University of Virginia Library

IV

Texts written in analog media are naturally auto-historical: that is, whatever their putative referent, they carry more or less legible traces of the history of their being inscribed, a history which can be understood in relation to the history of other events. In that respect such texts are like stone, which has its place in the geologic record. But texts written in digital bits are essentially ahistorical: as an ephemeral patterning of electrons they lack a fixed relation to any historical moment. They are like water. In Walter Benjamin's terms, they renounce the "aura" of historical authenticity in favor of an easy access afforded by the perfect, fungible reproduction.[36] Electronic text is naturally synchronic; and only the artifice of experts can authenticate it by binding it to the history of passing moments, through such devices as "digital timestamping" based upon "hashing" the code.[37]

If in principle the fear of forgery, of electronically cooked books, is a reasonable fear, in practice most people don't give it much thought, because they leave it to the experts to worry about. And there have always been experts,

custodians of our concerns, in positions of power over our understandings and our welfare. Illiterate monarchs[38] had to trust literate secretaries, who practiced their craft in secret; and ordinary people rely on the arcane verbal skills of lawyers and accountants. The computer technicians who would guarantee, with secret and invisible code, the authenticity and freedom-fromtampering of our electronic documents, do not have a categorically greater power than such bureaucrats.[39] But the scale of control is different, and the thralldom of a king to his clerk may be less disturbing politically than the dependency of millions on the document-security systems and assurances of a few. In the era of electronic writing, both of Blackstone's textual criteria— durability and security—can be satisfied, but only by entrusting a new secretariat of digital experts. What they do and how they do it will be a mystery: their guild will be a mystery; and that will be the source of their power.[40]

On 30 June 2000 President Clinton advanced electronic commerce in the United States by signing the Electronic Signatures in Global and National Commerce Act (15 U.S.C. 7001). He signed the bill twice, once by passing a signature card over a scanner (an act confirmed by his keyboarding his password,

"Buddy"—after his dog), and then again by inscribing his name on paper with a felt pen.[41] This law took effect on 1 October 2000. It provides, "with respect to any transaction in or affecting interstate or foreign commerce," that

"[W]ills, codicils, or testamentary trusts" are specifically exempted from the operation of this law, as are certain other documents (468). The law leaves it to federal or state regulators "to specify performance standards to assure accuracy, record integrity, and accessibility of records"—without, however, "requir[ing] use of a particular type of software or hardware" (470).

The terms "accuracy" and "record integrity" conceivably overlap each other, and so do not square precisely with Blackstone's concepts of durability and security. They are not among the terms explained in the Definitions section of the act. "Record integrity" presumably includes security against malicious alteration, as well as against casual decay; and accuracy presupposes a kind of self-identity and durability; so both of Blackstone's criteria remain in play. However, in the electronic domain there is no reason to foreground them by opposing them, as Blackstone does in the material domain: durability and security are not at odds in cyberspace, where threats to durability and security arrive from all sides. The E-Sign Act (to call it by its user-friendly nickname) favors no particular medium: like Faust in his indifference to paper, parchment, brass, or stone, the new law is open to using any tool at hand: whatever works, so long as it is electronic. "The term `electronic signature' means"—very broadly—"an electronic sound, symbol, or process, attached to or logically associated with a contract or other record and executed or adopted by a person with the intent to sign the record" (422).

This openness to possibility, this refusal to specify a narrow formal and technical definition of what will count as an electronic signature, is consistent with the long-standing informality of the law of signatures in Great Britain and the United States. In this respect law differs from lore. Despite common lore, often enforced by clerks, about the need to conform recognizably to one's so-called "legal signature" (a lay concept without legal standing), one can put one's name to paper in any form—and indeed not only one's name but any graphic symbol at all—in an indefinite number of ways: it will count as a signature, so long as it was affixed with the intent to sign. The longstanding Uniform Commercial Code defines the word "signed" so as to include "any symbol executed or adopted by a party with present intention to

authenticate a writing."[43] The new definition of "electronic signature" is comparably broad.

A key difference, though, is that under the old regime of paper a signer would know experientially, and could choose, how she was signing. Under the new, electronic regime, others, more expert, will know better and make the choice for her beforehand, and will afterwards determine whether she has in fact signed. The market, and the experts, will decide how we best will sign our names in electrons. Similarly for the construction of the electronic text or record as a whole: it will say what the experts assure.[44]

When T. S. Eliot complimented John Donne for being "expert beyond experience" the preposition "beyond" marked an intensification. That is, Donne, "[w]ho found no substitute for sense, / To seize and clutch and penetrate," experienced more than the ordinary person, and more thoroughly, but in the same modality, through physical "sense."[45] His expertise, though hyper-experience, beyond the limits of ordinary experience, was still something that he experienced. But the expertise of a postmodern technician is "beyond" experience in a simpler, less rigorous way: it deals with a disembodied reality inaccessible to and unassessable by the laity.[46]

The problem is one of scale—"proper magnitude," as Aristotle explained in a different context: we can directly assess only what is not too large or too small.[47] Superstrings of galaxies and strings of DNA are alike outsized and beyond experience.[48] So are the carriers of bits and bytes. We must entrust them all to the experts.

But don't we trust experts all the time? The reader who buys a copy of King Lear trusts that the editor prepared the text in an expert way—if she thinks about the question at all. (If not, the trust lies even deeper.) The patient who accepts her physician's prescription trusts that it was framed with appropriate expertise. Furthemore, the facts and principles that inform both disciplines (textual criticism, medicine) are more or less available, should one care to investigate them: the reader can learn enough about the text of King Lear to second-guess her editor, or the patient can learn enough about her ailment and treatment to provide "informed consent."[49] Is the case not the same with electronic text?

It is true that a producer or consumer of electronic text can educate herself in the mysteries that would guarantee its authenticity, as a general matter. But she will hardly be in a position to examine the coded electronic arrangements that would secure a given transaction. Concerning the particulars she must take much on faith. And to participate at all she will have no choice in the matter, for the expertise of electronic experts is now a cultural given, not an option.

Contrast the expertise of the "handwriting expert," or "examiner of questioned documents," which over a century ago gained prestige and authority in courts of law.[50] On Jennifer Mnookin's analysis, testimonial expertise in the identification of handwriting has been, if not quite a legal fiction, a judicial construct; furthermore, recent decisions leave its future very much in doubt.[51] That is, the expertise of the handwriting expert is contingent, not

necessary: courts can make do without it. But it is impossible to imagine secure electronic inscription without imagining also the supervision (direct or indirect) of an electronic expert, whose procedures and actions are beyond ordinary inspection.

"Believe one who experienced it": "experto credite," urged Diomedes, who had personally experienced combat against the hero of the Aeneid (11.283). We, however, must believe our new secretary, the electronic expert, not for what she has experienced, which we cannot assess, but for her credentials, which we can, more or less.[52] Such essential trust, circular, would probably have disquieted Blackstone, an expert of the old school; but now, in the new millennium, regarding littera scripta, there is no alternative to it.

I've presented versions of this argument to several academic conferences over the last few years, and the reception has varied in instructive ways.[53] Sometimes objections have been raised to Blackstone's ideologically suspect preoccupation with questions of authority and property. However, once there was a quick, impatient response from a law professor who is expert in internet law. Though I can only paraphrase his response, the gist of what he said is clear in memory: "You don't have to worry," he said; "trust us."