I am currently involved at the amateur level in a wide-ranging scholarly debate, one element of which consists of several affidavits or other signed witness statements that I believe were ghost-written or otherwise strongly prompted by a single, heavily biased individual. What I’m looking for is either end-user (and affordable) software to perform an analysis to determine that, or, failing that, an affordable service that could do the same thing for me.
I know there are services available to detect plagiarism, but that’s not what I need.
You’re looking for software that will determine if two documents, one written by person A and one written by person B, were written at the urging of of person C?
I’ve heard of “wordprint” software that is supposed to be able to tell if Document A and Document B were written by the same person, but I’ve also heard criticism that it isn’t at all reliable.
There’s a professor somewhere, Don Foster, who apparently does this kind of thing. He wrote a book, Author Unknown, about some cases he had. However, he’s gotten things wrong before, and some people strongly dislike him.
Not quite. I’m not seeking anything nearly that specific.
What I want is to evaluate five documents to try to get an estimate as to how likely it is that they were all written by the same person. It doesn’t even have to use computer analysis (but I think a computer analysis would be cheapest).
The idea would be to examine the writing style and word usage patterns, for example. I understand that the reliability of such an analysis would depend on how large the example texts are and so forth.
I’ve read a pretty fair amount of this kind of thing being done more or less routinely these days.