^{page 537 note 1} This essay revises and substantially enlarges upon material originally presented in my doctoral thesis: ‘The Law and the Spirit: An Investigation into the Theology of Galatians’ (PhD. dissertation, Princeton Theological Seminary, 1985)Google Scholar. For a thorough articulation of the global interpretation of Galatians in which the present essay is to be situated, see my forthcoming monograph, The Cross and the Spirit: A Study in the Argument and Theology of Galatians (Macon, GA: Mercer University Press, c. 1989)Google Scholar.

^{page 538 note 1} In addition to the commentaries, see especially the following: Conrat, Max, ‘Das Erbrecht im Galaterbrief (3:15–4:7)’, ZNW 5 (1904) 204–27;CrossRefGoogle ScholarBammel, Ernst, ‘Gottes ΔIAΘHKH (Gal. III 15–17) und das jüdische Rechtsdenken’, NTS 6 (1959–60) 313–19;CrossRefGoogle Scholar and Taylor, Greer M., ‘The Function of ΠIΣTIΣ XPIΣTOY in Galatians’, JBL 85 (1966) 58–76Google Scholar.

^{page 538 note 2} According to Greek and Roman law, a testament could be changed by the testator at his own discretion. For the Greek institution see Taubenschlag, Raphael, The Law of GrecoRoman Egypt in the Light of the Papyri, 2nd ed. (Warsaw: Panstwowe Wydawnicto Naukowe, 1955)Google Scholar. For the Roman testament see Conrat, ‘Das Erbrecht im Galaterbrief (3:15–18)’. The complete absence of the term πίστις from Gal 3. 15–18 speaks against the theory of Greer Taylor that the Roman fidei commissum is in view (‘The Function of ΠIΣTIΣ XPIΣTOY in Galatians’). Bammel has suggested the Jewish , ‘die Verfügung (Geschenk) eines Gesunden’ (‘Gottes ΔIAΘHKH [Gal. III 15–17] und das jüdische Rechtsdenken’, 315). This testament did not depend on the death of the testator but became effective immediately and was irrevocable. On the geographical distribution of this form of disposition, see Betz, H. D., Galatians: A Commentary on Paul's Letter to the Churches of Galatia (Hermeneia; Philadelphia: Fortress Press, 1979) 155, note 23Google Scholar.

^{page 539 note 1} See, for example, Bring, Ragnar, Commentary on Galatians, tr. Wahlstrom, Eric (Philadelphia: Muhlenberg Press, 1961) 152–5Google Scholar.

^{page 539 note 2} See, for example, Becker, Jürgen: ‘Zwischen Gesetz und Verheissung gibt es also keine Beziehung, weil das Gesetz ausserhalb des an das Abraham-Christus-Testament gebundenen Segens steht’ (Der Brief an die Galater [NTD 8; Göttingen: Vandenhoeck & Ruprecht, 1981] 39)Google Scholar. Compare Dülmen, Andrea van, Die Theologie des Gesetzes bei Paulus (SBM 5; Stuttgart: Verlag Katholisches Bibelwerk, 1968) 37–8;Google ScholarMussner, Franz, Der Galaterbrief, 4th ed. (HTKNT 9; Freiburg/Basel/Vienna: Herder, 1981) 241Google Scholar. As we shall see, this way of construing the relationship between the two, the law and the Abrahamic covenant, does represent Paul's own view, but it does not agree with the way Paul puts their relationship in v. 17.

^{page 540 note 1} See Oepke, Albrecht, Der Brief des Paulus an die Galater, 4th ed. posthumously published and enlarged by J. Rohde (ThHK 9; Berlin: Evangelische Verlagsanstalt, 1979) 116;Google ScholarSchlier, Heinrich, Der Brief an die Galater, 5th ed. (MeyerK 7/14; Göttingen: Vandenhoeck & Ruprecht, 1971)158Google Scholar.

^{page 540 note 2} A clear example in Paul is 1 Cor 15. 29 (a proof from the practice of baptism on behalf of the dead), where the voice in the third person plural, a distancing device, marks the argument as ad hominem.

^{page 540 note 3} The citation is from The Institutio Oratoria of Quintilian, vol. 3 (LCL) 405Google Scholar. The Greek term is παρομολογία, which Rutilius defines as follows: cum aliquot res adversario concedimus, deinde aliquid inferimus, quod aut maius sit quam superiora, aut etiam omnia quae posuimus infirmet. See Lupi, P. Rutilii, De figuris sententiarum et elocutionis, ed. with prolegomena and commentary by Edward Brooks, Jr. (Mnemosyne Suppl. 11; Leiden: E. J. Brill, 1970) 23Google Scholar.

^{page 540 note 4} See Lausberg, Heinrich, Handbuch der literarischen Rhetorik: Eine Grundlegung der Literarwissenschaft, vol. 2 (Registerband), 2nd ed. (2 vols.; Munich: Max Huber, 1973) 716Google Scholar. Although Lausberg associates this figure with what Quintilien calls argumenta a persona (Inst. 5.10.23), Quintilian himself does not speak explicitly of argumentum ad hominem.

^{page 540 note 5} The type of argument ad hominem we have in view here, which builds on the premises of the interlocutor, is to be distinguished from the abusive argument ad hominem, in which one attacks not the ideas but the person of the opponent. On the distinction, see Perelman, Ch. and Olbrechts-Tyteca, L., The New Rhetoric: A Treatise on Argumentation, tr. Wilkinson, J. and Weaver, P. (Notre Dame/London: University of Notre Dame Press, 1969) 110–12Google Scholar. Cf. Fischer, David Hackett, Historians’ Fallacies: Toward a Logic of Historical Thought (New York/Evanston: Harper Torchbooks, 1970) 290–1;Google ScholarCohen, Morris R. and Nagel, Ernest, An Introduction to Logic and Scientific Method (New York: Harcourt, Brace and Company, 1934) 380Google Scholar.

^{page 541 note 1} The finite construction with the legal term άκυροṺν should have sufficed, were Paul not treating the law as an άκυροûν of the covenant. As the sentence stands it says that the law's nullification of the covenant does not in fact result (εἱς τ) in the destruction () of the promise, the reason being that it is illegal to nullify a ‘ratified covenant’ (3. 15; cf. ‘a covenant previously ratified by God’, v. 17).

^{page 541 note 2} Compare Schlier, Der Brief an die Galater, 149Google Scholar (‘Verheissung und Gesetz widersprechen sich’).

^{page 542 note 1} As I have pointed out already, this suggestion was in fact made by Isaac Watts in his book, Logicke: or, the Right Use of Reason in the enquiry after truth (London, 1725)Google Scholar. Recently Folker Siegert has identified Paul's expression, λέγω κατ ἃνθρωπον, with argumentum ad hominem, but he has in mind a kind of rhetorical condescension on the apostle's part to his hearers' own level and cites Bjerkelund's work (see note 2, page 543 below) apparently with approval. See Argumentation bei Paulus gezeigt an Röm 9–11 (WUNT 34; Tübingen: J. C. B. Mohr [Paul Siebeck], 1985) 228Google Scholar.

^{page 542 note 2} It is not listed among the Greek entries in the index to Rhetores latini minores, ed. Halm, Karl (Leipzig: Teubner, 1863)Google Scholar. Nor does it occur in Ernesti, J. C. T., Lexicon technologiae graecorum rhetoricae (Hildesheim: Georg Olms, 1962 reprint of 1795 Leipzig ed.)Google Scholar.

^{page 542 note 3} I have not found the phrase in Quintilian or Cicero. It is not listed in the index to Halm's Rhetores latini minores, nor is it registered in Ernesti, J. C. T., Lexicon technologiae latinorum rhetoricae (Hildesheim: Georg Olms, 1962 reprint of 1797 Leipzig ed.)Google Scholar. Although Lausberg discusses it under the heading of argument a persona, which he derives from Quintilian, he is unable to adduce the technical phrase, argumentum ad hominem, from antiquity (see note 4, page 540 above). The same goes for Whately, Richard, Elements of Logic, rev. ed. (Boston/Cambridge: James Munroe and Company, 1859) 237–8Google Scholar. The OED (s.v. ‘argument’) gives the earliest instance of which I am aware when it cites Locke, John, An Essay Concerning Human Understanding (1695 ed., Book IV. chap. 21. par. III)CrossRefGoogle Scholar.

^{page 542 note 4} The New Testament and Rabbinic Judaism (London: Athlone, 1956) 394–400Google Scholar.

^{page 542 note 5} Daube makes the following observation: ‘In Tannaitic literature the technical apology for an overbold statement is kibheyakhol, “if one could (say so)” … Kibheyakhol, then, as far as sense is concerned, is very near Paul's “I speak after the manner of men.” But it is manifestly another expression. Paul's phrase is not a rendering of it’ (The New Testament and Rabbinic Judaism, 396–7)Google Scholar.

^{page 542 note 6} Galatians, 155Google Scholar. Not long after Watts, Karl Ludwig Bauer apparently also recognized κατά ἃνθρωπον as a technical expression in Paul, classifying it as e concessis, by which he means ‘not requiring demonstration’ (i.e., to argue from what is commonly accepted). See Logica Paullina (Halae: Impensis Orphanotrophei, 1774) 299Google Scholar. Although Bauer discusses 1 Cor 9. 8, together with passages where the phrase itself does not appear, he neglects to mention Ga1 3. 15 and Rom 3. 5 (see Logica Paullina, 299–302)Google Scholar.

^{page 543 note 1} Ibid., 154–5. He cites Burton, E. D. approvingly: ‘I draw an illustration from everyday human practice’ (A Critical and Exegetical Commentary on the Epistle to the Galatians [ICC; Edinburgh: T. & T. Clark, 1921] 178)Google Scholar. Some of the examples listed by Betz are discussed below.

^{page 543 note 2} Carl J. Bjerkelund seeks to link Paul's expression with the rabbinic use of worldly comparisons (specifically parables) to expound the Torah. The rabbis are said to justify this practice by pointing out that the Torah speaks after an earthly fashion. See ‘“Nach menschlicher Weise rede ich”: Funktion and Sinn des paulinischen Ausdrucks’, Studia Theologica 26 (1972) 63–100CrossRefGoogle Scholar. But this theory cannot account for Rom 3. 5 and 1 Cor 9. 8, where κατά ἃνθρωπον λέγω (λάλω) does not refer to earthly comparison (on 1 Cor 9. 8 see note 4, below).

^{page 543 note 3} See Malherbe, Abraham J., ‘The Beasts at Ephesus’, JBL 87 (1968) 71–80Google Scholar.

^{page 543 note 4} As the construction with μή indicates, Paul's point is not ‘Am I giving human examples?’ (which he is doing), but ‘I am not speaking on merely human authority, am I?’

^{page 543 note 5} Rom 3. 5 offers the only exact verbal parallel to Gal 3. 15. Paul ‘apologizes’ for even voicing the blasphemous question he has just formulated.

^{page 543 note 6} The index used was Mayer, Günter, Index Philoneus (Berlin/New York: Walter de Gruyter, 1974)CrossRefGoogle Scholar. A search of the Septuagint, Josephus, and Epictetus produced no examples. The following concordances were consulted: Hatch, Edwin and Redpath, Henry A., A Concordance to the Septuagint, vol. 1 (2 vols.; Oxford: Clarendon Press, 1897);Google ScholarRengstorf, K. H., A Complete Concordance to Flavius Josephus, vol. 1 (3 vols.; Leiden: E. J. Brill, 1973);Google ScholarEpicteti: Dissertations ab Arriano Digestae, ed. Schenkl, H. (Stuttgart: Teubner, 1965)Google Scholar (includes index verborum). The quotations are from the Loeb Classical Library (translations mine).

^{page 544 note 1} For some of the examples listed below, the following indices and lexicons were used: Lampe, G. W. H., A Patristic Greek Lexicon (Oxford: Clarendon Press, 1961);Google ScholarSauppe, G., Lexilogus to Xenophonteus (Hildesheim/New York: Georg Olms, 1971);Google ScholarPowell, J. E., A Lexicon to Herodotus (Hildesheim: Georg Olms, 1962);Google Scholar the indices provided within the various volumes of The Oxyrhynchus Papyri, ed. Grenfell, B. P. and Hunt, A. S. et al. (49 vols. to date; Great Britain: Egyptian Exploration Society, 1898–1982)Google Scholar. Except where indicated, the quotations are from the Loeb Classical Library editions of the Greek texts.

^{page 545 note 1} The text is reproduced in Denis, A. M., Fragmenta Pseudepigraphorum Quae Supersunt Graeca (PVTG 4; Leiden: E. J. Brill, 1970) 62Google Scholar.

^{page 545 note 2} For the text see Porphyrii Opuscula Selecta, ed. Nauck, A. (Leipzig: Teubner, 1886)Google Scholar.

^{page 545 note 3} Another example, not included in the preceding and difficult to date, is the variant to 2 Cor 11. 17 in the Leicester Codex (MS. 69), where in place of κύριον we have ἃνθρωπον. Here speaking κατά ᾃνθρωπον is the opposite of speaking ώς έν άφροσύνη.

^{page 546 note 1} In cap. III ep. ad Galat. comm. (PG 61.654) (my translation). Theodor Zahn came to the same conclusion as Chrysostom, upon examining other instances of the phrase in Paul. See Der Brief des Paulus an die Galater (Leipzig: A Deichert, 1905) 160–1Google Scholar.

^{page 546 note 2} Schlier, Der Brief an die Galater, 150;Google ScholarBetz, , Galatians, 159;Google ScholarBruce, F. F., The Epistle to the Galatians (NIGTC; Grand Rapids: William B. Eerdmans, 1982) 174;Google ScholarRidderbos, Hermann, The Epistle of Paul to the Churches of Galatia (NIC; Grand Rapids: Wm. B. Eerdmans, 1953) 135Google Scholar.

^{page 547 note 1} The inheritance terminology is found in 3. 18; 3. 29; 4. 1; 4. 7; 4. 30. The promise terminology is found in 3. 14; 3.16, 17, 18; 3.19; 3.21, 22; 3.29; 4.23; 4.28. The language of ‘blessing’ occurs in 3.8, 9; 3.14. It must be observed that in 3. 6–9 Paul does not identify ‘justification’ with ‘the blessing’ but rather uses both Gen 15. 6 and Gen 12. 3 to link ‘faith’ with ‘the blessing’.

^{page 547 note 2} The form γινώσκετε in 3. 7 is either the imperative or the indicative. Since Paul does not announce the words from Gen 15. 6 (Gal 3. 6) as Scripture, he apparently expects his readers to recognize the sentence, in which case they ‘know’ that ‘those from faith are sons of Abraham’.

^{page 547 note 3} The switch to the first person plural (‘our hearts’) is no doubt to be explained by the fact that Paul quotes from tradition here or at least adopts traditional language.

^{page 548 note 1} Through baptism one ‘puts on’ Christ and is therefore ‘in’ Christ and ‘of’ Christ. In specifying this condition for being in Christ, Paul does not argue for it but introduces it as familiar tradition in order to go on to make his point in v. 29: ‘And since you are of Christ, you are seed of Abraham, heirs according to the promise.’ That is to say, ‘being in Christ’ is the sole condition of ‘(realized) heirship’, and ‘being in Christ’ is the form in which all believers are related to Christ.

^{page 548 note 2} See Sanders, E. P., Paul, the Law, and the Jewish People (Philadelphia: Fortress Press, 1983)Google Scholar. According to Sanders, ‘The debate in Galatians is a debate about “entry” in the sense of what is essential in order to be considered a member at all’ (20).