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 15.0815  Monday, 5 April 2004

[1]     From:   Wayne Shore <
 This e-mail address is being protected from spambots. You need JavaScript enabled to view it
 >
        Date:   Friday, 2 Apr 2004 09:49:02 -0600
        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:   Friday, 2 Apr 2004 06:55:07 -1000
        Subj:   Re: SHK 15.0809 Stylometrics

[3]     From:   D Bloom <
 This e-mail address is being protected from spambots. You need JavaScript enabled to view it
 >
        Date:   Friday, 2 Apr 2004 11:36:53 -0600
        Subj:   RE: SHK 15.0809 Stylometrics

[4]     From:   Ward Elliott <
 This e-mail address is being protected from spambots. You need JavaScript enabled to view it
 >
        Date:   Friday, 02 Apr 2004 14:54:29 -0800
        Subj:   RE: SHK 15.0809 Stylometrics

[5]     From:   Mac Jackson <
 This e-mail address is being protected from spambots. You need JavaScript enabled to view it
 >
        Date:   Thursday, 1 Apr 2004 17:57:58 +1200
        Subj:   RE: SHK 15.0785 Stylometrics


[1]-----------------------------------------------------------------
From:           Wayne Shore <
 This e-mail address is being protected from spambots. You need JavaScript enabled to view it
 >
Date:           Friday, 2 Apr 2004 09:49:02 -0600
Subject: 15.0785 Stylometrics
Comment:        RE: SHK 15.0785 Stylometrics

 >To Ward Elliott:
 >
 >I'm suspicious of 2 H4 as wholly Shakespeare's. What's your
 >mechanical
 >opinion? Gerald E. Downs<

Ward seemed to respond to this as if the question were about 2H6. Is
that what you meant, Gerry?

As for 2H4, my analyses, although not as nearly definitive as Elliott
and Valenza's, consistently show 2H4 to be mainstream Shakespeare.
Mechanically speaking.  As to the question of this or any other play
being "wholly" Shakespeare's, there's a limit to how well minor
contributions by another author can be detected.  I'm positive about the
usefulness of stylometrics, but I doubt that it can detect minor hands,
absent a hypothesis specifying parts written by a specified author.

Wayne Shore

[2]-------------------------------------------------------------
From:           Michael Egan <
 This e-mail address is being protected from spambots. You need JavaScript enabled to view it
 >
Date:           Friday, 2 Apr 2004 06:55:07 -1000
Subject: 15.0809 Stylometrics
Comment:        Re: SHK 15.0809 Stylometrics

Prof. Elliot says: 'The H6 series as a whole is loaded with indicators
of other- or co-authorship. ' But is it? Or do his data merely show that
Shakespeare was various enough as an author not to fit Elliott's
preconceptions of how he 'ought' to write?

If the Henry VI series was rewritten or revised by Shakespeare, as the
external evidence suggests, the style differentials may merely reflect
his own evolution as a playwright.

Another unaddressed problem is editorial. For example, Elliott writes of
Arden of Faversham: 'We used Lou Ule's text...' But  judging by Ule's
version of 1 Richard II/Thomas of Woodstock, with which I am intimately
familiar, such a text would be grossly misleading at best and completely
useless at worst. Ule's version so denatured the original as to be
barely recognizable as the same play. I'm unclear how stylometrics deals
with the problem of second- or third-party interventions, from known or
unknown changes wrought by a copyist, to those introduced by printers
and subsequent editors.

Stephen Orgel argues that the chimera of a pure or authentic Shakespeare
text is merely that, an illusion, 'something bestowed, not inherent,' a
kind of 'platonic idea.' (Orgel and Keilen (eds.): Shakespeare and the
Editorial Tradition (New York and London: Garland Publishing, Inc. 1999)
pp. 101, 95, 102.) Is a shadow on the wall enough to permit confident
ascriptions of authorship/co-authorship?

--Michael Egan

[3]-------------------------------------------------------------
From:           D Bloom <
 This e-mail address is being protected from spambots. You need JavaScript enabled to view it
 >
Date:           Friday, 2 Apr 2004 11:36:53 -0600
Subject: 15.0809 Stylometrics
Comment:        RE: SHK 15.0809 Stylometrics

Michael Egan writes:

 >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?"

I am no expert in this field, but I have a feeling that there is a
misunderstanding here. ME seems to be thinking of style as the way the
conscious artist makes decisions about his materials and adapts them to
fit different forms, situations or characters. The stylometrists
(stylometricians?) think of it as quirks, unconscious patterns that a
writer (or other artist) doesn't adapt, because he or she doesn't think
of them.

With care, these unconscious patterns can be identified, and they will
be revealed in studies of individual works no matter what the genre, no
matter if the author (like WS) is replicating speech patterns of various
classes and even nationalities, or doing a deliberate parody of another
writer's work. Like every other human being, WS has this unconscious
patterning and the only issue is whether it has been adequately
identified in any given study. That can only be determined by a close
review of the study.

(I hope I have not muddled this badly.)

Cheers,
don

[4]-------------------------------------------------------------
From:           Ward Elliott <
 This e-mail address is being protected from spambots. You need JavaScript enabled to view it
 >
Date:           Friday, 02 Apr 2004 14:54:29 -0800
Subject: 15.0809 Stylometrics
Comment:        RE: SHK 15.0809 Stylometrics

  -- Michael Egan in SHK 15.0803 speaks of supposed stylometric
"axioms," (1) "that writers write  in habitual ways and these can be
statistically measured;" (2) "that none of them evolve and change;" (3)
that all genres are equally susceptible to such analysis; and 4) "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."

I doubt that any stylometrist would consider any of these propositions
axioms.  They are all hypotheses which can be tested empirically, and
some seem to me plausible after testing, others not so plausible.  We
subscribe to 1), some writers' habits are measurable.  It's testable,
and our tests say it's so.  We have never subscribed to 2), measurable
habits never change, but we have tested it.  Our tests say that about a
seventh of our 48 play tests are time-sensitive and require tighter
chronological matching than the others to be useful.  We have never
subscribed to 3) but confined our tests to poems, plays, and play verse.
  At least one of our tests, grade level, is genre-sensitive and, hence,
requires control for genre to be useful.  We have never subscribed to
4), that Shakespeare is "in almost all respects not like the rest" as an
axiom, but that's testable too, and we have found 48 tests that say that
in 48 ways he differs enough from the rest for the test to help
distinguish him from non-Shakespeare. We've probably done more than most
to show that there could be a grain of truth in 4, leaving aside the "in
almost all respects" part.  But my guess is that we could easily have
found 48 other ways in which Shakespeare resembled the rest, had there
been any good reason to do so.

  -- Mr. Egan asks: "What are the experimentally verified and testable
data behind these assumptions? Or are we simply dealing with articles of
faith?"

For the bones of an answer, see our "And Then There Were None: Winnowing
the Shakespeare Claimants," 30 Computers and the Humanities 191 (April
1996).

Ward Elliott

[5]-------------------------------------------------------------
From:           Mac Jackson <
 This e-mail address is being protected from spambots. You need JavaScript enabled to view it
 >
Date:           Thursday, 1 Apr 2004 17:57:58 +1200
Subject: 15.0785 Stylometrics
Comment:        RE: SHK 15.0785 Stylometrics

I take it from Terrence Hawkes's vivid little email about the
"soul-gelding, buttock-clenching tedium" of "stylometrics" that in the
unlikely event of my ever coming across an article that addresses a
question of attribution and is published under or over the name "T.
Hawkes" I can safely conclude that either (a) it has been misattributed,
or (b) it is by some other "T.  Hawkes", or (c) it is by the man whose
books we have all read and enjoyed but is intended as burlesque. If the
article should contain either of those present-participial compounds
I'll be inclined to opt for explanation "c".  Did some "overriding and
unassignable imperative" of the email text produce "soul-gelding" and
"buttock-clenching", or are those characteristic flourishes Terrence
Hawkes's own?

Mac Jackson

_______________________________________________________________
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.