¹ Christian Worship in the Primitive Church (Edinburgh: T. & T. Clark, 1934) 85–86.Google Scholar The fascinating controversy over the age and reliability of the sermon reports in Acts will be ignored here.

² Tannaitic Parallels to the Gospels (SBLMS 6; Philadelphia: Society of Biblical Literature, 1951) 84–114, esp. 88.Google Scholar

³ On the homiletic midrashim, see Heinemann, Joseph, “Preaching (in the Talmudic Period),” EncJud 13 (1971) 994–98Google Scholar, and literature cited there.

⁴ E.g., Borgen, Peder, Bread from Heaven (Leiden: Brill, 1965)Google Scholar; Bowker, J. W., “Speeches in Acts: A Study in Proem and Yellamedenu Form,” NTS 14 (1967) 96–111CrossRefGoogle Scholar; and Smith, Tannaitic Parallels. These studies have been criticized by Donfried, Karl Paul, The Setting of Second Clement in Early Christianity (Leiden: Brill, 1974) 19–48Google Scholar; and Davies, W. D., The Setting of the Sermon on the Mount (Cambridge: Cambridge University Press, 1977) 5–9.Google Scholar It should be noted that Borgen's method is less open to the charge of anachronism since he compares John to Philo as well as to Rabbinic sources.

⁵ Bultmann, Rudolf, Theology of the New Testament (2 vols.; New York: Scribner's, 1951–1955) 1. 99, 105–6Google Scholar; Dahl, Nils, “Form-Critical Observations on Early Christian Preaching,” in idem, Jesus in the Memory of the Early Church (Minneapolis: Augsburg, 1976) 30–36Google Scholar; and Wilckens, Ulrich, Die Missionsreden der Apostelgeschichte (3d ed.; Neukirchen-Vluyn: Neukirchener Verlag, 1974) 72–81.Google Scholar Bo Reicke's attempt to classify all the purported fragments of early Christian preaching, though interesting, is too broad and speculative to provide definitive form-critical categories (“A Synopsis of Early Christian Preaching,” in Fridrichsen, Anton, ed., Root of the Vine: Essays in Biblical Theology [Westminster: Dacre, 1953] 128–60).Google Scholar

⁶ On the Epistle of Jeremiah, see Pfeiffer, Robert, History of New Testament Times (New York: Harper, 1949) 248.Google Scholar For Hebrews, see Kümmel, Werner Georg, Introduction to the Ntew Testament (London: SCM, 1975) 398.Google Scholar

⁷ Introduction to the New Testament, vol. 2: History and Literature of Ancient Christianity (Philadelphia: Fortress, 1982) 273.Google Scholar

⁸ Donfried, Second Clement, 26.

⁹ According to Morton Smith (Tannaitic Parallels, 98–101, esp. 100), the synoptic “sermons” sometimes show a pattern not unlike this, but they lack such a neat formulaic division. The relation between synoptic sermons and the pattern I suggest here is vague at best. On the use of οὖν in NT paraenesis, see the short notice by Nauck, Wolfgang, “Das οὖν-paraeneticum,” ZNW 49 (1958) 134–35.CrossRefGoogle Scholar

¹⁰ Other similar usages of the term λόγος παρακλήσεως should also be noted. It is used in 1 Macc 10:24, evidently as “persuasive words,” and at 2 Macc 7:24 and 15:11 as “words of encouragement.” Cf. also 1 Mace 12:9. These passages may indicate that the term is already a fixed expression in the second century BCE, even if the meaning is somewhat general. Cf. also Acts 2:40 below.

¹¹ Of the various structuring principles of Hebrews that have been offered by scholars, some emphasize divisions by content, others by formal considerations, such as paraenetic passages. My proposed outline has more in common with the latter group. Of the many content divisions in print, see esp. Vanhoye, Albert, La structure littéraire de l'Epître aux Hébreux (StudNeot 1; Paris: Desclée de Brouwer, 1963)Google Scholar, and Koester, Introduction, 2. 274–16. Formal divisions are found in Haering, Th., “Gedankengang und Grundgedanken des Hebräerbriefs,” ZNW 18 (1917) 153–63Google Scholar; Nauck, Wolfgang, “Zum Aufbau des Hebräerbriefs,” in Eltester, Walther, ed., Judentum, Urchristentum, Kirche, Festschrift für Joachim Jeremias (BZNW 26; Berlin: Topelmann, 1960) 199–206Google Scholar; and MacRae, George, “A Kingdom that Cannot Be Shaken: The Heavenly Jerusalem in the Letter to the Hebrews,” Tantur Yearbook (1979–1980) 30–31.Google Scholar George MacRae offered many helpful suggestions during the writing of this article for which I am very grateful.

¹² MacRae (Kingdom, 31) notes this tendency in Hebrews.

¹³ The only difficulty here is the presence of two imperatives at 3:12–13, in the middle of the exempla section. It is possible that we have here two loosely structured cycles of the word of exhortation in a row, with the first ending at 3:12–13. This would account for the apparent alteration of Ps 95:10 by the addition of “therefore” (διό), which could have the effect in an oral sermonic context of transforming the last part of the scriptural quotation into a conclusion.

¹⁴ It could be argued that vs 9 is a second apodosis for the conditional sentence that begins in vs 8; in this case vss 8–10 should stand together as the conclusion. See BAG, 103, and cf. 1 Cor 15:17–18.

¹⁵ This larger pattern was suggested by a consultant of this journal.

¹⁶ Koester, Introduction, 2. 287.

¹⁷ According to Dionysius, bishop of Corinth, quoted by Eusebius Hist. eccl. 2.23.11.

¹⁸ As seen in 13:13–41, this theme is an integral part of salvation history for the author of Acts.

¹⁹ The first is attributed to Epimenides by Bruce, F. F. (The Acts of the Apostles [2d ed.; London: Marshall, 1954] 338)Google Scholar, but Haenchen, Ernst disagrees (The Acts of the Apostles [Philadelphia: Westminster, 1971] 524).Google Scholar The second is from Aratus.

²⁰ The call to repent that follows is likewise put in the form of an indirect report of God's proclamation. See below for Aristobulus's similar use of the same quotation from Aratus.

²¹ Betz, Hans Dieter, “2 Cor 6:14–7:1-Anti-Pauline Fragment?” JBL 92 (1973) 88–108Google Scholar; Fitzmyer, Joseph, “Qumran and the Interpolated Paragraph in 2 Cor 6,14–7,1,” CBQ 23 (1961) 271–80.Google Scholar

²² See Conzelmann, Hans, I Corinthians (Hermeneia; Philadelphia: Fortress, 1975) 165Google Scholar, and Barrett, C. K., A Commentary on the First Epistle to the Corinthians (London: Black, 1968) 220–21.Google Scholar Ulrich Luz has argued on grounds independent of mine (Das Geschichtsverständnis des Paulus [München: Kaiser, 1968] 118–19, 122–23)Google Scholar that the passage is based on an earlier midrash; Conzelmann agrees and Barrett allows this possibility.

²³ Another passage which may reflect our pattern is 2 Cor 3:7–4:1. According to Georgi, Dieter (Die Gegner des Paulus im 2. Korintherbrief [WMANT 11; Neukirchen-Vluyn: Neukirchener Verlag, 1964] 258–82)Google Scholar, 2 Cor 3:7–18 is based on a midrash of Exodus 34, which Paul has essentially “inverted” and satirized. The midrash could have been in the form of a word of exhortation, because hints of it can be detected. The first person plural indicatives differentiate it from my proposed pattern, but it is interesting to note that in each case a change to a hortatory subjunctive is all that is needed to arrive at a fairly regular repeating cycle of exempla/conclusion/exhortation. If, as Georgi suggests, Paul in 3:7–18 is undercutting the intention of an older midrash, this might have led him to emphasize indicatives where the audience would have expected exhortations.

²⁴ A comparison with the household codes in Eph 5:22–6:9 and Col 3:18–4:1 shows that 1 Peter has inserted a good deal more expository material between exhortations than either of the others, and has a stricter separation of indicative and imperative.

²⁵ With certain modifications, I follow the structural divisions of Käsemann, Ernst, “An Apologia for Primitive Christian Eschatology,” Essays on New Testament Themes (Philadelphia: Fortress, 1964) 178, 185–94.Google Scholar

²⁶ See Koester, Introduction, 2. 277–78, and Kraft, Robert, Barnabas and the Didache, in Grant, Robert M., ed., The Apostolic Fathers (5 vols.; New York: Nelson, 1965) 3. 1–3, 19–22.Google Scholar

²⁷ In this last instance, the mild imperative in 6.18b is more a command to perceive (αἰσθάνεσθαι), while the exhortation at 7.1 is stronger, calling on the hearer to choose to understand (νοεῖν) See Kraft, Barnabas, 3. 24–25.

²⁸ Ibid., 5–6, 134–36.

²⁹ Ibid., 5.

³⁰ On Barnabas and Hebrews, see Koester, Introduction, 2. 277–78 and 273–74. Besides the early Christian works treated in this section, several others could also be considered. The word of exhortation can be found in 2 Clement (e.g., 12.2–13.1, 13.2–14.1) and Polycarp's letter to the Philippians (e.g., 9.1–10.1), but both are so overburdened with exhortations that the pattern seems to break down. The sermon of John in the Acts of John (87–105) reflects the word of exhortation clearly.

³¹ See Eissfeldt, Otto, The Old Testament (New York: Harper & Row, 1965) 12Google Scholar, 15–17, and the literature cited there, esp. Rad, Gerhard von, “The Levitical Sermon in I and II Chronicles” in The Problem of the Hexateuch and Other Essays (New York: McGraw-Hill, 1966) 267–80.Google Scholar

³² Von Rad, “Levitical Sermon, “270.

³³ The Covenant Formulary (Philadelphia: Fortress, 1972) 34–36.Google Scholar Baltzer's anaylsis of Joshua 24 appears on pp. 19–27. The temptation to see a relation between the three-part covenant form and the word of exhortation should be avoided, despite the fact that the influence of the former can be found in many of the documents anaylzed here. The covenant formulary and its successors constitute a theological topos, while the word of exhortation is an oral rhetorical pattern that is not related to any one theme.

³⁴ Hartwig Thyen's study of the Hellenistic synagogue sermon, Der Stil der Jüdisch-Hellenistischen Homilie (Gottingen: Vandenhoeck & Ruprecht, 1955)Google Scholar is valuable in pointing out parallels with the Cynic-Stoic diatribe, but does not deal with the structuring pattern proposed here. Recently Stanley Stowers has argued (The Diatribe and Paul's Letter to the Romans [SBLDS 57; Chico: Scholars, 1981] 175–84)Google Scholar that the Sitz-im-Leben of the diatribe is school teaching and not preaching.

³⁵ The Theodotionic version has replaced the old LXX version of Susanna in most editions. Both are printed in Alfred Rahlfs edition (Stuttgart: Württembergische Bibelanstalt, 1935) the old LXX version above the Theodotionic one. The translation used here is from APOT.'.

³⁶ This was suggested by Professor John Strugnell in lectures at Harvard Divinity School, spring 1983.

³⁷ Collins's, John J. book Between Athens and Jerusalem (New York: Crossroad, 1983)Google Scholar was published after this article had been written. In his analysis of homiletical patterns in Testaments of the Twelve Patriarchs he outlines one pattern (158–59) which is identical to the word of exhortation. I completely endorse his view, although he does not relate this pattern to other literature as I have tried to do here. Collins also warns (159) against explaining this homiletical pattern in terms of the covenant form.

³⁸ See Charlesworth, James, The Pseudepigrapha and Modern Research with a Supplement (Chico: Scholars, 1981) 142–43, 211–20.Google Scholar

³⁹ Eusebius Praep. ev. 13.12, 664c–667a. My translation is that of Gifford, Edwin, Preparation for the Gospel (2 vols.; Grand Rapids: Baker, 1981).Google Scholar

⁴⁰ Quoted also in Acts 17:28 in the exempla of a word of exhortation treated above. The exhortations in these two words of exhortation are quite similar also. See Haenchen, Acts, 525.

⁴¹ Cf. Heb 2:9, another conclusion in a word of exhortation.

⁴² In addition to those examined above, 4 Macc 16:16–22 should be considered, as well as speeches from Josephus's Antiquities, e.g., 2.140–51.

⁴³ See the literature cited in nn. 3 and 4.

⁴⁴ Aristotle Rhetorica 1.3, Pseudo-Cicero Ad Herennium 1.2, and Quintilian 3.3–4.

⁴⁵ Haering (“Gedankengang,” 153–63) analyzes Hebrews into a similar four-part rhetorical schema, but not convincingly. Cf. the attempts to do the same regarding some of Paul's letters by Betz, Hans Dieter, “The Literary Composition and Function of Paul's Letter to the Galatians,” NTS 21 (1974–1975) 353–79CrossRefGoogle Scholar, and Church, F. Forrester, “Rhetorical Structure and Design in Paul's Letter to Philemon,” HTR 71 (1978) 17–33.CrossRefGoogle Scholar

⁴⁶ See, e.g., 8.140 and the various speeches in 9 42–48.

⁴⁷ See esp. Harding, W. F., The Speeches of Thucydides (Lawrence, KS: Coronado, 1973) 177–79Google Scholar, and Richard C. Jebb, “The Speeches of Thucydides,” in Harding's book, 264–66. It is assumed here that literary speeches reflect to a great extent oratorical conventions. See Kennedy, George, The Art of Persuasion in Greece (Princeton: Princeton University Press, 1963) 47–48.Google Scholar

⁴⁸ Kennedy, Persuasion, 158–64.

⁴⁹ Donfried, Second Clement, 35–36.