3:6 kî. Commentators are divided in their
understanding of the emphatic adverb kî
(cf. WO’C § 39.3.1d). If kî is a
coordinating conjunction introducing a causal subordinate clause (cf. WO’C §
38.4), then the particle may be translated “for, because” (so Calvin [1979:
666]; Luther [Oswald 1975: 413]; Keil [2:461]; Baldwin [1972a: 245]; R. L.
Smith [1984: 330]; Glazier-McDonald [1987a: 173]; cf. LXX dióti, Syr mšlʾd, T ʾry, NRSV “for”).
However, if kî is a clausal adverb modifying the
clause in itself, then the adverb may be translated emphatically “indeed,
surely, truly” (cf. WO’C § 39.3.4). Such emphatic adverbs modifying clauses
related to the act of speaking are disjuncts
(cf. WO’C § 39.3.4b). Here the issue is the content of the utterance,
permitting the rendering “indeed, surely” (so NAB “surely,” V enim; cf. von Bulmerincq [1932: 392] who
regarded kî as an “affirmative
particle” [but left it untranslated, so NEB and NIV]; Rudolph [1976: 281] and
NJB = “no”; Chary [1969: 268] = “yes”; and Verhoef [1987: 299] = “truly”).
Other
interpretive options have been proposed, including J.M.P. Smith (1912: 66) who
read kî as an adversative (“but”) and
Dentan and Sperry (p. 1139) who deleted kî
as an “artificial connector.” Here the clausal kî is a disjunct, an
emphatic adverb modifying the clause in itself. Since the referent is the
content of the utterance, the translation “indeed” or “surely” is preferred.
The messenger formula ʾāmar YHWH ṣĕbāʾôt
in 3:5 (typically used in Malachi to begin and/or conclude disputations) and
the independent pronoun ʾănî, also
used emphatically with YHWH after the
adverb kî, confirm this reading. This
does not deny that some coordination may exist between the fourth and the fifth
disputations. They are related in theme (judgment) and Waltke and O’Connor (§
39.3.4e) have noted the emphatic use
of kî and the logical (or causal) use of kî
are not to be “strictly separated.”
ʾănî YHWH. The combination of the independent pronoun (ʾănî) and the divine name (YHWH) constitute the self-introduction
formula, a formula by which a speaker reveals personal identity to an addressee
by announcing his or her name (cf. Hals [1989: 362–63]). The self-introduction
formula in Mal 3:6 is understood alternately as an appositional construction
(“I, Yahweh”; e.g., Glazier-McDonald [1987a: 173]; cf. NIV, NJB, NRSV) or as a
predicate construction (“I am
Yahweh,” e.g., Calvin [1979: 666]; Henderson [1980: 458]; R. L. Smith [1984:
330]; cf. KJV, NEB). Given the placement of the self-introduction formula after
the disjunct kî and before the
first-person suffixing conjugation form šānîtî,
the formula is better understood as a predicate construction (cf. kî ʾănî YHWH in Isa 43:3; 45:3; Jer 9:23
[24]; 24:7; Ezek 6:7, 10, 13; 7:4, 9).
The phrase is
related to the predicate construction ʾănî
YHWH (“I am Yahweh”) used liturgically in Lev 19:3, 4, 10, 12, 14, 16, 18,
25, 28, 30, 31, 32, 33, 36, 37 as a refrain marking legal pericopes (see von
Rad [1962: 1:197]; cf. Wenham [1979: 263]). The self-introduction formula
introduces the prophet’s next disputation. More than a vestige of priestly
terminology, this formula proclaims the identity of Yahweh much like qādôš ʾănî YHWH in Lev 19:2 (cf. Hartley
[1992: 309]). The formula also confers upon Yahweh “the authority and right …
to make known his will” to Israel, whether by gracious redemption (Exod 20:2)
or by righteous judgment (Ezek 7:5, 27; 20:42; Childs [1974: 401]).
The
contrasting independent pronouns in verse 6 (ʾănî // ʾattem)
constitute an example of logical
structure with finite verbs, involving explicit
antithesis (“Yahweh” vs. “descendants of Jacob,” WO’C § 16.3.2d). The
initial position of the pronoun adds emphasis to the subject of each clause
(GKC § 135a). See further the discussion of ʾănî
in 1:4 above.
lōʾ šānîtî. The verb šnh
means “to repeat” (e.g., 2 Sam 20:10; Prov 17:9) or “to change” (Job 14:20;
Prov 31:5). The use of šnh in Mal 3:6
is unique to the Haggai-Zechariah-Malachi corpus; however, the word occurs
elsewhere in the Prophets only in Jer 2:36 (a description of “fickle” Judah!)
and Jer 52:33 (Jehoiachin’s release from prison in Babylonian captivity meant a
“change” of garb). The suffixing form of šnh
here conveys both the sense of the indefinite
perfective (“I have not changed,” so Verhoef [1987: 297]; Petersen [1995:
212]) and the instantaneous perfective
(“I do not change,” so R. L. Smith [1984: 330]; cf. WO’C § 30.5.1b, d).
The prophet’s affirmation that Yahweh “has not changed” should not be
construed primarily as a metaphysical statement, a theological commentary on
the nature of God’s being. Rather, Malachi attests the faithfulness of Yahweh
to his covenant agreement with Israel. God has not changed the terms of the
pact, but has remained constant in his oath of loyalty. It is for this reason
that Israel has not been destroyed.
Waldman (pp.
543–44) supports a similar reading of verse 6 by extending the nuance of šnh (“to change”) to “go back on one’s
word” or “renege” (or even “betray”) on the basis of the Akkadian enû (adopted by Glazier-McDonald [1987a:
179–80]). While the issue of Yahweh “reneging” on or “honoring” his word is
paramount, Waldman misses the intent of the self-introduction formula. The
point ultimately is the identity or character of Yahweh in which his actions
and responses toward Israel are rooted, not simply the action or response
itself. The real issue is not so much Yahweh reneging on his word, as it is the
person of Yahweh who is incapable of any such behavior (curiously
Glazier-McDonald [1987a: 178–79] expounds on the holy character of Yahweh
implicit in the self-revelation formula and then proceeds to interpret šānîtî restrictively in the sense of
reneging on the word of his covenant).
A further word
critiquing the supposed parallel between the Akkadian enû and Mal 3:6 is warranted. The examples of the intransitive use
of enû cited by Waldman are limited
exclusively to Akkadian legal texts referring to written contracts and
documents. While I grant Waldman and Glazier-McDonald the argument of the
immediate and wider covenant context of Mal 3:6 (including mention of the
Yahweh’s “statutes,” mēḥuqqay, 3:7),
the prophet’s chief concern is Israel’s violation
of Yahweh’s covenant—not the validation
of the document (cf. Ps 89:32–33 [31–32]).
Waldman also
failed to note that in Akkadian literary texts, the sense of enû as “going back on one’s word” is
more clear because the verb enû is
always accompanied by a verb of speaking (e.g., qabû or zakāru, “to
speak” or even a noun like awātu(m)/amātu(m), “word”; cf. CAD 4:175–76). The MT adheres to this
pattern in Ps 89:35 [34] where Yahweh avers, “I will not invalidate my
covenant, nor alter the utterance of my mouth” (lōʾ-ʾăḥallēl bĕrîtî ûmôṣāʾ śĕpātay lōʾ ʾăšanneh).
Given the
clear purpose of the self-introduction formula (with emphatic adverb kî), the very specific legal genre of
this enû idiom (“renege”) in
Akkadian, and the lack of any complementary vocabulary referring to the speech
act in Mal 3:6–12 (a literary text), Waldman’s proposal to equate Malachi’s use
of šnh in verse 6 with the Akkadian enû idiom is dubious. It seems better to
rely on the context of the disputation for the understanding that Yahweh “has
not changed” with respect to his unwavering commitment to his covenant, nor has
he changed the terms of that agreement.
wĕʾattem. The conjunction waw attached to a non-verb is disjunctive, an inter-clausal waw of the type
where the scene or action shifts (WO’C § 39.2.3a, b). The clause describes both
a contrast between “Yahweh” and “the descendants of Jacob” and the relationship
of logical consequence (“I am Yahweh; I have not changed. And so, you, O
descendants of Jacob …,” so Calvin [1979: 579]; Verhoef [1987: 297]; cf. NRSV,
“therefore”; NIV, “so”; and Petersen [1995: 212], “moreover”).
No consensus
exists here among the commentators or the versions, because some render the waw as a simple
copulative (e.g., LXX, V, Keil [2:461]; Baldwin [1972a: 245]; cf. NEB, NJB) and
others read the conjunction with the adversative “but” (e.g., Rudolph [1976:
281]; R. L. Smith [1984: 330]; Glazier-McDonald [1987a: 173]; cf. NAB, “nor”).
On the use of the independent pronoun involving logical structure with explicit
antithesis, see WO’C § 16.3.2d.
bĕnê-yaʿăqōb. The expression “descendants of Jacob” designates the
national entity or collective community of Israelites living in postexilic
Yehud, who are literally sons of
Jacob, the eponymous ancestor of Israel (so NAB, NEB, NJB; cf. TDOT 2:151). The plural bānîm can signify individuals of both
sexes (TDOT 2:150), hence the
rendering “children” (NRSV) or “descendants” (NIV). The NIV and NRSV construe
the construct-genitive phrase bĕnê-yaʿăqōb
as a vocative construction (“O descendants of Jacob”), because it stands in
apposition to the second-person pronoun (ʾattem;
cf. WO’C § 8.3d). I concur with this reading despite the lack of a preceding
imperative verb characteristic of vocative constructions (cf. WO’C § 4.7d). The
vocative here is in keeping with the emphatic purpose of the antithetical
structure of the pronouns (ʾănî and ʾattem; cf. Rudolph [1976: 282]) and the
hortatory nature of the disputation (cf. ʾăhabtem
bĕnê yiśrāʾēl, “for so you love to do,
O descendants of Israel,” Amos 4:5).
Redditt (1994: 248) has resurrected Sellin’s (p. 588) hypothesis that Mal 3:6–7 was originally the conclusion to the first disputation, because both address “Jacob.” The issue of redaction in Malachi is broached in the introduction to this disputation (III. F.) and in I. C. 2. Unity, and 4. Structure above. . . .
lōʾ kĕlîtem. The verbal root klh
can mean “perish, be destroyed” or “be consumed” (BDB: 477–78). The verb occurs in the Haggai-Zechariah-Malachi
corpus only here and Zech 5:4 (“destroy,” Meyers and Meyers [1987: 287]) and in
the Twelve prophets only in Hos 11:6 (“finish off,” Andersen and Freedman
[1980: 585–86]) and Amos 7:2 (“devour … entirely,” Andersen and Freedman [1989:
xxxvii]). Malachi equates klh with
divine judgment much like Zech 5:4, the wicked totally “consumed” (NRSV) or
“destroyed” (NIV) by God’s just wrath (cf. Ezek 22:31). Perhaps Malachi had in
mind those earlier oracles announcing both Yahweh’s intent to “finish off”
Israel (klh, Isa 1:28; Jer 5:3; 16:4;
Ezek 5:13; 6:12; 7:8; etc.) and Yahweh’s reluctance to actually act so for the
sake of his name (klh, Ezek 20:13;
see vv 13–17). Indeed, Yahweh had not changed his posture toward Israel.
Whether he judged their sin and preserved a remnant (Ezek 5:12–13) or
threatened to judge them but relented (Exod 32:12; see v 14), Israel continued
to exist as a sociopolitical entity and remained the people of God in Malachi’s
day. What greater demonstration of Yahweh’s faithfulness to his covenant could
the prophet offer?
The verb klh can also mean “to cease” or “to come
to an end” (BDB: 477), permitting the
alternative translation “nor do you cease to be sons of Jacob” (NAB; so NEB).
This reading downplays ʾattem and bĕnê-yaʿăqōb as appositives and
emphasizes the pun on the name “Jacob” with the verb qbʿ “to cheat” in a predicate construction (cf. Chary [1969:
268–69]; Mason [1977: 155]; and Deissler [1988: 332]; the LXX pternízete [“heel” or “cheat”] = Hebrew ʿōqĕbîm from ʿqb). Still others have retained the appositive structure of ʾattem bĕnê-yāʿăqōb but read klytm (whether Qal stem or repointed as
a Piel or Pual stem) as “cease, come to an end” with an implied object in
keeping with the “lawbreaking” mentioned in verse 7 (see Glazier-McDonald’s
[1987a: 175–77] discussion of van Hoonacker; cf. Rudolph [1976: 282] and CHAL: 158, “do not come to an end =
remain the same”).
Emphasizing
direct coordination between the fourth dispute (2:17–3:5) and the fifth
disputation (3:6–12), J.M.P. Smith (1912: 68) understood lōʾ as emphatic lā and
translated “you will surely be
destroyed.” However, Glazier-McDonald’s (1987a: 177) criticism is apt here, in
that such a reading makes the call to repentance in the fifth disputation
meaningless. I am inclined to agree with Verhoef (1987: 300) who opted to
render klh in verse 6 with “destroy”
instead of “cease” or “come to an end” because “the translation … stresses the
fact of God’s unchangeableness as the reason for Israel’s continued existence.”
Quite apart from either translation (i.e., an explicit understanding of Israel
as a perpetual covenant violator deserving judgment or an implicit understanding
of Israel as “unceasing cheats”), Baldwin’s (1972a: 245) concession remains
cogent: “The fact is that neither God nor Israel had changed.”
The symmetry
of the bicolon in verse 6 and the LXX rendering of the MT kĕlîtem with a form of the verb apéchō
(“abstain, avoid” in middle voice instead of the expected sunteléō; cf. Zech 5:4, wĕkillattô
// kaì suntelései) has spawned
numerous alternative (and sometimes inferior) translations of the MT of Mal
3:6b. The conjectured lî kullĕkem
(“you sons of Jacob, you all belong to me”)
of the BHK has been dismissed by
J.M.P. Smith (1912: 69) and properly deleted from the textual apparatus of the BHS. The difficulty of coordinating the
verbs šnh and klh in a parallel construction is especially manifest in the ancient
versions.
The LXXB
combines verse 6b with the first two words of verse 7a and paraphrases kaì humȇis hoí huiȏi lakṑb, ou̓k a̓péchesthē
a̓pò tōn a̓dikiōn tōn patérōn humōn (“and you, sons of Jacob, have not
abstained from the sins of your fathers”). The Syr deviates from the influence
of the LXX (“you have not departed from your injury,” Kruse-Blinkenberg [1966:
104–5]); and T makes šnh transitive
by adding a direct object (“For I, Yahweh, have not changed my ancient covenant”) and verse 6b
introduces a pious expansion about those who believe divine judgment is
confined to life in this world (a jibe at Sadducean doctrine?; see the
discussion in Gordon [1994: 58–61]). Only the V renders verse 6 in a literal
fashion. (Andrew E. Hill, Malachi:
A New Translation with Introduction and Commentary [AYB 25D; New Haven:
Yale University Press, 2008], 294-98, emphasis in bold added)