First Rowling, then Shakespeare… who’s next?

The Times today has an article about the possibility that Shakespeare wrote a passage in an edition of The Spanish Tragedy, an early Elizabethan play by Thomas Kyd.  The original computer analysis (by Brian Vickers) was very similar to that used to suggest that J. K. Rowling was the author of The Cuckoo’s Calling, which we talked about here.  Big Think describes what Vickers did:

Sir Brian has employed software called Pl@giarism–a free program developed by Maastricht University to catch law students cheating on their written work–to search a database of the 58 different plays performed in London between 1580 and 1595. But Sir Brian isn’t looking to catch anyone cheating. Rather, he is looking for examples of so-called “self-plagiarism.” The Pl@giarism software identifies every occasion that a sequence of three words appears in Shakespeare’s known works, and then looks for repetitions of these sequences in an unattributed text. Some of these word sequences are common, everyday collocations such as “by the way” or “Yes, my lord.”

Excluding those phrases, Sir Brian focuses on word sequences that are unique to Shakespeare. For instance, the word sequence “eyebrows jutty over” appears only twice in all of Elizabethan drama. One instance is in Shakespeare’s Henry V, written in approximately 1599. The only other instance is found in the fourth edition of Thomas Kyd’s “The Spanish Tragedy,” published in 1602. This version contains additions to five scenes, totaling 320 lines. In these short passages, Sir Brian found 46 collocation matches that are completely unique to Shakespeare’s poems and plays written before 1596. That evidence is hard to argue with.

What got the Times’ attention was another paper that focuses on Shakespeare’s handwriting and how that helps explain oddities in the passage:

In a terse four-page paper, to be published in the September issue of the journal Notes and Queries, Douglas Bruster argues that various idiosyncratic features of the Additional Passages — including some awkward lines that have struck some doubters as distinctly sub-Shakespearean — may be explained as print shop misreadings of Shakespeare’s penmanship.

“What we’ve got here isn’t bad writing, but bad handwriting,” Mr. Bruster said in a telephone interview.

What I couldn’t find in a cursory Google search was the actual passage in question.  It’s easy enough, though, to find the standard sample of Shakespeare’s messy handwriting–the passage from the manuscript of the play Sir Thomas More that is generally agreed to be by Shakespeare:


Next time, should J. K. Rowling disguise her writing style?

Here’s an interesting interview on Science Friday with Patrick Juola, the guy who’s computerized analysis helped identify J. K. Rowling as the author of the mystery The Cuckoo’s Calling.  He gives much more detail about exactly what kind of analysis he did over at Language Log.  Essentially, he compared the novel to works by Rowling, P.D. James, Ruth Rendell, and Val McDermid on four linguistic variables: distribution of word lengths, use of the 100 most common English words, and two other tests based on authorial vocabulary.

So, the final score? The results look “mixed,” but pointing strongly to Rowlng. There were certainly a couple of likely losers: nothing at all pointed to Rendell as a possible author, and only one test, and an unreliable one at that, suggested James. McDermid could be a reasonable candidate author, but the word length distribution seemed almost entirely uncharacteristic of her. The only person consistently suggested by every analysis was Rowling, who showed up as the winner or the runner-up in each instance.

One of the comments to Juola’s Language Log post suggests that a determined author can defeat analyses like these.  This is referred to as “adversarial stylometry.”  There are two basic approaches: obfuscation, where you try to simply hide your own style, and imitation, where you try to copy someone else’s style.  (A third approach is machine translation, where you translate an original passage using machine translation services.)  I doubt that any of this is worth Rowling’s time, but you might consider it if, say, you’re a whistleblower who wants to remain anonymous.

Of course, all of this analysis is overshadowed by the Onion’s shocking revelation that J.K. Rowling’s books were really written by Newt Gingrich:

“Assuming a fake identity really gave me a lot of freedom to build out the world of Hogwarts and flesh out the characters without drawing unwanted attention to myself or having the novels associated in any way with my political career,” Gingrich said in a statement, confirming reports he wrote the first four books in the fantasy series while still in office, but wrote the remainder before his 2012 presidential run.

Why do people rely on anything besides the Onion for their news?