William Shakespeare. Big fan of his.

You’ve got to love reading also-ran Elizabethan dramatists. Sometimes, if you’ve huffed too much cynicism about Great Men in the Literary Canon, you can start to convince yourself that Shakespeare was just in the right place at the right time. Then you read some Henry Howard, and think, “Nope, nope, one of these things is definitely not like the other!”

And while your blogger enjoyed indirectly summarizing her experience of a first-year seminar long ago, this is actually a topical and timely opener, because Petr Plecháč at the Czech Academy of Sciences in Prague has just used machine learning to suss out exactly where Shakespeare’s work begins and ends in the cobbled-together manuscript of Henry VIII.

Scholars have long believed that large portions of Henry VIII are the work of John Fletcher, a prolific and famous playwright in the 17^th century.

The good old-fashioned way of proving such a thing was identifying verbal tics. As the MIT Technology Review puts it:

Fletcher often writes ye instead of you, and ’em instead of them. He also tended to add the word sir or still or next to a standard pentameter line to create an extra sixth syllable.

Ah, so delightfully mortal and non-genius. Were we forced to blog in iambic pentameter, we almost certainly would be filling out stray iambs with a “huh” or a “yep” here and there.

Not satisfied with simply identifying the clunky and less good stuff as “not Shakespeare,” Plecháč got computers to do exactly what the human brain once did, only with an, uh, well with an algorithm. (Starting to strongly suspect that as laypeople, we overuse and generalize the term “algorithm.” Anyone?)

They continue:

The technique uses a body of the author’s work to train the algorithm and a different, smaller body of work to test it on.

Apparently, Plecháč’s algorithm was able to identify where Fletcher ended and Shakespeare began to the line.

This is an exciting new step for literature. Machine learning has been used on visual art to such an extent that computer scientists are already able to learn a visual style and then apply it to other photos. Thankfully, the un-bylined writer at Technology Review jumped to the next logical question:

Might it be possible to transform an essay, or indeed an article for MIT Technology Review, into the style of Shakespeare or John Fletcher, for example?

A John Fletcher literary filter? We tremble with anticipation!