Make a Donation

Consider making a donation to support SHAKSPER.

Subscribe to Our Feeds

Current Postings RSS

Announcements RSS

Home :: Archive :: 2004 :: January ::
Stylometrics
The Shakespeare Conference: SHK 15.0118  Thursday, 15 January 2004

[1]     From:   Mac Jackson <
 This e-mail address is being protected from spambots. You need JavaScript enabled to view it
 >
        Date:   Thursday, 15 Jan 2004 12:53:12 +1300
        Subj:   RE: SHK 15.0083 Stylometrics

[2]     From:   Alexander Ourbook <
 This e-mail address is being protected from spambots. You need JavaScript enabled to view it
 >
        Date:   Thursday, 15 Jan 2004 13:29:49 +0300
        Subj:   Re: SHK 15.0096 Stylometrics


[1]-----------------------------------------------------------------
From:           Mac Jackson <
 This e-mail address is being protected from spambots. You need JavaScript enabled to view it
 >
Date:           Thursday, 15 Jan 2004 12:53:12 +1300
Subject: 15.0083 Stylometrics
Comment:        RE: SHK 15.0083 Stylometrics

Two comments on Marcus Dahl's interesting reservations about Hugh
Craig's work on the Additions to The Spanish Tragedy:

1. Craig's investigation of "rare pairs" is supplementary to his
analysis of high-frequency words in the manner pioneered by John
Burrows. In that study he shows what results his tests (which are
entirely objective and replicable) deliver on texts of known authorship
of the same size as the Additions, and then applies them to the
Additions. Shakespeare emerges as the strongest candidate.

2. Searching electronically for "rare pairs" can yield useful
information when the texts to be searched are specified in advance, when
all these texts are searched equally methodically, and when the numbers
of collocations linking the anonymous text with the works of the various
candidates, relative to the amount of text by each candidate, are
calculated and subjected to analysis.

Craig is reported to have defined his "rare pairs" as those occurring in
the Additions and not more than ten times in a large database. The
crucial result is the number of these "rare pairs" that are shared by
the Additions and, in turn, Shakespeare, Jonson, Webster, and so on,
particularly when the different sizes of each author's searched corpus
are taken into account. So long as the investigator's sense of what
constitutes a meaningful collocation remains reasonably consistent,
problems of definition are not disabling in practice. The Additions to
The Spanish Tragedy, as distinct from the original play, are unlikely to
have been performed before The Rape of Lucrece was written. But the
problem of knowing whether particular shared collocations indicate
common authorship or the influence of one work/author on another, or are
simply due to independent drawing on an early modern stock of poetical
associations tends, at any rate, to disappear when patterns in the data
collected are intelligently analysed. If the number of "rare pairs"
shared by the Additions and Shakespeare is larger than the numbers for
other dramatists, in proportion to the amount of each dramatist's text
searched, this is worth something as evidence for Shakespeare's
authorship of the Additions. In order to explain such a predominance of
Shakespeare links in terms of influence, one would have to assume that
either (a) the author of the Additions was more influenced by
Shakespeare than by any other dramatist, including himself - unless he
happens not to be represented in the extensive database searched, or (b)
Shakespeare was influenced more than any other dramatist by the
Additions, or (c) a combination of "a" and "b". If the predominance of
Shakespeare links is with his works written both before and after the
probable composition date of the Additions, explanations in terms of
"influence" become especially complex.

It is also possible to check how the "rare pairs" test performs on
material of known authorship. I imagine Craig will do this.

The main point is that Hugh Craig's test of "rare pairs" is entirely
independent of his test of high-frequency words. If both tests point
towards the same author, it will be hard to set them aside.

Mac Jackson

[2]-------------------------------------------------------------
From:           Alexander Ourbook <
 This e-mail address is being protected from spambots. You need JavaScript enabled to view it
 >
Date:           Thursday, 15 Jan 2004 13:29:49 +0300
Subject: 15.0096 Stylometrics
Comment:        Re: SHK 15.0096 Stylometrics

 >Marcus Dahl (in the midst of a very sage review of database searches and
 >authorship attributions) comments: "Virtually all old Shakespeare
 >scholars interested in the attribution of early Shakespeare texts (etc)
 >used matching examples of rare vocabulary . . . "
 >
 >I suddenly became worried about what age you became an old scholar at.
 >How close am I now? Are there are any benefits, such as senior discounts?
 >
 >Just wondering,
 >don

I am not sure I should give any response. I am afraid to seem uncivil so
I am 49. No discounts as far as I can see. Do you mean that you have
read my text? Yours Ourbook.

_______________________________________________________________
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.