Computational stylistics
Posted: 26 June 2012 12:42 AM   [ Ignore ]
RankRankRankRank
Total Posts:  1190
Joined  2007-04-28

I know this has been used before with minor Elizabethan dramatists like Cyril Tourneur and for determining which parts of The Two Noble Kinsmen Shakespeare and Fletcher wrote but I have no idea how reliable it is. Now it’s been applied to Dickens.

Profile
 
 
Posted: 26 June 2012 03:37 AM   [ Ignore ]   [ # 1 ]
Administrator
Avatar
RankRankRankRankRank
Total Posts:  4474
Joined  2007-01-03

Computational stylistics can be very reliable. But of course, there are various techniques, and these vary in reliability and function. But a key factor is the length of the text analyzed; you generally need samples of several thousand words to get statistically meaningful results. The Guardian piece doesn’t say how long the suspected Dickens article is.

Diegogarcity alert: the latest episode of Slate’s Lexicon Valley podcast is on computational stylistics and is well worth a listen. They use authorship of one of the Wizard of Oz books as their example. (Another writer carried on after Baum’s death, and their has been debate about who wrote the book that was first published after Baum’s death.) They make the excellent point that the “fingerprint” metaphor is not a good one as style is not necessarily unique. (Style can be imitated—not easy, but possible; writers may naturally be very similar in style; and a writer’s style changes over time.) “Handwriting analysis” is a better comparison, only computational stylistics is more reliable than that.

Profile
 
 
Posted: 26 June 2012 04:55 AM   [ Ignore ]   [ # 2 ]
RankRankRankRankRank
Total Posts:  3335
Joined  2007-01-29

Is it really?  Have enough studies been done, and evaluated with a gimlet eye by non-believers, to be able to claim such reliability?  I remind you that fingerprints have turned out not to be unique, and fingerprint evidence has started to be thrown out of court; the human will to believe in the latest magic is strong.  I’m not saying computational stylistics isn’t interesting and possibly useful, but just because somebody applies the technique and says confidently “Yep, this is Dickens” (or whoever) doesn’t mean they’re right.

Profile
 
 
Posted: 26 June 2012 05:24 AM   [ Ignore ]   [ # 3 ]
Administrator
Avatar
RankRankRankRankRank
Total Posts:  4474
Joined  2007-01-03

Computational stylistics is a broad term. Some of the techniques are sketchy, if not down right bogus. Others have been shown to be highly reliable. Also different techniques are applicable to different types of analysis. If the researchers know what they’re doing, it can be conclusive to a high degree of confidence; if not a good researcher will state the appropriate caveats. I wouldn’t trust newspaper articles to accurately assess the reliability of any particular study.

I remind you that fingerprints have turned out not to be unique

Is this the case? I thought the status was that the uniqueness of fingerprints had never been quantified in peer reviewed literature (not helped by the fact that the FBI, with the world’s largest collection of fingerprints, refuses to fund or participate in such research, presumably for fear that it might be found to be so), but that no one seriously doubts that fingerprints, when the technique is properly applied, is not a valid and useful method for high-confidence identification.

From what I’ve read, the cases where fingerprints have been tossed out of court have to do with inadequately trained collectors and analysts and the fact that there is no accepted standard for the number of points of comparison needed for a reliable identification—which goes back to the point that no one has ever systematically quantified the discipline. An identification with twelve points of comparison is almost certainly unique for all practical purposes, one with only four has a pretty good chance of matching several people.

Profile
 
 
Posted: 28 June 2012 08:45 AM   [ Ignore ]   [ # 4 ]
RankRankRankRank
Total Posts:  1190
Joined  2007-04-28

It would be interesting to run parodies and original writers’ texts through these programs and see what disparities are thrown up.
Fingerprint is a metaphor as Dave says; you could say stylistic markers (from DNA or genetics or wherever). It’s a false analogy amounting to “stuff has been shown to be wrong so this might be, too” though I’m sure this wasn’t what LH was saying and I haven’t had time to listen to the link yet.

I looked up The Revenger’s Tragedy and new authorship seems to have been assigned by human agency - imagery, vocab, theme, style, etc http://en.wikipedia.org/wiki/The_Revenger's_Tragedy#Authorship rather than computers though I’d imagine computer analysis is particularly helpful with metrical patterns in Elizabethan plays.

Profile
 
 
Posted: 17 July 2012 01:34 AM   [ Ignore ]   [ # 5 ]
Avatar
RankRankRankRankRank
Total Posts:  2884
Joined  2007-02-26

Would be made trickier by the fact that, presumably, Fletcher and Shakespeare would have been _trying_ to write in a consistent style so that TTNK didn’t seem like an obvious chimera.

Profile
 
 
   
 
 
‹‹ HD: 1990 Words      HD: 1991 Words ››