I am unable to concur in the opinion of Judge CARDOZO or fully in the opinion of Judge CRANE for the following reasons:
By the eighth article of the testator's will he gave his residuary estate to trustees named therein to divide into three parts, the first part to consist of 45% and each of the other two parts to consist of 27½%. He directed that the income of the first part be paid to his wife during her life and upon her death one-half of such 45% was to be disposed of by the trustees in the manner directed in such will as she might leave. Then follows a provision to the effect that if she did not make a testamentary disposition of such half the same was to be divided into two equal parts, one of which was to be held in trust for each of his daughters by a former wife for life with remainder to their respective issue. He appreciated that unless his wife did survive him she could not make a testamentary disposition of such half and having this *Page 244 thought in mind he sought by the ninth clause of his will to do what, as I understand, we are all agreed he could not do, namely, prevent the court from ascertaining if it could whether or not his wife survived him. It not being established that his wife did survive, the power of disposition given to her lapsed, since it could only come into existence by his death before hers. If the power never existed, obviously it could not be exercised, no matter what the testator said or what his wishes were on that subject.
But it is suggested that notwithstanding the power of appointment lapsed, the testator nevertheless could, and did, prevent the consequences of a lapse by providing in effect that the estate of which his wife was given the power of appointment, in case it could not be determined whether she survived him or not, should go to whom she designated in her will; and that this result can be accomplished (a) by reading into his will the will which she left, and (b) then construing his will as a gift by implication to the legatees named in her will, and (c) as a "ratification of any execution of the power, however premature."
There is, as it seems to me, a complete answer to the suggestion. (1) There is nothing in his will which in any way refers to her will and it cannot be read into his without disregarding the rule as settled by this court that an unattested paper which is of a testamentary character cannot be taken as a part of a will even though referred to by that instrument. (Booth v. Baptist Church of Christ, 126 N.Y. 215, 247;Matter of Will of O'Neil, 91 N.Y. 516, 523; Williams v.Freeman, 83 N.Y. 561, 566; Langdon v. Astor's Executors,16 N.Y. 9, 26.)
(2) I do not think his will can be construed in such a way as to constitute a legacy by implication to the legatees named in her will. There is no reference in his will to the legatees named in hers nor is there anything *Page 245 to indicate that he ever knew or had any knowledge of such legatees. In Bradhurst v. Field (135 N.Y. 564) this court held that to uphold a legacy by implication the inference from the will of the testator's intention must be such as to leave no hesitation in the mind of the court and to permit of no other reasonable inference. This rule was re-stated and approved inBrown v. Quintard (177 N.Y. 75). To sustain a bequest by implication there must be so strong a probability of the intention in this respect that the contrary cannot be supposed. (Post v. Hover, 33 N.Y. 593.) And especially is this true when the implication will result, as it does here, in the disinheritance of an heir at law. (Scott v. Guernsey, 48 N.Y. 106. ) An heir will not be disinherited unless by plain and cogent inference arising from the will. (Quinn v. Hardenbrook,54 N.Y. 83; Lynes v. Townsend, 33 N.Y. 558.)
(3) The method of construction suggested cannot be treated in my opinion "as a * * * ratification of any execution of the power, however premature." The power, so far as appears, never came into existence and there can be no ratification of its execution under such circumstances. A ratification of an act is always predicated upon the existence of the act ratified. There is no proof here that Mrs. Fowles ever exercised any power of appointment; hence, her assumed act is impossible of ratification.
It is said the construction suggested is justified by Matterof Piffard (111 N.Y. 410). I do not so understand that case. I think that authority is to the contrary. There certainly is a plain distinction between this case and that. In the Piffard case the will contained an express devise to the executors named in the daughter's will, and this provision was re-affirmed in the codicil executed after the daughter's death. The language in the will was: "I direct that such share or shares shall *Page 246 be paid over by my said executors to the executors or trustees named in and by the several wills of my said daughters in the case of the death of them or either of them in my lifetime instead of to my said daughter or daughters." In the codicil, the daughter having previously died, the language was: "Such share or shares shall be paid over by my said executors to the executors or trustee named in the several wills of my said daughters in case of the death of them or either of them in my lifetime instead of to my daughter or daughters." The property, therefore, passed in that case, not by the exercise of the power of appointment given or by anything said in the will to prevent the consequences of a lapse, but by the act of the testator himself.
In the present case the testator, if he had so desired, could have done precisely what was done in the Piffard case. He could have directed that if his wife predeceased him or if they both died under such circumstances that that fact could not be ascertained, then and in that event the property of which she was given the power of appointment should be paid to the executors named in her will. But he did not do this, and the court is now asked by a forced rule of construction to do it for him in order to avoid the consequences of the lapse. I am unwilling to do this, and, therefore, concur in the conclusion reached by my brother CRANE as to the construction to be put upon the will so far as the same relates to the one-half of the 45% of the residuary estate. I am, however, unable to concur in his opinion that the bequest to Mrs. Fowles of $5,000, and the furnishings of Fairmile Court upon the death of Mr. Fowles became the property of her estate. There is not a word in his will that indicates he ever intended such disposition. These bequests are absolute in form to her — not to her estate — and the rule is fundamental that a bequest lapses unless the legatee be alive at the time the testator *Page 247 dies. If these legacies became a part of her estate, then it is solely because she survived her husband, and the title to the property given to her vested prior to her death. We cannot so hold without assuming the existence of facts of which there are no findings or proof. These legacies, therefore, as it seems to me, lapsed and became a part of his residuary estate.
The fact that the testator, when he made his will, foresaw the possibility that he and his wife might meet the tragic end which they did, appeals so strongly to one's desire to carry out his supposed intention that we are apt to lose sight of the fact that the court can construe but cannot make wills, and in this connection it must be borne in mind that we are dealing, not with this one case, but with the law relating to the construction of wills which does not permit conjecture or possibility to take the place of proof. It is much better in my opinion to adhere to well-settled rules of construction rather than to make for a particular case new ones, the effect of which when put into practice cannot be foreseen.
For the reasons stated I think the order appealed from should be affirmed.
HISCOCK, Ch. J., CHASE and ANDREWS, JJ., concur with CARDOZO, J.; CRANE, J., reads dissenting opinion, except as to specific legacies, and CUDDEBACK, J., concurs; McLAUGHLIN, J., reads dissenting opinion.
Ordered accordingly. *Page 248