Abraham, Faith, And Works: Paul’s Use Of Scripture In Galatians 3:6-14

by Moisés Silva

Moiss Silva has taught Biblical Studies at Westminster Theological Seminary, Westmont College, and Gordon-Conwell Theological Seminary.

Few passages in the Pauline literature have received as much attention as vv. 10–14 in chapter 3 of Galatians. Thus the numerous exegetical problems we face here are greatly compounded by the increasingly large number of attempts to solve them. If I hope to say anything meaningful at all, selectivity is key. And beyond selectivity, an effort must be made to have as well-focused a goal as possible. Oddly enough, my decision to extend the limits of the passage by including vv. 6–9 helps to define the discussion more narrowly, for it makes even clearer that I cannot hope to treat all the questions that surface here like bristles on a porcupine. Although some of those details must be carefully considered, my primary purpose is to focus on the basic question of how and why Paul brings these quotations together as he does.[1]

I. The Structure of the Passage

6Consider Abraham: “He believed God, and it was credited to him as righteousness.” [Gen 15:6] 7Understand, then, that those who believe are children of Abraham. 8The Scripture foresaw that God would justify the Gentiles by faith, and announced the gospel in advance to Abraham: ‘All nations will be blessed through you.” [Gen 12:3; 18:18] 9So those who have faith are blessed along with Abraham, the man of faith. 

10All who rely on observing the law are under a curse, for it is written: “Cursed is everyone who does not continue to do everything written in the Book of the Law.” [Deut 27:26] 11Clearly no one is justified before God by the law, because, “The righteous will live by faith.” [Hab 2:4] 12The law is not based on faith; on the contrary, “The man who does these things will live by them.” [Lev 18:5] 13Christ redeemed us from the curse of the law by becoming a curse for us, for it is written: “Cursed is everyone who is hung on a tree.” [Deut 21:23] 14He redeemed us in order that the blessing given to Abraham might come to the Gentiles through Christ Jesus, so that by faith we might receive the promise of the Spirit. [NIV]

The decision to include vv. 6 9 in the present discussion is easily justified. Without denying that v. 10 introduces a new thought, we should do justice to the function of γάρ in that verse. It is true that Paul can use this conjunction as a simple transitional particle, and thus without a clear and specific logical connection.[2] Such a use, however, is not typical, and in this particular case it is highly unlikely, as the discussion below should confirm.

Moreover, the sheer number of citations within such short compass—six of them in the course of seven or eight sentences—is worthy of note.[3] The first two citations come from Genesis (15:6 and 12:3/18:18); the third and last are taken from Deuteronomy (27:26[4] and 21:23); the fourth and fifth constitute the (in)famous coupling of Hab 2:4 with Lev 18:5. Notice, moreover, that both the initial statement (v. 6) and the concluding comment (v. 14) highlight the person of Abraham and the function of faith. While I am not ready to argue against the conventional wisdom of beginning a new paragraph at v. 10, it does seem to me that more attention should be given to the coherence of the larger section.

One initial exegetical question that cannot be ignored is whether v. 6 constitutes the beginning of a new paragraph (so Nestle-Aland) or whether it is more closely connected with the section that began at v. 1 (so UBSGNT, which begins a new paragraph at v. 7). The question hangs largely on the function of καθώς here. Longenecker, after pointing out that nearly everyone takes the adverb as an abbreviation for καθώς γέγραπται, remarks approvingly that some versions (such as JB, NEB, and NIV) “have begun to treat καθώς here as an exemplum reference,” and so he opts for the rendering, “Take Abraham as the example.”[5] His use of the Latin term gives the impression that such a use of καθώςwas an established rhetorical feature, but he gives no parallels and I have not found any such function for the adverb elsewhere. While his rendering may be fully justified on the grounds that English style frowns on sentence fragments, we should avoid attributing to the Greek syntax a function that occurs to us because a suggested English translation seems felicitous.

The simplest explanation is that given by Bauer, who notes that sometimes the accompanying clause has to be supplied from the context.[6] If so, what needs to be supplied is obvious, as Lightfoot recognized: “The answer to the question asked in the former verse [i.e., ‘of works or of faith?’ v. 5] is assumed, ‘Surely of faith: and so it was with Abraham.’”7 My preference, therefore, is to begin the new paragraph at v. 7 rather than v. 6, but I would not insist on this point, so long as we recognize how close is the connection between vv. 5 and 6.

At any rate, we are now in a position to lay out the structure of the argument, using the OT citations as the logical markers. As the chart below indicates, I am suggesting that the biblical citations provide the grounds for five different theses. I also wish to stress, however, that in each case the logical connection is unclear.

first thesis (implied): [ὁ θεὸς ἐπιχορηγεῖ… ἐξ ἀκοῆς πίστεως]
grounds (v. 6): ᾿Αβραὰμ ἐπίστευσεν τῷ θεῷ, καὶ ἐλογίσθῃ αὐτῷ εἰς δικαιοσύην

second thesis (vv. 7, 9): οἱ ἐκ πίστεως, οὗτοι υἱοί εἰσιν ᾿Αβραάμ …
οἱ ἐκ πίστεως εὐλογοῦνται σὺν τῷ πιστῷ ᾿Αβραάμ
grounds (v. 8): ἐνευλογηθήσονται ἐν σοὶ πάντα τὰ ἔθνη

third thesis (v. 10a): ὅσοι ἐξ ἔργων νόμου εἰσὶν ὑπὸ κατάραν εἰσίν
grounds (v. 10b): ἐπικατάρατος πᾶς ὃσ οὐκ ἐμμένει πᾶσιν τοῖς γεγραμμένοις ἐν τῷ βιβλίῳ τοῦ νόμου τοῦ ποιῆσαι αὐτά
[assumed premise: all are disobedient (?)]

fourth thesis (v. 11a): ἐν νόμῳ οὐδεὶς δικαιοῦται παρὰ τῷ θεῷ
grounds (v. 11b): ὁ δίκαιος ἐκ πίστεως ζήσεται
stated premise (v. 12a): ὁ νόμος οὐκ ἔστιν ἐκ πίστεως
grounds of stated premise (v. 12b): ὁ ποιήσας αὐτὰ ζήσεται ἐν αὐτοῖς

fifth thesis (v. 13a): Χριστὸς ἡμᾶς ἐξηγόρασεν ἐκ τῆς κατάργας του νόμου γεμόμενος ὑπὲρ ἡμῶν κατάρα
grounds (v. 13b): ἐπικατάρατος πᾶς ὁ κρεμάμενος ἐπὶ ξύλοου

conclusion (v. 14): ἵνα εἰς τὰ ἔθνη ἡ εὐλογία τοῦ ᾿Αβραὰμ γένηται ἐν Χριστῷ ᾿Ιησοῦ, ἵνα τὴν ἐπαγγελίαν τοῦ πνεύματος λάβωμεν διὰ τῆς πίστεως.

(1) The first thesis—only implied in v. 6, but elaborated earlier in vv. 2–5—is that the Galatians received the Spirit through (the hearing of) faith, and this affirmation is supported by citing Gen 15:6. Curiously, though, this OT passage says nothing about the Holy Spirit.

(2) Paul next, in v. 7, states that the true children of Abraham are “the ones of faith,” and in support cites Gen 12:3 (but using the words πάντα τὰ ἔθνη from 18:18 in place of πᾶσαι αἰ φυλαι), although that passage says nothing about faith. Immediately Paul restates the thesis in v. 9, repeating the phrase “the ones of faith” and describing such people as the ones who are blessed along with Abraham.

(3) Third, the apostle directs our attention in v. 10 to “as many as are of the works of law” and affirms that they are under a curse, a thesis supported by an appeal to Deut 27:26. As is well known, much of the current controversy regarding the passage as a whole focuses precisely on the logical problem presented by Paul’s use of this citation. On the face of it, Deut 27:26 (which curses those who disobey the law) seems to state the precisely opposite point that Paul wants to make (those intent on doing the law are cursed). For the moment, let us simply note that the traditional—though hotly disputed way of understanding the argument is to suggest that Paul is assuming the universality of disobedience.

(4) The fourth thesis (v. 11), which Paul describes as “evident” (δῆλον), is that through the law no one can be made right with (or considered righteous by) God, and the grounds for this denial is Hab 2:4, a passage that says nothing about the law. It soon becomes clear, however, that in Paul’s mind what proves the thesis is not the Habakkuk reference by itself but the combination of that passage with Lev 18:5. I have chosen to treat the Leviticus citation as the grounds, not for a separate thesis, but for a premise that Paul now realizes he needs to spell out, namely, that the law cannot be viewed as belonging to the category “of faith.” At any rate, questions about Paul’s logical abilities or even about his honesty are most likely to arise from his use of Habakkuk and Leviticus here.

(5) The last thesis (v. 13), which is introduced without the use of a conjunction or any other transitional particle, relates Paul’s argumentation to gospel history by affirming that, through the crucifixion, Christ provided redemption from the curse of the law (presumably the curse mentioned in v. 10). The grounds he offers for this interpretation of the cross is Deut 21:23, a passage that of course says nothing about Christ or about redemption or even about crucifixion as such.

Finally, the two purpose clauses of v. 14 are intended to take the argument to its climax, and they do so by bringing together six crucial concepts: Abraham (vv. 6–9), Christ (vv. 1, 13, both of which verses mention the crucifixion), the Gentiles (v. 8), the promise/blessing (vv. 8–9), the reception of the Spirit (vv. 2, 5), and faith (vv. 2, 5, 6–9, 11–12). It is difficult not to be impressed, on the one hand, by the care and effectiveness with which these various themes have been interweaved and, on the other hand, by the glaring gaps in the argumentation.

The decision to look at this passage as consisting of five theses, each of which is supported by an OT citation, has certain advantages, but it can also obscure other features.8 In particular, it could be argued that what I am calling the first thesis is much more than just one item out of several; it is not adequate even to say that this thesis is the most important in the group. In effect, what Gen 15:6 is intended to demonstrate is the overarching burden of the whole passage (or even the whole chapter), and so it might be a more accurate representation to view theses two through five as subordinate arguments.[9]

In spite of these concerns, however, the proposed structure can prove useful. As long as we do not think of such an outline (or any other outline) as being right in some exclusive sense,[10] it does highlight important logical elements and can therefore serve us well as a preliminary framework within which to understand Paul’s use of Scripture. And in any case there is always the option of altering the outline after further reflection.

II. Light from Paul’s Hermeneutical Milieu

1. Abraham in Second Temple Judaism

That Paul, as a Jew, should appeal to Abraham in support of his teaching is hardly surprising. Given the significance of the patriarch in the Genesis narrative, it is understandable that the figure of Abraham looms large in all expressions of Judaism.[11] And a major emphasis found in the literature is, of course, Abraham’s faithfulness in the midst of trials.

Interestingly, however, Gen 15:6 does not play as prominent a role in that literature as one might expect. The Midrash on Genesis, a fifth-century homiletical commentary, has only a very brief paragraph on that verse: merely three sentences that say nothing about Abraham’s believing or about his faith being counted as righteousness (Gen. Rab. 44:13). Moreover, a search for references to Gen 15:6 in the whole of Midrash Rabbah yielded precious few instances.[12] This rabbinic literature, of course, is not a precise reflection of Judaism during the Second Temple, but if Gen 15:6 had played a significantly greater role during that period, it seems unlikely that the later rabbinic tradition would not have preserved it.

In any case, allusions to this verse are consistently tied to Abraham’s faithful obedience, particularly as expressed in the Akedah (the binding of Isaac), and one recurring concern is to point out that Abraham obeyed the Torah even though it had not yet been given to the Israelites. This fact immediately suggests a difference between Judaism and Paul. As Hansen puts it, “In contrast to the use of Abraham in much contemporary Jewish literature, Paul dissociates the Abrahamic promise and its blessing from the law and works of the law. This dissociation is designed to explode any attempt to use Abraham as an example for circumcision and law-observance.”[13]

Similarly, Garlington points out that Paul places Abraham “in the same arena as the Gentiles,” even though Abraham, having converted from paganism, “was the perfect model for the Jewish missionaries.” He concludes: “It is just this un-Jewishness of Paul’s use of Abraham that provides the bridge into his cursing of his opponents; that is, in the salvation-historical purposes of God, the paradigm of eschatological justification is provided not by the Torah, but by Abraham, who had nothing to do with Torah.”[14]

2. Deuteronomy in Second Temple Judaism

Recent research into the use of Deut 21:23 and 27:26 by Paul’s contemporaries has focused on two fairly specific issues. One is the question whether the first passage was understood with reference to capital punishment by crucifixion. Whereas rabbinic tradition understood the hanging as public exposure subsequent to death, the Temple Scroll from Qumran alters the word order of 21:22 (from “he will be put to death, and you will hang him on a tree” to “you shall hang him on the tree, and he shall die”; 11Q19 LXIV, 7–13) and thus appears to understand the hanging as a reference to the execution itself. To be sure, the evidence is not conclusive that crucifixion as such was in view.[15] Even without that evidence, however, many scholars are inclined to think that the reason Paul quotes Deut 21:23 is that it was already used by Jews in their anti-Christian polemic.[16]

The second issue has to do with the way in which the curses of Deut 27–32 were understood by postexilic Jews. James M. Scott, following M. Knibb, has argued forcefully that Daniel’s prayer of confession, which alludes to those curses (see especially Dan 9:11), gives expression to a point of view widely held in Second Temple Judaism, namely, that as a result of Israel’s disobedience, the nation would remain under the curse of exile until the eschatological restoration. Then, building on the work of N. T. Wright, Scott infers that this is the point of view Paul—for whom restoration had already come—is reflecting in Gal 3:10.[17] If so, it would follow that the phrase “those of the works of the law” refers to the nation of Israel as a whole, while the wrongdoing that brings a curse is not some specific sin but simply disobedience to God in a more general sense.

This solution, however, labors under some difficulties. One of them is the ambiguity of the evidence that Israel’s exile functions as a substantive hermeneutical paradigm for Paul. Although this approach is becoming increasingly popular and some scholars seem to regard it as proven fact, the arguments are far from conclusive.[18] In my opinion, it is not unreasonable to suspect that preoccupation with the exile may have played a role in Paul’s thinking, but when drawing exegetical conclusions one should hesitate to lean heavily on a concept that the apostle never mentions explicitly and to which he does not even clearly allude. But even if it could be proven that Paul shares this understanding, can we assume that he must be alluding to it in this specific quotation? After all, the apostle frequently uses scriptural texts in ways that differ from, or even contradict, those of his contemporaries, so we can hardly assume that his use of Deut 27:26 here conforms to theirs.

3. Hab 2:4 and Lev 18:5 in Second Temple Judaism

It goes without saying that Paul’s contemporaries would have viewed Lev 18:5 in a totally positive way, and the thought of pitting this verse against Hab 2:4 must have appeared to them every bit as surprising as it does to us. For the latter text we now have, of course, the pesher from Qumran, which interprets the passage as a reference to the doers of the law in Judah, whom God will deliver from condemnation because of their suffering and their faithfulness to the Teacher of Righteousness (1QpHab VIII, 1–3). Although James Sanders has sought to highlight the similarities between this interpretation and Paul’s,[19] the truth is that on the most crucial question raised by the apostle’s use of this text, the sectarians appear to take exactly the opposite position. For them, Habakkuk does not even suggest a tension between faith(fulness) and obedience to the law; on the contrary, that text serves as grounds for legal observance.

4. Summary

It would of course be possible to multiply references from Second Temple Judaism that may have some relevance for our understanding of how these OT passages functioned during that period. I am not aware, however, of any evidence that would alter the picture emerging from the material already cited. The value of that material is largely negative in character, that is, it highlights the differences between Paul and his contemporaries. But “negative” here does not have a pejorative nuance. Sometimes we learn more from discovering contrasts than similarities. In this particular case, appreciating how distinctive is Paul’s handling of the Bible can significantly enhance our understanding of the text.

III. As Many as Are of the Works of the Law

When one looks over the recent literature on Gal 3:10–14, it becomes increasingly apparent that the heart of the disagreement—and therefore a key to the proper understanding of Paul’s thought here—lies in the identification of ὅσοι ἐξ ἐργων νόμου εἰσίν. It should give us pause that Paul makes no effort to define the phrase; he assumes that his readers will have little difficulty understanding it. That alone is something of a clue. But before we can make progress on this question, we need to note that scholars, in their efforts to identify the “meaning” of the phrase, have not been careful to draw some important distinctions, especially the contrast between reference and sense.[20]

(1) Who are the possible referents? Jews in general? Certain kinds of Jews only? The Judaizers in particular? Gentiles who adopt Judaism? People (whether Jewish or not) who are “legalistic”?

(2) As for the sense, two different issues need to be addressed. (a) First is the meaning of the phrase ἔργα νόμου by itself.[21] Is it a neutral term indicating the requirements of the law generally? Does it have the specific meaning of ceremonialism or even “legalism”? Is it a way of speaking of sociological identity badges? (b) The second concerns the force of the construction οἱ/ὅσοι ἐξ. Does the preposition ἐκ retain its frequent meaning, “out of”? Does this particular construction convey the idea of “reliance on” (cf. NIV)? Should we view it as merely descriptive?

Concerning (2.b), the function of οἱ/ὅσοι ἐξ, I would suggest that a minimalist approach should be our starting point. There is no value in prejudging the question by translating the construction in a way that not everyone would accept. Perhaps the least prejudicial rendering would be something like “as many as are characterized by works of law.” It may well be that the phrase has a more specific semantic content than that, but let us not try to decide the question just yet.

As for (2.a), the sense of ἔργα νόμου by itself, here again it may be wise initially to look for “the least meaning,” that is, the most general sense. It is evident that the quotation from Deut 27:26, which immediately follows, has in view general obedience to the things written in the law. To be sure, Paul would expect his readers to understand the phrase in the light of how he has already used it in 2:16, that is, in a context where circumcision and the dietary laws are prominent.[22] It may well be the case, then, that in 3:10 Paul has in mind those specific features of Judaism that most obviously separate Jews from Gentiles. But the evidence is hardly conclusive. After all, one may readily agree that such features are present and even prominent in 2:16 without concluding that other, more general, elements should be excluded even there. We should therefore, at least for the time being, work with the general notion of obedience to the law and wait to see whether indeed we find the need to be more specific than that.

However, the simple fact is that the phrase ἔργα νόμου does not appear in 3:10 by itself but in a particular syntactical construction. Even assuming that Paul can elsewhere speak of “the works of the law” in a neutral or positive sense, the syntagmatic combination in this particular context clearly gives the phrase a negative nuance.[23] Scholars have not always appreciated that the meaning of the whole construction is to a large extent determined by its opposition to οἱ ἐκ πίστεως, that is, “the good guys.” The latter, Abraham’s true children, are blessed; in distinction from them, those who are of the works of the law are cursed.

Now the opposition between faith and works of the law—more precisely; between “of faith” and “of works of the law”—had been established clearly in 2:16.[24] It was then repeated in 3:2 and 5, becoming the operative principle for the rest of the passage. In effect, Paul defines those who are of the works of the law negatively, namely, as those who are not of faith. As obvious as that may appear, it has not always affected the scholarly discussion as it should have (probably because of the failure to see the integral connection between vv. 6–9 and 10–14).

This consideration alone helps us to eliminate one of the options regarding the referent of ὅσοι ἐξ ἔργων νόμου, namely, the view that it refers to Jews as a whole.[25] It is implausible that Paul would indiscriminately describe his fellow-Jews as people not characterized by faith. And it is simply out of the question to suggest that Paul would have thus regarded Abraham and David.[26]

But before we can make further progress on the question of the phrase’s reference, more needs to be said regarding the sense of the prepositional construction. Since prepositions are most frequently ruled by verbs, we could rephrase the question: What is the verbal idea implicit in the construction? Here again the context gives us the help we need, for Paul immediately quotes Hab 2:4, ὁ δίκαιος ἐκ πίστεως ζήσεται, and the verb ζάω is repeated in the quotation from Lev 18:5.[27] In other words, οἱ ἐκ πίστεως are those who, like Abraham, live by faith and are blessed; while ὅσοι ἐξ ἔργων νόμου) are those who live by the things commanded in the law and are cursed. And in view of the well-established correlation between the concepts of life and justification,[28] it is difficult to avoid identifying these people with those who seek to be justified by the law (5:4, οἵτινες ἐν νόμῳ δικαιοῦσθε). It turns out, then, that a rendering such as “those who rely on works of the law,” although much maligned by some recent scholars,[29] is hardly inimical to the context.

It might appear from this way of stating the matter that we have an answer to the question of what the referent is, and that the answer is “legalists.” In fact, however, we are not at all ready to draw that conclusion. To begin with, the term is riddled with ambiguity.[30] Second, we have not yet resolved the question whether “the things written in the book of the law” has a general meaning (which we have used only as the initial working hypothesis) or a more restricted sense. Third, we have not dealt with the difficulty that has greatly exercised some scholars regarding the citation from Deut 27:26, namely, the fact that this verse seems to place a curse on antinomians (those who do not obey the law) rather than on “legalists” (those who go out of their way to obey every detail). We move on to this last question as a means of reaching firmer conclusions.

IV The Riddle of Gal 3:10

If we wish to make sense of Paul’s argument in this verse, there are a couple of preliminary considerations that should be kept in the forefront. In the first place, we ought not to assume (as most students of this verse seem to do) that we have here an exceptional logical problem. The truth is that, as we noticed earlier, every single citation in vv. 6 14 is characterized by some kind of logical gap; that is, Paul does not trouble himself with spelling out the premises that make his thinking cogent. One of the most significant gaps is the lack of an explicit connection between the giving of the Spirit and Abraham’s faith, yet, strangely, commentators and scholars seldom even mention the problem.[31] The only time Paul bothers to give some sort of explanation is after quoting Hab 2:4 and in that case it has seemed to most commentators that Paul’s cure is worse than the disease!

I should add parenthetically that this feature in Paul’s use of Scripture has important points of contact with that of the rabbis. Even the most skeptical rabbinic scholar will agree that the kind of compressed argumentation found in the Mishnah has a very long prehistory. The same is true, mutatis mutandis, of the numerous citations in other rabbinic documents, where the connection between the scriptural passage on the one hand and the point being addressed on the other is not immediately obvious. As I have argued elsewhere, it is a grave mistake to infer that in every case the connection is artificial. That may well be true in some instances, but typically the gap is the result of assumed and agreed-upon premises that need not be spelled out.[32] (We ourselves, in ordinary conversation with family and close friends, use this “technique” far more often than we realize.)

The second consideration that must guide our thinking here is the pivotal significance of v. 10 for the argument as a whole. Earlier I suggested that the γάρ in this verse very likely preserves its causal function; in fact, however, the point needs to be made more strongly than that. Back in v. 7, Paul’s emphatic form of expression—the word order and, especially, the inclusion of οὗτοι—implied an opposition between the ones of faith and some other group. It was back then that Paul had thrown down the gauntlet, putting us on notice that there are people around who are not of faith and who therefore do not partake of the Abrahamic blessing. But now the apostle must make good on his implicit claim, and so it is not very likely that at this point he would let his guard down or become either careless or arbitrary.

At any rate, these two considerations create the presumption that Paul was counting on his readers’ ability to fill in the gaps. Undoubtedly they shared certain items of information that did not need to be explicitly formulated. Some of these items may have included general beliefs common to early Christianity; others may have consisted of specific communication, especially during Paul’s visit(s) to Galatia. Implicit data in any text can usually be inferred from the context—understanding context in the most general sense (that is, including additional knowledge we have about authors based on other writings).

Moreover, the very paradox that troubles scholars in this verse (those who do the works of the law are cursed because Deuteronomy says that those who do not do the works of the law are cursed!) should tell us something about what is motivating Paul. The fact is that the apostle nowhere (in Galatians or in his other letters) characterizes his opponents as people who are obedient to the law. He will admit to no such thing. In this very epistle, as many have pointed out, he specifically accuses them of not keeping the law (6:13).[33] And in Phil 3:2–4, when describing a group of opponents who, to say the least, had some affinities with the Judaizers in Galatia, he deliberately depicts them as pagans.[34] That general conviction could hardly have been foreign to the Galatian Christians. There is in fact every reason to believe that when they heard Paul describing his opponents as being of the works of the law, these Galatians knew that by that phrase he did not mean something like, “these are the people who fulfill the law”! Or to put it differently, the Galatians could perfectly well understand (whether they agreed or not) why Paul would think of his opponents as people who did not “remain in all the things written in the book of the law to do them.”

It would appear then that the assumed premise is not precisely the principle that all people fail to keep the law. That Paul believed in such a principle—and that the Galatians knew it—seems to me beyond dispute (after all, the nature and urgency of the gospel’s call to repentance makes little sense if there are people around who do succeed in keeping the law).[35] Moreover, the principle is not totally irrelevant to the understanding of this verse. But I want to argue that the specific item of information that supplies the missing premise (since it was information shared by Paul and his readers) was Paul’s conviction that his “faith-less” opponents in particular were the ones who failed to fulfill the requirement of Deut 27:26.[36] We could even say that the premise is built into the way Paul introduces the citation, namely, by describing the false teachers as being characterized by works (and therefore as not being children of faithful/ believing Abraham). In short, the quotation functions as Paul’s way of informing or reminding the Galatians of how the Judaizers should be regarded.

So we have now reached a conclusion regarding the reference of ὅσοι ἐξ ἔργων. Paul had primarily in mind not people in general (Jews and Gentiles) nor Judaism as a whole, but rather the Judaizers he opposes and, therefore, anyone—Jew or Gentile—who followed the Judaizers’ teaching.[37] As to the sense of the phrase, we had earlier reached the preliminary conclusion that it means, “as many as seek to live [= be justified] by the things commanded in the law.” But more needs to be said.

V. Justification and Faith

The quotation from Deuteronomy is intended to prove the accusation that Paul’s opponents are under a curse and are thus to be distinguished from those who “are blessed with faithful/believing Abraham” (v. 9). But the apostle has so far only assumed, without demonstration, that these opponents are not characterized by faith. That is why he needs to formulate the fourth thesis, “no one is justified by the law.” Notice that v. 11, which begins with δέ, introduces an additional piece of information by revealing the principle that allows Paul to characterize his opponents as “faith-less.” Thus it would be possible (as mentioned earlier) to view v. 11 a not as a distinct thesis but as a corollary of the thesis in v. 10.

If so, it may well be that Paul understands v. 10 as already giving expression to the concept spelled out in v. 11a. In other words, to say that those who seek to live or be justified by the works of the law are under a curse is in effect to say that it is not possible to be justified by the law. This negation, embedded in v. 10, is brought into prominence in v. 11a and supported by the combination of Hab 2:4 and Lev 18:5.

By regarding v. 12 as a stated premise, I am really suggesting that it would have been in character for Paul to omit that verse, in which case we would be facing the same problem we do in v. 10—a gap in the argumentation. One is tempted to wish the apostle had in fact left it out here as well. It might have been easier to fill the gap ourselves than to figure out what he had in mind by his statement that the “law is not of faith”!

This paper would be extended many times over if it were to deal adequately with the problems raised by that comment. Moreover, the discussion would turn into a treatment of Pauline theology rather than of Paul’s use of Scripture. I would suggest, however, that if we wish to understand the logic of his argumentation here, it is not really necessary to solve the conceptual problem of how he viewed the larger question of the relationship between law and faith.

In one sense, we ought not to be surprised by v. 12, since it appears to be one more expression of the opposition between works of law and faith an opposition initially formulated in 2:16, repeated in 3:2 and 5, alluded to in 3:3 (ἐναρξάμμενοι πνεύματι vs. σαρκὶ ἐπιτελεῖσθε), and restated in the contrast between 3:9 and 3:10. The restatement in v. 10 includes, of course, the quotation from Deut 27:26, which has the effect of linking the substantive ἔργων with the verb ποιῆσαι. Therefore, when in v. 12 Paul tells us that the “law,” which has to do with “doing,” is not “of faith,” perhaps he does not intend to say much more than what the context has already expressed up to this point.

Admittedly, it is difficult to read this bald statement and reduce it to a mere opposition between works of law and the hearing of faith. But there is also an exegetical difficulty involved in thinking that Paul, for no obvious reason, jeopardizes the persuasiveness of his argument by dumping out of the blue a star-fling, programmatic comment about the non-faith (or even anti-faith) character of the law. Only a few verses later Paul will forcefully deny the inference that the law is against the faith-based promise (and in Rom 3:31 he affirms just as forcefully that faith establishes the law).

If v. 12 does add semantic content to what the previous verses have already expressed, let me suggest this new datum may be no more than an anticipation of the chronological, redemptive-historical distinctions that he begins to develop in v. 15 (the law came centuries after the Abrahamic covenant), climaxing in his affirmation that the time of the law preceded the time of faith (vv. 19 and 23–25).[38] In other words, to say that “the law” (if this means more than “the works of the law”) is not of faith is to claim that the Sinaitic Covenant belongs to a different redemptive-historical epoch than the gospel.[39]

VI. Conclusion

Although this study of Gal 3:6–14 has not dealt with every exegetical problem raised by the passage, it may serve to shed light on Paul’s use of the OT. Insofar as we may speak of the apostle’s hermeneutics, it is clear that his choice of citations is determined by the polemics in which he is engaged, and not (at least not primarily) by a dispassionate exegesis of the texts. But that is not to say that his interpretative decisions are arbitrary. A recognition that he typically does not spell out his logical moves can help us appreciate the legitimacy of his approach. And close attention to the broader context of the apostle’s argumentation (e.g., reading v. 10 in the light of v. 7) reveals a more tightly reasoned discussion than is generally thought.

In addition, the present study has some theological implications, for it would appear to support, in some important respects, the so-called Lutheran interpretation of Paul’s teaching on the law, and in particular the traditional understanding of the contrast between faith and works. However, lest it be thought that these ideas originated with some Protestant Reformer oppressed by his introspective conscience,[40] I conclude with two quotations. One comes from as early as the second century:

For the whole human race will be found to be under a curse. For it is written in the law of Moses, “Cursed is every one that continueth not in all things that are written in the book of the law to do them.” And no one has accurately done all, nor will you venture to deny this [καὶ οὐδεὶς ἀκριβῶς πάντα ἐποίησεν, οὐδ᾿ ὑμεῖς τολμήσετε ἀντειπεῖν]; but some more and some less than others have observed the ordinances enjoined. But if those who are under this law appear to be under a curse for not having observed all the requirements [διὰ τὸ μὴ πάντα φυλάξαι], how much more shall all the nations appear to be under a curse who practice idolatry, who seduce youths, and commit other crimes?[41]

The second quotation, which brings out the significance of the prepositional construction in the key phrase, comes from the Angelic Doctor himself:

And it should be noted that the Apostle does not say, “As many as observe the works of the Law are under a curse,” because this is false when applied to the time of the Law. He says rather: as many as are of the works of the Law, i.e., whosoever trust in the works of the Law and believe that they are made just by them are under a curse. For it is one thing to be of the works of the Law and another to observe the Law. The latter consists in fulfilling the Law, so that one who fulfills it is not under a curse. But to be of the works of the Law is to trust in them and place one’s hope in them. And they that are of the Law in this way are under a curse,…, therefore, inasmuch as the Law begets a knowledge of sin and offers no help against sin, they are said to be under a curse, since they are powerless to escape it by those works.[42]

Notes

1. This paper was presented as the first DuToit-CUM Prestige Lecture at the University of Pretoria, South Africa, on February 22, 1999, and I am delighted now to offer it for publication in honor of Professor A. B. duToit, to whom I am greatly indebted for his contributions to NT scholarship. The paper was prepared originally for the New Testament Colloquium of the Boston Theological Institute (November 1997), and later versions were read at Concordia Theological Seminary (Fort Wayne, Indiana, January 1999) and Westmont College (Santa Barbara, California, February 2000). I am thankful for the valuable comments I received from colleagues at these institutions.
2. Cf. M. Silva, Explorations in Exegetical Method: Galatians as a Test Case (Grand Rapids: Baker, 1996), 82–83.
3. A comparable density is found only in the catena of Rom 3 and in the torrent of Rom 9–10. However, the way the quotations are strung together in Gal 3 has its own distinctiveness.
4. The LXX text of Deut 27:26 has πᾶσιν τοῖς λόγοις τοῦ νόμου τούτου (= MT את־דברי התורה־הזאת) instead of πᾶσιν τοῖς γεγαμμένοις ἐν τῷ βιβλίῳ τοῦ νόμου, which Paul apparently draws from Deut 29:19 (cf. 28:58). Notice also that although the word כל is missing in the MT of 27:26, it occurs in 28:58, which is conceptually parallel. For more textual details on these quotations, see especially Dietrich-Alex Koch, Die Schrift als Zeuge des Evangeliums. Untersuchungen zur Verwendung und zum Verständnis der Schrift bei Paulus (BHT 69; Tübingen: J. C. B. Mohr [Paul Siebeck], 1986), 163–65 et passim.
5. Richard N. Longenecker, Galatians (WBC 41; Dallas, Tex.: Word Books, 1990), 112. He comments that the absolute use of καθώς, which is rare, appears “only here in Paul,” but that depends on precisely how one defines “absolute”; cf. in particular Phil 1:7. On the use of exempla in ancient rhetoric, cf. Quintilian, Institutio oratoria 5.11.6.
6. BDAG s.v. καθώς1; the parallel given, 1 Tim 1:3, illustrates the principle well.
7. Joseph Barber Lightfoot, Saint Paul’s Epistle to the Galatians (10th ed.; London: Macmillan, 1898), 67. Cf. also Christopher D. Stanley, Paul and the Language of Scripture: Citation Technique in the Pauline Epistles and Contemporary Literature (SNTSMS 69; Cambridge: Cambridge University Press, 1992), 235.
8. For example, the connection between v. 10 and v. 13 — namely, the way in which the word κατάρα seems to bracket this section—tends to get buried. Similarly, my indentation obscures the parallel structure of Hab 2:4 and Lev 18:5: note that the antithesis between those two citations is highlighted by their structural similarity, that is, the combination of a definite substantive (ὁ δίκαιος in one, the substantival participle ὁ ποιήσας in the other) plus a simple prepositional phrase (ἐκ πίστεως, ἐν αὐτοῖς) plus the future verb ζήσεται. My outline, furthermore, may be too symmetrical, giving the impression that the link between, say, any one pair of verses is of the same sort as every other pair; it seems likely, however, that the connections are not quite so neat. (For example, the second thesis may be viewed as a corollary of the first, and the fourth as a corollary of the third. If so, my treating the second and the fourth as separate affirmations could distort the argument.)
9. There are other instructive parallelisms that do not show up in my chart, such as v. 8a with v. 11a, v. 9 with v. 11b, and v. 10b with v. 12b. See G. Walter Hansen, Abraham in Galatians: Epistolary and Rhetorical Contexts (JSNTSup 29; Sheffield: JSOT Press, 1989), 120.
10. Cf. Silva, Explorations, 96.
11. For a readable synthesis of Jewish traditions about Abraham, see especially Louis Ginzberg, The Legends of the Jews (Philadelphia: Jewish Publication Society of America, 1925–38), 1:185–308. See also H. L. Strack and P Billerbeck, Kommentar zum Neuen Testament aus Talmud und Midrash (5 vols.; Mtmchen: Beck, 1922–56), 3:186–201 (on Rom 4:2) and Hansen, Abraham, 175–99. It should however be pointed out that in the literature from Qumran (other than the Genesis Apocryphon) the name of Abraham does not appear frequently. On the significance of 4Q225, cf. Manfred Oeming, “Der Glaube Abrahams. Zur Rezeptionsgeschichte yon Gen 15, 6 in der Zeit des zweiten Tempels,” ZAW 110 (1998): 27-30.
12. I found a passing reference in Exod. Rab 3:12 (on Exod 4:1) and a more significant use of the verse in 23:5 (on Exod 15:1 = pp. 282–83 in the Soncino edition), which comments that because of Abraham’s faith, the Israelites were permitted to sing the Song by the Sea. There are some ten other passages where Abraham’s righteousness is spoken of, sometimes at length. The search was performed on The CD ROM Judaic Classic Library (Chicago: Davka, n.d.)
13. Hansen, Abraham, 99. Cf. also Dirk U. Rottzoll, “Gen 15, 6—Ein Beleg far den Glauben als Werkgerechtigkeit,” ZAW 106 (1994): 21-27. Building on M. Oeming’s suggestion that Abraham is the subject of the verb ויחשבה in Gen 15:6, and then appealing to Job 19:11 and 33:10, he translates, “Und Abraham glaubte Gott und rechnete sich das [sc. sein Glauben] zur/als Gerechtigkeit an” (pp. 25–26). Moreover, on the grounds that צרקה is nomen actionis, Rottzoll argues that here faith has “die Qualität einer konkreten Handlung, eines yon ihm vollbrachten Werks!” (p. 26). He concludes (referring to Strack-Billerbeck, Kommentar, 3:186–87), that one can appreciate why Jews understood the verse as they did. But what evidence is there that Jewish interpretation took Abraham as the subject of the verb?
14. Don Garlington, “Role Reversal and Paul’s Use of Scripture in Galatians 3.10-13, ” JSNT 65 (1997): 94. With regard to Gen 12:3/18:18, one does find considerable material on Abraham as recipient of and channel for God’s blessing, but even here the contrast with Paul is especially striking. While Paul highlights the significance of the promise for Gentiles, the midrashic tradition does not at all focus on this specific element of the Abrahamic promise.
15. See B. Hudson McLean, The Cursed Christ: Mediterranean Expulsion Rituals and Pauline Soteriology (JSNT Sup 126; Sheffield: Sheffield Academic Press, 1996), 132–33.
16. More speculative—but suggestive and worth considering is the theory that Paul is reflecting a Christian midrash on the Akedah based on link-words. See M. Wilcox, “‘Upon the Tree’ Deut 21:22–23 in the New Testament,” JBL 96 (1977): 85-99.
17. James M. Scott, “‘For as Many as Are of Works of the Law Are Under a Curse’ (Galatians 3.10),” in Paul and the Scriptures of Israel (ed. C. A. Evans and J. A. Sanders; JSNTSup 83, Sheffield: JSOT Press, 1993), 198-201,213-14. The implications of this approach for chaps. 3–4 of Galatians more generally are explored by Scott.]. Hafemann, “Paul and the Exile of Israel in Galatians 3–4, ” in Exile: Old Testament, Jewish, and Christian Conceptions (ed. J. M. Scott; JSJSup 56; Leiden: Brill, 1997), 329–71.
18. Note the perceptive discussion by Mark A. Seifrid, “Blind Alleys in the Controversy over the Paul of History,” TynBul 45 (1994): 73-95. So far as Judaism is concerned, the evidence “suggests not a widespread conviction, but a range of views on the status of Israel, which varied with time, place, and argumentative purpose” (p. 89). Moreover, “the ‘exile’ motif appears to have functioned for Paul in a manner precisely opposite to that which Wright and Scott have proposed” (p. 91).
19. James A. Sanders, “Habakkuk in Qumran, Paul, and the Old Testament,” in Evans and Sanders, Paul and the Scriptures of Israel, 98–117.
20. This is the old question of whether, for example, the Morning Star and the Evening Star “mean” the same thing; of course they do (they have the same referent—Venus), and of course they don’t (they convey different senses—one heavenly body that is visible at dawn and one that is visible at dusk).
21. Much has been made of the fact that the apparently corresponding Hebrew expression, מעשׂי התורה, is used in the literature from Qumran to designate a sectarian understanding of the law, leading to behavior that distinguished the Qumran community from other Jewish groups. Michael Bachmann has argued that the term refers specifically to Qumranian halakah (“4QMMT und Galaterbrief, מעשׂי התורה und ΕΡΓΑ ΝΟΜΟΥ,” ZNW 89 [1998]: 91-113). It should be pointed out, however, that since the relevant passage speaks of “some of the works of the law,” it is evident that the expression “works of the law” encompasses a broader set of regulations than the ones actually treated in 4QMMT. Note also that the similar expression in 1QS V, 21, מעשׂיו בתורה, is paralleled by the expression in the next line, “all his statutes which he commanded to do.”
22. And it surely is significant that although Paul can speak of keeping and fulfilling the law in very positive terms (Rom 8:4; 1 Cor 7:11), one looks in vain for a similarly positive context where the phrase ἔργα νόμου appears.
23. Contra Scott, “For as Many as Are of Works of the Law Are Under a Curse,” 190. He argues that “ὅσοι ἐκ denotes origin,” but we need only recall Rom 2:8 (τοῖς δὲ ἐξ ἐριθείας) to realize that one cannot press the meanings of prepositions in that way. This reference, incidentally, also argues against Ardel Caneday’s view (“‘Redeemed from the Curse of the Law’: The Use of Deut 21:22–23 in Gal 3:13, ” TrinJ 10 [1989]: 192-94) that the construction identifies people by their orientation as opposed to their actions (a doubtful dichotomy in any case). J. Louis Martyn’s rendering is better, “those whose identity is derived from observance of the Law” (Galatians: A New Translation with Introduction and Commentary [AB 33A; New York: Doubleday, 1997], 307). Cf. Kari Kuula, The Law, the Covenant and God’s Plan. Vol 1: Paul’s Polemical Treatment of the Law in Galatians (Publications of the Finnish Exegetical Society 72; Helsinki: Finnish Exegetical Society; G6ttingen: Vandenhoeck & Ruprecht, 1999), 66.
24. James D. G. Dunn (Jesus, Paul and the Law: Studies in Mark and Galatians [Louisville, Ky.: Westminster John Knox, 1990], 195–96) has argued, unconvincingly, that in the first part of 2:16, the clause ἐὰν μὴ διὰ πίστεως should be understood not adversatively but exceptively (thus, “no one is justified by works of the law unless [he is also justified] by faith in Jesus Christ”). Even he, however, recognizes that the last part of the verse establishes the opposition, and this is the theme that permeates chap. 3.
25. For example, Joseph R Braswell argues, “The reference is not to legalists, Judaizers, or all of unredeemed humankind, but to Jews in their special identity and distinctiveness provided by a torah lifestyle” (“‘The Blessing of Abraham’ versus ‘The Curse of the Law’: Another Look at Galatians 3:10–13, ” WTJ 53 [1991]: 77). My objection applies also to the view that restricts the reference only temporally, as in the following quotation: “Paul was making an assumption here that was fairly common in Jewish sources of the Second Temple period: the curses of Deut. 27–32 had indeed fallen upon Israel in (722 and) 587 BCE, and would remain upon the nation until the time of the messianic redemption and the restoration” (Scott, “For as Many as Are of Works of the Law Are Under a Curse,” 221).
26. It is quite baffling to read that “the only children that Abraham has are the Christians” and that therefore “Paul did not interpret the Christ-event as God’s way of bringing Gentiles into the children of Abraham, since before the ‘Christian’ faith there could be no people of faith and promise” (Kuula, The Law, the Covenant and God’s Plan, 65, his emphasis). Paul’s most basic thesis is that Christians believe after the pattern of Abraham; they are his children precisely because they believe as he did. What sense then does it make to say that prior to Christ’s coming “people of faith and promise” did not exist?
27. My argument assumes that ἐκ πίστεως in Gal 3:11 should be construed with the verb rather than with the substantive, a view strongly supported by the clear formal parallel between the Habakkuk and Leviticus citations (with regard to the latter, no one would argue that ἐν αὐτοῖς should be construed with ὁ ποιήσασς. If it had been important for Paul to link the prepositional phrase with the substantive, he could have easily, as pointed out by many, changed the word order to ὁ ἐκ πίστεως δίκαιος; and certainly Paul had no misgivings about altering the LXX wording when he wished to make a point. Norman H. Young, in a very interesting article (“Who’s Cursed and Why? [Galatians 3:10–14],”JBL 117 [1998]: 79-92, esp. 89), tries to preserve the parallelism by inserting the words ἐν αὐτῇ at the end of the Habakkuk citation, but this solution seems quite artificial and only highlights the problem involved in construing “by faith” with the substantive rather than with the noun. On the legitimacy of the Habakkuk citation more generally, see M. Silva, “Old Testament in Paul,” in Dictionary of Paul and His Letters (Downers Grove, Ill.: InterVarsity, 1993), 630–42.
28. In this very passage, δικαιοῦται stands in relation to ζήσεται (v. 11). Notice also the parallelism between 2:21 (εἰ γὰρ νόμου δικαιουσύνη) and 3:21 (εἰ γὰρ ἐδόθη νόμος ὁ δυνάμενος ζῳοποιῆσαι).
29. Garlington (“Role Reversal,” 106) affirms that such “time-honored terminology…, is wide of the mark”; in fact, it appears so only to someone who has already been persuaded by E. P Sanders’s analysis of Paul and Judaism. Normand Bonneau (“The Logic of Paul’s Argument on the Curse of the Law in Galatians 3:10–14, ” NovT 39 [1997]: 73) explicitly argues that the use of the word rely “slants the text in the direction of works-righteousness, which has nothing to do either with Judaism in Paul’s day or with Paul’s judaizing opponents, as Sanders has decisively demonstrated.” Incidentally, it should be appreciated that the translation “those who rely on works of the law” is not all that far from, for example, Bachmann’s view that the phrase refers to “diejenigen, die sich von solchen Regelungen her definieren” (“4QMMT und Galaterbrief,” 111).
30. Cf. M. Silva, “The Place of Historical Reconstruction in New Testament Criticism,” in Hermeneutics, Authority, and Canon (ed. D. A. Carson and.]. D. Woodbridge; 1986; repr., Grand Rapids: Baker, 1995), 117–21.
31. I am therefore not very impressed by the argument that the traditional interpretation of Gal 3:10 cannot be right because it has to supply a premise; so Daniel P. Fuller, Gospel and Law: Contrast or Continuum? The Hermeneutics of Dispensationalism and Covenant Theology (Grand Rapids: Eerdmans, 1980), 90–91. For the perception that an implied premise (about perfect obedience being impossible) would be “unintuitive,” see Michael Cranford, “The Possibility of Perfect Obedience: Paul and an Implied Premise in Galatians 3:10 and 5:3, ” NovT 36 (1994): 258. This is a curious comment; contrast the fact that the earliest attested interpretations we have do assume such an implied premise (see the quotations at the end of the present essay).
32. Cf. M. Silva, “The New Testament Use of the Old Testament: Text Form and Authority,” in Scripture and Truth (ed. D. A. Carson and J. D. Woodbridge; 1983; repr., Grand Rapids: Baker, 1992), 159–61. Young (“Who’s Cursed,” 86–87) gives an ingenious twist to the traditional interpretation by suggesting that Paul is referring only to a potential consequence: those who are of the works of the law are cursed if they abandon the laws of the covenant. In Young’s view (for which the evidence seems weak), it is the Judaizers who accuse Paul and his Gentile converts of being under a curse, and the apostle is defending himself from that charge.
33. It is also quite possible that the same idea is implied in 5:3, though some dispute the point. Incidentally, it is obvious that Paul, if pressed, would have had to admit that his Jewish opponents obeyed many specific commands, but that is a different issue from the question whether they could be appropriately characterized as people who obey the law.
34. Cf. M. Silva, Philippians (BECNT; Grand Rapids: Baker, 1992), 169.
35. On Phil 3:6, which is so frequently misused, see Silva, Philippians, 174–76.
36. It is an interesting question what it was about the Judaizers’ behavior that Paul considered grievous disobedience of the law. On the basis of Gal 5:14–26, Gaffing, ton says, “Proof positive to Paul that the opponents are apostate is their lack of love.” In the conclusion to his article, however, he points to a more fundamental issue by arguing that 3:10 involves an irony. “That is to sag in their very keeping of the law, the opponents have not kept it, because they have not ‘upheld’ [קוּם] it in its eschatological design, that is, to point Israel to Jesus of Nazareth as the one who has done away with the barriers of separation between nations … In a word, the opponents are apostates in a newly defined eschatological sense” (Garlington, “Role Reversal,” 109, 120). There is some merit in this approach, which could be supported by the way Paul describes his Jewish opponents in Phil 3:2–3.
37. Although scholars are not always explicit about it, this view is held widely (see most recently Martyn, Galatians, 308 n. 76). One difficulty with this position is Paul’s use of ἡμᾶς in v. 13 (although the shift to the first person is a problem for almost any interpretation). If the ones under the curse are primarily Paul’s opponents, how can the apostle say that Christ has redeemed “us” from the curse? Here too, I think, we need to avoid false dichotomies. Although the Judaizers are presented as the prime expression of people who fall under the curse, we need not infer that Paul could not see Deut 27:26 as having a broader application. After all, the apostle goes so far as to include the Gentiles at the end of the paragraph (v. 14), even though in the strict sense Gentiles could hardly be conceived of as being under the curse of the law (note especially also the shift in 4:4–7 between the first and second person). In support of this view, see among others Bernard M. Levinson, Deuteronomy and the Hermeneutics of Legal Innovation (New York and Oxford: Oxford University Press, 1997), 139–40.
38. Cf. Silva, Explorations, 177–78.
39. Garlington (“Role Reversal,” 101), though coming at this passage from a different perspective, reaches a similar position. “To say that the law is ‘not of faith’ is to affirm that the law and faith belong to distinctly different historical realms: the former does not occupy…, the same turf in the salvation-historical continuum as the latter.”
40. Cf. K. Stendahl, “The Apostle Paul and the Introspective Conscience of the West,” HTR 56 (1963): 199-215.
41. Justin Martyr, Dialogue 95.1 (ANF 1.247; I am indebted to Frank Thielman for bringing this passage to my attention). Cf. also Chrysostom, Commentary on Galatians, on 3:12: “You see how he proves that they are under the curse who cleave to the Law, because it is impossible to fulfill it” (NPNF 13.66; PG 61.652, lines 26–28: Εἶδες πῶς ἔδειξεν, ὅτι οἱ τῷ νόμῳ προσέχοντες, διὰ τὸ ἀδύνατον εἶναι πληρωθῆναι, ὑπὸ κατάραν ἐγέοντο).
42. St. Thomas Aquinas, Commentary on Saint Paul’s Epistle to the Galatians (trans. E R. Larcher; Aquinas Scripture Series 1; Albany, N.Y.: Magi, 1966), 79. After the present article was completed, I received in electronic form a very helpful doctoral dissertation by my former student Andrew H. Carver, “Means or Meaning: The Logic of Paul’s Rhetoric in Galatians 3:10–14” (University of Durham, 2000). I regret not having been able to interact with his extensive discussion of this passage.