Abstract
In 2 Sam 11 and Num 22, the direct speech of the first speaker (Joab and Balak, respectively) is passed on and quoted by others. However, the messenger does not convey the message exactly as it had been formulated by the first speaker but instead adapts it to his own point of view. In this short contribution I show that the differences in the Hebrew text between the original message (the first speaker’s direct speech) and the message as conveyed (the embedded direct speech within the messenger’s own direct speech) have a specific function in the narrative. We should respect these differences when we translate. They should therefore not be treated as text-critical problems and be harmonized on the basis of variant readings from ancient versions, as has been done from time to time.
Keywords
The first time I met David Clark was at a conference in Dallas in 1993, where he gave a paper on discourse units in Zech 1−6 (Clark 1994). As David demonstrated, it is important not only to look at how the beginning and the end of discourse units are marked, but also to look at how the language can be different in the various levels of embedding of direct speech in a unit, whether in a vision or oracle (as in Zechariah) or a narrative unit.
In 2 Sam 11 and Num 22, for instance, the direct speech of the first speaker (Joab and Balak, respectively) is passed on and quoted by others. I would like to show that the differences in the Hebrew text between the original message (the first speaker’s direct speech) and the message as conveyed (the embedded direct speech within the messenger’s own direct speech) are fitting within the narrative. We should respect these differences when we translate. They should therefore not be treated as a text-critical problem and be harmonized on the basis of variant readings from ancient versions, as has been done from time to time.
2 Sam 11.22-24
Our first example is taken from 2 Sam 11. After Uriah has been killed in battle, the Masoretic Text (as translated by NRSV) reads as follows: 19 and he instructed the messenger, “When you have finished telling the king all the news about the fighting, 20 then, if the king’s anger rises, and if he says to you, ‘Why did you go so near the city to fight? Did you not know that they would shoot from the wall? 21 Who killed Abimelech son of Jerubbaal? Did not a woman throw an upper millstone on him from the wall, so that he died at Thebez? Why did you go so near the wall?’ then you shall say, ‘Your servant Uriah the Hittite is dead too.’” 22 So the messenger went, and came and told David all that Joab had sent him to tell. 23 The messenger said to David, “The men gained an advantage over us, and came out against us in the field; but we drove them back to the entrance of the gate. 24 Then the archers shot at your servants from the wall; some of the king’s servants are dead; and your servant Uriah the Hittite is dead also.” (2 Sam 11.19-24 NRSV)
The king may well be angry that the army made themselves an easy target, but the messenger is clever. Against and beyond Joab’s instructions (vv. 19-21) he does not wait for the king’s angry outburst concerning the battle but hastens to conclude his report before the king can react. The messenger justifies the mistake of approaching too close to the wall (v. 23) and gets straight to the point that will please the king: Uriah is dead (Alter 1999, 255).
In the LXX the narrative is different: the messenger passes on his message just as Joab had instructed him to do and then waits for the king to be angry instead of astutely continuing as in the Hebrew. So the questions as Joab anticipated them in vv. 20-21 are repeated almost verbatim by David at the end of v. 22. LXX (as translated in NETS) reads as follows: 22 And the messenger of Ioab went to the king in Jerusalem and came and told Dauid all that Ioab had told him, all the news of the fighting. And Dauid was angry with Ioab and said to the messenger, [the verbatim recapitulation starts:] “Why did you draw so near the city to fight? Did you not know that you would be struck from the wall? Who struck Abimelech the son of Ierobaal? Did not a woman throw a piece of millstone on him from the wall, and he died at Thamasi? Why did you draw so near the wall? 23 And the messenger said to Dauid, … (2 Kgdms 11.22-23 NETS)
Only after this insertion or “editorial intervention” (Tov 2015, 236) does the messenger continue with vv. 23-24. In this way the conversation between the messenger and the king is made to correspond with Joab’s earlier instructions. The LXX translator wanted a smooth flow of the narrative, with David’s words being in line with what Joab expected him to say (Goslinga 1962, 210). The LXX translator may not have been satisfied with the fact that Joab’s prediction of David’s anger did not come to pass, and may not have understood the cleverness of the messenger (Pisano 1984, 54; Fernández Marcos 1987, 292; Barthélémy 1982, 260–61).
This makes LXX a quite different narrative from MT. There is no ground, then, for using the extra part in the LXX account to “restore” the Hebrew text (on the assumption that this part had been removed from the Hebrew) and include it in one’s translation, as the Russian Synodal Version (in square brackets), REB and DHH (both with a note), and NBJ have done. It appears that these translations have treated this difference between LXX and MT as textual variants (i.e., variant readings of the same text) rather than literary variants (i.e., an indication of different editorial versions) of the narrative that should not be mixed together. 1
Num 22.5-6, 11
Our second example is taken from Num 22. In vv. 5b-6, Balak’s direct speech is conveyed to Balaam by messengers. In v. 11, Balak’s direct speech of vv. 5b-6a is embedded as a second-degree quotation within Balaam’s direct speech to God, and this affects the exact wording; it is not repeated verbatim. Balaam is quoting Balak rather freely (Gispen 1964, 81) and is putting some distance between himself and Balak’s perspective and intention.
Balak’s message to Balaam of vv. 5b-6 is conveyed by Balaam to God in vv. 10-11: 5b “Look (הִנֵּה), a people (עַם) has come out (יָצָא) of Egypt. Look (הִנֵּה), it covers (כִסָּה) the face of the earth and it is sitting over against me. 6 You must (נָא) come then, curse (אָרָה) this people for me, since it is mightier than I; perhaps I shall succeed that we may defeat (נַכֶּה)
2
it and I may drive it (וַאֲגָרְשֶׁנּוּ) out of the land;
3
for I know that whomever you bless is blessed, and whomever you curse is cursed.” (vv. 5b-6, my translation) 10b “Balak … has sent to me: 11 ‘Look (הִנֵּה), the people (הָעָם) that has come (הַיֹּצֵא) out of Egypt covers (וַיְכַס) the face of the earth. Come then, hex (קָבָה) it for me; perhaps I shall succeed to fight (לְהִלָּחֶם) against it and drive it out (וְגֵרַשְׁתִּיו).’”
4
(vv. 10b-11, my translation)
In v. 5b, the two הִנֵּה + qatal clauses are typical: Balaam, the addressee, is not expected to be familiar with these events; the speaker assumes that they are new to Balaam: “Look, a people … .” In v. 6, the yiqtol נַכֶּה and weyiqtol וַאֲגָרְשֶׁנּוּ denote volitivity; Balak is really keen to drive Israel out of the land.
In v. 11, Balaam quotes what was Balak’s message in vv. 5b-6a, but it is already clear from the start that the quotation is not quite verbatim. The differences make good sense and fit well in their new context:
By leaving out וְהוּא יֹשֵׁב מִמֻּלִי “and it is sitting over against me” and כִּי־עָצוּם הוּא מִמֶּנִּי “it is mightier than I,” Balaam reduces Balak’s personal perspective (Alter 2004, 797–98), thereby distancing himself from it.
By saying לְהִלָּחֶם “to fight” instead of נַכֶּה “we may defeat,” Balaam leaves out Balak’s fluctuation between singular and plural—another reduction of Balak’s perspective.
Balaam (“Come”) leaves out the entreating particle נָא that marked the urgency of Balak’s message (“You must come,” “Please come”). 5 Unlike some translations, NLT shows the difference: “Please come and curse” (v. 6); “Come and curse” (v. 11). TOB does this as well: “Viens donc, je t’en prie, et maudis-moi ce peuple” (v. 6); “Viens donc et maudis-le pour moi” (v. 11.)
Balaam quotes Balak’s message as saying הָעָם הַיֹּצֵא “the people that has come out” (with two definite articles and a qotel) even when, in fact, Balak had said עַם יָצָא “a people has come out” (with a qatal). Balaam is referring to the Israelites and their coming out of Egypt as information that is already known, even though he does not refer to them explicitly by name (de Regt and Wendland 2016, 493). 6 The equivalent of the rendering “the people that has come out” in v. 11 is found in, for example, FC, NBJ, PdV, NBS, CEI, and GuNB.
Indeed, the quotation in v. 11 actually comes with different verb forms: “Look, the people that has come out (הָעָם הַיֹּצֵא) of Egypt covers (וַיְכַס) the face of the earth.” Instead of the qatals in v. 5b, Balaam now uses the qotel and the wayyiqtol of the same verbs, so that these actions are no longer presented as unfamiliar to the addressee (as they still were in v. 5b) but as accessible to the addressee. After all, Balaam’s addressee at this point is God, who, he expects, will already know (and so will the reader). 7 There is no need, then, to accept the recommendation in BHS and follow the readings in 4QNumb (עם יצא) and ancient versions (Samaritan Pentateuch, LXX, Peshitta) by reading עַם יָצָא (v. 5) here in v. 11 as well, as Budd does (1984, 249, 254).
The verb at the end of v. 11 is a weqatal (וְגֵרַשְׁתִּיו) instead of the volitive weyiqtol (וַאֲגָרְשֶׁנּוּ) of v. 6. Balaam is not involved: he quotes Balak in the first person but without sharing his interest in driving Israel from the land. The contrasts between the verbs in vv. 5-6 and v. 11 make for an instructive example in the teaching of the Hebrew verbal system (for more on this see de Regt 2008), even if these contrasts will not always lead to different renderings of these verbs in the target language.
Concerning the imperative, “curse,” Balaam substitutes קָבָה for אָרָה (v. 6). קָבָה is the first of ten instances in the Balaam story of the verbal root קבב, which occurs nowhere else in the Pentateuch (Sherwood 2002, 175). It “may imply that the curse would be accompanied by acts of magic or sorcery” (de Regt and Wendland 2016, 522). Perhaps the narrator is “playing on usual forms … in an effort to emphasize the theme of malediction” in the Balaam story (Levine 2000, 152). The renderings “hex” (Alter 2004), “verwünsche” rather than “verfluche” (Buber), “verwens” (Gispen 1964, 82), and “voue-le pour moi à la malédiction” (condemn it to the curse for me) rather than “maudis … pour moi” (curse … for me; NBS), and “call down troubles upon” (de Regt and Wendland 2016, 522) are rare attempts to differentiate between these two synonyms in translation.
As illustrated above, a number of translations show how the differences between vv. 5b-6 (Balak’s perspective) and v. 11 (Balak’s embedded direct speech, but from Balaam’s perspective) can be expressed.
Conclusion
Second Samuel 11.22-24 and Num 22.5b-6, 11, illustrate that the differences in the Hebrew text between the original message (the first speaker’s direct speech) and the message as conveyed (the quoted direct speech embedded within the messenger’s own direct speech) have specific functions in the narrative. In both cases, the messenger adapts the message to his own situation and point of view. Such differences in the text should therefore not be treated as text-critical problems and be harmonized on the basis of variant readings from ancient versions, as some translations have done. Rather, we should respect and express those differences when we translate.
Footnotes
1
On the distinction between textual and literary variants, see Tov 2012 (chapter 7, “Textual and Literary Criticism,” 283–326, esp. 324–26) and 
(236–38).
2
, 363) comment on LXX applies to MT here as well: “The fluctuation of singular and plural might seem odd, but it reflects the identification of king and people. The plural is hardly the royal ‘we’ or plural of majesty, but is simply a case of the king viewing himself as absolute monarch, and therefore incorporating in himself the will and action of the people.” So already the Jewish medieval commentator Rashi, “I and my people, we will defeat them.”
3
As Buber translates, “dass ich es aus dem Lande treibe.”
4
As Buber translates, “und ich vertreib es.”
6
These features of the Hebrew of v. 11 contradict (276 n. 51) when he writes, “In Num 22,11, Balaam refers only generically to ‘a people’ that has come out of Egypt and now covers the land of Moab, indicating that he has no direct knowledge of the Israelites. In fact his description of the ‘a people’ is merely a repetition of the message conveyed to him in vv. 5-6.”
7
In , esp. 82–87) I refer to such accessibility of an action to the addressee (as indicated by a wayyiqtol form) as “cognitive proximity”: the action is tightly linked to what the addressee is expected to know about the situation. A qatal indicates that the action is “non-proximate” (less accessible) to the addressee: the action is loosely linked to what the addressee is expected to know about the situation.
Abbreviations
BHS Biblia Hebraica Stuttgartensia (1977)
CEI Versione Conferenza Episcopale Italiana (2008)
DHH Dios Habla Hoy DC Estándar (1994)
FC Bible en français courant (1997)
GuNB Gute Nachricht Bibel (1997)
LXX Septuagint
MT Masoretic Text
NBJ Nouvelle Bible de Jérusalem (1998)
NBS Nouvelle Bible Segond (2002)
NLT New Living Translation (2007)
NRSV New Revised Standard Version (1989)
PdV Parole de Vie (2000)
REB Revised English Bible (1989)
TOB Traduction oecuménique de la Bible (2010)