Big Data Reveals Shakespeare Had a Co-Author
The playwright Christopher Marlowe, a contemporary of Shakespeare, will soon be designated as a co-author in three of the Bard's early plays.
For centuries, William Shakespeare got all the credit. But now, thanks in part to big data, scholars decided it was time to acknowledge the contributions of a man they say served as Shakespeare's co-author in three early works.
Christopher Marlowe will be listed as co-author of the three Henry VI plays in the New Oxford Shakespeare, due to be published in the coming weeks by the Oxford University Press.
The debate over whether Shakespeare was the sole author of all of his plays has raged since the 18th century. Marlowe, who was born in the same year as Shakespeare, was an established playwright and poet himself, and was also known to have at least influenced the famous English bard.
Some have argued Marlowe was actually the main writer of many of Shakspeare's plays, though most scholars reject that theory.
"You can say, well this character really seems like a Marlowe character or this scene really reads like a Marlowe scene. Those kinds of literary suspicions underlined arguments in the past," Gary Taylor, a Florida State University professor of English and general editor of the latest Complete Shakespeare volume, told Seeker. "But critics can disagree - that's why these debates have not been resolved all of these years."
Enter: big data.
To determine whether Marlowe's hand was behind passages in the "Henry VI" plays, scholars used three kinds of tests that analyze word use and word sequences within the text.
One test tracked the use of very common words such as "and" or "the" or "of." The frequency of these words appearing in a text can be linked to a unique identity based on previous known works. The effect is too subtle to be picked up by the human eye or ear, but algorithms are able to identify a clear pattern.
"If you're listening to what I'm saying, you can't actually count the prepositions that pass by you," Taylor explained. "It's something that is true of all of us, but it is unconscious and a person's unique use of these words can only be established by a lot of data."
Another test looked at the way more significant words, known as substance words, are used in combination within a play's text. Certain combinations of nouns, adjectives, verbs and adverbs can be linked to an individual writer.
A third analysis examined the order and grouping of the minor function words within the text and linked it to known works by Marlowe or Shakespeare.
"In some cases, we know that, say, Marlowe is likely to use two of these prepositions within five words of each other," said Taylor. "While Shakespeare is unlikely to use those two words within five words of each other."
The results from all three independent tests identified Marlowe as the author of various passages within the "Henry VI" plays, parts 1, 2 and 3. So in the new volume, Marlowe will be credited as a co-author for those plays.
"The fact that different tests were getting the same results in the same parts of the plays is the best kind of clue that it is, in fact, Marlowe," Taylor said.
This isn't the first time collaborators have been identified as working with Shakespeare. During Elizabethan times, play writing was often a collaborative process.
The New Oxford Shakespeare includes 44 plays, 17 of which are labeled as collaborative with other writers. Previous editions of Shakespeare's complete works identified eight of 39 plays as collaborative.
Taylor hastens to add, though, that even if Shakespeare occasionally teamed up with other playwrights to create some of his works, analysis suggests that Shakespeare was always at the helm as the dominant writer.
And, according to Taylor, Shakespeare's sole authorship of most of his best-known and most-loved works is not in doubt.
"All of the really really famous plays are, in terms of our research, clearly, entirely by Shakespeare," Taylor said. "'Julius Caesar,' 'A Midsummer's Night Dream,' 'Hamlet' - those plays are safe."
WATCH VIDEO: The Future of Big Data