The Evidence from Qumran - The Story of David and Goliath (1 Sam 17

3. The Story of David and Goliath (1 Sam 17–19)

3.2. The Evidence from Qumran

Unfortunately, only small portions of these chapters are preserved in the Qumran manuscripts. To investigate which version the Qumran scrolls support, I will analyse the Qumran variants in 1 Sam 17–18.

3.2.1 The Manuscript 4QSam^a

Of the fragments of 4QSam^a,, 5 include 1 Sam 17–18: Frags. 12, 13, 14, 16 and 17. Fragment 12 is the largest of these, containing parts of 1 Sam 17:3–6. In this fragment, five line-endings and the left margin of the column are visible (see PAM 43.111).

Cross’s reconstruction of the text is as follows:

]

֯ל[א םיד] ֯מ[וע םייתשׁלפו

17:3

]

֯ם[היניב איגהו הזמ רהה לא םידמע לארשיו הזמ רהה]

ׄשיא אצ ׄי[ו

תרזו תומ[א ]עבר ֯א[ והבג תגמ ומש תילג םייתשלפ תוכרעממ םינבה]

]

ןירשה לקשמ[ו שובל אוה םישקשק ןירשו ושאר לע תשחנ עבוכו

תשחנ םילקש םיפלא תשמח]

וי ֯פ ֯ת ֯כ ֯ן ֯י ֯ב [ ןדיכו וילגר לע תשחנ תחצמו

Only traces of the letters of the first and last line of the fragment are visible, but the medial three lines are preserved well enough to determine that the fragment contained the text of 1 Sam 17:3–6. With this reconstruction, the line-lengths for lines 2–5 are 54, 54, 52 and 57 characters, respectively.³⁷⁴Compared with the Masoretic text, the only quantitative difference in the reconstructed portion is the omission of

תשחנ

after

ןדיכו

, making the line-length closer to that of the previous three and thus fulfilling the vertical alignment.

373. The longer text is preferred, e.g., in Barthélemy 1986; Gooding 1986; Gordon 1986; Rofé 1987; Van der Kooij 1992; Dietrich 1996; Tsumura 2007. The shorter text is preferred, e.g., in McCarter 1980; Klein 1983; Lust 1986; Tov 1986; Trebolle 1990; Nelson 1991; Auld & Ho 1992; Tov 1999; Campbell 2003.

374. The line-lengths in 4QSam^a varies widely. The shortest average line-length is 45.4 characters (Column II), while the longest is 69.3 characters (Frags. 29–33).

Variants

17:4

תוֹמּ ַא שׁ ֵשׁ

}

]תומ[א ]עבר ֯א[ Q^a G (τεσσάρων πήχεων).

In the Masoretic text, the height of Goliath is six cubits and a span, while the Septuagint and the Qumran scroll give the number as ‘four’ instead of ‘six’. Two opposing explanations for these readings have been posited: either the editor of the text intentionally increased Goliath’s height or lowered it. The reason for increasing the height would be to make David’s challenge seem harder, thus increasing David’s glory after the victory.³⁷⁵The opposite arguments claim that the height of the Goliath was intentionally lowered to make it more moderate, increasing the verisimilitude of the story.³⁷⁶ That is, lowering Goliath’s height would make him an extremely big man—not an impossible figure. Thus, the rationalization of the height of Goliath would be a critique of Saul, ‘the tall Israelite’ (1 Sam 9:2) who did not have courage enough to enter the duel with an equal fighter.³⁷⁷ This argument is weak, however, since Saul’s situation in the story is already extremely poor already; lowering or increasing Goliath’s height does not change the situation. In addition, it is hard to believe that someone has intentionally decreased the height of Goliath to make the story more believable. Rather, it is more likely to decrease the verisimilitude of the story and to increase David’s glorious victory. At the same time, this increases the power and ability of God, since David wins only with the help of God. Moreover, the Hebrew Bible brims with stories with little verisimilitude, from which no similar such desire to increase the verisimilitude of these stories has been observed.

One further possibility must be taken into account. Namely, the reading

תוֹמּ ַא שׁ ֵשׁ

, ‘six cubits’, in the Masoretic text have emerged from an unintentional error. Three verses later, it is said that Goliath’s spear’s head weighted six hundred shekels. Graphically, ‘six cubits’ and

‘six hundred’ are extremely similar in the unvocalized text,

תומא שש

and

תואמ שש

. Although there is a considerable amount of text between these two expressions, the confusion might have happened in a manuscript with line-length such that the expressions

תומא שש

and

תואמ שש

happened to be vertically aligned.³⁷⁸ Against such a background, an unintentional error seems very plausible.³⁷⁹All in all, the evidence suggests that the reading of the Masoretic text is secondary, whether intentional or not.

17:5

ןוֹי ְר ִשּׁ ַה ל ַק ְשׁ ִמוּ

M Q^a (

ןירשה לקשמ

]

ו

)

} ונירש לקשמו

(?) G (καὶ ὁ σταθμὸς τοῦ θώρακος αὐτοῦ).

375. Hays 2005, 707; Conybeare & Stock 1980, 252; Cross 1979, 54n2. Cf. McCarter 1980, 286; DJD XVII, 376. Gehman 1950, 295; Stoebe 1973, 317; Tsumura 2007, 441; cf. Klein 1983, 175; Smith 1977, 155.79.

377. Johnson 2012, 540.

378. In the Masoretic text, the number of characters between the expressions is 163. With a line-length of c. 80 characters, the expressions would be vertically aligned with only one line intervening. With a line-length of c. 55 characters, there would be two lines intervening.

379. Cf. DJD XVII, 79

The Greek text has the pronoun αὐτοῦ, which may reflect the personal suffix

ו

- in its source text, though it may well be the translator’s style to supply the pronoun. Even if the personal suffix is derived back to Hebrew, it is probably secondary and may have derived from the context (cf.

וֹשׁאֹר

, earlier in the same verse). All in all, I regard the variants as independent, not belonging to the editorial layer of short/long texts.

Fragment 13 (PAM 43.113 4.2) has traces of a few letters from three consecutive lines.

Despite the small size of the fragment, it fits nicely into the reconstruction made with the help of fragment 12, totalling the same line-lengths (Frag. 12, l. 4 = Frag. 13, l. 1; Frag. 12, l. 5 = Frag. 13, l. 2).

The reconstruction is as follows:³⁸⁰

17:5] [ןירשה לקשמו שובל אוה םישקשק ]

֯ן ֯י[רשו ושאר לע תשחנ עבוכו

֯ת[שחנ םילקש םיפלא תשמח]

[ויפתכ ןיב ןדיכו וילגר לע תשחנ תחצמ]ו 6

7] [ותנצ אשנו לזרב םילקש תואמ שש ות]בה[לו םיגרא רונמכ ותינח ץעו Variant

17:5

ת ֶשֹׁח ְנ

M Q^a (

֯ת

]

שחנ

)

} לזרבו תשחנ

G (χαλκοῦ καὶ σιδήρου)

At the end of 1 Sam 17:5, the Septuagint has, against the Masoretic text and the Qumran scroll, the plusκαὶ σιδήρου‘and iron’, which indicates

לזרבו

in its source text. This is clearly an addition anticipating the similar expression in v. 7. An unintentional scribal mistake, this variant is not dependent on the variants that belong to the short/long edition of the story.

Fragment 14 (PAM 43.124 8.2) has letters on two consecutive lines. In DJD XVII, this is placed just one line after fragment 13. With such an alignment, it fits well into the column, making the reconstructed line-lengths 57 characters (frag. 13, l. 2) and 55 characters (frag.

14, l. 1).

The reconstruction is as follows:³⁸¹

וינפל ךלה]

[ואצת המל םהל רמאיו לארשי תכרעמ לא ארקי]

֯ו דמעי[ו

17:8

[ יכנא או]

ׄלה ה[מחלמ ךורעל]

380. DJD XVII, 78.

381. DJD XVII, 78.

Variant

17:8

אוֹל ֲה ה ָמ ָח ְל ִמ

M Q^a (

או] ׄלה ה[מחלמ

)

} אולה ונתארקל המחלמ

G (πολέμῳ ἐξ ἐναντίας ἡμῶν; οὐκ)

The Septuagint has the plusἐξ ἐναντίας ἡμῶνin comparison with the Masoretic text and the Septuagint. The expression is Hebraistic, indicating

ונתארקל

in the source text. Cross considers

ונתארקל

to have been omitted from the text due to a graphical error, since

המחלמ

ends in the same letter, he, as

אולה

begins.³⁸² This is possible, though, in that case, more usual would be then to continue the text without the letter he, with the following word

אול

(i.e.,

אול המחלמ

), unless one assumes also a dittography for the letter he. Another explanation is that the idea of the phrase

ונתארקל

was picked up from v. 2, where a similar expression (sc.,

םיתשלפ תארקל

) occurs. In either case, the variant readings are not dependent on the significant pluses found in the Masoretic text.

Fragment 16 (PAM 43.113 5.2) has letters on three consecutive lines. In the DJD edition, these are positioned as parts of 1 Sam 17:40–41. One cannot know whether the fragment belongs to the same column as fragments 12–14 or if it is from the following column, but the reconstruction, which follows mainly the Masoretic text, gives corresponding line-lengths:

the first line has 52 characters and the second 51.

The reconstruction is as follows:

]

[ילכב םתא ]

֯ם[ישיו לחנה ןמ םינבא יקלח השמח ול רחביו ודיב

ית]שלפה[ לא שגיו ודיב ועלקו טוקליב ול רשא םיעורה]

[ךליו41

[ וינ]

֯פל ה[נצה אשונ שיאהו דיוד לא ברקו ךלוה יתשלפה]

Variant

1S17:41

וי ָנ ָפ ְל ה ָנּ ִצּ ַה

M Q^a(

וינ] ֯פל ה[נצה

) L (τὸν θυρεὸν αὐτοῦ ἔμπροσθεν αὐτοῦ)

}

> G.

The Qumran scroll contains this plus, which is not present in the Septuagint.

Fragment 17 (PAM 43.113 6.5) contains letters from two consecutive lines. Again, one cannot be sure to which column the text belongs, but it fits well into 1 Sam 18:4–5, as reconstructed in DJD. With such an alignment, the reconstructed line-length, 55 characters, is similar to those of previous fragments.

The reconstruction is as follows:

ורגח דעו ותשק דעו ובר] ֯ח ד ׄע[ו וידמו דודל והנתיו וילע]

[אציו5

[ישנא לע לואש והמשיו לי]

֯כשיו ל ׄו[אש ונחלשי רשא לכב דוד]

382. DJD XVII, 79

Although containing only a few letters, this fragment turns out to be extremely interesting.

That is, while 1 Sam 17:55–18:5 is missing from the original Septuagint (Old Greek),³⁸³ if fragment 17 has been placed correctly, as it seems it is, this would indicate that manuscript 4QSam^a contained 1 Sam 18:4–5 and most likely the whole plus 1 Sam 17:55–18:5.

3.2.2 The Manuscript 1QSam

1QSam was not dated in the DJD series, but recently it is paleographically dated approximately the first half of the first century BCE.³⁸⁴ The manuscript comprises eight fragments, of which fragments 2–8 are from 2 Samuel and fragment 1 from 1 Samuel.

Fragments 2–8 clearly belong together, since, at the time of their discovery, they were attached to one another, forming one segment of a rolled-up scroll. Thus, fragments 2–8 have more or less similar shapes, while fragment 1 has a different one.

Since fragment 1 is separate from other fragments and contains only few letters, its placement is more uncertain than those of fragments 2–8. It is reconstructed as a part of 1 Sam 18:17–

18:

[יהתו וב ידי יהת לא רמא לואשו הוהי תו]

ׄמחלמ םח[להו ליח ןבל יל היה ךא השאל ךל]

םיתשלפ די וב]

18:18

[היהא יכ לארשיב יבא תחפשמ ייח ימו יכנא ימ ]לוא ֯ש[ לא דוד רמאיו

With such a placement, the line-length would be c. 62 characters, if the scroll had contained approximately the same amount of text as the Masoretic text in these verses. In comparison, the medial three lines in the best-preserved fragment (8) contained 68, 62 and 64 characters per line. Thus, the placement in 1 Sam 18:17–18 seems to be reasonable. Assuming the identification of 1 Sam 18:17–18 is correct, the case is interesting, since 1 Sam 18:17–19 does not belong to the original Septuagint.³⁸⁵Thus, it seems that manuscript 1QSam contains the same plus as the Masoretic text.

On the identification of the text, according to DJD XVII, the fragment 1 was found apart from other fragments, in spite of which, it is claimed as belonging to the same scroll and that the identification given is certain.³⁸⁶ I have tried to find other possible placements for the reconstruction. Exactly the same letters, ]מחלמ םח[ …1 line missing… ]לואש[ does not appear elsewhere in the Books of Samuel, but some similar cases do appear. In the following, the Masoretic text is divided into two lines and a possible placement of the fragment marked with a grey background:

383. According to Lust (1986, 7–9, 14) verses 1 Sam 18:1, 3–4 have been part of the Original Septuagint. The key point of his argumentation is that references to these verses are found in 2nd Century CE writer, Hip-polytus’ text De Davidi et Goliath 16. In subsection 4.3.4 I will show that it is more likely, that the allu-sions in Hippolytus’ text originate from other Greek translations (e.g. οἱ γ’) than the Septuagint.

384. Ulrich 2016, section 5.1.1.

385. The passage 1 Sam 18:17–19 is parallel to vv. 20–26 (Saul gives his daughter to David as a wife).

386. DJD I, 64: ‘Le f. I a été trouvé à part, mais son appartenance et son identification semblent certaines.’

Position 1°: 1 Sam 8:20–21.

Line-length of c. 75 characters. The Masoretic text has +

ב

. The first letter in 1QSam frag. 1, l. 1, could be also he instead of ḥet.

Line-length of c. 62 characters. The Masoretic text has +_ל. The first letter in 1QSam frag. 1, l. 1, could be also he instead of ḥet.

All in all, the placement of fragment 1 in 1 Sam 18:17–18 seems still to be the best possible option, albeit not the only one.

3.2.3 Conclusion

It seems that manuscripts 4QSam^aand 1QSam share at least some of the pluses found in the Masoretic text. However, the evidence for the pluses is slight and dependent on only a few letters. Furthermore, these observations indicate only that the pluses existed at the time of the copying of the scrolls. The question of their originality must be determined by other arguments (see sections 3.3–3.6). Interestingly, 4QSam^a not only agrees with one plus of a Masoretic text; it also agrees with the Septuagint against the Masoretic text (‘four’pro ‘six’

regarding Goliath’s height in cubits).

In document The Hebrew Text of Samuel : Differences in 1 Sam 1 – 2 Sam 9  between the Masoretic Text, the Septuagint,  and the Qumran Scrolls (sivua 124-129)