Would a Canadian Judge Say This?

In C.H. Giles & Co. Ltd. v. Morris, [1972] 1 W.L.R. 307, [1972] 1 All E.R. 960 at 971 (Ch.D., Megarry J.) said:

… In this judgment I have referred to a number of authorities not cited in argument. On the procedural point I have reached no final conclusion, and so the citation of additional authorities in that respect does not raise any particular difficulty. But it is otherwise in relation to the question of specific enforceability. On this, the only authority cited to me by either side was Fry, cited by counsel for the defendants. Wilson v West Hartlepool Railway Co came into the picture because looking at Snell’s Equity during the argument, I found the case cited there on what seemed to be the relevant point, and I drew counsel for the plaintiffs’ attention to it; and at the conclusion of leading counsel for the plaintiffs’ argument his junior read me certain passages from the case in support of the plaintiffs’ cause. The case was thus both mentioned and cited before counsel for the defendants began his reply; and at an earlier stage I had put questions to him based on there being a distinction between a contract for personal services and a contract to enter into a contract for personal services. At the conclusion of the argument, I intimated that I might require further argument, particularly as my copy of the report of Wilson’s case showed that the decision had been referred to in a number of subsequent cases. I have not pursued these references, and find that they relate to issues in Wilson’s case other than that of the specified enforceability of the agreement. In those circumstances, I have not thought it right to involve those concerned in the trouble and delay that would result from restoring the case for further argument. The point and the leading authority on it were fairly before counsel, and the other authorities seem to me in the main merely to support and amplify that authority. Indeed, Granville v Betts, Stocker v Wedderburn and Wilson v Wilson were all cited in Wilson’s case. Fortescue’s case, it is true, exemplifies a different aspect, but even if in some way it could be shown to be nihil and rem, that would not destroy either the authority of Wilson’s case or the principle. Fortescue’s case, I may say, is duly cited in Fry, two pages on from the passage which counsel for the defendants cited to me. I am dealing explicitly with this point because, of course, this is a committal case, with all that that implies, and the citation in a judgment of authorities not discussed during argument is a matter that must be approached with caution. Accordingly, although I do not require further argument, if counsel on either side desire that there should be any further argument, I shall readily listen to any application for the purpose.

The process of adjudication is characterized by the right of the parties to present proofs and reasoned arguments. This right presupposes that, inter alia, the standards of reasoned argument are known and that the judge will not consider points on which counsel have not had an opportunity to present either proofs (of facts) or to make representations on the law.

Megarry J. obviously takes the need to protect this right very seriously. I wonder how many Canadian judges would be so careful.

Perhaps the larger question is whether judges today should be so careful and, if they are not, what the consequences are.


  1. I’m not aware of any Canadian comment, and it’s not mentioned in the Canadian Judicial Council’s advice to judges. But across the border, Judge Frank Easterbrook commented on this issue, and tied it back to contract theory in Frank H. Easterbrook, Afterword: On Being a Commercial Court, 65 CHI.-KENT L. REV. 877, 880 (1989).

    [He deplored] judges who seize the moment to write essays about issues the parties did not present. Just as parties may choose the terms of their contract, they may choose the subjects of their litigation. Resolving a case on a ground not presented denies the parties this autonomy and increases the risk that an uninformed opinion will impede rather than promote commerce. It is hard enough to navigate when the court sticks to questions fully ventilated by counsel.

    The other side was put most strongly by Judge Tate of the Louisiana Court of Appeal, Third Circuit, in an article called SUA SPONTE CONSIDERATION ON APPEAL 9 Trial Judges J. 68 (1970)

    “Any theoretical unfairness to the litigant in deciding on the basis of authorities not cited is outweighed by the court’s general duty in the administration of justice. If an appellate court must decide cases in accordance with law, it must also decide them as quickly and efficiently as possible. Any delay in decision to allow for reargument and restudy will defer decision for the present litigants and cause additional counsel expense. It will not produce any different result, for a court will not sua sponte decide on the basis of uncited authorities without having studied them thoroughly; the very reason for diligently going beyond the briefs is that the court has concluded that independent and more laborious research on its part is required.”

  2. There is certainly law that says a trial judge isn’t allowed to decide a case in favour of the plainitiff on a theory not argued by the plaintiff and not in evidence ant trial, even if the trial judge raised it after the evidence so (arguably) gave the defence a chance to make submissions. There’s been a number of those cases sent back for a new trial, in this province, over the past two decades.

    The Shekhdar case Angela mentioned in the “what is a judgment thread” is an example, I suppose, of a case decide in the defendant’s favour on a point raised by the judge on his own volition. The only case that I recall, of hand, of an action being dismissed on an point not argued at all but appearing only in the judges reasons is the (ahem) classic Joly v Pelletier & Shoppers Drug Mart case in which the motion judge dismissed the action on a “standing” argument the moving defendants didn’t raise. For those who don’t know the case, the “standing” argument was that the plaintiff didn’t qualify as a plaintiff under Ontario rules, therefore didn’t have standing to sue, since (according to the judge) he claimed he was a “Martian” and not a human being. Actually, he didn’t claim he wasn’t human, but we won’t go there.

    Then there’s always the cases where the lawyers don’t help the judge enough, or at all: “Since neither counsel cited the binding decisions of the Court of Appeal (or indeed, any authority at all), neither party is entitled to costs of this motion.” Leave it that there were a raft of authorities. (No, I wan’t involved in this one.)

  3. I think that the proper response to the issue raised by Megarry J. requires one to consider the level of the court and, perhaps, the kind of issue being raised. A trial judge in, say, an “ordinary” breach of contract case, should adopt (and should be required to adopt) Megarry’s point of view.

    What about the Court of Appeal and Supreme Court in a constitutional matter? The problem of balancing fairness to the parties (on the basis that I set out in my post) with the proper development of the law is not easy and needs to be considered and discussed. Apart from the references found by Simon, is anyone aware of any Canadian discussion? Anecdotal evidence suggests that the Supreme Court will decide cases on grounds not raised by counsel. I am not aware that that court has discussed the problems this practice creates.

  4. Not as lengthy as Megarry J. but in Carling Developments Inc. v. Aurora River Tower Inc., 2005 ABCA 267, O’Brien J.A.’s concurring reasons included the following:

    I do not think it necessary for the disposition of this appeal to make the analysis of the true
    nature of trust conditions or otherwise to deal with the issues as set out in the judgment of my brother Côté, J.A. I am particularly reluctant to do so in the absence of argument by counsel with respect to some of the specific points developed in his judgment.

  5. As Legal Research Counsel to the Manitoba Court of Appeal, this is obviously an issue very near and dear to my heart. While courts are understandably reluctant to decide cases on grounds other than those raised by counsel, I think they are more likely to refer to additional case law on subjects raised (but perhaps not fully researched) by counsel. For example, see the comments of the majority of the Alberta Court of Appeal in R. v. Rhyason, 2006 ABCA 367 at para. 28:

    The parties had submitted written argument and had every opportunity to cite the law they thought was relevant. A trial judge cannot be faulted for relying on other relevant case law. Nor is he obliged to give the parties a chance to comment on such cases, particularly where they are germane to the issues. The appellant has submitted no authority to suggest otherwise.

    In my experience, if a court wishes to make a decision on a point not raised by counsel, they will request further submissions on that subject. However, if counsel have correctly identified the issue but failed to research the law thoroughly, they will not be given an opportunity to comment on cases that were available to them at the time their arguments were prepared. And, in my view, that is how it should be.

  6. I’ve been told by a Canadian appellate judge that the situation is as Melanie described it, and that if counsel just fail to put their finger on the real issue at stake, the court may well just fill the gap.

    The distinction is in cases where the court sua sponte perceives a constitutional point, where the court would invariably write to counsel to request submissions.

  7. In the context of an appeal from a criminal sentence, Doherty J.A. discussed the issue and concluded that by introducing material not in evidence and not put forward by the parties at the sentencing hearing the trial judge had “assumed the combined role of advocate, witness, and judge” R. v. Hamilton (2004), 72 O.R. (3d) 1 (C.A.) at para. 65.

    Doherty J.A. identified four potential problems:

    “First, by assuming the multi-faceted role of advocate, witness, and judge, the trial judge put the appearance of impartiality at risk, if not actually compromising that appearance” (para. 71)

    “The second problem arising from the trial judge’s approach is that it produced a fundamental disconnect between the case on sentencing presented by counsel … and the case … constructed by the trial judge.” (para. 73)

    “A third problem with the trial judge’s conduct of the proceedings is that it created a real risk of inaccurate fact-finding.” (para. 75)

    The fourth problem identified by Doherty J.A. was that one particular inquiry independently introduced by the trial judge “consumed a good deal of time and effort”, played little or no role in determining the appropriate sentence and “contributed to the impression that the trial judge had decided to conduct an inquiry into matters that concerned him rather than conduct a sentencing hearing to determine the appropriate sentence” (paras. 82 and 83).

    There are further comments in the case about the judge’s role as “neutral, passive arbiter” (para. 67) and the basically adversarial nature of the sentencing phase of a criminal proceeding (ibid.). Although the case has been cited by a civil court just once to date (Nicolazzo v. Princess Cruises (2009), 250 O.A.C. 4 (Div. Ct.) at para. 9), there would appear to be little reason to restrict its reasoning to criminal sentencing proceedings.

  8. On the issue of judicial notice versus judicial knowledge, one may refer to the comments of Lamer J. (as he then was) in R. v. Sioui, [1990] 1 S.C.R. 1025 at p. 1050:

    I am of the view that all the documents to which I will refer, whether my attention was drawn to them by the intervener or as a result of my personal research, are documents of a historical nature which I am entitled to rely on pursuant to the concept of judicial knowledge.

  9. I’d have thought that the realization that most reasons for judgment get reported, or at leasgt that there’s a very real risk of the reasons getting reported, make lawyers, at least, concerned about not giving the judge case law on point that’s easily enough found and is relevant. Some of use, here, know how often it is that the judge doesn’t get the case law.

    I’ll give you a recent example. The judge wrote: “Since the parties were not able to point me to any case which had considered the issue of the existence of a duty of care by a volunteer in circumstances where the volunteer’s act allegedly had caused, or contributed to, damage to the property of the person to whom he was volunteering his services, I conclude that the question of law raised by this case is a novel one, or at least one in which the case law has not been fully settled.”

    There’s no indication in the reasons whether anybody consulted the index of a tort text to see if there was anything there. Even if there wasn’t, those who remember their first year tort classes will recall the rescuer & duty of care issues.

    Apart from that, less than 5 minutes on CANLII produces this excerpt from a BCCA decision, quoting from a BCSC case.

    It is clear that where one person undertakes an activity the law does not ordinarily impose on another the duty to assist him in that regard; however, where a promise is given by one person that he will assist, performance of that promise is embarked on, and the other person relies on that performance, liability may lie: Coggs v. Bernard (1703). [At para. 6.]

    The author of that quotation? McLachlin J (as she then was)

    1703, you say? Hmmm….

    Oh, well.



  10. That BCCA case, itself, is a 2007 decision which is easily found. It’s Goodwin v. Mainroad North Island Contracting Ltd., 2007 BCCA 81, that’s also reported in the leading paper reports. It includes the following gem: [26] … there is a long and respectable line of authorities which have imposed liability where an actor, particularly one with public duties, has voluntarily undertaken to provide some kind of assistance to another, but does so negligently and causes foreseeable harm to that other person. The authorities go back as far as Coggs v. Bernard (1703) 2 Ld. Raym. 909, 92 E.R. 107 (K.B.)”

    A CANLII search for case law using – negligence volunteer /p “duty of care” – produces 19 hits. I din’t check them all. Goodwin is the 4th listed chronologically.

    Plus ca change …


  11. Interesting turn this discussion has taken. I recall (vaguely) from law school the rescuer cases – no duty to rescue (unlike in the civil law) but duty not to be negligent if one undertakes it. Horsley and McLaren was brand new back then, though as David C points out, the principle was not. But in general there are very few if any Canadian cases where a volunteer has been held liable for negligence in his/her voluntary activities, outside the domain of rescue.

    I note in McLaughlin’s quote above, “an actor, particularly one with public duties…”. Private volunteers are just about never liable. Despite this, the insurance companies jack up their rates atrociously, leading to requests to government for statutory immunity from liability. The insurance companies seem to be just charging what the market can (barely) bear, or profiting from US horror stories (a lot of that going around), or they just don’t care enough about the market to evaluate the risks, so they charge a nuisance premium.

    To return to the original theme, I recall statutory amendments made in Ontario that were intended to reverse appellate decisions that were totally unanticipated by any of the parties who submitted argument. I recall doing the amendments, but I don’t offhand recall which ones ….

    It seems to me that the ratio decidendi of the Eurig case in the Supreme Court in about 1998 came out of the clear blue sky: that a fee that recovers more than the cost of providing the service for which it is charged is a tax, and thus must satisfy the legal requirements of taxation, namely that it be authorized by legislation and for provincial purposes, that it be a direct tax. I don’t think that the revenue = cost principle had ever been stated in law before or that counsel argued it. (No doubt there are some readers who can correct me if required!)

  12. John,

    I might have created one of those situations for you, where you had to draft the amendments, but the situation wasn’t “totally unanticipated”. I was involved in a case that resulted in the amendment in the early 1980s to the Personal Property Security Act that added the specific provision that, for the most part, the act does not apply to interests in real property such as leases or mortgages. (For the non-lawyer and civil law readers, the amendment was necessary because leases and mortgages, although interests in land, are personal property not real property.)

    One of the good? consequences of judge’s frolics of the sort we’re talking about is that it provides academics, academic-manqués, and others, more grist for their writing mills.

    For example, I’ve spilled a good quantity of ink (real and electronic), published and unpublished, since the mid 1990s complaining about a pronouncement on a point of law that the Ont CA made that had not a thing to do with the issue it had to decided. The Court knew it. They decided the issue they had to that resolved the appeal, then went on to say a few words about something (borrowing from Monty Python) completely different which was never an issue in the appeal.

    The ONCA recently interred the obiter, after a few cases in which it whittled away at it – using distinctions that were factually correct but strained principle – because there would have been obvious injustice if the obiter applied. When they interred the obiter, the ONCA said both that it had been “overtaken” by the subsequent decisions and they thought it wrong. (They didn’t cite my articles. Should have, though.)

    I’ve not spilled quite as much real ink (online is different) about the recent SCC decision in Resurfice v. Hanke which has one of those “out of the blue” pronouncements of law that had nothing to do with the disposition of the issues in case: the SCC’s declaration that fault and increased risk may sometimes be enough to satisfy tort’s causation requirement, even though the injured person cannot establish, on the balance of probability, that the fault was a (factual) cause of the injury.