Scott Lawson wrote:Perhaps someone can cut and paste the relevant section:
BDF §§ 300, 2 introducing a direct question pages 157-58.
These are some of the texts cited but the discussion is involved:
1 Chronicles 17:6, Job 27:12, 2 Samuel 12:9, Jeremiah 22:28; 33 (26):9, 2 Kings 8:14
Here's the section under consideration:
BDF §300 (2) wrote:(2) Ὅ τι introducing a direct question is especially Markan: 2:16 (διὰ τί SDW, τί Θ, τί ὅτι AC al.); 9:11 (τί οὖν WΘ lat); 9:28 (διὰ τί D 33 al.); to these should be added Mk 2:7 (with BΘ, WH margin; cf. Black, Aramaic Approach2 47, 88) and possibly 8:12 (with C Or; C. H. Turner, JTS 27  58; Taylor, The Gospel of St Mark (1953) 61, 362, Black, Aramaic Approach2 89 and Katz take it as exclamatory מָה here). It is also possible in A 11:3 ‘Why did you go …’ (RSV; s. Beginnings III 102f., IV 124; Moule 159; Haenchen12 299 takes it as ὅτι recitativum). In Jn 8:25 ὅ τι may be taken in the sense of מָה ‘that I speak with you at all!’ (τὴν ἀρχήν = ὅλως), or interrogatively ‘why do I speak with you at all?’ Blass interprets acc. to class. usage (not attested in the NT): ‘(You ask,) why (an indefinite relative is commonly used when a question is repeated by the respondent before his reply; scil. ἐρωτᾷς [Smyth §2670]) do I speak to you at all?’ (but they have not asked this question). Or ‘(You reproach me,) that (ὅτι) I speak with you at all?’ Cf. the direct question in Homil Clem 6.11.4 τί καὶ τὴν ἀρχὴν διαλέγομαι; 19.6.6 ἐπεὶ τί καὶ τὴν ἀρχὴν ζητεῖ; R. W. Funk, HTR 51 (1958) 95–100 and E. R. Smothers, HTR 51 (1958) 111–22 independently adopt the reading of p (from the margin): ειπον υμιν την αρχην … ‘I told you at the beginning’, which gives good grammar and sense; rejected by C. K. Barrett, ET 66 (1957) 176. Cf. Bultmann (Das Evangelium des Johannes [Meyer Kom.] 1941) 267f. for full discussion. Barn 7.9 has the sequence τί …; καὶ ὅ τι …; In 8.5 ὅ τι δέ is parallel to διὰ τί δέ in 8.4 and 8.6, but in the latter SH read καὶ ὅ τι as do all MSS in 7.9 (gap in H), the διότι of the acephali being corrupt as everywhere in the LXX (Ziegler, Beiträge zur Jeremias-Septuaginta 15). All known examples in the LXX are in direct questions. Representing לָמָּה: 2 Km 7:7 ὅ τι B, τί ὅτι L(ucian), τί O(rigen), with the par. 1 Chr 17:6; Job 27:12 (Katz, ThLZ 1957, 114 n. 4). מַדּוּעַ: 2 Km 12:9 ὅ τι BO-A, τί A, τί ὅτι pc.; Jer 22:28; 33(26): 9. מָה: 4 Km 8:14; Jer 2:36; in 30 (49): 4 ὅ τι is a variant of τί, just as in 22:23 Lucian adds ὅ τι and Origen τί (Katz in Ziegler, op. cit. 15). In Gen 18:13 A alone has ὅ τι, the remainder τί ὅτι for לָמָּה זֶּה (= τί τοῦτο in 25:22); this τί ὅτι is frequently used to render מָה … כִּי (§299(4)) and is not therefore to be read as τί ὅ τι. W.–S. §24, 18a observes that to explain ὅτι as an abbreviation of τί ὅτι is odd because it leaves out what is essential; moreover, in τί ὅτι the second word is ὅτι, not ὅ τι. מַדּוּעַ in Judg 5:28 twice introduces questions which are answered; A correctly has διὰ τί in both cases, while διότι B is corrupt, for it is not to be interpreted as διʼ ὅ τι since this διότι is always a poor variant in the LXX. Interrogative ὅ τι was postulated for Jn 8:25 by Lachmann in his edition of 1832 I, praef. xliii, then by Buttmann 218 (for the LXX also); it is accepted by Mlt. 94; Rob. 729f.; C. H. Turner, JTS 27 (1926) 58–62 (who finds in it a characteristic of Markan usage); Katz, ThLZ 1957, 114; 1958, 318. As the parallels in the LXX and early Christian literature indicate, ὅ τι with direct questions is a piece of ‘biblical Greek’.—Controversial Mt 26:50 ἑταῖρε ἐφʼ ὃ (inferior v.l. ἐφʼ ᾧ) πάρει: hardly a direct question ‘For what?’; the easiest solution is to take it as a painful, ironic reminiscence of a toast like the one attested on a goblet from Syria: εὐφραίνου ἐφʼ ὃ (ᾧ) πάρει ‘Enjoy yourself! for that’s why you are here’ (on the aposiopesis cf. Iambl., VP 145 ὅσα βούλει, παρὰ τῶν θεῶν [scil. γένοιτό σοι], cf. ibid. βουλοίμην μᾶλλον, ὅσʼ ἄν μοι παρὰ τῶν θεῶν γένηται). Biblio.: Debrunner, Jahresb. Altertumsw. 236 (1932) 220; 261 (1938) 189; Deissmann, LO 104 [LAE 125–31]; Schwartz, ByzZ 25 (1925) 154f.; Crönert, Gnomon 4 (1928) 90 n. 3 (who sees in it an expression of eastern colloquial speech without sufficient reason); Klostermann, ZNW 29 (1930) 311 (he compares Acta Carpi 44 ἐγὼ δὲ ἐφʼ ὃ πάρειμι scil. ποιήσω); Sedgwick, ClR 46 (1932) 12 (he also compares ὅσον ζῇς φαίνου [error for εὐφραίνου?] on an epitaph [Musici scriptores ed. Jan p. 452]). Abbott §2231e is against interrogative ἐφʼ ὅ; but ἀνθʼ ὅτου and the like appear in direct questions in the Church Fathers (Jannaris §2038; Usener, Der heilige Tychon 50). Ὧν ἕνεκα Eus., Praep. Ev. 6.7 p. 257 D (I 316.10 Mras) is probably not interrogative, but ‘why did I come to speak about it—the reason is that …’ (Abbott, loc. cit.). Also cf. Ljungvik 4; Deissmann, LO4 101 n. 4 [LAE 126 n. 4]; Bauer s.v. ὅς 2a, 2b β, and 9b with references cited there; Zerwick, Graec. bibl. 51.
Blass, F., Debrunner, A., & Funk, R. W. (1961). A Greek grammar of the New Testament and other early Christian literature (157–158). Chicago: University of Chicago Press.