¹ Ecclesia semper reformanda. E. Wolf zum 50. Geburtstag (München, 1952), pp. 41–5Google Scholar; also in Bultmann, , Exegetica. Aufsätze zur Erforschung des Neuen Testaments (Tübingen, 1967), pp. 394–9.Google Scholar

² See especially Klein, G., ‘Individualgeschichte und Weltgeschichte bei Paulus. Eine Interpretation ihres Verhältnisses im Galaterbrief’, Ev. Theol. 24 (1964), 126–65, pp. 132–3Google Scholar; also in Klein, , Rekonstruktion und Interpretation. Gesammelte Aufsätze zum Neuen Testament, Beitr. Ev. Theol. 50 (München, 1969), pp. 180–224, pp. 186–7.Google Scholar Cf. also Feld, H., ‘“Christus Diener der Sünde”. Zum Ausgang des Streites zwischen Petrus und Paulus’, Theol. Quartalschr. 153 (1973), 119–31Google Scholar, p. 125, n. 21 and p. 127; Schlier, H., Der Brief an die Galater, Krit.-exeg. Komm. N.T. (Göttingen, ¹² 1962), p. 96 n. 2Google Scholar; Mussner, F., Der Galaterbrief, Herders Theol. Komm. N.T. (Freiburg-Basel-Wien, 1974), p. 176 n. 45.Google Scholar

³ Cf. Lagrange, M.-J., Épître aux Galates, Ét. Bibl. (Paris, ² 1925), p. 48Google Scholar: ‘Ce verset est extrêmement dificile…le v. 17 qu'il est très difficile de faire rentrer dans l'argumentation’; Hahn, F., ‘Das Gesetzesverständnis im Römer- und Galaterbrief’, Z.N.W. 67 (1976), 29–63, p. 53 n. 76Google Scholar: ‘Die eigentliche Schwierigkeit des Abschnitts liegt in 2, 17 f.’ For an example of a complicated reconstruction of the line of thought in vv. 17–20, see e.g. Schlier, , op. cit. p. 98.Google Scholar

⁴ Paul presents Gal. 2. 14b–21 as a short discourse addressed to Peter at Antioch in the presence of other Christians. Cf. Mussner, , op. cit. p. 178Google Scholar: ‘…der ganze Abschnitt 2, 14b–21 will eine “Rede” darstellen, deren Adressat formal Petrus ist (vgl. V 14a)’. In the context of this study there is no need to deal with the question of the relation between the two levels, that of the past historical event at Antioch and that of the actual crisis in the Galatian churches for whose benefit Paul recalls this event.

⁵ Cf. e.g. De Witt Burton, E., The Epistle to the Galatians, Int. Crit. Comm. (Edinburgh, 1921), p. 125Google Scholar: ‘Sinners’ is to be taken here in the sense suggested by v. 15, ‘men outside of the law, violators of the law’, ‘having reference to the disregard of the statutes of the law, especially those concerning clean and unclean meats, which statutes Paul, and for a time Peter also, had violated…’.

⁶ Cf. Vielhauer, P., Oikodome. Das Bild vom Bau in der christlichen Literatur vom Neuen Testament bis Clemens Alexandrinus (Heidelberg, 1939), p. 89Google Scholar, who refers to the opinion of M. Dibelius: ‘Die Parenthesis V. 18 bringt also sozusagen einen Midrasch zu άμαρτωλοι V. 17, die wohl dasselbe bedeuten wie παραßάτης V. 18.’ Likewise Lietzmann, H., An die Galater, Handb. N.T. (Tübingen, ³ 1932), pp. 16–17Google Scholar, considers v. 18 as a parenthesis: ‘eine Bejahung von’ v. 17a. Klein, Rekonstruktion, pp. 196–8Google Scholar, rejects the parallelization of vv. 17 and 18.

⁷ Cf. I Cor. 6. 16: this verse, introduced by ή οὐκ οιδατε, explains, not the immediately preceding μή γένοıτο (v. 15c), but the interrogative clause of v. 15b. Gal. 1. 18 likewise refers back to v. 17a.

⁸ Cf. Messner, , op. cit. p. 179Google Scholar: ‘Begründet wird also mit V 19 die Unsinnigkeit eines πάλıνοικοδομείν, konkret: einer Wiederaufrichtung des Gesetzes.’

⁹ For the view that v. 19 explains v. 17, see e.g. Klein, , Rekonstruktion, p. 199Google Scholar; Lietzmann, , op. cit. p. 16Google Scholar; Schlier, , op. cit. p. 96.Google Scholar

¹⁰ Bultmann, , Exegetica, p. 399Google Scholar, is also of this opinion.

¹¹ For a good discussion of this topic cf. Burton, , op. cit. pp. 128–9.Google Scholar

¹² Cf. Oepke, A., Der Brief des Paulus an die Galater, Theol. Handkomet. N.T. (Berlin, ² 1957), p. 60.Google Scholar

¹³ Although Klein, , Rekonstruktion, pp. 191–2Google Scholar, rejects the alternative – the expression εὑρέθημενάμαρτωλοι means that through our believing in Christ the sinful condition is either recognized as existing or created – he nevertheless does use phrases which point in the direction of the second of the two alternatives: ‘Im Glauben an Christus werden auch Juden…άμαρτωλοι, so daß Paulus mit der Möglichkeit rechnet, daß man den Ursprung dieses Geschehens als den Ursprung der Sünder selbst, Christus als Diener der Sünder, denunzieren könnte’. Berger, K., ‘Abraham in den paulinischen Hauptbriefen’, Münch. Theol. Zeitschr. 17 (1966), 47–89, p. 48Google Scholar n. 3, criticizes Klein on this point.

¹⁴ Exegetica, pp. 395–6. Sieffert, F., Der Brief an die Galakr, Krit. exeg. Komm. N.T. (Göttingen, ⁹ 1899), pp. 150–1Google Scholar, affirms that Christ's justification appears ‘als eine bloße Beschönigung und Bestärkung der nach wie vor bestehenden Sünde’. Cf. also Baljon, J. M. S., Exegetisch-kritische verhandeling over den brief van Paulus aan de Galatiërs (Leiden, 1889), p. 164Google Scholar; Loisy, A., L'Épître aux Galaks (Paris, 1816), p. 139.Google Scholar For a good discussion of Loisy' view, cf. Lagrange, , op, cit. p. 48.Google Scholar

¹⁵ Cf. Klein, , Rekonstruktion, p. 199Google Scholar: ‘eine Begründung a contrario’. Schmithals, W., Paulus and Jakobus, FRLANT 85 (Göttingen, 1963), p. 63Google Scholar, sees the opposition in a different way: Not I, Paul, make Christ an agent of sin. ‘Aber tatsächlich machst Du, Petrus, ihn dazu, wenn Du die erlangte Gesetzesfreiheit wieder preisgibst and zu den Werken des Gesetzes zurückkehrst; denn damit erklärst Du, bei der Bejahung der Gesetzesfreiheit als Sünder gehandelt zu haben.’

¹⁶ Klein, , Rekonstruktion, p. 199Google Scholar, adds: ‘übertretend oder gehorchend’.

¹⁷ Cf. Schneider, G., Der Brief an die Galater, Geistl. Schriftlesung (Düsseldorf, 1964), pp. 61–2.Google Scholar

¹⁸ Also in Rom. 6. I and 15, which passages are called in question by Feld, art. cit. p. 130 n. 39. In addition to Bultmann and Feld, Ibid. pp. 125–8, cf. also Moule, C. F. D., An Idiom-Book of New Testament Greek (Cambridge, ³ 1963), p. 196Google Scholar: ‘…in Gal. ii. 17, 18 I should propose to read άρα (not άρα) and a full stop (not a question mark), and to assign μή γένοıτο to the (imagined) St Peter; St Paul will then continue to the effect: “You may repudiate the position with a μή γένοıτο, but that is the position you logically place yourself in by your action; for, etc.”’

¹⁹ Feld, art. cit. pp. 127–8, maintains that an objection-question cannot be introduced here by άρα since the opponent expects a positive answer; otherwise, there should be an οὐ after άρα. But see Humbert, J., Syntaxe grecque (Paris, ³ 1972)Google Scholar, no. 684, who states that, although an οὐ is often present in cases where a positive answer is expected, the άρα of itself does not determine what kind of answer will follow.

²⁰ Cf. e.g. Hahn, , art. cit. p. 54 n. 76Google Scholar; Sieffert, , op. cit. pp. 147–8.Google Scholar Sieffert refers to Paul's use of άρα in an interrogative sentence in II Cor. 1. 17 (but the question is here introduced by μήτı). Burton, , op. cit. p. 126Google Scholar, adds in a note: The sentence is certainly a question, ‘but άρα being frequent in Paul, whereas there is no clear instance of άρα in his writings, the presumption is perhaps slightly in favour of the former’.

²¹ Cf. Feld, art. cit. p. 130 n. 38.

²² As such the realis only points to a simple supposition. It is possible that the supposition is contrary to fact (cf. e.g. I Cor. 15. 13), but the one employing a realis is not concerned with the possible ‘unreal’ aspect of the supposition. Cf. Zerwick, M., Biblical Greek (Rome, 1963), no. 306.Google Scholar

²³ This is explained in a convincing way inter alios by Sieffert, , op. cit. pp. 148–51Google Scholar; Lagrange, , op. cit. pp. 48–9Google Scholar; Lightfoot, J. B., Saint Paul's Epistle to the Galatians (London, ⁷ 1881), pp. 116–17.Google Scholar An irrealis does not supply a good sense, for in this case Paul would himself either consciously be formulating an absurdity (cf. the position of Bultmann who, however, wrongly understands v. 17a as ‘still sinners after justification’) or he would in v. 17a be citing the opinion of his opponents: the Judaizers hold that table fellowship with the Gentile Christians is sinful (for a recent view along this line, cf. Mussner, , op. cit. pp. 176–7)Google Scholar. Besides the poor nexus of v. 17 with v. 16 (and v. 15) it is to be doubted whether the άμαρτια in v. 17b can have the same weak sense as άμαρτωλς in v. 17a, which sense the latter term would have in the second alternative.

²⁴ Cf. Lagrange, , op. cit. p. 50Google Scholar: ‘Il est dans la nature des choses que μή γένοıτο, protestant contre l'interrogation captieuse qui précéde, proteste contre la fausse conséquence, non contre une sup-position fausse d'où l'on tirerait une conclusion juste.’

²⁵ Klein, , Rekonstruktion, p. 187Google Scholar, rightly points out: ‘die Passivformen von εὑρισκεıν meinen bei Paulus sonst durchweg die Entbergung bzw. Sicherstellung eines bis dahin verborgenen bzw. ungewissen Sachverhalts’. With Burton, , op. cit. p. 125Google Scholar, the question can even be raised whether the passive form is really perceived as such so that the reader would be led to ask ‘by whom’ we were found (God, our conscience, cf. Messner, , op. cit. p. 177Google Scholar; ‘les partisans de Jacques’, cf. Wagner, G., ‘Le repas du Seigneur et la justification par la foi. Exégèse de Galates 2, 17’, Ét. Théol. Rel. 36 (1961), 245–54, pp. 247–8)Google Scholar. Burton, however, reckons with the possibility that Paul was brought to the use of this verb by his opponents’ way of speaking.

²⁶ Cf. Schlier, , op. cit. p. 95.Google Scholar

²⁷ Ibid. pp. 95–6.

²⁸ Cf. n.23.

²⁹ The arguments, both formal and those bearing on the content, with which Hasler, V., ‘Glaube and Existenz. Hermeneutische Erwägungen zu Gal 2, 15–21’, Theol. Zeitschr. 25 (1969), 241–51Google Scholar, p. 246 n. 8, tries to prove v. 18 to be secondary, are not convincing.

³⁰ Wörterbuch N.T. (Berlin-New York, 1971 = ⁵1963)Google Scholar, sub voce, no. 4. He notes that certain modern readers would like to replace this type of γάρ by a δέ: ‘überflüssigerweise, da der Gramm. Trypho Alexander… feststellt, daß γάρ unter Umständen εις οὐν ήστıν άντι το δέ = ein and dasselbe wie δέ ist’. Bauer refers to Rom. 1. 18; 2. 25; 4. 3, 9; 5. 7; 12. 3; 14. 5; I Cor. 10. 1; II Cor. 1. 12; 10. 12; 11. 5; Gal. 1. i I (v.1. δέ); 5.13 (we italicize the passages which appear most evident to us). Cf. also Zerwick, , op. cit. no. 473Google Scholar; Revised Standard Version: ‘but’; Grossouw, W., De brief van Paulus aan de Galaten (Bussum, 1974), pp. 100Google Scholar, 107 and n. 70 (with reference to Bauer); Bonnard, P., L'Épître de saint Paul aux Galates, Comm. N.T. (Neuchâtel, ² 1972), p. 55Google Scholar: ‘Le γάρ = en effet renvoie aux mots “ministre du péché” (v. 17): ce n'est pas Jésus-Christ qui “sert” le péché en effet, le transgresseur, c'est le légaliste’. Like Bonnard, Grossouw less correctly accepts an opposition between Christ and ‘I’ here. On the asseverative force of γάρ, see Thrall, M. E., Greek Particles in the New Testament, N.T. Tools and Studies 3 (Leiden, 1962), pp. 43–5Google Scholar (three cases, but our text is not mentioned).

Two remarks are in place here. (a) Robertson, A. T., A Grammar of the Greek New Testament in the Light of Historical Research (London, 1914), p. 1190Google Scholar, notes that γάρ ‘does not always give a reason. It may be explanatory…It is a mistake…to approach the study of γάρ with the theory that it is always or properly an illative, not to say causal, particle. It is best, in fact, to note the explanatory use first.’ (b) That γάρ appears to possess a meaning which is not explanatory nor motivating may partly be due to the elliptical way of speaking (brachylogy): in an ellipsis the γάρ, in fact, explains or motivates a thought which is not expressed or omitted (cf. e.g. Moule's opinion on Gal.) 2. 17–18 in n. 18. On ellipsis see Denniston, J. D., The Greek Particles (Oxford, 1970 = ²1954), p. 56Google Scholar and p. 61: ‘Compression of thought is often the source of difficulty, and formal exactitude can then be achieved by supposing an ellipse…But this, though a convenient method of exposition, is psychologically somewhat misleading.’ Liddell, H. G.Scott, -R., Lexicon, sub voce, 1, 3Google Scholar: ‘in elliptical phrases, where that of which γάρ gives the reason is omitted, and must be supplied’; Humbert, , op. cit. no. 692Google Scholar: ‘en sautant fréquemment des intermédiaires de pensée dont l'explication n'est pas nécessaire’. It would seem that the opinion of Grimm, C. L. W.Thayer, J. H., A Greek-English Lexicon of the New Testament (Edinburgh, ⁴ 1908)Google Scholar, sub voce, II, 8, is too radical: ‘The particle is everywhere used in reference to something expressly stated’; ‘It is often said that the sentence of which γάρ introduces the cause, or renders the reason, is not expressed, but must be gathered from the context and supplied in thought’; ‘this ellipsis is wholly imaginary’. In this connection we should also mention Winer, G. B.Moulton, -W. F., A Treatise of the Grammar of New Testament Greek… (Edinburgh, ³ 1882), p. 365Google Scholar, who oppose ‘the doctrine of the “weakening” of conjunctions, which teaches that even particles with a sharply defined meaning, such as for, but, are in many cases altogether redundant, or are mere particles of transition’.

³¹ Cf. Lagrange, , op. cit. p. 145Google Scholar: ‘γάρ ne veut pas dire que la doctrine du sujet de ce verset résulte du précédent; c'est une reprise du sujet pour lui dormer un développement en harmonie avec le thème dogmatique’ (emphasis added). Cf. also Gal. 1. 11, where many manuscripts read a δέ.

³² Gal. 2. 17 and 3. 21 are the only cases in Paul where μή γένοıτο is followed by a conditional period (in 3. 21 it is an irrealis). It may be of interest in this connection to examine the continuations given to the remaining eleven Pauline uses of μή γένοıτο: Rom. 3. 4 (δέ); 3. 6 (έπει πς); 3. 31 (άλλά); 6. 2 (no particle); 6. 15 (no particle); 7. 7 (άλλά); 7. 13 (άλλά); 9. 14 (γάρ); II. I (γάρ); II. II (άλλά); I Cor. 6. 15 (no particle; cf., however, n. 7). The δέ and άλλά conjunctions point to a progression of thought.

³³ Cf. e.g. Grossouw, , op. cit. p. 144.Google Scholar

³⁴ We differ on this point from Mussner, , op. cit. p. 177 n. 46Google Scholar: ‘Ohne den Blick auf die unmittelbar folgenden Ausführungen des Apostels in den VV 18–21 kann der V 17 nicht richtig ausgelegt werden.’

³⁵ Klein, , Rekonstruktion, pp. 195–6 and 201–2Google Scholar, presents an intricate theory in which he distinguishes between the collective and empirical ‘we’ and the individual ‘gemeinschaftslose’ ‘I’. We may also refer here to the treatment of the ‘I’ in vv. 18–21 by Mussner, , op. cit. pp. 177–9Google Scholar, who, in discussion with Klein, points out that the ‘I’ is indeed more than a literary stylistic device – so e.g. Oepke, , op. cit. p. 61Google Scholar; Schlier, , op. cit. p. 96Google Scholar; Bonnard, , op. cit. p. 55Google Scholar – and that the ‘Übergang zur “Ich”-Form’, with its different nuances in v. 18 and vv. 19–21, has to be explained ‘aus sachlich-theologischen Gründen’.

³⁶ Cf. inter alios Bonnard, , op. cit. p. 55Google Scholar; Mundle, W., ‘Zur Auslegung von Gal 2, 17. 18’, Z.N.W. 22 (1924), 152–3Google Scholar; Schlier, , op. cit. pp. 97–8.Google Scholar Schlier also mentions the first explanation, but explicitly notes that the second is to be preferred. See also Hahn, , art. cit. pp. 53–4Google Scholar and n. 76: v. 18 refers to the present (and future) situation. If the Law were built up again, ‘dann würde allerdings erneut gelten, daß…niemand das Gesetz der Werke wahrhaft erfüllen, vielmehr nur an ihm schuldig werden kann’ (p. 53). Hahn admits a motivating sense for the γάρ in v. 18. ‘Die Wendung άρα ξρıστς άμαρτιας δıάκονος darf wegen des mit γάρ eingeleiteten V. 18 keinesfalls nur auf die Vergangenheit des Gläubigwerdens bezogen werden, sondern besitzt ebenso Relevanz für die Gegenwartssituation, umdiees in V. 18 geht’ (n. 76). But is such a double reference (past and present – future) for the phrase ‘agent of sin’ in v. 17b likely?

³⁷ Cf. n. 17 and see also Duncan, G. S., The Epistle of Paul to the Galatians, Moffatt N.T. Comm. (London, 1944 = 1934), p. 69Google Scholar: ‘doing violence to the Will of God as clearly revealed in Christ’; van Dülmen, A., Die Theologie des Gesetzes bei Paulus, Stuttg. Bibl. Monogr. 5 (Stuttgart, 1968), p. 21Google Scholar n. 28, states: ‘nicht zuletzt ist die Rückkehr zum Gesetz schon deshalb Sünde, weil damit die ausschließliche Gerechtmachung durch Christus in Frage gestellt wird’.

³⁸ V. 15 is most probably not an independent sentence. Together with v. 16 it forms a long, very solemn period. Otherwise, e.g., Kertelge, K., ‘Zur Deutung des Rechtfertigungsbegriffs im Galaterbrief’, Bibl. Zeitschr. 12 (1968), 211–22, p. 218Google Scholar, who, however, rightly notes: ‘V. 15 erscheint von V. 16 aus als Ausdruck der im Christusglauben erkannten und zugleich überwundenen heilsgeschichtlichen Voraussetzung.’

³⁹ Cf. Klein, , Rekonstruktion, p. 191Google Scholar: ‘Der Unterschied zwischen V. 16 and V. 17a besteht also nicht in einem sachlichen Gedankenfortschritt, sondern lediglich darin, daß die dort als Gleichheit der Glaubenden ausgearbeitete Identität der Judenchristen und Heidenchristen hier als Gleichheit der Sünder geltend gemacht…wird.’