An Impossible Shrine

This article was originally published in Vetus Testamentum 15:4 (1965), pp. 405–420.

The Masoretic Text of those chapters in the Third Book of Reigns which describe the temple is certainly corrupt in places, and we eagerly await any light that Qumran may shed upon it. Meanwhile we can say with even more certainty that however bad the MT is, the LXX is far worse.

Nowadays it is quite a worn out sermon that the analysis of the Greek texts is ‘a study which must be rigorously pursued and for its own sake first, before the Greek texts are applied to the criticism of the Hebrew’;¹ but in the days when the sermon was needed, none preached it more vigorously or more effectively than the late J. A. Montgomery. Although, therefore, in analysing a part of the LXX temple-chapters this article will call in question one of Montgomery’s own text-analyses, the present writer dares to hope that Montgomery himself would have approved of its methods, if not of its conclusions.

The passage to be dealt with, 3 Reigns 6:16–22 (MT), 17–21 (LXX), is the one which describes the building, decoration, function and dimensions of the debir, the innermost compartment of the temple, together with the making and ornamentation of the altar of incense which stood outside, but close to, the debir. The major textual differences will become at once apparent if the LXX version is placed beside the MT:

MT LXX v.16 And he built twenty cubits on the hinder part of the house with boards of cedar, from the floor unto the walls; he even built them for it within for an oracle, even for the most holy place. v.17 καὶ ᾠκοδόμησεν τοὺς εἴκοσι πήχεις ἀπʼ ἄκρου τοῦ τοίχκου, τὸ πλευρὸν τὸ ἓν ἀπὸ τοῦ ἐδάφους ἕως τῶν δοκῶν· καὶ ἐποίησεν (αὐτῷ) ἐκ τοῦ δαβειρ εἰς τὸ ἅγιον τῶν ἁγίων. v.17 And of forty cubits, was the house that is the temple before it (לפני :? לפניו) v.18 καὶ τεσσαράκοντα πηχῶν ἦν ὁ ναὸς κατὰ πρόσωπον τοῦ δαβιρ v.18 And there was cedar on the house within, carved with knops and open flowers: all was cedar; there was no stone seen. v.19 And an oracle in the midst of the house within he prepared to set there the ark of the covenant of the Lord. ἐν μέσῳ τοῦ οἴκου ἔσωθεν, δοῦναι ἐκεῖ τὴν κιβωτὸν διαθήκης Κυρίου. v.20 And before the oracle twenty cubits length, and twenty cubits breadth, and twenty cubits the height thereof; and he overlaid it with pure gold. And he covered the altar with cedar. v.19 εἴκοσι πήχεις μῆκος καὶ εἴκοσι πήχεις πλάτος καὶ εἴκοσι πήχεις τὸ ὕψος αὐτοῦ· v.20 καὶ περιέσχεν αὐτὸν χρυσίῳ συγκεκλεισμένῳ. καὶ ἐποίησεν θυσιαστήριον v.21 So Solomon overlaid the house within with pure gold and he drew chains of gold across before the oracle; and he overlaid it with gold. v.21 κατὰ πρόσωπον τοῦ δαβειρ. καὶ περιέσχεν αὐτὸ(ν) χρυσίῳ ²² καὶ ὅλον τὸν οἶκον περιέσχεν χρυσίῳ ἕως συντελείας παντὸς τοῦ οἴκου. v.22 And the whole house he overlaid with gold, until the completion of all the house; also the whole altar that belonged to the oracle he overlaid with gold.

The first noticeable feature is that the MT has much more material than the LXX, and scholars have been generally inclined to regard this as evidence that several glosses have intruded into the MT, and that the LXX, although by no means perfect, represents an earlier and better stage in the transmission of the text. Burney commented: ²

The passage as it stands is remarkably involved and appears to exhibit a double stratum of glosses . . . . . Here we notice the omission of הוא ההיכל, also lacking in Vulg., explanatory of הבית in v. 17; and the entire absence of v. 18, which contains details of the wood-carving of the house. These clearly are insertions made by R ^P . . . . But the account, even as simplified by LXX, cannot stand in its original form . . .

Montgomery’s view is similar; his analysis is worth quoting in full: ³

“15. And he built the walls of the house on the inside with cedar planks from the floor of the house to the ⸢beams⸣ [with OGrr.; Heb. walls] of the roof; he panelled with wood within, and he laid the floor of the house with cypress planks. 18. And cedar for the house on the inside, carved work of gourds and flower-calyxes; ⸢the whole was cedar, no stone was seen⸣ [OGrr. om]. 16. And he built off 20 cubits at the rear of the house with cedar planks from the floor up to the ⸢beams⸣ [with OGrr.; Heb. walls], and he built within [with correction of Heb.] for a shrine [Heb. + for the holy of holies]; 17. and 40 cubits [long] was [Heb. + the house, that is; OGrr. om] the hall ⸢in front of⸣ [with correction of Kr] ⸢the shrine⸣ [plus with Grr., Vulg.]; 20. and ⸢the shrine⸣ [with Vulg.; Heb. in front of the shrine] 20 cubits in length, and 20 cubits in width, and 20 cubits its height; 19. And a shrine within the house, deep within, he prepared, to set there the ark of the covenant of YHWH.__and overlaid it with refined gold. And he ⸢made⸣ [with Grr.; Heb. overlaid] an altar of cedar (21b.) in front of the shrine, and overlaid it with gold. 21a. And Solomon overlaid the house within with refined gold, and he drew chains of gold across. 22. And all the house he overlaid with gold, until at last the house was finished. ⸢And all the altar that belonged to the shrine he overlaid with gold⸣ [OGrr. om].

The above display presents in the second column a number of extensive additions that have been interpolated in the text, as also many glosses to the earlier form in the first column. The criticism is largely supported by the OGr. texts, and may in general explain itself. V. 19 parallels v. 16, setting forth the shrine as the depository of the ark. The plus of ‘the holy of holies, (v. 16—the Semitic = the holiest) is a current term peculiar to Ρ in the Pentateuch and to the latest Biblical books. Vv. 21a, 22a are wondrously extravagant with the gold-plating of the whole house. The original specifications concerned the house as a whole; cf. ‘the altar in front of the shrine’ (vv. 20, 21b) and the later item of ‘the altar of the shrine’ (v. 22b).

So far Montgomery.

In this analysis higher and lower criticism are closely intertwined. The analysis is, of course, based primarily on higher critical considerations: verse 19 is regarded as a late addition largely because it appears to parallel verse 16; verses 21a, 22a are excised as late intrusions because ‘they are wondrously extravagant with the gold-plating of the whole house’, and because, as Montgomery says later

The gilding of the furnishings, as of the altar, is reasonable, but not that of the whole interior . . . . Such extravagant description appears to be a step forward in the process of exuberant imagination, continued by the Chronicler . . . (p. 152)

But Montgomery does claim that these results of the higher criticism are largely supported by the OGr. texts. It will be interesting, therefore, before we turn to investigating the LXX text for its own sake, to trace in detail what relationship the LXX does in fact bear to Montgomery’s analysis.

First there are a number of places where the LXX reading seems to make better sense than the MT: verse 15, ‘beams’ instead of ‘walls’, and the same again in verse 16; verse 17, the addition of the words ‘the shrine’, completing the prepositional phrase ‘in front of; and at the end of verse 20 the verb ‘made’ instead of the verb ‘overlaid’. These LXX readings Montgomery prefers as being more likely to represent the original Hebrew and he bases his translation on them—understandably and with good reason. But then comes the question of the LXX’s attitude to the verses which Montgomery would excise as late insertions; we get the following picture:

1. Verse 18 is excised as an intrusion which ‘early disturbed the text’ (p. 154) and it is placed opposite verse 15 in Montgomery’s arrangement of the text (p. 149: see my page 6), presumably to suggest that it is a doublet of verse 15. Then it is observed that the LXX omits the verse ⁴ and this omission is regarded as confirmation that the verse was an early intrusion in the MT, not found in the LXX’s vorlage.
2. Verse 19 is excised, too, as an intrusion which ‘early disturbed the text’ (p. 154) and which parallels verse 16; but the LXX, contrarily enough, contains the whole of this verse (at the end of its v. 18) except the words, ‘and a shrine he prepared’.
3. The phrase ‘for the holy of holies’ (v. 16) is regarded as a plus, attributable to a redactor of the Ρ school, and therefore deserving of excision. The LXX has the phrase.
4. Verse 21a is omitted because it is ‘wondrously extravagant with the gold-plating’; and certainly the LXX omits this part of the verse. But whether its omission supports the view that verse 21a in the MT is a late, extravagant addition is very doubtful: verse 22a is the most extravagant of all—it talks of overlaying the whole of the inside of the temple with gold—and yet the LXX has this verse (its v. 21).
5. Verse 22b says ‘And all the altar that belonged to the shrine he overlaid with gold’. Montgomery excises it, not because he thinks that it is too extravagant with the gold-plating (for on p. 152 he says that the gilding of the altar is reasonable), but because, when verse 21b is allowed to follow immediately upon verse 20b, as Montgomery thinks it did originally, we get a verse—‘And he made an altar of cedar in front of the shrine, and overlaid it with gold’—of which verse 22b—‘And all the altar that belonged to the shrine he overlaid with gold’—then seems to be a doublet or an unnecessary repetition. Now the LXX does in fact omit MT’s verse 22b; and, what is more, it has no equivalent of MT’s verse 21a. Its own verse 21b represents a running together of MT verse 20b with verse 21b. Montgomery notices the fact and presumably regards these omissions as supporting his analysis of the Hebrew. But things are not quite so simple. Admittedly, as the Greek now stands, its verse 20b, καὶ έποίησεν θυσιαστήριον κατὰ πρόσωπον τοῦ δαβειρ καὶ περιέσχεν αὐτὸ χρυσίῳ seems to be a homogeneous whole, all referring to the incense altar. And, moreover, being understood in this way, it may well have led to the omission of what now seems a superfluous doublet, MT 22b, LXX 21b. But when the LXX verses 20 and 21 are laid out carefully beside the MT (see above, p. 4) it begins to look as if the LXX’s positioning of the words κατὰ πρόσωπον τοῦ δαβειρ καὶ περιέσχεν αὐτὸ χρυσίω immediately after the words καὶ έποίησεν θυσιαστήριον is due to nothing more than an accidental omission from the Greek of the words ‘with cedar. So Solomon overlaid the house within with pure gold and he drew chains of gold across’. At least, prima facie it is just as likely (perhaps even a bit more likely) that these words should have fallen out of the Greek, as that they should have been inserted in the Hebrew.

In view of all this there is surely something very unsatisfactory in taking the LXX in this piecemeal fashion, claiming the support of those pieces which seem to favour a given analysis of the Hebrew text and passing over other pieces which do not support it. One can, of course, plead that neither the MT nor the LXX is everywhere homogeneous, but each has secondary strata in its text; and that therefore it is legitimate to pick out individual secondary strata in the LXX and to use their presence or absence as confirmatory evidence of this or that source-analysis of the Heb. text, without necessarily investigating the context in the LXX in which they are embedded, and without necessarily expecting the adjacent strata to be contemporaneous or to supply consistent, corroboratory evidence. But this is a hazardous procedure, as we shall see when we now turn to investigate the Greek on its own account.

II

The first thing we notice, when we read the Greek through as a connected whole, is that in three places it makes factual nonsense. The most glaring instance is its verse 18 (see above p. 3): ‘And of 40 cubits was the temple (naos) in front of the debir in the midst of the house within, to set there the ark of the covenant of the Lord’. As this verse stands it tells us two things:

1. that the naos in front of the debir was 40 cubits (that is, in length), which is quite correct. The overall length of the temple was 60 cubits (see MT 6:2); it was divided into two parts: the first part, the naos in front of the debir, was 40 cubits, the second part, the debir itself, was 20 cubits.
2. that the purpose of this naos was ‘to put there the ark’. This, of course, is nonsense; the ark was put in the debir as both the MT and the LXX of 8:6 explicitly say ‘And the priests brought in the ark . . . into its place, into the debir of the house, into the holy of holies, under the wing the cherubim’. But nothing else can be made of the present grammatical structure of LXX verse 18: ‘the naos . . . was 40 cubits . . . to set there the ark’. No-one, who did not know his Hebrew or the temple-plans well, would take the Greek to mean that the ark was to be put in the debir.

Next let us scrutinise carefully exactly what the Greek of LXX verse 19 says: Twenty cubits the length twenty cubits the height of it. To what does the ‘it’ refer? If grammar is allowed to decide, the answer must be that it refers to the subject of the previous sentence, i.e. the naos. Only a very strained interpretation could construe it as referring to the debir. And yet, in fact, the dimensions 20 x 20 x 20 cubits are the dimensions of the debir and not of the naos. Indeed, the previous sentence says that the naos was 40 (not 20) cubits long. And so, as the Greek now stands, verses 18 and 19 happily contradict each other: Of 40 cubits (in length) was the naos in front of the debir 20 cubits was the length . . . and 20 cubits was the height of it(!)

Thirdly there is a very awkward phrase at the end of LXX verse 17: καὶ ἐποίησεν ἐκ τοῦ δαβειρ εἰς τὸ ἅγιον τῶν ἁγίων. The ἐκ τοῦ δαβειρ, out of the debir, is very difficult. What does it mean: and he made out of the debir for the holy of holies? The debir was the holy of holies. Admittedly the MSS disagree; but their variants are but the results of attempts to improve the sense. boc₂ e₂ La have: καὶ ἐποίησεν ἐκ τοῦ δαβειρ τὸν τοῖχον εἰς τὸ ἅγιον τῶν ἁγίων· but this does not improve the sense, for the debir was the holy of holies and not a name for the wall dividing the holy of holies from the holy place. Thdt reads: καὶ ἐποίησεν ἕως τοῦ δαβειρ τὸν τοῖχον εἰς τὰ ἅγια τῶν ἁγίων· but again τὸν τοῖχον betrays his reading as secondary, while ἕως appears to be an invention to accommodate τὸν τοῖχον in the sentence more easily. ΑΜΝ and a number of minuscules present: καὶ ἐποίησεν αὐτῷ (αὐτὸ) ἔσωθεν τοῦ δαβειρ εἰς τὸ ἅγιον τῶν ἁγίων. This is clearly an approximation to the MT, αὐτῷ representing לוֹ and ἔσωθεν מִבַּיִת. But ‘he made for it within the debir for the holy of holies’ still makes no strict sense: the holy of holies was not inside the debir, it was the debir. The Armenian, on the other hand, makes admirable sense. Instead of ἐκ τοῦ δαβειρ it has ex interiore latere eius dabir—at its (i.e. the temple’s) inner end (the) debir . . . . But of course, the Armenian is everywhere secondary and Origenic; its ex interiore latere eius is a clear approximation to the MT, לוֹ מִבַּיִת. We are, therefore, left with the awkward sentence with which we started, καὶ ἐποίησεν (αὐτῷ) ἐκ τοῦ δαβειρ εἰς τὸ ἅγιον τῶν ἁγίων as representing the nearest we can now get to the original LXX.

In view, then, of the factual nonsense of verse 18, the contradiction between verses 18 and 19, and the awkwardness of verse 17,⁵ Burney’s description of these verses as ‘the account . . . as simplified by the LXX’ sounds a trifle odd. The LXX’s account here is no simplification but rather a complete confusion of the original story, which has blurred the distinction between the naos and the debir, and shows no clear understanding of what the debir was. But the difference between the naos and the debir and their respective functions is such an elementary matter that it is difficult to think that the present confusion of the LXX arose originally from anything else than an accident, whatever subsequent attempts may have been made to import some meaning into it. Can we find any evidence that such an accident did in fact take place? I think we can.

The LXX verse 18 answers to the MT verses 17 and 18; let us place the two texts side by side.

MT		LXX
v. 17	וְאַרְבָּעִ֥ים בָּאַמָּ֖ה	v. 18	καὶ τεσσαράκοντα πηχῶν
	הָיָ֣ה הַבָּ֑יִת ה֖וּא		ἦν
	הַהֵיכָ֥ל לִפְנָֽי׃		ὁ ναὸς κατὰ πρόσωπον
	/		/
	v. 18		NIL
	/		/
v. 19	וּדְבִ֧יר		τοῦ δαβειρ
	בְּתוֹךְ־הַבַּ֛יִת מִפְּנִ֖ימָה		ἐν μέσῳ τοῦ οἴκου ἔσωθεν
	הֵכִ֑ין
	לְתִתֵּ֣ן שָׁ֔ם אֶת־אֲר֖וֹן		δοῦναι ἐκεῖ τὴν κιβωτὸν
	בְּרִ֥ית יְהוָֽה׃		διαθήκης Κυρίου.

It will be seen that the LXX omits the whole of MT verse 18 and runs verse 17 on to verse 19 reading לפנָי(v. 17) as לפני and ודבר as הַדביר; it omits also the words⁶ הַבַיִת הוא from verse 17 and the verb הכין from verse 19. Apart from that, however, it has a word-for-word equivalent of the MT and even follows the MT’s word-order without variation; so its vorlage, apart from the question of verse 18, was almost identical with the MT. But now we encounter a most significant difference which will help us decide whether the MT is an expanded form, or the LXX a shortened form, of the original. If verse 18 is excised from the MT, the MT still makes factual sense: its verse 17 says that the naos was 40 cubits (i.e. long) and its verse 19 says that Solomon built the debir to put therein the ark; both statements yield faultless sense, which even the difficult לִפְנָי at the end of verse 17 does not disturb.⁷ But the Greek, as we have already seen, makes factual nonsense. The reason is that the MT, if verse 18 is simply excised, is left with two sentences, each with its own main verb and complete in itself; whereas the Greek has only one long sentence all dependent on one main verb, which corresponds to the MT’s first main verb; the MT’s second main verb has no equivalent in the LXX. Nor can it have. If one runs the two sentences together, as the LXX does, it is impossible to have a second main verb. One cannot say: And of 40 cubits was the naos in front of the debir in the midst of the house within he made to place there the ark of the covenant of the Lord. So the LXX has no second main verb, with the result that its sentence construes grammatically. But it is precisely the absence of a second main verb and the welding together of the two parts into one sentence that is responsible for making the sentence yield factual nonsense; for now the purpose clause—to set there the ark—depends not as in the Hebrew on the clause—he made a debir—but on the clause—of 40 cubits was the naos in front of the debir.

Moreover, if the LXX did have, as the MT has, two sentences with an independent main verb in the second, the dimensions which the LXX gives in its verse 19 would refer quite properly to the debir: A debir . . . . he made . . . . 20 cubits its length . . . and 20 cubits its height. But because the LXX has only one sentence with one main verb, its verses 18 and 19 contradict each other, since the dimensions, as has been pointed out above (p. 8), must now grammatically refer to the subject of verse 18: ‘And of 40 cubits was the naos in front of the debir . . . 20 cubits was its length . . .’.

The issue to be decided, then, is whether (1) the LXX, for all its factual nonsense, represents an earlier form of the textual tradition, which the MT has relieved of its nonsense by interposing its verse 18 between the two parts and by raising the second part to the status of an independent sentence by the addition of a main verb; or (2) whether the MT represents the older form of the textual tradition, which in the LXX was shortened, when the MT verse 18 was omitted and so allowed the two parts—MT verses 17 and 19—to be welded together into one; which in turn necessitated the omission of the main verb from the second part for the sake of making grammatical sense, but all at the (perhaps unnoticed) expense of producing factual nonsense.

The present writer would choose the second explanation; but if after all he ought to choose the first, and the LXX represents a stage earlier than the MT, still it remains certain that that earlier stage was sadly corrupt, while the facts provided by the later MT are nearer the facts as the original must have given them, than is the nonsense of the LXX. The LXX’s account of the shrine is impossible.

III

But to choose the second explanation brings in its train a number of implications.

1) If the LXX’s omission of the MT’s verse 18 is secondary, an accident, perhaps, on the part of the original translator or in the course of the subsequent transmission, and if the presence of verse 18 in the MT is an original feature, what can be said about the difficult לפנָי which is left standing seemingly high and dry at the end of MT verse 17? And, in addition, what can be said about the difficult ולפני הַדביר at the beginning of the MT’s verse 20? To read And before the oracle 20 cubits length and 20 cubits breadth, etc., certainly seems odd, witness the difficulties of the EVV: And the oracle in the forepart was 20 cubits in length (AV); And within the oracle was a space of 20 cubits in length (RV). Moreover the LXX does not have the difficult words; it begins its verse (v. 19) simply: 20 cubits the length.

Montgomery’s solution is attractive for its very simplicity. Satisfied himself (as we are not) that verses 18 and 19 are late intrusions in the text, he can with likelihood argue that the end of verse 17 and the beginning of verse 20, which once stood next to each other, have become corrupt through the insertion of verses 18 and 19. His words are (p. 154):

לגנָי the pointing is impossible (= before me), indicative of the dilemma of the Masoretes; n.b. 1 MS deR., לפנין = Targum; cf. EVV ‘before the oracle/Sanctuary’. The introduction to v. 20, ולפני הדביר, an impossible clause in the connexion (cf. the attempts of Evv, exc. JV), is survival of original לפני הדביר at the end of this v., represented by the fragment לפני, and this by early corruption affected the original beginning of v. 20 = והדניר.

Yet attractive as this explanation is, we ought to notice that the LXX does not altogether support it. At the beginning of its verse 19 (the equivalent of MT. v. 20) it does not have καὶ τὸ δαβειρ to correspond with Montgomery’s conjectured והדביר; it simply omits the difficult ולפני הדביר altogether. With the other differences we have already dealt.

But it is possible at least to suggest another explanation, though even a tentative suggestion in such fine points of Hebrew text-history must be made with all due reserve and diffidence, while we wait for Qumran to speak. Perhaps after all לפני at the end of verse 17 and ולפני הדביר at the beginning of verse 20 are not so far off what is required to make explicit the dimensions which the Hebrew is trying to give. Its usage in verses 2 and 3 will help. In verse 2 it gives the dimensions of the temple as 60 in length and 20 in breadth. In the next verse it gives the dimensions of the porch which was joined to the front of the temple: 20 cubits in length and 10 cubits in breadth. But this time the length and breadth are measured in different directions from what they were with the temple. The matter is best explained by a diagram:

This change in the direction of the measurements is of course quite natural. Both temple and porch were oblong and hence the longest dimension of each is called the length. But the important thing to notice is how, when it comes to the porch, the Hebrew secures that the direction of the length and breadth shall be clearly understood: the porch is described as עַל־פני היכַל הַבַית ‘on the face of’, ‘before’ the temple of the house, i.e. lying across its breadth. Similarly its length (אָרכו) is described as עַל־פני רחַב הַבָיִת ‘on the face of’, ‘before’ the breadth of the house. Having thus determined which way the length lies, it then describes the breadth. Strangely enough, to our way of thinking, it uses the same preposition עַל־פני and says that its breadth (רָחבו) was 10 cubitsעַל־פני הַבָיִת . But there is no doubt what it means: עַל־פני is here describing what we should call the depth of the porch in front of the house.

If then, we allow the preposition לפני a similar slight variation in meaning in vv. 17 and 20, it will yield us excellent sense in both places. We need only assume that the MT’s לפנָי at the end of verse 17 is a misreading forלפנָיו , and we get the following: verse 16 And he built 20 cubits at the rear of the house . . . . . for an oracle . . . . . verse 17 And 40 cubits was the house, that is the temple, before it (i.e the oracle). In other words verse 17 is giving us the depth of the naos before, in front of, the debir, just as verse 3 gives the depth of the porch before the house. Verse 20, then, gives the dimensions of the oracle, which were 20 x 20 x 20 cubits. Actually since these dimensions make a perfect cube, it does not matter which is said to be the length and which the breadth. But the Hebrew troubles to tell us which way the length runs and which way the breadth:ולפני הַדביר עשרים אַמָה ארך ועשרים אַמָה רחַב i.e. the length is to be taken as running across as one stands in front of the debir, facing it. לִפני is here being used in a way which precisely parallels the use of עַל־פני to describe the porch as on the face of, across the front of the temple. And, moreover, this is the direction in which the length is measured in 2 Chronicles 3:8: And he made the most holy house (i.e. the shrine) ; the length thereof according to the breadth of the house (i.e. the naos), was 20 cubits ; and the breadth thereof twenty cubits . . . .

Understood, then, in this way the Hebrew not only makes sense, but also provides the exact information on the dimensions which we need to know. It is easy to see, of course, how later scribes and translators could readily fail to understand these somewhat intricate matters, and try to simplify them. But that is no reason for preferring the imprecisions and the omissions of the LXX to the present Hebrew, which, unless and until Qumran provides us with an even better text, presents the best reading we have. 2) If we cannot trust the LXX’s omission of the whole of the MT’s verse 18 and part of its verse 19, as a feature of the original text, can we trust its omissions of the MT’s verses 21a and 22b (see pp. 3–5)? Montgomery’s excision of all the Hebrew verses which mention gold overlay, except the overlay of the altar, on the grounds that they are extravagant with gold, rests on a very subjective judgment, particularly when ‘overlaid’ need not mean more than gold-leaf inlay work. But at least Montgomery’s excision is carried out consistently. The omissions in the Greek show no such consistency (see above p. 6). Moreover the Greek—and Montgomery—are left with a text that fails to mention of what material the altar was made—a most unusual omission, if it is original, in a context that everywhere freely and in detail mentions the materials things are made of. And then there is the question what Hebrew the second part of LXX verse 20 represents (see p. 4 and p. 6). As it stands καὶ ἐποίησεν θυσιαστήριον κατὰ πρόσωπον τοῦ δαβειρ καὶ περιέσχεν αὐτὸ χρυσίῳ seems to be a straightforward sentence, all referring to the altar of incense; in actual fact the words from κατὰ πρόσωπον onwards were originally translated from a Hebrew that was not talking about the altar at all. This can be seen, if we notice first that in general throughout this paragraph the LXX, apart from its omissions, agrees remarkably closely with the word-order of the MT and secondly that the closing words of LXX verse 20 and the whole of LXX verse 21 correspond word for word to the end of MT verse 21 and the first half of MT verses 22, thus:

MT LXX v. 21 …..וַיְעַבֵּ֞ר בְּרַתּ֯יּק֤וֹת זָהָב֙ לִפְנֵ֣י הַדְּבִ֔יר v. 20b κατὰ πρόσωπον τοῦ δαβειρ וַיְצַפֵּ֖הוּ זָהָֽב׃ καὶ περιέσχεν αὐτὸ χρυσίῳ. v. 22 וְאֶת־כָּל־הַבַּ֛יִת καὶ ὅλον τὸν οἶκον צִפָּ֥ה זָהָ֖ב περιέσχεν χρυσίῳ עַד־תֹּ֣ם ἕως συντελείας כָּל־הַבָּ֑יִת παντὸς τοῦ οἴκου.

---

This correspondence cannot be by chance; the LXX is simply following a text closely resembling the MT. But the closing words of MT verse 21, which LXX verse 20b so exactly reproduces, do not refer to the incense-altar at all, but to the golden chains which were drawn across the outside wall of the debir and to the gold overlay of that outside wall. The conclusion must be that the LXX has by accident omitted its equivalent of the Heb. ‘with cedar. So Solomon overlaid the house within with pure gold and he drew chains of gold across’, and has run together the end of MT verse 20 with the end of MT verse 21. The result appears to give a sensible account of the altar, but actually if we consider the Heb. which the Greek is, in fact, translating, the LXX lacks altogether both the cedar framework of the altar and its overlay; for, perhaps under the impression that its verse 20 referred to the gold overlay of the altar, the LXX has omitted from its verse 21b a counterpart to the MT verse 22b which does properly mention the gold overlay. Whether, then, these LXX omissions derive from some Hebrew text, or whether they are the work, deliberate or accidental, of the original translators or some subsequent copyist, it is most unlikely that the omissions represent either the original Hebrew or even some stage in its transmission preferable to the MT; although in some small points the LXX, as mentioned above (see p. 6) does seem to be superior to the MT.

IV

The lower criticism of the LXX, therefore, has not proved to support Montgomery’s higher critical analysis. In view of this, it might be permissible to question one of his judgments on his own higher critical grounds. He excises MT verse 19 on the grounds that it ‘parallels v 16, setting forth the shrine as the depository of the ark’. But a careful reading of the Hebrew will show that verse 19 is no mere repetition of verse 16. The purpose of verse 16 is to tell us that 20 cubits at the back of the temple were reserved for an inner shrine. The intention of verse 19 is not to repeat the announcement that a shrine was made, but, given the existence of a shrine, to tell us the purpose of its existence. And this intention accounts for the word order and phrasing of verse 19. It does not say ‘And he prepared a shrine to set there the ark . . . .’, but ‘And a shrine in the midst of the house within (i.e. such as has already been described in the previous verses; and then follows the new information) he prepared in order to set there the ark . . . .‘.

Moreover, whether the MT’s arrangement is original or secondary, it does show a perfectly intelligible, logical progression throughout its paragraph, verses 14–22. The earlier verses in the chapter have described the structural shell of the temple; now the present paragraph proceeds:

1. Verse 15. Internal wood-covering of shell: floor and walls.
2. Verse 16. Internal division of length into shrine and v. 17. naos.
3. Verse 18. Decoration motifs of shrine.
4. Verse 19. Purpose of shrine.
5. Verse 20a. Dimensions of shrine.
6. Verse 20b. Pure gold overlay of shrine.

(And then, coming out of the shrine, the first thing a visitor would see)

1. Verse 20c. Making (with LXX) of an ⁸ altar of cedar.
2. Verse 21a. Pause for summary and contrast between what the visitor has just seen inside and what he will now see outside the shrine. The shrine (the house within) was overlaid with pure gold; contrast the naos outside the shrine, which is overlaid with plain gold (i.e. not the more expensive pure gold.)
3. Verse 21b. Gold chains on outside wall of shrine.
4. Verse 21c. Gold overlay of outside of shrine (it)
5. Verse 22a. Gold overlay of all the house i.e. the naos.
6. Verse 22b. Gold overlay of all the altar.

One might at a stretch argue that the very orderliness and logic of this scheme shows it to be late and artificial, but to fault it on the grounds of being repetitive is to show a misunderstanding of what it is trying to say. To go further and to quote as witnesses against its supposed repetitiveness the irresponsible omissions of the LXX, which have involved the LXX in such factual nonsense, is surely very questionable methodology.

Additional Note

It may be that the Talmud can contribute something to our understanding of the variant readings of the difficult verse 17 (LXX). Some rabbis held that the wall between the naos and the debir was formed of two cedar partitions with a vacant space, one cubit wide, between them. But the rabbis could not agree whether this cubit space was to be deducted from the 40 cubits of the naos or the 20 cubits of the debir. Crucial to the discussion was the punctuation of the verse (MT v. 19, LXX v. 18b): And a debir in the midst of the house from within he prepared to set there the ark of the covenant of the Lord. The Gemara of the Babylonian Talmud (Yoma 52a) comments: ‘The question was asked [in the Academy]: What does Scripture mean to say? [Does it mean] “a debir in the midst of the house; from within he prepared to place the ark there”; or “a debir in the midst of the house from within”?’ Rabbi Dr L. Jung, whose translation (Soncino Press, London, 1938) I have just quoted, points out in his notes on this passage that the word debir is here being used in an unusual way to denote not the holy of holies, but the partition between the holy of holies and the holy place. The question at issue, then, is whether the ‘from within’ belongs to the first part of the verse and so refers to the debir—in which case the debir would be reckoned as included in the holy of holies—or whether it belongs to the second part of the verse and so refers to the holy of holies. In this case the phrase And a debir in the midst of the house; (note the position of the semi-colon) would imply that the debir (i.e. in this context, the partition) would be counted as being in the holy place and not in the holy of holies.

In the light of this uncertainty it is interesting to compare the reading of boc2e2 La with that of other witnesses. The former have: κὰ ἐποίησεν ἐκ τοῦ δαβειρ τὸν τοῖχον εἰς τὸ ἅγιον τῶν ἁγίων. This is open to two interpretations:

1. Allow δαβειρ its normal meaning, which makes it synonymous with the holy of holies. Then the sentence tells us that the partition-wall was made ‘out of the holy of holies’ i.e. the partition was counted as belonging to the holy of holies. Thdt, on the other hand, has καὶ ἐποίησεν ἕως τοῦ δαβειρ τὸν τοῖχον εἰς τὰ ἅγια των ἁγίων. Now ἕως + Genitive in the sense of ‘usque ad’ is quite normal in this part of Reigns; and so Thdt’s reading would imply that the partition-wall reached as far as the holy of holies but was not included therein.
2. Allow δαβειρ its unusual meaning, as denoting the cubit-deep partition. Then boc2e2 La are saying that the wall between the holy and most holy place was formed out of this cubit-deep partition.

It is in this unusual meaning that AMN and a few other minuscules use the word. They have καὶ ἐποίησεν αὐτω (αὐτὸ) ἔσωθεν τοῦ δαβειρ εἰς τὸ ἅγιον τῶν ἁγίων. Here the holy of holies is ‘from within the debir’; and this would probably be intended to mean that the cubit-deep partition was reckoned as belonging to the holy of holies and as partaking of its sanctity.

The interest of the variants lies, then, in the fact that they represent, not the original text, but the exegetical problems of later rabbis. (See also The Jerusalem Talmud, Yoma, 5:1).

---

Footnotes

¹ Montgomery, ZAW 50 (1932), p. 129.

² Notes on the Hebrew text of the Books of Kings, Oxford, 1903, pp. 71–2.

³ ICC Kings, 1951, pp. 149–50.

⁴ On p. 149 (see my page 6) Montgomery seems to indicate that the OGrr. omits only the last part of v. 18; but on p. 154 he makes it clear that the whole verse is omitted in the OGrr.

⁵ This is not the only strange feature about v. 17; but the others must be dealt with elsewhere.

⁶ Burney seems to have thought that the LXX omitted הוא הָהיכָל not הַבָית הוא (see his Notes, p. 72); and since he also thought that הוא הָהיכָל was an insertion made by R^P in the Hebrew text, he regarded the omission by the LXX and Vulgate as confirmatory of his analysis. But in III Reigns בית is always οἶκος while ναός represents היכל, so that ὁ ναὸς κατὰ πρόσωπον stands for ההיכל לפני and the words which the LXX has omitted are הבית הוא.

⁷ And perhaps it is not so difficult after all, as we shall presently see.

⁸ Notice the indefinite article here, as is appropriate for the first mention of this piece of furniture. When it is mentioned again in v. 22 it has the definite article.