In re Estate of Charles H. PRATT, Deceased.
CHILDREN'S HOME SOCIETY OF FLORIDA, the Berkshire Industrial Farm, the New York Infirmary for Women and Children, the Seaman's Church Institute and the Tuskegee State Normal and Industrial Institute, Appellants,
v.
Helen Aubrey PRATT, Drusilla Pratt Gjoerloff, Martha Pratt Shermund, Robert Hyde Pratt, by and through his next friend and natural guardian, Helen Aubrey Pratt and W.E. Winderweedle, Appellees.

Supreme Court of Florida, Special Division A.

March 2, 1956. Rehearing Denied July 16, 1956.

*500 G.W. Botts, Jacksonville, J.R. Spector, New York City, and Fleming, Jones, Scott & Botts, Jacksonville, for appellants.

J. Thomas Gurney, Orlando, for appellees.

HOBSON, Justice.

Charles H. Pratt departed this life on September 8, 1953, leaving a will executed three days before his death. The will commenced with a conventional revocation clause "hereby expressly revoking all other Wills and Codicils by me at any time made" and gave certain directions pertaining to funeral expenses, costs of administration and taxes. Under Item III of the will, the rest, residue and remainder of the property was to be divided into three equal portions and disposed of as prescribed in subdivisions A, B, and C of Item III. Under A, the first equal third portion of the residuary estate was bequeathed to the widow absolutely. Under B, another equal third was bequeathed to the testator's three children, share and share alike. Under C, the third equal portion was to be devoted first to certain minor specific legacies, and the balance as follows:

"The balance of said third portion of my residuary estate remaining after the payments of the said legacies, I give, devise and bequeath absolutely and forever in equal shares to the following:

"New York Infirmary for Women and Children, of New York City.

"Tuskegee State Normal and Industrial Institute of Tuskegee, Alabama.

"Berkshire Industrial Farm of Canaan, Columbia County, New York.

"Seamens Church Institute of New York, located at 25 South Street, New York City, N.Y.

"The Children's Home Society of Florida, Jacksonville, Florida."

This will was admitted to probate. Thereafter, the widow and three children of the testator petitioned the County Judge's Court, praying that gifts to the charitable corporations named in the will be avoided under F.S. Sec. 731.19, F.S.A., which provides that a will making charitable bequests shall be invalid as to such bequests unless it was duly executed at least six months before the testator's death.

The charitable corporations defended upon the ground that the testator, by a will validly executed more than six months before his death, had made the same charitable bequests; that by the probated will he had merely reiterated a dispositive scheme conceived more than six months before his death; and that under these circumstances F.S. Sec. 731.19, F.S.A. could not be invoked to invalidate the gifts to the charities. The charities also sought to compel the attorney for the executor to produce the earlier will. On motion, the defenses were stricken and the attorney for the executor was held not required to produce the prior will. The County Judge declared the charitable bequests void and ordered that the portion of the estate which was the subject of the invalid bequests be distributed as intestate property.

This judgment was appealed to the Circuit Court, by which it was affirmed, and *501 the charitable corporations now appeal to this court. Appellees are the widow and children of the testator and the attorney for the executor.

F.S. Sec. 731.19, F.S.A. so far as it is applicable here, reads as follows:

"If a testator dies leaving issue of his body, or an adopted child, or the lineal descendants of either, or a spouse, and if the will of such testator devises or bequeaths the estate of such testator, or any part thereof, to a benevolent, charitable, literary, scientific, religious or missionary institution, corporation or association or purpose, * * * such will as to such devise or bequest shall be invalid unless it was duly executed at least six months prior to the death of the testator."

We had occasion to construe this statute in Taylor v. Payne, 154 Fla. 359, 17 So.2d 615, 618, 154 A.L.R. 677. There we said that the purpose of the statute is "to prevent testators who may be laboring under the apprehension of impending death from disposing of their estates to the exclusion of those who are, or should be, the natural objects of the testator's bounty." We also held that the statute must be so construed as to secure full protection to the shielded class, at the same time giving as much effect as possible to the intention of the testator insofar as it is not inconsistent with some settled rule of law or public policy. Thus the statute is to be neither strictly nor narrowly construed. But its plain language, from which we cannot deviate, is broad enough to encompass many cases where a testator, by no means in extremis or even contemplating death, makes a charitable bequest and dies shortly thereafter. The statutory coverage is much more comprehensive than its apparent purpose would warrant, but the restriction of its scope is a problem for the legislature.

In the case before us, it is obvious that since the statutory prohibition has been raised by the testator's widow and children it operates to avoid the charitable gifts under the will from which we have quoted, and the courts below properly so held. The controlling question is whether the charitable bequests allegedly made in the earlier will should have been revived under the doctrine of "dependent relative revocation", as appellants have vigorously contended thoughout this litigation. The application of this doctrine would give rise to the rebuttable presumption that the testator would have preferred to revive his earlier charitable bequests rather than let the property go by intestacy.

The doctrine of dependent relative revocation becomes less abstruse and more comprehensible under other names. In the most celebrated article on the subject, Warren, Dependent Relative Revocation, 33 Harv.L.R. 337, the author suggests "revocation under a mistake". Thus if the testator by codicil, revokes a portion of a prior testamentary instrument and makes a substituted disposition under a mistake of fact or of law with the result that the later disposition is invalid, the prior disposition is revived on the theory that had the testator not been mistaken in his belief he would not have revoked the original gift. As explained by the Supreme Court of Iowa in Blackford v. Anderson, 226 Iowa 1138, 286 N.W. 735, 746:

"The basis for the doctrine of dependent relative revocation, or conditional or mistaken revocation, as it is sometimes called, is that there was never any revocation of the earlier instrument, or real intention to revoke, because of a mental misconception of the effect of his act, on account of mistake, or ignorance, or some other error."

The doctrine is called "conditional revocation" upon the theory that the testator has revoked the former gift only on condition that the later gift will be operative. Atkinson on Wills, 2d Ed., pp. 452-463.

In Stewart v. Johnson, 142 Fla. 425, 194 So. 869, we applied the doctrine to give effect to a prior will when a later will, containing an express revocation clause, failed for lack of sufficient witnesses. The *502 same result could have been achieved by applying Laws 1933, c. 16103, Sec. 14, which provided, as F.S. Sec. 731.13 F.S.A. presently provides, that a will may be revoked by a written instrument, provided such instrument is executed with the same formalities required for the execution of wills. Since the second will in the Stewart case attempted to revoke the first but did not comply with the statute, the revocation was ineffective and the first will stood unaltered. The situation in the Stewart case was similar to that in the classic case of Onions v. Tyrer, 2 Vern, 742, 23 Eng.Rep. 1085 (1717) where the dependent relative revocation doctrine was applied to save an earlier will when the later revoking will failed for want of proper attestation. Onions v. Tyrer, however, was decided under the common law of the period, which permitted revocation by unattested writing. Atkinson on Wills, supra, p. 461, n. 46. Under modern law, codified in F.S. Sec. 731.13, F.S.A., supra, Atkinson gives the following explanation at p. 461:

"If * * * invalidity [of a subsequent gift] is due to a defect in execution, no question of dependent relative revocation arises for the clause of revocation would likewise be ineffective since the same formalities are required for a revoking as for a disposing instrument."

In another class of case, the subsequent instrument is a properly executed codicil which revokes, expressly or by implication, a gift in a will or previous codicil, and itself attempts a new disposition of property which is ineffective. Here the doctrine may be applied to save the earlier gift. The will and its codicils if properly executed must by their very nature be presented to the probate court, by which they must be examined, and the problem calling for the application of the doctrine will then generally be apparent to the court without resort to evidence extrinsic to the documents themselves. Although execution of a codicil has the effect of republishing the will [in Florida, this rule is codified in F.S. Sec. 731.17, F.S.A., as construed in Massachusetts Audubon Soc. v. Ormond Village Imp. Ass'n, 152 Fla. 1, 10 So.2d 494], it has been held, under statutes similar to F.S. Sec. 731.19, invalidating charitable bequests if made within a certain period before the testator's death, that such gifts are effective if the will precedes the prescribed period although the codicil does not. In re Pence's Estate, 117 Cal. App., 323, 4 P.2d 202; In re McCauley's Estate, 138 Cal. 432, 71 P. 512; Appeal of Carl, 106 Pa. 635.

One of the clearest examples of a codicil case where the "dependent relative revocation" doctrine was in our opinion, properly applied is the English case of In re Bernard's Settlement [1916] 1 Ch. 552. There the testator, by will, exercised a power of appointment, but expressly revoked it by a third codicil in which a new appointment was made which violated the rule against perpetuities. In its opinion, the court thus stated the question to be resolved [1 Ch. 560]:

"Did the testator intend by the second appointment to revoke in any case the prior appointment, or did he really only intend to revoke it for the purpose of carrying out the alteration made in his second appointment and without having any intention of revoking the previous gift except for the purpose of the altered appointment?"

It was held, (solely on the basis of the testamentary papers necessarily before the court, so far as appears) that the probable testamentary intention was to revoke only in favor of a valid new appointment, and that the original appointment was preferable to intestacy and should therefore be made effective.

The differences between the instant case and the improper execution cases discussed above is too plain for comment. The codicil cases, however, are another matter.

It might appear that we could treat the later will as a codicil. But wills and codicils have different legal incidents which overlap but by no means coincide. And no test now occurs to us which would be useful in a case like this for determining when a testamentary document which appears in *503 all respects to be a will should be treated as in fact a codicil by the Probate Court.

The case before us is one wherein the testator, by an unambiguous, complete testamentary instrument, has disposed of all of his property, expressly revoking all prior wills. There is no suggestion of fraud or undue influence. There are no conflicting provisions of testamentary papers necessarily before the court, as in the case of a will and codicils, which might form a basis for the admission of parol evidence for its bearing upon the testamentary intent. We know of no principle of law which would authorize us to look beyond the probated will for testamentary intent in such a case. We do not understand that the "dependent relative revocation" doctrine, useful and salutary as it is in a proper case, carries with it the authority to disregard so well-established a rule as that which forbids us to write a new will for the testator in the face of a clear intent expressed in a proper instrument.

In the much-cited article to which we have alluded, Warren, Dependent Relative Revocation, the author states in part [33 Harv.L.R. at pp 348-9]:

"A will cannot be set aside for mistake in either the United States or in England where the testator knew and approved its contents. By the same token a revocation absolute on its face, the words of which the testator knows and approves, will not be set aside because the reasons which induced it are found to be based on false assumptions of fact or law. The testator is dead, and it is too dangerous to inquire what motives induced his action. This was to be done for the construction of the document; it should not be resorted to for alteration of it."

Two courts, in cases almost identical with the one before us, have held that an earlier charitable bequest is revived in spite of an express revocation clause in a later will. The Supreme Court of California so held in In re Kaufman's Estate, 25 Cal.2d 854, 155 P.2d 831. In that case, the earlier will and the circumstances surrounding the execution of the later one had been examined at length by the probate court, but the propriety of conducting such examination and admitting the evidence incident thereto does not seem to have been questioned, since it is not discussed by the California Supreme Court, nor was it discussed in the opinion of the District Court of Appeal (see In re Kaufman's Estate, 151 P.2d 37). The case is therefore not helpful upon this basic objection to the application of the doctrine. A second feature of the case perplexes us. The attorney who drafted the last, revoking will testified that he had consulted the California Code, discovered the provision invalidating charitable bequests made within thirty days of death, and advised the testator to make the desired change by codicil. The attorney testified in part [155 P.2d at page 833]:

"`I said to him, "if I were you, I would draw a Codicil to this Will because should you die within 30 days it would not be effective." He said, "No, I want a new will. I don't want a Codicil," and I drew the Will of 1941.'"

Thus the testator, having been advised of the effect of the statute, nevertheless drew a new will expressly revoking all others, under no misapprehension of fact or of law. So far as appears, he fully understood that if he died within thirty days the later charitable bequest would be invalid, yet he did not expressly make the revocation of the charitable bequest in his previously executed will conditional upon the validity of the later one. It is thus impossible to apply the "mistake" theory, and the "condition" theory is, to say the least, extremely tenuous under the circumstances. Insofar as the intent of the testator to leave money to the charity is concerned, such intent exists in every case where a charitable bequest is made but is frequently thwarted by statute in favor of the heirs. And we can give effect to the intent of the testator only insofar as it is not inconsistent with law, as we have observed. It is only the intent to revoke which is important in this discussion, and it appears that if, as in the Kaufman *504 case, the testator revokes absolutely and with full knowledge of the consequences there is nothing further to be considered. Compare the position of the testator in Re Bernard's Settlement, supra, where, by mistake of law, the latest codicil was invalid for violation of the rule against perpetuities. And note that there the Probate Court, presented with the will and codicils, had the problem thrust upon it, whereas in the Kaufman case the problem was unearthed only after a search behind the will. The Kaufman situation illustrates the danger of admitting and considering parol evidence to modify the unambiguous words of a complete testamentary instrument.

The second case in this line of authority is Linkins v. Protestant Epsicopal Cathedral Foundation, 87 U.S.App.D.C. 351, 187 F.2d 357, 28 A.L.R.2d 521. The testatrix had made three wills, making identical charitable bequests. The last will contained an express revocation clause. The District of Columbia Code, Sec. 19-202, provided that no such bequest "shall be valid unless the same shall be made at least one calendar month before the death of the testator." The testatrix died within the statutory period. The United States Court of Appeals, District of Columbia Circuit, held, in the main opinion by Judge Prettyman, that the charitable bequests in the will immediately preceding the last one were revived under the doctrine of dependent relative revocation. The Kaufman case was the authority for this decision. In the Linkins case, the court disposed of the extrinsic evidence problem in the following words:

"We hold that evidence extrinsic to the latest will may be adduced to show that the execution of the revocatory clause was intended to be upon a condition. Thus we find the solution in a well-established pattern of law. That pattern applies generally to contracts and even to deeds; e.g., delivery of an executed contract may be shown to have been in escrow, or a deed may be shown to have been in fact a mortgage." 87 U.S.App.D.C. 351, 353, 187 F.2d 357, 359, 28 A.L.R.2d 521, 524.

In this way, by analogy to the law of deeds and contracts, it was held proper to admit evidence to explain an unambiguous will. But the considerations are different in these apparently analagous fields of law, the principal difference being that they do not necessarily contemplate that the principal party, whose intention controls if it is possible to ascertain and give effect to it, is dead. The law of wills is calculated to avoid speculation as to the testator's intent and to concentrate upon what he said rather than what he might, or should, have wanted to say. Wright v. Sallet, Fla., 66 So.2d 237; Martin v. Shands, Fla., 49 So.2d 598; State v. North, 159 Fla. 351, 32 So.2d 14; Iles v. Iles, 158 Fla. 493, 29 So.2d 21; In re Block's Estate, 143 Fla. 163, 196 So. 410; Howe v. Sands, 141 Fla. 813, 194 So. 798, rehearing denied 142 Fla. 625, 195 So. 609; Rewis v. Rewis, 79 Fla. 126, 84 So. 93. Moreover, in probate matters statutory mandates hedge us in closely. Under F.S. Sec. 731.09, F.S.A., for example, revocation of a will is void if such revocation "is procured by fraud, duress, menace or undue influence * * *." At the very least, this statute, in laying down grounds for setting aside a revocation, narrows the allowable ambit of our movement in this field. See Hill v. Morris, Fla., 85 So.2d 847.

In a separate concurring opinion in the Linkins case, Judge Edgerton expressed the view that it would be possible to hold that the statute restricted only the making of new bequests and not the restatement of existing ones. Thus a bequest in a will executed within the statutory period would not be deemed "made", within the language of the D.C.Code, if it restated, or perhaps substantially restated, an earlier bequest. This result would, as pointed out in the note on the Linkins case at 64 Harv.L.R. 686, come closer to effectuating the testamentary intent than the use of the dependent relative revocation doctrine, because, by holding the statute inapplicable, the later and not the earlier bequest would be made effective. We cannot, however, apply this construction in Florida, because our statutory prohibition is hinged not upon the time *505 the bequest was "made" but upon the time the will was executed. There is no room for construction under our statute in a case of this nature.

In fine, although our labors over this question have been extensive, we can find no plausible theory upon which to proceed in permitting the examination of the revoked will in the case at bar with a view to further proceedings under the "dependent relative revocation" doctrine. We therefore conclude, as we must, that no error has been shown, and the judgment appealed from must be, and it is hereby

Affirmed.

DREW, C.J., and TERRELL and O'CONNELL, JJ., concur.