First, I won't pretend to have all of the answers (or any of the answers, for that matter), but I would like to weigh in on how syntax searching of the OpenText.org Syntactically Analyzed Greek New Testament might help one get a grasp of the problem and the options.
Disclaimer: I work for Logos, and have blogged extensively on the OpenText.org Syntactically Analyzed Greek New Testament on the Logos Blog. I'm not writing this post to say that "danielandtonya" are right or wrong (though their approach seems sound to me and I'm guessing their results will be too), or to say that their data sets are somehow in error. I just want to try to assemble their data sets using syntactic searching (and any advantage that relying on syntactic relationship gives) instead of relying on proximity + agreement relationships, with or without exclusion—and see what differences there are and how important or unimportant they might be.
For the uninitiated, this debate is concerning Gal 2.16 (go ahead, read it, and make sure to check the Greek too).
Specifically, I'd like to interact with Hebrew and Greek Reader's three data sets delineated in their "The Jesus Faith - Vol. 3" post (but also described in the Vol. 2 post).
Here are their descriptions of their three data sets, from their Vol. 3 post:
- Data Set 1 - How many times in the GNT is πιστις (in any case) followed by Χριστος (in any case) within four words of each other?
- Data Set 2 - How many times in the GNT is a genitive noun followed by another genitive noun within four words?
- Data Set 3 - How many times in the GNT is πιστις in the genitive case followed by another genitive noun within two words (and words in between are not conjunctions or non-genitive nouns)?
On the first data set, danielandtonya report that the following references are included: Ro 3.22; Gal. 2.16 (2x); 3.22, 26; Phil 3.9; Col. 1.4; 1Tim 3.13; and 2Tim 3.15. My syntax search located one additional hit, James 2.1. The syntax search looks like this:
I'm searching for where πιστις is the "head term" of the word group (loosely, the "phrase"), and where it is either directly or indirectly modified by χριστος. The James 2.1 instance has four words between πιστις and χριστος, so the intervening range is larger than danielandtonya accounted for. I'm not sure that it makes any difference to the argument, and they may have known about it but weeded it out. I just mention it because it was in my results.
On the second data set, danielandtonya report 1,431 hits (in their Vol. 3 post). That's a lot of hits. The syntax search I created narrows it down to 452 hits.* The difference is that the syntax search locates where the second genitive is in some sort of direct relationship with the first genitive, not just where two genitives happen to be within two words of each other. Again, it is relying on the relationship, not the proximity of words (which essentially serves as a loose approximation of syntactic relationship). Whether this makes any difference for danielandtonya's argument I have no idea. But here's the search:
I should note that I'm constraining to nouns because that is the wording that danielandtonya's specification uses; I might also want to consider adjectives in one or both slots, but that's left as an exercise for the reader to complete.
On the third data set, danielandtonya haven't yet reported (at the time of my initial posting), so I'll have to wing it. In OpenText-ese, what they appear to be looking for is when a genitive noun is in close relationship with πιστεως (genitive form of πιστις), hence the two-word proximity constraint, and the further specification that no conjunctions or pronouns intervene. With a syntactic analysis, there is no necessity to consider the exclusion of certain intervening types (such as conjunctions or pronouns) because one is really searching for the relationship between things no matter what may intervene. Here again, between two nouns, a simple "modification" relationship fits the bill (from what I understand of danielandtonya's intent). So my third search is relatively similar to the previous, I've just added that πιστις should be the lemma of the first word in the series:
What did I get for results? Five hits: Ro 3.22; Ga 2.16(2x); 3.22; Col 2.12. All but Col 2.12 were in danielandtonya's first dataset.
And this is where I leave you. I don't have a dog in the "objective or subjective genitive" argument. I don't like any of the labels because they (at least to me) seem to be geared toward answering the "how do I translate it?" question instead of the "how was it understood?" question. Yes; the two are somewhat related, but the primary difference is the end. One seems to think about and try to understand Greek in terms of English; the other at least tries to think about Greek in terms of Greek. Thus, I'm not a fan of labeling things like this. From my view, the obvious ones are, well, obvious; and the debatable ones are debated ad nauseum to little ultimate benefit.
* If you allow for variation in the order of the head term word and modifier, then the count is 458. But as danielandtonya's specifications rely on order, I figured these should too.