In Irving v. DiVito, the Virginia Supreme Court determined the circuit court did not err in refusing to admit to probate a writing offered as a holographic codicil.

The writing at issue in Irving is clear about the intended dispositive outcome—to disinherit someone. But the question at the core of this case is whether the testator intended the writing to have testamentary effect, i.e., whether it is authentic. The circuit and Virginia Supreme courts determined the writing was not intended as a codicil.


Declan Irving executed an attested will prepared by a law firm. The law firm gave the original will to Declan. They also gave him a binder with a copy of the will and his other estate planning documents. The binder included a 6-page “Estate Planning Reminders” document, which advised Declan not to make changes to the face of his will without contacting an attorney.

(Date Not Provided)
Declan’s marriage of 4 years ended in divorce. The property settlement agreement says, “no children were born of the marriage” and “Patrick D. Irving is not the child of [Declan].”

Declan wrote the following note in cursive on one of the tab dividers in his estate planning binder:


I wish to remove Patrick named as my son entirely from this will—no benefits.

[Declan’s initials]

Declan died after a battle with cancer.

Two notes were found in the hotel where Declan was staying (to receive treatment at a nearby hospital) when he died. The notes were addressed to his brother, Donal Irving.

  • One note, written on an advertisement for the law firm that prepared the will, reads “Brother Donal Montreal” with Donal’s phone number. On the reverse side is written “For my Brother Donal This is where my will is kept. Declan.”

  • On the other note, also on an advertisement for the law firm, is written “Donal my will is at [the law firm].” The other side of this note says “Call Donal Irving – Brother” and gives Donal’s address and phone number.

Despite what the notes indicate, Declan’s original will was in a briefcase in Declan’s storage unit. The binder was found elsewhere in the storage unit.

The 2000 will was admitted to probate. Donal was named and qualified as executor.

Donal offered the 2003 writing for probate as a holographic codicil. The clerk entered an order refusing to admit the writing. Donal appealed to the circuit court.

The circuit court agreed with the clerk, finding the writing did not comply with the requirements of Va. Code Sec. 64.2-403. The circuit court determined it was not manifest that the name in the document was intended as a signature. The circuit court also found the writing “establishes a thought or plan of the decedent to make a change to [the] will, and is precatory and tentative in nature.” Therefore, testamentary intent was not established by clear and convincing evidence, and the writing was not admissible to probate under Sec. 64.2-404. Donal appealed to the Virginia Supreme Court.

On appeal, Donal made two alternative arguments for the admissibility of the writing:

  1. It was a valid holographic codicil under Virginia Code § 64.2-403; and
  2. Any deficiency in the formalities was harmless error under the dispensing statute, Virginia Code § 64.2-404.

The Virginia Supreme Court unanimously upheld the circuit court.


Notwithstanding the different bases for Donal’s two arguments, underlying the Court’s analysis of both is a concern that Declan did not intend the instrument at issue to alter the disposition of his property at his death.

A. Was the document a valid holographic codicil?

Virginia Code § 64.2-403 sets out the requirements for a valid holographic will or codicil. It provides as follows:

A. No will shall be valid unless it is in writing and signed by the testator, or by some other person in the testator’s presence and by his direction, in such a manner as to make it manifest that the name is intended as a signature.

B. A will wholly in the testator’s handwriting is valid without further requirements, provided that the fact that a will is wholly in the testator’s handwriting and signed by the testator is proved by at least two disinterested witnesses.

Critical to the Supreme Court’s analysis is the reading of Va. Code 64.2-403(A) to require not only that a writing be signed but also that it be signed “as and for a will” (or codicil).

According to the Court, the circuit court’s determination that the writing was not “signed by the testator. . . in such a manner as to make it manifest that the name [wa]s intended as a signature” was a determination (by a preponderance of the evidence) that the signature did not authenticate the document “as and for a codicil.” [^1] The circuit court’s determination was factual, and it is reviewed for plain error or the absence of supporting evidence. The Court cited the following facts to conclude that there was sufficient evidence for the circuit court to find as it did: (1) Declan used his full signature on legal and testamentary documents, i.e., the property settlement agreement and his attested will; and (2) Declan’s notes mentioned the will but not the codicil.

1. The Signature.

Declan used his initials, his full name, or something in between to sign various documents:

  • He tended to sign documents of legal significance with his full name, though the legal documents referenced in the opinion were probably signed under a lawyer’s supervision (a point which was made during oral argument).
  • He signed his checks (presumably without a lawyer present) “D.P. Irving.”
  • He used his initials as a signature at work, according to colleagues. (Declan was a medical doctor, so he presumably used his initials to sign things like charts and prescriptions.)

From this evidence, the Court determined Declan’s use of only his initials on the writing raised “a doubt as to whether he signed the writing with the intent to authenticate it ‘as and for’ a codicil.” [^2]

I am not convinced Declan’s use of his initials supports the lack of authenticity. I do not see people using their initials rather than signing their full names to signify the documents should not be given legal effect. But perhaps I’m just unfamiliar with the convention. Do people who use their initials regularly do so as the equivalent of fingers crossed behind their backs? I would also be interested in knowing whether there is any evidence Declan engaged in that practice.

To me, the document seems authentic, i.e., intended as and for a codicil. The application of the initials and the date (in the absence of evidence that Declan initialed and dated notes to himself) suggest the document was intended to be given effect. The document’s placement in Declan’s estate planning binder suggests it was meant to be read with the other estate planning documents. Furthermore, the instrument refers to “this will,” suggesting Declan considered the instrument to be part of his will.

2. The Notes to Donal.

The notes found in Declan’s hotel room directed the public to Donal and directed Donal to Declan’s lawyer. About the notes, the Supreme Court said “[o]ur doubt is strengthened by the fact that Declan’s notes directed Donal only to his will without mentioning the writing, which suggests that Declan did not consider his initials to have authenticated the writing as a testamentary document.” In other words, in referring Donal to the law firm, Declan intended only the will prepared by the law firm to be given effect.

I’m not convinced the notes support any useful conclusion here.

  • It’s hard to believe Declan would refer Donal to the law firm to avoid the probate of his holographic codicil.[^3]
  • As noted above, it seems Declan considered the instrument to be part of his will. If that’s so, I’m not sure we should expect the codicil to be mentioned separately.
  • Given that the notes were wrong about the location of the will, perhaps we should only infer from them that Declan (like many smart people) wasn’t sure which copy of his will was the important one or where the original was.

B. Were any deficiencies in the document’s execution harmless error?

Virginia Code § 64.2-404, the dispensing statute or harmless error rule provides as follows:

“Although a document. . . was not executed in compliance with § 64.2-403, the document or writing shall be treated as if it had been [so] executed if the proponent of the document or writing establishes by clear and convincing evidence that the decedent intended the document or writing to constitute. . . the decedent’s will. . . [or] an addition to or an alteration of the will.”

Donal’s argument under this section also failed to convince the Supreme Court. Relying on its prior analysis, the Court said, “[i]f [Declan] intended and believed the writing to be an effective codicil, it is reasonable to conclude that he would have mentioned it in his notes to Donal. His failure to do so, along with his failure to sign the writing in the same manner as his will, suggests that he did not consider the writing to have binding testamentary effect.”

In dicta, the Supreme Court pointed out the writing expresses a wish. The Court said, “the circuit court’s conclusion that the writing itself does not actually effect a change to Declan’s will, but establishes only a ‘thought or plan’ to change his will that he ‘wish[ed]’ to implement at some point, is also supported by the evidence. The proposed codicil specifically states, ‘I wish to remove Patrick named as my son from this will—no benefits.’ (emphasis added).

I am reluctant to ascribe the Court’s interpretation of “wish” understanding to a non-lawyer. “I wish to remove…” is not everyday speech. In my observation, “I wish to remove…” is the sort of stilted language non-lawyers think lawyers use to express directives in wills.


I am not sure I would have reached the same conclusions as the courts in this case. However, I am not in possession of all the facts. (I don’t even have the briefs.) The circuit court was in the best position to weigh the evidence presented, and the Supreme Court’s standard of review accorded due deference.

There are some planning lessons to be taken from Irving, e.g., that people should sign codicils in the same way they sign their wills. However, most of those lessons are not particularly useful to T&E practitioner, because the Wills Act formalities allow us to readily distinguish authentic attested wills from inauthentic ones.

The recent trend in the law has been toward more freely admitting wills, perhaps particularly homemade wills, to probate. The enactment of the harmless error rule reflects that trend. Irving may be a step in the opposite direction.[^4] Either way, the case gives us cause to wonder whether current law (Wills Act formalities, harmless error rule, etc.) strikes the right balance between giving effect to authentic wills while denying effect to inauthentic writings.

  1. Apparently, “without further requirements” in B does not make A inapplicable to holographic wills.
  2. Based on the Court’s analysis and the precedent cited, it seems the conclusion would have been different had there been other evidence of whether the instrument was intended as a codicil. In this case, the signature seems to be one of the only available pegs (because of the statute and the apparent lack of other evidence of authenticity.
  3. The circumstances under which that might be logical, e.g., the testator is paranoid, don’t seem to be present here.
  4. I have not researched the extent to which the Virginia courts have been following this trend.