1 Phenomena related to authority have also intrigued scholars working with non-biblical ancient writings; cf. the case studies in the recent edited volumes O n Good Authority: Tradition, Compilation and the Construction of Authority in Literature from Antiquity to the Renaissance (ed. R. Ceulemans and P. de Leemans; Lectio 3; Turnhout: Brepols, 2015); Shaping Authority: How Did a Person Become an Authority in Antiquity, the Middle Ages and the Renaissance? (ed. S. Boodts, J. Leemans and B. Meijns; Lectio 4; Turnhout: Brepols, 2016).

2 Pace Rothschild, C. K. (New Essays on the Apostolic Fathers (WUNT 375; Tübingen: Mohr Siebeck, 2017)CrossRefGoogle Scholar), who argues for an allegorical and theological interpretation of Rome and Corinth: ‘Rome might connote those whom the author associated with proto-orthodoxy in Rome’ and Corinth ‘those whom the author associated with the wayward behavior Paul addresses in his first letter to the Corinthians’ (63). According to Rothschild, the letter is based on a ‘historical fiction’ and was ‘neither written from a church in Rome nor intended for (let alone delivered to) a Church in Corinth’ (66). However, the arguments Rothschild raises against viewing 1 Clement as a true letter from Rome to Corinth are weak. She mentions ‘the lack of historical evidence’ concerning churches in Corinth and Rome, and makes much of the fact that the epistolary elements were not accepted as the title of the letter. Furthermore, I am far from convinced that ‘Corinth’ would function as a self-evident symbol for wayward behaviour and ‘Rome’ for ‘proto-orthodoxy’ and ‘Paul's legacy’.

3 See e.g. Jaubert, A., Clément de Rome: Épître aux Corinthiens. Introduction, texte, traduction, notes et index (Sources Chrétiennes 167; Paris: Éditions du Cerf, 2000 ²) 20Google Scholar; Lindemann, A., Die Clemensbriefe (Handbuch zum Neuen Testament 17; Tübingen: Mohr Siebeck, 1992) 12Google Scholar; Lona, H. E., Der erste Clemensbrief (Kommentar zu den Apostolischen Vätern 2; Göttingen: Vandenhoeck & Ruprecht, 1998) 75–8Google Scholar.

4 For problems of the traditional dating, see Welborn, L. L., ‘On the Date of First Clement’, BR 29 (1984) 35–54Google Scholar. Gregory argues that there are no convincing arguments for a more exact dating than between ca. 70 and ca. 140 CE; see Gregory, A., ‘Disturbing Trajectories: 1 Clement, the Shepherd of Hermas and the Development of Early Christianity’, Rome in the Bible and the Early Church (ed. Oakes, P.; Carlisle: Paternoster/Grand Rapids, MI: Baker Academic, 2001) 142–66Google Scholar.

5 On the use of deliberative rhetoric in the letter, see Bakke, O. M., ‘Concord and Peace’: A Rhetorical Analysis of the First Letter of Clement with an Emphasis on the Language of Unity and Sedition (WUNT ii/143; Tübingen: Mohr Siebeck, 2001) 320–1Google Scholar.

6 While the letter itself names no author, the attribution of the letter to ‘Clement’ is relatively early and consistent (see below, n. 62). Lona argues that there is no reason to doubt the attribution of the letter to a person called Clement, but his function in Rome is a different matter. According to Lona, the literary quality of the writing and the influence and reception of the letter suggest that this Clement was an important figure in Rome (Clemensbrief, 71–2). Holmes similarly views Clement as ‘a (if not the) leading figure’ among Roman presbyters and bishops; see Holmes, M. W., Apostolic Fathers: Greek Texts and English Translations (Grand Rapids: Baker, 2007 ³) 35Google Scholar.

7 However, see Brooke's structural analysis of ‘actantial’ authority that ‘seems to inhere within a text as it is passed from one generation to another’ (Brooke, G. J., ‘Authority and the Authoritativeness of Scripture: Some Clues from the Dead Sea Scrolls’, Revue de Qumran 100 (2012) 507–23Google Scholar, at 509–14). Brooke argues that ‘texts have authority in the dynamics of their construction’. Brooke's analysis is enlightening, and I would only express the minor reservation that ‘texts make claims to authority’ (rather than ‘have’ it). As for connections between the writtenness of texts and their authority in a Jewish context, see H. Najman, ‘The Symbolic Significance of Writing in Ancient Judaism’, Past Renewals: Interpretative Authority, Renewed Revelation, and the Quest for Perfection in Jewish Antiquity (Supplements to the Journal for the Study of Judaism 53; Leiden: Brill, 2010) 3–38, at 23: ‘In the Second Temple period, a text's being authored or dictated (rather than, for example, an idea, vision, or law being orally communicated) by this or that figure, and a text's being transmitted by a line of faithful tradents, became marks of its authority.’

8 von Weissenberg, H., ‘Defining Authority’, In the Footsteps of Sherlock Holmes: Studies in the Biblical Text in Honour of Anneli Aejmelaeus (ed. de Troyer, K. et al. ; Leuven: Peeters, 2014) 679–95Google Scholar, at 690. For a model of authority as a complex relation with several variables, see J. Opsomer and A. Ulacco, ‘Epistemic Authority in Textual Traditions: A Model and Some Examples from Ancient Philosophy’, Shaping Authority, 21–46, at 32.

9 Brooke, ‘Authority and the Authoritativeness’, 515: ‘[A]s far as matters of authority go there is commonly a two-way process: authority is sometimes simply imposed on hearers or readers by authors and their texts (as with papal decrees), but generally it is also conferred through the assent of the hearer of reader.’ Thus, ‘any text's authority is also a construct of the interplay that runs through from author to editor to audience or readership and back again’ (519).

10 Von Weissenberg, ‘Defining Authority’, 685. Several other scholars describe the same process with different concepts; see the following footnote as well as Opsomer and Ulacco, ‘Epistemic Authority’, 36.

11 For this dynamics in the Dead Sea Scrolls, see Brooke, G. J., ‘Between Authority and Canon: The Significance of Reworking the Bible for Understanding the Canonical Process’, Reworking the Bible: Apocryphal and Related Texts at Qumran (ed. Chazon, E. G. et al. ; STDJ 58; Leiden: Brill, 2005) 85–104Google Scholar, at 85, 98; Martínez, F. García, ‘Rethinking the Bible: Sixty Years of Dead Sea Scrolls Research and Beyond’, Authoritative Scriptures in Ancient Judaism (ed. Popović, M.; Supplements to the Journal for the Study of Judaism 141; Leiden: Brill, 2010) 19–36Google Scholar, at 28–9. For examples in the Wisdom of Solomon, see Glicksman, A. T., ‘“Set Your Desire on My Words”: Authoritative Traditions in the Wisdom of Solomon’, Scriptural Authority in Early Judaism and Ancient Christianity (ed. Kalimi, I. et al. ; Deuterocanonical and Cognate Literature Studies 16; Berlin: de Gruyter, 2013) 167–82Google Scholar, at 180–1.

12 For Clement's ways of referring to ‘the scriptures’, see below, p. 132.

13 The only clear example of allegory is when Rahab's scarlet scarf is connected with Christ's blood (12.7–8); cf. Hagner, D. A., The Use of the Old and New Testaments in Clement of Rome (NovTSup 34; Leiden: Brill, 1973) 130CrossRefGoogle Scholar.

14 A. van der Kooij, ‘Authoritative Scriptures and Scribal Culture’, Authoritative Scriptures in Ancient Judaism, 55–71, at 57.

15 See the discussion of Sirach's prologue in van der Kooij, ‘Authoritative Scriptures’, 56–7, 60.

16 It has been argued that education is a setting that promotes the processes of canonisation in antiquity, since education necessitates a curriculum and a curriculum defines the core writings (see Brooke, G. J., ‘Canonisation Processes of the Jewish Bible in the Light of the Qumran Scrolls’, ‘For It Is Written’: Essays on the Function of Scripture in Early Judaism and Christianity (ed. Dochhorn, J.; Early Christianity in the Context of Antiquity 12; Frankfurt am Main: Peter Lang, 2011) 13–35Google Scholar, at 29–33). This observation is also highly interesting for the development of early Christianity. Yet at the time of 1 Clement, one can hardly speak of a curriculum of Christian teaching.

17 For comparison, see Rom 10.18–21, where the quotation formulae are somewhat ambiguous and Paul never concludes what one should learn from the quotations. The audience is left to deduce the point.

18 The translations of primary texts are my own.

19 For the practices concerning the circulation of letters among Christians in the first and second centuries, see Gamble, H. Y., Books and Readers in the Early Church: A History of Early Christian Texts (New Haven: Yale University Press, 1995) 96–101Google Scholar, 108–12.

20 G. H. van Kooten, ‘Ancestral, Oracular and Prophetic Authority: “Scriptural Authority” according to Paul and Philo’, Authoritative Scriptures in Ancient Judaism, 267–308.

21 For examples, see H. Najman, ‘Interpretation as Primordial Writing: Jubilees and its Authority Conferring Strategies’, Past Renewals: Interpretative Authority, Renewed Revelation, and the Quest for Perfection in Jewish Antiquity, 39–73, at 40.

22 J. N. Bremmer, ‘From Holy Books to Holy Bible: An Itinerary from Ancient Greece to Modern Islam via Second Temple Judaism and Early Christianity’, Authoritative Scriptures in Ancient Judaism, 327–60, at 328.

23 ἐνκεκύφατε εἰς τὰς ἱερὰς γραφάς, τὰς ἀληθεῖς, τὰς διὰ τοῦ πνεύματος τοῦ ἁγίου (45.2). Ἐπίστασθε γὰρ καὶ καλῶς ἐπίστασθε τὰς ἱερὰς γραφάς (53.1).

24 Though see the quotation formulae in Rom 9.15, 25; 2 Cor 6.2, in which God is the implicit subject; for discussion, see Kujanpää, K., The Rhetorical Functions of Scriptural Quotations in Romans: Paul's Argumentation by Quotations (NovTSup 172; Leiden: Brill, 2019) 98CrossRefGoogle Scholar.

25 22.1: Ταῦτα δὲ πάντα βεβαιοῖ ἡ ἐν Χριστῷ πίστις· καὶ γὰρ αὐτὸς διὰ τοῦ πνεύματος τοῦ ἁγίου οὕτως προσκαλεῖται ἡμᾶς (Ps 33.12–18, 20 LXX (34.11–17, 19 MT) follows).

26 Cf. Heb 3.7; 10.15. The mediation of the Spirit alone is not, however, an indication of the status of a text as sacred scripture, for Clement uses it of his own letter as well: ‘For you will give us great joy and gladness if you obey what we have written through the Holy Spirit (διὰ τοῦ ἁγίου πνεύματος)’ (63.2).

27 Nicklas makes an enlightening observation concerning the authority-conferring strategies in Revelation in relation to quotations: ‘Apocalypse does not offer explicit quotations from the Torah or from the Prophets; someone who claims to record a revelation received from God and Christ respectively, who directly envisions heaven, does not need to quote Scripture as an authority. Actually he cannot do it, in a manner of speaking, unless he wants to destroy at the same time the fiction of the immediacy of what was revealed to him’ (T. Nicklas, ‘“The Words of the Prophecy of this Book”: Playing with Scriptural Authority in the Book of Revelation’, Authoritative Scriptures in Ancient Judaism, 309–26, at 322).

28 For Philo's distinction between different kinds of oracles, see van Kooten, ‘Ancestral, Oracular and Prophetic Authority’, 293–6.

29 Van Kooten, ‘Ancestral, Oracular and Prophetic Authority’, 298.

30 Van Kooten, ‘Ancestral, Oracular and Prophetic Authority’, 305. For an example of Paul assuming various degrees of authority in different passages of the scriptures, see the discussion of Rom 10.5–8 in Kujanpää, Rhetorical Functions, 161–2.

31 Van der Kooij, ‘Authoritative Scriptures’, 55; van Kooten, ‘Ancestral, Oracular and Prophetic Authority’, 267.

32 As Carleton Paget argues, ‘the atmosphere of his epistle is both Jewish and strongly scriptural, and is so in an untroubled and confident manner with no sense that the Christian appropriation of the scriptures is a problem’ (Paget, J. Carleton, ‘1 Clement, Judaism, and the Jews’, Early Christianity 8 (2017) 218–50CrossRefGoogle Scholar, at 224).

33 ‘We are the portion (μερίς) of the Holy One’, Clement writes, just after quoting Deut 32.8–9 (‘his people, Jacob, became the Lord's portion’; see 1 Clem 29.1–30.1).

34 Jaubert, Clément de Rome, 30. For what this might tell of the Jewish-Christian relations in Rome, see Carleton Paget, ‘1 Clement’, 248–50.

35 Textual stability is demanded by some biblical texts themselves: the Textsicherungsformel in Deut 4.2 and 12.31/13.1 (often unhelpfully called ‘canon formula’) forbids any additions or omissions to the commandments. Based on these passages, Revelation declares that God himself will punish anyone making changes to the book (Rev 22.18–19).

36 Similarly, von Weissenberg, ‘Defining Authority’, 681–4. To take an example from the Christian transmission of the Septuagint, the edition that bears the name of Lucian of Antioch makes stylistic changes to the wording of Septuagint around 300.

37 For Paul, see Koch, D.-A., Die Schrift als Zeuge des Evangeliums: Untersuchungen zur Verwendung und zum Verständnis der Schrift bei Paulus (BHT 69; Tübingen: J. C. B. Mohr 1986) 186–7Google Scholar; Stanley, C. D., Paul and the Language of Scripture: Citation Technique in the Pauline Epistles and Contemporary Literature (SNTSMS 74; Cambridge: Cambridge University Press, 1992) 259–60CrossRefGoogle Scholar; Kujanpää, Rhetorical Functions, 334–6, 339–40. For Clement, see Hagner, Use of the Old and New Testaments, 78–9.

38 For the quotation technique of Paul's and 1 Clement's contemporaries, see Stanley, Paul and the Language, 267–350.

39 The Greek text of 1 Clement is quoted according to Holmes, Apostolic Fathers.

40 No Greek variants that would offer support for 1 Clement's wording are preserved. The Greek text is cited according to J. W. Wevers's edition Exodus. Vetus Testamentum Graecum: Auctoritate Academiae Scientarium Gottingensis editum (Göttingen: Vandenhoeck & Ruprecht, 1991).

41 Clement of Alexandria quotes the same passage in verbatim agreement with 1 Clement (Stromateis 4.19 (118.3)), but since he was a great admirer of the letter, his quotation does not support a hypothesis of an otherwise unattested reading of the passage.

42 Lona views this as a case of indirect quotation that demonstrates the author's ability to create a new independent formulation (Clemensbrief, 551). The letter, however, introduces Moses's reply as a direct quotation.

43 ἐξαλειφθήτωσαν ἐκ βίβλου ζώντων (Ps 68.29 (69.29 MT)). Similarly, Lindemann, Die Clemensbriefe, 151.

44 In addition, as Rothschild (New Essays, 89) argues, the substitution may create a contrast between those who rebelled against Moses and went down to Hades where ‘death is their shepherd’ (1 Clem 51:4), and Moses as the shepherd who keeps his flock in the book of the living.

45 The Greek text is cited according to J. Ziegler's edition Ezechiel. Vetus Testamentum Graecum: Auctoritate Academiae Scientarium Gottingensis editum (Göttingen: Vandenhoeck & Ruprecht, 2015⁴).

46 The following witnesses read here ἁμαρτωλοῦ, just like 1 Clement: Codex Alexandrinus and the Alexandrian manuscripts 26 544 106-410, part of the catena manuscripts and a handful of mixed codices (C-403′ 456), the Arabic translation, Theodoret's commentary, Apostolic Constitutions and Pseudo-Cyprian.

47 Lona, Clemensbrief, 187; Hagner, Use of the Old and New Testaments, 54 (as one option). Yet the arguments for this appear weak. The main argument for the apocryphon hypothesis is that the following quotation in 8.3 does not agree with any passage of the canonical Ezekiel. However, it could be Clement's own compilation, a free paraphrase of Ezek 33 (cf. Holmes in his edition). Lona observes that when Clement of Alexandria quotes the same passage in 1 Clem 8.2, he does not name the source. Yet this is hardly convincing evidence for postulating a new source text.

48 Similarly, Lindemann, Die Clemensbriefe, 46.

49 In the Greek Ezekiel, both ἀσεβής and ἁμαρτωλός are used to translate רָשָׁע, which explains how such a variant may have arisen.

50 For example, in 1 Clem 53.3 the long quotation of God's speech follows Deut 9.13–14 almost verbatim. Yet in Deuteronomy, Moses does not answer. It is possible that Clement consulted Deuteronomy 9 but quoted Moses's answer in Exodus from memory.

51 See further Maier, H. O., The Social Setting of the Ministry as Reflected in the Writings of Hermas, Clement and Ignatius (Studies in Christianity and Judaism 12; Waterloo, ON: Wilfrid Laurier University Press, 2002) 120–1Google Scholar, 131.

52 Clement also uses the indefinite πού in quotation formulae in 15.2; 21.2; 26.2; 28.2; cf. Heb 2.6; 4.4.

53 The Masoretic text makes a rather different statement, not referring to human agents at all: ‘I will appoint peace as your overseer and righteousness as your taskmaster’ (וְשַׂמְתִּי פְקֻדָּתֵךְ שָׁלוֹם וְנֹגְשַׂיִךְ צְדָקָֽה).

54 No Greek variants that would offer support for 1 Clement's wording are preserved. The Greek text is cited according to J. Ziegler's edition Isaias. Vetus Testamentum Graecum: Auctoritate Academiae Scientarium Gottingensis editum (Göttingen: Vandenhoeck & Ruprecht, 1983³).

55 Hagner, Use of the Old and New Testaments, 67, 106, 228. Evans develops Hagner's arguments further and suggests that the quotation ‘has been heavily influenced (perhaps unconsciously) by ecclesiastical tradition concerning church offices’, reflected in Acts 6.1–6 (C. Evans, ‘The Citation of Isaiah 60:17 in I Clement’, VC 36 (1982) 105–7, at 106). Jaubert, Clément de Rome, 43, for her part, argues that Clement relies on a translation used in his social context. Yet it is questionable whether Clement's version can be called a translation. I wonder, moreover, why it is better to attribute the misquotation to some unknown circle in Rome rather than to Clement. It is undeniable that Clement modifies the wording of quotations at least in a handful of cases.

56 Lindemann, Die Clemensbriefe, 127 similarly comments on the convergence between the substitutions and Clement's intention.

57 ‘By presenting the rejection of the authority of the deposed leaders as a rejection of divinely established community structures, Clement made those structures explicit and by showing how the group's beliefs committed them to support for the leaders, he strengthened the normative character of those structures’ (Maier, Social Setting of the Ministry, 119).

58 Irenaeus also cites Isa 60.17 in connection with presbyters, but accurately according to the Septuagint (Adversus haereses 4 fragment 13). This shows that Clement could likewise have made his point with the Septuagint's wording, but the modified wording makes it clearer and more straightforward.

59 See Letteney, M., ‘Authenticity and Authority: The Case for Dismantling a Dubious Correlation’, Rethinking Authority in Late Antiquity: Authorship, Law, and Transmission in Jewish and Christian Tradition (ed. Berkovitz, A. J. and Letteney, M.; London: Routledge, 2018) 33–56CrossRefGoogle Scholar. Letteney suggests that textual authority and authenticity need to be decoupled and offers fascinating examples from late antiquity: a text may be considered authentic and authored by an important figure (in Letteney's example Jesus!) and still lack authority.

60 Najman, H., Seconding Sinai: The Development of Mosaic Discourse in Second Temple Judaism (Supplements to the Journal for the Study of Judaism 77; Leiden: Brill, 2003) 12Google Scholar; see further 9–17. As for the reception of a pseudonymous work, see Najman, ‘Interpretation as Primordial Writing’, 66: ‘We cannot assume that Second Temple readers of Jubilees, such as the author(s) of the Damascus Document who cite(s) Jubilees as Scripture, took Moses’ transcription of Jubilees to be an historical fact, because we cannot assume that they shared the relevant conception of history.’

61 Including 1 Clement in Codex Alexandrinus highlights an important theme that cannot be discussed in this article, namely authority and manuscripts. Brooke calls attention to the varying degrees of authority assigned to individual copies; see Brooke, ‘Authority and the Authoritativeness’, 520–1. In the case of Codex Alexandrinus, the inclusion of 1 and 2 Clement in the fine codex reflects the high status of these writings.

62 See Irenaeus, Adversus haereses 3.3 for the succession. According to Irenaeus, Clement even met Paul and Peter in Rome. While Irenaeus does not directly say that Clement wrote the letter, he clearly connects Clement's time as the bishop with the sending of the letter.

63 Origen, Commentarii in evangelium Ioannis 6.279, on John 1.29 (Sources Chrétiennes 157, p. 340). For further primary texts and discussion, see Lona, Clemensbrief, 67.

64 In the manuscript tradition, 1 Clement and the so-called 2 Clement are linked. For the Pseudo-Clementine Homilies (or Klementia) and Recognitions that recount Clement's life, adventures and travels with Peter, see Jones, F. S., ‘Introduction to the Pseudo-Clementines’, Pseudoclementina Elchasaiticaque inter Judaeochristiana (OLA 203; Leuven: Peeters, 2012) 7–49Google Scholar.

65 Didymus the Blind and the Syriac work the Apostolic Canons (both from the second half of the fourth century) seem to view 1 Clement as authoritative scripture. See Metzger, B. M., The Canon of the New Testament: Its Origin, Development, and Significance (Oxford: Clarendon, 1987) 214Google Scholar, also 222; Gregory, A., ‘1 Clement: An Introduction’, ExpTim 117 (2006) 223–30Google Scholar, at 224.