BW10 can help us to get the frequency counts we need to normalize so we can make more accurate comparisons. According to principles of corpus linguistics, he should have normalized ("normed") the frequency counts (see now Biber, Conrad, and Reppen, Corpus Linguistics) before comparing. In my opinion, he ends up comparing apples to oranges rather than apples to apples, as the saying goes, so his results are skewed. But this begs the methodological question of what he's actually comparing when he does this. To calculate this, Varner simply counts the number of imperatives in each book of the NT and then divides that by the total number of words used in that book. In that book, Varner claims that James has a "higher ratio of imperatives to total words" than any other book in the New Testament. For example, I'm currently (as of this post) co-editing a book on linguistics studies in the epistle of James, and one contributor to the book cites William Varner's The Book of James: A New Perspective. BW10 makes it easy for me to do a variety of corpus analyses.
0 kommentar(er)
