EXCURSUS: PAUL, "WORKS OF THE LAW," AND FIRST-CENTURY JUDAISM

From "The Epistle to the Romans," NICNT

By Douglas J. Moo

(Grand Rapids:  Eerdmans, 1996), 211-217.

Two significant issues are raised by Paul's assertion in 3:20 that "no one is justified by works of the law."

The first is the question of the logic of this denial.  Why can't "works of the law" justify?  We might take our starting point in the juxtaposition of 2:13 -- "doers of the law will be justified" -- and 3:20 -- "no one will be justified by works of the law."  There appear to be only five ways of explaining this contrast.  First, one could argue that 2:13 states only a hypothetical promise:  only for the sake of argument does Paul suggest such a possibility.  As I have noted in the exegesis of 2:13, this is unlikely.  Second, one could find a salvation-historical distinction between the two:  in the "old age," doing the law could justify, but in the "new age" it no longer can.  Not only does this make Paul's argument in 2:13 irrelevant to his present readers (who are in the "new age"), but it contradicts the express denial of Paul that the law could justify (see Gal. 3:21).  Third, one could argue that "justify" in 2:13 refers to the judgment, or to a "second justification,"

while "justify" in 3:20 relates to the initial entrance into salvation. [68]  But, while there is some precedent in Judaism and in James for the use of "justify" to depict the judgment, evidence from Paul's usage for this meaning is lacking.  Fourth, one could define "doing the law" and "works of the law" in different ways, so that the former means "fulfilling the law in Christ," or "obeying God as a response to grace" while the latter connotes simple doing of the law or doing it in a legalistic spirit.  We have also found reason for rejecting this interpretation:  lexical and contextual evidence does not justify any sort of distinction.

We are left, then, with the supposition that one must insert a step in the argument between the two statements, to the effect that "no one can do the law."  Not only does this make the best sense of both statements in their contexts, but it is, in effect, the assertion that Paul inserts between the two verses:  "all are under the power of sin. . . there is no one righteous, no, not one" (3:9-10). We should add that a view like Dunn's, according to which "works of the law" are:  defined as Jewish identity markers and do not justify because the covenant that they represent cannot justify simply moves the question back one step further:  Why doesn't the covenant justify?  Paul has shown why in 2:1-3:19:  the law cannot be "done" to the extent necessary to secure justification through that "law" covenant.  It is far more likely, then, that this is the ultimate logic undergirding his denial that works of the law can justify here -- a logic rooted not (or at least not only) in salvation history, or in concern about social barriers, but, more deeply, in the human condition itself.

The second issue, related to the first, has to do with the viewpoint that Paul is opposing with his statement.  Traditionally, it has been understood as a denial that a person can "earn" salvation by doing anything:  no "works," however "good" -- even those done in obedience to God's holy law -- can bring a person into relationship with God.  It has, furthermore, usually been assumed that this thesis was directed against Jews in Paul's day who believed that, indeed, they could get into relationship with God by obedience to the law.  Many modem interpreters (some of whom label this traditional view the "Lutheran orthodox" interpretation) question this explanation of the situation.  I think, however, that, properly nuanced, the traditional view remains the best explanation of the Pauline polemic.  The attentive reader will recognize in the commentary on texts such as 2:11 and 13 and 3:20 our general endorsement of this view.  But here I would like to explore perhaps the most attractive counterposition and explain why I do not find it convincing.

We must begin with the work of E. P. Sanders. His monograph, Paul and Palestinian Judaism (1977), is the stimulus for most of the recent reconstructions of Paul's interaction with Judaism on the law.  To understand his position, then, we must understand his basic thesis about Palestinian Judaism.

Sanders argues that Palestinian Judaism in Paul's day was characterized by a "pattern of religion" that he labels "covenantal nomism."  In contrast to the "traditional" picture of Judaism as a religion that required works as a means of entry into salvation, this pattern of covenantal nomism features obedience to the law as the means to maintain the Jews' status in the covenant, a status that was freely granted to Jews through their election.  Thus, obedience to the law was not the means of "getting in" but of "staying in."  While this portrait is drawn on the basis of the surviving literature of early Palestinian Judaism, Sanders insists that, for lack of conflicting evidence, this must have been the common first-century Jewish belief.

If Sanders's view of first-century Judaism is adopted as accurate -- and it has met with wide acceptance -- then it is necessary to ask:  "Why has Paul so often and so emphatically denied that 'works of the law' can justify"?  If no one in first-century Judaism really believed that a person could be justified by doing the law, then why deny it?  To this question Sanders in his original monograph gave no clear answer beyond asserting that Paul ruled the law out of court on the basis of his "exclusivist soteriology":  because Christ was the only means of salvation, the law could not be.  But others quickly rushed in to fill the void. Two basic approaches can be distinguished.

First are those who think that both Sanders's view of Judaism and the traditional understanding of Paul -- that he implied that Jews sought justification through the law -- are correct and that the consequence is that Paul either misunderstood or deliberately misrepresented Judaism for polemical purposes.  Jews did not really believe that "a person could be justified by works of the law"; but Paul implies that they did as a result of an overreaction against his previous religion and with the purpose of distancing Christianity from Judaism. [69]  We think so extreme a conclusion is unwarranted.  Not only does it presume to judge Paul on the basis of questionable claims about what Jews believed in Paul's day (see below), but it also makes the unlikely assumption that Paul would misrepresent a position that some of the people he was trying to convince would know very well.

The second general attempt to explain Paul against the background of Sanders's interpretation of Judaism is found among scholars who think that Paul can be "reinterpreted" in such a way that his teaching meshes with Jewish "covenantal nomism."  Several specific proposals that take this general tack have emerged. One, associated especially with J. Gager and L. Gaston, argues that virtually all Paul's negative statements about the law are directed against those who wanted to impose the law on Gentiles. [70]  But the best-supported and most reasonable proposal is the one that J. D. G. Dunn has made prominent and which we touch on in the commentary on 3:20:  that "works of the law" connotes Jewish obedience to the law in its function as establishing a boundary around the Jewish people, marking them off from the Gentile peoples around them.  Jews did not see these works as bringing them into relationship with God but as maintaining their "national righteousness," their peculiar status as God's covenant people.  Throughout Rom. 2-3, then, Paul argues against this presumption of "favored status," showing that Jews cannot rely on their supposed special relationship with God for salvation.  "No one is justified by works of the law" means, then, that no Jew can expect to be justified through those works "by which a member of the covenant people identified himself as a Jew and maintained his status within the covenant." [71]  What the Jews, need, then, is a more inclusive understanding of salvation history and of the law, by which Gentiles can be allowed to take their rightful place alongside Jews within God's covenant community. [72] 

Despite its popularity, Dunn's proposal suffers from some serious drawbacks.  First, while Rom. 2 certainly has a great deal to say about Jewish, complacency because of a misunderstanding of the covenant, it must be questioned whether we can confine Paul's polemic to this one issue.  Dunn consistently downplays the role that transgression of the law -- not just adherence to certain ethnic identity markers -- plays in Paul's argument.  Again and again, Paul insists in 2:1-29 that it is not dependence on the law or circumcision as such that renders the Jews liable to judgment, but their disobedience of the law.  Transgressions of the law are the reason why the Jews cannot presume on the covenant for salvation. [73]  And these transgressions are said to involve the "same things" that Gentiles do (2:2-3) -- clearly making it a matter not of "inner" Jewish issues but of sin against God generally.  It is this larger and more basic problem of transgression of the law that informs Paul's conclusion to this section in 3:20: "no human being will be justified by works of the law."  The "works" mentioned here must, as Dunn says, be the "works" Paul has spoken of in chap. 2. [74]  But it is not circumcision -- let alone other "identity markers" that are not even mentioned in Rom. 1-3 -- that the Jew "does" in Rom. 2; it is, generally, what is demanded by the law, the "precepts" (v. 26; cf. vv. 22-23, 25, 27). [75]  Therefore, 3:20 must deny not the adequacy of Jewish identity to justify, but the adequacy of Jewish works to justify.  Belonging to the Jewish people does not justify because no Jew does the law sufficiently to give to that identity salvific power.  It is this root anthropological issue -- human inability -- that informs 3:20, and justifies its application to the circumstance of any person. [76] 

This, however, brings us back to the question:  Against whom is the polemic in 3:20 directed?  Two things can be said here.

First, as Sanders argues, Paul's denial that works of the law can justify may signify simply an attack on the covenant as understood by Jews.  As Sanders points out, Jews regarded the intention to obey the commandments as sufficient to maintain one's covenant status. [77]  Paul, however, insists that only what is actually done counts.  This argument is an outright attack on the "covenantal nomism" that Sanders has sketched. [78]  The denial of special status to the Jews is an implicit rejection of the election that was the foundation for "covenantal nomism," [79] and coheres closely with the polemic of John the Baptist (cf. Matt. 3:7-10) and of Jesus. [80]  How does this critical attitude toward the covenant cohere with the OT itself?  As we have seen, Paul is close to the prophets in his criticism of those who rely on the covenant as an automatic protection from judgment.  And, while he is more explicit, Paul's polemic against the Mosaic covenant is in keeping with the pessimistic attitude expressed toward that covenant in Deuteronomy (cf. chap. 32) and many of the prophets (e.g., Jer. 31:31-34).  Paul stresses that the Abrahamic promise, to which one must respond with faith, and which is now fulfilled in Christ, is the true locus of salvation (Rom. 4).  He therefore does not deny the promise of salvation given in the Scriptures to the Jews, but attaches it not to the Mosaic covenant, as did Judaism, but to the Abrahamic. [81]  One could, at the least, make a very good case for finding Paul's interpretation of the OT to be more accurate than that of "covenantal nomism."  "Works of the law" -- those things done by Jews in obedience to the law by which they sought to maintain their covenant status -- cannot justify because the covenant within which they performed those works was inadequate to bring justification.  Jewish works, then, are no different from Gentile works, once the larger framework of the covenant -- as usually understood in first-century Judaism -- is eliminated.

But a second possibility should also be considered -- that Palestinian Judaism was more "legalistic" than Sanders allows, and that Paul is also responding to Jews who did, in some sense, think to be justified by doing the law.  Even in Sanders's proposal, works play such a prominent role that it is fair to speak of a "synergism" of faith and works that elevates works to a crucial salvific role.  For, while works, according to Sanders, are not the means to "getting in," they are essential to "staying in."  When, then, we consider the matter from the perspective of the final judgment -- which we must in Jewish theology -- it is clear that "works," even in Sanders's view, playa necessary and instrumental role in "salvation." [82]  But this is what Paul denies, by equating "initial" justification with the final verdict of salvation and by stressing faith alone as the necessary corollary to the grace of God.  In effect, then, while not denying the role of faith, Jews were insisting on works as a means of justification.  But this is just what Paul denies in 3:20, and why he distinguishes in principle between faith and works (3:27-28; 4:1-5).

Moreover, there is some question as to whether Sanders's reconstruction of first-century Judaism is to be accepted in toto. [83]  While agreeing with Sanders that many interpreters have ignored the importance of election and covenant in Palestinian Judaism, and have been guilty, thereby, of caricaturing Jewish theology, I think that there is reason to conclude that Judaism was more "legalistic" than Sanders thinks.  In passage after passage in his scrutiny of the Jewish literature, he dismisses a "legalistic" interpretation by arguing that the covenantal framework must be read into the text or that the passage is homiletical rather than theological in intent.  But was the covenant as pervasive as Sanders thinks?  Might not lack of reference in many Jewish works imply that it had been lost sight of in a more general reliance on Jewish identity? [84]  And does not theology come into expression in homiletics?  Indeed, is it not in more practically oriented texts that we discover what people really believed?  Sanders may be guilty of underplaying a drift toward a more legalistic posture in first-century Judaism. [85]  We must also reckon with the possibility that many "lay" Jews were more legalistic than the surviving literary remains of Judaism would suggest.  Certainly the undeniable importance of the law in Judaism would naturally open the way to viewing doing the law in itself as salvific.  The gap between the average believer's theological views and the informed views of religious leaders is often a wide one.  If Christianity has been far from immune from legalism, is it likely to think that Judaism, at any stage of its development, was? [86]  Finally, and, I think, most important, I am convinced that the teaching of Paul -- and of Jesus and Matthew and Luke and Mark and Peter -- cannot satisfactorily be explained without the assumption that some Jews, at least, had drifted from a biblical conception of the primacy and sufficiency of God's grace into a belief that accorded their own works done in obedience to the law as basic to their justification/salvation.

We conclude, then, that Paul criticizes Jews for thinking that the Mosaic covenant is adequate without that perfection in "works" without which any system of law must fail to bring one into relationship with God.  The Jews become, as it were, representative of human beings generally.  If the Jews, with the best law that one could have, could not find salvation through it, then any system of works is revealed as unable to conquer the power of sin.  The "bottom line" in Paul's argument, then, is his conviction that sin creates for every person a situation of utterly helpless bondage.  "Works of the law" are inadequate not because they are "works of the law" but, ultimately, because they are "works."  This clearly removes the matter from the purely salvation historical realm to the broader realm of anthropology.  No person can gain a standing with God through works because no one is able to perform works to the degree needed to secure such a standing.  This human inability to meet the demands of God is what lies at the heart of Rom. 3.  On this point, at least, the Reformers understood Paul correctly. [87]

ENDNOTES

68. C. H. Cosgrove ("Justification in Paul: A Linguistic and Theological Reflection," JBL 106 [1987], 653-78) argues that the ek following in 3:20 denotes the instrument through which justification takes place, while statements such as 2:10 and 2:13 indicate that works are "the evidential basis" for justification. But the phrase "evidential basis" conceals a crucial ambiguity -- basis for justification or evidence of justification? -- and it is questionable whether dikaiow ek can be defined as narrowly as this.

69. Several scholars find some degree of misrepresentation in Paul, but the clearest and strongest presentation is that of Raisanen.

70. See esp. J. G. Gager, The Origins of Anti-Semitism: Attitudes toward Judaism in Pagan and Christian Antiquity (New York: Oxford University Press, 1983); L. Gaston, Paul and the Torah (Vancouver: University of British Columbia, 1987), esp. pp. 15-34.

71. Romans, 1.158.

72. The first scholar, apparently, to advance this general idea was N. T. Wright (cf. "The Paul of History and the Apostle of Faith," TynBul 29 [1978], 61-88). Sanders himself, in his second monograph on the general topic, Paul, the Law and the Jewish. People (1983), suggests that Paul's denial that "works of the law" could justify was directed against the "covenantal nomism" of Palestinian Judaism: "The argument is that one need not be Jewish to be 'righteous' and is thus against the standard Jewish view that accepting and living by the law is a sign and condition of favored status" (p. 46).

73. See, e.g., Schreiner, " 'Works of the Law,' " pp. 226-28.

74. See his Romans, 1.158.

75. Significantly, B. W. Longenecker, who embraces Dunn's viewpoint, admits that in Rom. 2 Paul sets up a "straw man," arguing against a view that did not in fact exist (Eschatology and the Covenant: A Comparison of4 Ezra and Romans 1-11  [JSNTSup 57; Sheffield: JSOT, 1991], p. 279).

76. M. Silva ("The Law and Christianity: Dunn's New Synthesis," WTJ 53 [1991], 350-51) criticizes Dunn for an "all-or-nothing" approach that sets up an unjustifiable mutual exclusion between individual and corporate concerns in Paul.

77. Sanders, Paul and Palestinian Judaism, pp. 157-82.

78. As both Dunn and Wright ("Paul of History") emphasize.

79. Sanders, Paul, the Law and the Jewish People, p. 47.

80. Cf. D. C. Allison, Jr., "Jesus and the Covenant: A Response to E. P. Sanders," JSNT 29 (1987), 61-63.

81. Cf. M. D. Hooker, "Paul and Covenantal Nomism," in Paul and Paulinism, p. 51.

82. See esp. Seifrid, Justification, pp. 56-57, 71-81 (he finds a clear emphasis on the importance of works for eventual salvation in Psalms of Solomon and IQS); Laato, Paulus und das Judentum, pp. 73-75, 195-211; Gundry, "Grace, Works, and Staying Saved," pp. 19-20, 35-36; Westerholm, 143-50; Byrne, 230; K. T. Cooper, "Paul and Rabbinic Soteriology," WTJ 44 (1982), 137-38.

83. Note in this regard the warning of J. C. Beker: "We might wonder whether the work of Krister Stendahl and E. P. Sanders influences our treatment of Judaism so heavily these days that their important contributions are unduly exaggerated and -- as it were -- considered to be dogmatic, unassailable truth" ("Echoes and Intertextuality: On the Role of Scripture in Paul's Theology," in Paul and the Scriptures of Israel [ed. C. A. Evans and J. A. Sanders; JSNTSup 83; Sheffield: Sheffield Academic Press, 1993], p. 68).

84. J. Neusner raises a similar question about Sanders's treatment (cf. "The Use of the Later Rabbinic Evidence for the Study of Paul," in Approaches to Ancient Judaism, vol. 2 [ed. W. S. Green; Brown Judaic Studies 9; Chico, CA: Scholars Press, 1980], pp. 47-52) -- and cf. Sanders's response in the same volume- ("Puzzling Out Rabbinic Judaism," pp. 69-75).

85. G. B. Caird, Review of Paul and Palestinian Judaism, by E. P. Sanders, JTS 29 (1978), 539-40; D. A. Carson, Divine Sovereignty and Human Responsibility (Atlanta: Knox, 1981), pp. 86-95; Snodgrass, "Justification by Grace," p. 77.

86. See, e.g., Seifrid, Justification. pp. 56-57; D. A. Hagner, "Paul's Quarrel with Judaism," in Anti-Semitism and Early Christianity: Issues of Polemic and Faith (ed. C. A. Evans and D. A. Hagner; Minneapolis: Fortress, 1993), pp. 138-39; idem, "Paul and Judaism. The Jewish Matrix of Early Christianity: Issues in the Current Debate," Bulletin for Biblical Research 3 (1993), 118-19; T. F. Best, "The Apostle Paul and E. P. Sanders: The Significance of Paul and Palestinian Judaism," RestQ 25 (1982), 72-73; Silva, "Dunn's New Synthesis," pp. 349-50. Sanders goes so far as to question whether "the notion of 'legalism' -- an attitudinal sin which consists in self-assertion -- goes back to the first century" ("Paul on the Law, his Opponents, and the Jewish People in Philippians 3 and 2 Corinthians II," in Paul and the Gospels, vol. 1 of Anti-Judaism in Early Christianity [Studies in Christianity and Judaism I; ed. P. Richardson; Waterloo, Ont.: Wilfrid Laurier University (Corporation for Studies in Religion}, 1986], pp. 78-79). Longenecker's suggestion (pp. 68-85) that Judaism probably featured at least two kinds of approaches - an "acting legalism" and a "reacting nomism" -- has (despite Sanders's criticisms) much to be said for it. Scholars are giving more and more recognition to the diversity of first-century Judaism, to the point of speaking of "Judaisms" (e.g., J. Neusner, Judaic Law from Jesus to the Mishnah [South Florida Studies in the History of Judaism 84; Atlanta: Scholars Press, 1993}, pp. 49-53; B. Longenecker, Eschatology and the Covenant, pp. 32-33). Note also the careful criticisms of Sanders in Laato, Paulus und das Judentum, pp. 38-82; cf. also Schreiner, The Law and Its Fulfillment. pp. 92-121.

87. Westerholm (pp. 221-22) comes to a similar conclusion.