Make a Donation

Consider making a donation to support SHAKSPER.

Subscribe to Our Feeds

Current Postings RSS

Announcements RSS

Home :: Archive :: 2004 :: April ::
Stylometrics
The Shakespeare Conference: SHK 15.0809  Friday, 2 April 2004

[1]     From:   Ward Elliott <
 This e-mail address is being protected from spambots. You need JavaScript enabled to view it
 >
        Date:   Wednesday, 31 Mar 2004 23:45:08 -0800
        Subj:   RE: SHK 15.0785 Stylometrics

[2]     From:   Michael Egan <
 This e-mail address is being protected from spambots. You need JavaScript enabled to view it
 >
        Date:   Thursday, 1 Apr 2004 09:10:49 -1000
        Subj:   Re: SHK 15.0803 Stylometrics

[3]     From:   R. A. Cantrell <
 This e-mail address is being protected from spambots. You need JavaScript enabled to view it
 >
        Date:   Thursday, 01 Apr 2004 13:21:35 -0600
        Subj:   Re: SHK 15.0803 Stylometrics


[1]-----------------------------------------------------------------
From:           Ward Elliott <
 This e-mail address is being protected from spambots. You need JavaScript enabled to view it
 >
Date:           Wednesday, 31 Mar 2004 23:45:08 -0800
Subject: 15.0785 Stylometrics
Comment:        RE: SHK 15.0785 Stylometrics

To Gerry Downs:  2H6 is one of the plays we purged from our baseline
this spring to make it as free as possible from suspicion of being
co-authored, and we haven't yet worked out a revised scoring system to
deal with it.  The H6 series as a whole is loaded with indicators of
other- or co-authorship.  1H6 has 11 rejections by new count; 3H6 has 8.
  The highest core-Shakespeare rejection count is two; 2H6 has three.
That's only one rejection worse than the farthest core Shakespeare
outlier.  It falls within the Shakespeare envelope by Valenza's
Continuous Composite Probability test, and only one order of magnitude
outside of it on Discrete Composite.  We're not talking about gazillions
here.  My guess is that it's almost all by Shakespeare and that we shall
come up with some kind of safety factor rules which will say more
formally that it belongs in the ballpark, even if it's not in our
testing baseline.

To Mac Jackson:  Closer examination of the hyphenated compound word
counts in our Arden of Feversham text could well eliminate that
rejection.  We used Lou Ule's text, discovered from spot checks on other
plays that his HCW counts, mostly taken from Tucker Brooke, can be lower
than those found in our baseline Riverside Shakespeare, and warned our
readers of the problem in our Chum 1996 "Final Report" of the
Shakespeare Clinic.  I haven't tried to re-edit the Arden text, but I
would guess from the alternative counts supplied that our HCW rejection
for it is highly doubtful. On the other hand, it seems to me also highly
doubtful that eliminating that one rejection would be enough to make
Arden a likely Shakespeare play, since it would still have 9 rejections
in 48 tests, which would put it maybe six or seven orders of magnitude
outside the ballpark by my odds, 12 or 13 by Valenza's.  2H6 is as close
as a play could get to our Shakespeare infield and still justify a grain
of suspicion of co-authorship; Arden would be on a distant planet even
without the HCW rejection.

[2]-------------------------------------------------------------
From:           Michael Egan <
 This e-mail address is being protected from spambots. You need JavaScript enabled to view it
 >
Date:           Thursday, 1 Apr 2004 09:10:49 -1000
Subject: 15.0803 Stylometrics
Comment:        Re: SHK 15.0803 Stylometrics

A number of interesting points have so far emerged. However I note that
many of the key theoretical axioms of stylometrics have still not been
addressed. One of them is that writers write  in habitual ways and these
can be statistically measured (ignoring 'counting wobble' of course),
then predicted.  Another is that  none of them evolve and change
sufficiently to alter this assumption (try telling that students of
Henry James!) A third is that all genres, expository prose, poetry and
the drama, etc., are equally susceptible to such analysis. A fourth is
the fact that  Shakespeare may be handled on the basis of the same
assumptions as all other writers despite overwhelming evidence that he
is in almost all respects not like the rest.

What are the experimentally verified and testable data behind these
assumptions? Or are we simply dealing with articles of faith?

--Michael Egan

[3]-------------------------------------------------------------
From:           R. A. Cantrell <
 This e-mail address is being protected from spambots. You need JavaScript enabled to view it
 >
Date:           Thursday, 01 Apr 2004 13:21:35 -0600
Subject: 15.0803 Stylometrics
Comment:        Re: SHK 15.0803 Stylometrics

 >"they have overriding and unassignable imperatives and styles of their
own"
 >
 >"Where'd they get them?"
 >
 >Aristotle.

Yaar Matey

_______________________________________________________________
S H A K S P E R: The Global Shakespeare Discussion List
Hardy M. Cook, 
 This e-mail address is being protected from spambots. You need JavaScript enabled to view it
 
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.
 

©2011 Hardy Cook. All rights reserved.