defence can be illustrated by the M'Naghten Rules on insanity at common law. The central proposition in the M'Naghten Rules was that the defence would be available to someone who, because of a defect of reason' resulting from disease of the mind', did not know the nature and quality of the act he was doing; or if he did know it, . . . he did not know he was doing what was wrong'. The first alternative deals with the situation where insanity negatives mental elements in the definitions of offences. The second alternative establishes a special exculpatory defence which is based on lack of capacity for normative understanding. There is, in effect, an exception to the general rule that ignorance of the law is no excuse'. The third way in which insanity could be relevant to criminal culpability is through a cognitive breakdown leading to a mistaken belief in a matter of contextual permission. This situation was covered in the M'Naghten Rules by a ruling that responsibility would be determined as if the facts were as they were believed to be. [Citations omitted.]

The foregoing examples illustrate that the insanity defence can be raised in a number of different ways, depending on the mental condition of the accused. All of these examples have one thing in common however. Each is based on an underlying claim that the accused has no capacity for criminal intent because his or her mental condition has brought about a skewed frame of reference. When a person claims insanity, he or she may well be denying the existence of mens rea in the particular case or putting forward an excuse which would preclude criminal liability in the particular case; but he is also making a more basic claim which goes beyond mens rea or actus reus in the particular case -‑ he is claiming that he does not fit within the normal assumptions of our criminal law model because he does not have the capacity for criminal intent. Such a claim may or may not be successful. If the incapacity is such that it fits into the defence of insanity encompassed in s. 16, it will preclude a conviction.

Based on the foregoing, I prefer to characterize the insanity defence as an exemption to criminal liability which is based on an incapacity for criminal intent. I note, however, that this basic claim for an exemption will usually be manifested under s. 16 either as a denial of mens rea in the particular case or as an excuse for what would otherwise be a criminal offence; this is because of the way that s. 16 is worded.

The definition of "legal insanity", or insanity which will preclude a criminal conviction, is found in ss. 16(2) and 16(3) of the Code. The first part of s. 16(2) has come to be known as the "first arm" or "first branch" of the insanity defence:

[A] person is insane when the person is in a state of natural imbecility or has disease of the mind to an extent that renders the person incapable of appreciating the nature and quality of an act or omission . . .

In R. v. Abbey, 1982 25 (SCC), [1982] 2 S.C.R. 24, this Court held (per Dickson J., as he then was) that "appreciat[e] the nature and quality of an act or omission" refers to an accused's ability to perceive the consequences, impact, and results of a physical act and not to an accused's ability to appreciate that the legal consequences of an act are applicable to him or her. Dickson J. stated, in coming to this conclusion, that a delusion falling under the "first branch" of the insanity defence negatives the element of mens rea, with respect to either the circumstances or consequences which form part of the actus reus. Given the wording of the "first branch" of s. 16(2), this is true in the vast majority of cases. A claim of insanity under the "first branch" will be manifested as a denial of mens rea in the particular case. It is possible, however, that such a claim could also manifest itself as a denial of voluntary actus reus in the particular case. This was clearly established by the decision of this Court in Rabey which discussed, inter alia, the defence of insane automatism.

This Court also held in Abbey that the "second branch" of s. 16(2) "is concerned with cognitive capabilities, with knowledge, and not with appreciation of consequences" (p. 36). This provision is also based on incapacity for criminal intent. Such incapacity arises because the accused, due to his mental condition, is incapable of distinguishing between right and wrong. This claim of incapacity does not manifest itself as a denial of mens rea in the particular case. The criminal law is not concerned with whether a sane accused knew that his act was wrong. Knowledge of wrongness is not part of the requirement of mens rea. This is because sane people are presumed to have the capacity to distinguish between right and wrong ‑- if a sane person is of the opinion that murder is not wrong, his opinion makes him "bad" (as opposed to "sick") because he has the capacity to distinguish right from wrong. However, if an accused makes a claim of insanity under the "second branch" of s. 16(2), he is challenging the assumption that he is capable of distinguishing between right and wrong. If it is proved that his mental condition brought about such incapacity, he will be excused from criminal liability despite the elements of actus reus and mens rea being established in the particular case. Thus, while a claim of insanity under the "second branch" is based on the same basic denial of criminal capacity upon which a claim under the "first branch" is based, it is manifested not as a denial of actus reus or mens rea, but rather as a claim to be excused for what would otherwise be criminal behaviour.

Based on the foregoing analysis, I owe it to a certain number of parties in this appeal to address a particular argument which they raised, to the effect that the presumption of innocence is not violated by s. 16(4) because an accused has the benefit of s. 11(d) until the Crown has proved both actus reus and mens rea, and it is only after this point that the claim of insanity arises. A number of parties have concluded that this means that if an accused fails to satisfy the jury that he or she was insane within the meaning of s. 16, it is not that failure which leads to a conviction, but rather the other evidence about which the jury must be satisfied beyond a reasonable doubt.

(I pause here to note that this appeal was heard along with a number of other related appeals. This case and R. v. Romeo, [1991] 1 S.C.R. 000, and R. v. Ratti, [1991] 1 S.C.R. 000, all raise the issue of whether s. 16(4) violates the Charter. Accordingly, in addressing this constitutional issue which is common to them all, I have considered and will refer to the arguments made by all parties (including the various interveners) who submitted argument on this point.)

The factum of the Attorney General for New Brunswick in Romeo expresses the argument as follows:

67. While undoubtedly there is a connection between mens rea and the presumption of sanity, the function each serves in the criminal law process is fundamentally different. The insanity plea is raised only when mens rea has been established. In such circumstances, it becomes necessary for the defendant to establish that the intent was that of a diseased mind. The intent remains, it is not negated. The mental state giving rise to that intent is explained.

68. The presumption of innocence means that the prosecution must establish, beyond a reasonable doubt, both the actus reus and mens rea of the particular crime. The burden arising on an insanity plea assumes that the prosecution has established these elements of the offence.

69. ... All the presumption does is to require the accused to establish a particular mental disorder -‑ after the prosecution has established mens rea. If the accused fails to satisfy the jury of that disorder on a balance of probability [sic], he will be convicted not because he is presumed sane but because the prosecution has otherwise proved the essential elements of the offence -‑ including mens rea.

In my respectful opinion, this argument should be rejected for two reasons. First, it is not necessarily true that the insanity plea is raised only after mens rea has been established by the Crown. What if an accused's mental condition is such that it operates to negate mens rea in the particular case? The example set out above, in which a person consciously and voluntarily chops off the victim's head, but believes, because of his mental condition, that he is chopping a loaf of bread in half, illustrates that evidence of insanity could go to negate mens rea in a particular case. In such a case, if the accused were to raise evidence of his mental condition (thereby putting his mental capacity in issue), the trial judge would be entitled to charge the jury on s. 16. In these circumstances, it is only when the trier of fact has rejected the defence of insanity that it may consider the evidence of his mental condition solely with respect to mens rea; this, in turn, has only been allowed in cases where an accused is seeking to deny either the element of planning and deliberation or the specific intent for murder, and to instead be found guilty of a lesser, included offence (i.e., second degree murder or manslaughter). Thus, it is not true that evidence of insanity can be raised by an accused simply to deny mens rea for an offence independent of the insanity defence.

Accordingly, it is not correct, in my respectful view, to state that the issue of insanity arises only after both actus reus and mens rea have been proved by the Crown. If the accused's insanity puts the existence of mens rea for the particular offence into question, it cannot be said that mens rea has been proved by the Crown before the issue of insanity arises. The Crown cannot be said to have proved anything beyond a reasonable doubt until the end of the trial. Evidence raised by the accused may well give rise to a reasonable doubt about the existence of an essential element, despite the fact that the Crown has raised enough evidence such that a no‑evidence motion would fail.

Thus, I cannot agree that "the insanity plea is raised only when mens rea has been established" (emphasis added). In cases where a claim of insanity is manifested as a denial of the requisite mens rea, it cannot be said that mens rea has been established until after the insanity claim has been raised and has failed.

The second problem with this argument is that even if it were true that the insanity plea is raised only when mens rea has been established, it does not mean that a conviction following a failed insanity claim arises because the Crown has proved the essential elements of the offence and not because the accused could not discharge the burden to prove his or her insanity. Such an argument assumes that the only necessary ingredients for guilt are the proof of actus reus and mens rea. However, this is not true even if s. 16 is ignored for the time being. If evidence is raised in the course of a trial for murder that the accused may have been acting in self‑defence, the Crown must disprove the existence of the self‑defence justification beyond a reasonable doubt in order for a conviction to follow. If the Crown is able to sufficiently negate the claim of self‑defence, it cannot be said that the resulting conviction follows from the proof of actus reus and mens rea and not from the failed self-defence claim. Thus, it could not be said that a reverse onus clause on self-defence would not intrude on s. 11(d); yet such a conclusion would logically follow from the argument put forward by the Attorney General for New Brunswick and others.

The above reasoning also applies to the defence of insanity. Section 16 makes clear that sanity is essential for guilt (that is, it is essential for guilt that an accused not fall within the definition of insanity in s. 16). If a person was insane at the time the offence was committed, he or she shall not be convicted. Thus, even if the Crown has proved actus reus and mens rea, if a plea of insanity is raised and fails because the accused cannot discharge the presumption of sanity, guilt must follow from the presumption of sanity and not merely from proof of the "essential elements". The practical implications of this last point will be discussed in greater detail below.

While it is helpful to consider the nature of a claim arising under the insanity provisions in order to gain perspective on the issues in this case, the question arises whether a conclusion on this issue is determinative in the s. 11(d) analysis.

Does the Nature of s. 16 Determine the Charter Issue?

The Crown has argued that sanity is not an essential element of an offence (as is mens rea) and, therefore, does not engage the presumption of innocence.

The Crown takes the position that insanity raises a question of mental capacity, which is not the same as denying mens rea in a particular case. The insanity defence denies that the accused is a "moral agent" and asserts that he or she should therefore be exempt from liability under the criminal law. This position largely accords with the foregoing analysis of the insanity defence as a denial of criminal capacity. However, I am not satisfied that this characterization of the insanity provisions necessarily leads to the conclusion that the presumption of innocence is not violated by the presumption of sanity.

The Crown argues that insanity does not "disprove" an offence (nor, I assume, an essential element of an offence) and that the presumption of innocence is therefore inapplicable to s. 16. The Crown relies on the judgment of McIntyre J. in R. v. Holmes, 1988 84 (SCC), [1988] 1 S.C.R. 914, to the effect that if a statute requires the accused to prove a defence on a balance of probabilities where the Crown has already proved the offence beyond reasonable doubt there is no violation of s. 11(d). Implicit in this argument is the assumption that the insanity provisions provide an exemption for an accused who, but for his insanity, would be found guilty beyond a reasonable doubt. This may or may not be true, depending on the way in which the claim of insanity is raised ‑- as can be seen from the examples set out above. The Crown submits that the insanity claim is a means of escaping criminal liability "above and beyond" the presumption of innocence which applies in respect of the specific offence. The Crown is essentially arguing that the presumption of innocence only operates with respect to essential elements of the offence and with respect to common‑law defences. A claim of insanity under s. 16 raises a claim for an exemption to criminal liability; it neither negates an essential element of the offence nor raises a common law defence (justification or excuse). Therefore, the requirement embodied in s. 16(4) that the accused prove insanity on a balance of probabilities does not violate the presumption of innocence.

In my view, this argument does not accord with the principles enunciated by this Court in R. v. Whyte, 1988 47 (SCC), [1988] 2 S.C.R. 3. In Whyte, the constitutionality of s. 237(1)(a) of the Code was in issue. Section 237(1)(a) raised a presumption of care or control of an automobile upon a finding of occupancy of the driver's seat, with respect to the offence of impaired driving. Dickson C.J. stated (at p. 18):

The real concern is not whether the accused must disprove an element or prove an excuse, but that an accused may be convicted while a reasonable doubt exists. When that possibility exists, there is a breach of the presumption of innocence.

The exact characterization of a factor as an essential element, a collateral factor, an excuse, or a defence should not affect the analysis of the presumption of innocence. It is the final effect of a provision on the verdict that is decisive. If an accused is required to prove some fact on the balance of probabilities to avoid conviction, the provision violates the presumption of innocence because it permits a conviction in spite of a reasonable doubt in the mind of the trier of fact as to the guilt of the accused. The trial of an accused in a criminal matter cannot be divided neatly into stages, with the onus of proof on the accused at an intermediate stage and the ultimate onus on the Crown. Section 237(1)(a) requires the accused to prove lack of intent on a balance of probabilities. If an accused does not meet this requirement the trier of fact is required by law to accept that the accused had care or control and to convict. But of course it does not follow that the trier of fact is convinced beyond a reasonable doubt that the accused had care or control of the vehicle. [Emphasis added.]

In my view, the principles enunciated in Whyte are applicable to this case and establish that the presumption of sanity embodied in s. 16(4) violates the presumption of innocence. If an accused is found to have been insane at the time of the offence, he will not be found guilty; thus the "fact" of insanity precludes a verdict of guilty. Whether the claim of insanity is characterized as a denial of mens rea, an excusing defence or, more generally, as an exemption based on criminal incapacity, the fact remains that sanity is essential for guilt. Section 16(4) allows a factor which is essential for guilt to be presumed, rather than proven by the Crown beyond a reasonable doubt. Moreover, it requires an accused to disprove sanity (or prove insanity) on a balance of probabilities; it therefore violates the presumption of innocence because it permits a conviction in spite of a reasonable doubt in the mind of the trier of fact as to the guilt of the accused.

Before leaving this point, let us examine the recent decision of this Court in R. v. Schwartz, 1988 11 (SCC), [1988] 2 S.C.R. 443. In Schwartz, this Court held that s. 106.7(1) of the Code, which required an accused to prove that he had a firearms acquisition certificate in order to avoid a conviction for the offence of possession of a restricted weapon for which the possessor has no registration certificate, did not violate s. 11(d) of the Charter. The Crown has relied on the majority reasons of McIntyre J. to support its argument that s. 16(4) does not violate the presumption of innocence because it sets up an exemption rather than a defence.

In my view, the reasons of McIntyre J. in Schwartz indicate that the principles enunciated by this Court in Whyte were not operative in that case. The result in Schwartz stemmed from a finding, based on the particular facts of the case, which could not have been made in Whyte; thus, the decision in Schwartz in no way overrules the decision in Whyte. In Schwartz, McIntyre J. set out the ratio of Whyte, and stated (at p. 485):

In my view, however, these principles cannot be of assistance to the appellant here. There is no reverse onus imposed upon the accused by s. 106.7(1), despite the words which are employed in the section. The holder of a registration certificate cannot be made subject to a conviction under s. 89(1). He is not required to prove or disprove any element of the offence or for that matter anything related to the offence.

Subsequent statements of McIntyre J. indicate that the conclusion that s. 106.7(1) imposed no reverse onus was based on the finding that, because of the particular factual nature of the offence, it could not be said that an accused who was unable to establish that he was the holder of a registration certificate could be convicted under s. 89(1) despite the existence of a reasonable doubt on that issue. In other words, the majority's view was that production or non‑production of the registration certificate was conclusive of whether the accused had a registration certificate.

Conversely, in Whyte, this Court held that proof that an accused occupied the driver's seat of a motor vehicle was not determinative of whether the accused had care and control of the vehicle. Therefore, the statutory presumption contained in s. 237(1)(a) of the Code was such that an accused could be convicted in spite of a reasonable doubt in the mind of the trier of fact as to his guilt. Consequently, s. 237(1)(a) was found to violate the presumption of innocence.

Accordingly, the principles enunciated in Whyte remain intact and are applicable to this case, unless this case falls within the narrow ratio of Schwartz. In my view, this case does not fall within the ratio of Schwartz. The presumption of sanity embodied in s. 16(4) operates differently than do the statutory presumptions which were at issue in Whyte and Schwartz. The presumption of sanity is not an inference drawn from the existence or proof of an underlying fact (such as being unable to produce a registration certificate for a restricted weapon, or being found in the driver's seat of a motor vehicle). Rather, sanity is something which is assumed from the outset (see Charles, Cromwell and Jobson, Evidence and the Charter of Rights and Freedoms (1989), at pp. 130‑31). Thus, in enacting s. 16(4), Parliament has not designated any basic "fact" which, when established, removes all possibility of reasonable doubt as to lack of insanity (and therefore removes all reasonable doubt as to guilt). Parliament has simply reversed the onus on a factor which is essential for guilt. If an accused cannot discharge the persuasive burden with respect to his insanity, the trier of fact may well be obliged to convict the accused despite the existence of a reasonable doubt as to sanity, and therefore, as to guilt.

A number of parties have raised the argument that a successful claim of insanity under s. 16 does not result in a finding of "innocence" and in the accused going free, but rather results in a special verdict and in the accused being detained in a psychiatric institution under a warrant of the Lieutenant Governor via s. 614(2) (formerly s. 542(2)) (the "L.G.W. system"). Because the result of a s. 16 claim is not a "true acquittal", the argument is made that the presumption of innocence is inapplicable.

The Attorney General for New Brunswick has made this argument as has the Attorney General of Quebec and the Attorney General for Alberta. The Attorney General for Alberta argues that a finding of insanity under s. 16 is neither a determination of guilt nor innocence, but rather is a qualification of a "not guilty" verdict. The Attorney General for Alberta notes that an accused who is found to be insane is put into the "treatment phase" of the Code rather than the "punishment phase" and argues that because a verdict of "not guilty by reason of insanity" is "clearly not a full acquittal in the traditional sense", it is necessary for an accused to prove his insanity so that a "positive finding" will underlie the special detention. The Attorney General for Alberta argues that the burden of proof embodied in s. 16(4) does not operate so as to leave a reasonable doubt remaining as to "innocence". He states "section 16(4) does not raise issue [sic] that the accused may be convicted despite a reasonable doubt that his actions were lawful".

This argument seems to amount to a claim that the detention under the "L.G.W. system", which follows from a finding of "not guilty by reason of insanity", is a result which an accused who raises s. 16 is positively seeking. In other words, an accused who is asking for the "benefit" of a finding of insanity properly bears the burden of proving that he falls within the criteria for special detention. In my view, this argument misses the essential point. An accused who raises s. 16 is raising it in order to avoid being convicted of a criminal offence. He is negating criminal culpability, on the basis of his mental condition, and is not necessarily "seeking" the special detention. The special detention is something which the criminal law imposes upon him for various policy reasons. It is incorrect to say that the accused is choosing special detention; he is seeking to negate his criminal culpability and the result of doing that (over which the accused has no control) is that he is subjected to the "L.G.W. system".

The Attorney General for Alberta has stated that a finding of "not guilty by reason of insanity" is not a finding as to "innocence". In my view, an accused raising insanity is asserting that he is incapable of having criminal intent and is, therefore, not criminally culpable ‑- this is why the verdict under s. 16 is "not guilty by reason of insanity". The accused is alleging that he is "innocent" vis‑à‑vis criminal culpability; he is alleging that he is not criminally responsible. Under s. 16, the criminal law may well be determining what sort of response is appropriate in the circumstances (as the Attorney General for Alberta has argued), but surely we are, in the process, determining that the response of attaching criminal culpability is not appropriate. In this sense, an accused raising s. 16 is, indeed, seeking a "true acquittal" vis‑à‑vis the attachment of criminal culpability and is, therefore, entitled to the presumption of innocence guaranteed by s. 11(d) of the Charter.

Finally, a number of parties have argued that s. 16(4) does not violate the Charter because the presumption of sanity is necessary to protect the accused when the Crown raises the issue of insanity. The appellants have argued that such reasoning erroneously assumes that the burden of proof for insanity must necessarily be the same no matter who raises s. 16. Assuming, without deciding, that the Crown may constitutionally raise the issue of insanity, I agree with the appellant Chaulk that there is no reason why the burden of proof on the accused and on the Crown must necessarily be the same. The argument put forward by the Crown, along with other interveners, is valid only if such an assumption is made.

In summary, I find that the presumption of sanity embodied in s. 16(4) of the Code limits the presumption of innocence guaranteed by s. 11(d) of the Charter. I turn now to a consideration of whether s. 16(4) can nonetheless be justified as a reasonable limit under s. 1 of the Charter.

Is s. 16(4) a reasonable limit under s. 1 of the Charter?

There is no question that the presumption of innocence, guaranteed by s. 11(d) of the Charter, is a fundamental legal right which plays a very important role in our criminal justice system. It has been referred to as "the golden thread" of English criminal law (Woolmington v. Director of Public Prosecutions, [1935] A.C. 462 (H.L.), at pp. 481‑82). However, like the other rights and freedoms guaranteed by the Charter, it is subject to limitations under s. 1 of the Charter. The procedure to be followed when the state is attempting to justify a limit on a right or freedom under s. 1 was set out by this Court in Oakes, supra:

1. The objective of the impugned provision must be of sufficient importance to warrant overriding a constitutionally protected right or freedom; it must relate to concerns which are pressing and substantial in a free and democratic society before it can be characterized as sufficiently important.

2. Assuming that a sufficiently important objective has been established, the means chosen to achieve the objective must pass a proportionality test; that is to say they must:

(a) be "rationally connected" to the objective and not be arbitrary, unfair or based on irrational considerations;

(b) impair the right or freedom in question as "little as possible"; and

(c) be such that their effects on the limitation of rights and freedoms are proportional to the objective.

Objective

The parties have put forward several suggestions as to the characterization of the objective of s. 16(4). The Attorney General for New Brunswick argues that the presumption of sanity embodied in s. 16(4) ensures that blame or responsibility is not placed on individuals who are not aware of or do not appreciate the nature of the acts that they have carried out, and provides the means by which insane individuals can be treated, rather than punished. "The insanity provisions also provide a shield to enable the individual to avoid the stigma attached to criminal conviction". The Attorney General for Ontario makes a similar argument, stating that the presumption of sanity plays an integral part in recognizing the distinction between those accused persons with a blameworthy state of mind and those individuals who, due to a condition of mental disorder, are in need of treatment.

In my opinion, with respect for contrary views, these characterizations of the objective miss the point. The factors cited by these parties are more connected to the objectives of s. 16(1) than to s. 16(4). In enacting s. 16(1), Parliament has made a law that no person shall be convicted of an offence committed while that person was insane; the objective of this law is to avoid stigmatizing and punishing, as criminals, people who are "sick" as opposed to "bad". In enacting s. 16(4), Parliament has made a law that any accused who alleges that he was insane at the time of the offence must prove his insanity on a balance of probabilities. Section 16(4) was not enacted to further the objective of s. 16(1); if anything, the onus imposed by s. 16(4) means that some "sick" people will be incorrectly stigmatized and punished as "bad" and, therefore, s. 16(4) actually hinders the objective cited by the Attorneys General for New Brunswick and Ontario.

Before characterizing the relevant objective, I feel that it is necessary to focus on what it is about s. 16(4) which violates s. 11(d). The problem with s. 16(4) is two‑fold. First, the provision allows sanity (something which is essential to guilt) to be presumed. This violates the basic principle (set out in Oakes) that the state bears the burden of proving guilt beyond a reasonable doubt. Second, the provision requires an accused to prove his or her insanity, on a balance of probabilities, in order to rebut the presumption of sanity. This gives rise to a reversal of the burden of proof such that an accused could be found guilty of a criminal offence despite a reasonable doubt in the mind of the trier of fact about the accused's insanity.

It is therefore necessary to focus on the presumption and on the reversal of the burden of proof and ask: what was Parliament's specific objective in forcing an accused who raises s. 16 to prove his or her insanity on a balance of probabilities? The answer can only be that Parliament wished to avoid placing on the Crown the impossibly onerous burden of disproving insanity and to thereby secure the conviction of the guilty (who are not "sick") by defeating acquittals based on a doubt as regards insanity. Drawing an analogy to my (dissenting) reasons in Schwartz, supra, s. 16(4) is no more "particular or essential" to insanity legislation than s. 106.7(1) is to weapons legislation. Section 16(4) is a purely evidentiary section whose objective is to relieve the prosecution of the tremendous difficulty of proving an accused's sanity in order to secure a conviction.

The above characterization of the objective of s. 16(4) becomes clearer if one considers what would be the case if society's knowledge of the brain and of the nature of mental illness was suddenly and enormously expanded. What if our state of knowledge and technology was such that a doctor could perform a simple test to conclusively determine whether a person was insane within the meaning of s. 16 at the time of the offence? If this were the case, s. 16(4) would not be justified (and would not pass even the first branch of the Oakes test). All that would be necessary would be a provision requiring an accused alleging insanity to submit to this test. The Crown would have the full burden of disproving insanity beyond a reasonable doubt in any case where insanity was raised by an accused. Note, however, that the remaining provisions in s. 16 (or very similar provisions) would still be necessary and desireable. The objective of avoiding the stigmatization and punishment of an individual under the criminal law and of maintaining the distinction between "bad" and "sick" offenders would still exist and would continue to be met by s. 16(1). If the objective of s. 16(4) is as stated by the Attorneys General for New Brunswick and Ontario, the section would prevail whatever society's state of knowledge. The fact that it would disappear if it became reasonably possible for the Crown to meet a claim of insanity illustrates that the essential objective of s. 16(4) is to relieve the Crown of the nearly impossible task of disproving insanity whenever it is raised.

Many of the parties have identified the potential difficulties faced by the Crown in proving sanity. The Attorney General for New Brunswick states that without the cooperation of the accused, evidence of mental illness would be virtually impossible for the prosecution to obtain. If the burden were on the Crown, there would be no way to ensure such cooperation. The Attorney General of Manitoba has described, in some detail, the impracticality of requiring the Crown to prove sanity. The Code provides no method of forcing an accused to submit to psychiatric examinations unless fitness to stand trial is in issue. The Attorney General of Canada has described, in detail, the difficulties involved in obtaining a conclusive psychiatric opinion as to an accused's sanity, and characterizes the objective of s. 16(4) as "to allocate the burden of proof on the issue of insanity in a workable fashion". A further difficulty arises because the Crown would be required to prove sanity not as of the time of trial, but as of the time of the offence. Given that the Crown will often not know that insanity is going to be raised until some time after the offence takes place, the difficulties mentioned above are compounded.

Accordingly, the objective of s. 16(4) is to avoid placing an impossible burden of proof on the Crown and to thereby secure the conviction of the guilty. In my view, this objective is sufficiently important to warrant limiting constitutionally protected rights and s. 16(4) passes the first branch of the Oakes test.

Proportionality Test

1. Rational Connection

The question to be addressed at this stage of the Oakes analysis is whether there is a rational connection between the objective, which was identified above under the first branch of the test, and the means which have been chosen to attain this objective ‑- namely, the reverse onus provision embodied in s. 16(4).

In my view, many of the parties have misdirected their arguments on this portion of the Oakes test. The Crown has correctly formulated the rational connection test, but because it adopted the wrong objective for s. 16(4), it has, in my view, applied the test incorrectly to the facts of this case. The Crown correctly stated the rational connection test as follows:

Unlike Oakes and Whyte, supra, this is not a case of a presumption based on a proved fact. The presumption of sanity applies without proof of basic fact, and applies until the contrary is proved.

Given this distinction, the "rational connection" in the present case must be drawn directly between the objective and the measures employed.

If the Crown's formulation of the rational connection test is employed using the objective suggested above (i.e., to avoid placing a next to impossible burden on the Crown), the result is that s. 16(4) passes the rational connection test. Placing a burden on an accused who raises s. 16 to prove his or her insanity on a balance of probabilities certainly furthers the objective of not putting a burden on the Crown which is virtually impossible to meet. Thus, the presumption of sanity and the reverse onus embodied in s. 16(4) are rationally connected to the objective.

Therefore, it is necessary to turn to the next part of the proportionality branch of the Oakes test: does s. 16(4) impair s. 11(d) "as little as possible"?

2. As Little as Possible

The Attorneys General for New Brunswick, Ontario, and Quebec, argue that s. 16(4) impairs the presumption of innocence as little as possible because the persuasive burden is placed on an accused only after the Crown has proved, beyond a reasonable doubt, both actus reus and mens rea. While the Crown is discharging its initial burden, an accused has the full benefit of s. 11(d).

As stated above, it is not necessarily true that the accused's burden under s. 16 arises only after the Crown has proved both actus reus and mens rea beyond a reasonable doubt. While this will sometimes be the case, insanity will often be manifested as a denial of actus reus or mens rea. Accordingly, I am not convinced by this argument that s. 16(4) impairs the presumption of innocence as little as possible.

Numerous other arguments were made under this branch of the proportionality test (i.e., that the reverse onus will not arise very often, that the burden placed on the accused is "not a heavy onus", etc.). I am not persuaded by any of these arguments. In my view, the question to be addressed at this stage of the s. 1 inquiry is whether Parliament could reasonably have chosen an alternative means which would have achieved the identified objective as effectively.

Recent judgments of this Court (R. v. Edwards Books and Art Ltd., 1986 12 (SCC), [1986] 2 S.C.R. 713; Irwin Toy Ltd. v. Quebec (Attorney General), 1989 87 (SCC), [1989] 1 S.C.R. 927; and Reference re ss. 193 and 195.1(1)(c) of the Criminal Code (Man.), 1990 105 (SCC), [1990] 1 S.C.R. 1123) indicate that Parliament is not required to search out and to adopt the absolutely least intrusive means of attaining its objective. Furthermore, when assessing the alternative means which were available to Parliament, it is important to consider whether a less intrusive means would achieve the "same" objective or would achieve the same objective as effectively.

The appellants have argued that s. 16(4) does not impair the presumption of innocence as little as possible because the objective of the provision could be obtained in a manner which is less intrusive on Charter rights. For example, Parliament could have enacted a provision which merely requires an accused alleging insanity to meet an evidentiary burden (i.e., raise a reasonable doubt), at which point the burden would shift to the Crown to disprove insanity (or prove sanity) beyond a reasonable doubt. A number of parties have countered this argument with the contention that placing an evidentiary burden on the accused would not be sufficient to achieve the objective. In other words, s. 16(4) impairs s. 11(d) rights as little as possible because any less intrusive means would not achieve the "same" objective or would not achieve the same objective as effectively.

The Attorneys General of Quebec and Ontario both argue that an evidentiary burden would be ineffective because it is very easy for an accused to "fake" insanity and to raise a reasonable doubt. Thus, lowering the burden on the accused to a mere evidentiary burden would defeat the very purpose of the presumption of sanity. The appellants answer that only an accused who is willing to be incarcerated under the "L.G.W. system" will be inclined to raise insanity. Given the severe loss of liberty which corresponds to a successful insanity plea, it is unlikely that accused people will raise s. 16 on a routine basis.

It is true that s. 16 will be seldom raised, given the substantial constraint on liberty which follows a successful insanity plea. Nonetheless, I have concluded that the objective of the current provision is "pressing and substantial", given the next to impossible burden which would be placed on the Crown if s. 16(4) did not exist. If insanity were easier for an accused to establish, the defence would be successfully invoked more often (even if, statistically, it is still infrequently raised). Thus, putting a lesser burden on the accused would not have achieved the objective which is achieved by s. 16(4).

It is, of course, possible to imagine some combination of insanity provisions which might achieve the identified objective and which would violate s. 11(d) to a lesser extent then does the extant provision. For example, Parliament could have enacted a provision which provided that when an accused raises insanity, he is required to raise a reasonable doubt as to his sanity in order to shift the burden to the Crown to disprove insanity (prove sanity) on a balance of probabilities. The provision could also state that if an accused wishes to raise insanity he must submit to psychiatric examinations at the request of the Crown. These hypothetical provisions would also have to address the problem of obtaining evidence regarding insanity as of the time of the offence. Whether such a combination of insanity provisions would achieve the identified objective as effectively as does s. 16(4) is, of course, a matter of speculation. It is impossible to know what the effects of this alternative would be until it is put into practice.

Some of the parties have argued that any alternative to s. 16(4) which would meet the identified objective could very well bring about violations of other Charter rights. For example, the Charter might be invoked by an individual who was incarcerated under the "L.G.W. system" when only a reasonable doubt existed as to his or her insanity. Some of the parties have also suggested that putting a greater burden on the Crown could lead to violations of s. 8 during the process of gathering evidence. While these arguments are necessarily speculative, they do point out the difficulty faced by Parliament in choosing among the alternative means to meet its "pressing and substantial" objective. As was stated in Reference re ss. 193 and 195.1(1)(c) of the Criminal Code (Man.), supra, it is not the role of this Court to second‑guess the wisdom of policy choices made by Parliament. In enacting s. 16(4), Parliament may not have chosen the absolutely least intrusive means of meeting its objective, but it has chosen from a range of means which impair s. 11(d) as little as is reasonably possible. Within this range of means it is virtually impossible to know, let alone be sure, which means violate Charter rights the least.

In summary, I conclude that s. 16(4) violates s. 11(d) as little as possible in achieving its objective. It is therefore necessary to address the last part of the proportionality test: are the detrimental effects of s. 16(4) on the presumption of innocence proportional to the objective?

3. Proportionality Between Effects and Objective

The presumption of sanity and the reversal of onus embodied in s. 16(4) exist in order to avoid placing a virtually impossible burden on the Crown. The burden on the accused is not the full criminal burden; rather, the accused is required to prove his or her insanity on a balance of probabilities. If an accused were able to rebut the presumption merely by raising a reasonable doubt as to his or her insanity, the very purpose of the presumption of sanity would be defeated and the objective would not be achieved. Any other means of achieving the objective could also give rise to violations of other Charter rights.

Section 16(4) represents an accommodation of three important societal interests: avoiding a virtually impossible burden on the Crown; convicting the guilty; and acquitting those who truly lack the capacity for criminal intent. The result of this compromise is that some guilty people will be acquitted and will be dealt with via the "L.G.W. system", and some insane (and therefore not guilty) people will be convicted and will be stigmatized and punished as criminals. Of course, this would still be the case if the burden on the accused were lowered so as to require him or her to raise a reasonable doubt as to insanity; some insane accused may be unable even to raise a reasonable doubt as to insanity. This result is the inevitable consequence of the uncertainty of our scientific knowledge and of our commitment (as expressed in s. 11(d)) not to convict those who were insane at the time of the offence. The alternatives to this compromise raise their own Charter problems and give no guarantee as to whether they will achieve the objective. As I have mentioned above, the Charter does not require Parliament to "roll the dice" in its effort to achieve "pressing and substantial" objectives in order to adopt the absolutely least intrusive legislative provision.

While the effect of s. 16(4) on the presumption of innocence is clearly detrimental, given the importance of the objective that the Crown not be encumbered with an unworkable burden and given that I have concluded above that s. 16(4) limits s. 11(d) as little as is reasonably possible, it is my view that there is proportionality between the effects of the measure and the objective.

Accordingly, s. 16(4) is a reasonable limit on the presumption of innocence which can be upheld under s. 1 of the Charter.

Having dealt with the constitutional questions, I turn now to a consideration of the other issues raised by the appellants.

The Meaning of "Wrong" in s. 16(2)

The appellants submit that the trial judge erred in his direction to the jury in stating that the word "wrong" as used in s. 16(2) means "contrary to the laws of Canada" and that the Court of Appeal erred in holding that the trial judge had made no mistake in this regard. It is argued that the word "wrong" in s. 16(2) of the Code should be interpreted by this Court to mean "morally wrong" and not simply "legally wrong". Section 16(2) reads as follows:

16. ...

(2) For the purposes of this section, a person is insane when the person is in a state of natural imbecility or has disease of the mind to an extent that renders the person incapable of appreciating the nature and quality of an act or omission or of knowing that an act or omission is wrong.

At trial, the jury was asked to consider whether the appellants were not guilty by reason of insanity under both branches of the test in s. 16(2). The jury rejected the plea of insanity and the appellants were convicted of first degree murder. The question of the correct interpretation and application of the first branch of the test for insanity in s. 16(2), namely, whether the appellants were capable of appreciating the nature and quality of the act committed, is not before this Court. However, the correct interpretation of the second branch of the test, that is, whether the appellants were capable of knowing that their act was wrong, is directly in issue in this appeal.

In his directions to the jury, Ferg J. clearly stated the meaning that was to be given to the term "wrong" for the purposes of s. 16(2). He directed:

Next as an alternative even if the accused did appreciate the nature and quality of what he was doing and it's for you to decide, as I've said, is also insanity if the accused was laboring under a disease of the mind that rendered him incapable of knowing that his act was wrong. [sic] By wrong I mean that it was a criminal act or legally wrong.

Again, if you are satisfied on the balance of probabilities that either accused was incapable of knowing that his act was legally wrong because of a disease of the mind from which he was suffering, you will find him not guilty by reason of insanity.

After deliberating for some time, the jury asked the trial judge the following question:

Re: Knowledge of the laws of Canada:

(a) does this refer to simple knowledge of the rules as evidenced by recognition of the consequences, i.e.: police and jail term or;

(b) does it refer to their awareness of whether the laws apply to them at the time of the murder . . .

Ferg J. answered their question by repeating substantially what he had stated in his original direction. He summarized:

This part is the alternative, the second part, and this has to do with knowledge capable of knowing his act was wrong. As an alternative even if the accused did ‑‑ if you find that it's for you to decide ‑‑ did appreciate the nature and quality of what he was doing, and it's for you to decide as I just said, it is also insanity if the accused was laboring under a disease of the mind that rendered him incapable of knowing that his act was wrong and by the word wrong I mean that it was a criminal act or illegally wrong, if you will, the laws of Canada. [sic]

The Court of Appeal found no error on the part of the trial judge in directing the jury, stating that it was clear that the appellants "knew and fully appreciated the nature and consequences of their acts and they knew that what they were doing was legally wrong".

As noted above, the appellants submit that the term "wrong" for the purposes of s. 16(2) of the Code should be interpreted to mean "morally" wrong and not "legally" wrong. The respondent made no submission with respect to this issue in its factum, but argued orally that, regardless of the meaning that is to be given to the word "wrong" in this case, the difference between morally wrong and legally wrong where a very serious offence such as murder is involved "is so narrow as to be hardly worth the effort of deciding between them." It was further argued in oral pleadings that the meaning of "wrong" in s. 16(2), having its source in the seminal judgment of the House of Lords in M'Naghten's Case (1843), 10 Cl. & Fin. 200, 8 E.R. 718, incorporates not only a moral but also a legal dimension in such a way that the two cannot be divorced, with the result that a person who knows that an act is legally wrong must also comprehend that the act is morally wrong.

The meaning of the term "wrong" for the purposes of s. 16(2) was determined by this Court in Schwartz v. The Queen, 1976 165 (SCC), [1977] 1 S.C.R. 673. Speaking for the majority, Martland J. held that the capacity to know that an act is wrong in this context means no more than the capacity to know that what one is doing is against the law of the land.

The majority reviewed the principles enunciated in M'Naghten's Case and considered their subsequent application by the Court of Criminal Appeal in England and by the High Court of Australia. In M'Naghten's Case, Lord Tindal C.J. stated (at pp. 722‑23):

. . . to establish a defence on the ground of insanity, it must be clearly proved that, at the time of the committing of the act, the party accused was labouring under such a defect of reason, from disease of the mind, as not to know the nature and quality of the act he was doing; or, if he did know it, that he did not know he was doing what was wrong.

If the question were to be put as to the knowledge of the accused solely and exclusively with reference to the law of the land, it might tend to confound the jury, by inducing them to believe that an actual knowledge of the law of the land was essential in order to lead to a conviction; whereas the law is administered upon the principle that every one must be taken conclusively to know it, without proof that he does know it. If the accused was conscious that the act was one which he ought not to do, and if that act was at the same time contrary to the law of the land, he is punishable; and the usual course therefore has been to leave the question to the jury, whether the party accused had a sufficient degree of reason to know that he was doing an act that was wrong: and this course we think is correct, accompanied with such observations and explanations as the circumstances of each particular case may require. [Emphasis added.]

The majority continued by considering the application of the M'Naghten rules by the Court of Criminal Appeal in R. v. Codere (1916), 12 Cr. App. R. 21, and R. v. Windle, [1952] 2 Q.B. 826. In Codere, Lord Reading C.J. held (at pp. 27-28):

It is conceded now that the standard to be applied is whether according to the ordinary standard adopted by reasonable men the act was right or wrong. There may be cases where it is difficult to decide that question, but that is not the case here.

That is the accepted test, and applying it in this case there can be no room for doubt; once it is clear that the appellant knew that the act was wrong in law, then he was doing an act which he was conscious he ought not to do, and as it was against the law, it was punishable by law; assuming, therefore, that he knew the nature and quality of the act, he was guilty of murder, and was properly convicted.

The difficulty no doubt arises over the words "conscious that the act was one which he ought not to do," but, looking at all the answers in M'Naghten's case, it seems that if it is punishable by law it is an act which he ought not to do, and that is the meaning in which the phrase is used in that case.

In Windle, Lord Goddard C.J. stated clearly that the word "wrong" as used in the M'Naghten rules meant "legally wrong" (at p. 833):

Courts of law can only distinguish between that which is in accordance with law and that which is contrary to law.... The law cannot embark on the question, and it would be an unfortunate thing if it were left to juries to consider whether some particular act was morally right or wrong. The test must be whether it is contrary to law.

Lastly, the majority in Schwartz considered the judgment of the High Court of Australia in Stapleton v. The Queen (1952), 86 C.L.R. 358. In Stapleton, the High Court refused to apply Windle. Instead, citing directions that Dixon C.J. had given to a jury as a trial judge, the High Court held that the capacity of the accused to make moral judgments must be examined and that "wrong" was not to be defined narrowly as a legal wrong (at p. 367):

The question is whether he was able to appreciate the wrongness of the particular act he was doing at the particular time. Could this man be said to know in this sense whether his act was wrong if through a disease or defect or disorder of the mind he could not think rationally of the reasons which to ordinary people make that act right or wrong? If through the disordered condition of the mind he could not reason about the matter with a moderate degree of sense and composure it may be said that he could not know that what he was doing was wrong.

The majority in Schwartz rejected the application of the M'Naghten rules in Stapleton in favour of the decision of the Court of Criminal Appeal in Codere. In their view, the effect of s. 16(2) is to make the defence of insanity available to an accused who, because he suffered from a disease of the mind, did not know that he was committing a crime. The majority did not believe that the test adopted by the High Court of Australia in Stapleton differed from the "legally wrong" test since, "according to the ordinary principles of reasonable men, it is wrong to commit a crime" (p. 701). Even if there was a difference between the two tests, they could not accept that an insane person, committing an act that he knows to be a crime, could be acquitted if he believed that the act was acceptable according to the standards of reasonable men when a sane person believing such would not be acquitted. Furthermore, the majority rejected the test adopted in Stapleton on the ground that it entailed a subjective approach by examining the individual capacity of an accused to reason as to the wrongfulness of an act (at pp. 701‑2):

In my opinion the test provided in s. 16(2) is not as to whether the accused, by reason of mental disease, could or could not calmly consider whether or not the crime which he committed was morally wrong. He is not to be considered as insane within s. 16(2) if he knew what he was doing and also knew that he was committing a criminal act.

Dickson J., as he then was, dissented in Schwartz. He noted that the word "wrong" as used in s. 16(2) is ambiguous and is capable of meaning either "legally" or "morally" wrong. He also noted that the issue had given rise to conflicting lines of authority in England, Australia and Canada.

In order to resolve this question, Dickson J. first examined the internal structure of the Code in order to determine the meaning that Parliament intended to give the term. It would have been internally coherent, he submitted, for Parliament to use the word "unlawful" if it had intended "wrong" to mean "contrary to law". Furthermore, the use of the word "mauvais" in the French version of s. 16(2) suggests that Parliament intended the term to have a meaning broader than merely "unlawful". Lastly, s. 13 formerly provided that no child between seven and thirteen years of age could be convicted of a criminal offence "unless he was competent to know the nature and consequences of his conduct and to appreciate that it was wrong"; it would be insupportable, in this context, to equate knowledge that an act is wrong with the knowledge that the act is contrary to law.

Dickson J. then considered jurisprudential and doctrinal authorities antedating M'Naghten's Case and concluded that the historical common law test to determine the criminal responsibility of insane persons was whether the particular accused had the capacity to distinguish between conduct that was good or evil, right or wrong. M'Naghten's Case, in his view, did not depart from this standard. In fact, the case drew a clear line between knowledge that an act is illegal and knowledge that the act is one that a person ought not do; this distinction is revealed in the following passage in M'Naghten's Case (at p. 723):

If the accused was conscious that the act was one which he ought not to do, and if that act was at the same time contrary to the law of the land, he is punishable.

This passage indicates clearly that an accused will only be convicted if he commits an act which he knows he ought not do and which, at the same time, is contrary to law. As a result of his interpretation of M'Naghten's Case and the antecedent common law, Dickson J. concluded that Windle was decided wrongly and that the judgment in Codere was to be restricted to the particular facts of the case. The reasoning in Stapleton was preferred.

More fundamentally, Dickson J. concluded that a reading of s. 16(2) as a whole leads to the conclusion that "wrong" must mean contrary to the ordinary moral standards of reasonable men and women. The object of s. 16(2) is to protect individuals who do not have the capacity to judge whether an act is wrong; the inquiry as to the capacity of an accused to reason must not end simply because it is determined that the accused knew that the act was a crime. He argued that this would not serve to protect amoral persons since any incapacity must result from a disease of the mind (at p. 689):

Section 16(2) must be read in toto. One looks at capacity to reason and to reach rational decisions as to whether the act is morally wrong. If wrong simply means "illegal" this virtually forecloses any inquiry as to capacity. The question for the jury is whether mental illness so obstructed the thought processes of the accused as to make him incapable of knowing that his acts were morally wrong. The argument is sometimes advanced that a moral test favours the amoral offender and that the most favoured will be he who had rid himself of all moral compunction. This argument overlooks the factor of disease of the mind. If, as a result of disease of the mind, the offender has lost completely the ability to make moral distinctions and acts under an insane delusion, it can well be said that he should not be criminally accountable.

The interpretation of "wrong" as meaning "morally wrong" would not, in his opinion, have the effect of opening up the insanity defence to a far greater number of accused persons. First, what is illegal and what breaches society's moral standards do not often differ. Secondly, ""[m]oral wrong" is not to be judged by the personal standards of the offender but by his awareness that society regards the act as wrong" (p. 678). He concluded that an accused is not therefore free, as a result of such interpretation, to substitute at will his own sense of morality for that of society, but is to be acquitted by reason of insanity if, by reason of disease of the mind, he is incapable of knowing that society generally considers a particular act to be immoral.

With respect for contrary views, it is my opinion that Schwartz was wrongly decided by this Court and that the dissenting opinion of Dickson J. (concurred in by Laskin C.J., Spence and Beetz JJ.) is to be preferred. The majority judgment fails, in my respectful view, to appreciate the manner in which insanity renders our normal principles of criminal responsibility inapplicable to an individual as well as the particular objectives of s. 16 of the Code.

I do not dispute the principle that this Court should not easily overrule its prior judgments. In this regard, I refer to the words of Dickson C.J., in which I concurred, in R. v. Bernard, 1988 22 (SCC), [1988] 2 S.C.R. 833, at p. 849 (Dickson C.J. was in dissent, although the Justices who disagreed with his conclusions did not disagree with him on this point):

Let me say immediately that, even if a case were wrongly decided, certainty in the law remains an important consideration. There must be compelling circumstances to justify departure from a prior decision. On the other hand, it is clear that this Court may overrule its own decisions and indeed, it has exercised that discretion on a number of occasions.

In my opinion, it is appropriate in this case to overrule the majority decision in Schwartz with respect to the meaning of "wrong" in s. 16(2). Dickson C.J. described in Bernard four separate factors that would support a decision by this Court to overrule an earlier judgment. These factors were not held to be a comprehensive list nor was it claimed that they must all be present in a particular case to justify overruling a prior decision. They are instead guidelines to assist this Court in exercising its discretion. The fourth factor that was discussed by Dickson C.J., and that is directly relevant to the case at bar, is whether the prior judgment in the context of penal law is unfavourable to an accused in that it expands the scope of criminal responsibility beyond acceptable limits. Dickson C.J. stated (at pp. 860‑61):

Respect for the principle of certainty and the institutional limits imposed upon the law‑making function of the courts should constrain the Court from overruling a prior decision where the effect would be to expand criminal liability. It is not for the courts to create new offences, or to broaden the net of liability, particularly as changes in the law through judicial decision operate retrospectively. The same argument does not apply, however, where the result of overruling a prior decision is to establish a rule favourable to the accused.

In my view, Schwartz had the effect of expanding the scope of criminal responsibility unacceptably to include persons who, by reason of disease of the mind, were incapable of knowing that an act was wrong according to the normal and reasonable standards of society even though they were aware that the act was formally a crime. It is now necessary for this Court to reconsider its decision in Schwartz in order to redefine the scope of criminal liability in a manner that will bring it into accordance with the basic principles of our criminal law.

The rationale underlying the defence of insanity in Canada, as discussed above under the rubric "The Nature of the Insanity Provisions", rests on the belief that persons suffering from insanity should not be subject to standard criminal culpability with its resulting punishment and stigmatization. This belief, in turn, flows from the principle that individuals are held responsible for the commission of criminal offences because they possess the capacity to distinguish between what is right and what is wrong.

Section 16(2) of the Code embodies this conception of criminal responsibility by providing that no person shall be convicted of an offence who, at the time of committing the act in question, is in a state of "natural imbecility" or has disease of the mind to such a degree as to render him incapable of "knowing that an act or omission is wrong." The principal issue in this regard is the capacity of the accused person to know that a particular act or omission is wrong. As such, to ask simply what is the meaning of the word "wrong" for the purposes of s. 16(2) is to frame the question too narrowly. To paraphrase the words of the House of Lords in M'Naghten's Case, the courts must determine in any particular case whether an accused was rendered incapable, by the fact of his mental disorder, of knowing that the act committed was one that he ought not have done.

Viewed from this perspective, it is plain to me that the term "wrong" as used in s. 16(2) must mean more than simply "legally wrong". In considering the capacity of a person to know whether an act is one that he ought or ought not to do, the inquiry cannot terminate with the discovery that the accused knew that the act was contrary to the formal law. A person may well be aware that an act is contrary to law but, by reason of "natural imbecility" or disease of the mind, is at the same time incapable of knowing that the act is morally wrong in the circumstances according to the moral standards of society. This would be the case, for example, if the person suffered from a disease of the mind to such a degree as to know that it is legally wrong to kill but, as described by Dickson J. in Schwartz, kills "in the belief that it is in response to a divine order and therefore not morally wrong" (p. 678).

In applying s. 16(2) to a particular set of facts, it may be established that the accused who attempts to invoke the insanity defence is capable of knowing that he ought not do the act because he knows, first, that the act is contrary to the formal law or, secondly, that the act breaches the standard of moral conduct that society expects of its members. In this regard, subject to the qualification discussed below, I approve of the interpretation of s. 16(2) put forward by Professor Alan Mewett in "Section 16 and `Wrong'" (1976), 18 Crim. L.Q. 413, at pp. 415‑16:

The question that ought to be asked, it is submitted, is whether the accused, because of a disease of the mind (first hurdle) was rendered incapable (second hurdle) of knowing that this act was something that he ought not to do (third hurdle). If he was capable of knowing that the act was contrary to law and that he ought not to do an act contrary to law, then the defence should not apply. If he was incapable of knowing that it was contrary to law, but capable of knowing that it was an act condemned by people generally, then again the defence should not apply. But if he was incapable of knowing that the act was contrary to law and incapable of knowing that it was an act condemned by people generally, then the defence should apply. This only leaves a situation where he was capable of knowing that the act was contrary to law but incapable both of knowing that to act contrary to law was condemned by people generally and of knowing that this particular act was condemned by people generally. I would have thought that such an accused (who must be the rarest of all individuals) is precisely one who ought to be found not guilty by reason of insanity.

The qualification that I would make of Professor Mewett's comments is that the insanity defence should not be made unavailable simply on the basis that an accused knows that a particular act is contrary to law and that he knows, generally, that he should not commit an act that is a crime. It is possible that a person may be aware that it is ordinarily wrong to commit a crime but, by reason of a disease of the mind, believes that it would be "right" according to the ordinary morals of his society to commit the crime in a particular context. In this situation, the accused would be entitled to be acquitted by reason of insanity.

The test articulated in s. 16(2) is directed, as emphasized above, at an analysis of the capacity of the accused to reason and to understand the meaning of the terms "right" and "wrong", concepts that demand a moral judgment on the part of every individual in order to be applied in practice. It cannot be determined that an accused does not have the necessary capacity to engage in such moral reasoning simply because he or she does not have the simple ability to retain factual information, for example, the ability to know that a certain act is a crime in the formal sense. As Professor Colvin points out in Principles of Criminal Law, op. cit., at pp. 253 and 255:

The capacity to know the law in a formal sense involves typically no more than the capacity to absorb and retain normative information. . . . For the most part, formal knowledge of law does not call into play the capacity to analyze and assess an assortment of data which may present divergent and competing messages.

Full normative understanding surely encompasses comprehension of what a rule violation will mean to other people and how they will respond to it. If a psychotic person kills, knowing that her action is a crime, but deludedly believing that it will be lauded by everyone and that the law will not be enforced against her, then her capacity for normative understanding is so severely impaired that the attribution of criminal culpability is inappropriate. It could also be said that she is incapable of knowing that the act is a moral wrong. It could also be said that she is incapable of knowing that the act is contextually a legal wrong. Nothing turns on the choice between these modes of expression. The crucial point is that, although she is capable of knowing that her act is wrong in a formal sense, she is incapable of knowing that it is wrong in a contextual sense.

Professors Jacques Fortin and Louise Viau discuss a comparable principle in their Traité de droit pénal général (1982), at p. 184:

[TRANSLATION] Criminal responsibility is based historically and culturally both on the notion of moral blame and on the rule that ignorance of the law is no excuse.

It does not seem right to us to determine capacity using a rule which, by definition, applies only to those of sound mind. Capacity is a purely subjective and personal dimension of the individual, whereas the rule excluding ignorance of the law is purely utilitarian in nature and allows an accused to be convicted where he or she commits an offence in ignorance of the legal prohibition.

See also Don Stuart, Canadian Criminal Law (2nd ed. 1987), at pp. 340‑41, and Anne‑Marie Boisvert, "Psychanalyse d'une défense: réflexions sur l'aliénation mentale" (1990), 69 Can. Bar Rev. 46, at pp. 61‑62.

An interpretation of s. 16(2) that makes the defence available to an accused who knew that he or she was committing a crime but was unable to comprehend that the act was a moral wrong will not open the floodgates to amoral offenders or to offenders who relieve themselves of all moral considerations. First, the incapacity to make moral judgments must be causally linked to a disease of the mind; if the presence of a serious mental disorder is not established, criminal responsibility cannot be avoided. Secondly, as was pointed out by Dickson J. in Schwartz, supra, ""[m]oral wrong" is not to be judged by the personal standards of the offender but by his awareness that society regards the act as wrong" (p. 678). The accused will not benefit from substituting his own moral code for that of society. Instead, he will be protected by s. 16(2) if he is incapable of understanding that the act is wrong according to the ordinary moral standards of reasonable members of society.

In the case at bar, the trial judge directed the jury that the insanity defence was not available to the appellants pursuant to the second branch of the test set out in s. 16(2) if it reached the conclusion that the appellants knew, at the time of committing the offence, that the act was contrary to the laws of Canada. Of course, he cannot be faulted for having followed the decision of this Court in Schwartz. Nevertheless, for the reasons discussed above, our interpretation of s. 16(2) in Schwartz was not correct. As a result, I would order a new trial.

Section 16(3): Specific Delusions

The appellants argue that s. 16(3) of the Code provides an accused with a defence that is independent of s. 16(2) and that they should be entitled to an acquittal under s. 16(3) even if they fail to meet the conditions set out in s. 16(2). Section 16(3) provides as follows:

16. ...

(3) A person who has specific delusions, but is in other respects sane, shall not be acquitted on the ground of insanity unless the delusions caused that person to believe in the existence of a state of things that, if it existed, would have justified or excused the act or omission of that person.

Specifically, they submit that s. 16(3) should be interpreted in the following manner:

(1)Specific delusion should be interpreted in accordance with ordinary language and should mean a non‑ambiguous fixed unshakeable belief.

(2)The words "in other respects sane" should be interpreted to mean "does not fall within s. 16(2)".

(3)The final clause of s. 16(3) should be interpreted to give to the accused the right to raise the common law defences if the accused believed in a state of things that would at common law give rise to these defences.

The appellants submit that the trial judge should have clearly instructed the jury to consider and apply s. 16(3) in accordance with this interpretation. They argue that they suffered from specific delusions such that, even if they did not fall within s. 16(2), they should have been acquitted under s. 16(3) on the grounds that they believed in a state of things that, if true, would have justified or excused their actions. In their view, while the trial judge did not exclude the application of s. 16(3), he did not state adequately that, in accordance with the appellants' interpretation of the section, it could provide an alternative defence if the conditions of s. 16(2) were not satisfied.

The question of interpreting and applying s. 16(3) has given rise to no shortage of difficulties since its enactment in the Criminal Code of 1892. Courts have been required to consider whether s. 16(3) provides a separate and independent insanity defence for persons suffering from specific delusions, whether it simply qualifies the application of s. 16(2) by fixing a stricter test for such persons, or whether it merely confirms that such persons are eligible to claim the benefit of the defence under s. 16(2). Professor Dennis Klinck explores the various plausible interpretations of s. 16(3) in ""Specific Delusions" in the Insanity Defence" (1983), 25 Crim. L.Q. 458, at pp. 464‑67.

Many of the problems that are involved in the interpretation of s. 16(3) result from the uncertainty of the terms used. As a number of authors have duly noted, the preliminary obstacle that must be overcome is to define the nature of a specific delusion that does not render a person generally insane. This task is not a simple one, since it is not absolutely clear when a specific delusion becomes a general one or when a basically sane person becomes insane. In addition, it can be questioned whether the phrase "in other respects sane" refers to the test set out in s. 16(2) or to non‑legal notions of sanity. Lastly, it is not certain whether s. 16(3) refers to technical common‑law justifications and excuses for criminal law purposes or justifications and excuses in a more general, practical sense.

These issues have caused many writers to question the practical importance of s. 16(3). The Report of the Royal Commission on the Law of Insanity as a Defence in Criminal Cases (McRuer Report) (1956), cited by Professor Stuart in his textbook, Canadian Criminal Law, op. cit., at p. 342, recommended that the provision be repealed on account of redundancy:

The preponderance of medical evidence condemned the wording of this subsection on the ground that it describes a person who could not exist. The opinion of these witnesses was that no one who has "specific delusions" could be "in other respects sane".

The Law Reform Commission of Canada has likewise recommended the elimination of the rule in Working Paper 29 (affirmed by Report 31, Recodifying Criminal Law (1987)) at p. 48:

[Section 16(3)] has been much criticized and rarely applied. Medical opinion rejects the idea of partial insanity and legal scholarship stresses the injustice and illogicality of applying to the mentally abnormal a rule requiring normal reactions within their abnormality: a paranoiac killing his persecutor will be acquitted only if the imagined persecution would have justified the killing by way of self‑defence ‑‑ the law requires him to be sane in his insanity.

Mewett and Manning are also in accordance with this view (Criminal Law (2nd ed. 1985), at pp. 255‑56).

Section 16(3) has rarely been jurisprudentially considered. In R. v. Budic (No. 3) (1978), 1978 2404 (AB SCAD), 43 C.C.C. (2d) 419, the Alberta Court of Appeal held that s. 16(3) could not deprive an accused of the insanity defence if the conditions of s. 16(2) were met. In other words, an accused suffering from a specific delusion who meets the conditions of the defence set out in s. 16(2) cannot be deprived thereof by the apparently more restrictive condition stipulated by s. 16(3). Similarly, this Court confirmed in Abbey, supra, that s. 16(3) does not add to or qualify the general defence under s. 16(2). Dickson J. stated at p. 37:

I agree with the submission of counsel for the Crown that s. 16(3) of the Code can be of no assistance to Abbey as his delusion could not in any way "justify" or "excuse" his actions. In any event, any defence which could be raised under subsection (3) could also be raised under subsection (2). See the McRuer Report [citation omitted], p. 36. [Emphasis added.]

In my view, it is not necessary for this Court to engage in the difficult and perhaps impossible task of deciphering the plain meaning of s. 16(3) or of fathoming the intention of Parliament in enacting the provision. As a result of this Court's reconsideration of the meaning of the word "wrong" in s. 16(2), there can be no doubt that any successful attempt to invoke the insanity defence under s. 16(3) would also succeed under s. 16(2). Furthermore, if an accused fails to satisfy the conditions set out in s. 16(2), he or she will not be able to benefit from s. 16(3). It would not, therefore, assist an accused in any way if s. 16(3) was indeed held to constitute a separate and independent defence.

An accused will be able to invoke the insanity defence successfully under s. 16(2) if he can establish that he was incapable of knowing that the act in question was morally wrong in the particular circumstances. If it is instead proven that he was capable of knowing or did in fact know that the act was morally wrong in those circumstances, his claim to the defence will fail. An independent defence under s. 16(3) would only be relevant if the accused suffered from a specific delusion that, if true, would have justified the act, notwithstanding that the accused knew or was capable of knowing that the particular act was wrong in those circumstances.

In my opinion, such a case would be a logical absurdity. Consider, for example, the following situation. An individual suffers from the delusion that another person is attempting to kill her and that she will die unless she herself kills her apparent attacker. She therefore acts upon her delusion and kills that person. It cannot be denied that her conduct is excused or justified within the confines of the delusion, since the act of killing was necessary in order to preserve her own life. Accordingly, it cannot be said that she knew or was capable of knowing that the act was morally wrong in the circumstances. Although she may well have known that her act amounted to murder and that murder, as a general rule, is morally wrong, she was prevented by the specific delusion from being capable of knowing that the act of murder was wrong in the particular circumstances in which it was committed. In fact, the delusion caused her to believe that, while murder normally breaches the moral standard of society, it was morally right to commit murder in her case.

Conversely, this logic can be extended in the other direction. If the specific delusion had the effect that the individual did not know or was incapable of knowing that the offence was wrong in the circumstances, it must be admitted that the act would have been excused or justified had the delusion been true. If an act is not known to be morally wrong in a specific situation, an individual is justified in engaging in such conduct.

Similarly, Professor Klinck, op. cit., poses the following question in his analysis of s. 16(3), at p. 469:

On the other hand, to say that a person is capable of knowing that an act is wrong, but may still have a deluded belief in a state of things, which, if real, would excuse or justify his act is more problematical. To extend the example above: A might strike B, knowing what he is doing and knowing that assault is illegal, or that injuring people is morally wrong. Would subsec. (2) be unavailable to him if, knowing these things, he was deluded into believing that he was acting in self‑defence? Could it not be said that his delusion led him to perceive circumstances that would make his act neither morally nor legally wrong? Or would he have to resort to subsec. (3), which seems to allow a person to do a "wrong" act if his delusion affords him a justification or excuse?

There can be no doubt as to the answer to this question. An accused will be able to bring his claim within the scope of the second branch of the test set out in s. 16(2) if he proves that he was incapable of knowing that his conduct was morally wrong in the particular circumstances, for example, if he believed that the act was necessary to protect his life. If he is not able to establish this fact, it must be concluded that he either knew or was capable of knowing that the act was wrong in the circumstances. He cannot then possibly succeed in claiming that the act would have been justified or excused had the perceived facts been true.

I therefore find no flaw in the order of the trial judge except to the extent that he left open the possibility that the appellants may have succeeded under s. 16(3) if they did not qualify for the defence under s. 16(2). Again, he cannot be faulted given the high degree of uncertainty in the law as to the effect of s. 16(3).

The Rule Against Splitting the Case

The appellant Morrissette submits that the trial judge erred in permitting the Crown to split its case by presenting its evidence with respect to his sanity in rebuttal.

The appellant claims that the insanity defence became the dominant issue in the case from the outset, soon after the counsel for the appellants were retained. As a result, the Crown knew long before the trial that the insanity defence would be invoked and were thus able to consult several psychiatrists before the trial was held. The appellants argue that the Crown should have submitted this evidence as part of its case‑in‑chief. The appellant Morrissette states at paragraph 158 of his factum:

In the present case, the defence was well known and the Crown made a clear decision to try to split its case. It is submitted that there was a resulting injustice to the accused, who was effectively denied the opportunity of hearing the entire case against him before presenting his defence. [Emphasis added.]

The appellant Morrissette further claims to have taken this position at trial. Indeed, a motion was made unsuccessfully by the appellant for particulars of the Crown's evidence with respect to the issue of insanity. The Court of Appeal did not address this issue, although the Notice of Appeal by the appellant Chaulk to the Manitoba Court of Appeal, dated July 14, 1987, indicates that the issue was raised before the Court.

It is my opinion that the trial judge correctly permitted the Crown to adduce evidence with respect to sanity in rebuttal. There is no doubt that the Crown must tender as part of its case the evidence that it relies on to support a charge against an accused. It cannot split its case by presenting such evidence in rebuttal since the accused must be given a fair chance to meet the case against him. In R. v. Bruno (1975), 1975 1240 (ON CA), 27 C.C.C. (2d) 318 (Ont. C.A.), at p. 320, MacKinnon J.A. summarized the rule preventing the Crown from splitting its case as follows:

A statement made by Mr. Justice McPherson in R. v. Drake (1970), 1970 577 (SK QB), 1 C.C.C. (2d) 396 at p. 397 . . . is apposite:

There is a well‑known principle that evidence which is clearly relevant to the issues and within the possession of the Crown should be advanced by the Crown as part of its case, and such evidence cannot properly be admitted after the evidence for the defence by way of rebuttal. In other words, the law regards it as unfair for the Crown to lie in wait and to permit the accused to trap himself. The principle, however, does not apply to evidence which is only marginally, minimally or doubtfully relevant.

It is therefore apparent that the Crown must tender, as part of its case, evidence that will establish the existence of all elements of the offence with which the accused is charged.

The principle that the Crown is obliged to adduce, as part of its case, only evidence that is relevant to an element of the offence that the Crown must prove is affirmed by the corollary principle that the Crown need not adduce evidence in chief to challenge a defence that an accused might possibly raise. I approve of the analysis in this respect of Peter K. McWilliams, Q.C., in Canadian Criminal Evidence (3rd ed. 1990), at p. 31‑5:

In R. v. Campbell (1977), 1977 1191 (ON CA), 38 C.C.C. (2d) 6 (Ont C.A.), Martin, J.A. said at p. 26:

Rebuttal evidence by the prosecution is restricted to evidence to meet new facts introduced by the defence. The accused's mere denial of the prosecution's case in the witness‑box does not permit the prosecution in reply to reiterate its case, or to adduce additional evidence in support of it. In practice, however, it may often be difficult to distinguish between evidence, properly the subject of rebuttal, and evidence of facts relevant to prove guilt which should have been proved in the first instance by a full presentation of the prosecution's case: see Wigmore on Evidence, 3rd ed. (1940), vol. VI at pp. 510‑1.

Clearly relevant issues may arise during the case for the defence which could not have been reasonably anticipated by the prosecution before it completed its case. Indeed, that is the test for the admissibility of rebuttal evidence. The prosecution may have some anticipation that the defence will raise a defence be it innocent intent, accident, mistake, necessity or alibi or that the defence may be calling some expert witness as to the state of mind of the accused or as to the cause of the injury as in R. v. Campbell . . . and so on. The defence may give some hint of the line of its defence in the cross‑examination of Crown witnesses but in many cases it would be speculative and presumptuous for Crown counsel to anticipate: R. v. Perka et al., (1982), 1982 472 (BC CA), 69 C.C.C. (2d) 405 (B.C.C.A.). It would be, moreover, difficult and wasteful in time because Crown counsel could not be sure of just the exact issue or evidence it had to meet. None the less the prosecution must adduce all the evidence to prove the material ingredient of the offence.

In my view, the trial judge in this case correctly allowed the Crown to adduce rebuttal evidence with respect to the issue of insanity. First, the Crown cannot be expected to adduce evidence as part of its case‑in‑chief to meet a defence that the accused might possibly raise; this principle is not altered by the fact that the accused may warn the Crown that it intends to raise a particular defence. Secondly, a requirement that the Crown adduce evidence in chief to establish the sanity of the accused would defeat the presumption contained in s. 16(4). Section 16(4), as discussed above, posits that every person, for the purposes of the Criminal Code, is presumed to be sane until the contrary is proven. As a result, the sanity of a person accused of committing a criminal offence is not an issue until it is raised by that accused. The Crown is not obliged to prove the sanity of an accused as part of its case against such person. McWilliams makes the following point in this regard (at p. 31-8):

Section [16] of the Code imposes the persuasive burden of proof to prove the defence of insanity on the defence and therefore the prosecution may rebut evidence of insanity. Similarly, where the defence of insanity was raised the prosecution is permitted to call rebuttal evidence. R. v. Kasparek (1951), 1951 71 (ON CA), 101 C.C.C. 375 (Ont C.A.); see also R. v. Browne (1943), 29 Cr. App. R. 106.

Lastly, the arguments raised by the appellant Morrissette in his factum to the effect that the Crown should have been required to adduce the evidence in question as part of its case‑in‑chief are not persuasive. Since the appellants were given the opportunity for surrebuttal in this case, they are unable to make a serious argument that they suffered prejudice as a result of the fact that the Crown's evidence was adduced in rebuttal rather than in chief.

Disposition

I would answer the constitutional questions as follows:

1.Is s. 16(4) of the Criminal Code of Canada inconsistent with s. 11(d) of the Canadian Charter of Rights and Freedoms?

Answer:Yes.

2.If the answer to question 1 is yes, is s. 16(4) justified by s. 1 of the Canadian Charter of Rights and Freedoms and therefore not inconsistent with the Constitution Act, 1982?

Answer:Yes.

Given the interpretation that I have given to the word "wrong" in s. 16(2), I would allow the appeal and order a new trial.

//Wilson J.//

The following are the reasons delivered by

Wilson J. -‑ I have had the advantage of reading Chief Justice Lamer's reasons in this appeal. I agree with him that s. 16(4) of the Criminal Code, R.S.C., 1985, c. C-46, infringes s. 11(d) of the Canadian Charter of Rights and Freedoms and that the key issue on this appeal is whether s. 16(4) constitutes a reasonable and demonstrably justified limit on the presumption of innocence under s. 1 of the Charter. I take a different approach, however, from my colleague to the pressing and substantial concern that s. 16(4) is designed to address and to the standard of review that this Court should bring to bear on the "proportionality" aspect of the Oakes test.

Before turning to s. 1 I propose to address briefly the conflict which currently exists between the presumption of innocence guaranteed in s. 11(d) of the Charter and the presumption of sanity embodied in s. 16(4) of the Criminal Code. It may seem strange at first blush to speak of conflict between these two presumptions in light of the fact that they appear to have happily coexisted at common law for centuries. But in order to appreciate how the conflict has come about it is necessary to advert briefly to the history of each.

The Presumption of Sanity

Section 16(4) is the current version of a provision that has been in the Criminal Code of Canada since its first enactment in 1892. This Court discussed its proper interpretation in Clark v. The King (1921), 1921 603 (SCC), 61 S.C.R. 608. At that time the section (s. 19(3)) read:

Everyone shall be presumed to be sane at the time of doing or omitting to do any act until the contrary is proved.

As is noted in Clark the inspiration for the statutory presumption was M'Naghten's Case (1843), 10 Cl. & Fin. 200, 8 E.R. 718 (H.L.), which has provided the model for insanity defence legislation throughout the Commonwealth ever since. In that case, Lord Tindal C.J. said at p. 722:

[T]he jurors ought to be told in all cases that every man is to be presumed to be sane, and to possess a sufficient degree of reason to be responsible for his crimes, until the contrary be proved to their satisfaction; and that to establish a defence on the ground of insanity, it must be clearly proved that, at the time of the committing of the act, the party accused was labouring under such a defect of reason, from disease of the mind, as not to know the nature and quality of the act he was doing; or, if he did know it, that he did not know he was doing what was wrong. [Emphasis added.]

Although Lord Tindal C.J. did not elaborate on what standard of proof was involved in "to their satisfaction" other than to say that it must be "clearly proved", it was subsequently established that the standard was proof on a balance of probabilities (Clark v. The King, supra, per Duff J., at p. 621). Thus, the presumption of sanity in the criminal law has been part of our statutory and common law tradition for at least 150 years.

The Presumption of Innocence

The presumption of innocence has traditionally been viewed as the cornerstone of our criminal justice system. It is reflected in the axiom that "It is better that a guilty person go free than that an innocent person be convicted of a crime". In Woolmington v. Director of Public Prosecutions, [1935] A.C. 462 (H.L.), Viscount Sankey L.C. said at pp. 481-82:

If, at the end of and on the whole of the case, there is a reasonable doubt, created by the evidence given by either the prosecution or the prisoner, as to whether the prisoner killed the deceased with a malicious intention, the prosecution has not made out the case and the prisoner is entitled to an acquittal. No matter what the charge or where the trial, the principle that the prosecution must prove the guilt of the prisoner is part of the common law of England and no attempt to whittle it down can be entertained. [Emphasis added.]

The presumption of innocence has been carried forward into the Charter.

Section 11(d) of the Charter has frequently been the subject of judicial interpretation in the context of what have come to be known as reverse onus clauses. In R. v. Oakes, 1986 46 (SCC), [1986] 1 S.C.R. 103, Dickson C.J. explained the rationale of s. 11(d). He said at pp. 119-20:

The presumption of innocence is a hallowed principle lying at the very heart of criminal law. Although protected expressly in s. 11(d) of the Charter, the presumption of innocence is referable and integral to the general protection of life, liberty and security of the person contained in s. 7 of the Charter ... The presumption of innocence protects the fundamental liberty and human dignity of any and every person accused by the State of criminal conduct. An individual charged with a criminal offence faces grave social and personal consequences, including potential loss of physical liberty, subjection to social stigma and ostracism from the community, as well as other social, psychological and economic harms. In light of the gravity of these consequences, the presumption of innocence is crucial. It ensures that until the State proves an accused's guilt beyond all reasonable doubt, he or she is innocent. This is essential in a society committed to fairness and social justice. The presumption of innocence confirms our faith in humankind; it reflects our belief that individuals are decent and law-abiding members of the community until proven otherwise. [Emphasis added.]

In Oakes the constitutional validity of a statutory provision that placed the burden on an accused to prove that he was not in possession of narcotics for the purposes of trafficking was challenged. The majority concluded that this burden violated s. 11(d). Dickson C.J. stated at pp. 132-33:

[A] provision which requires an accused to disprove on a balance of probabilities the existence of a presumed fact, which is an important element of the offence in question, violates the presumption of innocence in s. 11(d). If an accused bears the burden of disproving on a balance of probabilities an essential element of an offence, it would be possible for a conviction to occur despite the existence of a reasonable doubt. This would arise if the accused adduced sufficient evidence to raise a reasonable doubt as to his or her innocence but did not convince the jury on a balance of probabilities that the presumed fact was untrue. [Emphasis added.]

In Oakes the scope of the presumption of innocence was confined to the "essential elements" of the offence. The prosecution had to bear the burden of proof of all the essential elements. A reversal of the onus of proof of any of these essential elements violated the presumption of innocence.

Since Oakes there has been a gradual broadening of the scope of s. 11(d). The decisions of this Court in R. v. Vaillancourt, 1987 2 (SCC), [1987] 2 S.C.R. 636, R. v. Holmes, 1988 84 (SCC), [1988] 1 S.C.R. 914, and R. v. Whyte, 1988 47 (SCC), [1988] 2 S.C.R. 3, have removed the qualification present in Oakes that limited the applicability of the presumption of innocence to the essential elements of the offence. In Whyte, Dickson C.J. (writing for the Court) unequivocally rejected any such qualification. He said at p. 18:

The short answer... is that the distinction between elements of the offence and other aspects of the charge is irrelevant to the s. 11(d) inquiry. The real concern is not whether the accused must disprove an element or prove an excuse, but that an accused may be convicted while a reasonable doubt exists. When that possibility exists, there is a breach of the presumption of innocence.

The exact characterization of a factor as an essential element, a collateral factor, an excuse, or a defence should not affect the analysis of the presumption of innocence. It is the final effect of a provision on the verdict that is decisive. If an accused is required to prove some fact on the balance of probabilities to avoid conviction, the provision violates the presumption of innocence because it permits a conviction in spite of a reasonable doubt in the mind of the trier of fact as to the guilt of the accused. [Emphasis added.]

The Court decided, in other words, that if we are to take the presumption of innocence seriously, it should not matter whether the reverse onus relates to an essential element of the offence as defined in the Criminal Code or to an independent defence, excuse or justification. To hold otherwise would be to make the ultimate determination of the guilt or innocence of an accused depend upon the contingencies of legislative drafting rather than upon the more basic philosophical underpinnings of the presumption of innocence. It is, as Dickson C.J. emphasized, "the final effect of a provision on the verdict that is decisive".

The respondent relies, however, upon the reasons of the majority of this Court in R. v. Schwartz, 1988 11 (SCC), [1988] 2 S.C.R. 443, to support its submission that s. 16(4) does not infringe the presumption of innocence because it creates an exemption rather than a defence. I believe that the respondent's reliance on Schwartz is misplaced. It seems to me that in Schwartz McIntyre J., writing for the majority, refused to apply to regulatory or licencing schemes the same strict standard in respect of the presumption of innocence as applies to criminal prohibitions even although the regulatory or licencing scheme arises in the criminal law context. Quoting from his reasons at p. 486:

The theory behind any licencing system is that when an issue arises as to the possession of the licence, it is the accused who is in the best position to resolve the issue. Otherwise, the issuance of the certificate or licence would serve no useful purpose.

McIntyre J. accordingly did not perceive the requirement that the accused establish that he had a licence as involving the imposition of a reverse onus in the Whyte sense. Schwartz, in my view, is distinguishable from the Vaillancourt, Holmes and Whyte line of cases on the ground that it deals with regulated not prohibited activity.

As already mentioned, Whyte stands for the proposition that it is the net effect of a reverse onus provision on the final guilt or innocence of the accused rather than the precise nature of the provision that must be examined under s. 11(d). It is this expansion of Oakes in Whyte that has given rise to the constitutional challenge to the presumption of sanity under s. 11(d) of the Charter. The presumption of sanity requires the accused to establish his or her insanity on a balance of probabilities and, however one conceives the plea of insanity, whether as an exemption, a defence, a justification or an excuse, the persuasive burden imposed on the accused by s. 16(4) permits him or her to be convicted of a crime despite the existence of a reasonable doubt as to his or her guilt. This offends the presumption of innocence on the authority of Whyte.

The constitutional validity of s. 16(4) depends therefore upon whether it can be saved under s. 1.

Section 1 of the Charter

The Oakes test requires first the identification of a pressing and substantial concern to which the impugned legislation is addressed. The government objective must be sufficiently important to warrant overriding a Charter right; otherwise the inquiry stops here and the legislation is unconstitutional. The first task is therefore to identify the government objective at which s. 16(4) of the Criminal Code is aimed and to assess whether it is of sufficient importance to justify an infringement on what is arguably one of the most, if not the most, fundamental tenet of our criminal justice system.

I start from the premise that the government must have been of the view that it was necessary to impose a persuasive burden on the accused to prove his insanity on a balance of probabilities in order to prevent perfectly sane persons who had committed crimes from escaping criminal liability on tenuous insanity pleas. In other words, the government must have concluded that the imposition of a purely evidentiary burden on the accused, i.e. the burden of adducing sufficient evidence to raise a reasonable doubt in the minds of the jury as to his sanity was not enough. Hence the presumption of sanity and the reverse onus on the accused to prove insanity, bringing s. 16(4) into conflict with s. 11(d) of the Charter as explained in Whyte.

If I am correct in my starting premise, then it would appear that under the first branch of Oakes the government would have to adduce evidence under s. 1 to show that this was a real social problem, that perfectly sane persons who had committed crimes were in significant numbers escaping criminal liability on tenuous insanity pleas and that something had to be done about it.

There is, however, a difficulty here because s. 16(4) merely reflects what, as we have seen, was already the common law and had been the common law for some time prior even to the enactment of the original section in Canada's first Criminal Code. There is therefore no historic experience in our jurisdiction with a purely evidentiary burden in order to show that such a burden was not adequate to achieve the government's objective.

My colleague sees the pressing and substantial concern somewhat differently. He says that the government objective was to relieve the prosecution of the "impossibly onerous burden" of proving the sanity of the accused beyond a reasonable doubt which it would have to do, he says, if the burden on the accused was merely an evidentiary one. In other words, my colleague narrows the characterization of the government objective to a purely procedural one arising in the context of the trial. He sees s. 16(4) as "a purely evidentiary section whose objective is to relieve the prosecution of the tremendous difficulty of proving an accused's sanity in order to secure a conviction," (p. 000). He does not identify any pressing and substantial concern to which this has in fact given rise. The provision is, in effect, a prophylactic measure designed to guard against a possible problem that might arise absent the reverse onus.

I am of the view that this represents a significant departure from the approach taken to s. 1 by this Court up until now. This Court has consistently evaluated laws subject to Charter challenge in terms of their justifiability as a response to existing social problems. The pressing social problem at which s. 16(4) is aimed must, it seems to me, be the one I have already identified, namely that guilty persons were escaping criminal liability on tenuous insanity pleas. Accordingly, if the accused wished to plead insanity it was essential that he be made to bear the persuasive burden. The prosecution must be given a "leg up" through the enactment of a presumption of sanity and the imposition on the accused of the burden of proving insanity on a balance of probabilities. Otherwise, as my colleague says, an accused whose sanity is in doubt might escape conviction.

I pause here to emphasize that, in my view, by analogy to the presumption of innocence, it is better that a guilty person be found not guilty by reason of insanity and committed for psychiatric treatment than an insane person be convicted of a crime.

Unlike my colleague, I do not believe that subs. (4) of s. 16 can be considered apart from subss. (2) and (3). The presumption in subs. (4) is accordingly operative to presume that the accused did not have disease of the mind (or natural imbecility) which made him incapable of appreciating the nature or quality of his act or of knowing that it was wrong at the time he did it. If the accused merely had the evidentiary burden and he or she succeeded in producing evidence sufficient to raise a doubt in the minds of the jury that one of these elements was present, then the prosecution clearly must dispel that doubt. Although I agree that the burden on the prosecution is not an easy one, I think it overstates the case to say that it is an "impossibly onerous" one. It is not, in my opinion, a matter of the prosecution's having to prove sanity in a vacuum; it is a matter of removing any doubt raised by the accused in the minds of the jury as to the presence of any of the elements spelled out in subss. (2) or (3). The extent of the burden on the prosecution will vary from case to case depending upon the evidence of insanity which the accused is able to produce, assuming that such evidence is sufficient to raise a doubt at all in the minds of the jury as to his sanity. The prosecution's task is simply to address any doubt raised by specific evidence adduced by the accused to support his or her insanity plea. The prosecution faces this kind of challenge all the time in cases where, for example, drunkenness is raised as a defence. It is my view that the burden is not an "impossibly onerous" one.

Be that as it may, I do not believe that Lamer C.J.'s characterization of the pressing and substantial concern succeeds in any event in overcoming the hurdle of the absence of any historic experience with a purely evidentiary burden.

The question posed by the Chief Justice's approach, it seems to me, is therefore whether s. 16(4) of the Code can be justified under s. 1 as a prophylactic measure designed to fend off a hypothetical social problem that might arise if accused persons pleading insanity had to meet only an evidentiary burden. This prompts me to ask: do we wish to go down this path and justify infringements of guaranteed Charter rights on a purely hypothetical basis? And, in particular, do we wish to go down this path where such a fundamental tenet of our justice system as the presumption of innocence is at stake? I have serious reservations about adopting such a course even in cases where it could be said that the hypothesis was a strong one which I do not think it is in this instance for reasons which I will discuss later. Before adopting such a course I would prefer to look elsewhere to see what the experience has been in other free and democratic societies which have had the experience of living with the purely evidentiary burden.

1. The United States Experience

Given the lack of available data in Canada and the Commonwealth, it makes good sense in applying s. 1 of the Charter to turn to other societies where the accused who pleads insanity faces the lower evidentiary burden. Indeed, this Court endorsed that approach in Law Society of Upper Canada v. Skapinker, 1984 3 (SCC), [1984] 1 S.C.R. 357, per Estey J., at pp. 367-68. The United States presents an excellent source for this type of information since American criminal law is a matter of state jurisdiction and the states are fairly evenly split on the issue of the appropriate standard of proof that an accused pleading insanity should have to meet.

(a) United States Jurisprudence

The starting point for an analysis of the American law concerning the appropriate standard of proof on the accused is the landmark ruling of the U.S. Supreme Court in Davis v. United States, 160 U.S. 469 (1895). The Assistant Attorney General in Davis argued that insanity is a fairly easy condition to feign or simulate and that there are strong policy reasons for imposing a persuasive burden. He identified the policy reasons as follows (at p. 473):

The probability of a jury finding an insane man guilty, under the rule that insanity must be established to their reasonable satisfaction, is very slight as compared with the evil that results to society from the application of the doctrine that a reasonable doubt as to whether the defendant is sane or insane must be followed by acquittal.

The "evil" to which he referred was that "[m]erely doubtful evidence of insanity would fill the land with acquitted criminals" (Ortwein v. Commonwealth, 76 Pa. 414 (1874), at p. 425). After an extensive review of the authorities, the Supreme Court squarely rejected this line of argument in a unanimous judgment (per Harlan J., at pp. 492-93 and 488):

It seems to us that undue stress is placed in some of the cases upon the fact that, in prosecutions for murder the defence of insanity is frequently resorted to and is sustained by the evidence of ingenious experts whose theories are difficult to be met and overcome. Thus, it is said, crimes of the most atrocious character often go unpunished, and the public safety is thereby endangered. But the possibility of such results must always attend any system devised to ascertain and punish crime, and ought not to induce the courts to depart from principles fundamental in criminal law, and the recognition and enforcement of which are demanded by every consideration of humanity and justice....

How then upon principle or consistently with humanity can a verdict of guilty be properly returned, if the jury entertain a reasonable doubt as to the existence of a fact which is essential to guilt, namely, the capacity in law of the accused to commit that crime? [Emphasis added.]

Harlan J. was of the view that the crime of murder "necessarily involves" the possession of a mental capacity by the accused that is sufficient to attribute criminal responsibility. He stressed that the risk of acquitting guilty people accompanies any criminal justice system that requires guilt to be proven beyond a reasonable doubt and that the insanity defence was not exceptional in this regard. As such, the policy considerations did not warrant a departure from the fundamental principle that the prosecution should bear the burden of establishing guilt beyond a reasonable doubt. This reasoning has been consistently approved in subsequent Supreme Court cases: see In Re Winship, 397 U.S. 358 (1970); Mullaney v. Wilbur, 421 U.S. 684 (1975); Jackson v. Virginia, 443 U.S. 307 (1979). Thus, the Davis case set down the rule which has governed federal criminal procedure in the United States for the past 95 years, namely that when there is evidence before the trier of fact tending to show that the accused was criminally insane at the relevant time, the burden shifts to the prosecution to show beyond a reasonable doubt that he in fact had the requisite mental capacity to be held criminally responsible.

In my view, the factors which were taken into account by the United States Supreme Court in Davis are precisely those which Lamer C.J. invites us to consider in the present appeal. Harlan J. had to decide whether placing the lower evidentiary burden on the accused would result in an impossibly onerous burden on the prosecution and therefore in unwarranted acquittals. He found that it would not. The fact that this decision has stood unchallenged for almost a century tends to persuade me that the lower burden on the accused has not created a pressing and substantial problem in American society.

It is noted, however, that the ratio decidendi in Davis turned on the supervisory powers of the Supreme Court and not, strictly speaking, on constitutional grounds. Thus, in Leland v. Oregon, 343 U.S. 790 (1952), it was explicitly held to be permissible for a state to shift the persuasive burden for insanity claims onto the shoulders of the accused. Accordingly, a number of states have imposed a persuasive burden on the accused for offences within their criminal jurisdiction. Leland was upheld in Rivera v. Delaware, 351 A.2d 561 (Del. 1976), appeal dismissed (Brennan and Marshall JJ. dissenting), 429 U.S. 877 (1976). The criminal trial process for those states that impose the persuasive burden is much the same as ours in Canada. It is incumbent on the accused to establish his or her insanity on a balance of probabilities.

In jurisdictions that impose only an evidentiary burden on the accused the procedure, including the burden on the prosecution, is explained by the First Circuit Court of Appeal in United States v. Pasarell, 727 F.2d 13 (1984), certiorari denied, 105 S. Ct. 107 (1984), at p. 14:

When a defendant introduces evidence of insanity, the prosecution has the burden of proving him sane. In United States v. Dube. . . , this court described the prosecution's burden in these terms:

"Insanity is a jury question unless a reasonable man viewing the facts and reasonable inferences therefrom in the light most favorable to the prosecution must necessarily possess a reasonable doubt as to the defendant's sanity. The nature and quantum of rebuttal evidence sufficient to present a jury question is to some extent determined by the strength of the case for insanity. There is no general principle that the prosecution must counter defendant's expert medical evidence with expert testimony of its own. The expert testimony is not conclusive even where uncontradicted; its weight and credibility are for the jury to determine..." [Emphasis added; citations omitted.]

In these jurisdictions the courts are very reluctant to interfere with the jury's finding of fact as to the mental state of the accused: see United States v. Voice, 627 F.2d 138 (8th Cir. 1980); United States v. Samuels, 801 F.2d 1052 (8th Cir. 1986); People v. Stockwell, 242 N.W.2d 559 (Mich. Ct. App. 1976). It is also important to note that the jury is not bound by expert witnesses:

The jury is the ultimate judge of defendant's sanity at the time of the crime, and in this case, since it had before it evidence of defendant's behaviour and state of mind upon the basis of which it could have found defendant sane at that time, it was not bound by the expert opinion testimony of the doctor.

(People v. Krugman, 141 N.W.2d 33 (Mich. 1966), at p. 35.)

Nor, as the Pasarell case makes clear, is there an obligation on the prosecution to "match" expert testimony. This would seem to undercut the argument often advanced by proponents of the persuasive burden that a merely evidentiary burden on the accused tends to produce a "battle of the experts" that leaves the jurors confused and often results in unjustified acquittals. It appears that in these jurisdictions leaving the determination of sanity, like the determination of other crucial questions of fact, to the good common sense and considered judgment of the jury has not presented any particular problem.

The American cases do not, in my view, support the conclusion that the higher burden on the accused is indispensable for the proper administration of criminal justice nor do they confirm the existence of the type of pressing and substantial concern that Lamer C.J. believes would arise if the persuasive burden on the accused were to be removed.

(b) Commentary on the American Experience

Canadian commentators have looked to the American experience for guidance in this matter. In "A Critique of Proposals to Reform the Insanity Defence" (1989), 14 Queen's L.J. 135, Professor G. Ferguson evaluates what he calls the "pragmatic considerations" in favour of keeping the persuasive burden on the accused. They are (at p. 148):

(i)to reduce the likelihood of successful fabrication of the insanity defence;

(ii)a reasonable doubt about the accused's sanity can be created too easily; [and]

(iii)proving the accused's sanity beyond a reasonable doubt is impossible.

He states at p. 148:

Upon closer examination, these pragmatic considerations lack validity. The experience in the United States is particularly revealing. As of 1982, in half of the States and in all federal courts, once there is some evidence of insanity, the prosecution has the burden of proving the accused's sanity beyond a reasonable doubt. Does that burden allow a throng of fabricated insanity pleas to succeed? Does it put an intolerable or impossible burden on the Crown? I sampled the reported cases in those jurisdictions for the year 1982. In almost all of the cases there was at least some expert evidence supporting the accused's insanity plea. But in twenty-eight of the thirty cases, the defence of insanity failed. The Crown proved its case; the accused failed to raise a reasonable doubt. If anything, these figures suggest that even raising a reasonable doubt about insanity may be too difficult a standard to meet rather than one which is too facile. (Incidentally, in jurisdictions where the accused had the burden of proof on a balance of probabilities, the accused's insanity plea failed sixteen times in seventeen cases.) [Emphasis added.]

It would appear that in the United States the evidentiary burden has not resulted in a flood of accused persons being found not guilty by reason of insanity. Nor have the Americans witnessed a great rush of insanity pleas. As the American Psychiatric Association notes in its Statement on the Insanity Defence issued in 1982:

(S)uccessful invocation of the [insanity] defence is rare (probably involving a fraction of one percent of all felony cases). While philosophically important for the criminal law, the insanity defense is empirically unimportant.

(See R. J. Simon and D. E. Aaronson, The Insanity Defense: A Critical Assessment of Law and Policy in the Post‑Hinckley Era (1988), at p 8.)

Statistics gathered from follow-up studies on NGRI acquittees (Not Guilty by Reason of Insanity) in both countries tend to indicate that not many mistakes are being made. In Canadian statistical materials provided to this Court by the parties (S. Hodgins et al., Annual Report, Year 1 Canadian Data Base: Patients Held on Lieutenant Governor's Warrants (1989), Table 5, "A Description of Patients Under Lieutenant Governors' Warrant March 1, 1988"), the most common reported mental disorders among detainees were: schizophrenia (65.3%), personality disorders (11.9%), affective disorders (7.6%), delusional paranoid disorders (6.9%), and mental retardation (4.2%). On the other hand, only 0.2% of the patients were found to have "No Diagnosis". This figure means that only 2 of the 1007 detainees under the L.G.W. system in 1988 were found to have no mental disorder.

In the United States, a somewhat less recent study of NGRI Admissions to Saint Elizabeth's Hospital in Washington D.C. also tends to support the proposition that most acquittees are in fact in need of psychiatric treatment. Arthur Matthews Jr. consolidated the psychiatric diagnoses of NGRI acquittees over a 12‑year period (1954-1965). He found schizophrenia to be the most prevalent mental disorder (39.1%). His research indicates that only 5.5% of NGRIs were found to be "Without Mental Disorder": see Mental Disability and the Criminal Law: a Field Study (1970), at p. 35. Although significantly higher than the Canadian statistics, these figures, in my view, do not indicate that successfully fabricated insanity pleas have given rise to a pressing and substantial problem in the United States.

(c) Recommendations Respecting the Burden of Proof in American States

The American Bar Association's Standing Committee on Association Standards for Criminal Justice has done extensive research on the problems associated with mental disability and the criminal justice system. The First Tentative Draft: Criminal Justice Mental Health Standards (1983), reviews the state of the law and sets out recommendations for allocating the burden of proof in insanity cases (at pp. 291-93):

Standard 7-6.9. Burden of production and burden of persuasion

(a)The defendant should have the burden of assuring that evidence of insanity is introduced.

(b) Once evidence of insanity has been introduced at trial, the burden of persuasion should be allocated as follows:

(i) in jurisdictions utilizing any test for insanity which focuses solely on whether the defendant, as a result of mental disease or was unable to know, understand or appreciate the wrongfulness of his or her conduct at the time of the offense charged, the prosecution should have the burden of disproving the defendant's claim of insanity beyond a reasonable doubt;

(ii) in jurisdictions utilizing the ALI-Model Penal Code test for insanity, the defendant should have the burden of proving the claim of insanity by a preponderance of the evidence.

(c) Nothing contained in paragraph (b) above relieves the prosecution of its burden of proving beyond a reasonable doubt all elements of the offense charged including the mental state required for the offense charged.

Commentary

... The difficult question concerns the allocation of the burden of persuasion once the issue has been raised....

At the present time, the states are closely divided on this question: twenty-six states (and the District of Columbia in nonfederal cases) place the burden of proving insanity on the defendant, (usually) by a preponderance of the evidence. In contrast, twenty-four states require the prosecution to prove the defendant's sanity beyond a reasonable doubt. In the federal courts, the burden is borne by the prosecution....

... Paragraph (b) provides that the burden should fall upon the prosecution to disprove a defendant's claim of insanity in those states using a test which focuses solely on whether the defendant, as a result of mental disease or defect, was unable to know, understand or appreciate the wrongfulness of his conduct. This embrace of the Davis doctrine recognizes that a narrowing of the insanity test requires that the "risk of error" remain with the prosecution. In contrast, the second part provides that jurisdictions employing the combined cognitive-volitional test should shift the burden of proof to the defendant, recognizing the right of jurisdictions to place the burden on the defendant when that defendant is the recipient of the possible benefits flowing from an expanded insanity test. The shift is thus a quid pro quo for the greater latitude implicit within the volitional part of the test.

The allocation formula is based on the view that mistakes in the administration of the insanity defense occur primarily when the volitional prong of the current ALI-Model Penal Code formulation is the test employed.... Conversely, because the risk of mistake is greatly reduced under a cognitive test, the reduction of risk in such cases supports the view that there is no justification for shifting the burden of proof to the defendant. [Emphasis added.]

I note that formulation of the insanity defence in s. 16(2) of the Code employs a cognitive test. As such, the ABA recommendations would indicate that ceteris paribus the lower evidentiary burden is appropriate in the Canadian context.

2. Recommendations in Canada

Canadian commentators have reached similar conclusions concerning the appropriate burden of proof for the insanity defence. In The Presumption of Innocence in the Draft Code of Substantive Criminal Law (1986), an unpublished paper prepared for the Law Reform Commission of Canada by Professor Patrick Healy, the author examines the insanity issue within the context of reverse onus clauses generally. He says (at paras. 29 and 32):

29. ... the two principal rationales for the reverse onus are animated by notions of efficiency and convenience, and thus suggest that reverse onuses are expedient devices.... [T]herefore, it must be asked whether considerations of expediency carry enough force to overtake the general rule regarding the burden of proof....

30. It is submitted that neither the difficulty of proving a negative or the difficulty of acquiring affirmative evidence of facts on which the accused might be the best witness is possessed of such overriding force that they justify a rule that would require the accused to prove his innocence....

He further notes (at paras. 35 and 36) that:

35. The reverse onus with respect to insanity is an anachronistic vestige of the era in which the accused bore the burden to disprove guilt once the prosecution had established a prima facie case with regard to actus reus of the offence. The allocation of the onus probandi in this instance, however, was fortified by the so-called presumption of sanity and the presumption that the accused intended the natural and probable consequences of his act. These two presumptions effectively created a presumption of mens rea and their combined force ensured the survival of this particular reverse onus at common law long after the end of the nineteenth century, by which time the courts had largely reversed the older doctrine that the prosecution need only establish the actus reus and that, to avoid conviction, the presumption of mens rea required the accused to prove his innocence by proving a viable justification, excuse or defence.

36. The general presumption of mens rea, at least as a mandatory presumption arising upon the prosecution's prima facie case, is now a thing of the past, and although the presumption of sanity has survived its sole utility is to state that the sanity of the accused is not an implied or necessary element in the prosecution's case. In more recent times the rationales advanced most frequently for the reverse onus in insanity are not materially different from the rationales given for other reverse onuses: the mental state of the accused is exceptionally difficult to ascertain and only the accused is best equipped to provide relevant evidence on the issue. If the most compelling justification for allocating the onus probandi to the accused on this issue is to ensure that it does not avail to his benefit without a secure evidentiary foundation, then neither a presumption of sanity nor the allocation is necessary or desirable in a new code. Accordingly, the draft ought to stipulate in specific terms the essential elements of each offence and sanity (or insanity) would therefore not be an issue unless specifically raised by either party. The salient interests in policy are satisfied not by requiring proof once the issue is raised, but by ensuring that the issue need only be negatived when the defence makes a sufficient case for doing so. Such a burden is not lightly undertaken nor easily discharged and there is no compelling reason of principle, or even of utility, to distinguish it from provocation, duress, and self-defence. [Emphasis added.]

Professor Colvin is also of the view that the lower burden deserves serious consideration. He states at p. 241 of Principles of Criminal Law (1986):

The placement of the persuasive burden should now be vulnerable to a constitutional challenge under s. 11(d) of the Canadian Charter of Rights and Freedoms.... [I]nsanity is the only defence of deficient mental capacity for which the onus of proof is now reversed. If the efficiency of the criminal justice system can withstand the prosecution carrying the persuasive burden for the other defences, why should it be supposed that a reverse onus is of crucial importance for insanity?

I would note that the Law Reform Commission embraced this lower burden of proof in Report 31, Recodifying Criminal Law (1987). The Commission states at p. 4:

In the Commission's view the prosecution should prove all necessary conditions of liability by evidence which is admissible and which in the judgement of the trier of fact proves them beyond reasonable doubt, while the accused should not have to prove any fact in issue in a criminal trial. This burden of proof obliges the prosecution, therefore, not only to prove all inculpatory elements of a crime but also to disprove any matter of exculpation for which the evidence adduced (irrespective of its source) discloses a foundation. By contrast, an accused, when seeking to plead a matter of exculpation not already raised by the evidence, need only produce sufficient evidence to disclose such a foundation. Meanwhile the prosecution need not disprove any matter of defence, justification or excuse before such foundation is disclosed.

In line with this approach, the Commission has refrained ... from casting a legal burden on an accused regarding any matter of exculpation. On the one hand such reverse onuses may well be contrary to the presumption of innocence and paragraph 11(d) of the Charter. On the other hand they are also unnecessary, because sufficient protection against unwarranted acquittals lies in the requirement that any matter of exculpation be based on a demonstrable evidentiary foundation. [Emphasis added.]

3. Recommendations in Great Britain

It is of note that the English Criminal Law Revision Committee has come to the same conclusion regarding the appropriate burden of proof. In the Eleventh Report of the English Criminal Law Revision Committee, Evidence (General) (1972), the eminent jurists state at pp. 88-90:

140. We are strongly of the opinion that, both on principle and for the sake of clarity and convenience in practice, burdens on the defence should be evidential only.... Our reasons are given below.

(i)In the typical case where the essence of the offence is that the offender has acted with blameworthy intent, and the defence which the accused has the burden of proving implies that he had no such intent but acted wholly innocently, it seems to us repugnant to principle that the jury or magistrates' court should be under a legal duty, if they are left in doubt whether or not the accused had the guilty intent, to convict him. For this is what the law requires. ... It may well be that this kind of situation does not occur at all commonly but that in practice the prosecution assume [sic] the burden of proving the whole case; for in our 6th report we mentioned (in relation to a suggestion to create an offence with a persuasive burden on the accused) that "the experience of members of the committee in the case of similar provisions as regards existing offences suggests that in practice provisions of this kind are less efficacious than may have been hoped when they were enacted, because juries are sometimes unwilling to convict unless the prosecution leads evidence sufficient to rule out the defence". But even if so, it seems to us undesirable that the possibility which we mentioned should continue to exist even in theory. If provisions of this kind have indeed little or no effect, the case for altering them seems, if anything, stronger.

(v)The change would get rid of the present need for the judge to give the jury the complicated direction on the difference between the burden on the prosecution of proving a matter beyond reasonable doubt and that on the defence of proving a matter on a balance of probabilities. Many judges have said that they find it difficult or impossible to direct juries on this in a way which the jury are likely to find satisfactory or even intelligible.

We propose also that the burden of proving insanity as a defence should be made an evidential one. The present rule that it is a persuasive burden is, in our opinion, anomalous and open to most of the objections stated in this paragraph to the statutory persuasive burdens.... The great majority of those whom we consulted in 1968 were in favour of the proposed change. A few, however, opposed it, chiefly on the ground that it would be an unwarranted departure from the policy of Parliament in a number of enactments. But we do not regard this as sufficient to outweigh the arguments mentioned above. [Emphasis added.]

They also stress that even the lower burden on the accused is not an inconsequential or trivial one (at pp. 87-88):

138. ...

The words "a live issue fit and proper to be left to the jury" show that it is not enough for the defence merely to allege the fact in question: the court decides whether there is a real issue on the matter. Therefore, when the burden is an evidential one, there is no need for the judge to mention the burden to the jury at all. If he decides that there is insufficient evidence to raise an issue on the matter, he directs the jury that they must take the matter as proved against the defence. If he decides that there is sufficient evidence to raise an issue on the matter, he simply directs the jury in the ordinary way that it is for the prosecution to satisfy them on the matter beyond reasonable doubt. [Emphasis added.]

In short, the lower burden on the accused who pleads insanity is seen as being both more consonant with fundamental principles of criminal law and as a sufficiently high threshold to prevent insanity pleas in cases where there is only tenuous support for such a plea.

The respondent, for the reasons discussed above, has not been able to establish that s. 16(4) of the Criminal Code was aimed at an existing pressing and substantial concern and, while it may be that the legislature need not necessarily wait until such a concern has arisen, I do not believe that the respondent has succeeded in establishing even a likelihood of its arising. I would conclude therefore that the first requirement of the Oakes test has not been met. However, in case I am wrong in this I propose to go on to consider whether s. 16(4) meets the proportionality aspect of Oakes.

4. The Proportionality Test

There are three parts to the proportionality test in Oakes. I do not take issue with Lamer C.J.'s conclusion on the first of these that s. 16(4) is rationally connected to the legislative objective being sought, whether that objective is perceived in his narrow terms or in my broader terms. The second part is commonly referred to as the "minimal impairment" test and poses more difficulty.

The issue under this part of the Oakes test is whether some other legislative provision could achieve the desired objective while impairing the Charter right "as little as possible". Lamer C.J. is of the view that Parliament is not required to seek and adopt "the absolutely least intrusive means of attaining its objective" (p. 000). He indicates that he is unwilling to embark on a course of "second-guessing" the wisdom of Parliament's choice of legislative means and cites some recent decisions of this Court as authority for this deferential attitude. In my view, this is not a case for deference. In one of the cases on which the Chief Justice relies, Irwin Toy Ltd. v. Quebec (Attorney General), 1989 87 (SCC), [1989] 1 S.C.R. 927, this Court indicated that there might be exceptions to the stringent review called for under this part of the Oakes test. Whether or not such an exception was warranted would depend upon the role Parliament was fulfilling in enacting the impugned legislation.

As I understand this aspect of Irwin Toy, an exception may be made where the legislature mediating between the competing claims of groups of citizens or allocating scarce resources among them is forced to strike a compromise on the basis of conflicting evidence. In such cases there will be a substantial policy component to the choice of means selected by the legislature and that choice should be respected even if it cannot be said to represent the "least intrusive means". In my view, Irwin Toy does not stand for the proposition that in balancing the objective of government against the guaranteed right of the citizen under s. 1 different levels of scrutiny may be applied depending upon the nature of the right. The prerequisite for the exception to the minimal impairment test in Oakes, as I understand Irwin Toy, is that the guaranteed right of different groups of citizens cannot be fully respected; to respect to the full the right of one group will necessarily involve an infringement upon the right of the other. In such a circumstance Irwin Toy holds that it is appropriate for the government to fashion a compromise on the basis of policy considerations.

In Irwin Toy the Court went on to contrast the "competing claims" situation with the situation in which no exception to the strict test in Oakes should be permitted. Lamer J. (as he then was) concurred in the following passage appearing at p. 994 of the majority opinion. It seems apposite to the case at bar:

In other cases, however, rather than mediating between different groups, the government is best characterized as the singular antagonist of the individual whose right has been infringed. For example, in justifying an infringement of legal rights enshrined in ss. 7 to 14 of the Charter, the state, on behalf of the whole community, typically will assert its responsibility for prosecuting crime whereas the individual will assert the paramountcy of principles of fundamental justice. There might not be any further competing claims among different groups. In such circumstances, and indeed whenever the government's purpose relates to maintaining the authority and impartiality of the judicial system, the courts can assess with some certainty whether the "least drastic means" for achieving the purpose have been chosen, especially given their accumulated experience in dealing with such questions... [Emphasis added.]

I re-emphasized this distinction in United States of America v. Cotroni, 1989 106 (SCC), [1989] 1 S.C.R. 1469, at pp. 1515-16:

In coming to this conclusion I am not unmindful of the comments of La Forest J. that a flexible approach should be taken in some cases to the proportionality test in R. v. Oakes. However, this does not seem to me to be one of those cases. It is one thing to temper scrutiny of legislation and relax the general approach to "fine tuning" when dealing with alternate forms of business regulation (see R. v. Edwards Books and Art Ltd., 1986 12 (SCC), [1986] 2 S.C.R. 713, per Dickson C.J., at p. 772) and another thing entirely to abandon careful scrutiny of a legislative scheme which directly abridges a guaranteed right particularly in relation to an aspect of the criminal law. [Emphasis added.]

In my view, this is not a situation calling for a departure from the strict standard of review set forth in Oakes. On the contrary, the issue on appeal seems to be the quintessential case of the state acting as the "singular antagonist" of a very basic legal right of the accused rather than in the role of "mediating between different groups" as discussed in Irwin Toy. This is, in my view, an appropriate case in which to apply the stricter standard of review on the "minimal impairment" issue.

The appellants submit that the government's objective can be met by imposing only an evidentiary burden on the accused and they cite the American experience in support. I agree that the American experience is relevant to the minimal impairment issue as well as to the existence of a pressing and substantial concern because a substantial number of states have the lesser requirement that once the issue of insanity is put into play by the accused, the prosecution must prove his or her sanity beyond a reasonable doubt: 24 states as of 1983: see American Bar Association's First Tentative Draft: Criminal Justice Mental Health Standards, op. cit., at p. 292. This is also the rule which has governed all federal crimes in the United States ever since the Davis case.

As mentioned earlier, the U.S. Supreme Court specifically rejected in Davis the argument that insanity is a fairly easy condition to feign or simulate and that "[m]erely doubtful evidence of insanity would fill the land with acquitted criminals". The case made by Harlan J., speaking for the unanimous court, may have become even stronger as the state of psychological and psychiatric knowledge has increased. The argument is sometimes advanced that feigning insanity is easy but in fact it appears that nothing is further from the truth. As the body of scientific and diagnostic knowledge about mental illness develops and is consolidated by interdisciplinary research, the disease becomes more and more clearly defined. As Professor Weihofen stated in Mental Disorder as a Criminal Defense (1954), at p. 46:

Actually, it would in most cases be extremely difficult for an imposter to mislead a competent psychiatrist...

The ordinary malinger does not realize that the various forms of mental disorder have their characteristic symptoms, and that displaying a hodge-podge of symptoms will not only fail to convince, but will rather clearly reveal his malingering.

He does, however, caution that in order to arrive at an accurate diagnosis of malingering in difficult cases, some opportunity for examination and observation by impartial experts should be put in place. Nevertheless, we can say as a general rule that, everything else being equal, as scientific knowledge about mental illness increases the possibility of a successful feigned insanity plea diminishes.

Assuming for the moment that the standard of proof on the accused is reduced to a purely evidentiary burden, what is the burden on the prosecution? Lamer C.J. has characterized it at pp. 000 and 000 as the "impossible burden ... of proving an accused's sanity in order to secure a conviction". (Emphasis added.)

As already mentioned, I believe that the Chief Justice has overstated the burden that would lie on the Crown in the absence of s. 16(4). If the plea of insanity pursuant to s. 16 of the Code is viewed as a defence or a justification or excuse, then the procedure set forth by Dickson C.J. in R. v. Holmes, supra, would govern. He said at p. 935:

The basic principle of the common law has been that the accused need not prove a defence. Once an accused raises the possibility that a defence exists, whether by pointing to some fact in the Crown evidence or by leading defence evidence, the Crown is required to disprove the defence beyond a reasonable doubt. The common law has not distinguished in this area between defences that challenge the existence of a necessary element of the offence and those defences that admit the mens rea and actus reus but avoid criminal liability because of circumstances that excuse or justify the conduct. [Emphasis added.]

In the absence of s. 16(4) the onus would still be on the accused to adduce evidence that makes insanity "a live issue fit and proper to be left to the jury" (R. v. Gill (1963), 47 Cr. App. R. 166, at p. 172). The burden on the Crown would then be to remove any doubt raised by the accused in the jury's mind as to the presence of any of the elements of insanity as set out in ss. 16(2) and 16(3).

Chief Justice Lamer acknowledges that the insanity defence is invoked infrequently given the significant constraint on liberty that follows a successful plea. Nevertheless, he is of the view that it would be invoked more often if the standard of proof were lowered to a purely evidentiary burden on the accused. There is no support for this proposition and the experience in the United States seems to be to the contrary. This is not surprising since a successful insanity plea is not an open door to freedom. It is a mandate for confinement in an appropriate institution for treatment rather than confinement simpliciter. The mean length of time of LGW detentions is six years and four months: see Hodgins et al., op. cit., Table 5. Moreover, the principle already stated that it is better that a guilty person be found not guilty by reason of insanity and committed for psychiatric treatment than that an insane person be convicted of a crime is, in my view, fully in keeping with both Charter values and the most basic tenets of our criminal justice system.

For these reasons, I am not persuaded that s. 16(4) impairs the accused's right to be presumed innocent as little as is reasonably possible. Rather, I am of the view that the government's objective could be quite readily met by imposing a purely evidentiary burden on the accused. The infringement on s. 11(d) of the Charter resulting from s. 16(4) is accordingly not saved by s. 1.

While striking down s. 16(4) may represent a substantial departure from our criminal law tradition, I believe that it is necessitated by the expansive reading this Court gave to s. 11(d) of the Charter in Whyte and by the absence of any hard evidence to show that the provision constitutes a reasonable and demonstrably justified limit under s. 1.

It follows, of course, from the above analysis that the common law principle which is reflected in s. 16(4) also infringes s. 11(d) and is not saved by s. 1.

Morally or Legally Wrong: s. 16(2)

On the issue of whether the word "wrong" in s. 16(2) of the Code should be interpreted narrowly as meaning "legally wrong" or more broadly as meaning "morally wrong", I agree with the Chief Justice that the word is properly construed as "morally wrong".

Specific Delusions: s. 16(3)

On the issue of "specific delusions" under s. 16(3) of the Code, while I agree with the Chief Justice that most cases of specific delusions will now be caught under the second branch of s. 16(2), I am reluctant to adopt his statement that "there can be no doubt that any successful attempt to invoke the insanity defence under s. 16(3) would also succeed under s. 16(2)" (p. 000). I am concerned that there may be "specific delusions" which are not necessarily attributable to "disease of the mind". I note that in R. v. Abbey, 1982 25 (SCC), [1982] 2 S.C.R. 24, this Court accepted (without comment) the trial judge's conclusion that the specific delusion from which the accused was suffering in that case constituted "disease of the mind" within the meaning of s. 16(2). That case is not, however, authority for the broader proposition that every specific delusion necessarily stems from disease of the mind. Since legislative provisions should not readily be held to be redundant, particularly ones which are favourable to the accused, I would prefer to leave s. 16(3) open to an accused who might not qualify for the insanity defence under the second branch of s. 16(2).

Splitting the Case

I agree with Lamer C.J. that no issue of the Crown splitting its case arises here and that the evidence adduced to rebut the defence of insanity was properly admitted by the trial judge.

Disposition

I would allow the appeal and order a new trial to be conducted on the basis of the legal principles set out above. I would answer the constitutional questions set by Dickson C.J. as follows:

1. Is s. 16(4) of the Criminal Code of Canada inconsistent with s. 11(d) of the Canadian Charter of Rights and Freedoms?

Yes.

2. If the answer to question 1 is yes, is s. 16(4) justified by s. 1 of the Canadian Charter of Rights and Freedoms and therefore not inconsistent with the Constitution Act, 1982?

No.

//McLachlin J.//

The reasons of L'Heureux‑Dubé and McLachlin JJ. were delivered by

McLachlin J. (dissenting) ‑‑ This case raises fundamental questions relating to the presumption of sanity in the Criminal Code, R.S.C., 1985, c. C-46. First, does the presumption of sanity offend the presumption of innocence embodied in s. 11(d) of the Canadian Charter of Rights and Freedoms? Second, how should the definition of insanity in s. 16(2) of the Code be interpreted: when the Code speaks of "knowing that an act or omission is wrong", does it mean legally "wrong" or morally "wrong"? I respectfully disagree with Lamer C.J. and Wilson J. on each of these issues.

Is s. 16(4) of the Code inconsistent with s. 11(d) of the Charter?

Construction of s. 16 of the Criminal Code

The question posed is whether s. 16(4) of the Criminal Code, which provides that every one is presumed sane until the contrary is proved, is invalid because it infringes the presumption of innocence found in s. 11(d) of the Charter.

The problem can be approached in two different ways. The first option is to treat sanity for the purposes of analysis as relating to an essential element of a criminal offence or to an exculpatory defence: Colvin, "Exculpatory Defences in Criminal Law" (1990), 10 Oxford J. Legal Stud. 381, at p. 394. This approach inexorably leads, as the Chief Justice's reasons demonstrate, to the conclusion that the presumption of sanity infringes s. 11(d) of the Charter. The question then becomes whether it is justified under s. 1 of the Charter.

The second option sees sanity as relevant, not to the essential elements of or exculpatory defences to a criminal offence, but rather as the basic pre-condition of criminal responsibility. On this view, the question of sanity relates not to guilt or innocence, but to the more fundamental issue of whether the accused can fairly be held criminally responsible for his or her acts or omissions.

Professor Alan Mewett identifies these alternate approaches to the problem in "Insanity, Criminal Law and the Charter" (1989), 31 Crim. L.Q. 241. He states, at p. 241:

One assumes that it will not be difficult to uphold the validity of [the presumption of sanity] either by using s. 1 of the Charter, it being impossible to prove sanity, or simply because the issue of insanity arises at a tangent to the basic principles of criminal liability and does not affect the prosecution's burden to prove beyond a reasonable doubt everything that constitutes guilt.

Lamer C.J. takes the view that the presumption of sanity in s. 16(4) violates the presumption of innocence found in s. 11(d) but that, as Professor Mewett suggests, the provision is saved under s. 1 of the Charter. I, on the other hand, favour the alternative route identified by Professor Mewett: because the question of insanity "arises at a tangent to the basic principles of criminal liability", the presumption found in s. 16(4) does not offend s. 11(d) of the Charter. In my opinion, this approach best accords with the rationale underlying the insanity provisions of our criminal law as well as the wording of s. 16 of the Criminal Code.

The Underlying Rationale

I turn first to the rationale underlying the insanity provisions in the criminal law.

At the heart of our criminal law system is the cardinal assumption that human beings are rational and autonomous: G. Ferguson, "A Critique of Proposals to Reform the Insanity Defence" (1989), 14 Queen's L. J. 135, at p. 140. This is the fundamental condition upon which criminal responsibility reposes. Individuals have the capacity to reason right from wrong, and thus to choose between right and wrong. Ferguson continues (at p. 140):

It is these dual capacities -- reason and choice -- which give moral justification to imposing criminal responsibility and punishment on offenders. If a person can reason right from wrong and has the ability to choose right or wrong, then attribution of responsibility and punishment is morally justified or deserved when that person consciously chooses wrong.

The requirement of moral blameworthiness for attribution of responsibility and punishment dates back to the origins of western ethical and legal thought: I. Keilitz and J. P. Fulton, The Insanity Defense and its Alternatives: A Guide for Policymakers (1984), at p. 5. Aristotle, for example, reasoned that capacity for choice was central to the issue of moral culpability: J. M. Quen, "Anglo-American Concepts of Criminal Responsibility: A Brief History", in S. J. Hucker, C. D. Webster, M. H. Ben-Aron, eds., Mental Disorder and Criminal Responsibility (1981), 1, at p. 1. Where a person lacks this capacity for choice because he or she is not capable of knowing that his or her acts are wrong, the moral justification for attribution of responsibility and punishment will be absent for, as Ferguson, op. cit., observes at p. 140, "It is immoral to punish those who do not have the capacity to reason or to choose right from wrong".

The long history of insanity provisions in Anglo-Canadian criminal law reflects this fundamental nexus between capacity for rational choice and legal blameworthiness. In explaining the rationale underlying the present s. 16 of the Criminal Code, the Law Reform Commission of Canada fastened on this consideration (Working Paper 29, Criminal Law: The General Part -- Liability and Defences (1982), at p. 42):

[T]he defence of insanity rests on the fundamental moral view that insane persons are not responsible for their actions and are not therefore fit subjects for punishment.

Parliament, through s. 16, has defined "fitness for punishment" in terms of capacity for the knowledge of right and wrong rather than volitional impairment (see Ferguson, op. cit., at p. 143).

These then are the historical and philosophical underpinnings of the universal notion that insane persons should not be held criminally responsible for their acts and omissions in the same way that sane persons are. They reflect a fundamental conviction that criminal responsibility is appropriate only where the actor is a discerning moral agent, capable of making choices between right and wrong. This is the fundamental pre-condition for imposition of criminal liability.

The existence of sanity as the fundamental pre-condition of criminal responsibility is not negated by the fact that insanity may also be seen as relevant to the question of whether the essential elements of a criminal offence have been established or whether an exculpatory defence is raised. As Lamer C.J. points out, insanity may be seen as negating the mens rea of the offence or the presence of a voluntary, conscious act, required for the actus reus of the offence: see, for example, Rabey v. The Queen, 1980 44 (SCC), [1980] 2 S.C.R. 513. Similarly, insanity might be relevant to an exculpatory defence in the nature of an excuse or justification, such as provocation or self-defence. But it does not follow from the fact that insanity may be relevant to the elements of a criminal offence or to defences, that it should be regarded as only relevant to the essential elements of, or exculpatory defences to, criminal offences, or that s. 16 of the Criminal Code should be read as focussing on them. The fact that insanity might be relevant to these matters does not preclude the notion honoured from Aristotle's time to the present that the question of guilt or innocence arises only where the actor is a sane, responsible agent capable of discerning right and wrong.

This point is made by Mewett and Manning, Criminal Law (2nd ed. 1985), who specifically counsel against viewing the insanity provisions of the Criminal Code merely in terms of the essential elements of, or exculpatory defences to, an offence (at pp. 254-55):

A problem arises if one attempts to fit insanity into the confining limits of actus reus and mens rea. While insanity may negative either actus reus or mens rea, it is suggested that this fact is merely incidental to the scope of the issue which, put shortly, states that whether or not the actus reus and mens rea are affected, a person within s. 16 is entitled to be acquitted.

The Language of s. 16

Having established that the underlying rationale of our insanity provisions is the broad concept that criminal responsibility should be confined to persons capable of discerning between right and wrong, I turn to s. 16 of the Code itself. In my view, an examination of the wording and functioning of s. 16 confirms that it should be read as relating to this fundamental pre-condition for the assignment of criminal responsibility rather than to the elements of an offence or to particular defences.

The language of the Code makes it clear that s. 16 is concerned only with capacity for criminal responsibility. Section 16(2) provides that a person is insane when, due to disease of the mind, the person is incapable of appreciating the nature or quality of an act or omission or of knowing that an act or omission is wrong. As Mewett and Manning point out at p. 234, the proper inquiry is thus not into what the accused actually appreciated but, rather, into what the accused's capacity was.

[I]n insanity the ultimate question is the accused's capacity -- it is not concerned with what he actually appreciated and is thus not concerned with actual mens rea or actual actus reus. Insanity is a defence because it affects the capacity of the accused.

By contrast, insanity as it relates to the essential elements of an offence or defences to it, on the other hand, is concerned not with capacity, but with the actual state of mind of the accused. To focus on insanity as somehow denying an essential element of the offence or establishing a defence is to fail to appreciate the proper operation of the insanity provisions found in s. 16 of the Code. It confounds the question of capacity for criminal responsibility with the quite different question of what the accused actually appreciated.

It is true, of course, that an accused who does not have the capacity to appreciate something cannot have appreciated it; the point, however, is that the insanity inquiry never looks beyond capacity to actual mens rea or actus reus. For practical purposes, where insanity becomes an issue at trial, there will be objective evidence from which, absent the claim of insanity under s. 16 of the Code, the trier of fact will be justified in inferring the existence of the essential elements of the offence, i.e. the actus reus and mens rea. The claim of insanity, however, pre-empts the traditional inference-drawing process on the ground that a person without the capacity for choice as defined in s. 16 of the Criminal Code is not morally culpable. Because of lack of capacity, therefore, the issue of actus reus and mens rea never arises.

This capacity for choice is a fundamental prerequisite to attribution of criminal responsibility and punishment. To limit application of the insanity provisions to situations where mental disorder within s. 16 negatives mens rea would fail to accord with this precept; moreover, and importantly, it is not what s. 16 of the Code directs. A person can be found insane within s. 16 even where mens rea in the strict sense can be inferred. As Ferguson, op. cit., explains, any other approach would be without moral justification (at p. 140):

An insane person, particularly a psychotic person who has delusions, may have mens rea in the strict sense, for example, an intent to kill, but it is a mens rea concocted in an irrational mind. For one who is insane the normal controls, beliefs and perceptions of reality which influence the right-minded citizen are absent or impaired. Thus the capacity to reason or choose are [sic] impaired even though mens rea in the narrow sense exists.

This logic is equally applicable to exculpatory defences as to the elements of mens rea and actus reus. A defence may be defined as "any claim which, if accepted, would necessitate an acquittal": Colvin, Principles of Criminal Law (1986), at p. 163. But a successful claim of insanity does not produce an unqualified acquittal. Professor Colvin, in "Exculpatory Defences in Criminal Law", op. cit., makes this point (at p. 392):

When insanity provides an exculpatory defence, the actor remains very much the concern of the criminal law. The insanity rules identify special mental conditions under which persons cannot be expected to ensure that their conduct conforms to the requirements of law; and therefore the general law of criminal culpability is unsuited. The actor is formally acquitted because mental impairment has made the standard penal sanctions inappropriate. Alternative coercive measures may, however, be taken because of the potential dangerousness of the condition. [Emphasis added.]

A regime which relieves a person from traditional criminal sanction but substitutes alternative coercive measures, as the Criminal Code does when a person charged with an indictable offence is found to be insane within the terms of s. 16 of the Code, is not one which treats insanity as a true exculpatory defence. As Professor Richard Mahoney writes in "The Presumption of Innocence: A New Era" (1988), 67 Can. Bar Rev. 1, at p. 14, implicit in the nature of a true defence is the notion that:

... an accused who has a true defence is just as "innocent" as an accused who has not been shown to have committed all the essential elements of the crime charged.

Were insanity under s. 16 of the Code a true defence, we would expect that a finding of insanity within the terms of s. 16 of the Code would result in the same disposition, from the point of view of the accused, as a finding that the accused has not been shown to have committed the essential elements of the offence. That, of course, is not the case; an accused who is found to be insane within the terms of s. 16 of the Code is committed for treatment under a Lieutenant-Governor's warrant. Implicit in this verdict is a determination that while the accused cannot be held morally culpable for his or her conduct, the accused is not absolutely acquitted in the same way as one who either has not been shown to have committed the essential elements of the offence or who does have a true defence.

I do not attach undue significance to the wording of the special verdict -- "acquitted on account of insanity" -- which s. 614 of the Criminal Code directs for persons charged with an indictable offence who are found to be insane within the terms of s. 16 of the Code. To refer again to the passage from Professor Colvin quoted above, while an insane person is "formally acquitted", "alternative coercive measures" are imposed -- namely, potentially indefinite detention under a Lieutenant-Governor's warrant. To equate an acquittal on account of insanity with an acquittal "on account of innocence" (by which I mean absence of the essential elements of the offence or presence of an exculpatory defence) is to deny the very different consequences which flow from those two verdicts and, in truth, to allow form to triumph over substance.

The conclusion that insanity does not qualify as a defence in the true sense of the term, finds support in the fact that the Crown may lead evidence of insanity, notwithstanding that the accused may have disavowed the claim and subject to a discretion in the trial judge to exclude evidence of insanity when the interests of justice so require: R. v. Simpson (1977), 1977 1142 (ON CA), 35 C.C.C. (2d) 337 (Ont. C.A.); R. v. Saxell (1980), 1980 1623 (ON CA), 59 C.C.C. (2d) 176 (Ont. C.A.). This principle reflects the view that insanity is not a defence per se but, rather, that sanity is a condition precedent to attribution of criminal responsibility and punishment; the accused must be sane before any consideration of the essential elements of the offence or exculpatory defences becomes relevant.

The place of the insanity provisions of the Criminal Code provides further support for the notion that Parliament viewed them as establishing the basic pre-condition for criminal responsibility rather than in terms of specific elements or defences relevant to criminal offences. Section 16 finds itself near the beginning of the Code. It may be viewed as logically related to s. 13 of the Code which provides that "No person shall be convicted of an offence in respect of an act or omission on his part while that person was under the age of twelve years". In each case, the issue is whether it is fair or just to hold the person in question criminally responsible. In neither case are the elements of the offence or potential defences relevant.

Finally, if insanity were concerned only with the essential elements of an offence or defences to it, s. 16 would be unnecessary. It would be sufficient to leave it as a matter of evidence to be considered in determining whether the Crown had discharged its burden of proving the guilt of the accused beyond a reasonable doubt.

I arrive then at this conclusion. To conceive the insanity provisions of the Criminal Code narrowly in terms of the essential elements of criminal offences or exculpatory defences ignores the historical and philosophical origins of the fundamental precept that sanity is a pre-condition of criminal responsibility. It violates the language of s. 16 of the Code, which refers to capacity for criminal responsibility rather than actual states of mind. It is at odds with the fact that insanity in s. 16 can be raised by the Crown in circumstances where neither the elements of the offence nor a defence are at issue. And it confuses true acquittal, the result of the absence of an essential element of an offence or the presence of a defence to it, with formal acquittal coupled with alternative coercive measures because mental impairment renders the imposition of true penal responsibility inappropriate. Rather than straining to confine the insanity provisions in the dual strait-jacket of the "elements" of or "exculpatory defences" to an offence, I prefer to view s. 16 as referring to a more basic precept of the criminal law system -- the notion that the attribution of criminal responsibility and punishment is morally and legally justifiable only for those who have the capacity to reason and thus to choose right from wrong.

Section 16 of the Code and the Charter

The next question is whether the presumption of sanity, viewed as the fundamental pre-condition of criminal responsibility, offends the presumption of innocence embodied in s. 11(d) of the Charter. The answer to this question must be negative. The presumption of innocence found in s. 11(d) of the Charter is merely another way of expressing the principle that the Crown must prove an accused's guilt beyond a reasonable doubt. That being the purpose of the presumption of innocence, it follows that the presumption of sanity cannot be contrary to s. 11(d) because, as Professor Mewett observed in the passage set out earlier, the issue of insanity "does not affect the prosecution's burden to prove beyond a reasonable doubt everything that constitutes guilt" (p. 241). The presumption of sanity in s. 16(4) of the Criminal Code merely relieves the Crown from establishing that the accused has the capacity for rational choice which makes attribution of criminal responsibility and punishment morally justifiable. The Crown must still prove the guilt of the accused -- i.e. the actus reus, the mens rea, and the absence of exculpatory defences raised on the evidence -- beyond a reasonable doubt.

I thus conclude that as a matter of logic, the presumption of sanity does not violate s. 11(d). But given that s. 11(d) reflects the principles of fundamental justice basic to criminal procedure referred to more generally in s. 7, it may not be amiss to consider whether this result will in fact create an injustice. In my view, such evaluation may serve as a useful check in considering the ambit of principles of justice, and is not confined to s. 1.

It is my conclusion that the presumption of sanity, viewed as I have suggested, does not violate the fundamental notion of procedural fairness which underlies the procedural guarantees of the Charter, of which s. 11(d) is one. The criminal law, from time immemorial, has presumed that persons are sane and responsible. This presumption reflects the dignity which the law accords to each human being, and avoids the practical difficulties associated with requiring the Crown to prove in each case that the accused was sufficiently sane to justify the imposition of criminal responsibility. The practical and fair approach is to presume sanity and responsibility, subject to certain exceptions -- such as juveniles -- and subject to the accused's right to rebut the presumption on grounds of insanity. Viewed thus, the provisions of the Criminal Code dealing with insanity, far from violating a fundamental principle of justice, reflect the fundamental precepts upon which our legal system and our Charter are based.

There is no suggestion that the present system results in injustice. The system as it exists in this country seems to be working well. Indeed, the debate between Lamer C.J. and Wilson J. is not as to whether the present law creates practical injustice which must be remedied, but whether changing the system would create an injustice.

The reality suggested by the Canadian experience is that accused persons appear to have no difficulty establishing insanity on a balance of probabilities where it exists. The diagnosis, depending as it does on the accused's state of mind, is peculiarly within the accused's power. Nor is there any suggestion that accused persons are unable to obtain whatever expert assistance as they may require to establish the plea. These features distinguish the presumption of sanity from the evil of requiring the accused to disprove the elements of an offence, upon which the presumption of innocence rests. A requirement that an accused disprove the elements of the offence, has long and universally been recognized by our legal system as raising a grave danger of injustice; notwithstanding current debate, the same cannot be said of the presumption of sanity.

Put at its most fundamental, the question posed by this case is this: should the presumption of innocence, a fundamental tenet of the criminal law, be extended to negate the equally fundamental principle of criminal law that the accused is presumed sane for purposes of determining whether he or she should be found capable of criminal responsibility? If it were demonstrated that the time-honoured presumption of sanity raised the same concerns as those addressed by the presumption of innocence or produced injustice by placing an undue burden on the accused, a case might be made that the presumption of innocence should be so extended. But the concerns are not the same, and no injustice is demonstrated. In the absence of such considerations, this Court should not, under the aegis of the Charter, rewrite the criminal law as it has existed for centuries by abrogating the presumption of sanity.

It remains to consider whether the view I propose is precluded by prior jurisprudence in this Court. In my opinion, it is not. In R. v. Oakes, 1986 46 (SCC), [1986] 1 S.C.R. 103, this Court ruled that requiring the accused to establish an essential element of an offence offended the presumption of innocence in s. 11(d) of the Charter. In R. v. Whyte, 1988 47 (SCC), [1988] 2 S.C.R. 3, the Court, per Dickson C.J., suggested that a violation might be established where the accused was required to disprove an essential element or to prove "a collateral factor, an excuse, or a defence" (p. 18). While the ratio of Whyte is confined to defences (which the presumption of sanity cannot be), it is not amiss to consider the import of the obiter reference to "collateral factor".

Can the presumption of sanity be considered a "collateral factor" within Whyte? I think not. The Court in Whyte was not concerned with the fundamental pre-conditions for the imposition of criminal responsibility. The fundamental precept that moral culpability depends on the existence of capacity for choice falls entirely outside the intended scope of the judgment in Whyte. Sanity is not merely a "collateral factor" which must be established in order to convict; it is, rather, a condition precedent to any attribution of criminal responsibility and punishment.

I conclude that the pronouncements of this Court on s. 11(d) of the Charter are consistent with the conclusion that s. 16(4) of the Criminal Code does not violate the presumption of innocence enshrined in s. 11(d) of the Charter.

In view of these conclusions, I need not consider the application of s. 1 of the Charter.

The Meaning of "Wrong" in s. 16(2) of the Code

Section 16(2) of the Criminal Code defines insanity. There are two requirements. The first is a "disease of the mind." The second is that the disease be to an extent that it renders the person incapable either, of (a) appreciating the nature and quality of an act or omission; or (b) of knowing that an act or omission is wrong.

The interpretation of the second arm of the second condition arose at the trial in this case. Following earlier decisions of this Court (in particular Schwartz v. The Queen, 1976 165 (SCC), [1977] 1 S.C.R. 673), the trial judge instructed the jury that this condition was filled if the appellants were capable of knowing that their act was legally wrong. The fact that the appellants may not have been capable of appreciating that the act was morally wrong, i.e., that they may have thought it was morally justifiable although legally wrong, was irrelevant.

Lamer C.J. has accepted the appellants' invitation to reconsider this Court's earlier conclusion that the capacity to know the act or omission was legally wrong suffices. In his view, an accused who is capable of knowing an act or omission is legally wrong is not subject to the criminal process if mental illness rendered him or her incapable of knowing the act or omission was morally wrong. I, on the other hand, take the view that it does not matter whether the capacity relates to legal wrongness or moral wrongness -- all that is required is that the accused be capable of knowing that the act was in some sense "wrong". If the accused has this capacity, then it is neither unfair nor unjust to submit the accused to criminal responsibility and penal sanction.

The latter position is supported in my view by: (a) the plain language of s. 16(2); (b) an historical view of our insanity provisions; (c) the purpose and theory underlying our insanity provisions; and (d) by practical difficulties related to the problem of determining what is "morally wrong". I will consider each of these arguments in turn.

The Language of s. 16(2)

I turn first to the wording of s. 16(2) of the Criminal Code. It uses the word "wrong" without modification. Had Parliament intended it to mean a specific kind of wrong, one would have expected Parliament to have said so. To know an act is wrong is to know that one ought not to do it. The reason for which one ought not to do it is a collateral and separate point -- a point which does not arise on a plain reading of s. 16(2). All that is required to negate the pre-condition for criminal responsibility of sanity is that mental disease has rendered the accused incapable of knowing that he or she ought not to have done the act or neglected the omission in question.

This is the view of Professor Mewett in "Section 16 and `Wrong'" (1976), 18 Crim. L.Q. 413. At page 416 he states:

The question, it seems to me, is not whether the accused thought an act was morally wrong or legally wrong but whether he was incapable of knowing that he ought not to do it. [Emphasis added.]

The History of our Insanity Provisions

An historical view of our insanity provisions supports the conclusion that "wrong" in s. 16(2) should be read simply in the sense of what ought not to be done or omitted. Section 16(2) finds its roots in M'Naghten's Case (1843), 10 Cl. & Fin. 200, 8 E.R. 718. Its wording is borrowed from the words of Lord Tindal C.J., who stated at p. 722:

...to establish a defence on the ground of insanity, it must be clearly proved that, at the time of the committing of the act, the party accused was labouring under such a defect of reason, from disease of the mind, as not to know the nature and quality of the act he was doing; or, if he did know it, that he did not know he was doing what was wrong. [Emphasis added.]

Lord Tindal C.J. went on to explain what he meant by "wrong" (at p. 723):

If the accused was conscious that the act was one which he ought not to do, and if that act was at the same time contrary to the law of the land, he is punishable... [Emphasis added.]

It is hardly surprising that Professor Colvin suggests that the original formulation of the M'Naghten rule supports neither the "legally wrong" nor the "morally wrong" standard: Principles of Criminal Law, op. cit., at p. 252.

This was the way courts subsequently interpreted M'Naghten. Thus, in R. v. Codere (1916), 12 Cr. App. R. 21, Lord Reading C.J. stated (at pp. 27-28):

That is the accepted test, and applying it in this case there can be no room for doubt; once it is clear that the appellant knew that the act was wrong in law, then he was doing an act which he was conscious he ought not to do, and as it was against the law, it was punishable by law....

The difficulty no doubt arises over the words "conscious that the act was one which he ought not to do," but, looking at all the answers in M'Naghten's case, it seems that if it is punishable by law it is an act which he ought not to do.... [Emphasis added.]

This language leaves no doubt that what is essential is that the accused know that he or she ought not to do the act in question. This condition is met if the accused knows that the act is legally wrong.

More than a century after M'Naghten, in R. v. Windle, [1952] 2 Q.B. 826 (C.C.A.), Lord Goddard C.J. held that "wrong" should mean only legally wrong. For all practical purposes, he was only saying what Lord Reading C.J. had said in Codere. The question is whether the accused was capable of knowing that he or she ought not to do the act, and knowledge that it is illegal suffices to meet this test, making further enquiry into moral awareness unnecessary. (It may be that Lord Goddard C.J. went too far in saying all that could ever be relevant is legal wrong. One can imagine a case, although it must be rare, where an accused is incapable of appreciating that an act is legally wrong but capable of appreciating that it is morally wrong. In such a case, he or she arguably should be criminally responsible on the M'Naghten "ought to have known" test.)

Debate about the reason for which a person is incapable of knowing what he or she ought not to do can be traced to Windle and the High Court of Australia's rejection of this approach in Stapleton v. The Queen (1952), 86 C.L.R. 358. It finds its place in the dissenting reasons of this Court in Schwartz. My point at this juncture is that the debate is a relatively recent one, and that for most of the period in which M'Naghten's Case has been the rule, the only question was whether the accused was capable of knowing that he or she ought not to do the act in question, for whatever reason.

The Purpose and Theory of our Insanity Provisions

I turn now to the purpose and theory underlying the insanity provisions. In my view, they too support the view that "wrong" in s. 16(2) of the Code means simply that which one "ought not to do." The rationale behind the insanity provisions, as discussed earlier in these reasons, is that it is unfair and unjust to make a person who is not capable of conscious choice between right or wrong criminally responsible. Penal sanctions are appropriate only for those who have the ability to reason right from wrong, people capable of appreciating what they ought and ought not to do. A person may conclude that he or she ought not to do an act for a variety of reasons. One may be that it is illegal. Another may be that it is immoral. The reasons for which one concludes that one ought not to do an act are collateral to the fundamental rationale behind the insanity provisions -- that criminal conviction is appropriate only where the person is capable of understanding that he or she ought not to do the act in question.

The wider rationale underlying the criminal law generally supports the same view. While other factors may figure, two main mechanisms function to keep people's conduct within the appropriate legal parameters: (1) a sense of morality, and (2) a desire to obey the law. In most cases, law and morality are co-extensive, but exceptionally they are different. Where morality fails, the legal sanction should not be removed as well. To do so is to open the door to arguments that absence of moral discernment should excuse a person from the sanction of the criminal law, and thus remove one of the factors which deters inappropriate and destructive conduct. That should not be done lightly. The fact that such arguments could not be entertained without establishing a "disease of the mind" is small comfort when one takes account of the difficulty of defining or diagnosing "disease of the mind." Recent research seems to suggest that the vast majority of forensic psychiatrists and psychologists, including those who have given evidence with respect to legal insanity in a large number of cases, have no effective understanding of the legal test about which they are expressing an opinion: R. Rogers and R. E. Turner, "Understanding of Insanity: A National Survey of Forensic Psychiatrists and Psychologists" (1987), 7 Health L. Can. 71. See also J. Ziskin and D. Faust, Coping with Psychiatric and Psychological Testimony (4th ed. 1988), vol. I, at pp. 389-408.

To hold that absence of moral discernment due to mental illness should exempt a person who knows that legally he or she ought not to do a certain act is, moreover, to introduce a lack of parallelism into the criminal law; generally absence of moral appreciation is no excuse for criminal conduct. When the moral mechanism breaks down in the case of an individual who is sane, we do not treat that as an excuse for disobeying the law; for example, in the case of a psychopath. The rationale is that an individual either knows or is presumed to know the law, and the fact that his or her moral standards are at variance with those of society is not an excuse. Why, if the moral mechanism breaks down because of disease of the mind, should it exempt the accused from criminal responsibility where he or she knows, or was capable of knowing, that the act was illegal and hence one which he or she "ought not to do". Why should deficiency of moral appreciation due to mental illness have a different consequence than deficiency of moral appreciation due to a morally impoverished upbringing, for example? I see no reason why the policy of the law should differ in the two cases.

The Difficulty of Determining What Is Morally Wrong

I turn finally to the practical difficulty imposed by permitting persons who know they ought not to do an act for legal reasons from escaping criminal responsibility on the ground that they were incapable of realizing that the act was immoral. As Colvin, Principles of Criminal Law, op. cit., observes at p. 253, "There is no system of authoritative moral pronouncements which is equivalent to that found in law".

In Mental Disorder and the Criminal Trial Process (1978), Marc Schiffer expands on this point (at p. 134):

If ... we were to judge wrongfulness by the moral standards of society (as the minority in Schwartz suggested), it is submitted that the right-wrong test would become virtually meaningless. In the case of certain crimes (e.g. abortion), even the most lucid individual would have trouble appraising society's views without conducting an opinion poll. In the case of other crimes (e.g. rape) the most severe psychotic might know that they are morally condemned by society.

Although abortion is no longer the subject of criminal sanction, it is not difficult to conceive of other criminally proscribed acts (e.g. euthanasia) where wrongfulness in the eyes of society may not necessarily coincide with wrongfulness in the eyes of the law.

The problem with making capacity to appreciate moral wrong the test for criminal responsibility where the incapacity is caused by mental illness is that of determining what society's moral judgment will be in every situation. What result is to obtain on those occasions where an accused claims an incapacity to know that his or her unlawful act was morally wrong and, objectively, the act was one for which the moral wrongfulness can be disputed? Certainly a court is in no position to make determinations on questions of morality, nor is it fair to expect a jury to be able to agree on what is morally right or morally wrong. The prospect of greater certainty, and the avoidance of metaphysical arguments on right and wrong is the chief advantage of adhering to the traditional M'Naghten test for criminal responsibility where causative disease of the mind exists -- whether the accused, for whatever reason, was capable of appreciating that his or her act is wrong.

The importance of certainty in the criminal law cannot be over-estimated. It should be relatively clear when criminal responsibility attaches and when it does not if the criminal law is to have the requisite deterrent effect and if it is to be seen to function fairly and equitably to all. A person's criminal responsibility should not hinge on questions of whether an act would be generally perceived as immoral.

Summary

For all these reasons I conclude that "wrong" in s. 16(2) of the Code ought to be interpreted simply in the sense of what one ought not to do, for whatever reason, legal or moral. In practice, this has the effect in most cases that where an accused is capable of knowing that his or her act is legally wrong, he or she will be held liable to the criminal process, regardless of what his or her moral appreciation may have been. In the end, I endorse the conclusion of Professor Mewett, "Section 16 and `Wrong'", op. cit., at pp. 415-16:

With the greatest respect for judges who have wrestled with this problem and for academics who have written on it, in fact this is a non-problem.... The question that ought to be asked, it is submitted, is whether the accused, because of a disease of the mind (first hurdle) was rendered incapable (second hurdle) of knowing that this act was something that he ought not to do (third hurdle). If he was capable of knowing that the act was contrary to law and that he ought not to do an act contrary to law, then the defence should not apply. If he was incapable of knowing that it was contrary to law, but capable of knowing that it was an act condemned by people generally, then again the defence should not apply. But if he was incapable of knowing that the act was contrary to law and incapable of knowing that it was an act condemned by people generally, then the defence should apply. This only leaves a situation where he was capable of knowing that the act was contrary to law but incapable both of knowing that to act contrary to law was condemned by people generally and of knowing that this particular act was condemned by people generally. I would have thought that such an accused (who must be the rarest of all individuals) is precisely one who ought to be found not guilty by reason of insanity. Far from opening any floodgates, such an interpretation scarcely affects the number of persons who would have a valid defence, but on the rare occasions when it is relevant seems extremely important.

The question, it seems to me, is not whether the accused thought an act was morally wrong or legally wrong but whether he was incapable of knowing that he ought not to do it.

Specific Delusions

Assuming that s. 16(3) relating to specific delusions can arise where the conditions of s. 16(2) are not met, I am satisfied that the trial judge left open this possibility in his instructions to the jury and that his instructions were adequate. I would not direct a new trial on this point.

Splitting the Case

The appellant, Morrissette, submits that the trial judge erred in permitting the Crown to split its case by presenting its evidence on insanity in rebuttal.

For the reasons set out earlier, I do not regard the evidence on insanity in this case as going to an essential element of the offence. The Crown was entitled to rely on the presumption of sanity and was under no obligation to present evidence on this issue as part of its case against the accused.

Conclusion

I would dismiss the appeal and answer the constitutional questions as follows:

1. Is s. 16(4) of the Criminal Code of Canada inconsistent with s. 11(d) of the Canadian Charter of Rights and Freedoms?

No.

2. If the answer to question 1 is yes, is s. 16(4) justified by s. 1 of the Canadian Charter of Rights and Freedoms and therefore not inconsistent with the Constitution Act, 1982?

No answer is required.

//Sopinka J.//

The following are the reasons delivered by

Sopinka J. (dissenting) -‑ I have had the advantage of reading the reasons of Chief Justice Lamer, Justice Wilson and Justice McLachlin herein. I agree with the disposition of this appeal proposed by McLachlin J. and with her reasons in respect of all issues except that, while I agree that s. 16(4) of the Criminal Code, R.S.C., 1985, c. C-46, is valid, I arrive at this conclusion for the reasons expressed by the Chief Justice. It follows that I would answer the constitutional question as proposed by the Chief Justice but would dismiss the appeal.

//Gonthier J.//

The following are the reasons delivered by

Gonthier J. -‑ I have had the benefit of reading the reasons of Chief Justice Lamer and Justices Wilson and McLachlin in the present appeal. For the reasons given by McLachlin J., I am of the view that the presumption of sanity embodied in s. 16 of the Criminal Code, R.S.C., 1985, c. C-46, does not offend the presumption of innocence enshrined in s. 11(d) of the Canadian Charter of Rights and Freedoms. I am on the other hand in agreement with the reasons of Lamer C.J. as to the meaning of "wrong" in s. 16(2) of the Criminal Code as signifying "morally wrong" as well as with his reasons on the other issues raised in this appeal.

Consequently, I would answer the constitutional questions as follows:

1. Is s. 16(4) of the Criminal Code of Canada inconsistent with s. 11(d) of the Canadian Charter of Rights and Freedoms?

Answer: No.

2. If the answer to question 1 is yes, is s. 16(4) justified by s. 1 of the Canadian Charter of Rights and Freedoms and therefore not inconsistent with the Constitution Act, 1982?

Answer: No answer is required.

As I join Lamer C.J. in his interpretation of the word "wrong" in s. 16(2), I would likewise allow the appeal and order a new trial.

Appeal allowed and new trial ordered, L'Heureux‑Dubé, Sopinka and McLachlin JJ. dissenting.

Solicitors for the appellant Chaulk: Wolch, Pinx, Tapper, Scurfield, Winnipeg.

Solicitors for the appellant Morrissette: Walsh, Micay and Company, Winnipeg.

Solicitor for the respondent: The Attorney General of Manitoba, Winnipeg.

Solicitor for the intervener the Attorney General of Canada: John C. Tait, Ottawa.

Solicitor for the intervener the Attorney General for Ontario: The Attorney General for Ontario, Toronto.

Solicitor for the intervener the Attorney General of Quebec: Jean-François Dionne et Jacques Gauvin, Ste-Foy.

Solicitor for the intervener the Attorney General for New Brunswick: The Deputy Attorney General of New Brunswick, Fredericton.

Solicitor for the intervener the Attorney General for Alberta: M. J. Watson, Edmonton.

• Chief Justice at the time of hearing.

** Chief Justice at the time of judgment.