Problems of Text and Midrash in the Third Book of Reigns

This text was originally published in Textus 7 (1969), 11–13 pp.

¹ For three good reasons the following observations on 3 Reigns should be regarded as little more than an interim report. In the first place my own investigations are as yet incomplete. In the second, Dr J. D. Shenkel’s researches into the chronology and the recensional development of the Greek text of Kings have not yet been published (though there is hope that they will be in the not too distant future), and the chronological peculiarities of 3 Reigns are so closely connected with all its other large, distinctive features, that no one would wish to commit himself finally to any particular view of the book as a whole, until Dr Shenkel’s findings are made known.² And in the third place, as long as the bulk of the material from Cave 4 at Qumran remains unpublished, one must reserve one’s judgment on the basic question whether the peculiarities of 3 Reigns are, or are not, based on a Hebrew text, remembering that 3 Reigns belongs to the same complex of books as 1 and 2 Samuel, and that F. M. Cross has warned us that the Hebrew MSS of Samuel ‘from Cave IV are all at wide variance with the Massoretic tradition’, and all have ties with ‘the tradition used in the Septuagint translation’.³

On the other hand, some, at least, of the conclusions that will be presented, can be regarded as definite. Given a peculiarity in the LXX, we may not be able to say definitely whether it is, or is not, founded on a Hebrew text differing from the MT. But if, for the argument’s sake, we suppose it is, we can proceed to consider whether this supposed Hebrew vorlage, or the MT, better represents the original story. And at this level decision is often possible.

Moreover, even where final decisions are not possible, an interim report can serve the useful purpose of surveying and assessing the problems, putting forward tentative solutions, and indicating the points at which further research is most needful and most likely to yield profitable results.

I propose, then, to divide this paper into two parts. In the first part, I shall attempt to demonstrate that the several re-orderings of the material in 3 Reigns, to which I have elsewhere called attention, along with some others, are not isolated phenomena, but are part of a deliberate scheme of re-ordering that covers the whole book and extends beyond it. In addition, I shall argue that the re-interpretations of the leading characters in this book, to which also I have called attention elsewhere, are bound up with this large-scale re-ordering of the material and come from the same hand.

And then I shall point to evidence that appears to suggest that some, at least, of this reinterpretation and re-ordering entered the Greek text, not at the time of its original translation, but as a result of some later revision of the Greek text. This revision was probably based, at least in part, on written Hebrew (or Aramaic) traditions of one kind or another.

In the second part of the paper I shall discuss the implications of all this for the use of 3 Reigns as a witness in the task of recovering the original Hebrew text of 1 Kings. And finally I shall ask, more than answer, the question: what can the treatment that 1 Kings has suffered at the hands of its reviser(s), translator(s) and translation-reviser(s) tell us about their attitude to Scripture? Did they have any concept of an inspired and sacrosanct original which might not be changed, and if so, what relation did they think the revised Hebrew text and the Greek translation and its revisions bore to that inspired and sacrosanct original?

I

A good starting point, from which to survey 3 Reigns’ re-ordering of the text, is provided by the duplicate translations. In 3 Reigns there are two prominent pairs of duplicate translations and in 4 Reigns there is a third pair, and they all have elements in common.

For two of them difference from the MT in the matter of chronology is the chief cause behind the duplication. Normally the LXX agrees with the MT in its well known narrative-pattern of so many kings of Judah with chronological cross-references to the contemporary king of Israel, followed by so many kings of Israel with chronological cross-references to the contemporary king of Judah. But in 3 Reigns 15 and 16 there are big differences in chronology between the LXX and the MT. Quite clearly the LXX is confused and inconsistent, and the storm centre of its troubles seems to lie in its confusion of Zimri with Omri, both of whom it calls Ζαμβρει (16:15 ff. and 21 ff.), and in its attributing to Zimri a reign of seven years instead of seven days. The result is that, according to its peculiar system, the Judaean king, Jehoshaphat, begins his reign during the reign of the Israelite king Omri, and so falls due for independent mention in chapter 16:28a ff.; whereas, according to the MT’s system, Jehoshaphat begins his reign during the reign of Ahab, and therefore falls due for independent mention in chapter 22:41. Accordingly the LXX, following its own peculiar system, has a translation of the summary of Jehoshaphat’s reign stationed in chapter 16:28 a–h, and another different translation of that same summary stationed according to the MT’s system in chapter 22:41–51. Similarly, the LXX, following its own chronological scheme, places a translation of the introduction to Jehoram b. Ahab’s reign in 4 Reigns 1:18 a–d, and then, following the MT’s scheme, it puts another different translation of this same introduction in 4 Reigns 3:1–3.

The third pair of duplicate translations, on the other hand, has nothing to do with chronology. They occur in the story of the execution of Shimei. This story is told, in both the LXX and the MT, in two parts. The first part, David’s dying charge to Solomon to find some pretext to execute Shimei, is told along with the rest of David’s last words, MT 1 Kings 2:8–9, LXX 3 Reigns 2:89. The second part, which tells how Solomon imposed an arbitrary restriction on Shimei and executed him when he exceeded it, is placed, at some distance from the first part, among the early acts of Solomon’s reign, MT 1 Kings 2:36–46, LXX 3 Reigns 2:36–46. But in the LXX, immediately in front of the second part, the first part is repeated, 2:35 l–o, in a translation that differs considerably from the translation given on the first occasion, 2:8–9.

Now, while the duplication of the translation of the Shimei episode is not connected with any difference in chronology between the LXX and the MT, the Shimei duplicates have this much in common with the Jehoshaphat and the Jehoram duplicates: in each pair the two translations are quite different and come quite obviously from different hands. Both cannot be original. And it is commonly agreed that the translation which stands in the position peculiar to the LXX is on each occasion the older translation, while the translation that stands in the position that agrees with the MT’s scheme is each time a translation that approximates more nearly to the MT’s text and shows signs of being the more recent translation.⁴

If this, then, is so, one could easily jump to the conclusion that the translations which stand in the positions which are peculiar to the LXX, are the original LXX translations and represent a non-MT Hebrew vorlage, and that the translations which stand in the positions which agree with the MT, are the result of a later revision of the LXX on the basis of an MT-type Hebrew. And one might further be tempted to suppose that the chronological scheme into which the original LXX translations fit, is the scheme followed by the original LXX (and its Hebrew vorlage).

But we would be wiser not to jump to these conclusions, until we have put the question of these LXX duplicates into the wider context of the LXX’s overall order of contents in 3 Reigns, and its other, smaller, but no less significant, duplicate translations. For the moment we should simply notice the range within which these three pairs of duplicates are found: the first pair in 3 Reigns 2, the second in 3 Reigns 16 and 22, and the third in 4 Reigns 1 and 3. In other words, they cover between them the whole of 3 Reigns and the beginning of 4 Reigns. What is more, if one accepts Thackeray’s analysis of the four Books of Reigns, then our three pairs of duplicates extend from 3 Reigns 2:8­–9, which belongs to Thackeray’s βγ section, through 3 Reigns 2:35^l–o , which belongs to his γγ section, to 3 Reigns 22:41–51, 4 Reigns 1:18^a–d and 3:1–3, which belong to his γδ section.

But, as has just been hinted, these three pairs of duplicates are not the only duplicate translations in 3 Reigns. Indeed, the Shimei story is, in the LXX, surrounded on both sides with two strange collections of textual variants, many of which are likewise duplicate translations of verses translated differently elsewhere in 3 Reigns.⁵ Their positioning is very odd. To grasp its significance let us recall that in the MT the story of Shimei is told in two parts, which stand at some distance from each other, Pt. I at 2:8–9 and Pt. II at 2:36–46. In the LXX there are two different translations of Pt. I, let us call them A and B. Pt. I A stands, as in the MT, at 2: 8–9; but Pt. I B stands immediately in front of Pt. II, thus:

Shim. Pt. I B 2:35^l–o , Shim. Pt. II 2:36–46.

---

The two Miscellanies are then placed thus:

---

Misc. 1 2:35^a–k (containing duplicate translations) Shim. Pt. 1 B 2:35^l–o (duplicate translation) Shim. Pt. II 2:36–46 Misc. 2 2:46^a–l (containing duplicate translations)

---

the four items running consecutively without interruption. It would be reasonable to suppose that the positioning of the duplicate translation, Shimei Pt. I B, (though not necessarily the translation itself), is due to the same mind that arranged the positioning of Miscellanies 1 and 2. with their duplicate translations.

But even more significant than positioning is the matter of leading theme. As I have pointed out at length elsewhere,⁶ Miscellanies 1 and 2, with a device and an intention which both patently belong to midrashic exegesis, contrive to give first place to the theme of Solomon’s wisdom.⁷ Moreover, not only is the story of Shimei’s execution an example of Solomon’s wisdom, but it is Pt. I of the story that explicitly lays stress on the wisdom that was involved in Solomon’s handling of the case.⁸ The consecution

puts it practically beyond doubt that the positioning of the duplicate translation, Shimei Pt. I B, along with Miscellanies 1 and 2, which both contain duplicate translations, is deliberately contrived to develop the leading theme of Solomon’s wisdom.

Next we must observe that this scheme is integrated with the LXX’s peculiar scheme of order in the following chapters.

Chapter 3

Miscellany 1, Shimei Pt. I B and Miscellany 2 have, of course, nothing to correspond to them in this position in the MT. But after Miscellany 2 the LXX begins ch. 3 and agrees with the MT’s order throughout the whole chapter with one notable exception. The MT has three items: Solomon’s marriage to Pharaoh’s daughter, his visit to Gibeon, and the incident of the two harlots. The LXX has the second two, but its equivalent of the first stands, at some remove, at 4:31. This secures for the LXX the following consecution

Now not only does this arrangement allow the theme of Solomon’s wisdom to proceed uninterruptedly, which it would not do if Solomon’s marriage to Pharaoh’s daughter came in between Miscellany 2 and the visit to Gibeon, but it removes from this wisdom context a marriage which later rabbis considered the most disastrously unwise thing that Solomon ever did. It may well be that the mind behind the LXX’s order would have agreed with the later rabbis’ judgement.

Chapter 4:1–30 (MT 4:1–5:14)

This passage in both the MT and the LXX is devoted to a description of Solomon’s rule, his officers of state, his supplies, the extent of his empire and his wisdom. Internally the MT and the LXX disagree over several details of order and content, and the LXX’s order certainly seems more logical. But the relevant point to grasp for our immediate purpose is that the leading themes of this section of the main body of the book are closely matched by the leading themes of Miscellany 2.

4:1-30		Misc. 2. 2:46a-l
καὶ ἦν ὁ βασιλεὺς Σαλωμων βασιλεύων ἐπὶ Ισραηλ.	46l.	Σαλωμων υἱος Δαυιδ ἐβασίλευσεν ἐπὶ Ισραηλ καὶ Ιουδα ἐν Ιερουσαλημ.
καὶ οὗτοι οἱ ἄρχοντες, οἳ ἦσαν αὐτοῦ + names	46h.	καὶ οὗτοι οἱ ἄρχοντες τοῦ Σαλωμων + names
καὶ ταῦτα τὰ δέοντα τῷ Σ. + list of provisions	46e.	καὶ τοῦτο τὸ ἀριστον τῷ Σ. + same list of provisions
ὅτι ἦν ἄρχων πέραν τοῦ ποταμοῦ, καὶ	46f.	ὅτι ἦν ἄρχων ἐν παντὶ πέραν τοῦ ποταμοῦ…..
ἦν αὐτῷ εἰρήνη ἐκ πάντων τῶν μερῶν …	46g.	καὶ ἦν αὐτῷ εἰρήνη ἐκ πάντων τῶν μερῶν …
	46b.	καὶ Σ. ἦν ἄρχων ἐν πάσαις ταῖς βασιλείαις
	46k.	καὶ ἦν ἄρχων ἐν πᾶσιν τοῖς βασιλεῦσιν….
… φρόνησιν … σοφίαν …	46)a.	καὶ ἦν ὁ βασιλεὺς Σ.
… φρόνησιν … φρονιμόυς		φρόνιμος σφόδρα καὶ
… ἐσοφίσατο bis.		σοφός
… σοφίας. bis.

This close agreement, of course, is neither an accident nor a mystery. It comes about because Miscellany 2 is made up largely of material that either duplicates (sometimes in a slightly different translation) verses which also stand in the section 4:1–30, or else supplies translations of verses which occur in the corresponding section of the MT (4:1–5:14) but are entirely lacking in the LXX section (4:1–30). But it means that the marshalling of this material in Miscellany 2, so that Miscellany 2 and the section 4:1–30 both deal with the same general topic (Solomon’s wisdom in government, empire, administration and supplies), confirms our earlier impression that the two Miscellanies and the Shimei story (with its duplicate translation) have been deliberately worked into a scheme of order that embraces the main body of the book. The scheme, as far as we have gone at the moment now looks like this: —

Of these seven sections Miscellany 1 is the only one whose description of Solomon’s wisdom we have not yet examined in detail. We are now about to see that its leading theme is matched by the theme of next great section of the main body of the book.

Chapters 4:31–9:9 (MT 5:15–9:9)

This section, both in the LXX and the MT, deals with Solomons building activities: the preparations, Hiram’s help, the building of the temple and the king’s palace, the dedication of the temple, and the second appearance of the Lord to Solomon upon the completion of the temple and palace. Likewise Miscellany 1 has as its main burden the theme of Solomon’s building activities. Montgomery was the first to see this and to show that Miscellany 1 is not just a jumble of variants haphazardly swept together in a heap.⁹ Specially significant, for our present purpose, is the fact, which I have elsewhere expounded at length,¹⁰ that the editor behind Miscellany 1 shows in his editorial remarks the same concern for the order in which the temple and the royal palace were built as does the main record in our present section. Unlike the MT, which follows the (doubtless, original) order, Temple, Temple, Palace, Temple, our section has Temple, Temple, Temple, Palace; while the editorial comments in Miscellany 1 go out of their way to stress that Solomon built the house of the Lord ἐν πρώτοις 2:35^c, and that he built the many cities he did build only after he had built the house of the Lord. Now the concern to maintain that Solomon put the building of the Lord’s house before his own house is shared by later rabbinic tradition.¹¹ The fact that Miscellany 1 and our present section both show this same concern, could, therefore, be nothing more than a coincidence. But in view of the other evidence already cited, it is much more likely that the editor of Miscellany 1 is also responsible for the LXX’s peculiarity of order in our present section.

The general scheme, then, now stands as follows:

This surely is enough evidence to show that the duplicate translation in the Shimei story and the two Miscellanies with their duplicate translations have been integrated with the LXX’s scheme of order in the main body of the text up to 9:9. We must now pursue the evidence that shows that the LXX’s peculiarities in this part of the main text are involved in the main LXX peculiarities throughout the rest of the Solomon story, and, indeed, the rest of the book.

First of all there is the positioning of the material on the marriage of Solomon to Pharaoh’s daughter. As we saw above, the announcement of the wedding in the LXX does not stand, as in the MT, in front of the Gibeon visit, for there it might conflict too obviously with the ‘wisdom theme’. It has been placed in front of the main section on Solomon’s building activities at 4:31–33, and closely integrated with it on the grounds of timetable. After all, the verse, both in the MT and the LXX, says that ‘Solomon took the daughter of Pharaoh to wife, and brought her into the city of David until he had completed the house of the Lord and his own house . . .’ What, then, could be more natural (or pedantic) than to place this announcement at 4:31 immediately before the preparations for these building operations? Moreover, in order further to integrate this marriage with the building theme, as soon as the building and the dedication with its accompanying vision are over, 9:9, the LXX puts the announcement, ‘then Solomon brought up the daughter of Pharaoh out of the city of David into the house which he had built for himself in those days’. It would appear that the marriage is not meant to stand by itself as a separate item—how could it properly do so in the middle of paragraphs on the theme of Solomon’s wisdom?—but rather to be taken as a detail closely connected with the building operations.

Moreover, this statement at 9:9 ‘Then Solomon brought up . . .’, which has nothing in this position in the MT to correspond to it, is very closely paralleled by a phrase in the MT at 9:24a, ‘But Pharaoh’s daughter came up out of the city of David unto her house which Solomon had built for her’.¹² In the same paragraph in the MT (it extends from 9:15­­–25) there stands at v. 16 the information ‘Pharaoh king of Egypt had gone up and taken Gezer, and burn tit with fire, and slain the Canaanites that dwelt in the city, and given it for a portion unto his daughter, Solomon’s wife. And Solomon built Gezer’. This information, cited practically word for word, the LXX puts at 4:32, 33, that is, immediately after the marriage announcement in 4:31, and so deliberately groups the marriage items together; and with this reference to the building of Gezer, the LXX ties the marriage item even more closely to the section on building operations.

Furthermore, the paragraph from which verses 16–17a have been taken to stand immediately before the building section, and verse 7a has been taken to stand immediately after the building operation, has most of it been transferred by the LXX to stand at chapter 10:23 ff. But in the LXX some sizeable parts of it have been removed to stand in one or other of the two Miscellanies! Thus

Misc. 1. 2: 35f	offers a translation of what in the MT is 9:24a. It follows the MT closely, whereas the translation of this verse which stands at LXX 9:9 follows a non MT-type text.
2:35f and 35g	offer a translation of MT 9:24b and 25, which are entirely absent from the LXX paragraph at 10: 23 ff.
2:35h	presents a translation of MT 9:23, though the numbers of the officers mentioned differs. Again this verse is completely absent from its expected place in LXX 10:23 ff.
2:35i	gives a translation of part of MT 9:15 and part of 17. This partly overlaps the LXX verse 10:23; but where they overlap, the translations are different.
Misc. 2. 2: 46h	offers information very similar to parts of MT 9:18 and 19. Most of it is missing from LXX 10:23 ff, and what is there—καὶ τὴν Ιεθερμαθ—differs in translation from the τὴν Θερμαι ἐν τῇ ἐρήμῳ of 2:46.

No more need be said to demonstrate that the two Miscellanies are deeply involved in the rearrangement of material that took place when the paragraph which corresponds to the MT’s 9:15–25, was uprooted and (what was left of it) removed to 10:23 ff. But the motives and methods of this removal are a very important link in the progress of our argument. I have already analysed them in detail elsewhere¹³ and there is no need to go over the same ground again here. It is sufficient to recall that the transposition of the paragraph was the work of some midrashic exegete intent on whitewashing Solomon’s character and justifying his questionable activities. Similar midrashic exegesis, accompanied by—indeed, dependent on —similar re-arrangement of the text, is found in both of the final two major sections of the book; and though, of course, this midrashic activity could be the work of several people, and is in fact probably based on traditions common to several interpreters, its integration here into the text of 3 Reigns is most likely to be the work of one editor. But before we proceed to the two final sections, we should first formally take notice of the last section of Solomon’s reign.

9:10­–10:33 (MT 9:10–10:29)

Strictly speaking, the paragraph about the removal of which we have just spoken, MT 9:15–25, LXX 10:23 ff, belongs to this section, though its rearrangement in the LXX is so closely connected with the building section and the Miscellanies, that for convenience we have already dealt with it under the building section. A few further details should be noticed. The main theme in this final section of Solomon’s reign is his wisdom and his riches: ‘So king Solomon exceeded all the kings of the earth in riches and in wisdom. And all the kings of the earth sought the presence of Solomon, to hear his wisdom, which God had put in his heart. And they brought every man his present . . . silver . . . gold . . . raiment . . .’ (MT 10:23–25, LXX 10:26–28). Actually, later rabbis were embarrassed by Solomon’s accumulation of riches, because it seemed to transgress the Deuteronomic prohibition (Deut 17:17); and certainly, if Solomon’s riches were a transgression of the law, they could not be cited as an evidence of his wisdom. The later rabbis, however, found ways of justifying Solomon’s accumulation of wealth,¹⁴ and so has the LXX. One of the LXX’s ways is the above-mentioned removal and adaptation of a whole paragraph. Its other devices are too detailed to be discussed here; they must be dealt with on another occasion. The main point is that in the LXX this whole section has so been re-arranged and re-interpreted that it can safely mention Solomon’s massive accumulation of riches as being unquestionably a result of Solomon’s wisdom. And so this section too fits in with the LXX’s general scheme which now stands as follows:

Chapter 11:1–13:34 (MT 11:1–14:20)

This section deals with Solomon’s decline, the adversaries God raised up against him, and the rebellion of Jeroboam and the ten tribes against Solomon’s successor, Rehoboam. The LXX has two separate and contradictory accounts of this rebellion. The first one, as I have demonstrated at length elsewhere,¹⁵ endeavours to whitewash Jeroboam’s character. In doing so, it shows the same spirit as the attempts to whitewash Solomon’s character in the previous sections. Moreover, it uses the same technique: whereas it is based on a Hebrew text very similar to the MT, it secures its re-interpretation in large part by a removal of a certain block of material to a different position in the narrative.¹⁶

The second account appears to be based on a Hebrew (or Aramaic?) tradition which differs completely from the MT. As again I have elsewhere pointed out,¹⁷ it seeks to blacken Jeroboam’s character. But I may here add to what I have written there, that the point at which it seeks to blacken Jeroboam is the very point at which the first LXX account seeks to whitewash him. The first account will have it that Jeroboam committed no act of rebellion and was not even present at the rebellious parliament at Shechem. The second account has it that not only was he the convenor and ringleader of the rebellious parliament (12:24ⁿ), but even before that he committed three highly treasonable acts: he amassed a very large retinue of horses and chariots, he ‘shut up’ (i.e. besieged) the city of David and ‘was lifted up against (or, aspired to) the kingdom’ (12:24^b), and he built an entrenched camp for the assembled tribe of Ephraim (12:24^f). From this it would appear that the second account is deliberately set over against the first account in order to balance the first account’s interpretation with an opposite interpretation. In just such a manner do the Talmud and Midrashim record differing and sometimes contradictory interpretations.

Moreover later in this section the LXX omits from the beginning of its chapter 14, twenty verses that are to be found at the beginning of the MT’s chapter 14. Eighteen of these verses relate Ahijah’s denunciation of Jeroboam on the occasion of the illness (and consequent death) of Jeroboam’s son, Abijah. The other two record Jeroboam’s death and the accession of his son Nadab. Now the absence of these twenty verses from the LXX is without any doubt secondary. Without them the LXX completely lacks for the end of Jeroboam’s reign the customary formula: ‘now the rest of the acts of So-and-So . . . and So-and-So slept with his fathers, and So-and-So, his son, reigned in his stead’. It is most unlikely that the original Hebrew omitted this formula here. The omission arose with the LXX or its vorlage; but whether it was an accidental omission or a deliberate excision is not so certain. It could of course be accidental, but there are signs that it is probably a deliberate excision made by the editor responsible for the introduction of the second account of Jeroboam’s revolt into the text.

The evidence is this. In the MT’s account Ahijah’s denunciation of Jeroboam (ch. 14) comes right at the end of his reign after his promotion to power had corrupted him and led him into idolatry, of which he refused to repent even after the most signal warning from God through a prophet (ch. 13). But the second account in the LXX is inserted in chapter 12 (vv. 24^a–z) and in accordance with its determination to vilify Jeroboam, this account makes out that the denunciation of Jeroboam on the occasion of his son’s illness and subsequent death occurred not only before he committed idolatry, but even before he ascended the throne (12:24^g–n). It was not a case of a reasonable man being corrupted by power; he was incurably wicked before he came to power! But with the denunciation deliberately stationed at this early point in the story, and the whole account stationed in chapter 12, it would be difficult for an editor to leave an account of the same denunciation standing at the end of Jeroboam’s reign in chapter 14. Therefore the denunciation was excised from chapter 14. But whoever it was that excised it seems not to have noticed that immediately following the denunciation in chapter 14 (vv. 1­–18) comes the formula for Jeroboam’s death and his son’s succession (vv. 19–20). Accidentally he has excised the formula along with the denunciation.

If, then, these explanations are right, the major LXX peculiarities in this section are shown to be linked with the LXX’s general scheme for the interpretation (not to say re-interpretation) of the whole book.

Chapters 14:21–16:28^h (MT 14:21–16:28)

This section comprises the reigns of Rehoboam, Abijam and Asa in Judah, and the reigns of Nadab, Baasha, Elah, Zimri and Omri in Israel. In the LXX it is chiefly noteworthy for the peculiar system of chronology which the LXX employs, and for the translation of the summary of Jehoshaphat’s reign which appears at 16:28^a–h with nothing to correspond to it in the MT in this position; whereas another translation of this same summary appears at LXX 22:41–51, where its position matches the position of the MTs summary.

Now the question of the LXX’s peculiar chronology, with all the recensional variants of it to be found in the Greek tradition, is very complicated. We need notice only two points: the position of the duplicate summary at 16:28^a–h is directly dependent on the LXX’s peculiar chronology; and the position of the summary here, while it may be due to the original translator, or may even 80 back to the LXX’s Hebrew vorlage, is most definitely secondary: it results from a revision of the original Hebrew work. The evidence for this is simple: the introduction of the Jehoshaphat summary here at 16:28^a–h breaks the regular pattern of the record.

Normally the record proceeds for so long to give the history of Judah as its main subject with relevant details from the contemporary reigns in Israel woven into the history of Judah. Then it will change over and take the history of Israel as its main subject and weave into it details from the contemporary reigns in Judah. The change-over point is determined by chronology, and to illustrate what happens, let us construct a hypothetical example. Two kings of Judah, Judah A and Judah B, come to the throne within the reign of a king of Israel, Israel 1. Judah B lives a long time. During his reign Israel 1 dies and is succeeded by three successive kings, Israel 2, Israel 3 and Israel 4. The record at this point is taking Judah B as its main subject, though it records details of Israel 2, Israel 3 and Israel 4 as subsidiary matters. When Judah B eventually dies, his successor is named, but then the record switches to the kings of Israel as main subject. It goes back over the reigns of Israel 2, Israel 3 and Israel 4 and fills in any important details that have not already been told in connection with Judah B. For Israel 2 and Israel 3 these details tend naturally to be few and their reigns are more or less simply summarised. But for Israel 4, whose reign is perhaps not long begun, many details remain, and his story is now taken up as the main subject, while details from Judah C, Judah D etc. are interwoven, when relevant, as subsidiary matters.

Now let us see what has happened to the record of Ahab of Israel and Jehoshaphat of Judah, over whose chronology and placings the MT and the LXX disagree. According to the MT Ahab is the last of five kings of Israel to come to the throne during the long reign of Asa of Judah. When Asa, who has been the main subject, dies, the record switches to the Israelite kings. It goes back over four of these kings of Israel and summarises the remaining details of their reigns which have not already been told in connection with Asa, It then takes the fifth king, Ahab, and tells his story in full as the main subject, chapter 16:29 ff. Very early in his reign Asa of Judah died and was succeeded by Jehoshaphat, and since Jehoshaphat had a lot to do with Ahab, a major story from Jehoshaphat’s reign is recorded here, but recorded as material subsidiary to the main subject, which is Ahab. Now Ahab died before any other king came to the throne of Judah. So upon his death the record switches back to Judah to give it its turn as the main subject. Jehoshaphat, of course, heads the list, but little of his reign remains that has not been already told at length in connection with Ahab. So Jehoshaphat has now to be content with a brief summary of ten verses, chapter 22:41–51, and then is followed by his son, Ahaziah.

But in the LXX the chronology is different: Omri, not Ahab is the last king of Israel to come to the throne during the reign of Asa. When Ahab comes to the throne, Asa is already dead and Jehoshaphat is already king of Judah. But Omri of Israel did not reign very long, and Jehoshaphat of Judah, who ascended the throne during his reign, outlived him. Consequently, if the normal pattern of the record is to be maintained, when Omri of Israel dies, (chapter 16:28) the record ought to switch to the kings of Judah and make them the main subject. And this the LXX does, for immediately after Omri it introduces the story of Jehoshaphat, the Judaean king, chapter 16:28^a-h, and so appears to maintain the pattern. But just here it betrays itself. If it represented the original, free, Hebrew composition, it would now proceed to tell the story of Jehoshaphat as the main story. All the main incidents in his life that it intended to relate, would here be related, and, in particular, his important and colourful dealings with Ahab of Israel would be recorded, and they would be recorded from Jehoshaphat’s point of view as the main subject. But in fact, all that the LXX gives us here of Jehoshaphat’s reign is a brief summary, and, upon inspection, this turns out to be, with certain convenient alterations, precisely the same summary that the MT has in chapter 22! Now to have merely a summary in chapter 22 is understandable, since by this time the major incident of Jehoshaphat’s reign has already been told in connection with Ahab. But to have nothing but a summary in chapter 16, when nothing at all has yet been told about Jehoshaphat, and when by position his reign ought to be told in full as the main subject, is most inappropriate.

Furthermore, having dismissed Jehoshaphat’s reign in a brief summary when it ought to be the main subject, the LXX proceeds at 16:29 to tell the story of Ahab, and tells it quite rightly as the main subject, for it is now Israel’s turn to be the main subject. But in the course of it, it relates the major item of Jehoshaphat’s reign at length (his alliance with Ahab, ch. 22), and tells it as a main matter of Ahab’s story.¹⁸ This is simply because the LXX is following a Hebrew text very similar to the MT, which does likewise. But in the MT this is appropriate since Ahab comes to power before Jehoshaphat. But in the LXX, where Jehoshaphat comes to power before Ahab, it is again most inappropriate.

The final absurdity is this. The MT, having related at great length in chapter 22:1­–40 Jehoshaphat’s famous alliance with Ahab, makes a brief reference to it again in the summary at 22:45 ‘And Jehoshaphat made peace with the king of Israel’. This reference it then follows immediately with the remark, ‘And the rest of the acts of Jehoshaphat, and his might that he showed and how he warred, are they not written in the book of the chronicles of the kings of Judah?’ Now both the Greek translations of the summary follow a text that is basically the same as the MT, as can be seen from the following comparison:

MT 22:45–46	LXX 22:45–46	LXX 16:28c
וַיַּשְׁלֵם	καὶ εἰρήνευσεν	καὶ ἃ συνέθετο
יהוֹשָׁפָט	Ιωσαφαθ	Ιωσαφαθ
עּם־מֶלֶךְ יּשְׂרָאֵל׃	μετὰ βασιλέως Ισραηλ	—
וֶיֶתֶר דִּבְרֵי	καὶ τὰ λοιπὰ τῶν λόγων	—
יהוֹשָׁפָט	Ιωσαφαθ	—
וּגְבוּרָתוֹ	καὶ δυναστεῖαι αὐτοῦְ	καὶ πάσα δυναστεία
אֲשֶׁר־עשׂה	ὅσα ἐποίησεν	ἧν ἐποίησεν
וַאֲשֶׁר נִלְחָם֑		καὶ οὓς ἐπολέμησεν
הֲלֹא־הֵם	οὐκ ἰδοὺ ταῦτα	οὐκ ἰδοὺ ταῦτα
כְּתוּבִים עַל־סֵפֶר	ἐνγεγραμμένα ἐν βιβλίῳ	γεγραμμένα ἐν βιβλίῳ
דִּבְרֵי הַיָּמִים	λόγων τῶν ἡμερῶν	λόγων τῶν ἡμερῶν
לְמִלְכֵי יְהוּדָֽה׃	τῶν βασιλέων Ιουδα;	τῶν βασιλέων Ιουδα;

---

The large omission in 16:28^c is simply a case of parablepsis from 1° יהושפט to 2° יהושפט, which probably already existed in the Hebrew vorlage. At any rate, the translator has as a result construed the opening phrase as one of the several things that are mentioned in the book of Chronicles. But then ‘the agreements which J. made’—ἃ συνέθετο Ιωσ.—quite clearly refers (or at least it did originally) to J.’s agreements with Ahab, as in the MT and the other Greek summary; and this being so, the position of the summary in chapter 16 is extraordinarily inept. If the summary stood at the end of chapter 22, it would at least make sense, for having described Jehoshaphat’s agreement with Ahab at length in the early part of chapter 22, one might sensibly remark in a summary at the end of chapter 22 ‘a (further) account of Jehoshaphat’s agreements is given in the book of Chronicles’. But for 3 Reigns to say in its chapter 16 ‘an account of Jehoshaphat’s agreements (that is, with Ahab) is given in the book of Chronicles’, when it is itself about to give a detailed account of the agreement in its own chapter 22, is very odd indeed.

Everything, then, points to the fact that the position of the summary in chapter 16 does not represent the Hebrew original. It is a secondary alteration; and while in this position the summary fits in with the LXX’s peculiar chronology, it breaks the basic pattern of the narrative and shows itself ill adapted to its context.

From this two things stand out as particularly relevant to our main thesis. First, the removal of the summary to chapter 16 is part of a major revision, based on chronology, which has covered the whole of 3 Reigns from at least chapter 15 to chapter 22 and is also responsible for the duplicate translations of the introduction to Jehoram b. Ahab’s reign in 4 Reigns. Secondly, the fact that the removal of the summary to chapter 16 for the sake of the ‘revised’ chronology breaks the basic pattern of the narrative, vividly recalls the fact that, to secure its re-interpretation of Jeroboam’s character, the LXX has at the end of chapter 11 disrupted the normally invariable end-of-reign formula,¹⁹ by inserting into it material removed from chapter 12.

Chapter 16:29 — the end of the book

This section is almost completely taken up with the story of Ahab’s reign. In the LXX it is noteworthy for its whitewashing reinterpretation of Ahab’s character.²⁰ This re-interpretation shows exactly the same spirit, and uses exactly the same techniques, as the re-interpretations of Solomon and Jeroboam earlier in 3 Reigns. Like them it involves a large-scale re-ordering of the material, namely the transposition of chapters 20 and 21.

With this we have achieved part of our first objective which was to demonstrate that the several re-orderings of the material in 3 Reigns are not isolated phenomena, but are part of a deliberate scheme of re-ordering that covers the whole book and extends beyond it; and also to show that the reinterpretations of the leading characters in this book are bound up with this large-scale re-ordering of the material and come from the same hand. But it is one thing to demonstrate that all this re-interpretation and re-ordering does not represent the original Hebrew work, but results from a rather gauche re-writing of the original, and it is quite another thing to decide from where and at what level it entered the text of 3 Reigns.

We shall here briefly discuss four theoretical possibilities.

1. It was already to be found in the LXX’s vorlage and the original translators took it over.
2. The original translators used a text basically the same as the MT; but later revisers revised and re-ordered the Greek on the basis of a Hebrew text far different from the MT.
3. The original LXX translators themselves in the process of translation did all the re-ordering and re-interpretation without reference to any Hebrew text different from the MT.
4. Not the original translators, but subsequent revisers did the re-ordering and re-interpretation, but, again, without reference to any Hebrew text.

Of these four possibilities 3 and 4 are most unlikely. The second story of Jeroboam’s revolt, for instance, seems quite clearly to be founded on a Hebrew text. Moreover, the duplicate translations come from different hands; they cannot all be original Septuagint.

1., on the other hand, must be reckoned a decided possibility, particularly in the light of the situation in Jeremiah, of which F. M. Cross writes:

From Qumran comes a fragmentary Hebrew manuscript, which, where preserved, follows the short text of Jeremiah found hitherto only in Greek. In Chapter 10, for example, the Septuagint omits no fewer than four verses, and shifts the order²¹ of a fifth. The Qumran Jeremiah (4 Q Jer.^b) omits the four verses and shifts the order²² in identical fashion.

If in the LXX Jeremiah the differences in order were introduced by the original LXX translators working from a non-MT-type Hebrew MS, so they could have been in 3 Reigns. If they were, then we must regard the duplicate translations of Shimei Pt. I A (2:8–9), Jehoshaphat (22:41–and Jehoram (4 R. 3:1–3) whose positions agree with the MT’s positions, as having been filled in by some subsequent revision on the basis of some MT-type text, but done before Origen, since these duplicate translations themselves show signs of later Origenic revision. The duplicate translations Shimei Pt. I B (2:35^l-o), Jehoshaphat (16:28^a-h) and Jehoram (4 R. 1:18 ^a–d) would then be the original translations and in their original positions. And certainly these translations have been regarded by scholars as the older translations.

But here we meet a snag. The Miscellanies likewise contain translations which duplicate passages translated differently in the main text, and some of these Miscellany translations have likewise the appearance of being older than their duplicates in the main text. But surely their position in the LXX is not original. Consider, for instance, the list of duplicates on p. 8 above. If the translations in the Miscellanies are in the original LXX position, were there once gaps in the main text where their duplicates now stand? Moreover if there were originally gaps, and these gaps were filled in later, then they were not filled in on the basis of an MT-type text, as the other three duplicates may have been, for the LXX’s arrangement of the material in this section of the main text is very decidedly different from the MT’s.

In view of these difficulties, it may be that 2 has a better chance of being the correct explanation. According to it, the original translators worked from a Hebrew text that was in major respects basically the same as the MT. The older duplicate translations, now displaced, were by them originally positioned as in the MT. But subsequently the LXX suffered a good deal of re-interpretation on the basis of non-MT-type manuscripts, which involved re-positioning of blocks of material. Moreover, some parts of the old translation were considered mistaken (as indeed they are) and unsatisfactory. They were replaced by better translations. But the old translations were not scrapped; they were assembled, together with the debris resulting from the re-positionings involved in re-interpretation, and organised into two, as coherent as possible, groups, the Miscellanies. And then these Miscellanies were worked into a major reordering of the text, especially the text of Solomon’s reign, done to achieve a more logically ordered presentation of the facts.

But this explanation, too, has a snag: it can well account for the positioning of the older duplicate translations which now stand in the Miscellanies, but not so easily for the positioning of the older Jehoshaphat and Jehoram duplicates. The Miscellany duplicates, including Shimei Pt. I B, will have been banished to these convenient assembly-points by the person responsible for the present overall order of 3 Reigns. But the positioning of the older Jehoshaphat and Jehoram duplicates involves the acceptance of a scheme of chronology that conflicts with the scheme of chronology followed by the younger Jehoshaphat and Jehoram duplicates and by the book as a whole. Now if the introductory chronological phrases in the older duplicates were the work of the original translators, so must the position of these duplicates be. But if their present position is due to our theoretical reviser, he must have altered their original introductory phrases to fit their chronology to their present position. But can he possibly have followed one system of chronology with them and another system for the younger duplicates (which he himself probably translated)? Moreover, as I have elsewhere demonstrated, the editor responsible for the detailed order of 3 Reigns, shows himself a stickler for the most pedantically exact time-tabling possible.²³ Would he have followed two different systems of chronology?

In answer to this difficulty we may perhaps appeal to the two different and mutually contradictory accounts of Jeroboam, which we find standing side by side. As we have seen above, they are both conscious, deliberate, ‘reinterpretations’. It is surely no accident that they stand side by side, and yet their difference is no less than the difference between the two systems of chronology. But does this strengthen the possibility that one and the same editor could have been responsible for both, or emphasise its improbability?

However, it may be, whether 1 or 2 is the correct explanation, both imply some revision of the original Greek translation. The main material difference would be that according to 1 the midrashic re-interpretation of Solomon, Jeroboam and Ahab entered the Greek at the time of the original translation; according to 2 it entered later, perhaps a century or two later.

In light, then, of the fact that two translators have been involved in 3 Reigns at different times, it is tempting to cite as analogous Barthélemy’s καίγε recension.²⁴ If that recension’s peculiar features entered the βγ section of Reigns at the level of a revision of the Greek based on Hebrew manuscripts, might not the peculiar features of the γγ section of Reigns have entered the Greek at a similar level? But even here there is a significant difference. The καίγε recension was motivated by concern for halakic midrash, and was by definition interested in making the Greek conform to the minutiae of Hebrew Bible manuscripts. But the peculiar re-interpretations of the γγ section are more in the spirit of haggadic midrash, which, again by definition, does not limit itself to re-producing the Bible, or any other tradition, exactly, but allows itself at times a homilist’s inventive freedom and imagination.

Here, then, is no ground for dogmatism. Maybe none of our explanations is correct. Maybe the matter is far more complicated than we think. Obviously, there is need for more research, and the fact that it is important for us to know, if we can, at what level, and, therefore, at what time, the ‘re-interpretations’ of Scripture entered the textual tradition, makes this research the more urgent. But one thing stands out clearly. The re-interpretation and the re-ordering of the text have proved to be so extensive and systematic, that 3 Reigns cannot adequately be described as representing the Hebrew original plus a few targumic glosses. Rather we have in 3 Reigns a deliberate thoroughgoing, interpretative re-organisation of the original.

II

The foregoing observations obviously carry implications for the use of 3 Reigns as evidence for the reconstruction of the original Hebrew text of 1 Kings. They also raise the question of the re-interpreter’s attitude to Scripture.

The widespread re-ordering and re-interpretation certainly complicate the use of 3 Reigns as evidence for the original, but they do not make it impossible. Indeed, there are many places where the LXX is quite clearly founded on a Hebrew text that is self-evidently better than the MT.

We shall take just one very small and simple instance, representative of many others that could be quoted. At 3 Reigns 6:16 (MT 6:15) the MT, talking of the covering of the Temple walls on the inside with cedar planks, says that Solomon performed this operation ‘from the floor of the house unto the walls of the ceiling’. The phrase, ‘walls of the ceiling’, is obviously difficult. The LXX has ἀπὸ τοῦ ἐδάφους τοῦ οἴκου καὶ ἕως τῶν δοκῶν καὶ ἕως τῶν τοίχων—’both up to the rafters and up to the walls’. Here ἕως τῶν δοκῶν represents in part the עַד קִירוֹת הַסִּפֻּן of the MT; ἕως τῶν τοίχων represents a different Hebrew, עַד קוֹרוֹת. And the result is that the LXX has a doublet, both parts of which are based on a Hebrew text, but one of which—the one that disagrees with the MT—better represents the original than the MT does.

Nevertheless the knowledge that 3 Reigns is prone to re-interpretation and re-ordering will demand the greatest caution before one arrives at definite conclusions. Take the following example of how the possible presence of midrashic re-interpretation complicates decision.

In chapter 15:5 comparison is made between evil kings and David, ‘who did that which was right in the eyes of the Lord and did not turn from anything he had commanded him all the days of his life except in the affair of Uriah the Hittite’. Now this last phrase is absent from 3 Reigns. Commenting on this fact Burney²⁵ says, ‘LXX omits. The words may perhaps be a qualification inserted by a later hand.’ Montgomery²⁶ says, ‘except in the case of Uriah the Hittite: a unique moralizing judgment, and a late addition, absent in LXX Eth.’ Gray²⁷ says: ‘G omits the clause, which, indeed, has all the appearance of a later gloss’.

The unspoken implication in these comments is that because the LXX omits the clause, this is some support for the view that it is a late addition. Maybe! But we have seen in the case of Solomon, Jeroboam, and Ahab, that 3rd Reigns has been at great pains to whitewash their characters and, in so doing, it exhibits the kind of exegesis that we later find in the Talmud. Now it so happens that in the Babylonian Talmud (Shabbath 56_a_) there is a similar attempt to whitewash king David, in the course of which our phrase—except in the matter of Uriah the Hitrite—comes up for discussion.

R. Samuel b. Nahmani said in R Jonathan’s name, Whoever says that David sinned is merely erring, for it is said, And David behaved himself wisely in all his ways: and the Lord was with him. Is it possible that sin came to his hand?

Faced, then, with the plain statements about David’s sin with Bathsheba,

R. Samuel b Nahmani said in R. Jonathan’s name: Every one who went out in the wars of the house of David wrote a bill of divorcement for his wife . . .’ so that in taking Bathsheba, David was not guilty of adultery! To counter this, apparently, Rab said ‘When you examine the life of David, you find nought “save only in the matter of Uriah the Hittite” . . . but the discussion was inconclusive.

And no wonder!!

In view of this we can no longer be so certain that the absence of the phrase ‘save only in the matter of Uriah the Hittite’ from 3 Reigns is necessarily evidence that this phrase is a late moralizing addition in the MT. It could be that its absence from the LXX is the result of deliberate excision in the interests of the whitewashing exegesis. Even using Montgomery’s ‘moralizing’ criterion, we cannot be sure that the original story saw nothing wrong with the Bathsheba incident, or regarded it as trivial, and that a later ‘moralizing’ age, with stricter standards, felt obliged to add the qualification—’but he did sin with Bathsheba’. It is possible, it is even perhaps more likely, that the original realistically admitted David’s sin, and a later age, concerned for the reputation of the great king and saint, excised the mention of his sin, in an attempt to whitewash him.

Next we may notice an example of how awareness that 3 Reigns abounds both with midrashic re-interpretation and with secondary changes of order complicates the use of its evidence for reconstructing the original Hebrew. In 8:12 the MT has:

Then spake Solomon,

The Lord has said that He would dwell in the thick darkness.

I surely have built Thee an house of habitation,

A place for Thee to dwell in for ever.

Now the LXX not only has an enlarged version of this, but it has it in a different position, 8:53a, and equipped with an introductory time-note and a citation of its source. That the enlarged version is founded on a Hebrew text is practically certain; but whether that Hebrew text represents the original text is not quite so certain. And when we remember the many changes of order which 3 Reigns everywhere shows, and its characteristic concern for time-tabling, we may well feel that the LXX’s evidence on this point only increases the uncertainty. Here, at any rate is the LXX’s version:

Τότε ἐλάλησεν Σαλωμων ὑπὲρ τοῦ οἴκου, ὡς συνετέλεσεν τοῦ οἰκοδομῆσαι

αὐτόν Ἥλιον ἐγνώρισεν ἐν οὐρανῷ Κύριος.

εἶπεν τοῦ κατοικεῖν ἐν γνόφου.

Οἰκοδόμησον οἶκόν μου, οἶκον ἐκπρεπῆ σαυτῷ,

τοῦ κατοικεῖν ἐπὶ καινότητος.

οὐκ ἰδοὺ αὕτη γέγραπται ἐν βιβλίῳ τῆς ᾠδῆς;

From this St. John Thackeray²⁸ attempted a reconstruction of the Hebrew and translated it to read:

Sun, (the) Glory clouds the heavens,

Jahwe hath promised to dwell without the thick darkness: (and then, the singer addressing Jahwe)

‘Build Thou my house, a celestial Palace for Thyself.’

For the Sabbath On Alamoth (i.e. for soprano voices).

Here, he concluded, we had a very early piece of Israelitic folklore, in which the speaker regards the sun as a Deity and addresses him, on the occasion of an eclipse. He then explained the shortened version of the song in the MT as deliberate manipulation, aimed at suppressing the reference to the sun, and to the Book of Jashar, which may have been suspected of unorthodoxy, and here came perilously near to sun-worship.

At the other extreme from Thackeray stands Montgomery,²⁹ who completely rejects all theories regarding the sun-cult in Jerusalem, and sees in the Greek ‘a fine theological contrast’ between the so evident sun and the Deity who will not be seen, On this theological ground he supposes the fragment in the Greek to be original, and asks whether it was not cautious censorship which deleted from the Hebrew the first hemistich, preserved in the Greek.

With him, at least in this theological judgement, stands Prof. John Gray³⁰ who comments

This poetic fragment, which by its very nature we take as early, indicates the foundation of the Temple and the installation of the ark in the inmost shrine as the symbol of an important religious truth. The God of Israel is at once the Creator of the universe, who has ‘established the sun in the heavens’, and the God of Israel who has condescended to fix His throne in the midst of His people Israel, but whose nearness and readily-experienced power and grace never exhaust His revelation (cf. v. 27). He dwells still ‘in thick darkness’ (cf. Pss. 18:12; 97:2) infinitely greater and more mysterious than the sum of this revelation. He is mysterium tremendum et fascinans.

This, theologically, is a very far cry from Thackeray’s interpretation, which would make of Solomon a sun-worshipper, and to me, at least, it is much more convincing. But, theology apart, from the point of view of establishing the original text, it is to be noted that both Thackeray and Gray put forward each his own reconstruction of the Hebrew on the basis of the Greek, as indeed have most of those who have pronounced upon the matter. This in itself shows what an uncertain procedure such reconstructions are. Nor is it cynicism or obscurantism to spell out the uncertainties in full:

1. The Greek text itself is uncertain e.g.
2. ἔστησεν/ἐγνώρισεν. ἐκ γνόφου/ἐν γνόφῳ.One has to decide whether the Greek text is a true or a mistaken translation: e.g. ἐπὶ καινότητος

a. Montgomery = בַחֲנֻכָּה = At the dedication. i.e. it is not part of the prayer, but a note appended to it.

b. Thackeray = עַל־עֲלָמוֹת = for the soprano voices —misread as עַל־עֲלוּמִים = ‘youthful vigour’ by the LXX; and misread as עַוֹלָמִים ‘for ever’ by the MT.

c. Gray, the RSV and others think that the MT’s עַוֹלָמִים ‘for ever’ original.

1. τοῦ κατοικεῖν = לְשִׁבְתְּךָ or, with Thackeray, לַשַּׁבַת ‘for the Sabbath.’
2. The source-book which the LXX names: is it סֵפֶר־השִּׁיר or סֵפֶר־הַיָשָּׁר which is mentioned elsewhere in the OT?

But suppose we could agree on a reconstruction of the Hebrew of the song, we are still left with two major problems: (1) the introductory note, and the citation of the source; (2) the difference of position. In the light of this Thackeray decided that neither the MT nor the LXX represented the original text, but that the whole passage was an ancient gloss which eventually entered the Hebrew text, was incorporated into the LXX translation and there fully preserved under a clumsy translation, but in the MT was manipulated and shortened for dogmatic reasons.

The Revised Standard Version, on the other hand, accepts the LXX’s additional first line, though it follows the MT for the rest. It seems to regard the MT’s positioning as original, and dismisses the LXX’s introductory phrase and its citation of the source as secondary. It is perhaps rather bold of the Revised Standard Version to commit itself like this. (Certainly it will be interesting to see what the New English Bible does with the passage). On the other hand it may well be right in its view that the LXX’s order, even if it goes back to a Hebrew text, is not original. In the first place there is something very odd about the content as well as the position of the introductory phrase, ‘then spake Solomon concerning the house when he had finished building It’. After all, we have already been told that the building was long since finished: verse 1 of chapter 8, in one of the LXX’s peculiar plusses, has been particularly concerned to tell us: ‘And it came to pass when Solomon had finished building the Lord’s house and his own house after twenty years, then king Solomon assembled all the elders of Israel’. Moreover, we have by this stage had the introduction of the ark, the descent of the Glory, Solomon’s first blessing of the people, and Solomon’s long prayer, verses 22–53. Verse 54 is about to say, most naturally, ‘and when Solomon had finished praying’, he did something else. But before it can do so, we get not only the Song, but the strange ‘time-note’ ‘when he had finished building’. It seems totally out of place.

It could be that considerations of piety lie behind the LXX’s order here, just as they do behind the order of the Temple and Palace sections. The MT, in placing the Song in verse 12, achieves the following consecution: verses 10–11 the Glory descends, verses 12–13 Solomon quotes the Song, and then, verse 14, he turns his face away . . . Piety might well have objected that for Solomon to remain looking at the Glory-filled temple, even just long enough to quote the Song, would have been the height of irreverence. The absence of the Song, as in the LXX, gives, of course, an unexceptionable order:

καὶ οὐκ ἠδύναντο οἱ ἱερεῖς στῆκειν λειτουργεῖν ἀπὸ προσώπου τῆς νεφέλης, ὅτι ἔπλησεν δόξα Κυρίου τὸν οἶκον. καὶ ἀπέστρεψεν ὁ βασιλεὺς τὸ πρόσωπον αὐτο . . . .

If, then, the Song was moved for these reason from verse 12 to the much later position at verse 53a, this would account for its having to be supplied with an introductory time-table note. At the same time it would proclaim the LXX’s order here secondary, as it is in the Temple-Palace section.

Moreover, motives of piety may also lie not only behind the position of the Song in the LXX, but also behind its peculiar reading in line 3 οἰκοδόμησον οἶκόν μου, οἶκον ἐκπρεπῆ σαυτῷ. As distinct from this 2nd pers. sing. imperat., ‘Build Thou my house’, the MT has the first person singular perfect בָּנֹה בָנִיתִי בֵּית זְבֻל לָֽךְ ‘I have surely built Thee a house . . .’ Now in the Midrash on Psalms,³¹ at 24:10, we are told that Solomon had a great deal of trouble in getting the gates of the Temple to open, when he wanted to bring in the Ark. The question is then asked ‘Why was Solomon put to such trouble? Because he was arrogant, having said: I have surely built Thee a house of habitation’ (1 Kgs 8:13). But what, we may ask, was there arrogant about this? Another comment a little further on makes it clear: ‘Another comment on Solomon’s saying I have surely built Thee a house of habitation (1 Kgs 8:13). R. Jacob, the son of R. Judah bar Ezekiel, said: The literal meaning of these words is: ‘I have built Thee a house, a house, indeed, which built itself’. How R. Jacob extracted this meaning out of the text we need not stay to consider, nor is the late date of this tradition particularly important for our purposes. What is relevant is the evidence that some people felt that the verse as it stood put in Solomon’s mouth an objectionably arrogant claim that he, a mere man, had built God a house. Exegesis, therefore, had in the interests of piety, to show that Solomon did not in fact build the house—it built itself. Now the LXX’s text does not indulge in the fancy that the temple built itself, but it does avoid all semblance of arrogance, for instead of representing Solomon as claiming to be the builder, it has him humbly praying to God to build the house. Clearly a concern for piety has been at work here too, and this means that the LXX represents a deliberate, secondary, exegetical reworking of the original.

It may well be, then, that the LXX in its first verse of the Song has preserved a line of the original, which the MT has lost. For the rest, it is clear that the LXX has suffered as much, if not more, secondary influence than the MT, and amid so much uncertainty it were, perhaps, prudent to suspend judgment and learn a lesson from Psalm 151.³² While everyone knew that it must have been based on a Hebrew text, no-one could really have suspected before the Hebrew texts turned up, how very complicated the textual history would prove to be. Indeed, even now they have turned up, full agreement on the topic has not yet been reached.

We come finally to what is perhaps the most important question raised by the observations in the first part of this paper: if the re-interpretation and re-ordering found in 3 Reigns are based on a Hebrew text, how should we regard that text? Indeed, what did the original translator (or subsequent reviser, if the re-interpretation and re-ordering entered the Greek text at the level of a revision based on a Hebrew text) think he was translating? Or, more fundamentally still, what did the compiler of the Hebrew text from which the Greek has taken its re-interpretation think his text was? Bible? Targum? Midrash?

To attempt a full answer to this question would lead us into many areas where I have no competence. I must, therefore, be content to ask, rather than answer, the question. But we can, perhaps, contribute towards an answer by adducing as evidence the methods employed by whoever was responsible for the reinterpretation.

The main difficulty is to account for the extremely different elements presented by 3 Reigns. On the one hand, the Greek for long stretches agrees with the MT very closely, and many of its differences are readily explainable as having come from Hebrew Biblical manuscripts belonging to text-traditions differing from the MT. On the other hand, the whitewashing re-interpretation of Ahab has so much in common with R. Levi’s re-interpretation of Ahab, recorded in the Jerusalem Talmud,³³ that it is difficult not to think that it comes from a similar source. And certainly R. Levi’s favourable re-interpretation of Ahab was not based on the discovery of some Biblical manuscript of a non-MT-type, still less on some extra-Biblical historical source. It was totally a matter of exegesis, dictated by theological considerations and achieved by shifting the weight of homiletic emphasis from one phrase of the Biblical text to another, all the while employing the same text. Similarly, the re-ordering of the Ahab chapters in 3 Reigns so that his doom does not follow immediately upon his repentance, but his repentance (ch. 20) is granted a period of respite, in which nevertheless he sins again (ch. 21) and only after that meets his doom (ch. 22), has a good deal in common with the view expressed in the Babylonian Talmud:

The repentance of the confirmed sinner delays punishment, even though the decree of punishment for him had been signed already. The careless ease of the wicked ends in calamity . . . It is not good to respect the person of the wicked. It is not good for the wicked that they are being favoured [by the Holy One, blessed be He] in this world. It was not good for Ahab that he was favoured in this world as it is said: Because he humbled himself before Me, I will not bring the evil in his days.

But if in fact the LXX’s re-interpretation and re-ordering of the Ahab story do belong to this theological and expositional level, this suggests that the Hebrew text on which 3 Reigns is based, was more of a very free targum or midrash than a straight-forward Biblical text.

Now the interest in Ahab’s repentance shown by 3 Reigns, the Jerusalem Talmud, the Babylonian Talmud, and works like the Pesikta de Rav Kahana, arose in the course of the general development of doctrinal theology among the Jewish rabbis. The fact that one re-interpretation is attributed to R. Levi³⁴ does not, perhaps, necessarily mean that he originated it. He might merely have given fresh impetus, or more vivid expression, to an idea that had already had widespread existence in oral tradition. If, then, the re-interpretations in 3 Reigns entered the text at the stage of a revision of the original translation, it is possible that this revision, too, was based not only on written Hebrew sources, but in part also on oral, exegetical traditions.

But here is where the dating of these features of re-interpretation and reordering in 3 Reigns becomes important. The later the date at which these features entered the text, the nearer we come to the time when according to the Talmud, the Jews understood quite clearly the difference between Bible and targum (let alone between Bible and midrash). The earlier the date, the less external evidence there is to show that the Jews would have understood any such difference. If then, these features entered 3 Reigns at the level of the original translators, it might perhaps be argued that the translators would have regarded the text from which they worked—supposing they had so much as noticed how thoroughly it re-interprets the original story—as having exactly the same status and authority as the original story itself—just as they would have seen no essential difference between Chronicles and Samuel–Kings. After all, there is a certain similarity between what the Chronicler has done with Samuel—Kings and what the author of 3 Reigns’ vorlage has done with 1 Kings. The Chronicler for long stretches follows closely the text of Samuel–Kings; but he also re-phrases it from time to time, omits parts, and adds other stories, and generally re-tells the history from a noticeably different point of view from that adopted in Samuel–Kings. Yet, Chronicles was accepted and eventually came to be regarded as Bible equally with Sam.–Kings. Could we not argue the same for the vorlage of 3 Reigns by analogy?

As I have said, I do not know enough of these matters to say whether there is any available external evidence to confirm or deny this suggestion. But the internal evidence of 3 Reigns goes very strongly against the suggestion and shows that 3 Reigns’ vorlage was built up on very different principles from Chronicles. Chronicles though it follows Sam.–Kings very closely in parts, is in its entirety a free, independent work, in the same way as Matthew and Luke, though they share a good deal of common material, are free, independent compositions. Luke is not a commentary on the text of Matthew, and Chronicles is not a commentary on the text of Samuel–Kings. But 3 Reigns is quite obviously a commentary on the text of 1 Kings. This can be seen in three ways:

1. At 20:27, 3 Reigns remarks ‘Now concerning the statement that Ahab was pricked in his heart before the Lord’, καὶ ὑπὲρ τοῦ λόγου, ὡς κατενύγη Αχααβ ἀπὸ προσώπου τοῦ κυρίου. But the statement referred to does not occur till verse 29, though there it occurs exactly as quoted in verse 27, ὡς κατενύγη Αχααβ ἀπὸ προσώπου μου, the only difference, μου/τοῦ κυρίου, being required to adapt the direct statement of verse 29 to the reported speech of verse 27. This means that verse 27 is a comment on verse 29, and, curiously enough, a comment made in advance. This certainly is not the manner of a free composition; but it is understandable in a commentary.
2. In its first story of Jeroboam’s revolt 3 Reigns achieves its re-interpretation of Jeroboam’s character by removing material from the beginning of chapter 12 and placing it in chapter 11. But in chapter 11 this material can do its job only by standing in a most awkward place right in the middle of the stock formula for the end of a king’s reign, where it leads to repetitiousness and altogether disturbs a formula which everywhere else is invariable.³⁵ This, we may be confident, would not have happened if 3 Reigns’ vorlage had been a free composition in the manner of Chronicles. But the compiler of that vorlage was clearly not attempting a free re-telling of the story. He was merely at pains to interpret the text of 1 Kings, by re-arranging it slightly—and very awkwardly—to say what he thought 1 Kings meant—or ought to mean.
3. Following the first story of Jeroboam’s revolt there stands in 3 Reigns, as we have seen, a second story that takes the completely opposite view of things from the first story. Chronicles never does this sort of thing. It may give a different slant on, say, Manasseh’s character from that which 2 Kings does, and record a repentance on Manasseh’s part on which Kings is silent (2 Chr 33). But it does not immediately follow this ‘favourable’ account of Manasseh with a directly ‘unfavourable’ account, calculated to contradict the main point which the ‘favourable’ account is making. Of course, it does not, nor does any Biblical book anywhere do such a thing. But it is in the very nature of midrash to do this, as, given a Biblical text, it seeks to bring out different possible interpretations, invents haggadoth to illustrate different homiletic points, and records them happily side by side, even if the haggadoth differ among themselves irreconcilably.

But commentary of this kind, as distinct from free re-writing of a still fluid tradition, pre-supposes an authoritative Biblical record, which is the starting point and inspiration for the homilist, and which it is the duty of the commentator to elucidate and expound.

Third Reigns, then, and its Hebrew Vorlage can best be described in the present, limited state of our knowledge as mongrels: part Bible-text, part haggadic midrash; to which mixture is added in 3 Reigns the element of targumic interpretation which inevitably enters into the translation of a Biblical text into another language, and which has entered 3 Reigns at least twice, first at the level of the original translation, and secondly at the level of the duplicate translations and their incorporation into the text.

(Since this article was sent to press, Dr Shenkel’s book has been published (Harvard University Press, 1968). For my review of it see JTS, October, 1969.)

Footnotes

¹ This paper represents an expanded summary of the Grinfield Lectures on the Septuagint in the University of Oxford for the year 1967­–8. An abridged version of it was read as a Tyndale Lecture in Cambridge on December 1st, 1967, and again as a Faculty of Theology Lecture in the University of Manchester on March 12th, 1968. I am indebted to those who on each occasion offered constructive criticisms and observations.

² Dr Shenkel is one of Prof. F. M. Cross’s pupils. He has hopes of publishing shortly a revised edition of his Harvard dissertation, Chronology and Recensional Development in the Greek Text of Kings, 1964. In view of this, the original dissertation may not be quoted.

³ HTR (1964) 287.

⁴ See Rahlfs, Septuaginta-Studien, 3, 265–7; Montgomery, ZAW, 50 (1932) 124­–9.

⁵ These two strange collections, which stand one at 3 R. 2:35^a–k and the other at 2:46^a–l , have been conveniently dubbed ‘Miscellanies’ by Montgomery. In addition to duplicate translations they contain translations of verses which occur in various places in the MT but have no counterpart in the corresponding places in the LXX, and also some editorial comments.

⁶ In an article entitled, ‘The Shimei Duplicate and its Satellite Miscellanies in 3 Reigns II’ (JSS, May, 1968).

⁷ Misc. 1 begins καὶ ἔδωκεν Κύριος φρόνησιν τῷ Σαλωμων καὶ σοφίαν πολλὴν σφόδρα καὶ πλάτος καρδίας ὡς ἡ ἄμμος ἡ παρὰ τὴν θάλασσαν. Misc. 2 begins καὶ ἦν ὁ βασιλεὺς Σαλωμων φρόνιμος σφόδρα καὶ σοφός· καὶ Ιουδα καὶ Ισραηλ πολλοὶ σφόδρα ὡς ἡ ἄμμος ἡ ἐπὶ τῆς θαλάσσης εἰς πλῆθος. . . . Notice also the repeated use of the sand-of-the-seashore simile, and for its significance here see Midrash Rabbah, Numbers, translated by J. J. Slotki (Soncino Press, 1951) 749.

⁸ ὅτι ἀνὴρ φρόνιμος σύ, καὶ γνώσῃ ἃ ποιήσεις αὐτῷ’ 2:35°.

⁹ ZAW 50 (1932) 124–9.

¹⁰ In ‘The Shimei Duplicate . . .’ See also VT XV (1965) 154–6.

¹¹ Babylonian Talmud, Sanhedrin, 104b.

¹² Owing to a drastic re-arrangement of the text, the LXX’s equivalent, if it had one, of the MT’s 9:24a would stand in chapter 10 between verses 25 and 26. But there the LXX has no equivalent. What it does have is the near equivalent at 9:9, and a more exact equivalent at 2:35 f.

¹³ VT XV (1965) 331­–4.

¹⁴ BT, Sanhedrin, 71b.

¹⁵ VT XVII (1967) 173–189.

¹⁶ Verses 2 and 3a are omitted from chapter 12; but the greater part of verse 2, together with some other material, is placed in the middle of 11:43.

¹⁷ VT XVII (1967) 173–189.

¹⁸ For this item told from Jehoshaphat’s side as a major item in his reign, see 2 Chronicles 18.

¹⁹ See VT XVII (1967) 178–9.

²⁰ See ZAW 76 (1964) 269–80.

²¹ The Ancient Library of Qumrân, Duckworth (London 1958) 139.

²² Italics mine.

²³ VT XV (1965) 153 ff.

²⁴ Les Devanciers d’Aquilla (Brill, Leiden 1963).

²⁵ Notes on the Hebrew Text of the Books of Kings (Oxford 1903) 196.

²⁶ ICC Kings, T. & T. Clark (Edinburgh 1951) 274,

²⁷ I & II Kings (S.C.M. Press, London 1964) 316 n.e.

²⁸ JTS 11 (1910) 525.

²⁹ ICC Kings, 190 f.

³⁰ I & II Kings, 197.

³¹ Translated by W. G. Braude, Yale Judaica Series, Vol. XIII: 1.

³² See J. A. Sanders, The Psalms Scroll of Qumrân Cave 11 (Oxford 1965) 54 ff.

³³ San. 51a.

³⁴ Yoma 87a.

³⁵ 27. VT XVII (1967) 179.

---