So I can't seem to get around the circular reasoning used to determine if a text is Paul or Deutero-Pauline. The argument goes something like this (albeit in simplified terms):
Step 1: Identify places (words, passages, etc.) x in text y that does not sound like Paul
Step 2: Since Paul wouldn't not sound like Paul, x gives you a good measure whether y was in fact written by Paul
Now, what seems missing is Step 1a, which is: Let us decide already what falls under the purview of "Paul," so that we have a collection of terms, thoughts, systems, theology(s), etc., that tells us what constitutes genuine Pauline thought. But if we have already decided that only a certain set of evidence falls under the "Paul" category, isn't this a form of loading the dice?
To take the stylistic issue for example, if one already decided Letters 1-7 are genuine but 8-13 are disputed, and then from 1-7 gather the evidence for some commonality in style (="Pauline"), of course as a self-fulfilling prophecy it is no surprise to find that Letters 8-13 do not sound "Pauline."
To broaden the scope a bit further, there is the question of "consistency" in Pauline thought, which as often is the case, scholars are divided to no end. Some argue inconsistency, some argue consistency, some argue some middle ground, etc. etc. etc. If in fact Paul has written letters spanning 5-10 years from the first to the last, is it so out of the question to think that he could change his mind (or develop) on a particular position? Must his theology remain static from beginning to end?
I suppose all of this is provoked by my reading through Ehrman's historical intro to the NT, and again the method of argumentation sounds strange to me. For example, in his discussion of Ephesians, he notes that it has roughly 100 complete sentences and 9 of them are over 50 words in length. He tells us that Gal/Phil are roughly the same length and we can note some astonishing differences. Phil has 102 sentences and "only one of them is over fifty words" (408); Gal has 181 sentences and only 1 over 50 words (He then notes similar statistics in Rom 1-4; 1 Cor 1-4; etc.) Then in terms of hapax, he notes that Ephesians uses 116 words not found in any of the undisputed letters while in comparison, Philippians at slightly shorter length has the highest number of unique words (among the undisputed) "but the total there is only 76" (408).
What if Gal/Phil had 30 other sentences that had over 40 words in length (NB: I did not count them, so this is just hypothetical), would that be okay? Or, what if they had 5 sentences over 50 words in length? Is that good enough? Or would they have to have at least 7 (and why)? What kind of specific criteria does one use to say "Okay, this one made the cut" when you are using these numbers? 15%? 25%? Or to look at the hapax question, what if Philippians had 86 unique words, does that then allow us to bring Ephesians into the fold or not? Why is 116 (or 76) used as a disqualifying #? Without any discussion of what constitutes statistical significance (Ehrman does not indicate if he has looked into the p-value of these #'s; or can something like that even be established here?), these numbers mean nothing besides one's "feeling" that something is amiss.
Now, I am not arguing here for a 7 letter corpus, 10 letter corpus, or even a 13 letter corpus, but I am concerned primarily with the methods of argumentation used to establish the categories in the first place (I would say this also applies for the opposite end of the spectrum, to assume a priori that all 13 letters are "genuine" Paul without clear argumentation to that end.) Finally, a scholar may argue that it is the cumulative case that allows for these conclusions, but again, I am not so sure that works; if I put together 5 questionable probabilities, does the conglomeration of them increase the overall probability of my original thesis?
