1 In this article, citations of Scripture follow the versification of the Rahlfs–Hanhart LXX (=MT for Joel).

2 For a helpful study of the Psalms texts in particular, see Moessner, D. P., ‘ Two Lords ‘at the Right Hand’? The Psalms and an Intertextual Reading of Peter's Speech (Acts 2.14–36)’, Literary Studies in Luke-Acts: Essays in Honor of Joseph B. Tyson (ed. Thompson, R. P. and Phillips, T. E.; Macon, GA: Mercer University Press, 1998) 215–32Google Scholar.

3 On the relatively secure texts of Joel 3.1–5, see Ziegler's, Joseph volume in the Göttingen LXX (Ziegler, J., ed., Septuaginta: Vetus Testamentum Graecum, vol. xiii: Duodecim Prophetae (Göttingen: Vandenhoeck & Ruprecht, 1943) 235–6)Google Scholar, and cf. also Origen's text in the Hexapla (PG 16.3.2953). The presence at Qumran of fragments of Joel (2.8–10, 10–23; 4.6–21) also demonstrates the stability of the text of Joel in general, although the lack of a witness for our particular passage renders this demonstration suggestive but nothing more than that. For the Qumran manuscripts and a critical apparatus, see Ulrich, E. et al. , Qumran Cave 4: The Prophets (DJD XV; Oxford: Clarendon, 1997) 243–6Google Scholar. Crenshaw, James (Joel: A New Translation and Commentary (AB 24C; New York: Doubleday, 1995) 53–4Google Scholar) suggests that the kinds of parallelism present in the text of Joel enabled the high level of consistency that text critics find in the manuscript tradition.

4 Cf. here the Targum on Joel (3rd–4th c.), which makes two noticeable, though still rather conservative, changes: ‘I will pour out my Holy Spirit’ (3.1) and ‘everyone who prays in the name of the Lord shall be delivered’ (3.5). Quoted from K. J. Cathcart and R. P. Gordo, The Targum of the Minor Prophets (ArBib 14; Wilmington, DE: Michael Glazier, Inc, 1989) 71. Although the Targum makes only slight changes to the text of Joel, the overall Tendenz of the changes it makes emphasises penitence as capable of restoring relationship with God. On this and other critical issues, see Flesher, P. V. M. and Chilton, B., The Targums: A Critical Introduction (Waco, TX: Baylor University Press, 2011) 199–200 CrossRefGoogle Scholar, 225–6.

5 Runge, S. E., ‘Joel 2:28–32a in Acts 2:17–21: The Discourse and Text-Critical Implications of Variation from the LXX’, Early Christian Literature and Intertextuality, vol. ii: Exegetical Studies (ed. Evans, C. A. and Zacharias, H. D.; New York: T&T Clark, 2009) 103–13Google Scholar.

6 The most significant alternative readings in both Joel 3.1–5 and Acts 2.17–21 reflect efforts to bring the two passages more closely in line with one another. This position, the current consensus, is well presented in Johnson, L. T., Septuagintal Midrash in the Speeches in Acts (Marquette, WI: Marquette University Press, 2002) 20–2Google Scholar.

7 Interpreters who note the changes but do not give them sustained attention include Haenchen, E., The Acts of the Apostles (Philadelphia: Westminster, 1971) 179 Google Scholar; Conzelmann, H., Acts of the Apostles (Hermeneia; Philadelphia: Fortress, 1987) 18–21 Google Scholar; Pervo, R., Acts (Hermeneia; Minneapolis: Fortress, 2008) 76–80 Google Scholar. Pervo notes the changes, but limits their significance to style. Also in this vein, Dodd describes them as ‘a few not very important variations [on the LXX]’. Dodd, C. H., According to the Scriptures (London: Nisbett & Co, 1952) 47 Google Scholar.

Descriptions of how Luke's changes resonate with the narrative and theology of Luke-Acts can be found in Barrett, C. K., Acts, vol. i (ICC; Edinburgh: T&T Clark, 1994) 135–8Google Scholar; Bruce, F. F., The Acts of the Apostles (NICNT; Grand Rapids: Eerdmans, 1988) 59–62 Google Scholar; Fitzmyer, J., Acts of the Apostles (AB; New York: Doubleday, 1997) 248–53Google Scholar; Johnson, L. T., The Acts of the Apostles (Sacra Pagina; Collegeville, MN: The Liturgical Press, 1992) 48–56 Google Scholar; Marguerat, D., Les Actes des apôtres (1–12) (CNT; Genève: Labor et Fides, 2007) 84–90 Google Scholar: Witherington, B. III, The Acts of the Apostles: A Socio-Rhetorical Commentary (Grand Rapids: Eerdmans, 1998) 142–43Google Scholar; Litwack, K. D., Echoes of Scripture in Luke-Acts: Telling the History of God's People Intertextually (JSNTSup 282; London: T&T Clark, 2005) esp. 155–79Google Scholar; and Moessner, ‘Two Lords at the Right Hand?’ 218–20. It should be noted that Johnson's commentary (pp. 49–50) offers a reading of ‘signs and wonders’ (Acts 2.19) that is exceptionally sensitive to the Mosaic background. See also Johnson, Septuagintal Midrash, 22.

8 Several of these images of restoration would lie close at hand for any reader of the Jewish Bible – for instance, the close alignment of the outpoured Spirit in Joel 3.1 with Isa 32.15 (the forsaken state of Israel will last ‘until the Spirit is poured on us from above …’) and Ezek 39.29 (‘I will never hide my face from them, when I pour out my Spirit upon the house of Israel’) (cf. Crenshaw, Joel, 163–72). In light of such intertextual connections, Luke need not be considered the source of all of the images of restoration, but, as I will show below, Luke does add to, heighten and reshape Joel's vision of restoration.

9 Dodd, According to the Scriptures, 62–4. On the trumpet call, cf. Joel 2.1, 15 with 1 Cor 15.52; 1 Thess 4.16; Rev 1.10; 4.1; 8.7, 8, 10, 12; 9.1, 13; 10.7. On κηρύσσειν, see Joel 1.14; 2.1, 15; 4.9. Dodd's argument is convincing provided one does not press it beyond its modest claims. Uses of similar terminology in Zech 9.9, 14 suggest that the imagery of two or more OT authors may have been combined in the process of shaping the imaginations of the first Christians. The circulation of the ‘Book of the Twelve’ as a single work during the first century, and perhaps earlier, further supports this possibility. On the ‘Book of the Twelve’ as a distinct corpus during this period, see Watson, F., Paul and the Hermeneutics of Faith (New York: T&T Clark, 2004) 80–8Google Scholar.

10 In this essay, citations of the NT are drawn from NA²⁷ and citations of the LXX are from Rahlfs–Hanhart. On the text-critical reasoning here, see above (nn. 2, 3). Translations are my own.

11 On these parallels, see Evans, C., ‘The Prophetic Setting of the Pentecost Sermon’, Luke and Scripture: The Function of Sacred Tradition in Luke-Acts (C. A. Evans and J. Sanders; Minneapolis: Fortress, 1993) 212–18Google Scholar. See also Strazicich, J., Joel's Use of Scripture and the Scripture's Use of Joel (Leiden: Brill, 2007) 275–7Google Scholar.

12 My way of putting the point here is indebted to Marguerat, Les Actes des apôtres, 86.

13 The verses immediately following Joel 3.5 demonstrate the essentially linear progression from repentance to restoration to judgement. After the assembly of people has repented, God says through the prophet, ‘Therefore, behold! In those days and at that time, when I bring back the captives of Judah and Jerusalem, I will gather all the nations and bring them down to the valley of Jehoshaphat, and there I will enter into judgement on them’ (Joel 4.1–2).

14 Haenchen reads against ‘in the last days’ and opts for ‘after these things’. The textual evidence for this is weak (B 076 C sa^mss). But the decision-maker for Haenchen is not textual – it is Lukan theology: ‘The last days do not begin as soon as the Spirit has been outpoured!’ (Haenchen, Acts, 179; see also id., ‘Schriftzitate und Textüberlieferung in der Apostelgeschichte’, ZTK 51 (1954) 162 Google Scholar). In an early response to Haenchen that anticipated the current consensus, Franz Mußner argued that the theology of Luke-Acts does, in fact, fit with the conviction of the church living within ‘the last days’ ( Mußner, F., ‘In den letzen Tagen (Apg 2,17a)’, BZ 5 (1961) 263–5Google Scholar).

15 For a similar view, see Zech 12.10 and cf. Crenshaw, Joel, 165.

16 Note, for example, God's judgement on the nations in Joel 4.4–8: those that sold the people of Judah and Jerusalem into slavery will now be sold by God into slavery.

17 Hays, R., Echoes of Scripture in the Letters of Paul (New Haven: Yale University Press, 1989) 20 Google Scholar.

18 Fisk, B., Do You Not Remember? Scripture, Story and Exegesis in the Rewritten Bible of Pseudo-Philo (JSPSup 37; Sheffield: Sheffield Academic Press, 2001) 98–101 Google ScholarPubMed.

19 Fisk, Do You Not Remember?, 207–17. Cf. e.g. LAB 15.5–6: ‘And I commanded the sea, and when the abyss was divided before them, walls of water stood forth. And there was never anything like this event since the day I said, “Let the waters under the heaven be gathered together into one place”, until this, day’ (trans. D. H. Harrington, OTP ii.323; emphasis added).

20 Fisk, Do You Not Remember?, 21. Emphasis original.

21 An extended discussion of this point can be found in Pao, D., Acts and the Isaianic New Exodus (Grand Rapids: Baker Academic, 2000) 123–9Google Scholar.

22 Here, Dietrich Rusam's observation regarding the necessity of the reconstitution of the Twelve (i.e. Israel) as a precondition for Pentecost is helpful: ‘Ebenso dient die Nachwahl des Matthias zum zwölfen Apostel der Vorbereitung auf Pfingsten … Nur wegen der Geistverleihung an Pfingsten in Jerusalem musste die Zwölfzahl der Apostel wieder hergestellt werden … Pfingsten ist für Lukas ein einmaliges Ereignis, und nur für Pfingsten war die Anwesenheit aller zwölf Apostel in Jerusalem als Zeichen für den Anspruch Jesu auf ganz Israel nötig’ ( Rusam, Dietrich, Das Alte Testament bei Lukas (BZNW 112; Berlin: de Gruyter, 2003) 287–8Google Scholar).

23 Prophecy in the first century ce had a particular construal, as the narrative of Acts suggests (visions, dreams, signs etc.) and as Rebecca Gray has helpfully described in her study of Josephus ( Prophetic Figures in Late Second Temple Jewish Palestine: The Evidence from Josephus (Oxford: Oxford University Press, 1993)Google Scholar).

24 Tannehill, Robert (The Narrative Unity of Luke-Acts: A Literary Interpretation, vol. ii (Minneapolis: Fortress, 1990) 29–33)Google Scholar helpfully describes how the promise of Joel is ‘realized progressively’ throughout Acts. Marguerat (Les Actes des apôtres, 89), following John Calvin, makes a similar observation about the extension of the meaning of ‘all flesh’ in commenting on the phrase πᾶς ὃς in Acts 2.21.

25 See Luke 2.29; 12.37, 43–7; 14.17–23; 17.7–10; 19.13–22; 20.10–11. Uses of these terms outside a context of discipleship might include Luke 7.2, 8, 10; 15.22; 22.50.

26 This is helpfully noted in Butticaz, S. D., L'identité de l'Église dans les Actes des apôtres: de la restauration d'Israël à la conquête universelle (BZNW 174; Berlin: de Gruyter, 2011) 106 Google Scholar.

27 Cf. 2 Kings 9.7; 17.13; Jer 7.25; 26.5; 29.19; 35.15; 44.4; Ezek 38.17; Zech 1.6. It is important to note that my argument here attends to the claim of a group of people as ‘my servants’. The singular phrase (‘my servant’) in its various Greek and Hebrew forms of expression is a more common designation for individuals such as David, Moses, Joshua, and others. See e.g. Acts 4.25.

28 Lev 25.42, 55. Note that ‘My servants’ also appears in Isa 65.7–15, a text rich in imagery of restoration. The Jubilee imagery also evokes Luke 4.18–19.

29 See also Runge, ‘Joel 2:28–32a in Acts 2:17–21’, 107–9.

30 A similar conclusion (not noting the OT background) is arrived at by means of a Bakhtinian carnivalesque reading in Strazicich, Joel's Use of Scripture, 281–2.

31 Cf. John Chrysostom, ‘Homily v on Acts ii.14’, Nicene and Post-Nicene Fathers, vol. xi (ed. P. Schaff; repr. Grand Rapids: Eerdmans, 1969) 33. I have altered the translations given by Chrysostom away from Schaff in consultation with the Greek in Comm. in Acta Apostolorum, PG lx.52.

32 Acts 2.40; 4.30; 5.12; 6.8; 14.3; 15.12; Philip (8.6, 13) is credited with signs but not wonders. See also Marguerat, Les Actes des apôtres, 88–9.

33 Cf. Exod 7.3, 9; 11.9; Deut 4.34; 6.22; 7.19; 11.3; 13.2; 26.8; 28.46; 29.2; 34.11; Esther 10.9; Ps 77.43; 104.27; 134.9 ; Wis 8.8; 10.16; Isa 8.18; 20.3; Jer 39.20; Bar 2.11; Dan 4.37.

34 Cf. Pao, Acts and the Isaianic New Exodus, 163–4.

35 Also note how Luke comes back to Joel 3.5 at Acts 2.39. Joel 3.5 reads: καὶ ἔσται πᾶς, ὃς ἂν ἐπικαλέσηται τὸ ὄνομα κυρίου, σωθήσεται· ὅτι ἐν τῷ ὄρει Σιων καὶ ἐν Ιερουσαλημ ἔσται ἀνασῳζόμενος, καθότι εἶπεν κύριος, καὶ εὐαγγελιζόμενοι, οὓς κύριος προσκέκληται. Luke echoes this closing phrase of Joel 3.5 in his climactic call to repentance in Acts 2.39, ὅσους ἂν προσκαλέσηται κύριος ὁ θεὸς ἡμῶν. The metalepsis here is noteworthy: by quoting the beginning and alluding to the end of Joel 3.5, Luke conspicuously omits the aspects of Joel 3.5 that limit salvation to Jerusalem. For observations along these lines, see Butticaz, L'identité de l'Église, 116; Marguerat, Les Actes des apôtres, 89–90.

36 Cf. Runge, ‘Joel 2:28–32a in Acts 2:17–21’, 105–7.

37 For an extended treatment of the significance of this move, see also Rowe, C. K., ‘Romans 10:13: What Is the Name of the Lord?’, HBT 22 (2000) 135–73Google Scholar; and for a treatment of this language in Luke and its theological implications, see id., Early Narrative Christology: The Lord in the Gospel of Luke (BZNW 139; Berlin/ New York: de Gruyter, 2006)Google Scholar.

38 A similar move occurs in Romans 10 with the reorientation of πᾶς to include not only Israel but all who believe in Jesus. Cf. Rowe, ‘What Is the Name of the Lord?’, 154–6.

39 Pao, Acts and the Isaianic New Exodus, 57 (emphasis added). I see Luke's universalism primarily as an extension of Isaiah's, and thus I hesitate to identify that universalism with an imperial ideology of universal domination such as the one identified in the recent, and very interesting, work of Butticaz (L'identité de l'Église, 52–3, 97–8).

40 This connection of Luke 19 with Acts 15 builds on the argument of Craig Hill: ‘[T]he rebuilding of David's “hut” (skenē) is a reference to the reestablishment of the Davidic dynasty, in line with the promise of II Samuel 7:16: “Your house and your kingdom shall be sure forever before me; your throne shall be established forever.” Because Jesus has taken the Davidic throne, James argues, the Gentile mission foreseen in Amos 9:11–12 … may now commence.’ Hill, C., ‘Restoring the Kingdom to Israel: Luke-Acts and Christian Supersessionism’, A Shadow of Glory: Reading the New Testament after the Holocaust (ed. Linafelt, T.; New York: Routledge, 2002) 196–7Google Scholar.

41 The way in which mission and eschatology co-exist for Luke is helpfully noted by Butticaz, L'identité de l'Église, 85.

42 Cf. here Joseph Fitzmyer: ‘[Luke] gives no specific references to the Torah or the Nebi'im and the modern reader will look in vain for the passages in the Old Testament to which the Lucan Christ refers …’ (here citing Luke 24.27, but see also 24.44). Fitzmyer, J., The Gospel according to Luke x–xxiv: Introduction, Translation, and Commentary (AB 28; Garden City, NY: Doubleday, 1985) 1558 Google Scholar.

43 Johnson, Septuagintal Midrash, 40.

44 On Luke's use of Scripture more generally, see Hays, R. B., Reading Backward: Figural Christology and the Fourfold Gospel Witness (Waco, TX: Baylor University Press, 2014) 55–74 Google Scholar, 99–100. On the specific use of Isa 58 as a co-text within Isa 61, see id., ‘The Liberation of Israel in Luke-Acts: Intertextual Narration as Countercultural Practice’, Reading the Bible Intertextually (ed. Hays, R. B., Alkier, S. and Huizenga, L. A.; Waco, TX: Baylor University Press, 2009) 101–16Google Scholar, esp. 107–8.

45 For more on this point, see Sterling, G. E., ‘Luke as a Reader of the LXX’, Biblical Interpretation in Early Christian Gospels, volume iii: The Gospel of Luke (ed. Hatina, Thomas R.; LNTS 376; London: T&T Clark, 2010) 161–79Google Scholar, esp. 174.