by In Jersey City, a woman can sit behind a desk with a ten-dollar embosser. She checks a driver’s license, watches a name move across a page, signs her own name, and presses a raised circle into paper. The tool is cheap. The ceremony may take three minutes. Yet the result can travel through banks, registries, agencies, and courtrooms with a dignity that a far more sophisticated digital signature does not automatically possess. Strictly speaking, the squeeze does not defeat cryptography. Federal evidence law recognizes both lawfully acknowledged documents and properly certified electronic records or digitally identified copies as potentially self-authenticating.1 The difference is cultural and procedural. The notarial certificate arrives as a familiar legal form, already attached to a public officer’s accountable act. A digital signature may provide strong evidence that data remained intact and that a particular private key generated a signature. It does not, by itself, prove who controlled the hand behind the key, what that person understood, or who else was standing in the room.
The notary’s stamp matters because it is the residue of an encounter. Trust is not merely attached to a document. Trust is performed by bodies, under rules, before someone who has the authority to stop the performance.
The Stamp Is an Office, Not an Ornament
A notary does not certify that every sentence in a contract is true. Depending upon the act, she certifies a narrower and more important event: a person appeared, was identified, acknowledged a signature, swore to a statement, or signed in the officer’s presence. New Jersey’s own notary manual makes the human duty explicit. A notarial officer may refuse to proceed when not satisfied that the signer is competent or has capacity, or that the signature is knowingly and voluntarily made. The same manual requires an official stamp for electronic notarization, but the stamp may itself be electronic and must be independently verifiable, under the notary’s control, and tamper-evident.2 The object is therefore not sacred. The accountable office is. That distinction explains why a cheap seal can carry such weight. The state has delegated a small portion of its witnessing authority to an identifiable person. That person follows identification procedures, dates the act, records it in a journal or authorized equivalent, and risks sanction for abuse.2 The stamp compresses this institutional story into a mark.
Federal Rule of Evidence 902(8) gives an acknowledged document a direct route to authentication when the certificate was lawfully executed by a notary or another authorized officer. The document remains open to challenge. The rule merely removes the ordinary need for an additional witness to establish that this recognized form is what it purports to be. Rules 902(13) and 902(14) offer parallel routes for certified electronic processes and digitally identified copies.1 That procedural economy is the notary’s true power. The stamp is a portable presumption of performed order.
Human witnessing is not magic. A notary can be inattentive, deceived, or corrupt, and a trembling hand can be overlooked. The ceremony does not guarantee truth. It creates a place where responsibility can be assigned and facts investigated: a named officer, a date, a record, an identification method, and a duty honored or breached. A frictionless click may leave abundant logs but no responsible observer. The law’s preference is not sentimental faith in human goodness. It is the sober distribution of accountability.
Rome’s Public Memory
Rome had its tabelliones, although they should not be lazily equated with modern American notaries. They were professional drafters of legal instruments, distinct from the Roman notarii and other documentary personnel. By the age of Justinian, the tabellio had become part of a regulated documentary practice. Imperial law addressed both the legality of the instruments they prepared and the evidentiary force those instruments could receive.3 Justinian’s Novel 73, issued in 538, is especially revealing. When a public document was challenged, the tabellio could be required to appear and testify under oath. An assistant who wrote the instrument and another participating official might also be called. Witnesses mattered because the public instrument did not become trustworthy through parchment alone. Its credibility remained connected to a human chain of preparation, presence, recognition, and sworn responsibility.4 The ancient lesson is not that old law hated innovation. Roman administration converted land, debt, inheritance, status, and obligation into portable records. Yet it understood that a record becomes dangerous when severed from the conditions of its making. The tabellio stood between private intention and public consequence.
Our notary still occupies that threshold. She is not merely verifying a signature. She is turning an event into institutional memory.
What a Key Cannot Know
The vocabulary matters. E-SIGN defines an electronic signature broadly as an electronic sound, symbol, or process associated with a record and adopted with intent to sign. It can be far simpler than cryptography. A digital signature is a narrower cryptographic mechanism.5 Modern cryptography is magnificent at answering technical questions. The National Institute of Standards and Technology describes digital signatures as tools for detecting unauthorized modification, authenticating a claimed signatory, and supporting evidence that a signature was generated by that signatory.6 These are genuine achievements. A properly managed digital-signature system can establish document integrity more rigorously than ink, paper, or embossing ever could. Yet integrity is not consent. Authentication is not capacity. Attribution is not voluntariness.
A private key cannot tell whether its owner understood the instrument. It cannot see the adult child hovering behind an elderly parent. It cannot hear a caregiver supplying answers. It cannot notice sedation, confusion, panic, rehearsed speech, or the small delay before a coerced person says yes. It cannot distinguish a final act from an accidental click unless the surrounding system gives that click a disciplined meaning. This is not a criticism of cryptography. It is a criticism of category error. We keep asking machines that secure data to certify human freedom. The machine can establish that a key acted upon a file. The law must still decide whether a person acted through the key. No key exchange has peripheral vision.
The Will Behind the Wall
Congress enacted the Electronic Signatures in Global and National Commerce Act on June 30, 2000. E-SIGN did not proclaim that every electronic mark was trustworthy. It established a rule against discrimination: a signature, contract, or record may not be denied legal effect solely because it is electronic. The statute preserved other substantive legal requirements and did not generally compel private parties to accept electronic records.5
Then Congress built a wall around wills, codicils, and testamentary trusts. E-SIGN’s general rule does not apply to laws governing their creation and execution.7
That exception was not mere nostalgia for fountain pens. A disputed contract usually leaves living participants who can testify. A disputed will arrives after the one indispensable witness has died. Probate law therefore depends upon formalities that perform several functions at once. They create evidence. They warn the testator that this act is final in a special legal sense. They channel intention into a form a court can recognize. They protect against substitution, fraud, and pressure. The classic scholarship describes these as evidentiary, channeling, cautionary or ritual, and protective functions.8 A will-signing ceremony says more than, “This file came from this account.” It says that the testator understood the document, adopted it as a will, and acted before witnesses. This was not a draft, a joke, or a momentary wish.
Electronic wills are no longer theoretical. The Uniform Law Commission approved the Uniform Electronic Wills Act in 2019, offering states a framework for electronic execution.9 New Jersey, however, still illustrates the old wall. Its notary manual expressly limits remote communication technology for records governed by laws concerning wills and codicils, except for tangible records. As of August 18, 2026, Senate Bill 1336, which would authorize electronic wills, had been introduced and referred to the Senate Judiciary Committee but had not become law.10 The persistence of ceremony is often described as technological backwardness. That accusation is too easy. The serious question is not whether a will lives on paper or in a database. The question is whether the execution system preserves the human protections that paper ceremony accidentally bundled together. The correct reform is not to abolish ritual. It is to design a better ritual.
Why Medicine Still Hears the Dial Tone
Medicine’s attachment to the fax machine appears to be institutional embarrassment. Hospitals possess electronic records, portals, secure messaging, interfaces, and national exchange networks. Yet paper still enters a machine, crosses a telephone network, and emerges elsewhere as paper or an image. The absurdity is partly written into federal regulation. Under the HIPAA definitions, certain transmissions of paper by facsimile, and voice by telephone, are not treated as transmissions by electronic media when the information did not exist electronically immediately before transmission. The information remains protected health information, but the channel is classified through an old technical distinction. The HIPAA Privacy Rule permits treatment information to be sent by fax, provided reasonable safeguards are used, such as confirming the number and securing the receiving machine.11
This does not make fax secure. A mistyped number can disclose a record to a stranger. A tray can sit in a public corridor. Pages can vanish. The telephone line is not trusted because it is technically superior. It is trusted because institutions know the ritual. There is a number, a cover sheet, a sending confirmation, a destination, a queue, and a human clerk who knows what “faxed records” means. The numbers remain striking. In the 2025 American Hospital Association information-technology survey reported by the federal health IT office, 74 percent of responding hospitals said that they often or sometimes sent summary-of-care records by mail or fax, and 81 percent often or sometimes received them that way. Those figures combine mail and fax, so they do not measure fax alone, but they show that supposedly obsolete channels remain embedded in ordinary exchange.12
In March 2026, the Department of Health and Human Services finalized standards for electronic health-care claims attachments and related electronic signatures. The rule itself acknowledged that claims attachments had remained largely manual, frequently relying on fax, mail, or portal uploads. It was published on March 24, became effective on May 26, 2026, and requires compliance by May 26, 2028.13 Twenty-six years after E-SIGN, federal health administration was still standardizing a class of electronic signatures and attachments that paper and fax had continued to carry. The fax survives because interoperability is not only a technical problem. It is a trust problem. Two institutions may each possess excellent software and still lack a shared directory, shared workflow, shared vendor, shared certificate authority, or shared responsibility for failure. The fax number is crude, but it is socially routable. Medicine does not need more reverence for fax. It needs digital systems that are as universally legible and operationally accountable as fax, without fax’s insecurity and waste.
The Screen Rebuilds the Room
Remote notarization provides the most honest answer to the conflict between convenience and presence. New Jersey permits many notarial acts through communication technology, but it does not simply substitute a password for an appearance. The remote session must permit simultaneous communication by sight and sound. The notary must confirm identity through specified procedures, connect the signer to the precise record, create an audio-visual recording, and preserve that recording for ten years.14 This is the law confessing what trust requires. When the physical room disappears, the legal system tries to reconstruct the room.
The reconstruction is imperfect. A coercer can stand outside the camera’s narrow frame, while latency, illness, and platform design can hide cues visible across a table. Yet a recorded ceremony can preserve evidence that an unrecorded visit loses. The answer is not to romanticize paper. It is to combine human observation with technical integrity. For high-stakes acts, we should build ceremonial computing. The system should bind the exact document to a cryptographic hash, verify identity through independent means, require a live acknowledgment in ordinary language, give a responsible witness the duty and power to stop, preserve the event with the record, and maintain a durable chain of custody. Technology should strengthen the ceremony rather than erase it.
The Human Presence Test
Before trusting any high-stakes signature system, ask six questions:
-
Who established the signer’s identity, and by what independent evidence?
-
How did the signer express present intent to adopt this exact record?
-
Who was responsible for noticing incapacity, confusion, or coercion?
-
Did that person have the authority to pause or refuse the act?
-
Is the ceremony securely linked to the precise document that was signed?
-
Will the evidence remain readable, reviewable, and accountable when the transaction is later disputed?
A system that answers only the fifth question is a document-security system. It is not yet a trust system.
A Brief Timeline of Performed Trust
538: Justinian’s Novel 73 regulates proof involving documents prepared by tabelliones and contemplates testimony from the drafter and associated participants.
1941: Ashbel Gulliver and Catherine Tilson publish their classic analysis of the functions served by testamentary formalities.
1975: John H. Langbein develops the evidentiary, channeling, cautionary, and protective account while arguing against mechanical invalidation for harmless defects.
August 21, 1996: HIPAA becomes law, establishing the federal framework that later distinguishes electronic health information from certain paper-fax and voice transmissions.
June 30, 2000: E-SIGN becomes law. Its principal provisions take effect on October 1, 2000, while wills, codicils, and testamentary trusts remain outside the general federal rule.
2019: The Uniform Law Commission approves the Uniform Electronic Wills Act.
2021: New Jersey modernizes its notarial law to authorize electronic and remote acts while retaining live communication, identity proofing, recording, and official responsibility.
2025: Federally reported AHA survey data show mail or fax still used often or sometimes by large majorities of responding hospitals for sending and receiving summary-of-care records.
January 13, 2026: New Jersey Senate Bill 1336, the proposed Uniform Electronic Wills Act, is introduced and referred to the Senate Judiciary Committee.
March 24, 2026: HHS publishes its final rule for electronic claims attachments and signatures. The rule becomes effective May 26, 2026, with compliance due May 26, 2028.
The Squeeze
The notary’s stamp is not wiser than mathematics. It is not more secure than a well-designed cryptographic system. Its advantage is moral and institutional: someone is responsible for looking. That responsibility is small, fallible, and often performed in an ordinary office by an ordinary person with an inexpensive tool. Precisely for that reason, it is democratic. The law does not always require a priest, judge, physician, or engineer. Sometimes it asks a commissioned neighbor to witness the boundary between intention and consequence. We should not preserve paper merely because our ancestors touched paper. We should preserve the human functions that paper rituals carried: attention, pause, publicity, accountability, refusal, and memory. Then we should use digital tools to make those functions stronger, more accessible, and harder to corrupt.
A society that replaces witnesses with keys may gain mathematically elegant evidence that a file was unchanged while losing the evidence that matters most: whether the person was free. No key exchange has ever watched a hand tremble. The future of trustworthy signatures begins when our systems are designed to admit that they cannot.
Leave a Reply