1 Unless otherwise stated, all translations of the Apostolic Fathers are from B. D. Ehrman, The Apostolic Fathers (2 vols.; LCL; Cambridge/London: Harvard University Press, 2003).

2 Porter, S. E., ‘Is “Dipsuchos” (James 1,8; 4,8) a “Christian” Word?’, Bib 71.4 (1990) 469–98Google Scholar. Porter's conclusions are cited approvingly by Moo, D. J., The Letter of James (PNTC; Grand Rapids: Eerdmans, 2000) 62Google Scholar; McCartney, D. G., James (BECNT; Grand Rapids: Baker Academic, 2009) 94Google Scholar; and Adam, A. K. M., James: A Handbook on the Greek Text (BHGNT; Waco, TX: Baylor University Press, 2013) 10Google Scholar, though none of them deal with the substance of Porter's argument.

3 Dates for James have ranged from the late forties through till 170 ce.

4 On the relationship between James and the Apostolic Fathers, see discussion below.

5 Porter, ‘Dipsuchos’, 484.

6 Porter, ‘Dipsuchos’, 483.

7 Porter, ‘Dipsuchos’, 484.

8 Porter, ‘Dipsuchos’, 483.

9 Porter, ‘Dipsuchos’, 485.

10 Stevens, C. S., ‘Does Neglect Mean Rejection? Canonical Reception History of James’, JETS 60.4 (2017) 767–80, at 775Google Scholar. Strangely, Stevens seems to treat 1 Clement and 2 Clement (an anonymous sermon) as if they were composed by the one author.

11 Stevens, ‘Neglect’, 775.

12 Stevens, ‘Neglect’, 775.

13 Bauckham, R., James (London: Routledge, 1999) 17Google Scholar.

14 The only work Hermas references explicitly is the elusive Eldad and Modad, using the introductory formula ὡς γέγραπται (‘as it is written’) (Herm. Vis. 2.3.4). James’ stronger tone can hardly be argued as a marker of authority as Stevens suggests (Stevens, ‘Neglect’, 775).

15 Furthermore, while it makes sense to assume that usage of the term went from low to high, the actual occurrences of δίψυχος are too few and far between to be able to tell us anything of use. 1 Clem. 23.2–3 and 2 Clem. 11.2, 5 are a citation from the same source, thereby limiting independent uses of δίψυχος κτλ. in 1 and 2 Clement to two (1 Clem. 11.2 and 2 Clem. 19.2). Likewise Did. 4.4 and Barn. 19.5 share a similar tradition. Therefore, rather than showing a progression from low to high usage, δίψυχος κτλ. appears once or twice in a few early texts, Hermas being the exception.

16 Porter, ‘Dipsuchos’, 475–6 discusses verbal and conceptual parallels.

17 Hagner, D. A., The Use of the Old and New Testaments in Clement of Rome (NovTSup 34; Leiden: Brill, 1973) 256CrossRefGoogle Scholar (emphasis added).

18 For a fuller treatment see Hagner, Clement, 248–56; Porter, ‘Dipsuchos’, 476.

19 See Allison, D. C., A Critical and Exegetical Commentary on the Epistle of James (ICC; New York: T. & T. Clark, 2013) 17 n. 83Google Scholar.

20 Gregory, A. F.,  ‘1 Clement and the Writings That Later Formed the New Testament’, The Reception of the New Testament in the Apostolic Fathers (ed. Gregory, A. F. and Tuckett, C.; Oxford: Oxford University Press, 2005) 129–57Google Scholar. Gregory's only reference to James appears in 154 n. 101.

21 Gregory, ‘Clement’, 154.

22 Porter, ‘Dipsuchos’, 476 concedes that ‘1 and 2 Clement are not apparently directly summarizing or paraphrasing James at this point’. Similarly for the Didache and Barnabas: ‘the verbal parallels with James are perhaps scantiest here’ (‘Dipsuchos’, 487). Even if one does conclude that a literary relationship is likely (see Johnson, L. T., The Letter of James (AB 37A; New York: Doubleday, 1996) 72–4Google Scholar), the direction of the intertextual borrowing must still be decided before a terminus ad quem can be reached. See Young, F. W., ‘The Relation of I Clement to the Epistle of James’, JBL 67 (1948) 339–45Google Scholar, who argues that James borrows from the earlier Clement.

23 Porter, ‘Dipsuchos’, 487. However, Niederwimmer, K., The Didache: A Commentary (Hermeneia; Minneapolis: Fortress, 1998) 106Google Scholar acknowledges: ‘What [διψυχήσεις] means concretely can no longer be determined with precision.’

24 See Gilmour, C., ‘Religious Vacillation and Indecision: Doublemindedness as the Opposite of Faith. A Study of Dipsychos and its Cognates in the Shepherd of Hermas and Other Early Christian Literature’, Prudentia 16 (1984) 33–42Google Scholar; Robinson, D., ‘The Problem of Dipsychia in the Shepherd of Hermas’, StPatr 45 (2010) 303–8Google Scholar. See also A. W. Strock, ‘The Shepherd of Hermas: A Study of his Anthropology as Seen in the Tension between Dipsychia and Harmartia’ (PhD diss., Emory University, 1984) 97–105, who thinks that James and Hermas make use of the term in quite similar (not conflated) ways.

25 J. Mutie, ‘The Identity of the Δίψυχος in the Shepherd of Hermas’ (unpublished paper delivered at the 2011 ETS Southwest Regional Meeting, 18 March 2011).

26 Mutie, ‘Identity’, 5–13.

27 Osiek, C., The Shepherd of Hermas: A Commentary (Hermeneia; Minneapolis: Augsburg, 1999) 185Google Scholar.

28 Mutie, ‘Identity’, 13–15.

29 Osiek, Hermas, 42.

30 Translation by J. B. Lightfoot, The Apostolic Fathers: Clement, Ignatius, and Polycarp, Part 1: Clement (London and New York: Macmillan, 1889²).

31 The existence of a literary relationship between James and Hermas has often been dismissed, with scholars viewing any similarities in light of a common religious background. See Ropes, J., A Critical and Exegetical Commentary on the Epistle of St. James (Edinburgh: T&T Clark, 1916) 89Google Scholar; Seitz, O. J. F., ‘Relationship of the Shepherd of Hermas to the Epistle of James’, JETS 63.2 (1944): 131–40Google Scholar; Dibelius, M., James (Hermeneia; Minneapolis: Fortress Press, 1988) 31–2Google Scholar; Metzner, R., Der Brief des Jakobus (Leipzig: Evangelische Verlagsanstalt, 2017) 18–9Google Scholar; though Allison, James, 23, admits that the exact relationship (or non-relationship) is very difficult to work out.

32 The eponym of this pseudepigraphon takes its referent from a short account in Num 11.26–30. The spelling of Eldad and Modad is based on the LXX Num 11.26 (Ελδάδ and Μωδάδ) rather than on Hermas (Μωδάτ) (cf. Heb אלדד and מידד).

33 Lightfoot, Fathers, 80–1. Cited in Allison, D. C., ‘Eldad and Modad’, JSP 21 (2011) 99–131, at 107Google Scholar.

34 Seitz, ‘Relationship’; idem, ‘Antecedents and Signification of the Term ΔΙΨϒΧΟΣ’, JBL 66 (1947) 211–19; idem, ‘Afterthoughts on the Term “Dipsychos”’, NTS 4 (1958) 327–34; idem, ‘Two Spirits in Man: An Essay in Biblical Exegesis’, NTS 6 (1959) 82–95.

35 Allison, ‘Eldad and Modad’.

36 Bauckham, R., ‘The Spirit of God in Us Loathes Envy: James 4:5’, The Holy Spirit and Christian Origins: Essays in Honor of James D. G. Dunn (ed. Stanton, G., Longenecker, B. W. and Barton, S. C.; Grand Rapids: Eerdmans, 2004) 270–81, at 280Google Scholar; idem, ‘Eldad and Modad’, Old Testament Pseudepigrapha, vol. i (ed. R. Bauckham, J. R. Davila and A. Panayotov; Grand Rapids/Cambridge: Eerdmans, 2012) 244–54.

37 Bauckham, ‘Spirit’, 280.

38 Allison, ‘Eldad and Modad’, 113.

39 Allison, ‘Eldad and Modad’, 113; Bauckham, ‘Spirit’, 281.

40 Bauckham, ‘Spirit’, 281.

41 Seitz, ‘Relationship’, 134; Tuckett, C., 2 Clement: Introduction, Text, and Commentary (Oxford: Oxford University Press, 2012) 217Google Scholar.

42 1 Clem. 23.3 reads τῇ ψυχῇ instead of τῇ καρδίᾳ. Seitz, ‘Relationship’, 135 argues that the author of 1 Clement has altered his source text: ‘Resuming our examination of the unknown “scripture”, as cited in I Clement, it is not difficult to understand the author's reason for substituting ψυχή in place of καρδία, since it simply brings into the quotation itself the root of the perhaps unfamiliar word δίψυχος, which is thus interpreted as διστάξων [sic; sc. διστάζοντες] τῇ ψυχῇ.’ Seitz is supported by Tuckett, 2 Clement, 217. Porter, ‘Dipsuchos’, 475, n. 21, dismisses this as conjecture.

43 Allison, ‘Eldad and Modad’, 112–14.

44 Allison, ‘Eldad and Modad’, 112.

45 Note that 2 Clem. 11.2 introduces the citation as ὁ προφητικὸς λόγος (‘the prophetic word’), which is suggestive of the source's eschatological orientation.

46 See Allison, ‘Eldad and Modad’, 100–1. However, little can be discerned about the pseudepigraphon by means of appealing to traditions about Num 11.26. Three distinct traditions exist concerning the content of Eldad and Modad's prophecy: (1) Joshua's succession of Moses; (2) prophecy about quails; and (3) Gog and Magog. All three are present in the Targum Pseudo-Jonathan and Targum Neofti, yet their presence in these Targums can easily be accounted for from their presence in the Talmud. Whereas in b. Sanh. 17a each prophecy presents an interpretation from different rabbis ((1) = R. Shimon; (2) = R. Eliezer; (3) = R. Nahman), the Targums affirms all three prophecies, assigning (1) to Eldad, (2) to Medad and (3) to both (however, in T. Neof. 1 Num 11.26, the assignment of (1) and (2) is reversed). Bauckham, ‘Eldad and Modad’, 248 concludes rightly: ‘It seems clear that this passage in the Targums must be dependent on the collection of three different opinions in the Talmud. So the set of three topics as such cannot be an older tradition.’ Thus Allison's use of rabbinic tradition (‘Eldad and Modad’, 102–6) in elucidating the character of the pseudepigraphon is flawed.

47 Bauckham, ‘Eldad and Modad’, 257; Allison, ‘Eldad and Modad’, 129. The primary reason for this is Jas 4.5, which Bauckham has argued is a Semitic citation from Eldad and Modad. See Bauckham, ‘Spirit’, 277.

48 Ever since J. Barr's devastating critique on the fallacious uses of etymology (The Semantics of Biblical Language (London: Oxford University Press, 1961), esp. 107–60), biblical scholars have been allergic to the word. This overreaction, while understandable, is unfortunate, given that the modern study of etymology is a robust linguistic field in its own right, and its legitimate use has much to offer to biblical studies.

49 Johnson, James, 69 concludes that no literary relationship exists between James and either document. Metzner, Jakobus, 19 n. 134 also notes that the Didache and Barnabas are not of Roman provenance, and the provenance of 2 Clement is also uncertain, thus problematising the view that the word arose within a confined geographical-linguistic locale (pace Marshall, S., ‘Διψυχος: A Local Term?’, SE 6 (1973) 348–51)Google Scholar.

50 Jackson-McCabe, M., of, review Matthew, James, and the Didache: Three Related Documents in Their Jewish and Christian Settings, CBQ 73 (2011) 212–13, at 213Google Scholar.

51 Schröter, J., ‘Jesus Tradition in Matthew, James, and the Didache: Searching for Characteristic Examples’, Matthew, James, and the Didache: Three Related Documents in Their Jewish and Christian Settings (ed. van de Sandt, H. and Zangenberg, J.; Atlanta: SBL, 2008) 233–55, at 237Google Scholar, writes, ‘it must remain an open question whether James can be put into the same framework as Matthew and the Didache’. See also M. Konradt, ‘The Love Command in Matthew, James, and the Didache’, Matthew, James, and the Didache, 271–88, at 288.

52 In fact, the issue of dependency never really comes up. Since the volume claims that the three documents attest a shared milieu of tradition, the stronger the shared background, the weaker claims of direct dependency become.

53 Osiek, Hermas, 27, notes one exception, Did. 1.5 and Herm. Mand. 2.4–6, though she suggests that ‘the best conclusion to draw is that there is a common written, or perhaps even oral, source behind the appearance of this one cluster of ideas in the two teachings on the Two Ways in these two otherwise quite different texts’. See also Niederwimmer, Didache, 51–2.

54 See Porter's attempt, ‘Dipsuchos’, 476, 485–6. Brox, N., Der Hirt des Hermas (Göttingen: Vandenhoeck & Ruprecht, 1991) 552Google Scholar thinks that one does not really need to deal with the use of διψυχέω in the Didache and Barnabas, since both ‘schreiben den Term … im Zusammenhang der Zwei-Wege-Lehre, beide innerhalb des schwer verständlichen Logions unbekannter Herkunft, so daß sie das Wort lediglich zitieren und nicht, wie 1 und 2 Klem und [Hermas], in ihren eigenen Sprachschatz aufgenommen haben’. Even so, one would still have to account for the relationship between the Two Ways document and the other works – an even more tentative task given the hypothetical reconstruction of the sources. For such a reconstruction, see Niederwimmer, K., ‘Der Didachist und seine Quellen’, The Didache in Context: Essays on its Text, History, and Transmission (ed. Jefford, C. N.; Leiden: Brill, 1995) 15–36Google Scholar.

55 Seitz, ‘Relationship’, 134–5: ‘In [LXX 1 Chron 12.34; Eng. 12.33], the Septuagint fails to reproduce the idea at all, substituting χεροκένως or some confusion of this word [Rahlfs LXX reads: ἑτεροκλινῶς], while [in Ps 11.3; Eng. 12.2] it translates quite literally [as] ἐν καρδίᾳ καὶ ἐν καρδίᾳ’ (cf. LXX Hos 10.2).

56 Seitz, ‘Afterthoughts’; Allison, ‘Eldad and Modad’, 130 n. 75. 1QH 12 iv, 14: ‘They [hypocrites] look for you [God] with a double heart (לבולב).’ 4Q542 1 i, 19: ‘holding on to the truth and walking in uprightness and not with a double heart (לבבולבב Aramaic)’.

57 See Wolverton, W., ‘The Double-Minded Man in Light of Essene Psychology’, AThR 38 (1956) 166–75Google Scholar.

58 Note that the Coptic renders the phrase ϩⲛⲟⲩⲙⲛⲧϩⲏⲧⲥⲛⲁⲩ (Sahidic), with the abstracter ⲙⲛⲧ forming a single noun (hence the indefinite singular article ⲟⲩ) from ϩⲏⲧ (‘heart’) and ⲥⲛⲁⲩ (‘two’). See Crum, W. E., A Coptic Dictionary (Oxford: Oxford University Press, 1939) 714Google Scholar.

59 Wisdom (Jas 1.5); law keeping (1.25); double-mindedness (1.8, 4.8); speech (3.5–12); exultation (4.10; cf. 1.9). The point here is not to put forward an argument for intertextuality between James and Ben Sira. My point is that a similar conceptual background accounts well for the similarities in theme. There are key differences between James and Ben Sira, including their understanding of ‘desire’. See Wold, B., ‘Sin and Evil in the Letter of James in Light of Qumran Discoveries’, NTS 65 (2019) 78–93, at 88CrossRefGoogle Scholar.

60 See Metzner, Jakobus, 67.

61 A. Paretsky, ‘The Two Ways and Dipsuchia in Early Christian Literature: An Interesting Dead End in Moral Discourse’, Ang (1997) 271–88, at 312, notes that ‘prior to its attestation in Sirach δίγλωσσος meant only “bilingual” or “interpreter”’. However, a similar sense is attested in LXX Prov 11.13: ‘A double–tongued man discloses counsels in a meeting, but a person loyal in spirit conceals matters’ (ἀνὴρ δίγλωσσος ἀποκαλύπτει βουλὰς ἐν συνεδρίῳ, πιστὸς δὲ πνοῇ κρύπτει πράγματα).

62 See Allison, ‘Eldad and Modad’, 130 n. 75.

63 Stuckenbruck, L. T., 1 Enoch 91–108 (Berlin/Boston: de Gruyter, 2007) 167–8CrossRefGoogle Scholar.

64 Stuckenbruck, 1 Enoch, 167 nn. 331 and 332. Stuckenbruck, 1 Enoch, 166 also notes, ‘The expression [double heart] does not stem from an understanding of human nature that is concerned with inner moral conflict, as found for example in the Two Spirits Treatise (1QS iii 13–iv 26, between truth and iniquity) and Philo (Gig. 56; Her. 183).’ It is not actually clear whether 1QS is concerned with inner moral conflict, as opposed to external angelic forces (J. L. Kugel, ‘Testament of the Twelve Patriarchs’, Outside the Bible, vol. ii (ed. L. H. Feldman, J. L. Kugel and L. H. Schiffman; Lincoln: University of Nebraska Press, 2013) 1697–1856, at 1810). A better candidate would be the use of διπρόσωπος in the Testament of Asher, which is clearly part of the Two Ways tradition that understands there to be ‘two inclinations’ (δύο διαβούλια) within the person. Cf. T. Ash. 1.5: ‘For there are two ways of good and evil, and with these are the two inclinations in our chests evaluating them’ (ὁδοὶ δύο, καλοῦ καὶ κακοῦ· ἐν οἷς εἰσι τὰ δύο διαβούλια ἐν στέρνοις ἡμῶν διακρίνοντα αὐτάς).

65 Browning, R., Medieval and Modern Greek (London: Hutchinson University Library, 1969) 47Google Scholar.

66 Witmer, S. E., ‘Θεοδίδακτοι in 1 Thessalonians 4.9: A Pauline Neologism’, NTS 52 (2006) 239–50, at 243CrossRefGoogle Scholar. In the following section, I develop an argument in a way similar to Witmer's application of Tov (see n. 72 below).

67 See Bauer, L., Introducing Linguistic Morphology (Edinburgh: Edinburgh University Press, 1988) 12Google Scholar. In etymological research, there is a technical distinction drawn between ‘affixation’ and ‘compounding’. The latter is the joining of two distinct word bases to form a new base, whereas in the former, the affix is not a lexeme in its own right, but modifies a base lexeme to form a new word.  While δίς is a lexeme, in Greek it actually comes to function as an affix, and thus I analyse δίψυχ- as a derivational affixation. In the following section I interact with the work of Emanuel Tov, who does not employ the technical distinction, and thus compounding is discussed in a non-technical sense.

68 Following Thayer, J. H., Thayer's Greek–English Lexicon of the New Testament (Harper & Brothers, 1889) §1374Google Scholar; pace B. M. Newman, A Concise Greek–English Dictionary of the New Testament (rev. edn; Stuttgart: Deutsche Bibelgesellschaft, 2010) s.v. δίψυχος, who divides the word as δύο + ψύχομαι. Note that δίς is usually reduced to δι- in compound forms, though this is not always the case (cf. δισ- in δισμύριοι, δισχίλιοι, δισθανής, δίσαβος, δισάρπαγος, δίσευνος, LSJ s.v. δίς).

69 But this is merely an assumption, and does not have much to substantiate it. The masculine compound πολυκέφαλος (‘many headed’), derived from the feminine base κεφαλή, has no intermediary feminine compound form (*πολυκεφαλη). Within the Apostolic Fathers, the majority attestations of  δίψυχ- are masculine (27; 18 feminine), with all but one feminine form (2 Clem. 19.2) coming from Hermas. See Metzner, Jakobus, 19 and 67, who takes διψυχία and διψυχέω as derivatives of δίψυχος.

70 This would be a case of ‘secondary derivation’. See Weiss, M., ‘Morphology and Word Formation’, A Companion to the Ancient Greek Language (ed. Bakker, E. J.; Chichester/Malden, MA: Wiley-Blackwell, 2010) 104–19, at 109Google Scholar.

71 Durkin, P., Oxford Guide to Etymology (Oxford: Oxford University Press, 2009) 107Google Scholar.

72 Tov, E., ‘Compound Words in the LXX Representing Two or More Hebrew Words’, Bib 58 (1977) 189–212Google Scholar.

73 Tov, ‘Compound Words’, 191.

74 Tov, ‘Compound Words’, 191.

75 We cannot be confident that the term was not commonly used in the oral culture: ‘a word described as a neologism on the basis of our present knowledge may, in fact, be contained in an as yet unpublished papyrus fragment or the word may never have been used in the written language’ (Tov, ‘Compound Words’, 199–200).

76 Translation mine. Greek text from H. Diels, Die Fragmente der Vorsokratiker, griechisch und deutsch, vol. i (Berlin: Weidmann, 1912³) 153.

77 See Cordero, N.-L., By Being, It Is: The Thesis of Parmenides (Las Vegas: Parmenides Publishing, 2004) 130Google Scholar.

78 δίκρανος (δίς, κάρα) is attested elsewhere, though with the meaning ‘pitchfork’ (Lucian, Tim. 12); this reference in Parmenides is the first attestation of the lexeme meaning ‘two-headed’ (cf. LSJ s.v. δίκρανος).

79 Brox, Der Hirt, 552 thinks that διπλοκαρδία presents itself as a more expected Greek form of לב ולב (‘Das etymologisch “natürlichere” Äquivalent zum (jüdischen) geteilten oder zwiespältigen Herzen’).

80 Niederwimmer, Didache, 91 n. 34 notes that the text is uncertain here: Barnabas and Codex Hierosolymitanus read δίγνωμων, Apostolic Constitutions reads δίγνωμος. I have followed Ehrman's text here. δίγνωμων is also found in the earlier Greek scholion on Euripides (633).

81 For δίγλωσσος, cf. LXX Prov 11.13; LXX Sir 5.9, 14–15; 28.13; Sib. Or. 3.36. Cf. T. Benj. 6.5: ‘The good mind does not have two tongues’ (ἡ ἀγαθὴ διάνοια οὐκ ἔχει δύο γλώσσας).

82 καρδία, πνεῦμα and νοῦς are categorised under the same semantic domain in Louw and Nida, 26 (‘psychological faculties’). See also E. A. Nida and J. P. Louw, Lexical Semantics of the Greek New Testament (Atlanta: Scholars, 1992) 81–2.

83 Or in the case of διχόνους and διπλοκαρδία, δίχα- and διπλοῦς- compounding, respectively.

84 M. Clarke, ‘Semantics and Vocabulary’, A Companion to the Ancient Greek Language (ed. E. J. Bakker; Chichester/Malden, MA: Wiley-Blackwell, 2010) 120–33, at 125.

85 Clarke, ‘Semantics’, 125.

86 Clarke, ‘Semantics’, 126: ‘For example, if the basic sense is something like “nourish, rear a child,” how could the word become applicable to salt drying onto the skin?’ Hence one meaning is not necessarily a simple metaphorical extension of another.

87 Clarke, ‘Semantics’, 126.

88 Clarke, ‘Semantics’, 126 (emphasis removed). Clarke explains: ‘The body literally thickens and fattens as we eat … the briny stuff from the sea cakes dry onto the skin, cheese rapidly solidifies when the fig juice is squirted into it; and, remarkably, there is evidence from Aristotle and the Hippocratics that the male's fertilizing act in conception was understood in a way that invited explicit comparison with the use of juice to curdle cheese’ (126–7).

89 See Clarke, ‘Semantics’, 127 (cf. 126) for a helpful representation of the prototype semantics of τρέφω.

90 We may decide that δίκρανος does not fit with the proposed prototype, since in Parmenides’ time, the head is associated not with thinking, but as ‘the receptacle for the principle senses: sight, hearing, smell, taste’ (Cordero, Being, 130 n. 557), though I think a prototype of perception and cognition could be broad enough to keep the example.

91 Durkin, Etymology, 106.

92 See Porter, ‘Dipsuchos’, 494–6; Gilmour, ‘Religious Vacillation’, 41–2.

93 ‘Institutionalised’ refers to the moment the form and meaning of a word is accepted within a linguistic community. ‘Blocking’, also a technical term, ‘refers to the non-existence of a derivative … because of the prior existence of some other lexeme’ (Bauer, Morphology, 66; see also Bauer, L., A Glossary of Morphology (Washington: Georgetown University Press, 2004) 22–3, 56–7Google Scholar).

94 Durkin, Etymology, 104.

95 Of course, this cannot function as evidence for a late date for James. It may still be the case that James predates 1 Clement or a sub-redactional layer of Hermas, but Porter's line of argumentation is insufficient to support this conclusion. Evidence for an early terminus ante quem must be sought elsewhere.