I have been slowly reading my way into the New Perspective on Paul (NPP) literature for the last few months as part of my current work in Romans. James Dunn’s keen observations about commonly overlooked–but important–words has been a bright spot in wading through the rat’s nest of detail. The next time I hear someone whine about how complex and detailed discourse grammar is, they’ll get a dope slap. Lots of fields and discussions within them are complex; I am 9re)learning that with the NPP literature. It is just a matter of gaining the needed background to understand all of the information that is assumed by those participating in the discussion.

A great example of that is ποτε in Gal 2:6; I had never noticed it there before, and few translations make note of it. The other big surprise was the presence of ἐὰν μὴ in Gal 2:16. I have not read closely through Galatians since writing the discourse grammar, and so had not been as keenly aware of the importance of conjunctions as I am now.

In doing some checking of how the major translations handle Gal 2:15-16, I was unpleasantly surprised at how consistently some clearly marked grammatical feature was translated as though it were something else. Here is the text from the SBLGNT:

15 Ἡμεῖς φύσει Ἰουδαῖοι καὶ οὐκ ἐξ ἐθνῶν ἁμαρτωλοί,

16 εἰδότες ⸀δὲ ὅτι οὐ δικαιοῦται ἄνθρωπος ἐξ ἔργων νόμου ἐὰν μὴ διὰ πίστεως ⸂Ἰησοῦ Χριστοῦ⸃, καὶ ἡμεῖς εἰς Χριστὸν Ἰησοῦν ἐπιστεύσαμεν,

ἵνα δικαιωθῶμεν ἐκ πίστεως Χριστοῦ καὶ οὐκ ἐξ ἔργων νόμου,

⸂ὅτι ἐξ ἔργων νόμου οὐ δικαιωθήσεται⸃ πᾶσα σάρξ.

The only content-bearing textual variant is the presence or absence of δὲ at the beginning of v. 16;  it is bracketed in both NA 27 and 28, but unbracketed in the SBLGNT. The rest concern the ordering within noun phrases or clauses, i.e. the content of these verses is not disputed. Most translations take v. 15 as a separate clause as opposed to a topical frame of reference for v. 16, which I would agree with. This means that the choice is between δέ versus asyndeton, not reading v. 15 as a complex subject for v. 16.

The more surprising issues I found had nothing to do with variant readings. Most translations render εἰδότες as a finite verb, either as though it were an indicative or an imperative. This decision then impacts the understanding of καί later in the same line. If the participle is rendered as a finite verb, then καί wants to be conjunctive instead of ascensive.

But the most surprising (and disturbing) thing was to see how many scholars I deeply respect argued that the exceptive construction ἐὰν μὴ is somehow exceptional and thus NOT exceptive. I had no idea how much ink had been spilled on this issue. Many trees gave their lives so we could have another contentious issue in the field!

Please save your vitriol and piling up of scholars in favor of your view on the matter. At least for now, this is not the time or place for that.

What I intend to do in preparation for a more thorough treatment of this verse is to start looking at exceptive clause in the GNT. I’ll begin with the prototypical ones and offer an expansion of the description I offer in the discourse grammar. 1 Then I’ll tackle the claims in the dead grammarians regarding the apparent substitution of ἀλλά for ἐὰν/ἐι μὴ. This is the basis for treating Gal 2:16 as an exceptional exceptive; well, that and the apparent horror over the theological implications of treating it as a true exceptive.

I’ll insert a preview here: if we are all happy to allow Paul to state a for-sake-of-argument position with which he likely disagrees in v. 15, then why can’t that same explanation extend into 16a? Especially since he resoundingly dispels this supposed “common ground” in 16c and 16d?? Just a thought, but I’ll leave that (and addressing any comments) for later. This would mean that I disagree with Dunn’s claim that 2:16 somehow represents a development in Paul’s own understanding of justification. I would also disagree with most disagreeing with Dunn by claiming this is not an exceptive clause. Be patient, we’ll get there, but not today.

The crux of the matter is understanding what exception clauses do, regardless of how they might be translated into English. If we can understand this, then we’ll be in a much better position to understand why cases like this one in Gal 2:16 have proven troublesome. There are several other similar cases, none of which carry the same theological freight.

Part 2 continues here.

  1. If you are not familiar with that, there is an excerpt available here as a PDF. If you have the book the relevant discussion is found in chapter 4, especially pp. 83-100.