Abstract
Textual variation between ול and אל occurs throughout the Hebrew Bible about thirty times. The initial impression might be that this is the result of unintentional scribal error, yet the vast majority of these occurrences of variation involve passages wherein both readings work equally well. It is also possible to determine in many cases the linear development from one reading to the other. The present article argues that the similar pronunciation of ול and אל became an occasion to introduce variant readings intentionally.
1. Introduction
Textual variation between ול and אל occurs in each of the three major divisions of the Hebrew Bible, about thirty times in all. 1 At first glance these occurrences of variation look like the product of faulty hearing on the part of scribes who were working from dictation. The phonological similarity would have made this an easy error. But unintentional scribal errors often create difficult readings, and the number of examples where אל and ול work equally well in the same context is disproportionately higher than that of those where one is a more difficult or impossible reading. Furthermore, it is possible to determine in many cases the linear development from one reading to the other or at least the origin of the textual variation and the intent behind it.
The text of Genesis 23 provides a good introductory example of the problem at hand in the present article:
According to the MT, Gen. 23.5 is an introduction to the reply of the sons of Heth to Abraham: ‘And the sons of Heth answered Abraham, saying to him (ֹול)’. This is also the way Ephron's reply to Abraham is introduced in Gen. 23.14. 2 The BHS apparatus, however, suggests that ול should be vocalized ּול (‘If only’) and placed at the beginning of the discourse in Gen. 23.6. This suggestion has warrant from Gen. 23.13, although ּול does not normally precede an imperative. The SP and LXX have אלֹ instead of ֹול. This negation comes as the first part of the discourse—‘“No, hear us …”’ (Gen. 23.6)—much like Gen. 23.10–11 (see also 2 Sam. 18.12; 19.7). Thus, each reading makes sense in context and has a parallel in the same passage. Without making a decision about the priority of one reading over the others, it is evident that there is sufficient motive for the intentional production of each one of them.
While it is possible to argue that all such instances of textual variation are due to confusion of sound, the relatively low number of difficult readings produced by this variation suggests that unintentional scribal error is not the most likely explanation. The following study argues that the similar pronunciation of ול and אל became an occasion to introduce variant readings intentionally. That is, during a period of textual fluidity before the stabilization process took hold in textual transmission, scribes saw an opportunity in the phonological similarity of ול and אל to introduce the reading that they considered more suitable. They could do this and go virtually unnoticed, especially to those who only heard the text. The textual variation itself is evidence enough that the pronunciation of ול and אל was similar if not identical at the time.
2. Examples of Scribal Error
Before proceeding to the majority of instances of textual variation between ול and אל, it is conceded here that there are a few occurrences where the best explanation for the variation is confusion due to similar sound. For example:
The kethiv of the MT does not make sense without a lamedh of possession. But even such a remedy could not fix the problem that the text would then have with the following verse in which creatures with jointed legs are permitted for consumption. The qere, followed by the Cairo Geniza reading and the versions, reads smoothly and works well with Lev. 11.22. The text of Lev. 25.30 has a very similar situation, although there the Samaritan Pentateuch preserves the qere along with the versions.
Hannah's prayer (1 Sam. 2.1–10) furnishes another example where only one of the readings is realistically possible: 3
The L reading does not fit in context, unless it is a question: ‘And are not deeds weighed?’ But this is awkward and unlikely. The prayer asserts that the proud should be aware of the fact that God knows what they do and say and will bring them down and exalt the lowly. Some manuscripts of the Septuagint have a quick fix for this text—a pronoun after the word ‘deeds’: ‘And his deeds are not weighed’. But again, regardless of the truth of such an orthodox claim, the point of the context is not the immeasurable nature of God's deeds but his judgment of the proud and vindication of the lowly. And so the qere, along with a multitude of medieval Hebrew manuscripts, preserves the original reading, which was corrupted via an unintentional scribal error of the ear, an error for which some manuscripts of the Septuagint (and perhaps their Vorlage) made compensation.
The book of Samuel, which is well known for its textual problems, provides two more examples of what is most likely unintentional scribal error due to phonological similarity (see also 2 Sam. 14.30):
It is difficult, if not impossible, to make sense of the L text of 1 Sam. 20.2. It appears to be the result of an accident. The L text of 2 Sam. 16.18 is entirely unsuitable to the context. Hushai is pledging his allegiance to Absalom.
3. Examples of Intentional Alteration
This section examines cases where either ול or אל will work in context, but one evidently derives from the other. The phonological similarity between the two has become the scribe's opportunity to make a subtle change in the text for some particular reason.
Even though these two readings give the exact opposite sense, there is nothing that precedes or follows that demands one or the other. The story concludes in Gen. 26.33 with the naming of the well, which is given as the reason for the name of the city. But it is hard to imagine that the MT reading would have ever been changed deliberately to the reading represented by the LXX, a decidedly anticlimactic reading. On the other hand, it is easy to see why a scribe would want to create a more suitable ending by changing the LXX reading to the MT reading. Perhaps the scribe found warrant for this alteration in Gen. 26.32a: ‘and they told him (ֹול ודגיו) about the well that they dug’. A somewhat comparable example comes from 1 Sam. 8.19:
Both the Masoretic readings give a similar sense, but the L reading makes the refusal of Samuel's advice clearer (see 1 Sam. 8.19a; cf. 1 Sam. 10.19). The reading of the LXX and the Syriac is a conflation of the two Masoretic readings.
The textual variation in Exod. 21.8 has a possible legal dispute behind it: 4
The case law in Exod. 21.7–11 outlines what is to happen when a man sells his daughter as a maidservant and one of three things occurs: (1) she is displeasing to her master/husband; (2) she becomes the wife of her master's son; or (3) her master/husband takes another wife. If the instructions for these three situations are not followed, then the woman is free to leave. According to the qere of Exod. 21.8, a master could designate a maidservant as his wife, find her displeasing, and then allow her to be redeemed. It is possible that this was deemed unacceptable to some because it appeared to open the door for frivolous divorce contrary to the law in Deut. 24.1–4 (cf. Mt. 19.3, 7). Thus, the kethiv indicates that the master does not designate the maidservant as his wife.
The original text of Josh. 5.14 creates another opportunity for a scribe's sleight of hand, this time in the service of a less difficult reading:
The captain responds to Joshua's either–or question in Josh. 5.13 with ‘No’ in the L text of Josh. 5.14. This likely means that the captain is not merely another man who is on one side or the other. 5 The following יכ could then be rendered as ‘for’, ‘but’, or ‘indeed’. But since אל does not ordinarily mean ‘neither’, a smoother reading was created with ול. The following יכ would then be rendered ‘indeed’, affirming that the captain is on Joshua's side. This adjustment has its warrant in the second half of Josh. 5.14 where the discourse introduction ול רמאיו occurs. Such encouragement from the context is consistent with other examples (Gen. 23.6; 26.32).
Two more cases involving what follows דמא occur in 1 Sam. 2.16 and 1 Kgs 2.30 (see also 1 Kgs 11.22):
The 1 Sam. 2.16 text works with either reading, but the L text looks like a foreign intrusion given the fact that the surrounding narrative uses וילא דמאיו (1 Sam. 2.16a, 27b; see also 1 Sam. 1.14), not ול דמאיו. The same might be said of 1 Kgs 2.30 where וילא דמאיו occurs earlier in the verse (but see 1 Kgs 2.36).
Another kind of variation appears in 2 Sam. 23.18. Once again the motive for the change is clear from the context:
According to the L text of 2 Sam. 23.18, Abishai had a name among the three men mentioned in the story of 2 Sam. 23.13–17. In fact, he was their leader (but see two Masoretic manuscripts [‘leader of the thirty’] and 2 Sam. 23.8). But according to 2 Sam. 23.8–12, the names of the three were Ishbaal (?), Eleazar, and Shammah. Furthermore, according to 2 Sam. 23.19, it seems that Abishai was honored by the three, but he did not attain to them. The same problem occurs with Benaiah in 2 Sam. 23.22–23. The solution in a few manuscripts and in 1 Chron. 11.20 is to change ול to אל. But while this avoids the issue of who was named among the three, it hardly creates a grammatically acceptable clause. One possibility is that השלשב should be read השלשכ. That is, Abishai (and Benaiah) had a name ‘like the three’ but not ‘among the three’. Confusion of ב and כ is well documented, but there is no textual evidence for the reading השלשכ. Another way to look at this is to ask why there would have been a change from אל to ול if אל had been the original reading. It seems more likely that there was a change from ול to אל in order to avoid the impression that there were two triads: (1) Ishbaal, Eleazar, and Shammah; and (2) Abishai, Benaiah, and Asahel (?). This leaves the question of what it would mean for Abishai and Benaiah to be named among the three and yet not attain to them. Perhaps the answer lies in the existence of two triads. They were named among their own triad, but they did not attain to the other. It is not clear then who the three are in the story of 2 Sam. 23.13–17. Perhaps Abishai and Benaiah were simply named ‘with’ the three.
The text of 2 Kgs 8.10 poses a theological problem that causes variation: 6
In the L text, Elisha instructs Hazael to inform Ben Hadad that he will not live. This is consistent with the second half of the verse: ‘And the LORD has shown me that he will surely die’ (2 Kgs 8.10b). But this runs into problems with the delivery of the prophecy in 2 Kgs 8.14b: ‘He said to me, “You will surely live”’. Hazael then murders Ben Hadad in 2 Kgs 8.15. On the other hand, according to the qere of 2 Kgs 8.10a, Elisha instructs Hazael to lie to Ben Hadad. Thus, Hazael gives Ben Hadad a false sense of security in 2 Kgs 8.14 and then conveniently assassinates him in 2 Kgs 8.15. It is possible that some felt that this reading in 2 Kgs 8.10 made Elisha look like a false prophet (Deut. 18.21–22; 1 Kgs 13.11–32). A change from ול to אל would absolve Elisha and make Hazael look like he took matters into his own hands in 2 Kgs 8.14–15. It is also possible that it was thought that Elisha's (and thus God's) instructions were in violation of the command in Exod. 20.16 and Deut. 5.20. But the words, ‘You will surely live’, are only part of what Elisha says (i.e., the setup). The real prophecy is that Ben Hadad will die. Furthermore, the prohibition in Exod. 20.16 and Deut. 5.20 is not against such a setup. It is against false accusation in a court of law (see Exod. 23.1; Lev. 19.11; Deut. 19.18; see also Exod. 1.15–22; Josh. 2; Heb. 11.31; Jas. 2.25). 7 Deception is sometimes a form of divine judgment (1 Kgs 22.19–23; Ezek. 14.9). Therefore, it would appear that the qere in this instance preserves the original reading. This reading was then ‘corrected’ on the false assumption that the text presented a theological difficulty. The easy switch from ול to אל made the perceived correction a very subtle one.
Both Isa. 8.23 and 9.2 furnish examples of variation between אל and ול in admittedly difficult texts:
Part of the difficulty with Isa. 8.23a is the decision whether to put the text with what precedes (1QIsaa) or with what follows (MT). Another problem is the antecedent of the third feminine singular pronoun (ץרא?). It is possible that הל should be vocalized הלֹ (‘to him’). 8 This is an instance where both אלֹ and ֹול work well in context, but the linear development from one to the other is not immediately obvious. Presumably a scribe either found the one that goes with what precedes and sought to construe it with what follows or vice versa.
The kethiv of Isa. 9.2 at first glance appears contrary to the context. But it is also possible to render it: ‘You increase the nation whose joy you did not magnify’. The qere, on the other hand, fits the context nicely and forms a good piece of parallelism. The BHS proposal—הליגה (‘the gladness’) for אל יוגה—has no manuscript support. So, does this mean that the qere is a ‘correction’ of the more difficult reading of the kethiv? Not necessarily. It is possible that a scribe saw the neat parallelism of the qere and saw an opportunity to continue the thought from Isa. 9.1 that the former darkness will give way to a great light. That is, the kethiv sets the increase of the nation (the light) over against the lack of joy in the past (the darkness).
The second servant song (Isa. 49.1–6) has a pertinent example of textual variation that relates to the identity of the servant:
The L text looks out of place given the fact that the preceding parallel says the purpose of the servant is ‘to restore Jacob to him (וילא)’. 9 But it might also be the result of an intentional change designed to preserve the interpretation that the servant is always Israel. According to Isa. 49.5–6, the servant is someone other than Israel who acts on behalf of Israel. Thus, the insertion of ‘Israel’ in Isa. 49.3b is perhaps secondary (> Ms). 10 Of course, the servant is sometimes Israel (e.g. Isa. 41.8), but even outside the book of Isaiah the servant is often an individual (e.g. Moses [Deut. 34.5], Joshua [Josh. 24.29], David [Ps. 18.1]).
One last example from the book of Isaiah comes from Isa. 63.9:
Brevard Childs noted the problem well:
The initial problem is signalled by the Kethib/Qere variant interpreting the Hebrew lōʾ. The
However, in response, it is also clear that there is a rather close parallel in Ex. 33:12ff. of the face or presence of God accompanying Israel as a sign of his continued commitment in spite of the sin of the golden calf. The extension of the term in the formula ‘angel of his presence’ is fully in accord with the other circuitous formulations of divine intermediaries who are the visible agents of the selfsame divine essence (e.g., ‘angel of Yahweh’, etc.). Although the
The final example from the Prophets features variation between the MT and LXX of Hos. 11.5. This variation has ties to two different but related themes in the book of Hosea and the Twelve.
The MT could be a question: ‘Will he not return to Egypt?’ In such a case, the sense would be the same as the LXX. But if it is not a question, then the denied return to Egypt and Assyria in Hos. 11.5a is a denial of political alliance and aid (cf. Hos. 5.13; 7.11). The LXX, however, states, ‘Ephraim will live in Egypt’. This is a new Egypt, a metaphor for a new place of bondage from which there will be a new exodus. Israel will recapitulate its history. The return to Egypt metaphor is well known from elsewhere in Hosea (Hos. 8.13; 9.3) and is likely the intended meaning of Hos. 11.5 given the fact that Hos. 11.11 speaks of a new exodus. Thus, the reference to the exodus in Hos. 11.1 has become an opportunity to anticipate a future deliverance (cf. Num. 23.22; 24.8; Mt. 2.15). 12 The MT has misunderstood the metaphor (and the context) and harmonized the passage with Hos. 5.13; 7.11.
Several examples of variation between ול and אל occur in the Writings. Two of them are in the book of Psalms:
The L text of Ps. 100.3 proclaims that Yahweh is God, not the people. But the qere understands the text with what follows and represents a harmonization with other texts that speak of God's people as the sheep of his pasture (Pss. 79.13; 95.7). The Ps. 139.16 text is not so straightforward. Most take the L text to mean that the psalmist's days were fashioned before one of them came to be (e.g.,
The book of Job is notorious for its difficult Hebrew text, and the text of Job 6.21 is no exception: 14
The LXX appears to read יל (cf. Syr. and Vulg.): ‘But yet you too trod on me (μoτ) without mercy; so now that you have seen my hurt, be afraid!’ (
The linear development of the textual variation in the witnesses to Job 9.33, where Job expresses his desire for a judge, is counterintuitive:
The L text states the absence of a judge, while the other witnesses make the connection with the second half of the verse smoother by expressing a wish (cf. Job 16.4). This would seem to suggest that the L text has the priority. But a closer look reveals that this would be the only occurrence of שי אל in the Hebrew Bible. The opposite of שי is not שי אל but ןיא. 17 While it is possible to make sense of the L text, it is perhaps better to see it as a change from an original ול designed to fit with Job 9.32a: ‘For he is not (אל) a man like me … There is no (אל) judge between us …’ (Job 9.32–33).
The variation between אל and ול in Job 13.15a is more visible than many of the other examples due to the fact that it introduces an important theological question. The English versions vary more because of this, whereas they tend to be more lopsided with the other examples (see also the discussion in m. Sota 5.5).
The sense of ןה here is likely ‘if, though’ rather than ‘behold’, given the Aramaic influence elsewhere in the book. Nevertheless, it is possible to render the kethiv in a manner different from the NLT: ‘Behold, he will slay me; I have no hope’ (RSV). For those who rule out a priori the possibility of a doctrine of resurrection in the Hebrew Bible, the qere is late and not even an option to explore. But there is no reason to bring such an assumption to the text. 18 There are indications from the context that Job could be expressing hope rather than despair (Job 14.14; 19.26). It is possible that a scribe found the kethiv to be too negative and changed it, but it is also possible that, like many modern scholars, a scribe considered the qere to be foreign to the theology of the book and thus in need of change (cf. Mt. 22.23; Mk 12.18; Lk. 20.27; Acts 23.8).
A final example from the book of Job gives evidence of a scribe misinformed by an alleged parallel elsewhere in the book:
It is true that the kethiv–qere system in the MT is a reading tradition and not a means of registering textual variants, 19 but it is also true that the variation between kethiv and qere often corresponds to actual textual variation in the witnesses (see the examples above). Sometimes the reading tradition prompts the textual variation. In other cases the relationship is reversed. It is also possible for the two to originate independently. In the case of Job 41.4 we are dealing primarily with the reading tradition. According to the kethiv of this text, God declares that he will not be silent about the members of Leviathan's body. This is because the greatness of Leviathan says something about the greatness of the one who controls Leviathan. The qere is likely based on a misunderstanding of the use of דב (‘member’) in this passage. In a similar expression elsewhere in the book of Job, the word דב is a homonym meaning ‘empty talk, boasting’: ‘Should your empty words (ךידב) silence (ושידחי) men?’ (Job 11.3a). This is Zophar's response to Job's words. The verse also employs a causative sense for the hiphil of שדח as opposed to the usual exhibitive sense, which occurs in Job 41.4. Such an understanding of דב and שדח with the kethiv of Job 41.4 yields an unsuitable meaning: ‘I will not silence his boastings’. The qere is thus designed to avoid a nonexistent problem.
Two instances from the book of Proverbs show the versions in favor of the kethiv of the MT:
The subject of the participle ףדךמ in Prov. 19.7b is presumably the poor man from Prov. 19.7a. The referent of the pronoun המה in Prov. 19.7b would then either be the family and friends of Prov. 19.7a or the ‘words’ of Prov. 19.7b. According to the kethiv, the text would mean that the poor man seeks to speak with his family and friends but is unsuccessful, precisely because they have rejected him and are far from him (Prov. 19.7a). According to the qere, the poor man is in fact successful in his pursuit despite the prior rejection. There is, of course, some question about the placement of Prov. 19.7b in the parallelism of the verse. 20 BHS sets Prov. 19.7a as a parallelistic line, but it leaves Prov. 19.7b without a parallel. It is possible to take it as a third colon or a separate comment on Prov. 19.7a, but it is also possible to understand Prov. 19.7b as a parallelistic line (see translation above). The LXX, on the other hand, has an expansion in Prov. 19.7b that gives the verse a different sense and leaves no doubt about the parallelism: ‘Insight will draw near to them who know it, and a prudent man will find it. He who does much evil perfects wickedness, and he who uses provoking words will not be saved’ (NETS). Because the LXX of Proverbs is well known for its contributions from the translator(s), the differences in its Vorlage, and its rearrangement of Proverbs 24–31, it seems best to see the expansion here as secondary. 21 Due to the fact that the kethiv of Prov. 19.7b basically continues the line of thought from Prov. 19.7a, it is difficult to imagine why a scribe would change the text from אל to ול. On the other hand, if a scribe found ול in his text, it is understandable that a change might have been made to אל.
The section Prov. 26.1–12 is about the ‘fool’ (ליםכ). Every verse in this section uses the term ליםכ except v. 2. The kethiv of Prov. 26.2 seems to be a separate proverb without relation to the surrounding content other than the parallels to the syntactical structure of Prov. 26.1. This likely prompted a change from אל to ול. The antecedent of the pronoun would then be the fool of v. 1, the point being that even a groundless curse comes to a fool as a providential consequence of his folly.
Ecclesiastes 10.10 is the only other example from the wisdom literature:
The LXX and Syriac have neither אל nor ול in this text. The textual variation represented by the western and eastern Masoretes turns on the understanding of the following רבגי םיליחו. If the western Masoretes are correct (L), then this must mean: ‘then he must increase strength’. If the eastern Masoretes are correct, then it would mean: ‘then (i.e. as a result) he increases strength’. The western text would seem to fit the context of Eccl. 10.8–11 much better, which would be an argument in favor of the eastern text, but it could also be argued that the western text was original, while the eastern text adjusted the verse without attention to the context. A third possibility is that the LXX and Syriac represent the original text. The Masoretic texts could then be equally pristine yet secondary readings introduced independently of one another. 22
The book of Ezra–Nehemiah offers one final example from the biblical corpus:
These words form part of what ‘the foes of Judah and Benjamin’ (Ezra 4.1) say to Zerubbabel and the leaders when they claim to be seekers of the same God and express their desire to join in the temple rebuilding project. It is clear from what follows that Zerubbabel and Joshua understand their words to be disingenuous (Ezra 4.3). According to the kethiv, the foes say that they have not been sacrificing since the days of Esarhaddon and therefore would like to help build the temple so that they can resume their religious practice. According to the qere, they claim to have been sacrificing all along, but apparently without a temple. Was the change from one reading to the other motivated by a desire to depict the foes as unfaithful despite their claim (i.e. lacking sacrifices or sacrificing outside of the temple), or was it simply to portray them as making a claim to faithfulness (refusing to sacrifice outside of the temple or sacrificing no matter what)? Here again it is difficult to make a definitive decision, but it is likely that some permutation of the above options is behind the present textual variation. One possible scenario is that the qere is original, the idea being that the foes are claiming to be faithful (sacrificing), but they are really putting their foot in their mouth (sacrificing without a temple). This could have been misunderstood by a scribe who thought it would only be appropriate for the foes to say that they had not been sacrificing, prompting a change to the kethiv. Of course, it would be possible to start with the kethiv and imagine a scenario in which there would be a change to the qere. For the purposes of the present article, it is of primary importance to show that the textual variation is intentional rather than unintentional.
4. Conclusion
Textual variation in the Hebrew Bible between ול and אל is indeed widespread. But the number of passages where the explanation for the variation is confusion due to phonological similarity is not as high as one might think at first. The majority of occurrences involve intentional change. This change is motivated by the context, but it is also true that the similar sound between ול and אל provided an opportunity too good to pass up. Scribes could make the adjustment with very little graphic or phonological alteration and yet completely alter the sense of the text.
Such an observation invites exploration of similar textual phenomena, such as that found in 2 Sam. 5.23 and 1 Chron. 14.14:
The Samuel text almost looks like an instance of scribal oversight where someone skipped from ול to אל. Or was the אל of the shorter text seen as an opportunity to expand the text? These and other matters can be the subject of future research.
Footnotes
1.
Bruno Ognibeni has shown that the alternation is ancient and that each context has its own element of ambiguity (Tradizioni orali di lettura e testo ebraico della Bibbia: Studio dei diciassette ketiv אל/qere ול [Studia Friburgensia,
2.
E.A. Speiser argues that the
3.
See S.R. Driver, Notes on the Hebrew Text and the Topography of the Books of Samuel (Oxford: Oxford University Press, 2nd edn, 1912), p. 25.
4.
See Brevard S. Childs, The Book of Exodus (OTL; Louisville, KY: Westminster/John Knox Press, 1974), pp. 442, 448, 469.
5.
Marten H. Woudstra, The Book of Joshua (NICOT; Grand Rapids: Eerdmans, 1981), p. 105: ‘The visitor does not fit Joshua's categories exactly’.
6.
See Marvin A. Sweeney, I & II Kings (OTL; Louisville, KY: Westminster/John Knox Press, 2007), p. 318.
7.
See J.G. McConville, Deuteronomy (AOTC; Downers Grove, IL: InterVarsity Press, 2002), p. 130.
8.
E.F. Kautzsch, Gesenius’ Hebrew Grammar (trans. A.E. Cowley; Oxford: Clarendon Press, 2nd edn, 1910), p. 36.
9.
It is possible that the L text is meant to say that Israel will not die, but the use of the idiom ‘to be gathered’ in this sense typically applies to individuals rather than nations. Furthermore, the use of this idiom elsewhere usually indicates to where the individual will be gathered—e.g., to one's fathers, people, or graves (see Gen. 25.8, 17; 35.29; 49.29, 33; Num. 20.24, 26; 27.13; 31.2; Deut. 32.50; Judg. 2.10; 2 Kgs 22.20; 2 Chron. 34.28). It is also not certain how well this sense would fit into the parallelism of Isa. 49.5.
10.
See John L. McKenzie, Second Isaiah (AB, 20; New York: Doubleday, 1968), pp. 103–106; Claus Westermann, Isaiah 40–66 (trans. David M.G. Stalker; OTL; Philadelphia: Westminster Press, 1969), p. 209. This decision does not have the weight of the external evidence. It is based primarily on internal evidence.
11.
Brevard S. Childs, Isaiah (OTL; Louisville, KY: Westminster/John Knox Press, 2001), p. 523.
12.
See Michael B. Shepherd, ‘The New Exodus in the Composition of the Twelve’, in Paul J. Kissling and Robert L. Cole (eds.), Text and Canon (Eugene, OR: Pickwick, forthcoming).
13.
John Goldingay, Psalms. III. Psalms 90–150 (Grand Rapids: Baker Books, 2008), p. 624.
14.
See Francis I. Andersen, Job (TOTC; Downers Grove, IL: InterVarsity Press, 1976; repr., 2008), pp. 141–42.
15.
See Emanuel Tov, Textual Criticism of the Hebrew Bible (Minneapolis: Fortress Press, 3rd edn, 2012), p. 138.
16.
Franz Delitzsch, The Book of Job (trans. Francis Bolton; Keil & Delitzsch Commentary on the Old Testament; Edinburgh: T. & T. Clark, 1866–91; repr., Peabody, MA: Hendrickson, 2001), p. 308.
17.
There is one occurrence of שי ןיא in the Hebrew Bible (Ps. 135.17b; cf. 1 Sam. 21.9) where שי probably retains its original sense as a noun (‘existence’): ‘there is no existence of breath in their mouth’ (see BDB, p. 441).
18.
See Walther Eichrodt, Theology of the Old Testament, II (trans. J.A. Baker; Philadelphia: Westminster Press, 1967), pp. 514–15; Gerhard von Rad, Old Testament Theology. I. The Theology of Israel's Historical Traditions (trans. D.M.G. Stalker; New York: Harper & Row, 1962), p. 406.
19.
See James Barr, ‘A New Look at Kethibh–Qere’, OTS 21 (1981), pp. 19–37.
20.
See Tremper Longman III, Proverbs (Grand Rapids: Baker Books, 2006), pp. 366–67.
21.
See Tov, Textual Criticism, pp. 304–305.
22.
See Shemaryahu Talmon, Text and Canon of the Hebrew Bible: Collected Studies (Winona Lake, IN: Eisenbrauns, 2010), pp. 171–266.