computers and authorship

Grant White ulgw at dewey.newcastle.edu.au
Thu Dec 28 06:04:47 CST 1995


Hi Jeff,

The work I mentioned is;

Burrows, J.F, Computation into Criticism; a study of Jane Austen's novels 
and an experiment in method. Oxford Uni. Press 1987.

In it he explains his methodology and the then state of the art 
techniques he used to analyse the text. At the time it was a pretty big 
undertaking involving typists, proofreaders and a sizeable chunk of the 
uni's vax. Now, almost embarrassingly you can do the same kind of 
analysis with any recent wp programme by using find and list and then 
graphing occurences of particular words or strings. 

Scanning is still irksome, and some kind of text translator is usually 
necessary to put the scanned material into a usuable ascii based 
format or whatever so the wp can recognise it.

I don't know of any off the shelf programmes which help to analyse 
though, but I seriously doubt that you could find a great enough 
deviation in the styles of TRP's stuff to really bother going through the 
scanning process; that is a predictable deviation accuring through 
experimentation and other artistic factors.                                
 
I really warm to the conspiracy theory, but if I were 
writing with them, I'd make pretty damn sure that the formula we were 
using was watertight. Especially when you look at the mass market, 
machine produced product which adheres to simple rules, trying to pin 
down different "signatures" within works as complex as the ones under 
discussion ain't gonna be easy.

But you never can tell.

Cheers,
Grant
--                                                                       -
Grant White                        |
Multi-media & Special Collections  |
Librarian                          | Internet:ulgw at dewey.newcastle.edu.au 
The Central Coast Campus           |
Information Resource Centre        | Ph  (intl+61+43) 484026
University of Newcastle, AUSTRALIA | Fax (intl+61+43) 484215




More information about the Pynchon-l mailing list