The Shakespeare Wars

by Ron Rosenbaum

To which one wants to say, huh? His only evidence for this is that “plots” according to the OED could, at the time, denote “birthmarks, age-spots, or any relatively darker patch of skin,” but Foster has no doubt the verse means “the woman described in ‘Shall I Die?’ wears her breasts bare.” But even if one accepts Foster’s dubious assertion that “plots” equal “nipples,” why is it necessary to believe she publicly bares her “nipples, exposed by a low-cut bodice”? Why not just as easily read the verse as a lover describing his lady’s private nakedness in bed, or in his mind? Why? Because Foster conjures up, in his imagination, a nipple-baring dress that will put the fashion-date of the poem fifteen years later than Gary Taylor wants it.

  This tendentious exegesis suggests a failure of close reading that would eventually cause Foster’s downfall, because the scholarly refutation that led Foster to retract his thesis was founded on one demonstration after another of Foster’s failure to read closely the context of what he claimed were “Shakespearean parallels” in the “Funeral Elegy.” His reliance on counting word usage, rather than careful contextualizing.

  The counting—which Foster first did by hand—may explain Foster’s turn to the computer; he was substituting a silicon chip for a tin ear. He turned to the computer after his historical research into the “external evidence” had reached a dead end. It’s evident from his 1989 book, although never mentioned again, that Foster had a working theory about why Shakespeare would have come out of virtual retirement to write six hundred tedious lines of abstract and windy piety about someone he hardly seemed to know (on the evidence of the poem anyway, which basically says nothing more than that the deceased was good and his death was bad).

  Foster’s initial working theory, which survives in traces in his 1989 book, was that the dead man was actually a part-time actor in Shakespeare’s theatrical troupe.

  To this end the industrious Foster, whom one must credit for investigative energy, however misguided, dug up the 370-year-old Oxford “buttery” (dining hall) records for the dead man, William Peter, when he was an Oxford student, and used them to try to correlate Peter’s absences from the buttery with the dates when Shakespeare’s company was playing or traveling nearby.

  Nothing much came of this, nothing definitive anyway, and the state of the “external evidence” that linked the “W.S.” on the Elegy pamphlet and William Shakespeare (and the dead man William Peter) was limited to the initials’ similarity—and undermined by the meretricious use of the initials “W.S.” in previous cases, and the existence of several religiously inclined writers of memorial verse with the real initials “W.S.”

  Foster did, however, demonstrate that the dead man’s path had, on numerous occasions, crossed that of a John Ford at Oxford who seems to have been the same John Ford who later became a playwright in Shakespeare’s company. Ford had also gone through an intense devotional period (of which there is little evidence in Shakespeare’s life) around the time of the “Funeral Elegy,” in which Ford himself wrote funeral elegies and religious poems with titles like “Christe’s Bloodie Sweat.” (I even discovered a line in one of Ford’s lesser-known plays, Love’s Melancholy, that reads: “a funeral elegy of tears.”)

  Foster examined the possibility of Ford’s authorship of the Elegy in his 1989 book, A Study in Attribution, and admitted that this John Ford and his brother were more likely, based on the record, to have known the dead man in the “Funeral Elegy” than Shakespeare was. And while there were stylistic similarities and echoed phrases in both the “Funeral Elegy” and Ford’s work, Foster concluded, largely on the basis of statistical evidence, that it was more likely that Ford borrowed from “Shakespeare’s” “Funeral Elegy” for his own work than that he wrote it himself.

  DON FOSTER’S FATAL CLEOPATRA

  Too bad that, his quest for external evidence frustrated, Foster was led astray by the siren song of statistics and the newly fashionable “science” of “stylometrics.” Because he might have gotten it right the first time with Ford, although there was little glory to be gathered from identifying a relentlessly mediocre poem by Ford, who is mainly admired for such lively Jacobean plays as ’Tis Pity She’s a Whore. The “Funeral Elegy”—“ ’Tis Pity He’s Dead” you might call it—is sub-sub-standard Ford.

  But Foster’s fate was sealed when he turned to statistics. They were, as Samuel Johnson said of Shakespeare’s puns, “his fatal Cleopatra.” The “science” of the statistical analysis of literary style and identity, sometimes called “stylometrics” (although Foster unconvincingly claims that his methods are really not stylometrics), was then in its largely pre-digitized infancy and relied a lot on hand-counting prepositions and conjunctions (“function words”) to arrive at comparative percentages of “ifs” and “buts” in a known author—which were then matched with an as-yet-unattributed work.

  Statistics and stylometrics implicitly argued that style could be atomized into numbers, that one could at the very least define an author’s identity, his “fingerprints” by counting his characteristic percentages of word use, especially “function words” such as “and” or “but.”

  So poor Don Foster spends years hand-counting “ands” and “buts” in Shakespeare’s work as well as in the work of scores of forgotten funeral elegy writers, working without computerized help—which perhaps explains his overreliance on SHAXICON, his computerized database, once he got it up and running. He confesses, toward the end of A Study in Attribution:

  I am nonetheless left with the impression that my methods have been terribly rudimentary in some respects due to the lack of available tools. The word-counts present in Table 1.9 are the partial product of many tedious weeks counting “nots” and “buts” and “so’s” and “that’s” … in the process I discovered a maddening tendency to arrive at a different count for my several variables no matter how many times I tallied a particular poem, and am not certain even now, that my tables are perfectly accurate. (Italics mine.)

  A maddening tendency indeed. One has to feel, once one has glanced at Foster’s astonishingly complex tables, that it was this, this maddening slipperiness more than anything that led to his fatal infatuation with SHAXICON: anger, perhaps rage at the slipperiness of Shakespeare’s language, the words that somehow just wouldn’t stay counted precisely, almost a metaphor for the way Shakespeare’s words resisted, in a larger sense, reduction to a single meaning. No matter how many times Foster set them down “in the tables of his memory” as Hamlet put it, they would “dizzy the arithmetic of memory” as Hamlet also put it.

  So Foster’s Fall, like Hamlet’s, could be construed as a revenge plot gone wrong: he’d show them, these slippery uncountable words—he’d turn them into numbers and crunch them, crunch the hell out of them in his computer, and make them say what he wanted to hear. He devised a “mousetrap” that would “catch the conscience of the king,” catch the presence of the Bard.

  Staring at Foster’s tables, I thought of another set of maddening tables, one that I had come across in researching my Hitler book. They were the tables compiled by Hitler scholar Rudolph Binion, a Brandeis professor, to defend his theory of the origin of Hitler’s evil, the grail of the Hitler explainers.

  Binion’s theory focused on young Adolf’s reaction to his mother’s fatal breast cancer. Binion believed, from his study of the medical records of Hitler’s mother’s Jewish doctor, one Eduard Bloch, that Bloch had over-applied the standard ineffective remedy of the time—the caustic and searing chemical iodoform—to Hitler’s mother’s cancerous breast tumor, causing her more agony than necessary. Suffering which, Binion believed, Hitler blamed on the Jewish doctor but repressed, because unbearable, and later projected upon the Jewish people as a whole.

  Then another writer, a psychoanalyst named John Kafka (a distant relative of Franz, and also a descendant of Hitler’s Jewish doctor) disputed Binion’s calculation of Bloch’s use of iodoform. Binion went back to his archival researches—into the price in Austrian kroners and the st
andard number of grams and meters of Austrian iodoform gauze prescribed in 1908—and constructed a table similar in complexity to Foster’s.

  It looks incomprehensible and a bit silly, in the sense that it reduces the vast question of the genesis of Hitler’s evil to kroners-per-meter-of-gauze. But it is another instance of the persistent longing to believe that something as numinous and awesome and bottomless and inexplicable as Hitler’s evil or Shakespeare’s genius could somehow be captured by reducing it to numbers. Captured and somehow tamed, made less unbearable because less incomprehensible.

  One can see why this has more of a maddening than illuminating effect on those who seek elusive truths through numerical calculations. At first, in Foster’s case, it seemed to have a salutary inhibiting effect as well. Foster’s awareness of the imprecision of his counting, the incompleteness of his database, the impossibility of getting a perfect hand-count of all the “ifs,” “ands” and “buts” in all the funeral elegies composed in the decades before and after 1612, may have inhibited Foster from claiming a certainty for his Shakespearean attribution in his 1989 book.

  1995

  Enter SHAXICON. Now it was no more Mr. Nice Guy. No more shrinking violet. Foster’s initial claim was that SHAXICON was a digitized database of “rare words” in Shakespeare’s plays and other texts from that era. Rare words, in this definition, are words that appear in Shakespeare’s entire canon less than twelve times. In Foster’s theory this would allow him to compare the frequency of rare words in Shakespeare with the frequency of those rare words in a disputed text such as the “Funeral Elegy.” A test which Foster claims the “Funeral Elegy” passed with flying colors. Then Foster went on to try to use SHAXICON to tell us which speaking parts Shakespeare played as an actor in his own plays, by “showing” that the frequency of rare words that had appeared in earlier plays would sometimes spike in later plays. Foster’s theory was that, while Shakespeare was writing the later play, he was acting in the earlier play and this accounted for the spike in early-play rare words in the late plays he was writing. And the early-play rare words could point us to what role in the early plays Shakespeare might be working on. He called this particular category of rare words “Egeon words,” after Egeon the father of the twins in The Comedy of Errors, one of the “old man” parts Shakespeare was said (in one uncorroborated apocryphal report) to have specialized in playing himself.

  Few serious scholars have Foster’s confidence in his ascriptions of Shakespeare’s speaking parts. But for the media it was not so much what Foster did with his digitized database as the fact that he used one at all. It made such a seductive story: wonkish scholar finds “new” Shakespeare poem overlooked by old-fashioned professors with new-fashioned techno-geek computer program. A program with the sleek James Bondish name of SHAXICON. It played to the techno-optimism of the age. A new age has dawned when the numinous secrets of literature will be served up on a silicon platter by our powerful computers.

  Armed with SHAXICON, Foster now had the confidence, or chutzpah, to abandon his previous hedgings and to declare that he had proved the Elegy was by Shakespeare. The subsequent worldwide headlines obscured the fact that Foster never made public for examination what was in his SHAXICON database. He kept claiming, for some seven years, that he was about to make it publicly available, but recurrently gave the excuse that he was too busy with his very important, very high-profile criminal work, claiming to have invented the science of “forensic linguistics.” A claim that apparently impressed some folks at the FBI who, despite—or because of—repeated embarrassing failures of their self-promoting “profilers,” believed Foster’s techniques would add to their arsenal of “scientific” crime-detecting techniques.

  Alas for his future crime-fighting prospects, Foster told Caleb Crain of Lingua Franca in 1997, “All I need is to get one attribution wrong ever, and it will discredit me, not just as an expert witness in criminal cases, but in the academy as well” (italics mine).

  Still, for a time, SHAXICON served its purpose. In December 1995 at the Modern Language Association convention in Chicago, Foster and Professor Richard Abrams made an unequivocal declaration that Shakespeare wrote the “Funeral Elegy,” a claim Foster backed with the elusive SHAXICON and Abrams supported with his “close reading” which “proved” that the dullness of the poem was deliberate. That the poem was “an intimate document from the poet’s final years,” Abrams argued, in which the death of William Peter in that drunken brawl caused Shakespeare to “reconsider theatricality” itself because it might be a source of murderous passion (the drunken brawl).

  Stephen Booth would later characterize Abrams’s reading as an example of what others have called “The Fallacy of Imitative Form”:

  “William Peter, a dull plodding man who lived a dull plodding life, was a good man. The dull plodding Funerall Elegye is imitative of its dull plodding subject, and is, if looked at in that light, a good poem.”

  But Booth did, alas, title his talk on the Elegy “A Long Dull Poem by William Shakespeare” (italics mine).

  THE MARKETING OF THE SHAKESPEAREAN CLAIM

  Looking back on it, it was remarkable to see the apparent coordination of allies, publishers, scholarly publications and conferences that Don Foster marshaled to bolster his Shakespearean claim for the Elegy.

  Another indication that, like Gary Taylor, some contemporary Shakespearean scholars are scarcely shrinking violets, but are forced, perhaps because of the economics of the profession, to become warriors (or entrepreneurs) for their cause.

  In Counterfeiting Shakespeare, his book on the Elegy question, Brian Vickers, one of the first to call Foster’s “discovery” into question, calls it an instance of “the politics of attribution.” Vickers documents elements of the campaign Foster waged for the Elegy, in particular his triumphalist efforts to delegitimize dissent.

  Politics is one way to characterize it. Marketing is another. In both cases the selling of the Elegy depended on Foster’s getting the most validation and canonization for the Elegy with the least possible skeptical scrutiny.

  The first target was the academic press. According to Vickers’s account Foster and Abrams made their decision to make a definitive claim for the “Funeral Elegy” after Foster was persuaded by Abrams to become more bold and forthcoming in April of 1994, at a meeting of the Shakespeare Association of America, and they both began to prepare papers for the Publications of the Modern Language Association (PMLA) from their statistical and “close reading” perspectives respectively.

  Vickers believed that it was the prospective entry of the British Norton Shakespeare into the American market at this moment that was crucial. The Norton decision to include the “Funeral Elegy” in the American edition, even though the edition distanced itself by describing it as “a poem … that raises important questions about the attribution of works to Shakespeare,” was the key to setting off a kind of marketing feeding frenzy over the Elegy among the other two American Complete Works that were also about to go to press with new editions: the Riverside and the Addison-Wesley.

  Once Foster got the Elegy under the tent, so to speak, in the guise of a poem “raising questions” about attribution in the Norton, the other Complete Works followed suit and included it.

  The Norton edition had a competitive edge because it engaged Foster himself to edit and introduce “his” Elegy for them. The editors of the rival editions were clearly not happy with their publisher’s decision to include the “long dull poem” in their editions as a “work of Shakespeare,” but, it seems, publishing and marketing people craved “The New Shakespeare Poem.” What resulted in both cases was a kind of inclusion hedged so strongly by its editors, it verged on an attempt to quarantine the Elegy, if not reject it entirely. As David Bevington, the respected University of Chicago scholar and editor of the Addison-Wesley edition concluded, after summarizing Foster’s case for the Elegy:

  To skeptics, on the other hand, A Funeral Elegy remains too piously conventi
onal in its treatment of the great themes of slander, death, and immortality through poetry, to be attributable to Shakespeare even as an occasion piece. Not all tests of vocabulary uniformly endorse his authorship. The attribution remains uncertain.

  But the poem entered the edition anyway. Similarly with the Houghton-Mifflin Riverside edition in which the poem’s editor, J. J. M. Tobin, also made clear that he thought there was something missing from the poem, something one could define as Shakespearean. He defined what was missing as “the philosophical tolerance and psychological profundity that we expect in Shakespeare’s work.” Take that, marketing department!

  But the damage had been done. The triple inclusion allowed Foster to make the inflated claim that his Elegy had been “accepted by the three major scholarly publishers of Complete Works editions.” For Foster, evidently, “accepted” was a flexible enough term to cover these emphatically unenthusiastic and distanced inclusions.

  And then came the headlines, first in the Chicago Sun-Times, then The New York Times, then the world. Despite widespread acceptance in the media, especially in America, Foster likes to portray the reception of his Elegy as deeply embattled. Indeed the first sentence of his retraction in 2002 reads, “In 1996, having ventured an attribution of W.S.’s A Funeral Elegy I was blasted in the pages of the TLS.” And in fact in the TLS, Stanley Wells (who had been himself the subject of attack by Foster for his endorsement of the inclusion of “Shall I Die?” in the Oxford edition) did have his (well-reasoned) revenge by heaping contempt on the attribution of the Elegy. As did several other U.K. scholars (such as Brian Vickers and Katherine Duncan-Jones) with no evident axe to grind other than their common horror at the idea that a soulless computer program identified this particularly soulless poem (on ostensibly soulful subjects) as “Shakespearean.”