Why Is David and Goliath’s Story 40% Longer in the MT Than in the LXX?

The LXX includes 17:1–11, 32–40, 42–48a, 49, 51–54; 18:6a*, 6b–8 [-ועוד לו אך המלוכה], 12a, 13–16, 20–21a, 22–26a, 27 [-וימלאום], 28–29a. The “pluses” in the MT thus comprise 17:12–31, 41, 48b, 50, 55–58; 18:1–5, 6a*, 8* [ועוד לו אך המלוכה], 9–11, 12b, 17–19, 21b, 26b, 27* [וימלאום], 29b–30. The precise assignment of the verses may vary slightly, depending in part on what the particular scholar assumes the Septuagint translator was doing with some of the Hebrew source text.

The MT describes Goliath’s height as שֵׁשׁ אַמּוֹת וָזָרֶת, “six cubits and a span,” while the LXX has τεσσάρων πήχεων καὶ σπιθαμῆς, “four cubits and a span.”

Unless otherwise indicated, biblical translations follow the NRSVue, with modifications.

The analysis here draws on Emanuel Tov, who has also called this narrative “Version 1,” and named the additional material in the MT “Version 2.” Emanuel Tov, “The Nature of the Differences between MT and the LXX in 1 Sam. 17–18,” in The Story of David and Goliath: Textual and Literary Criticism, ed. D. Barthélemy et al., OBO 73 (Fribourg: Éditions Universitaires; Göttingen: Vandenhoeck & Ruprecht, 1986), 19–46.

See e.g., Henry Preserved Smith, A Critical and Exegetical Commentary on the Books of Samuel, ICC 9 (Edinburgh: T&T Clark, 1904), 150–152; P. Kyle McCarter, I Samuel: A New Translation with Introduction and Commentary, AB 8 (New York: Doubleday, 1980), 306–309; Steven L. McKenzie, King David: A Biography (Oxford: Oxford University Press, 2000), 71; Antony F. Campbell, 1 Samuel, FOTL 7 (Grand Rapids: Eerdmans, 2003), 171, 174–176. At some point before the LXX was translated into Greek, around the 3^rd or early 2^nd century B.C.E., the MT and LXX textual traditions in the David and Goliath narrative parted ways, and the MT tradition received significant subsequent editing. Since the book of Samuel in the LXX otherwise shows much the same shape as the MT, including the Deuteronomistic redactions of the 6^th century and later, the split in the traditions could not have occurred before the end of the Babylonian Exile, or even the early Persian period.

See especially the recent publication by Simeon Chavel and Jessie DeGrado, “Text- and Source-Criticism of 1 Samuel 17–18: A Complete Account,” VT 70 (2020): 553–580. Other scholars suggest a supplementary, rather than a documentary, approach to explaining the differences, arguing that the Masoretic additions were composed by an editor of the book. See, e.g., A. Graeme Auld and Craig Y. S. Ho, “The Making of David and Goliath,” JSOT 56 (1992): 19–39; A. Graeme Auld, “The Story of David and Goliath: A Test-Case for Synchrony plus Diachrony,” in David und Saul im Widerstreit—Diachronie und Synchronie im Wettstreit: Beiträge zur Auslegung des ersten Samuelbuches, ed. Walter Dietrich, OBO 206 (Fribourg: Academic Press; Göttingen: Vandenhoeck & Ruprecht, 2004), 118–128, here 124–125; idem, I & II Samuel, OTL (Louisville: WJKP, 2011), 194, 198–199; Anneli Aejmelaeus, “Rewriting David and Goliath?,” in From Scribal Error to Rewriting : How Ancient Texts Could and Could Not Be Changed, ed. Anneli Aejmelaeus, Drew Longacre, and N. Mirotadze, De Septuaginta Investigationes 12 (Göttingen: Vandenhoeck & Ruprecht, 2020), 165–180.

It is frequently observed that this and other connective verses seem to contradict portions of the two main versions (e.g., that David was already ensconced in Saul’s court as a musician [16:14–23], as is already assumed by his speech to Saul in 17:32, which belongs to the LXX version).

For a fuller accounting, see Chavel and DeGrado, “Text- and Source-Criticism of 1 Samuel 17–18.”

The two friends further ratify and clarify their covenant in 1 Samuel 23:18, and David seems to refer to it in 2 Samuel 9:1.

^{LXX 1 Sam 20:17} καὶ προσέθετο ἔτι Ιωναθαν ὀμόσαι τῷ Δαυιδ, ὅτι ἠγάπησεν ψυχὴν ἀγαπῶντος αὐτόν. The translation follows NETS, with modifications.

The MT and LXX^B (i.e., the manuscript Vaticanus) identify Michal as Adriel’s wife in this passage, but this reading is probably a mistake on the part of MT and a secondary correction on the part of LXX^B, which contains a later Greek recension here. The MT reads:

‏^{שׁמואל ב כא:ח} וַיִּקַּח הַמֶּלֶךְ אֶת שְׁנֵי בְּנֵי רִצְפָּה בַת אַיָּה אֲשֶׁר יָלְדָה לְשָׁאוּל אֶת אַרְמֹנִי וְאֶת מְפִבֹשֶׁת וְאֶת חֲמֵשֶׁת בְּנֵי מִיכַל בַּת שָׁאוּל אֲשֶׁר יָלְדָה לְעַדְרִיאֵל בֶּן בַּרְזִלַּי הַמְּחֹלָתִי.

^{2 Sam 21:8} The king took the two sons of Rizpah daughter of Aiah, whom she bore to Saul, Armoni and Mephibosheth; and the five sons of Michal daughter of Saul, whom she bore to Adriel son of Barzillai the Meholathite.

(Similarly, the LXX^B reads: καὶ τοὺς πέντε υἱοὺς Μιχολ θυγατρὸς Σαουλ, οὓς ἔτεκεν τῷ Εσριηλ…, “and the five sons of Michol daughter of Saoul, whom she bore to Esriel…” [NETS]). Traditions seeking to alleviate the tension by translating “sons of Michal … whom she raised for Adriel” (e.g., KJV) are based on the Targumic attempt to reconcile the discrepancy, and should be ignored.

See, e.g., Simon J. De Vries, “David’s Victory over the Philistine as Saga and as Legend,” JBL 92 (1973): 23–36; Heda Jason, “The Story of David and Goliath: A Folk Epic?” Bib 60 (1979): 36–70, esp. 60–61, 67; David W. Gooding, “An Approach to the Literary and Textual Problems in the David-Goliath Story: 1 Sam 16–18,” in The Story of David and Goliath: Textual and Literary Criticism, Papers of a Joint Venture, ed. Dominique Barthélemy et al., OBO 73 (Fribourg: University Press; Göttingen: Vandenhoeck & Ruprecht, 1986), 55–86, here 75–82; Alexander Rofé, “The Battle of David and Goliath: Folklore, Theology, Eschatology,” in Judaic Perspectives on Ancient Israel, ed. Jacob Neusner, Baruch A. Levine, and Ernest S. Frerichs (Philadelphia: Fortress, 1987), 117–151, esp. 119–123; Arie van der Kooij, “The Story of David and Goliath: The Early History of Its Text,” ETL 68 (1992): 118–131, here 129–130.

See, e.g., Hans Wilhelm Hertzberg, I & II Samuel, trans. J. S. Bowden, OTL (Philadelphia: Westminster, 1964), 147–148; Stephen Pisano, Additions or Omissions in the Books of Samuel: The Significant Pluses and Minuses in the Massoretic, LXX and Qumran Texts, OBO 57 (Freiburg: University Press, 1984), 86; Baruch Halpern, David’s Secret Demons: Messiah, Murderer, Traitor, King (Grand Rapids: Eerdmans, 2001), 7 n. 3 (but cf. 280–281); and Rebecca Poe Hays, “A Problematic Spouse: A Text-Critical Examination of Merab’s Place in 1 Samuel 18:17–19 and 2 Samuel 21:8,” ZAW 120 (2017): 220–233.

See in particular Jason, “Story of David and Goliath”; Gooding, “Approach to the Literary and Textual Problems”; and Rofé, “Battle of David and Goliath.”

See the discussion of this issue in, e.g., Julius Wellhausen, Der Text der Bücher Samuelis (Göttingen: Vandenhoeck & Ruprecht, 1871), 104–105; Smith, Books of Samuel, 150; McCarter, I Samuel, 307; Johan Lust “The Story of David and Goliath in Hebrew and Greek,” in Barthélemy et al., Story of David and Goliath, 5–18, here 9; Emanuel Tov, “The Composition of 1 Samuel 16–18 in Light of the Septuagint,” in idem, The Greek and Hebrew Bible: Collected Essays on the Septuagint, VTSup 72 (Leiden: Brill, 1999), 333–362.

Although alternative explanations can be given for the Jubilees manuscript in the first example, the identification of the patch-job in the Temple Scroll in the second example seems relatively secure.

James C. VanderKam and J. T. Milik, “The First Jubilees Manuscript from Qumran Cave 4: A Preliminary Publication,” JBL 110 (1991): 243–270, here 244 (PAM 43.185); James C. VanderKam and J. T. Milik, “216. 4QJubilees^a,” in Qumran Cave 4. VIII: Parabiblical Texts, Part 1, ed. Harold Attridge et al., DJD 13 (Oxford: Clarendon, 1994), 1–22. Other fragments of the right sheet preserve portions of columns I and II (parts of Jubliees’s prologue as far as Jub. 1:28). Other fragments of the left sheet show that the sheet contained at least two more columns (VI–VII), encompassing parts of Jub. 2:1–24. For a summary, see esp. Eibert J. C. Tigchelaar, “The Qumran Jubilees Manuscripts as Evidence for the Literary Growth of the Book,” RevQ 26 (2014): 579–594, here 584.

Molly M. Zahn, Genres of Rewriting in Second Temple Judaism: Scribal Composition and Transmission (Cambridge: Cambridge University Press, 2020), 105. Some have argued that this manuscript offers evidence for the composition history of Jubilees. See, e.g., Tigchelaar, “Qumran Jubilees Manuscripts,” 584; Charlotte Hempel, “The Place of the Book of Jubilees at Qumran and Beyond,” in The Dead Sea Scrolls in Their Historical Context, ed. Timothy H. Lim (Edinburgh: T&T Clark, 2000), 187–196, here 190; and Monger, “4Q216 and the State of Jubilees,” 604. It was the assessment of the original editors, however, that this manuscript simply reflects a repair job (VanderKam and Milik, “First Jubilees Manuscript,” 246; idem, “216. 4Qjubilees^a,” 1).

Furthermore, the last few lines of the text on the right-side sheet overlap with the first few lines of column VI on the older, left-side sheet (Hempel, “Place of the Book of Jubilees,” 189; see subsequently Monger, “4Q216 and the State of Jubilees,” 604). It is somewhat difficult to explain why a scribe repairing the manuscript would have included material that was already at the top of the next column. Yigael Yadin, who first published the manuscript, suggested that the scribe “preferred to finish the whole sheet.” See Yigael Yadin, The Temple Scroll, 3 vols. (Jerusalem: Israel Exploration Society, 1983), 1:11–12, here 12; see also ibid., 2:17.

A recent study by Jong-Hoon Kim argues that the Greek text of 1–4 Reigns (= the books of Samuel and Kings) originally circulated as a set of “Short Scrolls (Kleinrollen).” Jong-Hoon Kim, “Vom hellenistischen Kleinrollensystem zum Kodex: Beobachtungen zur Textgestalt der griechishen Samuel- und Königebücher,” in XIV Congress of the IOSCS, Helsinki, 2010, ed. Melvin K. H. Peters, Septuagint and Cognate Studies 59 (Atlanta: Society of Biblical Literature, 2013), 231–242, esp. 238–240. This conclusion is very similar to earlier arguments by Menahem Haran and Alexander Rofé in their studies of Hebrew scrolls. See Menahem Haran, “Book-Scrolls at the Beginning of the Second Temple Period: The Transition from Papyrus to Skins,” HUCA 54 (1983): 111–122; idem, “Book-Size and the Thematic Cycles in the Pentateuch,” in Die Hebräische Bibel und ihre zweifache Nachgeschichte: Festschrift für Rolf Rendtorff zum 65. Geburtstag, ed. Erhard Blum, Christian Macholz, and Ekkehard W. Stegemann (Neukirchen-Vluyn: Neukirchener Verlag, 1990), 165–176; Alexander Rofé, “The Composition of Deuteronomy 31” (Eng. Title) = “The Question of the Composition of the Portion ‘וילך’ (Deut 31) in Light of the Hypothesis Regarding the Exchange of Columns in the Biblical Version,” Shnaton 3 (1978): 59–76. See also David M. Carr, “Rethinking the Materiality of Biblical Texts: From Source, Tradition and Redaction to a Scroll Approach,” ZAW 132 (2020): 594–621, and compare the subsequent study by Asaf Gayer and David Carr, “Use of Digital Imaging Software to Determine Shifts in Script Density Distribution in Egyptian and Levantine Scrolls,” unpublished paper delivered at the Annual Meeting of the SBL, Denver (November 21, 2022).

This solution differs somewhat from the proposal by Kim (“Vom hellenistischen Kleinrollensystem zum Kodex,” 231–242, esp. 238–240), who suggests that the first scroll in the Greek text of Samuel ended with the defeat of Goliath (after ca. 1 Sam / 1 Reigns 17:54). The second continued with David’s entry into Saul’s court, before 18:6b. At some point, these edges were damaged in the parent-text of LXX and supplemented with a variant version. Kim’s hypothesis is very perceptive, but it is based on a literary judgment: because of Kim’s conviction that each scroll has to contain a coherent story (with beginning, middle, and end), he is led to argue that the first scroll ended with the killing of Goliath. The problem, however, is that the differences between the texts of the MT and LXX do not end with the death of Goliath in chapter 17; they persist throughout chapter 18 as well. It is only at 19:1 that MT and LXX begin to read similarly again.

Versions 1 and 2 of the David and Goliath narrative each originally circulated as part of longer traditions about the history of David’s rise to kingship. These two separate traditions, HDR₁ and HDR₂, were later incorporated into the book of Samuel (ca. 1 Sam 16–2 Sam 6, 7, or 8). In repairing the LXX scroll, the scribe drew on the text of Version 1 from a manuscript of HDR₁, a tradition that had been preserved alongside the developing book of Samuel for a long time. See Jeremy M. Hutton, “Split Identities and Fragmented Texts: Considerations on Text, Source, and Materiality in 1Sam 17–18 MT and OG,” in Proceedings of the Samuel Colloquium, IOSOT 2022 (Zürich), ed. Johannes Klein and Walter Dietrich, BWANT (Stuttgart: Kohlhammer, forthcoming); and Jeremy M. Hutton, The Transjordanian Palimpsest: The Overwritten Texts of Personal Exile and Transformation in the Deuteronomistic History, BZAW 396 (Berlin: de Gruyter, 2009), 239–288, esp. 241–242.

Again, see Hutton, “Split Identities and Fragmented Texts.”

The Corruption-Replacement Hypothesis provides an adequate, if admittedly imperfect, way of accounting for the problems with the Addition and Omission Hypotheses. I discuss one of the problems with the theory in the Conclusion section of my article, “Split Identities and Fragmented Texts.”

Additional examples of similar texts with a common background that circulated simultaneously include: (1) the LXX incorporates both Esdras A′ and Esdras B′ (= 1 Esdras and Ezra-Nehemiah) which have substantial textual overlap; (2) the MT and Samaritan Pentateuchs; and (3) many similar-but-not-identical documents identified by studies in “Rewritten Scripture.” (See esp. Molly Zahn, Genres of Rewriting in Second Temple Judaism [Cambridge: Cambridge University Press, 2020].)