Winer’s warning about mistreating conjunctions

This is part 2 of a series begun here.

Diane Blakemore, working in the area of cognitive linguistics, has claimed that particles like conjunctions  each have their own unique cognitive constraint, i.e. a set of instructions regarding how to relate what precedes with what follows.¹  So from a functional standpoint, choosing to use one conjunction over another represents the choice to bring this cognitive constraint to bear over that one.

Although the work of Blakemore and others represents an advance in terms of its application of cognitive linguistics, the underlying assumption of a unique function or meaning is not new. Well over a hundred years ago Georg Winer made much the same point regarding conjunctions each having a distinct contribution to the context. He made this claim in the course of critiquing commentators who were quite comfortable with obscuring the uniqueness by claiming that conjunctions were regularly substituted for one another. This appeal to semantic overlap via substitution allowed them to avoid engaging unpleasant exegetical implications associated with the core constraint of the conjunction.

It is strange indeed to see how the commentators (up to a recent period) take the apostles to task again and again, and almost always supply them with a different conjunction from that which actually stands in the text. If a calculation were made, we should certainly find that in Paul’s Epistles, for instance, there are not more than six or eight passages in which the apostle has hit upon the right particle, and does not need the commentator to help him out. This has introduced great arbitrariness into N. T. exegesis. Are we to suppose that Paul and Luke knew Greek no better than many of their censors? The Hebrew usage cannot be appealed to here by any who do not take a wholly irrational view of the Hebrew language: indeed such an arbitrary use of quid pro quo is not possible in any human speech. The arbitrariness of the N. T. interpreters was rendered the more obvious by the fact that different commentators often assigned entirely different meanings to a conjunction in the same passage. … The translators of the N. T. books (not excepting even the excellent Schulz in the Epistle to the Hebrews) are also deserving of censure, since they render the conjunctions in the most arbitrary manner.²

Ouch! If we changed to a more vernacular wording, I can picture Mr. T saying making the same point: “I pity the commentator…” (Georg ‘Clubber’ Winer).

Winer’s warning about appealing to Hebrew (or Aramaic) will prove relevant when we finally get back to Gal 2:16. Why? Most commentators who appeal for an exceptional sense of ει μη in Gal 2:16 do so by appealing to grammarians, most of whom appeal to a comment by Wellhausen that ει μη and αλλα are used interchangeably for אִלָּא, an Aramaic conjunction.³

So does this mean that the two Greek connectives share significant semantic overlap? No! Nor is this a case of an “Aramaism” that has crept into Paul’s writing.⁴ This turns out not to be a case of an Aramaism, but one of the many instances of mismatches between languages.

Linguistic Mismatches

What do I mean by a mismatch between the languages? The mismatch involves Greek having two options available whereas Aramaic has only one. This means that the LXX translator is constrained to make a choice because of the two options available in Greek that are not present in Aramaic. In other words it’s a Greek problem rather an an Aramaic one. We find the same thing in Greek with translating και : a single word plays triple duty compared to English (and, also, even) and double duty compared to Hebrew (ו and גַמ). Put differently, there is a dilemma in English to choose between and or but for Greek’s conjunctive και. Why? Because English makes a semantic distinction between +semantic continuity (and) and -semantic continuity (but).⁵ This is an English problem, not a Greek problem.

“Boy, those French! They have a different word for everything. … They don’t have the decency to speak English.”
― Steve Martin

Why? Because the Greeks didn’t have the common decency to only have one meaning for each of their conjunctions like we do in English! Because our language forces us to make a choice which is not present in Greek by virtue of having English having a finer distinction available than Greek does in this particular context. The opposite would hold true as well.

Nevertheless, we find something of a language-centric snobbery at times, a grumbling that Greek conjunctions don’t have the “common courtesy” of having only one meaning like English conjunctions do. I am not making this up. Dana and Mantey seem to regard English as something of an anomaly in that its conjunctions only have one meaning when compared to other languages. They state, “In Greek, as in Hebrew and Latin, but unlike the English use, a conjunction may have several meanings, each requiring separate and careful study.”⁶, 240).] This, however, is not really the case. Each conjunction in each language has a unique function. If it did not it would eventually drop out of the language. Instead what we find is that the mismatch between the languages is treated from an English- or German-centric standpoint. The assumption is made that English words have a single meaning, whereas the Greek, Hebrew or Latin words have more than one meaning, based on how many different options there are for translating the latter into English. This assumption is just plain wrong. Despite the wrongness, we find the same presupposition undergirding claims of commentators and exegetes.

The different senses of conjunctions are not evidence of their having really broad or multiple meanings. Rather, these different senses highlight the mismatches between Greek and English. English conjunctions have a unique cognitive constraint that they bring to bear just as Greek or Hebrew ones have.

Function words

Claiming semantic overlap or substitution of the Greek conjunctions highlights one of the problems with a sense-based or translation-value understanding of a conjunction. Rick Brannan, in his killer treatment of αλλα, cited Funk’s treatment of conjunctions as function words which are semantically empty.⁷ Here is the quote:

Negatives, conjunctions, sentence connectors, and subordinators may be termed function words or structure signaling words. The point of these labels is that such words are nearly lexically empty, i.e. they have little or no dictionary meaning of their own. However, they are grammatically significant in indicating the structure of sentences and parts of sentences. Some of them are so common as to require acquaintance at the grossest level of the language. This simply means that one must learn how they function early in the process. One may guess at the meaning of lexically full words, or leave them blank when reading, but one must know the grammatical “meaning” of function words to be able to proceed at all.⁸

Summary

To be sure, cognitive linguistics has greatly advanced our understanding of conjunctions. The notion of each one providing a unique cognitive constraint  proves far more useful in bridging the mismatches between languages than a sense-based approach based on the different glosses. But the call for a more unified, functional understanding of these words is not new. Even though the older grammarians may have had a clunky way of describing things, Winer, Funk and others seem to be in fundamental agreement with the functional direction advocated by Blakemore using cognitive linguistics.

Now in claiming that each conjunction has a unique constraint, I am not saying that there aren’t any diachronic or extra-linguistic issues that need to be taken into account. There are. But there is little hope of being able to adequately handle these extra complexities if we can’t even clearly describe the core meaning/function from a synchronic, monolingual environment. The latter is where I am headed in my discussion of exceptive clauses; the former I’ll leave to more qualified folks like Randall Buth.

Happy new year, I hope you stick around for this ongoing series. Depending on the level of interest, it may turn into an SBL paper proposal.

Isn’t grammar wonderful?