• Previous message: Hardy M. Cook: "Solid Flesh Once More"
• Messages sorted by: [ date ] [ thread ] [ subject ]

The Shakespeare Conference: SHK 19.0193  Friday, 28 March 2008

From:		Ward Elliott <ward.elliott@claremontmckenna.edu>
Date:		Friday, 14 Mar 2008 12:58:16 -0700
Subject: 19.0176 Shakespeare's Style
Comment:	RE: SHK 19.0176 Shakespeare's Style

Elliott Stone [SHK 19.0176] notes our skepticism of the handy assumption 
that co-authors neatly divide their contributions by act and scene and 
wonders how it might affect attributions of Titus Andronicus. Great 
question! The short answer, not surprisingly, is that, in some cases, it 
makes it harder to make clear ascriptions scene by scene, but maybe 
easier to make sense of scenes with ambiguous or conflicting indicators.

We've been working for more than two years at applying our new-optics 
methods to co-authored Shakespeare plays, typically analyzing passages 
of about 1,500 words, which should be long enough to get our expected 
accuracy in distinguishing single-authored texts to 95% or better. 
We're still working on these and have only published some of our 
results. Our current results on Titus are broadly consistent with the 
old-optics consensus, magisterially described by Sir Brian Vickers in 
Chapter 3 of his Shakespeare, Co-Author (2002):  of the play's 13 
blocks, all nine of the old-optics "Shakespeare" blocks are what we 
cautiously call "Shakespeare could-be's," fitting more or less snugly 
into our Shakespeare test profiles. Our results support the consensus on 
these. Two of the old-optics "Peele" blocks, 1.01.1-257, and 4.01, look 
like "Shakespeare couldn't-be's" by our tests, again supporting the 
consensus. In other words, we are 85% in agreement with the old-optics 
consensus.

But two of the consensus "Peele" blocks - 1.01.258-end and 2.01-.02 -- 
look like Shakespeare could-be's on our regular tests, and still look to 
us like Shakespeare could-be's even after trying a few more new-optics 
tests. For these, it's not so clear that the old consensus is right. 
Some combination of further new-optics testing on our part and separate 
old-optics testing of the suspect blocks might help clear this up. We do 
get about 5% false positives from known non-Shakespeare blocks of this 
size, and most of the old-optics results have been studied and presented 
for all the "Peele" and "Shakespeare" blocks in aggregates, not just the 
two blocks in question. Focusing on them separately might help. But it's 
also entirely possible that further testing will not unmuddy the waters 
on these two blocks and leave conflicting evidence as to who could have 
written them. If so, it could be taken as an indicator that either 
old-optics or  new-optics tests still need further work-or that the 
muddied picture lies not so much in the optics as in the authorship 
itself. What might be a great mystery if you assumed that co-authorship, 
which can easily be found in whole plays, could never be found in 
subsections of plays, would be much less mysterious if you did not make 
such an assumption.

Yours,
Ward Elliott

_______________________________________________________________
S H A K S P E R: The Global Shakespeare Discussion List
Hardy M. Cook, editor@shaksper.net
The S H A K S P E R Web Site <http://www.shaksper.net>

DISCLAIMER: Although SHAKSPER is a moderated discussion list, the 
opinions expressed on it are the sole property of the poster, and the 
editor assumes no responsibility for them.

• Previous message: Hardy M. Cook: "Solid Flesh Once More"
• Messages sorted by: [ date ] [ thread ] [ subject ]