« Aggregators, Blog Readers and other Animals | Main | Michigan-Claremont Electronic BHS Available Online »

Concordance Compiled with the Aid of... Bible Software?

Andreas Köstenberger and Raymond Bouchoc are the authors of the fairly recent The Book Study Concordance of the Greek New Testament (Broadman & Holman, 2003). This is what the blurb of the book says:

A New Bible Study Tool and a New Venue of Academic Research. Aided by breakthroughs in computer technology, The Book Study Concordance of the Greek New Testament has compiled data in a format that has never before been available to Bible students. The result is a collection of twenty-seven concordances listing every word used in the Greek New Testament in alphabetical order book by book. Also provided are word totals, most-frequently-used words, and words set in relation to the New Testament as a whole. This is an absolutely invaluable new tool for all serious Bible students and for the scholarly community.

This mammoth work (viii/1528 pages) is based on the electronic version of NA27 developed by the Gramcord Institute. In the Preface, the authors state:

The Book Study Concordance of the Greek New Testament for the first time assembles concordances of each of the twenty-seven books of the New Testament. The concordance is a fresh effort, though of course standing on the shoulders of those who have gone before us. The textual base of the present concordance is the electronic version of the 27th edition of the Nestle-Aland Novum Testamentum Graece. The roots of the words were matched with their forms based on data developed and provided by the GRAMCORD Institute. In this regard we would like to acknowledge the foundational debt we owe to the previous work of the GRAMCORD Institute. The concordances themselves were generated by our own programs written to generate the concordance listings from the raw GRAMCORD data.

Since Raymond Bouchoc is research scholar for the GRAMCORD Institute, they no doubt had access to the latest version of the tagged database, but I wonder what kind of program is "our own programs", and why did they not use any of the Bible software packages currently available. Could it be that none of them had the kind of statistical features and flexibility they needed? Here is what H. Van Dyke Parunak had to say about the issue of statistical analysis in his recent review Windows Software for Bible Study (pp. 481-482):

At first glance, it seems natural to plot frequency statistics per chapter, but this approach has several weaknesses. Chapters do not necessarily correspond to the natural discourse units of the text, either in extent (a natural unit may be wider or narrower than a chapter) or in their limits (which may not correspond with those of natural units). The same can be said of fixed width windows that are sometimes used in plots of this sort (for example, plotting occurrences in windows ten verses wide). This mismatch results in a profile that distorts the actual structure of the text. A much better approach is to let the window width change dynamically with the distribution, an algorithm that could be easily implemented by any of these packages. With this refinement, plots such as these become powerful tools for visualizing the structure of texts (...) but these features are not visible with plots at the chapter level. It would be even more useful if software packages provided an option to generate a file containing, not verse references, but the index number of each hit in a search, together with the number of words per verse and per chapter, so that users could directly manipulate distributional information in a package such as Excel or Mathematica. A further refinement would be to let the user define and annotate a number of fields with each hit to capture contextual features (e.g. direct vs. indirect or human vs. divine speech, putative literary source), and provide a simple flat-file database function (sorting and searching) to help the user perform supplementary studies.

Well, if such an algorithm "could be easily implemented by any of these packages", I wonder why not a single one of them has already done so! Moreover, when I checked some of the stats given in the book against Accordance (which happens to include that particular database and has a pretty good statistical analysis feature) I soon came across a few discrepancies in the numbers. So, I would definitely like to know a little bit more about the tools and methodology followed by Köstenberger and Bouchoc. Their book is a welcome addition to the field of Greek reference tools, but I think some more information is in order if peer-review is to be pursued consistently.


This page contains a single entry from the blog posted on February 21, 2004 4:22 PM.

The previous post in this blog was Aggregators, Blog Readers and other Animals.

The next post in this blog is Michigan-Claremont Electronic BHS Available Online.

Many more can be found on the main index page or by looking through the archives.

Powered by
Movable Type 3.35