I am grateful for the long-awaited publication of Hayim ben Yosef Tawil’s An Akkadian Lexical Companion for Biblical Hebrew: Etymological-Semantic and Idiomatic Equivalents with Supplement on Biblical Aramaic (Jersey City: Ktav, 2009). The volume is clearly a labor of love, a pioneering effort in what is still a virgin field of inquiry. The volume bills itself as a systematic evaluation of “the parallels and cognates between these two branches of the Semitic family of languages” (ix). It comes with the highest possible recommendations: on the back cover, Moshe Bar Asher, William Hallo, and Peter Machinist all sing its praises.
Go ahead and buy it if you haven’t already (if you are an SBL/ASOR/ETS member, you can order it from Eisenbrauns through Dec 31 for $70.20, an unbeatable price). It’s impossible not to learn a thing or two from virtually every one of its xxiv + 503 pages.
Specifically, Tawil’s Companion compares the lexis of biblical Hebrew with that of Akkadian wherever etymological equivalents have been identified, with limited but nonetheless helpful coverage of semantic equivalents and idiomatic equivalents which do not have an etymological basis.
A strength of the Companion is its careful cataloguing of shared features like cognate accusative and other syntactic constructions; denotative and idiomatic usages; a-b parallelisms and sequencing; hendiadys, merismus, simile/metaphors, and transferred meanings. To be clear, Tawil, since he restricted himself to thoroughly canvassing the lexis of Hebrew and Akkadian in instances of etymological equivalence, merely scratches the surface. If all cases of semantic equivalence were canvassed, the volume would have stretched to thousands of pages. That’s how much work remains to be done.
Would that this Companion encourage others, especially those with training in linguistics, to finish the task. There is really no telling how many passages of the Bible will be illuminated and better understood, once the research project Tawil has pioneered is brought to completion.
Another plus: Tawil points out numerous cases in which NJPSV has a footnote saying “Meaning of Hebrew word uncertain” (in this sense, NJPSV is the only honest translation of the Hebrew Bible on the market; NJPSV so notes, I would guess, over a thousand times) whose probable sense is nonetheless certain based on comparative data the massive corpus of Akkadian literature provides.
An example from p. 225 may suffice to illustrate both the value and limits of Tawil’s Companion. The heading of the entry reads as follows:
מרץ = Akk. marāṣu v. OAkk. on (CAD M1 269a; AHw 609a). Akk. to fall ill, become troublesome; Heb. to be bad, painful.
The approach is straightforward. Tawil begins with a biblical Hebrew vocable and identifies its etymological equivalent in Akkadian. The chronological range of attestation in Akkadian is provided, along with page and column references to the location of the cited equivalent in the standard reference dictionaries, CAD and AHw. Then the principal meanings of the Akkadian and Hebrew vocables are listed.
This is all very helpful, but not quite as helpful as one might wish. The starting point is a Hebrew vocable, which is fine, but an apples-to-apples comparison of Hebrew to Akkadian is not possible without recourse to a complex algorithm. Specifically, verbs should not be compared with verbs sic et simpliciter, but G verbs with G verbs, Š verbs with H verbs, etc. (Tawil knows this very well, as other entries prove.) That too is simplistic, since the semantic counterparts of many verbs in Akkadian which occur in the G but not the N stem – like kabātu = כבד - will be neatly distributed across the G and N stems in Hebrew (this shows up in Tawil’s illuminating treatment of כבד = kabātu, #7 [153b], though one is not alerted to the fact in the heading of said entry, which takes note of the G and D stems in Hebrew and Akkadian alone). Furthermore, and this is a crucial point, attributive participles in Hebrew should be compared with attributive adjectives in Akkadian. Akkadian is rich in adjectives and Hebrew is rich in participles with identical attributive functions.
In point of fact, מרץ occurs in the N stem, as an N attributive participle, and in the H (= Š) stem. For that reason, it would have been preferable for the heading to read as follows:
מרץ = Akk. marāṣu v. OAkk. on (CAD M1 269a; AHw 609a); marṣu adj. OA, OB on (CAD M1 291a . . .). Akk. G to be/become ill, hurt, aggrieved; adj. ill, hurtful, grievous; Š make ill, trans. hurt, trans. grieve; Heb. N to be grievous; N ptp. grievous; H trans. grieve.
[Note that I have chosen the English glosses for the Akkadian and Hebrew items so as to do a number of things simultaneously: illustrate the semantic coherence *and* range *and* development of the items in question; illustrate the semantic coherence of the items across the two languages; and provide translation equivalents which, if deployed, "fit" in a broad number of cases and allow for a relatively high degree of concordance in translation across attested examples: the effort to do all of this simultaneously coheres with a long philological tradition alive and well in Gesenius and BDB and AHw far more than in CAD and HALOT; CAD and HALOT (and DCH) are better if the goal is to see what meaning one might assign to items in context without recourse to philological thought experiments (*controls,* I would argue, when used properly) of the kind just mentioned]
Tawil’s מרץ entry continues as follows:
The semantic equivalent of Akk. marāṣu “to be sick” is the Hebrew verb חָלָה.
Exactly right, and if Tawil had made a point-by-point comparison of the denotative and idiomatic usages of marāṣu and חלה, specific phrases, sequences, and a-b parallelisms in which they both occur, he would have treated his readers to a feast several columns in length.
It would be insane to complain that he did not. The scope of his work is narrower. But other Assyriologists have done broader spadework. Here is a relevant quote from a volume by Shalom Paul, his Amos commentary in the Hermeneia series, which is rich in observations of this kind:
The verb נֶחְלוּ [in Amos 6:6, JFH], which is the niph‛al of the root חלה, basically means “to be sick,” and then develops the meaning “to be sick about, grieved over, worried and concerned about.” Thus 1 Sam 22:8: “No one is concerned (חֹלֶה) about me.” [A footnote cites another example, with bibliography: Isa 57:10, JFH] (Compare similarly the same semantic development in Akkadian, in which the verb marāṣu [“to be sick”] also means “to be concerned about, care for”; for example, šumma ina kittim aḫi atta u tamarraṣa; “If you really are my brother and concerned about me, [send me barley].”
Tawil does not appear to have consulted Paul’s Amos commentary (it is not listed in his extensive bibliography). But he comes to analogous conclusions with respect to מרץ:
[N]ot unlike Akk. marāṣu, Heb. מָרַץ (attested four times) has the extended semantic development “to be painful, difficult, severe,” as can be seen from the following semantic equivalents: Akk: amatu marṣu “severe, painful (lit. sickening) word, matter” (CAD M1 273a c; 273b) may be compared to the Heb. phrase (considered by NJPS as “meaning of Heb. uncertain”) מַה־נִּמְרְצוּ אִמְרֵי־יֹשֶׁר “how (lit. sickening) painful, severe are ‘just’ words” (Job 6:25). More specifically, note the following Heb. and Akk. idiomatic equivalents: Heb. קְלָלָה נִמְצֶרֶת “severe curse,” e.g., וְהוּא קִלְלַנִי קְלָלָה נִמְצֶרֶת “and he cursed me a severe curse (1 Kgs 2:8); Akk: erreta marulta, e.g., erreta marulta ša nasāḫ išdi šarrūtišu u ḫalāq nišēšu littašqar “may he (Assur) pronounce an evil curse to uproot his kingship and destroy his people (CAD A2 305a); ar[rat] la napšuri marušta līrurūšu “may they (the gods) curse him with a grievous, indissoluble curse” (CAD M1 294a b).
To be sure, Driver anticipated Tawil with respect to Job 6:25. Clines notes in his Word commentary:
G. R. Driver (“Some Hebrew Words,” JTS 29 [1927–28] 390–96 ) sees מרץ [in Job 6:25) as cognate with Akk. marāṣu “be ill, displeasing,” and translates “are bitter”; hence probably neb “how harsh.”
Tawil does not list the cited article in his bibliography, though he lists others by Driver.
It is salutary that the adjective marṣu makes it into Tawil’s discussion (with a gloss of his own devising, not that of CAD), though of course the phrase in question ought to be awatu mariṣtu, with a reference and quotations from CAD M1 294b, not “CAD M1 273a c; 273b” (cognate expressions with G stative marāṣu are cited there).
The loci with מרץ Tawil does not discuss – Mic 2:10 (to which should be added a famous passage in 1QHodayot) and Job 16:3 - also have, in my estimation, excellent analogues in Akkadian, but this review is not the place to expound on them.
Typos turn up here and there in the Companion, but they are rare for a work of this complexity and detail. For example, on xxi, Heb. לֵב חֹלֵה = Akk: libbu marṣu should be לֵב חֹלֶה, perhaps, or rather, another example should be chosen, since that particular collocation is not attested in the Tanakh; in fact, it is not discussed in Tawil’s magnificent לֵב entry stretching over 14 columns.
 Shalom Paul, Amos (ed. Frank Moore Cross; Hermeneia; Minneapolis: Fortress, 1991) 209-210.
 David J. A. Clines, Job 1-20 (WBC 17; Dallas: Word, 1989) 161.