Mathematical Artifacts: "Shakespeare and Co.: Christopher Marlowe, Thomas Dekker, Ben Jonson, Thomas Middleton, John Fletcher and the Other Players in His Story" by Stanley Wells
As a Shakespeare dilettante, I find some of the attributions regarding collaborations slightly worrying. I'm not quite sure why this has been worthy of research. One of the more risible of 'evidence' put forward, I forget where, was that Middleton was co-author of “All's Well That Ends Well” (incidentally Wells also professes this attribution). The argument was: 'As an example, the word "ruttish" appears in the play, meaning lustful - and its only other usage at that time is in a work by Middleton' or something to that effect. So, creative writers are supposed never to have used a word only once in their entire oeuvre? This is quite typical of academics who have no idea how creative writers - and particularly dramatists - work. But the most preposterous of all must surely be their citing of the stage direction 'all': '"All" (preferred by Middleton) only occurs twice in the Folio - both times in All’s Well.' Playwrights were writing their plays on the hoof to impossible deadlines. Stylometric analysis is a method which has been seriously challenged and is evidently flawed because it takes no account of how writers write. Only a few obsessives really care, those of us who can bring ourselves to watch Shakespeare, generally just enjoy and don't really worry about whether he might have had assistance from this or that writer. We know he collaborated as a matter of habit, so one for the historians to mull over, the rest of us will focus on what is best, the often-astounding dialogue...
Statistics is a very dangerous tool for someone to use who is not experienced with the kind of mathematical artifacts which can be produced in complex analyses. It is VERY easy to amend the modelling parameters slightly to produce the answer you are hoping for, and few people will ever delve into the workings of a complex statistical algorithm to see whether the weights put on different variables are justifiable or not. In practice, skilled English professors are not going to have the mathematical experience to challenge the findings.
If you're into Shakespeare, read on.