Sumpton v Downing [1947] HCA 36; (1947) 75 CLR 76 (6 October 1947)

Sumpton and Others Defendants, Appellants; and Downing and Others Plaintiffs and Defendants, Respondents.

F. Downing, for the respondents H. P. and I. B. Downing, surviving trustees of the will of Thomas Statham.

Louch K.C. (with him Negus), for the respondents other than the surviving trustees of the will of Thomas Statham.

This is an appeal from an order of the Supreme Court of Western Australia (Wolff J.) made upon an originating summons submitting for the opinion of the court questions relating to the construction of the will of the late Thomas Statham, who died on 13th February 1918. By his will he gave his house and household furniture &c. to his wife. He left certain legacies. He had two married sisters, Mary Blood and Elizabeth Coneybeare. He provided that his trustees should invest £2,000 and pay the income to Mary Blood during her life and after her death pay and divide the said sum of £2,000 between such of her children as should be living at the time of her death with a substitutionary gift in the case of children who died before their mother leaving children living at her death. Mary Blood had children when the will was made. In the case of the other sister, Elizabeth Coneybeare, the trustees were also directed to set aside a sum of £2,000 and pay her the income and after her death to pay and divide the income equally between such of her children as should be living at the time of her death. In this case, however, there was no substitutionary gift to grandchildren, but it was provided that if Mrs. Coneybeare died without leaving issue £500 of the money was to be paid to her husband and the balance to the children of Mary Blood or their children if any of them (Mary Blood's children) died in the lifetime of Mrs. Coneybeare leaving children. Mrs. Coneybeare was aged forty-three in 1910 when she married. Thus the testator made a provision for the case of Mrs. Coneybeare dying without leaving issue, but made no such provision in the case of Mrs. Blood.

Under the will of the testator there was a residuary gift to the testator's wife.

By a codicil the gifts of the house and furniture &c. and of the residue to the widow were revoked, but she was given permission to hold the house and to use the furniture during widowhood. There was a direction that the income of the residue should be paid to his wife during widowhood. If she remarried she was to receive £5 per week. Upon the death or remarriage of the widow the trustees were to realize the house, furniture &c., and to invest the proceeds. The will proceeded: "I direct my trustees to pay the income derived therefrom (subject to the payment of the said sum of Five pounds per week to my said wife during her life in the event of her remarrying) to my sisters Mary Blood and Elizabeth Conybere in equal shares during their lives and subject thereto I direct my said Trustees to hold the moneys aforesaid upon trust for the children of the said Mary Blood and Elizabeth Conybere as shall be living at their death in equal shares absolutely." This is followed by a substitutionary gift to take effect if any of the children of Mary Blood or Elizabeth Coneybeare should die in the lifetime of their parent, leaving a child or children living at the death of "my sister."

The testator's widow did not remarry, and therefore she was entitled to all the income of the residuary estate until she died on 1st March 1942.

The testator's sister Elizabeth Coneybeare died without issue on 10th November 1932. The defendant C. E. Coneybeare is her personal representative. The testator's sister Mary Blood died on 15th October 1942, leaving surviving her two sons and three daughters. Another son of Mary Blood died in her lifetime and two of his children survived Mary Blood.

The questions which arise relate to the gift of income to the testator's sisters and to the gift of the corpus to the children of his sisters. The income is given to the sisters in equal shares during their lives. It was held by the learned judge that this provision gave an equal share to each sister during her life, and that when Elizabeth Coneybeare died her share of income fell into residue and was dealt with by the ultimate gift to the children of Mary and Elizabeth. As to the corpus of the residue, the gift is to the children of Mary Blood and Elizabeth Coneybeare as shall be living on their death in equal shares absolutely with a substitutionary gift to grandchildren. It was held that this was a gift to such children of Mary and Elizabeth as should be living at the death of the survivor of Mary and Elizabeth, the children to take in equal shares per capita, the grandchildren taking their parents' share. In the events which have happened the result of this interpretation of the will is that such of Mary's children and grandchildren as may survive Mary will receive the whole of the corpus.

It is contended for the appellants that the gift of income to the sisters in equal shares during their lives is a gift of one-half of the income to each of them, and that when Mrs. Coneybeare died there was no disposition of her share. Similarly, it is argued that the gift of the residue is a gift of one-half to the children of Mrs. Blood and another half to the children of Mrs. Coneybeare, so that there is an intestacy as to the half which any children of Mrs. Coneybeare, if living at the death of their mother, would have taken.

This argument in effect inserts the word "respective" in certain places in the provision which I have quoted. I requote it, inserting the word "respective" in the manner for which the appellant contends: "... to pay the income ... derived therefrom ... to my sisters Mary Blood and Elizabeth Conybere in equal shares during their" (respective) "lives and subject thereto I direct my trustees to hold the moneys aforesaid upon trust for the" (respective) "children of the said Mary Blood and Elizabeth Conybere as shall be living at their" (respective) "death in equal shares absolutely." This is a construction which has been adopted in a number of cases: see In re Hutchinson's Trusts[1]; In re Errington; Gibbs v. Lassam[2].

I consider first the gift of the corpus to the children of Mary Blood and Elizabeth Coneybeare. In the first place, this gift is given "subject thereto," that is, subject to the payment of £5 per week to the testator's wife and of income to both or one of the sisters. This provision suggests a retention of the whole fund and a disposition of the fund upon a single event, that is, the death of the survivor of the sisters, rather than a separate devolution of separate halves, each devolving upon a different event, namely, first, the death of the sister who dies first, and, next, the death of the surviving sister. Next, the final gift is a gift of "the moneys aforesaid." This, again, is a provision which deals with the fund as an entirety, and not as consisting of two separate halves. Further, the gift is to a composite class consisting of the children of Mary and Elizabeth. It was unlikely that Elizabeth would have children, as appears from the contrasted provisions in the will relating to the two sums of £2,000 to which reference has already been made. The fact that the gift is a residuary gift, when combined with the circumstances already mentioned and the presumption against intestacy, supports the decision of the learned judge that in the events which have happened the children and grandchildren of Mary living at her death are entitled to the whole of the corpus of the residue, the grandchildren to have their parents' share.

The gift of income to the sisters Mary Blood and Elizabeth Coneybeare is a gift to each of them during the life of each of them. The words "subject thereto" define the residue as consisting of the proceeds of realization ("the moneys aforesaid") less what is required to meet the payments of £5 per week to the testator's wife and of the balance of the income to his sisters. Moneys not required for those purposes are dealt with by the words "and subject thereto I direct my said trustees to hold the moneys aforesaid." Thus, when Mrs. Coneybeare died the income which she would have received if living became subject to the last-mentioned disposition. Accordingly, the income of half only of the residuary estate was payable to Mrs. Blood between 1st March 1942, the date of the death of the testator's widow, and 15th October 1942, the date of Mrs. Blood's death.

My brother Dixon has expounded the authorities dealing with this subject, and I agree with his reasoning and his conclusions.

I agree with the conclusion arrived at by Wolff J. with regard to the main question whether there was an intestacy in the will and codicil in default of issue of Elizabeth Coneybeare. And I also think that the appeal on the second question fails. In my opinion the codicil does not warrant an implication of a cross-executory limitation. In this appeal, also, I find myself in agreement with the reasons of my brother Dixon J.

This appeal concerns the construction and effect of some dispositions contained in the codicil to the will of Thomas Statham deceased. The testator died on 13th February 1918. He had made his will on 25th September 1917 and the codicil on 28th November 1917. The codicil revokes a specific devise and bequest to his wife and an absolute bequest to her of residue. A comparison of the will and the codicil suggests that the chief reason for making the codicil was to substitute for the absolute dispositions in favour of the testator's wife life interests which would determine upon her remarriage and to provide, in that contingency, that she should receive an annuity of £5 a week. The substitution of a life interest determinable upon the widow's remarriage meant that the testator must create limitations of the property expectant upon the widow's defeasible life interest. The codicil in fact contains limitations to take effect on the widow's death or remarriage and it is upon these that the questions we have to decide arise. The limitations are in favour of two sisters of the testator and their children. Both sisters were married. Their names were Mary Blood and Elizabeth Coneybeare. Elizabeth Coneybeare married at the age of forty-three, just seven years before the will was made. She had no children and died without issue on 10th November 1932. Mary Blood had children. She died on 15th October 1942 leaving her surviving five children. One of her sons died before she died. But he left two sons who survived their grandmother Mary Blood. The testator's widow did not remarry. She died on 1st March 1942. These are the circumstances in which the residuary bequest must be applied.

The bequest begins by a direction to the trustees, upon the death or remarriage of the widow, which ever event shall first happen, to sell the house she occupied and the household effects and to invest the proceeds. The clause goes on to direct that the income therefrom and the income of the rest, residue and remainder of the money arising from the sale and conversion of the estate shall be paid (subject to the payment of £5 per week to the testator's widow during her life in the event of her remarriage) to the testator's sisters Mary Blood and Elizabeth Conybere (so spelt in the codicil) in equal shares during their lives and, subject thereto, the testator directed his trustees to hold the moneys aforesaid upon trust for the children of Mary Blood and Elizabeth Conybere as (sic) should be living at their death (sic) absolutely. Then followed a substitutional gift to a child or children of any deceased child. It will become necessary further to refer to this clause.

The questions which have arisen upon the residuary gift are two. The first, which is the substantial question, is whether the five children of Mary Blood and the two children of her deceased son take the whole residue, there being no children of Elizabeth Coneybeare, or they take half only, the other half being distributable as on an intestacy for want of a gift over in default of issue of Elizabeth Coneybeare.

In the Supreme Court Wolff J. decided that the whole of the balance of the residuary estate is held in trust for the children of Mary Blood and the children of the son who predeceased her.

The second question, which is of little importance to the parties, is whether all or only half the income of residue accruing in the seven and a half months between the death of the widow and the death of Mary Blood was payable to Mary Blood and, if half only, to whom was the other half payable. Wolff J. held that half only of the income in that interval was payable to Mary Blood and the other half passed under the final gift of residue.

With respect to the first question I agree in the conclusion reached by Wolff J. The argument for the contrary conclusion, that which restricts the five children and two grandchildren of Mary Blood to a moiety of the residuary fund and leaves the other moiety undisposed of, is based upon an interpretation of the gift which, in effect, divides the residue between Mary Blood for life and Elizabeth Coneybeare for life with remainder to their respective children, if any, per stirpes.

In form, the limitation under which the children of Mary Blood claim is a gift, subject to precedent interests, to such children of two named persons as shall be living at their death(s) in equal shares. Prima facie, under a gift to the children of named persons as a class, the children take per capita and not per stirpes. It has been said that no man who was guided only by a knowledge of English speech would suppose that a direction to distribute money between the children of A and of B equally could mean anything but a division in which each child took a share equal with that of every other child, whether his parent was A or B. However this may be, it is enough that at least the prima-facie legal meaning of such a direction is that the distribution should be per capita. This is so whether in point of expression the class is described as the children of A and (of) B or as the children of A and the children of B.

"A bequest to the younger children of A and to the younger children of B means the same exactly as a bequest to the younger children of A and B": per Lord Eldon in Lady Lincoln v. Pelham[3]. "But this mode of construction will yield to a very faint glimpse of a different intention in the context": Jarman on Wills, 7th ed. (1930), p. 1688. The intention to the contrary has been discerned in gifts to several for life with remainder to their children when the form of the gift creates a tenancy in common in those taking for life and remainders expectant upon the determination severally of the interest of each tenant in common. Thus a gift to A, B and C for their lives and at their deaths to their children in equal shares is construed as a limitation to A, B and C for their respective lives as tenants in common with remainders severally expectant upon their respective deaths. It is easy to take the next step and say that the several remainders are to their respective children per stirpes and not to the children of all of them as a composite class taking per capita.

This method of construction is stated in the text books in forms that are not quite the same and are not uniformly definite: See Hawkins, 2nd ed. (1912), pp. 150-151; Theobald on Wills, 8th ed. (1927), p. 334; Jarman on Wills, 7th ed. (1930), p. 1690. It is well supported by authority, though as always in cases upon construction, every particular instrument construed, if it does not govern the decision, will at least afford grounds for explaining it. The principle was applied or illustrated in Arrow v. Mellish[4]; Willes v. Douglas[5]; Sutcliffe v. Howard[6]; and Wills v. Wills[7]. In Re Hutchinson's Trusts[8] many of the authorities were discussed by Kay J., who pointed out the necessity of reading "respective" more than once into such limitations, if it does not occur, in order to secure the result. His Lordship also placed some importance on the use of the word "at", or the word "after", or some like expression used in the same sense as "at", in limiting the remainders after the life interests. A gift to A, B and C for their lives as tenants in common and "at" their deaths remainder to their children suggests more strongly that a distinct future interest is expectant upon the death of each. An expression denoting no more than that the future interest takes effect at a time when the previous takers are dead is consistent with, if not indicative of, an intention that the subsequent estate or interest must await the death of all. Clear as is the distinction between the two intentions, the distinction between the expressions is perhaps an over-refinement. At bottom the principle must be that a division of the fund maintained or repeated through successive limitations is ground for inferring a final stirpital distribution and with other circumstances may justify the conclusion.

However, in Re Errington; Gibbs v. Lassam[9] Romer J. gave the following exposition of the mode of construction: "The rule, stated in its simplest way, is this: Where a testator gives the income of his estate to two people, A and B, for their lives and follows that gift by a direction that at their death, or at their deaths, or at or after the death or deaths of A and B the property is to go to their issue, the Court does not construe the gift as a gift only to take effect on the death of both in favour of the issue of both, but construes it as a gift, to take effect on the death of each, of the share to the income of which the deceased was entitled, to the issue of the deceased. So that in the simple cases to which I refer, on the death of A leaving issue, the issue of A would take one-half notwithstanding the fact that B still was living. On the death of B, B's issue in the same way would take the share in which B had a life interest"[10]. No doubt the same construction may be applicable though the limitations do not quite conform to the conditions laid down by Romer J. For in matters of interpretation rules must have a flexible application, one depending more upon their rationale than the recurrence of forms of expression. But in the present case I think that, if the substitutionary clause be left out of account, it is not easy to find in the limitation under consideration the substantial elements necessary to attract the operation of the rule. There are the prior life interests in income of Mary Blood and Elizabeth Coneybeare and a gift "subject thereto" which for all that appears is to await the death of both. The gift is of the moneys as an entirety and the donees are described as a composite class. There is no such attachment or connection between the prior and the subsequent gifts as Kay J. and Romer J. contemplate; nothing showing that distinct remainders are expectant on distinct particular estates or interests.

But I think it must be conceded that much may be found in the substitutionary clause to place upon the main gift meanings which bring the gift much closer to the operation of the rule, though in the end I think the rule does not control the meaning. The substitutionary clause is expressed very confusedly, and for clearness I shall set it out inserting explanatory identifications of the objects to which it refers. After the trust for (such of) the children of Mary Blood and Elizabeth Coneybeare as shall be living at their deaths in equal shares absolutely, it proceeds thus: But if any of the said children (of Mary and Elizabeth) shall have died in the life-time of their parent (that is, the lifetime of Mary or Elizabeth, as the case may be) leaving a child or children living at the death of my sister (Mary or Elizabeth, as the case may be) such child or children shall be entitled to the share, equally between them if more than one, which the deceased parent (that is, the deceased child of Mary or Elizabeth, as the case may be) would have been entitled to if living at my sister's death (that is, at the death of Mary or Elizabeth, as the case may be).

In so ill-drawn an instrument as this codicil there is much danger in drawing inferences from provisions such as the substitutionary clause, which bears every mark of a transmigration from some other document and of an imperfect adaptation to the environment in which it is now found. But the clause is framed essentially upon the assumption that a child of Elizabeth Coneybeare would take a vested interest if the child survived his mother, Elizabeth Coneybeare, and a child of Mary Blood would take a vested interest if the child survived his mother, Mary Blood. That means that no child of the testator's sisters need survive his aunt as well as his mother in order to obtain a vested interest. There is no sufficient reason for refusing to carry back into the gift this definite indication of its meaning. Whatever uneasiness courts may feel about agglutinative drafting, they must have better ground than mere misgiving before they reject the internal evidence of the instrument. Accordingly, I think that the appellants may properly say that, for the purpose of fixing some of the meaning of the disposition, we should rewrite the limitation as follows before we determine its effect: To my sisters Mary Blood and Elizabeth Coneybeare in equal shares during their respective lives and subject thereto upon trust for the children of Mary Blood who shall be living at her death and the children of Elizabeth Coneybeare who shall be living at her death in equal shares.

It must be confessed that so to rewrite the gift is to bring out a number of features which tend towards a stirpital distribution. The interest of each sister determines with her life. The class taking as children of hers who survive is ascertained at her death. The two classes of children are ascertained on separate events. During the lives of the first takers the interests in the fund are divided in a manner corresponding with the division which would result from a stirpital distribution.

But notwithstanding these features I think that a distribution per capita was intended and that the gift bears that interpretation. The words "subject thereto" do not suggest that the testator was limiting successive interests in distinct shares by way of particular estate and remainder or by way of any other limitation which would operate to pass a distinct undivided share or interest from one generation to another. These words indicate rather that the testator regarded the sisters' interests more as prior charges upon or as independent antecedent interests in his estate or the residue considered as a whole. The circumstance that the classes of children are ascertained on different events cannot be pressed far. It is consistent with their taking as a composite class and the language of the gift as distinguished from the deductions drawn from the substitutionary clause discloses no advertence to the distinction on which separate ascertainment depends. Finally, the language of the disposition literally means that the fund is to be kept together, "to hold the moneys aforesaid." There is a good deal of authority to warrant adherence in such a context to the prima-facie meaning, that is, division per capita.

Lord Alvanley in Malcolm v. Martin[11] had no doubt that a distribution of corpus per capita was required by a limitation to children of two named persons for life to be equally divided between them and at their decease the same to be divided betwixt the grandchildren of the persons, again naming them. In Smith v. Streatfield[12] Sir William Grant similarly construed a bequest of a sum of money in trust to pay half the interest to A and the other half to B during their lives and as their lives drop the principal and interest to be reserved and equally divided among their children when they severally become twenty-one. This notwithstanding that his Honour intimated at the hearing that it might be doubted whether the word "respectively" was not virtually in the clause and took time to consider. In Abrey v. Newman[13] the bequest was of property to be equally divided between four specified people, two married couples, for the term of their natural lives after which to be equally divided between the children, that is to say the children of the husbands, again naming them. Sir John Romilly decided that the children took per capita and on the deaths of one couple a half share became divisible between the children of both families. His Honour placed some importance on the naming of the parents. Theobald on Wills, 8th ed. (1927), p. 335 agrees in treating the naming of the parents in describing the class as a factor against stirpital division. "If the gift be to A and B for their lives, at their death, not to their children, but to the children of A and B, there seems less reason for contending that the children are to take per stirpes." The pronoun may be more readily understood as distributively referential, reddendo singulos singulis. To speak of the children of A and (of) B is merely to define a composite class.

In Re Stanley's Settlement; Maddocks v. Andrews[14] the limitation was to two named daughters for and during their natural lives as tenants in common and not as joint tenants and from and after the decease of the survivor for their respective child or children of the said daughters naming them, again share and share alike, as tenants in common and not as joint tenants. One daughter died childless. Sargant J. held that the children took per capita and the entire interest passed to the children of the other daughter. It is to be noticed that although the limitation of the prior estate was until the death of the survivor of the two daughters the word "respective" was used. Of this Sargant J. said: "There is nothing to point to the intention that one moiety is to go only to the children of the tenant for life of that moiety except the word respective, and that is not enough of itself to have the required effect"[15].

There is, in my opinion, no sufficient reason for departing from the prima-facie meaning of the gift which is that the division of the entire fund should be per capita among such of the children of the two daughters as survived their mother. But the conclusion that this represents the intention of the testator is much strengthened by the fact that when he made his will and codicil he could not have expected Elizabeth Coneybeare to have issue. It is a residuary gift and the presumption against intestacy is strong.

For these reasons I think that upon the main question the decision of Wolff J. should be affirmed.

The appellants alone appealed against his Honour's decision upon the second question. The respondents did not cross appeal. The contention of the appellants is that there was an intestacy as to a half share of the income after the death of Elizabeth Coneybeare. The appeal upon this point, in my opinion, must fail with the appeal upon the main question. In this view I do not think that we should entertain now the contention of the respondents representing the estate of Mary Blood deceased that the income for the short interval in question fell into that estate by virtue of an implied cross-executory limitation. The contention seemed to be made only to preserve consistency and coherence of argument and perhaps interpretation. In the case of Boyd's will (Gibb-Maitland v. Perpetual Executors Trustees and Agency Co. (W.A.) Ltd.[16]) I have discussed fully the rule of construction upon which the contention is based. It is enough to say that I am not satisfied that an implication is required by the limitation of the codicil.

Appeal dismissed. Costs of appeal of all parties to be paid out of the estate, those of the trustees as between solicitor and client.

Solicitors for the respondents, the surviving trustees of the estate of Thomas Statham, Downing and Downing.

[1] (1882) 21 Ch. D. 811.

[2] (1927) 1 Ch. 421.

[3] (1804) 10 Ves. 166, at p. 176 [1804] EngR 352; [32 E.R. 808, at p. 811].

[4] [1847] EngR 593; (1847) 1 De G. & Sm. 355; [63 E.R. 1102].

[5] [1846] EngR 1115; (1847) 10 Beav. 47 [50 E.R. 499].

[6] (1869) 38 L.J. Ch. 472.

[7] (1875) L.R. 20 Eq. 342.

[8] (1882) 21 Ch. D. 811.

[9] (1927) 1 Ch. 421.

[10] (1927) 1 Ch., at p. 425.

[11] [1790] EngR 2301; (1790) 3 Bro. C.C. 50 [29 E.R. 402].

[12] (1816) 1 Mer. 358 [35 E.R. 706].

[13] [1853] EngR 187; (1853) 16 Beav. 431 [51 E.R. 845].

[14] (1916) 2 Ch. 50.

[15] (1916) 2 Ch., at p. 56.

[16] [1947] HCA 35; (1947) 74 C.L.R. 579.

High Court of Australia

Sumpton v Downing [1947] HCA 36; (1947) 75 CLR 76 (6 October 1947)

HIGH COURT OF AUSTRALIA