The answer to the above question is obviously not that at some point in interpreting a will the presiding judge may sit in easy chair, put up her feet, and ‘veg out’. The sentence, ‘Sit in the testator’s armchair’, is not an invitation to relax. It is not cast in the indicative mood, but is hortatory in nature and therefore the sentence is cast in the imperative mood. In other words, it exhorts the judge to get to work.
The direction to sit in the testator’s armchair addresses the kind of evidence the judge may consider in interpreting the will. It refers specifically to her ability, indeed, her obligation to consider extrinsic evidence, that is evidence of the testator’s circumstances at the time he made his will.
But the question remains, when should she sit in the testator’s armchair and consider such evidence? In the past the law required that the court first read the will and determine what the language used meant. If that process led to a sensible meaning, that ended the interpretation, even if it was obvious that the testator intended a different result. Only if the initial process did not lead to a sensible conclusion was the court permitted to consider surrounding circumstances. A simple example will illustrate the problem. Suppose a testator makes a will which says only, ‘All for mother’. When he made his will, his mother was alive, but extrinsic evidence shows that he customarily referred to his wife, the mother of their children, as ‘mother’. Under the strict construction approach the testator’s mother would take ‘all’ and his wife would receive nothing. This approach is referred to as the 19th century, strict or objective approach to construction and it led to all kinds of erroneous interpretations of wills. Lord Denning, among others, fulminated against this approach and said in dissent in one case:
It seems to me that the fallacy in that argument is that it starts from the wrong place. It proceeds on the assumption that, in construing a will, ‘It is not what the testator meant, but what is the meaning of his words’. That may have been the nineteenth-century view; but I believe it to be wrong and to have been the cause of many mistakes. I have myself known a judge to say: ‘I believe this to be contrary to the true intention of the testator but nevertheless it is the result of the words he has used’. When a judge goes so far as to say that, the chances are that he has misconstrued the will.
He also said that what you (i.e., the judge) should do is, ‘place yourself as far as possible in [the testator’s] position, taking note of the facts and circumstances known to him at the time; and then say what he meant by his words’. He also mentioned that a Chancery judge was once said to have remarked that a group of dissatisfied testators, whose souls the ferryman Charon had carried across the river Styx, were waiting on the other side of the river to receive the judicial personages who misconstrued their wills, presumably to take out their wrath upon them.
Fortunately, we no longer have to put up with that strict construction approach. Today we follow the more liberal, subjective approach to construction. It allows the court to admit evidence of surrounding circumstances immediately when it begins to consider the meaning a will. In other words, the court does not wait to sit in the testator’s armchair until after it discovers that it cannot read the will sensibly without it. Still, there are some traces of the old approach in the cases, which sometimes suggest that the court should first read the disputed passage in a will to try to ascertain its meaning before it considers evidence of surrounding circumstances. Although I think that the Ontario Court of Appeal reached the correct conclusion in the recent case, Ross v. Canada Trust Company, it believe that the reasons contain hints of the old practice. I think that is regrettable and inconsistent with modern cases.
For example, in Haidl v. Sacher, the Saskatchewan Court of Appeal took the position that the court should sit in the testator’s armchair immediately, as Lord Denning suggested. So did Laidlaw J.A. in Re Burke, when he said:
Each Judge must endeavour to place himself in the position of the testator at the time when the will was made. He should concentrate his thoughts on the circumstances which then existed and which might reasonably be expected to influence the testator in the disposition of his property. He must give due weight to those circumstances in so far as they bear on the intention of the testator. He should then study the whole contents of the will and, after full consideration of all the provisions and language used therein, try to find what intention was in the mind of the testator. When an opinion has been formed as to that intention, the Court should strive to give effect to it and should do so unless there is some rule or principle of law that prohibits it from doing so.
Similarly, in Hicklin Estate v. Hicklin, the Alberta Court of Appeal stated:
67 When should a court admit and rely on extrinsic material?
68 Some judges believe that a court may not admit and review extrinsic evidence unless they have concluded that the text supports more than one plausible meaning or introduces an ambiguity.
. . .
75 Others are unwilling to impose any onerous preconditions to the admissibility of extrinsic evidence. They rely on extrinsic evidence if it assist in ascertaining the meaning a testator attached to a word used in the will. They understand that context … may influence the meaning a testator attributes to a word:
A second school of thought is willing to explore extraneous material without demanding that an initial assessment of the clarity of the words of the will be undertaken. It encourages a court to review proferred [sic] extrinsic evidence….
Supporters of this school believe that the meaning of words a testator has used may not be accurately divined without a grasp of the context in which they were expressed and an understanding that the same words may bear different meanings in different contexts.
76 The latter approach is correct – extrinsic evidence … is admissible. Its admissibility is not dependent on a finding that a word is capable of more than one meaning or that an ambiguity exists. This approach is theoretically sound and inherently practical….
77 It is theoretically sound because it promotes an accurate assessment of the testator’s wishes:
It is the court’s role to give effect to the testator’s intention. This is an indispensable function the exercise of which perfects the transferal process the testator commenced when she signed her will.
To be faithful to the testator’s will, a court must identify the meaning the testator wished to convey by her choice of words. This can only be done, in many cases, if the court has access to relevant evidence which records information, in existence at the time the testator signed her will, about the testator’s family and the nature of various family relationships, close friends, interests and many other facts which might influence the testator when engaged in the will-making process. A court, aware of important information about the testator, must carefully read the entire will, giving the words she selected or approved their ordinary meaning.
78 The correct approach is practical because the admissibility standard it introduces is not onerous and there will be few contests as to whether proferred [sic] extrinsic evidence is admissible.
79 To repeat, this approach allows “[p]arties who advance a claim to property the testator disposes under her will and others with a legitimate interest in ensuring that the testator’s intentions are honoured” to present evidence that they think will assist the adjudicator to discharge his or her judicial task.
Thus, modern case law endorses the view that to interpret a will properly, you have to sit in the testator’s armchair at the outset and read the entire will and the provisions that cause difficulty in light of the surrounding circumstances.
The Court of Appeal does say as much in Ross, for in paragraph 37 it quotes the above passage from Re Burke, which adopts this view, and see also paragraph 41. However, the quotations from Feeney’s Canadian Law of Wills, referred to in paragraphs 38 and 39, contain the suggestion that the first step is still to read the provisions that are in doubt to see if the court can find a clear meaning from them and should only resort to the armchair rule if it can’t. With respect, that is a relic of the old strict construction approach and should be abandoned, because it can lead to interpretations the testator did not intend, as in the Thorn case referred to above.
With that in mind, let us look at the facts of the case. The testator, Sarah Grafton, had two daughters, Margaret and Mary. Mary never married, but Margaret did and had five children: Jane, Gordon, Graeme, Grafton, and James. Sarah died in 1971, Jane died in 2015, Mary died in 2002, and Margaret died in 2015. Jane died intestate and her estate passed to her parents and thence to Margaret when her husband died. Margaret left her estate to Gordon.
Clause 3(C) of Sarah’s will left her cottage for the use and enjoyment of her daughters for the life of Mary. After the death of the survivor of Margaret and Mary. the executors were to transfer the cottage to my grandchildren as joint tenants, but if they did not agree unanimously to accept the transfer, the executors had to sell the property to strangers and to hold the residue in trust for my said grandchildren in equal shares, to be distributed to them in accordance with the residue provision in Clause 3(I)(iii). The executors also had power to sell the real property if they should think it advisable, to invest the proceeds, and pay the annual income equally to Mary and Margaret, and on the death of the survivor to distribute the residue of the proceeds among the said grandchildren in equal shares in accordance with the residue provision in Clause 3(I)(iii). (Emphasis supplied.)
Clause 3(I)(iii) provided that the residue should be divided into six equal parts and to divide one such equal part equally among the issue of Margaret, living at my death. (Emphasis supplied.)
Gordon argued that the proceeds of sale of the cottage should be divided into five equal shares, because they had to be distributed in accordance with Clause 3(I)(iii) of the will. Under that clause Jane was included, since she was living at Sarah’s death. And since Gordon had inherited Jane’s share via his mother, Gordon claimed to be entitled to two of the five shares. The other grandchildren argued that the proceeds of sale of the cottage should be divided into four equal shares and divided among the four surviving grandchildren, in accordance with Clause 3(C).
The motion judge, Broad J., concluded that the references to Clause 3(I(iii) in Clause 3(C) caused interpretational difficulties that did not permit him to ascertain Sarah’s intention solely from the plain meaning of the langue used in the Will (see paragraph 25, clause (vii) of the Court of Appeal’s summary of the reasons of Broad J.). Broad J. then applied the armchair rule. Based on the evidence of the surrounding circumstances and the language of Clause 3(C), in particular the italicised passages, all of which suggested that Sarah wanted to keep the cottage in the family, Broad J. concluded that Sarah did not intend that the beneficiaries of a deceased grandchild could take an interest in the cottage or its sale proceeds (para. 27). Thus, the procedure followed by Broad J., seems to have been first to determine if the provisions of the will were clear, and only when he found that they were not did he turn to the armchair rule.
The Court of Appeal reviewed and endorsed the interpretation and methodology of Broad J., including the fact that the motion judge first determined that the provisions of the will were not clear, but created a patent inconsistency, and only then resorted to the armchair rule (paras. 49-51).
For the reasons stated above, and with great respect, I believe that this is incorrect and that the court should interpret the will from the outset with the aid of extrinsic evidence. But in other respects I fully agree with the reasons for judgment.
The case raised three other issues that should be mentioned. The first was the standard of review for the interpretation of a will by a lower court. The court held, rightly, in my opinion, that, absent an error of law, today the standard is no longer correctness, but a deferential standard, citing Trezzi v. Trezzi, Zindler v. The Salvation Army, and Hicklin Estate v. Hicklin.
The second issue arose out of the respondents’ cross-appeal. They agreed with the judgment of the motion judge, but disagreed with how he reached his conclusion. As the court rightly noted, an appeal to the Court of Appeal lies only from a ‘final order’ of a judge of the Superior Court of Justice.
You cannot appeal a judge’s reasons for judgment. Thus the cross-appeal was misconceived.
The third issue was whether the motion judge failed to apply the presumption of early vesting, as Gordon argued. While the court agreed that there is such a presumption, it is only a presumption and it yields to a contrary intention in the will. It should be applied only if the court has doubts about the testator’s intention (para. 58). Since the motion judge had concluded that, properly interpreted with the aid of the armchair rule, the testator did not intend that all grandchildren share in the proceeds of sale of the cottage, only those living at the time of the sale were entitled to share in it. Thus the interests of the five grandchildren were not vested a morte testoris (i.e., at the death of the testator).
 The example is taken from Thorn v. Dickens,  W.N. 54. In fact, the testator’s mother predeceased him, so the will created an ambiguity and that let in the extrinsic evidence
 You will find a number of examples of such egregious interpretations in Oosterhoff on Wills, 9th ed. by Albert H. Oosterhoff, C. David Freedman, Mitchell McInnes, and Adam Parachin (Toronto: Thomson Reuters / Carswell, 2021), §13.6.2(a).
 Re Rowland,  1 Ch. 1 at 9-10 (C.A.).
 Ibid., p. 10.
 Ibid. In Perrin v. Morgan,  A.C. 399 (H.L.) at 415, Lord Atkin made the same comment.
 2021 ONCA 161.
 1979 CarswellSask 131, 7 E.T.R. 1 (C.A.).
 1959 CarwellOnt 98,  O.R. 26.
 Ibid., para. 5, emphasis supplied.
 2019 ABCA 136, 46 E.T.R. (4th) 1.
 Tottrup v. Patterson (1969),  S.C.R. 318 (S.C.C.), 322 (”if the meaning is clear, surrounding circumstances cannot be looked at to throw a doubt upon that meaning:); Smith v. Home of the Friendless,  S.C.R. 713 (S.C.C.), 719 (”where [the testator’s language] … is ambiguous, we are entitled to consider not only the provisions of the will, but also the circumstances surrounding and known to the testator at the time when he made the will”) ….
 Hanson v. Mercredi, 2014 ABCA 216 (Alta. C.A.), ¶ 52;  10 W.W.R. 41 (Alta. C.A.), 62 per Wakeling, J.A. (”To determine whether the testator intended to create a trust or a power of appointment … the Court must identify the meaning the testator attached to the words of the will, taking into account any other relevant evidence which may assist the Court to ascertain the testator’s intention”).
 Hanson v. Mercredi, 2014 ABCA 216 (Alta. C.A.), ¶¶ 9, 59 & 60;  10 W.W.R. 41 (Alta. C.A.), 70-71 per Wakeling, J.A. See also … Marks v. Marks (1908), 40 S.C.R. 210 (S.C.C.), 212 & 220 per Idington J. (”we are bound to read his language in light of all the circumstances that surrounded, and were known to him when he used it and gave effect to the intention it discloses when so read” & “I prefer to read the ordinary meaning … of the words used … in light of surrounding circumstances in accordance with common sense”) ….
 Hanson v. Mercredi, 2014 ABCA 216 (Alta. C.A.), ¶¶ 8 & 9;  10 W.W.R. 41 (Alta. C.A.), 50 per Wakeling, J.A. See also Re Burke (1959),  O.R. 26 (Ont. C.A.), 30 ….
 Hanson v. Mercredi, 2014 ABCA 216 (Alta. C.A.), ¶ 11;  10 W.W.R. 41 (Alta. C.A.), 51 per Wakeling, J.A.
 4th ed, by Ian Hull and Suzana Popovic-Montag (Toronto: LexisNexis, 2020)
 In the text at footnote 1, supra.
 2019 ONCA 978, para. 15.
 2015 MBCA 33, para. 10.
 Footnote 10, supra, paras. 10 and 94-95
 The court cited, inter alia, R. v. Sheppard, 2002 SCC 26, para 4, as authority for the proposition.
This paper is intended for the purposes of providing information only and is to be used only for the purposes of guidance. This paper is not intended to be relied upon as the giving of legal advice and does not purport to be exhaustive.