COVID-19 and Formal Wills

David Horton & Reid Kress Weisbord *


Most Americans do not have a will. 1 See 2020 Estate Planning and Wills Study,, (archived May 18, 2020) (reporting that 68% of survey participants did not have a will or another type of estate planning document). The reasons are easy to understand. Thinking about death is unpleasant, and hiring a lawyer is expensive. However, as COVID-19 sweeps through the country, some Americans urgently need an estate plan. 2 See Julie Jargon, Estate Planning Goes Digital as Many Families Explore Options, Wall St. J. (Mar. 25, 2020, 5:30 AM EST), (explaining that the pandemic “has unfortunately made [estate planning] a more urgent topic”).

Unfortunately, U.S. law makes it difficult to create a will during crises like these. Indeed, roughly half of the states recognize only one type of will: a “formal” will executed in compliance with the Wills Act. 3 See infra Part III; cf. Jesse Dukeminier & Robert H. Sitkoff, Wills, Trusts, And Estates 197 (9th ed. 2013) (illustrating through a map of the U.S. that about half of the jurisdictions recognize formal wills and half also validate holographic wills). Under this ancient statute, wills must be written, signed by the testator, and also witnessed by two people who were present at the same time. 4 Statute of Frauds 1677, 29 Car. 2 c. 3, § 4 (Eng. & Wales); Wills Act 1837, 7 Will. 4 & 1 Vict. c. 26, § 9 (Eng., Wales & N. Ir.). As journalists and lawyers are recognizing, the Wills Act’s insistence that the parties physically occupy the same space “creat[es] unprecedented roadblocks” during a time of widespread quarantine and shelter-in-place orders. 5 See Susan B. Garland, What to Know About Making a Will in the Age of Coronavirus, N.Y. Times (Mar. 30, 2020),

Yet the pandemic has also arrived during a period in which wills law is in flux. In the last two decades, a handful of jurisdictions have begun excusing harmless errors during the will-execution process. 6 See, e.g., David Horton, Partial Harmless Error for Wills: Evidence from California, 103 Iowa L. Rev. 2027, 2032 & n.37 (2018) (collecting harmless error statutes). This bold new rule is “the most significant change in what constitutes a will since [the] enactment of the Statute of Frauds.” 7 Bruce H. Mann, Formalities and Formalism in the Uniform Probate Code, 142 U. Pa. L. Rev. 1033, 1035 (1994). And, in an even sharper departure from the Wills Act’s stuffy norms, four states have recently authorized electronic wills. 8 See Ariz. Rev. Stat. Ann. § 14-2518 (West 2020); Fla. Stat. Ann. § 732.522 (West 2020); Ind. Code Ann. § 29-1-21-4 (West 2020); Nev. Rev. Stat. Ann. § 133.085 (LexisNexis 2019).

This Essay argues that COVID-19 vividly highlights the shortcomings of formal wills. Indeed, the outbreak has exposed the main problem with the Wills Act: it renders will-making inaccessible. As a result, we urge lawmakers in states that cling to the statute to liberalize the requirements for creating a will.

Our argument proceeds in three Parts. Part I details the social value of will-making. Part II describes the Wills Act and explains why it creates formidable obstacles for testators who are caught in the jaws of a pandemic. Part III explores four ways in which policymakers can solve this problem: by permitting holographic wills, adopting the harmless-error doctrine, enacting electronic will legislation, or temporarily suspending certain elements of the Wills Act during public health emergencies.

I. The Benefits of Will-Making

Courts often declare that there is an “established public policy favoring testacy.” 9 Estate of Reid, 145 Cal. Rptr. 451, 454 (Ct. App. 1978). This is because the downsides of dying intestate are well-documented.

For one, the intestacy laws generally distribute property to the decedent’s spouse or descendants. 10 See Unif. Probate Code §§ 2-102, 2-103 (2010). As a result, intestacy is not necessarily suitable for unmarried same-sex couples and other non-traditional families. Similarly, people who die intestate cannot use a will to nominate a guardian for minor children or appoint an executor to administer their estate. 11 See id. § 5-202 (“A guardian may be appointed by will or other signed writing by a parent for any minor child the parent has or may have in the future.”); Reid Kress Weisbord, Wills for Everyone: Helping Individuals Opt Out of Intestacy, 53 B.C. L. Rev. 877, 896 (2012).

Moreover, intestacy causes wealth to fractionate. 12 See Weisbord, supra note 11, at 895. For example, by dividing an asset like a family home equally among many different heirs—who may disagree about what to do with it—intestacy reduces its value. This problem is especially damaging in low-income communities, where the testator’s residence may be the most valuable asset in the estate. 13 See Palma Joy Strand, Inheriting Inequality: Wealth, Race, and the Laws of Succession, 89 Or. L. Rev. 453, 496-97 (2010).

In addition, intestacies fare worse in probate. One might expect the opposite to be true: Intestacies do not require courts to authenticate a will, ensure that the decedent had testamentary capacity, adjudicate allegations of undue influence, or interpret ambiguous language. But several factors make intestacies harder to administer. For one, the judge—not the decedent—must select a personal representative, which can be contentious and time-consuming. 14 See Weisbord, supra note 11, at 896. Also, the need to locate heirs can cause delays.

In fact, we found hard evidence of these procedural downsides in a dataset that we collected of 1,349 probate filings from San Francisco Superior Court between 2014 and 2016. We discovered that 90 of the 676 intestacies (13.3%) devolved into litigation, compared to only 65 of the 673 testacies (9.7%)—a difference that is statistically significant (p < 0.05). Likewise, intestacies took longer, lasting an average of 552 days, while testacies concluded in a mean of just 528 days (although, admittedly, this divergence is not statistically meaningful (p = 0.17).

Finally, making a will has psychological benefits. Most obviously, testators feel a sense of relief when they get their affairs in order. 15 See Mark Glover, The Therapeutic Function of Testamentary Formality, 61 U. Kan. L. Rev. 139, 144 (2012) (noting that “will formalities interact with the therapeutic qualities of estate planning and affect the overall therapeutic potential of the estate planning process”). But in addition, wills have long been seen as “vessel[s] of truth” that allow “one [to] sa[y], at last, exactly what one felt.” 16 Edward Champlin, Final Judgments: Duty and Emotion in Roman Wills, 200 B.C.–A.D. 250, at 9-10 (1991). As a result, creating a will can be cathartic for people who are grappling with their own mortality. 17 See David Horton, Testation and Speech, 101 Geo. L.J. 61, 89 (2012) (arguing that creating a will is a unique form of self-expression).

Nevertheless, the law of wills contains a paradox. As we explain next, despite the advantages of will-making, it is not easy to create a valid testamentary instrument.

II. COVID-19 and Formal Wills

Every American jurisdiction allows testators to create a formal will under the Wills Act. But people who are infected with COVID-19 or sheltering in place may have difficulty going down this path.

The Wills Act dates back to the English Statute of Frauds, which Parliament passed in 1677. 18 Statute of Frauds 1677, 29 Car. 2 c. 3 (Eng. & Wales); Wills Act 1837, 7 Will. 4 & 1 Vict. c. 26, § 9 (Eng., Wales & N. Ir.). It requires wills to be in writing and signed by the testator. 19 Wills Act 1837, 7 Will. 4 & 1 Vict. c. 26, § 9 (Eng., Wales & N. Ir.). In addition, unlike other transfers of property, such as contracts and gifts, the Wills Act also mandates that testators either sign the will or acknowledge a previously made signature in front of two witnesses who are present in the same room at the same time. 20 Id. The witnesses must then subscribe the will, too. 21 Id.

Some American states not only imported the demands of the English Wills Act, but also added their own idiosyncratic formalities. In New York, for instance, the witnesses must sign no more than thirty days after a ceremony in which the testator “declare[s] to each of the[m] . . . that the instrument to which his signature has been affixed is his will.” 22 N.Y. Est. Powers & Trusts Law § 3-2.1(a)(3)-(4) (McKinney 2019). Similarly, under West Virginia’s version of the Wills Act, which became effective in 1882, the witnesses and the testator need to be in physical proximity to each other both when the testator signs the will and when each witness signs the will. 23 W. Va. Code Ann. § 41-1-3 (West 2020) Florida imposes the same “simultaneous presence” requirement for attesting witnesses as West Virginia. 24 Fla. Stat. Ann. § 732.502(b)-(c) (West 2020).

Commentators have offered four justifications for the Wills Act’s components. First, the writing and signature elements help clarify the testator’s wishes (the evidentiary function). 25 See Ashbel G. Gulliver & Catherine J. Tilson, Classification of Gratuitous Transfers, 51 Yale L.J. 1, 6 (1941). Second, the fanfare of a witnessed document reinforces the gravity of testation (the ritual function). 26 See id. at 5. Third, the necessity of witnesses shields the testator from fraud and coercion (the protective function). 27 See id. at 9. And fourth, by shoehorning a decedent’s wishes into a recognizable format, “[c]ourts are seldom left to puzzle whether the document was meant to be a will” (the channeling function). 28 John H. Langbein, Substantial Compliance with the Wills Act, 88 Harv. L. Rev. 489, 493-94 (1975). There is no question that these are important purposes.

Nevertheless, courts have insisted on strict compliance with each step and annulled would-be wills even when all of the functions of the formalities were satisfied. For example, judges have refused to enforce documents that the testator forgot to sign 29 See Brown v. Fluharty, 748 S.E.2d 809, 810 (W. Va. 2013) (per curiam). or those signed by witnesses who heard the testator acknowledge his signature over the telephone. 30 See In re Estate of McGurrin, 743 P.2d 994, 1002 (Idaho Ct. App. 1987) (“a telephonic acknowledgment by the testator, without more, will not suffice”); accord In re Will of Jefferson, 349 So. 2d 1032, 1036 (Miss. 1977). Likewise, in a recent case that made headlines, a decedent attempted to devise his $7 million house to his same-sex partner after living there together for fifty-five years, but only secured the signature of one witness, making his attempt at estate planning “legally invalid.” 31 Sarah Maslin Nir, A Brownstone and the Bitter Fight to Inherit It, N.Y. Times (Oct. 23, 2016), Results like these are why the caselaw interpreting the Wills Act is infamous for its “harsh and relentless formalism.” 32 Langbein, supra note 28, at 489. Bruce Mann’s searing critique, published more than twenty-five years ago, still rings true today in many jurisdictions: “Courts have routinely invalidated wills for minor defects in form even in uncontested cases and sometimes even while conceding—always ruefully, of course—that the document clearly represents the wishes and intent of the testator.” 33 Mann, supra note 7, at 1036.

This unforgiving rubric makes formal wills impractical for those who are sheltering in place or have contracted COVID-19. First, testators who try to navigate the Wills Act themselves face several daunting barriers. Stay-at-home orders and social distancing can make it hard to find the requisite witnesses. In fact, this task is impossible for people who are living alone. Finally, because the witnesses must be physically present for the testator’s signature or acknowledgment, Zoom, FaceTime, and other videoconferencing methods are non-starters under traditional doctrine.

Second, because executing a formal will is so complicated, testators usually hire lawyers to shepherd them through the process. But like many businesses, law firms across the country are currently closed. To be sure, some states—including Idaho, Iowa, Maryland, Minnesota, Mississippi, Nebraska, Nevada, and New Hampshire—have sought to provide access to legal services by temporarily permitting remote notarizations. 34 See Garland, supra note 5; Emergency Remote Notarization and Witnessing Orders, Am. Coll. Tr. Estate Counsel (May 8, 2020), [hereinafter Emergency Orders]. Nevertheless, that relief is cold comfort for testators, because (with two exceptions we note in Part III.D) notarization is generally not a substitute for witnesses under the Wills Act.

Third, even welcome technological developments fail to cure this predicament. For several years, vendors such as Nolo, LegalZoom, and Rocket Lawyer have sold inexpensive fill-in-the-blank will templates over the Internet. 35 See David Horton, Do-It-Yourself Wills, 53 U.C. Davis. L. Rev. (forthcoming 2020) (manuscript at 6), Some scholars have touted this trend as a shift in estate planning away “from an elite universe to one by and for everyone.” 36 Iris J. Goodwin, Access to Justice: What to Do About the Law of Wills, 2016 Wis. L. Rev. 947, 951. Yet these do-it-yourself wills are designed to be filled out online, printed, mailed to the testator, and then executed in compliance with the Wills Act. Accordingly, they do not always bridge the gap in the age of COVID-19.

In sum, formal wills are not well-suited to an era in which “social distancing [is] the new normal.” 37 Garland, supra note 5. But as we explain next, there are several ways to overcome this hurdle.

III. Alternatives to Formal Wills

This Part evaluates four ways in which lawmakers in states that adhere to the Wills Act could loosen the rules of will-creation and empower lay testators. Although none are perfect, we contend that the benefits of each outweigh the costs in the current landscape.

A. Holographic Wills

One species of self-made will is already somewhat common: the holograph. About half of the states, including California, New Jersey, Pennsylvania, Texas, and Virginia, allow testators to make holographic wills. 38 See, e.g., Richard Lewis Brown, The Holograph Problem—The Case Against Holographic Wills, 74 Tenn. L. Rev. 93, 93 & n.2 (2006). Holographs do not need witnesses. 39 Id. at 93. Instead, they must be entirely in the testator’s handwriting and signed by her. 40 Id. at 116.

Because holographs are so easy to create, they have earned a reputation as “emergency room” wills. 41 Stephen Clowney, In Their Own Hand: An Analysis of Holographic Wills and Homemade Willmaking, 43 Real Prop. Tr. & Est. L.J. 27, 58 (2008). In one famous example, a farmer named Cecil George Harris became trapped under his tractor and used his pocketknife to scratch into the fender: “In case I die in this mess I leave all to the wife. Cecil Geo. Harris.” 42 Geoff Ellwand, An Analysis of Canada’s Most Famous Holograph Will: How A Saskatchewan Farmer Scratched His Way into Legal History, 77 Sask. L. Rev. 1, 1-3, 17-18 (2014). He died the next day, but the fender was removed from the vehicle and admitted as a valid holograph. 43 Id. at 6, 21, 24. As his case elucidates, holographs are invaluable for testators “who are unable or unwilling to secure the assistance of counsel.” 44 In re Estate of Teubert, 298 S.E.2d 456, 460 (W. Va. 1982).

The conventional argument against recognizing holographs is that they breed litigation. For example, because holographs lack the safeguard of witnesses, they arguably invite fraud, forgery, and undue influence. 45 See Mark Glover, Minimizing Probate-Error Risk, 49 U. Mich. J.L. Reform 335, 381 n.213 (2016) (observing that holographs are supposedly “particularly susceptible to fraud and forgery”). However, the empirical evidence on this point is mixed. For example, in a soon-to-be-published work, one of us found that holographs were correlated with a statistically significant increase in the odds of litigation in just one of two neighboring California counties. 46 See Horton, supra note 35, at 31. And in an even more encouraging sign, Stephen Clowney surveyed 145 holographs from Pennsylvania and discovered that only 6, or 4 percent, led to a dispute. 47 Clowney, supra note 41, at 27, 28, 59. As Clowney observed, this figure was on the low end of the 2 percent to 10 percent litigation rates that other scholars have uncovered for formal wills. 48 See id. at 60-61. Thus, although deterring will contests is not an idle concern, holographs may be a risk worth taking.

B. Harmless Error

Similarly, states could pave the way for do-it-yourself wills by adopting the harmless-error doctrine. Harmless error, which originated in Australia in the 1970s, allows a judge to enforce a writing that does not comply with the Wills Act if there is clear and convincing evidence that the testator intended it to be effective. 49 For the classic article introducing harmless error to an American audience, see generally John H. Langbein, Excusing Harmless Errors in the Execution of Wills: A Report on Australia’s Tranquil Revolution in Probate Law, 87 Colum. L. Rev. 1 (1987). Although only eleven states have enacted the rule either partially or in its entirety, 50 See Cal. Prob. Code. § 6110(c)(2) (West 2020); Colo. Rev. Stat. § 15-11-503 (2019); Haw. Rev. Stat. Ann. § 560:2-503 (LexisNexis 2019); Mich. Comp. Laws Ann. § 700.2503 (West 2020); Mont. Code Ann. § 72-2-523 (2019); N.J. Rev. Stat. § 3B:3-3 (West 2020); Ohio Rev. Code Ann. § 2107.24 (LexisNexis 2019); Or. Rev. Stat. Ann. § 112.238 (West 2017) (amended 2019); S.D. Codified Laws § 29A-2-503 (2020); Utah Code Ann. § 75-2-503 (West 2020); Va. Code Ann. § 64.2-404 (2020). it is worth a second look in the age of Coronavirus.

Harmless error provides a safe harbor for self-directed testators. For example, in Estate of Stoker, Steven Stoker decided to change his will and revoke his trust. 51 122 Cal. Rptr. 3d 529, 532 (Ct. App. 2011). Because he was dyslexic, he asked a third party to handwrite a new will for him, which he signed. 52 Id. See Dukeminier & Sitkoff, supra note 3, at 222 (providing more detail about Stoker). This document violated the Wills Act: although two of Stoker’s friends saw him sign it, they neglected to sign their own names. 53 Stoker, 122 Cal. Rptr. 3d at 533. Moreover, because it was not in Stoker’s handwriting, it failed to be a valid holograph. Nevertheless, a California appellate court salvaged Stoker’s wishes by applying harmless error. 54 See id. at 536. Accordingly, the doctrine ensures that testators who lack access to professional estate planning will not doom their efforts through a minor misstep.

In addition, harmless error can be invaluable for testators who suddenly fall ill. For example, in the New Jersey case Estate of Anton Jr., William Anton asked a lawyer to draft a will because he was worried that his wife, whom he was in the process of divorcing, would inherit his property in intestacy. 55 In re Estate of Anton Jr., No. BER-P-335-15, 2015 WL 6085394, at *1 (N.J. Super. Ct. Ch. Div. Oct. 6, 2015). Anton received the document from his attorney, reviewed it, approved of it, and scheduled a meeting to sign it. 56 Id. Sadly, a few hours before this appointment, he died. 57 Id. Citing the abundant proof that Anton wanted the unsigned and unwitnessed document to be his will, the court admitted it to probate. 58 Id. at *3. Thus, in stark contrast to the Wills Act, harmless error effectuates the intent of testators who set the wheels of the will-making process in motion but are unable to finish the task.

Like holographs, harmless error has drawbacks. The Wills Act may generate harsh results, but it makes cases predictable. A document that lacks the testator’s signature or features a single witness can never be a valid will. Erasing these bright lines “might tend to encourage carelessness and breed litigation, or open up avenues for fraud.” 59 See, e.g., Adam J. Hirsch, Formalizing Gratuitous and Contractual Transfers: A Situational Theory, 91 Wash. U. L. Rev. 797, 829 (2014).

Yet states that remain hesitant to relax the Wills Act could at least take a baby step in that direction. Some jurisdictions, such as California, Colorado, and Virginia, have adopted “partial” harmless error rules. 60 Cal. Prob. Code § 6110(c)(2) (West 2020); Colo. Rev. Stat. § 15-11-503 (2019); Va. Code Ann. § 64.2-404 (2020). Rather than curing any failure to comply with the Wills Act, these doctrines only apply to glitches that relate to witnesses. Critically, though, they can validate instruments that are completely unwitnessed if there is forceful proof that a person intended a writing to be her will. Thus, by dispensing with the requirement that the testator find third parties who are willing and able to help, partial harmless error could facilitate will-making by people who are alone.

C. Electronic Wills

The most dramatic departure from the Wills Act is electronic-will legislation. Since 2017, Arizona, Florida, Indiana, and Nevada have passed statutes that permit e-wills. 61 Ariz. Rev. Stat. Ann. § 14-2518 (West 2020); Fla. Stat. Ann. § 732.522 (West 2020); Ind. Code Ann. § 29-1-21-4 (West 2020); Nev. Rev. Stat. Ann. § 133.085 (LexisNexis 2019). In the same vein, in 2019, the Uniform Law Commission approved a Uniform Electronic Wills Act (UEWA). 62 Unif. Elect. Wills Act (Unif. Law Comm’n 2019). Finally, California, New Hampshire, Texas, and Virginia are considering similar laws. 63 See Adam J. Hirsch, Technology Adrift: In Search of a Role for Electronic Wills, 61 B.C. L. Rev. 827, 846 & n.120 (2020). These new and proposed laws encourage will-making by allowing testators to memorialize their last wishes in emails, text messages, and word processing programs.

E-wills are controversial. Arguably, they fail to serve the functions of Wills Act formalities. For example, saving or digitally signing an electronic document lacks the gravitas necessary to impress upon the testator the legal significance of creating a will. 64 See David Horton, Tomorrow’s Inheritance: The Frontiers of Estate Planning Formalism, 58 B.C. L. Rev. 539, 573 (2017). Moreover, safeguarding electronic documents against fraud poses new challenges because digital files can be created, altered, and deleted by someone other than the testator. 65 See id. Indeed, as Adam Hirsch has quipped, the ease with which e-wills can be altered makes them “proof-proof.” 66 Hirsch, supra note 63, at 865 (emphasis omitted).

But even states that are not prepared to give e-wills a full-throated endorsement should consider borrowing one innovation from some of the existing laws. Nevada’s statute and the UEWA permit remote witnessing. Indeed, the Silver State’s e-will legislation defines “presence” to include people who are in “[d]ifferent physical locations but can communicate with each other by means of audio-video communication.” 67 Nev. Rev. Stat. Ann. § 133.088(1)(a)(2) (West 2019). Likewise, the UEWA allows testators and witnesses to be in the “electronic presence” of each other. 68 Unif. Elec. Wills Act § 5(a)(3)(A) (Unif. Law Comm’n 2019). This feature might by an elegant way to retain the protections of the witnessing requirement in periods of extreme isolation.

D. Interim Measures

Finally, at a bare minimum, state lawmakers could relax the Wills Act during the COVID-19 outbreak or similarly exigent circumstances. 69 For an eerily prescient pre-coronavirus article proposing that states recognize electronic wills in emergencies, see Hirsch, supra note 63, at 873-88.

As of early April 2020, several jurisdictions had already done so. Between March 26 and April 9, Governors of states including Connecticut, Illinois, Kansas, Michigan, New York, and Tennessee issued Executive Orders that temporarily allow witnesses to appear by audio-video technology rather than in person. 70 N.Y. Exec. Order No. 202.14 (Apr. 7, 2020),; see also Emergency Orders, supra note 34. Likewise, Colorado and North Dakota—which are the only jurisdictions that allow notaries to serve as substitutes for witnesses—had authorized remote notarization. 71 See Colo. Rev. Stat. § 15-11-502 (2019); N.D. Cent. Code § 30.1-08-02 (2019); Emergency Orders, supra note 34. Finally, on April 10, the District of Columbia Mayor approved a bill that authorized both remote witnessing and electronic wills during periods in which the Mayor has declared a public health emergency. 72 See COVID-19 Response Supplemental Emergency Amendment Act of 2020, No. 23-286, 67 D.C. Reg. 4178 (Apr. 10, 2020).

Although these measures are much better than nothing, they are also riddled with caveats and exceptions. In New York, for instance, the testator must show a photo ID to the witnesses if she is “not personally known to [them].” 73 N.Y. Exec. Order No. 202.14 (Apr. 7, 2020), In Kansas, the testator must transmit a copy of the document to the witnesses within one day of its execution. 74 See Kan. Exec. Order No. 20-20 (Apr. 9, 2020), In Connecticut, remote witnessing of wills must be completed under the supervision of an attorney. 75 See Conn. Exec. Order. No. 7Q (Mar. 30, 2020), It would be a shame if these efforts to soften the rough edges of the Wills Act ended up creating additional intent-thwarting formalities for people who desperately need estate plans. Further, temporary measures that eventually will be rescinded once COVID-19 subsides should not stand in place of permanent legislative reforms that make it easier to express one’s last wishes.


The field of trusts and estates is notorious for “resist[ing] modernity . . . successfully.” 76 Bruce H. Mann, Self-Proving Affidavits and Formalism in Wills Adjudication, 63 Wash. U. L.Q. 39, 39 (1985). However, the COVID-19 crisis has exposed the downsides of the Wills Act. To bring the law in line with our newly fraught times, states should relax the rigors of this centuries-old statute.

*David Horton is a Professor of Law and Chancellor’s Fellow, University of California, Davis, School of Law. Reid Kress Weisbord is a Professor of Law and Judge Norma L. Shapiro Scholar, Rutgers Law School.

The featured image “Last Will and Testament” by Ken Mayer is licensed under CC BY 2.0.