The Paradox of Certainty: Retrospective Application of New Self-Defence Provisions in R v Evans

Courts have often struggled to interpret the application of newly enacted law on events that occurred prior to the law’s enactment. The British Columbia Court of Appeal (“BCCA”) tackled the interpretation of retrospective application in R v Evans, 2015 BCCA 46 [Evans]. This decision marks the first consideration by an appellate court on whether the new self-defence provisions, namely, sections 34 and 35 of the Criminal Code, RSC 1985, c C-46, could apply retrospectively. In ultimately determining that the new provisions did not apply retrospectively, the court weighed the balance between legal certainty and potential unfairness for the accused in these circumstances.

Justice Frankel for the BCCA approached this balance by underscoring the importance of clarity of laws which allow citizens to “govern themselves accordingly” (Evans, para 25). If the new self-defence provision could apply retrospectively, then the lines of criminal conduct may blur into absurdity, with the punishment of a criminal act depending “on when the determination of that issue [was] made” (para 27).

Facts and Issues Before the Court

The facts of the case are standard street brawl fare. The accused, Daniel Evans, was convicted of aggravated assault for punching the victim, Dereck Stevens, outside a nightclub in Victoria, BC. Stevens intervened in an argument between Evans and the bouncers after the bouncers evicted Evans and refused him re-entry to retrieve his jacket. Stevens’ intrusion was not appreciated and Evans responded to Stevens with an “expletive so common it does not bear repeating” (para 6).

The utterance of this expletive spurred Stevens to return to the street after retreating into the nightclub, this time with friends fanned out behind him. According to Evans’ testimony, Stevens now had six to eight friends with him. Evans, with his back against a fish and chips shop, felt increasingly threatened by Stevens’ new friends and encroaching presence. Evans swung the first punch, and connected with Stevens’ face.

The sole issue before the trial judge was self-defence under the old self-defence provision of the Criminal Code, which has now been repealed. The jury convicted Evans, dismissing his claim of self-defence under section 37. However, key to the appeal before the BCCA was the 2012 enactment of the Citizen’s Arrest and Self-defence Act, SC 2012, c 9. This legislation replaced sections 34-42 of the Criminal Code with two new consolidated sections under 34 and 35, dealing with defence of person and defence of property, respectively. Section 34 came into force on March 11, 2013 – the first day of Evans’ trial.

There were two issues before the Court of Appeal.

  1. Did the trial judge err in using the self-defence provisions in force at the time of the events in issue rather than the new provisions, in place by the time of the trial?
  2. Did the trial judge err by only instructing the jury on Mr. Evans defending against Mr. Stevens and not also instructing the jury on Mr. Evans defending himself against others present at the time?

In ten paragraphs, the court dismissed the second issue that the jury was not properly instructed on the defence. Prior jurisprudence on how the presence and involvement of a group can contribute to a claim for self-defence, such as R v AF, 2002 BCPC 545, were all distinguishable from these facts. Specifically, in this case it was unclear whether any of Stevens’ friends were actively involved in the confrontation. It was Stevens taking a step towards Evans that caused Evans to feel immediately threatened and to think that Stevens, specifically, was going to hit him. Thus, the Court dismissed Evans’ claim that the trial judge erred in failing to instruct the jury on whether Evans thought his force was reasonable to protect himself from a “hostile group” (Evans, para 36).

The court focuses the bulk of its analysis on the first issue, ultimately concluding that the trial judge did not err in either respect. The rest of this article will focus on the court’s analysis of retrospectivity.

The Court’s Analysis 

The recent jurisprudence from trial courts across Canada on whether this provision should apply retrospectively has been extremely inconclusive. The Court of Appeal was faced with three diverging approaches to retrospective application of section 34 and 35:

  1. New provisions apply prospectively only;
  2. New provisions apply retrospectively; or
  3. Both new and old provisions can be considered, giving the accused the benefit of both.

With little to help the court in analyzing this section specifically, the BCCA looked elsewhere. The Court relied on the 2012 Supreme Court of Canada (“SCC”) decision in R v Dineley, [2012] 3 SCR 272 [Dineley], a leading authority on retrospective application. For the SCC, Justice Deschamps wrote:

Because of the need for certainty as to the legal consequences that attach to past facts and conduct, courts have long recognized that the cases in which legislation has retrospective effect must be exceptional. More specifically, where legislative provisions affect either vested or substantive rights, retrospectivity has been found to be undesirable. New legislation that affects substantive rights will be presumed to have only prospective effect unless it is possible to discern a clear legislative intent that it is to apply retrospectively. (para 10) [emphasis added]

A foundation of statutory interpretation is the presumption that statutes are not to operate retrospectively “unless such a construction is expressly or by necessary implication required by the language of the Act” (para 28).

Counsel for Evans pointed to the judgment of Justice MacDonnell in R v Pandurevic2013 ONSC 2978 [Pandurevic], in order to rebut such a presumption. MacDonnell had written with respect to the amended provisions that: “Parliament’s aim was not to alter the essential nature of the defence of self-defence” (Pandurevic, para 29). Rather, Parliament’s aim was simply clarification of messy provisions that forced juries into “making decisions not because of the legal instructions they received but notwithstanding them” (para 29).

Justice Frankel agreed that one of the purposes of the amendments was clarification of the provisions, but disagreed that Parliament did not intend to “alter the essential nature” of self-defence. Justice Frankel points primarily to the substantive implications of the changed provisions. While the line between substantive rights and procedural rights is often blurry (see Cromwell J’s dissent in Dineley, para 66), the new provisions have quite clearly targeted substantive rights. Many elements in the old section that were conditions to the defence are now part of a non-exhaustive list of nine factors that can be taken into account.

Fundamentally, the old section focused on the use of “proportionate force.” In the new section, proportionality is only one factor, with the focus shifted to whether the act was “reasonable in the circumstances” (Evans, para 19). In fact, scholars have highlighted numerous differences between the two provisions, and the substantive effects of those differences (for an overview of other effects of the amendments, see Kent Roach, “A Preliminary Assessment of the New Self-Defence and Defence of Property Provisions” (2012) 16:1 Can Crim L Rev 275).

MacDonnell J’s position in Pandurevic does seem to agree that the changes were substantive, and also even agrees that the changes are substantive enough to fall under the presumption that Parliament should not be interpreted to have intended retrospective application. Instead, MacDonnell argued that this presumption can be rebuttable, and should be in this case. The necessary implications of Parliament’s aim in clarifying this area of law must be retrospective application; otherwise, the “evils that the legislation was intended to cure [could] linger, perhaps for years” (Pandurevic, para 29).

While Frankel JA agreed that clarification was the intent of the law, he disagreed with MacDonnell J’s argument that such an intent led by “necessary implication” to retrospective application of the sections. Frankel JA wrote that just because “a substantive law is more coherent today than it was before is not a reason to apply it to past events” (Evans, para 33).

Justice Frankel additionally took issue with MacDonnell J’s characterization of the difficulties in giving legal instruction to juries with regards to these provisions. Frankel JA emphasized that judges have found “practical ways to explain self-defence in terms juries can understand” (para 31). This seems to brush aside the long-standing criticisms of sections 34-42. Frankel JA’s characterization that the provisions “were complex and in need of legislative reform” (para 31) doesn’t quite capture the extent of it.

Pandurevic outlines the long history of criticism against the defence of persons sections, from scholars and benchers alike, in paras 10-16 of that decision. Chief Justice Lamer called the sections “internally inconsistent.” Justice Moldaver stated that “it is no secret that many trial judges consider their instructions on the law of self-defence to be little more than a source of bewilderment and confusion to the jury.” Scholar David Paciocco called sections 34 to 37 “the most confusing tangle of sections known to law” (see Pandurevic, paras 10-16).


Justice Frankel took a reasoned, if sometimes rigid, approach to the issue of retrospective application regarding sections 34 and 35. Though never explicit, the underlying repugnancy of retrospectivity certainly underlies much of the court’s reasoning – Frankel JA emphasized the importance of certainty and predictability in the law, where “everyone is presumed to know the parameters established by Parliament and is expected to govern themselves accordingly” (Evans, para 25).

This argument is understandable – the repugnance of retrospective application is pulled into sharp focus when you consider the possibilities of state abuse of power in attaching a punishment to an act which was legal at the time it was committed. McLachlin CJC’s powerful comments in R v Mabior, 2012 SCC 47, are particularly poignant on this front. She wrote:  “The rule of law requires that laws provide in advance what can and cannot be done… Condemning people for conduct that they could not have reasonably known was criminal is Kafkaesque and anathema to our notions of justice” (para 14).

However, Justice MacDonnell worded it well in Pandurevic when he wrote that this repugnance, or antipathy, “may be substantially attenuated … where the ex post facto change to the rules either does not prejudice the actor’s position or improves it” (Pandurevic, para 36). In other words, individual civil liberties are not put in the same state of peril when the retrospective application of laws would be “beneficial” to the accused (Evans, para 34).

Interestingly, Frankel JA noted in his decision that both Evans and the Crown agreed on the new provisions potentially being more favourable to an accused. There are circumstances where an accused will succeed under the new provisions and not under the old ones. While there is no direct discussion on the likely outcome of these facts applied under the new provisions, the emphasis on reasonableness and discretionary nature of the nine non-exhaustive factors under the new provisions may have worked in Evans’ favour.

Justice Frankel argued that even in “beneficial” circumstances such as these, the potential injustice of imposing the old provisions should not be weighed heavier than the importance of certainty and predictability afforded by measuring criminal accountability against the laws in force when the acts were committed. However, Justice Frankel seems to miss the paradox of this balance, given that the laws in question have been consistently criticized for their lack of clarity and uncertain application.

Justice Frankel wrote that it is “neither unfair nor unjust to decide whether an accused committed a crime on the basis of the law in force at the time of the events.” However, if we want the law at the time the crime was committed to apply because it is the law we would expect the accused to guide her actions by, then it must be assumed the law is capable of guiding the accused’s actions in a certain and predictable fashion. Given the varied, vociferous and inveterate criticism of this provision, that does not seem to be a fair assumption in this case and with regards to the application of this section.

Justice Frankel’s decision in Evans is sound in principle – it creates strong long-lasting law that upholds the presumption against retrospective application. The “Kafka-esque” retrospective application of law that punishes past actions under present morality is a dangerous potential that should be avoided at all costs. However, in an ironic twist of application, Frankel JA rigidly applied confusing old law under the guise of upholding certainty in law, in order to allow people to “govern themselves accordingly.”

How the accused would be meant to guide their actions under “the most confusing tangle of sections known to law” is never discussed. As such, not allowing the accused the benefit of the new sections does in fact seem both unfair and unjust. The decision in Evans produces safe law. However, the routes taken to get to there are circuitous and may strike the wrong balance in a case where judicial discretion could have pronounced a fairer outcome for the accused.

You may also like...

Join the conversation

Loading Facebook Comments ...