The Radical Argument of the New Oxford Shakespeare

A new edition gives Christopher Marlowe credit as a co-author of “Henry VI,” Parts 1, 2, and 3.GL ARCHIVE / ALAMY

In 1989, a young professor named Gary Taylor published “Reinventing Shakespeare,” in which he argued that Shakespeare’s unrivalled literary status derives less from the sheer greatness of his plays than from the cultural institutions that have mythologized the Bard, elevating him above equally talented Renaissance playwrights. “Shakespeare was a star, but never the only one in our galaxy,” Taylor wrote. The book was his second major attempt to counter the view of Shakespeare as a singular genius; a few years earlier, he had served as one of two general editors of the Oxford Shakespeare, which credited co-authors for five of Shakespeare’s plays. In “Reinventing Shakespeare,” Taylor wrote that the Oxford Shakespeare “repeatedly shocks its readers, and knows that it will.”

Late last year, Taylor shocked readers once again. The New Oxford Shakespeare, for which Taylor serves as lead general editor, is the first edition of the plays to credit Christopher Marlowe as a co-author of Shakespeare’s “Henry VI,” Parts 1, 2, and 3. It lists co-authors for fourteen other plays as well, ushering a host of playwrights—Thomas Nashe, George Peele, Thomas Heywood, Ben Jonson, George Wilkins, Thomas Middleton, and John Fletcher, along with Marlowe—into the big tent of the complete works. This past fall, headlines around the world trumpeted the Marlowe-Shakespeare connection, and spotlighted the editors’ methodology: computer-aided analysis of linguistic patterns across databases of early modern plays. “Shakespeare has now fully entered the era of Big Data,” Taylor announced in a press release.

It’s no longer controversial to give other authors a share in Shakespeare’s plays—not because he was a front for an aristocrat, as conspiracy theorists since the Victorian era have proposed, but because scholars have come to recognize that writing a play in the sixteenth century was a bit like writing a screenplay today, with many hands revising a company’s product. The New Oxford Shakespeare claims that its algorithms can tease out the work of individual hands—a possibility, although there are reasons to challenge its computational methods. But there is a deeper argument made by the edition that is both less definitive and more interesting. It’s not just that Shakespeare collaborated with other playwrights, and it’s not just that Shakespeare was one of a number of great Renaissance writers whose fame he outstripped in the ensuing centuries. It’s that the canonization of Shakespeare has made his way of telling stories—especially his monarch-centered view of history—seem like the norm to us, when there are other ways of telling stories, and other ways of staging history, that other playwrights did better. If Shakespeare worshippers have told one story in order to discredit his contemporary rivals, the New Oxford is telling a story that aims to give the credit back.

In the past few decades, scholars have settled on a standard narrative of Shakespeare’s professional life: he gets to London by the early fifteen-nineties and learns the business by collaborating, perhaps on “Titus Andronicus” and the “Henry VI” plays; he becomes the in-house dramatist for the Lord Chamberlain’s Men and the King’s Men, and he writes, alone, his masterpieces; then he passes the baton, around 1613, to a new crop of up-and-comers, such as John Fletcher, with whom he works on the plays that ease him into retirement (“Henry VIII,” “The Two Noble Kinsmen”). The New Oxford attributions unsettle this story. Taylor’s team identifies Shakespeare as a collaborator well into the middle of his career, proposing that he co-wrote an early draft of the tragedy “Sejanus” with Ben Jonson and served as a script doctor for Thomas Kyd’s “The Spanish Tragedy” and the anonymous play “Sir Thomas More” right around the time he was writing “Hamlet”—and that Middleton revised several of the plays Shakespeare wrote around “King Lear.” The most surprising claim is that Marlowe—the bad boy of the English Renaissance, an alleged spy, heretic, and homosexual, the star playwright of the London stage until he was stabbed to death in a tavern brawl in 1593, just as Shakespeare’s career was starting to take off—collaborated on the scripts that popularized the English history play.

Scholars have long seen Marlowe and Shakespeare as admiring rivals. Shakespeare may have emulated Marlowe’s charismatic overreachers, Tamburlaine and Faustus, in “Richard III,” and taken up Marlowe’s “Jew of Malta” in “The Merchant of Venice,” another dark comedy with a villainous stage Jew who exposes the hypocrisy of the Christians around him. Marlowe, in turn, may have looked to Shakespeare’s history plays as a model for his own royal tragedy, “Edward II.” But the New Oxford’s claim that Marlowe and Shakespeare actually collaborated on the “Henry VI” plays is far from mainstream. And the idea that statistical analysis is behind the conclusion does not inspire universal confidence.

Scholars have been trying to quantify and tabulate Shakespeare’s style for more than a century. In 1901, a meteorologist in Ohio named T. C. Mendenhall hired two women to count the number of letters in every word of the works of Shakespeare and several other famous writers. (Mendenhall was funded by a wealthy Bostonian who wanted to test the hypothesis that Francis Bacon was the true author of Shakespeare’s plays.) Contemporary stylometric methods are more sophisticated, but the difference is one of degree, rather than kind. Gary Taylor and Stanley Wells’s “Textual Companion to the Oxford Shakespeare,” from 1997, relied heavily on tabulating the relative frequency of linguistic features such as “function words”—the little bits of sentence glue like “by,” “so,” and “from,” which seem to be used unconsciously and are thus deemed unlikely to reflect deliberate imitation of another author. The curious result is that Shakespeare’s fingerprints appear less in the lyricism of a line like “Lilies that fester smell far worse than weeds” (which has been used to link the anonymous play “Edward III” to Shakespeare’s sonnets) than in his subliminal preference for “you” and “them,” which allegedly differentiates his writing from John Fletcher’s more frequent use of “ye” and “em.”

Stylometrics has had its share of high-profile flameouts. In the nineteen-nineties, Donald Foster made the front page of the Times by claiming that his SHAXICON database could identify Shakespeare as the author of a little-known funeral elegy, which was subsequently included in several editions of the complete works. Then a French scholar, using old-fashioned close reading, showed that the funeral elegy more closely resembled the style of another seventeenth-century writer, John Ford, and Foster conceded his mistake. The first Oxford Shakespeare included a poem beginning “Shall I die? Shall I fly?” that Taylor found in a manuscript collection bearing Shakespeare’s name. He bolstered the attribution with stylistic parallels to the rest of Shakespeare’s work, but the collection is dubious, the parallels are thin, and very few scholars today accept the poem as Shakespeare’s. Taylor also once ruled out Marlowe as a co-author for the “Henry VI” plays; now he says he based that judgment on evidence that has since been disproved. “If you’re an empiricist, when you get new data, you change your mind,” he told me. “Unlike politicians, it’s a good thing for a scholar to be a flip-flopper.”

For the Marlowe attributions in the New Oxford, Taylor told me that the editorial board was convinced by an article that appears in the current issue of Shakespeare Quarterly. The article examines how likely it is that a playwright will use one function word near another, a group of tendencies they dub a “word-adjacency network.” In a line like “With mirth in funeral and with dirge in marriage,” from “Hamlet,” the decisive words are “and with,” a combination of function words more often found in Shakespeare’s known plays than in, say, Thomas Dekker’s. Marlowe’s characteristic verbal strings appear in a few scenes from each of the “Henry VI” plays. The authors grant that their method isn’t perfect; on a validation test, it misattributed several plays whose authorship isn’t in dispute, and there are playwrights, such as Thomas Nashe, who didn’t leave behind enough single-authored plays to generate a meaningful word-adjacency network. And in order to establish a baseline for each playwright to test uncertain cases, you need to assume that you know which plays Shakespeare and Marlowe wrote alone, an assumption that the New Oxford is designed to question. The fact that some other scholars have also attributed the same “Henry VI” scenes to Marlowe, however, gives the authors confidence: “The presence of Marlowe in these plays … is now undeniable.”

That attribution might stick, but it’s unlikely to convert stylometric skeptics. For Emma Smith, a professor whose work with Laurie Maguire is cited in the New Oxford press release, Taylor’s attachment to Big Data is part of a larger attempt to claim the prestige of science for Shakespeare scholarship. “We have always wanted English studies to look more scientific and more objective,” she told me. “But there are really just stories we tell, and some are more convincing than others.” Crediting Marlowe as a co-author allows Taylor to tell a different story about Shakespeare—one that might help us reimagine the shape of history.

Taylor has been accused of hating Shakespeare, which isn’t true. He admires Shakespeare. But he thinks that our veneration of the playwright’s genius has blinded us to the brilliance of other writers, such as Marlowe and Middleton—and to the political alternatives they envisioned. “Shakespeare’s favourite subjects are monarchy, monogamy, and monotheism; not coincidentally, his most famous speeches and sonnets are monologues,” Taylor and a co-author, the performance scholar Terri Bourus, write in the volume’s opening essay. It’s no coincidence that the essay is co-written: the New Oxford lists twenty-five editors and board members, and, rather than include single-authored scholarly introductions (or “critical monologues,” as they call them) for each play, this edition offers snippets from a range of critical responses. “You might think of this as tapas Shakespeare,” the general editors’ preface suggests. This approach upsets conventions that the editors associate with Shakespeare himself. Shakespeare, they argue, “specializes in one-and-onliness.” Taylor, on the other hand, believes “in a democracy of readers who are not being told what they have to like,” he told me over the phone.

For a long time, scholars used Shakespeare’s uniqueness to devalue other playwrights. When critics found a clumsy passage or inconsistency in one of Shakespeare’s plays, especially if it came toward the beginning or end of his career, an attractive solution was to blame it on an inferior co-author. Shakespeare’s sanctity was reinforced by Biblical terms used to distinguish his legitimate creation—“the Canon”—from doubtful material—“the Apocrypha.” The complementary move was to identify superior passages from other people’s plays as Shakespeare’s own. A powerful scene from the anonymous manuscript of “Sir Thomas More,” for instance, has been matched in certain features of its handwriting to Shakespeare’s six extant signatures. It’s a slender basis for attribution, but the allure of finding a Shakespeare manuscript is compelling, as is the desire to associate Shakespeare with a liberal passage where Thomas More refutes an anti-immigrant mob. Is it good because it’s Shakespeare, or do we want it to be Shakespeare because it’s so good?

This assumption made it common for editors to palm off parts of the “Henry VI” plays, which have often been deemed crude and faulty, on other playwrights, like Thomas Nashe or George Peele. But it also makes the New Oxford’s attribution to Christopher Marlowe an odd fit. “You could have Peele, Nashe, anybody else as a dumping ground,” Eric Rasmussen, who co-edited the R.S.C. editions of Shakespeare’s complete works and his collaborative plays, told me over the phone. “But to say Marlowe, who’s almost never anything but genius—that’s really kind of surprising and energizingly cool.” Rasmussen, who also co-edited an Oxford edition of Marlowe’s plays, said that he didn’t think there was any real evidence to support Marlowe as a co-author on the “Henry VI”s, but he credited Gary Taylor with divorcing claims about authorship from judgments about artistic merit. “When he was championing ‘Shall I die,’ everybody said, ‘But this sucks! If it’s Shakespeare, it’s supposed to be good, and this is aesthetically bad.’ And that’s just wrong! You can go to any number of plays and find aesthetically bad things that we attribute to Shakespeare.”

For Taylor, that might be the point of changing the story. If we see Shakespeare as but one star among many, that could open us up to other models of literary value, instead of assuming that Renaissance playwrights are bad if they’re not like Shakespeare. For a couple of decades, Taylor has championed Middleton as “our other Shakespeare.” Marlowe, a sexier star, may now look like a better bet, as the barrage of recent press coverage confirms. And Marlowe’s possible contributions to the “Henry VI” plays, in particular, suggest an alternate way to tell history.

The New Oxford refers to “Henry VI,” Part 2, by its earlier quarto title, “The First Part of the Contention of the Two Famous Houses of York and Lancaster,” claiming that “The First Part” only became a Part 2 when Shakespeare reworked a prequel a few years later. The theory is that Shakespeare decided to organize his history plays as a nation-building sequence with monarchs in the titles, which is how they were subsequently printed in the First Folio. (The so-called first tetralogy covers the Wars of the Roses through the three parts of “Henry VI” and “Richard III,” while the second tells the civil-war backstory from “Richard II” through “Henry V.”) “If you look at ‘Henry VI,’ Part 2, as a free-standing play, written in collaboration, you see a different notion of what a history play can and should be,” Taylor told me.

That notion, which Taylor’s team identifies with Marlowe, focusses less on male monarchs than on powerful female characters, like Joan of Arc in Part 1, and on rebellious commoners, like the fearsome Jack Cade in Part 2. (It’s Cade’s henchman who says, “The first thing we do, let’s kill all the lawyers.”) “Shakespeare didn’t invent the history play; he reinvented it and gave it a much more monarchical, great-man emphasis,” Taylor said. (The idea echoes the story Taylor told in 1989, about Shakespeare himself being reinvented as the monarch of drama.) “The first tetralogy is misunderstood because people think it’s trying to be like the second but not doing it right,” Taylor explained. “Once you realize that there are two visions of politics duking it out in the ‘Henry VI’s, especially in Part 2, they become much more interesting.”

Taylor thinks these competing visions of politics have implications beyond “Henry VI,” Part 2. “There’s not only a political civil war going on in the play but an aesthetic civil war going on. ‘The First Part of the Contention’ is the first great English history play. Both Marlowe and Shakespeare are young geniuses; both are sons of craftsmen, who did not grow up in the metropolis or around court, and both are clearly ambitious. It’s fascinating to see two brilliant but very different writers imagine what this new form of drama could be like,” he continued. “One of the ways to tell the story is that Shakespeare wins the competition because Marlowe gets assassinated. At the start of his career, it’s by no means clear that Shakespeare is the greater writer. If Marlowe had an additional twenty years, we could imagine him providing very different canonical models of histories, tragedies, and comedies.”

The Marlovian model, as Taylor sees it, asks us to examine the role of literature in politics. There’s a scene in “The First Part of the Contention” where the rebel Jack Cade beheads a lord for printing books and setting up a grammar school to teach young men to read instead of leaving them to tally their business accounts. The New Oxford attributes this scene to Marlowe, and Taylor thinks it raises a timely question about the relationship between books and history, about whether a society should value imaginative reading. “There were clearly people in Marlowe’s world who answered that question violently,” he said, “and there still are.”