Comparing Greek Morphologies – Part I


Over the next couple weeks, I’m going to be comparing the various morphology tagging systems that I have access to. Much of this is related to my BibleTech presentation, which will devote a large amount of time to discussing morphological analysis.

Since I’m not one of those independently wealthy people I only have one Bible software package – Logos. This limits me somewhat in that I cannot examine the morphological databases provided by either Accordance or Bibleworks (If either company would be willing to provide a review copy of their software, I would be more than happy to provide a review).

Even still, I do have five difference morphological databases within Logos – I think all they have available:

  • Friberg
  • Gramcord
  • Logos (their proprietary morphology)
  • Robinson
  • Swanson

In this first post, there is more general issue that I want to deal with first that has troubled me for some time. The rest of this series will discuss the various other morphological options beginning with parts of speech. From there, we’ll examine available for each part of speech in the rest of the series.

The Criticism:

For a good year and a half now I’ve grown more and more frustrated by how linguistic/grammatical databases for Biblical Greek has been developed and published. Now normally, I am rather careful not to call the Greek found in the New Testament “Biblical,” since it is the same Greek we find in plenty of non-biblical texts of the same period. But this criticism requires the word simply because the databases in question were made specifically for Biblical scholars and students in mind. Frankly, the majority of our morphological databases have exegesis as the end goal rather than grammar. Now one might say such a statement is not very fair since grammar is an important prerequisite for morphological analysis.

But that is exactly my point. Doing the grammar is a prerequisite. But with only one exception (Friberg), no other Greek morphological database that I am aware of actually documents their grammatical and morphological analysis in a thorough manner. (CATSS for the LXX provides a very basic description). If anyone knows of any other publically available descriptions of different morphological analyses, I would be very interested. I searched around the Accordance site as well as Bibleworks for such a description, but came up empty.

Friberg, in a way, documents their work twice. They first provide an extensive explanation of their morphological work in the appendix to the Analytical Greek New Testament and then again more than a decade later in the Analytical Lexicon, where there are some profound differences in perspective. The appendix on Deponency by Neva Miller, in particular, represents a massive shift; one that if implemented into the database itself would greatly increase the value of the work. And as much work as it would take, I think it would be worth it to do so.

I can think of a good dozen reasons (and probably more) why morphological analysis should be documented. Let me put some of them in the form of questions that I have asked myself a few times when doing morphological searches in various databases.

  • How does this databases define a lexeme/lemma?
  • What other kinds of lexical information is represented and how?
  • If the database uses the tag “deponent,” how do they determine what is and what is not “deponent”?
  • What are the criteria for deciding how many Parts of Speech to use?
  • What are the criteria for deciding whether a give word is an Adjective? An Adverb? A Particle?
  • To what extent are semantic categories used rather than morphological or syntactic categories?

I’ll let you mull over those questions for a bit. There’s plenty of food for thought there without going overboard. Other issues will come up later on as I write other posts. The point is that considering the amount of work that goes into creating such a database, grammars (or at least grammar sketches) should result every time. I want answers to these questions without having to spend what would likely become several months in each database doing search after search after search looking through the analysis. And I’m sure I’m not the only one.

Now, onto Part II