I wrote some more code today to compare “tri-logs” (a set of three adjoining words) in the Pastoral Epistles with tri-logs that appear in the Pauline Epistles.

One thing I noticed is that I forgot Philemon; so I need to regenerate some data. But I thought I'd post sans-Philemon anyway. Here's the link:

A Concordance of Tri-Logs Held in Common Between the “Genuine” Paulines and Pastorals.

The interesting bit: Out of 27,166 unique tri-logs in the combined corpus, 2408 occur more than once. Of those 2408, 280 are found in both the Pastoral Epistles and in the “genuine” Pauline epistles. So, right around 1% of tri-logs repeat across corpora. The Pastoral Epistles themselves, remember, have 3270 tri-logs with 141 occuring more than once.

(this, of course, assumes I don't have any nasty bugs like I had before)

How'd we get 280 in common? Apparently, many of the tri-logs that occur only once in the Pastorals also occur in the Paulines. That's only to be expected.

Math is not my strong suit, so I haven't done any actual statistical analysis beyond just looking at numbers and comparing. Maybe later.

Update: I should note that some of the more interesting areas of the data involve common words. If you examine the tri-logs that begin with a conjunction or a preposition like διὰ or ἐν or καί, you can see some interesting things.

Update II: Data has been updated to include Philemon. 27,422 tri-logs, 2436 occur more than once. The Pastorals and “genuine” Paulines have 280 in common.

Update III: In response to Eli's question, I must've mis-communicated. My basic process to compare the “genuine” Paulines to the Pastorals has been, in brief:

1. Start with entire listing of tri-logs in the “genuine” Paulines.
2. Compare Pauline tri-log to the entire Pastoral tri-log list.
3. If there is a match, then it is recorded and the reference lists for the Paulines and Pastorals are merged. The resultant node is dumped into a new document (linked to above).

This includes the “non-repeaters”. If a tri-log only occurred once in the Paulines, it is evaluated against all of the Pastoral tri-logs. If a tri-log only occurred once in the Pastorals and wasn't found in the mass comparison, then it must not occur in the Paulines. For example, the very first hit in the comparison concordance is αἰών ἀμήν ἀσπάζομαι. This occurs once in the Paulines (Php 4.21) and once in the Pastorals (2Ti 4.19).

Unless I'm missing something blindingly obvious (which is quite possible).