¹ Farrer, A. M., ‘On Dispensing with Q’, Studies in the Gospels: Essays in Memory of R. H. Lightfoot (ed. Nineham, D. E. (Oxford: Blackwell, 1955) 55–88Google Scholar. Farrer's most formidable defenders include Goulder, M. D. (Luke: A New Paradigm (2 vols.; Sheffield: Sheffield Academic, 1989))Google Scholar; Goodacre, M. (The Case Against Q: Studies in Markan Priority and the Synoptic Problem (Harrisburg: Trinity Press International, 2002))Google Scholar; and Watson, F. (Gospel Writing: A Canonical Perspective (Grand Rapids: Eerdmans, 2013))Google Scholar.

² Kümmel, W. G., Introduction to the New Testament (trans. Kee, H. C.; rev. edn; Nashville, TN: Abingdon, 1991) 64Google Scholar.

³ On the biographical nature of the gospels, see Talbert, C. H., What Is a Gospel? The Genre of the Canonical Gospels (Philadelphia: Fortress, 1977)Google Scholar; Burridge, R., What Are the Gospels? A Comparison with Graeco-Roman Biography (25th anniversary edn; Waco, TX: Baylor University Press, 2018)Google Scholar; and Keener, C. S., Christobiography: Memory, History, and the Reliability of the Gospels (Grand Rapids: Eerdmans, 2019)Google Scholar. For a dissenting voice, see A. Yarbro Collins, ‘Genre and the Gospels’, The Journal of Religion 75 (1995) 239–46.

⁴ See Cadbury, H. J., The Making of Luke-Acts (London: SPCK, 1958) 4Google Scholar; H. Koester, Introduction to the New Testament, vol. i: History, Culture, and Religion of the Hellenistic Age (Philadelphia: Fortress, 1982) 108. While Luke's style and vocabulary are superior to that of other Gospels, this is not to say that his rhetoric reaches the heights of true Greek classics. For differences, see Wifstrand, A., Epochs and Styles: Selected Writings on the New Testament, Greek Language and Greek Culture in the Post-Classical Era (ed. Rydbeck, L. and Porter, S. E.; trans. Searby, D.; Tübingen: Mohr Siebeck, 2005) 35Google Scholar.

⁵ The way in which interruptions function in Luke (and the other Gospels) may correspond to certain theological aims. On this, see Smith, D. L., The Rhetoric of Interruption: Speech-Making, Turn-Taking, and Rule-Breaking in Luke-Acts and Ancient Greek Narrative (Berlin: de Gruyter, 2012), esp. 186–210CrossRefGoogle Scholar.

⁶ All NT word counts are based on the Nestle Aland 28th edition.

⁷ The word count would total 2,137 if one were to include the preceding section, Matt 23.2–39. To do this, however, one would have to overlook a brief aside involving the disciples in Matt 24.1–3. So, in accordance with the criteria of counting a subject change as an interruption, I proceed with 1,494 as the relevant figure.

⁸ E.g. Luke 1.46–55; 14.7–24; 16.19–31; 18.9–14.

⁹ Though not completely: Mark 13.15–19 and Matt 24.17–21 are in near-complete agreement, Luke omits the entire section except for 21.23.

¹⁰ The parables in Mark 4.3–32, for example, consist of roughly 500 words, and Luke (8.5–16) reduces his corresponding section by nearly 40 per cent.

¹¹ Here I include one minor change in subject at Luke 17.37.

¹² It is possible, of course, to alter the criteria for counting words. For example, one might examine passages more broadly and overlook changes in subjects. In that case, Luke's middle discourse, which is interrupted by a reaction from the Pharisees, would add up to roughly 1,100 words. However, we would then need to modify Matthew's second speech plausibly to include the prior section about Jesus’ denunciation of the Scribes and Pharisees and his Lament over Jerusalem (Matt 23.2–39). In that case, the length of Matthew's second passage would add up to over 2,200 words. In order to avoid the need to adjudicate what is considered a narrative break, I leave the criterion as described.

¹³ B. H. Streeter, The Four Gospels: A Study of Origins, Treating of the Manuscript Tradition, Sources, Authorship, & Dates (London: MacMillan and Co., 1924) 183. In many ways, Streeter should be credited for setting the agenda for English scholarship on the Synoptic problem.

¹⁴ H. J. Holtzmann, Die synoptischen Evangelien: Ihr Ursprung und geschichtlicher Charackter (Leipzig: Engelmann, 1863) 130. I translate ‘muthwillig’ as ‘wantonly’. Interestingly, Holtzmann thinks it more likely that Matthew has built his ‘walls’ out of Luke's ‘pile of stones’: ‘Was ist an sich wahrscheinlicher: dass Lucas die grossen Bauten muthwillig zerschlagen und die Trümmer nach allen vier Winden auseinandergesprengt, oder dass Matthäus jene Mauern aus den Steinhaufen des Lucas erbaut habe?’

¹⁵ G. Stanton, ‘Matthew, Gospel of’, A Dictionary of Biblical Interpretation (ed. R. J. Coggins and J. L. Houlden; London: Press, 1990) 432–6, at 434. For similar statements, see Kümmel, Introduction, 64; R. H. Gundry, Matthew: A Commentary on his Handbook for a Mixed Church under Persecution (Grand Rapids: Eerdmans, 1995²) xvi; J. S. Kloppenborg, Excavating Q: The History and Setting of the Sayings Gospel (Minneapolis: Fortress, 2000) 41.

¹⁶ Goulder, Luke, 40–1.

¹⁷ Goulder, Luke, 347.

¹⁸ Part of the intuitive appeal of Matthew's version is due undoubtedly to the immense reverence towards the Sermon of the Mount throughout Christian history. Luke Timothy Johnson notes: ‘In the history of Christian thought – indeed in the history of those observing Christianity – the Sermon on the Mount has been considered an epitome of the teaching of Jesus and therefore, for many, the essence of Christianity’ (L. T. Johnson, ‘The Sermon on the Mount’, The Oxford Companion to Christian Thought (ed. A. Hastings; Oxford: Oxford University Press, 2000) 654).

¹⁹ Farrer declares: ‘All we have to show is that St. Luke's plan was capable of attracting St. Luke. You do not like what I have done to the garden my predecessor left me. You are welcome to your opinion, but I did what I did because I thought I should prefer the new arrangement. And if you want to enjoy whatever special merit my gardening has, you must forget my predecessor's idea and try to appreciate mine’ (Farrer, ‘Dispensing’, 65).

²⁰ This point applies especially to the Sermon on the Mount's second half, which Stanton labels a ‘rag-bag’ of sayings and Fitzmyer refers to as a ‘series of loosely related sayings’ (G. Stanton, A Gospel for a New People: Studies in Matthew (Louisville, KY: Westminster John Knox, 1993) 298; J. A. Fitzmyer, The Gospel According to Luke i–ix (ABC 28; New York: Doubleday, 1981); see also Goodacre, Case Against Q, 99–100.

²¹ For details see Watson, Gospel Writing, 156–216.

²² Mark Matson also notes that Luke's ordering has a ‘distinctive theology’ (M. Matson, ‘Luke's Rewriting of the Sermon on the Mount’, Questioning Q (ed. M. Goodacre and N. Perrin; London: SPCK, 2004) 43–70, at 64; Goulder, Luke, 40.

²³ M. Goodacre, ‘Re-Walking the “Way of the Lord”: Luke's Use of Mark and his Reaction to Matthew’, Luke's Literary Creativity (ed. M. Müller and J. T. Nielsen; Library of New Testament Studies 550; New York: Bloomsbury, 2016) 26–43.

²⁴ What I have described, of course, does not exhaust FH theorists’ responses. For example, Goodacre and Matson highlight the fact that Luke often omits material from the Sermon on the Mount that would be less relevant or unappealing to Luke's gentile audience (Goodacre, Case against Q, 95–6; Matson, ‘Luke's Rewriting’, 49).

²⁵ See n. 3 for references. Charles Talbert is on record for claiming that Burridge's volume ‘ought to end any legitimate denials of the canonical Gospels’ biographical character’ (C. Talbert, ‘Review of Richard A. Burridge, What Are the Gospels?’, JBL 112 (1993) 714–15, at 715.

²⁶ M. R. Licona, Why Are There Differences in the Gospels? What We Can Learn from Ancient Biography (New York: Oxford University Press, 2016); Keener, Christobiography.

²⁷ Some of these family resemblances include introductory material, an ancestry/genealogy, a birth narrative, tales of early childhood that anticipate the protagonist's later character qualities, sayings and a death narrative. See Burridge, Gospels, 105–23.

²⁸ See n. 4. Of course, Luke's elevated literary status in no way detracts from its Jewish content and concerns.

²⁹ Alex Damm, for instance, has written an entire monograph showing how ancient rhetorical features can shed light on the Synoptic Problem (A. Damm, Ancient Rhetoric and the Synoptic Problem: Clarifying Markan Priority (Leuven; Walpole, MA: Peeters, 2013). Damm's valuable contribution, however, does not focus extensively on the narrower subject of this paper. For this reason, I engage in more detail with Heather Gorman's work below.

³⁰ See C. S. Keener, The Gospel of Matthew: A Socio-Rhetorical Commentary (Grand Rapids: Eerdmans, 2009) 16–24.

³¹ One might respond that Luke includes longer speeches in Acts. I address this objection later.

³² H. M. Gorman, ‘Crank of Creative Genius? How Ancient Rhetoric Makes Sense of Luke's Order’, Marcan Priority Without Q: Explorations in the Farrer Hypothesis (ed. J. C. Poirier and J. Peterson; New York: Bloomsbury, 2015) 62–81.

³³ Theon, Prog. 79; Aphthonius, Prog. 22; Nicolaus, Prog. 14. Quintilian (Inst. 4.2.31) correspondingly prescribes that court speeches should be lucid (lucidus), brief (brevis) and plausible (verisimilis). Citations found in Gorman, ‘Crank’, 71.

³⁴ Theon, Prog. 80. Translation by George Alexander Kennedy in his Progymnasmata: Greek Textbooks of Prose Composition and Rhetoric (Society of Biblical Literature, 2003) 30. For a discussion on Luke's implementation of rhetorical principles found in Quintilian, see Robert Morgenthaler, Lukas und Quintilian: Rhetorik als Erzählkunst (Zürich: Gotthelf, 1993), esp. 191–363.

³⁵ Theon, Prog. 83 (translation from Kennedy, Progymnasmata, 32). See Inst. 4.2.40 for similar remarks by Quintilian.

³⁶ Gorman, ‘Crank’, 67–73. Here Gorman is principally following M. W. Martin, ‘Progymnastic Topic Lists: A Compositional Template for Luke and Other Bioi?’, NTS 54 (2008) 18–41.

³⁷ Burridge, for example, notes briefly that Luke often places Matthew's speeches in narrative contexts in which Jesus answers his interlocutors’ questions. This, Burridge suggests, ‘may be evidence … of Luke's attempt to conform his gospel more closely to Bioi’ (Burridge, Gospels, 191).

³⁸ Among the works Burridge examines, the only one I leave out is Satyrus’ Euripides since it is too fragmentary for the present analysis. While Burridge includes only one of Plutarch's Lives (Cato Minor), I have added two more (Antony, Pompey). Still more could be added, but the same tendency is found throughout Plutarch's other biographies.

³⁹ Most word counts for the book lengths can be found electronically through the Thesaurus linguae Graecae and Perseus Catalogue. In other cases, I have counted the words using the most current editions in the Loeb series. Word counts from speeches are my own.

⁴⁰ Initially, Lucian's Passing of Peregrinus may seem like a counter-example since it contains a speech (by an unnamed speaker) that is roughly half the length of the work. However, the speech itself functions as part of the narrative which drives the account forward. Part of the reason why the work is structured in this way is that Lucian seems to depart purposefully from certain biographical conventions. On this point, see J. König, ‘The Cynic and Christian Lives of Lucian's Peregrinus’, The Limits of Ancient Biography (ed. B. McGing and J. Mossman; Swansea: Classical Press of Wales, 2006) 227–53. Consequently, my analysis includes only Lucian's Alexander and his Demonax. The latter in particular, Keener notes, can be considered a bios with much more confidence (Keener, Christobiography, 92).

⁴¹ The original 1984 album contained lengthier songs which were later shortened and rearranged for the 1986 stage version of the musical. Thanks to Mark Goodacre for pointing me to this.

⁴² See above for references.

⁴³ Burridge, Gospels, 156, 165.

⁴⁴ E. L. Bowie, ‘Apollonius of Tyana: Tradition and Reality’, Teilband Religion (Heidentum: Römische Religion, Allgemeines) (ed. W. Haase; ANRW ii.16.2; Berlin: de Gruyter, 1978) 1652–85, at 1652.

⁴⁵ Burridge, Gospels, 164–5.

⁴⁶ Burridge, Gospels, 165. Compare with Plato's Republic at 90,000 words and Xenophon's Cyropaedia at 80,000 words. In this vein, the Philostratus specialist Graham Anderson notes, ‘It is futile in the end to try to “explain” Apollonius in terms of any single genre.’ Anderson nevertheless seems somewhat more accepting of the label ‘sophistic biography’. G. Anderson, Philostratus: Biography and Belles Lettres in the Third Century ad (new edn; Routledge Revivals; New York: Routledge, 2014) 235.

⁴⁷ See P. V. I. Varneda, The Historical Method of Flavius Josephus (Arbeiten zur Literatur und Geschichte des Hellenistischen Judentums xix; Leiden: Brill, 1986) 92.

⁴⁸ Josephus, J.W. 2.345–401; 5.376–420; 7.341–88. The last of these is in fact the second of a two-part speech. If we include the first part, the total word count becomes 1,676.

⁴⁹ Josephus, Ant. 2.140–59; 4.177–93; 15.127–46; 19.167–84.

⁵⁰ So Goulder asserts that Luke ‘tends to favor short speeches’ (Goulder, Luke, 39). ‘He does not like long units’ (Luke, 346).

⁵¹ Tuckett, C. M., Q and the History of Early Christianity: Studies on Q (Peabody, MA: Hendrickson, 1996) 27Google Scholar.

⁵² Tuckett, Q, 27. We might note that Peter's interjection in verse 41 constitutes a change in subject, but we will ignore this.

⁵³ For a different response to Tuckett's argument, see Goodacre, Case Against Q, 94–6.

⁵⁴ For extensive argumentation, see C. S. Keener, Acts: An Exegetical Commentary. Introduction and 1:1–2:47 (Grand Rapids: Baker Academic, 2012) 51–89; Keener, Christobiography, 222.

⁵⁵ Keener, Acts, 57; C. W. Fornara, The Nature of History in Ancient Greece and Rome (London: University of California Press, 1988) 34–6.

⁵⁶ Thanks to Mark Goodacre from drawing my attention to this point.

⁵⁷ Acts 22.3–21; 26.2–23; 13.16–41; 2.14–26; 7.2–53, respectively. It is also odd that Tuckett appeals to the story of Cornelius’ conversion since its longest speech is relatively short at 181 words.

⁵⁸ Compare them with Luke 21.8–36 (442 words); 17.20–8.14 (486 words); 6.20–49 (569 words); and 15.4–16.13 (778 words).

⁵⁹ Special thanks to Mark Goodacre, Caleb Friedeman, Mark Jeong, Vince Archer and an anonymous reviewer for valuable feedback on this article. I am grateful, in addition, to members of the Duke NT graduate student colloquium who supplied many helpful comments.