The Current Usefulness And Limitations To Computer Textual Analysis

Recently, some scholars have made use of computer analysis to address the questions of who wrote the Federalist Papers and who wrote the Epistle to the Hebrews:

According to the article, the main way in which they pursued this task was to analyze vocabulary within the documents, which was then compared to the vocabulary within documents by known authors.  That is quite useful.  When I authored my book on Sarapion of Thmuis, I made use of a similar approach when arguing that he did, in fact, write the Letter to the Monks.  It is important to remember, however, that it takes more than just common vocabulary.  A stronger case can also be made if rhetorical patterns are common, for instance.  In the case of Hebrews, although the computer-assisted scholars determined St. Paul was the most likely author of the candidates investigated, one needs to utilize other aspects of the text beyond vocabulary.  Of course, computers can help with that too.  I think it’s fair to say that “digital humanities” has a long future before it.

2 Responses

  1. 123

    I’ve always thought analysis of such things to be founded on a faulty assumption regarding a person’s vocabulary, rhetorical patterns, etc.: that they remain fixed. I can write wildly differently depending on the audience and purposed of the piece, and people can change styles over years as any cursory glance of one’s high school or college papers reveal.

    1. Certainly, people can and do change over time, it is true, but statistically, we each do tend to have “writing voices,” no less so in an era when rhetoric was important and so highly regarded. I don’t dismiss such approaches, but merely note that as important as vocabulary is, other factors have to be analyzed as well.

Comments are closed.