1 E.g. Green, S. G., Handbook to the Grammar of the Greek Testament (revised and improved edn; London: The Religious Tract Society, 1886) 344–5Google Scholar; Winer, G. B., A Treatise on the Grammar of New Testament Greek (trans. Moulton, W. F.; Edinburgh: T. & T. Clark, 1882 ⁹) 551Google Scholar; Kühner, R., Ausführliche Grammatik der Griechischen Sprache, vol. ii (ed. Gerth, B.; Hannover und Leipzig: Hahnsche Buchhandlung, 1904 ³) 264Google Scholar; Le P. F. and Abel, M., Grammaire du Grec Biblique suivie d'un choix de papyrus (Paris: J. Gabalda et Fils, 1927 ²)Google Scholar §78k; Robertson, A. T., A Grammar of the Greek New Testament in the Light of Historical Research (Nashville, TN: Broadman, 1934) 1151Google Scholar; Dana, H. E. and Mantey, J. R., A Manual Grammar of the Greek New Testament (Upper Saddle River, NJ: Prentice Hall, 1957) §232Google Scholar; Denniston, J. D., The Greek Particles (rev. Dover, K. J.; Oxford: Clarendon, 1959 ²) 359, 369–84Google Scholar; Blass, F. and Debrunner, A., A Greek Grammar of the New Testament and Other Early Christian Literature (trans. Funk, R.; rev. edn; Chicago: The University of Chicago Press, 1961) §447Google Scholar; Smyth, H. W., Greek Grammar (rev. Messing, G. M.; Harvard: Harvard University Press, 1984)Google Scholar §§2903–15; Louw, P. and Nida, E. A., Greek–English Lexicon of the New Testament Based on Semantic Domains, vol. i (New York: United Bible Societies, 1989)Google Scholar §§89.104, 89.136, 91.3; Young, R. A., Intermediate New Testament Greek: A Linguistic and Exegetical Approach (Nashville, TN: Broadman & Holman Publishers, 1994) 200Google Scholar; Wallace, D. B., Greek Grammar Beyond the Basics (Zondervan, 1996) 672Google Scholar; BDAG s.v. μέν; Porter, S. E., Idioms of the Greek New Testament (London: Continuum, 2005 ²) 212Google Scholar; Danker, F. W., The Concise Greek–English Lexicon of the New Testament (Chicago: University of Chicago Press, 2009)Google Scholar s.v. μέν; Muraoka, J T., A Greek–English Lexicon of the Septuagint (Leuven: Peeters, 2009)Google Scholar s.v. μέν.

2 Levinsohn, S. H., Discourse Features of New Testament Greek: A Coursebook on the Information Structure of New Testament Greek (Dallas: SIL International, 2000 ²) 170Google Scholar. See also Levinsohn, S. H., Textual Connections in Acts (SBLMS 31; Atlanta: Scholars, 1987) 143–4Google Scholar.

3 Runge, S. E., Discourse Grammar of the Greek New Testament: A Practical Introduction for Teaching and Exegesis (Lexham Bible Reference Series; Peabody, MA: Hendrickson Publishers, 2010) 75Google Scholar. Runge notes in n. 7 of the same page that the related clause is ‘most often introduced by δέ’ (emphasis added).

4 Bakker, E. J., ‘Boundaries, Topics, and the Structure of Discourse: An Investigation of the Ancient Greek Particle dé ’, Studies in Language 17 (1993) 301CrossRefGoogle Scholar.

5 C. J. Fresch, ‘Discourse Markers in the Septuagint and Early Koine Greek with Special Reference to The Twelve’ (PhD diss., University of Cambridge, 2015) 209.

6 Credit for this description is due to Margaret Sim, who wrote it in a private correspondence to me on 22 August 2014. To this description, I would add, based on my research in the LXX and papyri, that μέν also communicates that optimal relevance with regard to a preceding or established topic is attained only by understanding the corresponding materials as an intimately connected, cohesive unit.

7 This use is traditionally labelled μέν solitarium.

8 So also Beekes's etymological dictionary (though it covers more than the Koine period). Beekes provides an entry on μέν that simply states ‘emphatic particle’ and points the reader to the entry on μήν, which then describes μέν as having the same function as μήν, though being slightly weaker. R. Beekes, with the assistance of van Beek, L., Etymological Dictionary of Greek (Leiden Indo-European Etymological Dictionary Series, vol. x/1; Leiden: Brill, 2010)Google Scholar s.v. μέν.

9 Porter, Idioms of the Greek New Testament, 212. So also Young, New Testament Greek, 200, though Young does add that there may be an inherent discourse function to link thoughts together.

10 Chamberlain, W. D., An Exegetical Grammar of the Greek New Testament (New York: The MacMillan Company, 1941) 161Google Scholar.

11 Louw and Nida, Greek–English Lexicon, §91.6.

12 Danker, Concise Greek–English Lexicon, s.v. μέν. So also BDAG s.vv. μέν, μήν 1.

13 Blass and Debrunner, Greek Grammar, §447.

14 Wallace, Greek Grammar, 672; Green, Grammar of the Greek Testament, 344–5; Moulton, J. H. and Turner, N., A Grammar of New Testament Greek, vol. iii: Syntax (London: T&T Clark International, 2005) 331–2Google Scholar; Muraoka, A Greek–English Lexicon of the Septuagint, s.v. μέν. Mayser also does not mention a confirmatory or emphatic μέν, not even in his section on μέν without δέ ( Mayser, E., Grammatik der Griechischen Papyri aus der Ptolemäerzeit: Satzlehre, vol. ii.3 (Berlin und Leipzig: de Gruyter, 1934) 129–32Google Scholar). While Abel does mention a confirmatory μέν, he only does so with regard to the construction μὲν οὖν (with which I disagree, as the anticipatory function of μέν can be observed when this collocation is used – there is no need to posit more plurality than necessary). So also Wakker, G. C., ‘“Well I Will Now Present My Arguments”: Discourse Cohesion Marked by ΟΥΝ and ΤΟΙΝΥΝ in Lysias’, Discourse Cohesion in Ancient Greek (ed. Bakker, S. and Wakker, G.; Leiden: Brill, 2009) 70–1Google Scholar; Levinsohn, Textual Connections, 141–50; idem, Discourse Features, 170–1; Runge, Discourse Grammar, 78–80). In addition, Abel does not mention an emphatic or confirmatory use in his section on μέν without a corresponding δέ (Abel, Grammaire du Grec Biblique, §78k, l).

15 I am grateful to Geoffrey Horrocks for interacting with me and offering invaluable insights on the content of this section.

16 Robertson, Grammar of the Greek New Testament, 1150–1.

17 Jannaris, A. N., An Historical Greek Grammar: Chiefly of the Attic Dialect (Hildesheim: Olms, 1987) §1744Google Scholar.

18 Robertson, Grammar of the Greek New Testament, 1152.

19 Denniston, Particles, 359.

20 Denniston, Particles, 359.

21 Jannaris, Historical Greek Grammar, §1744 (emphasis original). It would seem that Smyth is one of the only old grammarians who views μέν solitarium as categorically emphatic, at least in Attic Greek, though he also notes that μέν did gradually lose its asseverative force and is not clear whether this applies only to μέν followed by a corresponding particle or also to μέν solitarium (Smyth, Greek Grammar, §§2895–2901).

22 ‘Discourse marker’, simply put, is a functional category that comprises any linguistic item that directs the reader in the building of his or her mental representation of the text. See Fresch, ‘Discourse Markers’, §1.1; Fischer, K., ‘Towards an Understanding of the Spectrum of Approaches to Discourse Particles: Introduction to the Volume’, Approaches to Discourse Particles (ed. Fischer, K.; SIP 1; Amsterdam: Elsevier, 2006) 6Google Scholar; M. M. Hansen, ‘A Dynamic Polysemy Approach to the Lexical Semantics of Discourse Markers (with an Exemplary Analysis of French toujours)’, Approaches to Discourse Particles, 25; A. Drummen, ‘Discourse Cohesion in Dialogue. Turn-Initial ΑΛΛΑ in Greek Drama’, Discourse Cohesion in Ancient Greek, 135.

23 D. M. Lewis, ‘Discourse Markers in English: A Discourse-pragmatic View’, Approaches to Discourse Particles, 52. Similarly, Richard Waltereit states: ‘Discourse markers are historical relics of speakers' strategies for manipulating the structure of the discourse or the interaction. Speakers discover that some word forms may have a certain appeal for textual and interpersonal purposes. They then start to employ these forms in communicative contexts that do not properly justify their primary use. Hearers will then discover that the form is being used “abusively”, thereby reanalyzing it as a discourse marker. Many of the properties of this new discourse marker can be shown to be related to the initial rhetorical strategies’ (R. Waltereit, ‘The Rise of Discourse Markers in Italian: A specific Type of Language Change’, Approaches to Discourse Particles, 66.)

24 Denniston, Particles, 359. Denniston's comparison of this to the word ‘certainly’ is apt, as it seems to have undergone similar change. Instead of two phonological forms developing as in the case of μέν and μήν, though, English speakers use alternative prosodies for its affirmative and anticipatory functions (affirmative: no pauses on either side of the word – falling intonation; anticipatory: brief pause before and after the word – usually rising intonation). A helpful, brief discussion on the discourse-marking capability of linguistic items and gradual nature of diachronic evolution can be found in Hansen, ‘A Dynamic Polysemy Approach’, 27–8.

25 In Attic Greek, before the fourth century bce, μήν would have been written μέν, owing to the fact that eta had not yet entered the Attic alphabet. So, when one says that μέν originally was emphatic, it could be that μέν itself was emphatic or μέν – pronounced μήν – was the emphatic original from which μέν developed. On the introduction of eta to the Attic alphabet and subsequent spelling shifts, see Horrocks, G., Greek: A History of the Language and its Speakers (Chichester: Wiley-Blackwell, 2010 ²) 27Google Scholar, 38, 40; Caragounis, C. C., The Development of Greek and the New Testament: Morphology, Syntax, Phonology, and Textual Transmission (WUNT 167; Tübingen: Mohr Siebeck, 2004) 28, 355–9Google Scholar.

26 Thus, the following representative sample was chosen on the basis of what occurrences of μέν are often treated as emphatic in Greek lexica and grammars.

27 Young, New Testament Greek, 200.

28 Morris, L., The Gospel according to Matthew (Grand Rapids: Eerdmans, 1992) 60Google Scholar (emphasis original).

29 For a similar argument, see Runge, Discourse Grammar, 76–7.

30 ‘The presence of μέν only serves to highlight and strengthen what was already present, ensuring that the reader or hearer does not miss the speaker's intended connection’ (Runge, Discourse Grammar, 77).

31 Dana and Mantey, Manual Grammar, 261.

32 Cf. Runge, Discourse Grammar, 81–2; Denniston, Particles, 370, 374–6; Robertson, Grammar of the Greek New Testament, 1151–3; Blass and Debrunner, Greek Grammar, §447(6); Winer, New Testament Greek, 553; Abel, Grammaire du Grec Biblique, §78k (mentions ἀλλά but not καί); Mayser, Grammatik der Griechischen Papyri, §164.4 (mentions καί but not ἀλλά); BDAG s.v. μέν 1β. For an example of μέν corresponding with καί, see the example in UPZ i.125 in section 1 above.

33 See Fresch, ‘Discourse Markers’, 188–212; Runge, Discourse Grammar, 83. Blass and Debrunner, Greek Grammar, §447(6); Abel, Grammaire du Grec Biblique, §78l; and Mayser, Grammatik der Griechischen Papyri, §164.4 also note that μέν can correlate with asyndeton.

34 BDAG s.v. μέν 2a.

35 So Robertson, Grammar of the Greek New Testament, 1152.

36 This second option is given partially by F. F. Bruce. He writes: ‘Formally, μέν implies a following δέ, in the shape of some such phrase as οἱ δὲ ἄλλοι, “but as for the others (they may speak for themselves)”’ ( Bruce, F. F., 1&2 Thessalonians (WBC 45; Waco: Word Books, 1982)Google Scholar 55.

37 Denniston, Particles, 380. Cf. Levinsohn, Discourse Features, 170; Robertson, Grammar of the Greek New Testament, 1152; Green, Grammar of the Greek Testament, 345; BDAG s.v. μέν 2; LSJ s.v. μέν 2.

38 Another example of μέν implying its corresponding clause may be found in Acts 1.1, Τὸν μὲν πρῶτον λόγον ἐποιησάμην περὶ πάντων, ὦ Θεόφιλε, ὧν ἤρξατο ὁ Ἰησοῦς ποιεῖν τε καὶ διδάσκειν. The anticipated δεύτερον (δέ) λόγον is implied and clear due to it being the present letter. Robertson, Grammar of the Greek New Testament, 1152 also acknowledges this, as does Danker, Concise Greek–English Lexicon, s.v. μέν b. So also in LXX Exod 4.23, where the Lord says: Ἐξαπόστειλον τὸν λαόν μου, ἵνα μοι λατρεύσῃ· εἰ μὲν οὖν μὴ βούλει ἐξαποστεῖλαι αὐτούς, ὅρα οὖν ἐγὼ ἀποκτενῶ τὸν υἱόν σου τὸν πρωτότοκον. The μέν anticipates and thereby implies the corresponding εἰ δέ clause: ‘But if you do want to send them out, I will not kill your firstborn son’. (See Fresch, ‘Discourse Markers’, 202–3.) I do not consider the μέν to be anticipating the apodosis of the conditional as I have observed no other instance in the LXX or third–first century papyri of that happening. Rather, in almost every case of εἰ/ἐὰν μέν, the μέν anticipates a corresponding conditional clause, usually introduced by εἰ δὲ μή.

39 BDAG s.v. μέν 2a.

40 See nn. 32 and 33 above, respectively.

41 Blass and Debrunner, Greek Grammar, §447(4).

42 E.g. Moo, D. J., The Epistle to the Romans (NICNT; Grand Rapids: Eerdmans, 1996)Google Scholar 631 n. 4; Cranfield, C. E. B., A Critical and Exegetical Commentary on the Epistle to the Romans, vol. ii (ICC; Edinburgh: T. & T. Clark, 1979; reprinted 1981) 512–13Google Scholar; Dunn, J. D. G., Romans 9–16 (WBC 38B; Dallas: Word Books, 1988) 585Google Scholar.

43 Blass and Debrunner, Greek Grammar, §447(4).

44 Another example of μέν … καί with a narrow scope can be observed in the papyrus UPZ i.125, which was given as an example in section 1 above.

45 Louw and Nida, Greek–English Lexicon, §91.6.

46 Ibid.

47 Sanders, J. N. and Mastin, B. A., The Gospel according to St John (Black's New Testament Commentaries; London: Adam & Charles Black, 1968; reprinted 1975) 265Google Scholar.

48 Sanders and Mastin, The Gospel according to St John, 265.

49 So also Barrett, C. K., The Gospel according to St John: An Introduction with Commentary and Notes on the Greek Text (London: SPCK, 1955; reprinted 1970) 324Google Scholar. In addition, BDAG s.v. μέν 2b, though not treating this μέν as correlative, at least notes that there is a contrast expressed between the two verses and that a δέ would be expected after ἔπειτα.

50 Louw and Nida, Greek–English Lexicon, §91.6.

51 Collins, Contra R. F., First Corinthians (SPS 7; Collegeville, MN: The Liturgical Press, 1999) 234Google Scholar. According to Collins, μὲν οὖν heightens the force of Paul's rejoinder.

52 This is noted also by Fee, G. D., The First Epistle to the Corinthians (rev. edn; NICNT; Grand Rapids: Eerdmans, 2014) 263Google Scholar n. 217; Kistemaker, S. J., Exposition of the First Epistle to the Corinthians (NTC; Grand Rapids: Baker Academic, 1993; reprinted 2007) 187Google Scholar; Danker, Concise Greek–English Lexicon, s.v. μέν b.

53 So Porter, Idioms of the Greek New Testament, 212; Dana and Mantey, Manual Grammar, 261.

54 Robertson, Grammar of the Greek New Testament, 1152; Green, Grammar of the Greek Testament, 345. Robertson also states this could be the original use of μέν and thus offers the rendering ‘first of all in truth’. (This is the only instance in which Robertson posits an emphatic reading of μέν in the New Testament. He comes close to doing so again in his reading of Luke 11.48 (p. 1153), but he stops just shy of claiming it outright.) This is a good example of the current situation and how ‘emphatic μέν’ has been propagated. Because of instances such as this in which maintaining the synchronically and diachronically established function of μέν is difficult or provides a less exciting reading, an historically defunct function is attributed to the particle.

55 Moo, Romans, 57; Morris, L., The Epistle to the Romans (Grand Rapids: Eerdmans, 1988) 55–6Google Scholar.

56 Runge, S. E., High Definition Commentary: Romans (Bellingham: Lexham, 2014) 15–17 Google Scholar. Moo also considers an option similar to Runge's (Moo, Romans, 57).

57 This third explanation was suggested to me by Margaret Sim in a private communication on 22 August 2014.

58 BDAG s.v. πρῶτος 2b.

59 Porter, Idioms of the Greek New Testament, 212; Dana and Mantey, Manual Grammar, 261.

60 For example, see Runge's discussion of Heb 9.1–11 in Runge, Discourse Grammar, 78–80.

61 It may be a case of etymological confusion as well, as μέν is regularly regarded as having developed from μήν. As I argued in section 2 above, however, even if μέν did come from μήν, the prospective function of μέν probably developed from an earlier emphatic function (that was then left to μήν) and that development would seem to attest to a functional divergence between the two. Moreover, a word's etymology does not necessarily indicate its meaning.