Make a Donation

Consider making a donation to support SHAKSPER.

Subscribe to Our Feeds

Current Postings RSS

Announcements RSS

Home :: Archive :: 2004 :: March ::
Stylometrics
The Shakespeare Conference: SHK 15.0739  Monday, 22 March 2004

[1]     From:   Sean Lawrence <
 This e-mail address is being protected from spambots. You need JavaScript enabled to view it
 >
        Date:   Friday, 19 Mar 2004 10:48:09 -0800
        Subj:   RE: SHK 15.0729 Stylometrics

[2]     From:   Ward Elliott <
 This e-mail address is being protected from spambots. You need JavaScript enabled to view it
 >
        Date:   Friday, 19 Mar 2004 13:37:25 -0800
        Subj:   RE: SHK 15.0729 Stylometrics


[1]-----------------------------------------------------------------
From:           Sean Lawrence <
 This e-mail address is being protected from spambots. You need JavaScript enabled to view it
 >
Date:           Friday, 19 Mar 2004 10:48:09 -0800
Subject: 15.0729 Stylometrics
Comment:        RE: SHK 15.0729 Stylometrics

Michael Egan writes,

 >To pursue Elliott's astronomical analogy (which
 >in his version is not remotely relevant), it's as though we're given
 >wildly varying estimates of the distances of Mars and Venus from the
 >sun, but assured that it doesn't matter just as long as we agree that
 >the planets are out there somewhere. Which is nearer to the earth
 >however is of some importance to those actually making the trip.

Not necessarily, to extend an irrelevant analogy.  We might disagree
about the size of the solar system as a whole, but be fairly certain of
the relationship of its parts.

It strikes me that what stylometrists are trying to measure in
Shakespeare is seldom absolute numbers anyway, but ratios and
relationships.  These may remain constant while the absolute numbers
rise and fall.

Differences in counts, in other words, don't seem to me to call into
question stylometrics in general, only the occasionally exaggerated
claims by some practitioners to determine authorship or other questions
with the precision usually reserved for things like the calculations of
civil engineers.

Yours,
Sean.

[2]-------------------------------------------------------------
From:           Ward Elliott <
 This e-mail address is being protected from spambots. You need JavaScript enabled to view it
 >
Date:           Friday, 19 Mar 2004 13:37:25 -0800
Subject: 15.0729 Stylometrics
Comment:        RE: SHK 15.0729 Stylometrics

Have I got this straight?  Michael Egan earlier told us that, because he
and Alfred Hart got different counts of "un-" words, Hart's work was
therefore "not always reliable" and was "not good enough for a
methodology claiming the authority and precision of a science." (SHK
15.0663).  Now (SHK 15.0729) he says he "made no slighting remarks"
about Hart and was "certainly not concerned to 'put him down,' nor
anyone else, least of all Prof. Elliott." On the other hand, in the same
posting, he also talks of my "testy and defensive response," my "ad
hominem remarks,"  and my "minatory tone," which "hardly conduces to
objective and/or collegial discussion," and concludes that, if what I
say is so, it just proves that stylometric data are "woefully
subjective, variable, and insecure,"  and that Brian Vickers'
"Shakespeare, Co-author" and Mac Jackson's "Defining Shakespeare" are
"not worth the paper they are printed on."

Strong talk, I would say. Some might call it testy and minatory, or even
puzzling.  But I'm only a mild-mannered political scientist, and I leave
it to my lit-prof betters to make such judgments.  But it's a bit shy of
the mark in implying that my acknowledgement that there is such a thing
as counting wobble casts doubt on the brilliant stylometric
accomplishments of Vickers and Jackson.  Stylometry is often less cut
and dried than Mr. Egan supposed, but "counting wobble," as I pointed
out in my last posting is neither universal nor impossible to control.
To discredit someone's stylometric work you have to do more than perform
a recount, using a different set of rules, and proclaim that your
different outcome invalidates the first count.  That's the "my way"
fallacy, and it is a fallacy.

Could it be that Mr. Egan at bottom shares the deep computerphobia that
I've encountered in prior exchanges with another prolific Shaksper
poster?  And that Mr. Egan, too, is looking forward to another of those
interminable, feisty wrangles that luxuriate on Shaksper either because
people enjoy them or because they are so confrontational that you can't
decline or bow out without its being taken as a concession that the
other side was right?  I don't like such exchanges and would prefer to
avoid them, though not to the point of never making a Shaksper posting
that might arouse someone's ire.

In the other poster's case, his argument was that computers can only
tell you what you already know, not what you don't know.  I got out of
it by offering him a $1,000 bet that, if we tested any one previously
untested non-Shakespeare play of his choice, our tests would reject it.
  He did not take me up.  The amount was purposely set high because our
testing costs are not trivial, but also because his testing costs, if he
took us up, would not be trivial either.  Nothing would please me more
than a serious, deep attempt to challenge our own rules by trying to
understand them and apply them in different ways and to different texts.
  Even if we lost, $1,000 of incentives to do so would be money well
spent on our part.  But my guess is that we would win, and offering it
as a bet, rather than a reward, is a filter against shallow, frivolous
takers who might otherwise consider our time a free good.

I am prepared to make the same offer to Mr. Egan.  Our $1,000 even-odds
bet says he can't find us an untested, non-Shakespeare play that our
computer won't reject with room to spare.  But with this addition:  He
may recount any Shakespeare indicators he considers wobbly, both in the
play he chooses and in our core Shakespeare baseline of 30+ Shakespeare
plays with no suspicion of co-authorship, by appropriate, documented
counting rules of his own choice.  I would be surprised if it made much
difference.

Wayne Shore seems to me spot-on in his skepticism that Shakespeare's
vocabulary was double or triple Milton's.  We've got three tests, two of
them novel, that, with proper allowance for corpus size, say Milton's
was bigger, and by more than a nose, but we haven't written it up yet.
Shakespeare's genius was in how he used words, not how many words he knew.

_______________________________________________________________
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.