OFFPRINT FROM

Birkat Shalom Studies in the Bible, Ancient Near Eastern Literature, and Postbiblical Judaism Presented to Shalom M. Paul on the Occasion of His Seventieth Birthday

Edited by

CHAIM COHEN, VICTOR AVIGDOR HUROWITZ, AVI HURVITZ, YOCHANAN MUFFS, BARUCHJ. SCHWARTZ, and JEFFREY H. TIGAY

Copyright 2008 by Eisenbrauns. All rights reserved.

Winona Lake, Indiana EISENBRAUNS 2008

On rahasum I, II, III and on Akkadian rihsum = Hebrew I NATHAN WASSERMAN The Hebrew University of Jerusalem

More than half a century ago, W. von Soden (1947) published a short lexicographical note entitled "rahdsu I und II." This preliminary note was the basis for his AHw entries on rahdsu I and rahdsu II. Akkadian dictionaries now distinguish three or even four seemingly homonymous rahasum verbs. CDA 295 lists three different verbs: rahasum I 'to flood; wash' (said mainly of Adad) (i/i) and 'to wash, cleanse' (fl/u); rahasum II 'to trust' (i/i, Bab. u/u); and rahasum III 'to hold a debate' (a/a, i/i). CAD R 69-75 splits rahasum I into two and records four different verbs: rahasum A 'to trample, to kick, to destroy, to devastate' and also 'to hurry(?)' (i/i); rahasum B 'to wash, to bathe' (a/u); rahasum C 'to trust, to rely' (u/u); and rahasum D 'to gather(?)' (i/a, i/i). As mentioned above, AHw 942-43 registers only two verbs: rahasum I, which has been divided, based on the root vowel, into A 'iiberschwemmen' (i/i) and B 'spiilen' (a/u); and rahasum II Vertrauen'. Thus, rahasum I in AHw and CDA comprises rahasum A and rahasum B in CAD. This lexical situation calls for investigation, because it is highly improbable that three or even four diachronic homonymous verbs, unambiguously distinctive from a semantic point of view and by their root vowel, Author's note: In addition to the abbreviation found in the CAD and the AHw, the following abbreviations are specific to this essay. AEL Lane, E. W. Al-Qamusu: An Arabic-English Lexicon. London, 1863-93 CDA Black, J.; George, A.; and Postgate, N. A Concise Dictionary of Akkadian. SANTAG 5. Wiesbaden: Harrassowitz, 1999 DJBA Sokoloff, M. A Dictionary of Jewish Babylonian Aramaic. Ramat Gan: BarIlan University Press / Baltimore: Johns Hopkins University Press, 2002 DULAT del Olmo Lete, G., and Sanmartin, J. A Dictionary of the Vgaritic Language in the Alphabetic Tradition. Leiden: Brill, 2003 I am grateful to Dr. J. N. Ford, the Hebrew University of Jerusalem, who agreed to read a draft of this essay and offered me his helpful remarks. 705

NATHAN WASSERMAN

On rahasum I, II, III and on Akk. rihsum = Hebrew mX57

would historically derive from the same verbal stem.1 In this essay, I will argue that each of the homonymous rahasum verbs has a different etymology and different origin (that is, their homonymy is only synchronic rather than diachronic), and consequently their respective appearances in the various chronolects of Akkadian exhibit an exact complementary distribution. I will also suggest that rihsum (from rahasum III), which designates 'public debate' or 'gathering', corresponds etymologically and semantically to BH rmS7 and HISS? 'assembly'.2 This short essay, dealing with both Akkadian and Hebrew, is offered with appreciation to Prof. Shalom Paul, whose scholarship and teaching have bridged these two kindred languages for many years. The hrst of the verbs under consideration, rahasum I meaning 'to flood, to wash' and more abstractly 'to devastate', represents the common Semitic root *RHD. It is found in Heb. fm and Ug. rhs, in Arab, rahada, and in Aram. 27m.3 In later Aram., probably influenced by the Heb. cognate verb, it is attested also as 2 f m (DJBA, 1071). This verb is widely attested in all chronolects of Akkadian, from the OAkk period to late first-millennium texts. The verb rahasum II meaning 'to trust' is equated in an ancient commentary with the more common verb takfllum. 4 Unlike rahasum I, no Sumerogram is used for rahasum II. Etymologically and semantically, rahasum II has a clear counterpart in Aram. 1 fm 'to trust'.5 Various lexicographers have mentioned Arab, rahhasa 'to permit, allow; to authorize, license; to empower' (AEL 1058C, 2) as a possible candidate for the etymon of Aram. 1 fm. 6 Whether *RHS can thus be considered the etymon of Akk. rahasum II is far from certain, but whatever the Semitic root of this verb may be, Aram. 1 f m is etymologically and semantically equivalent to Akk. rahasum II. It is important to note that all the references to rahasum 'to trust' collected in CAD are found only in NA and NB sources and in SB royal inscriptions-

that is, in first-millennium texts (CAD R 74-75). In fact, all known references to rahasum II are post-750 B.c.E.7 This fact, together with its etymological and semantic ties to Aram. 1 f m indicate that the verb rahasum II 'to trust' is in all likelihood a direct lexical import from Aramaic-another example of the decisive influence that Aramaic had on Akkadian during the Sargonid period.8 The late entrance of this verb into Akkadian and its foreign origin also explain the absence of any logographic writing for rahasum II. The last of the verbs under consideration is rahasum III, which denotes 'to hold a debate' (CDA 295b) or 'to gather(?)' (CAD R 75). This verb (unknown to AHw) is entirely restricted to OB Mari sources. A thorough discussion of rihsum, a nominal derivative of this verb, was offered by J.-M. Durand. Based on a careful examination of the Mari material, Durand was able to ascertain the meaning of this noun as 'palabre', expanding on the political and ceremonial functions of this institution that were typical in various Hana tribal groups.9 Durand's lexical investigation was followed by the dictionaries, which render this lemma 'gathering' (CAD R 336-37, s.v. rihsu B) or 'public debate' (CDA 304a, s.v. rihsu II). Heimpel, in his study on Mari letters (2003: 199-200 n. 3) defined rihsum as 'parley' and remarked that: "the terms 'parley' (rihsum) and 'parleying' (rahasum') designate the function of an assembly whereby individuals could voice their opinions frankly. The expenditure of a lamb for a 'parley' indicates that it had a ceremonial counterpart." I propose here that the etymology of rahasum III is a hitherto unrecognized Amorite root,*cSR, corresponding to Heb. 1SV*.10 More specifically, I

706

1. This case of three synchronic homonymous verbs is of course not unique. See, for example, enenum I (a/u) 'to grant favour, be favorable', enenum II (i/i), 'to punish', and enenum III (e/u) 'to sin' (CDA 73a'b). 2. Durand 2004: 189 n. 428: "Ce terme . . . est, aujourd'hui, sans etymologic semitique sure." 3. Ugaritic: Leslau 1968: 364 (no. 2323); DULAT 738. Arabic: AEL 1052. Aramaic. Greenfield 1960: 100; 1992: 592-93. 4. Izbu Comm. V 254c (cf. CAD R 74, lex. sec.). 5. DJBA 1070-71; HALOT 1981b. 6. BOB 1113a; HALOT 1981b.

707

7. AHw 943a: "bab. u ass. Nach 750": Note that von Soden lists here one OAkk attestation of this verb, UM 1/2, 1:15. The classification of this text, however, needs to be changed, because it is not an OAkk but an early OB letter (see Stol, AbB 11,1 with earlier references). Furthermore, this source must be cited not under rahasum II 'to trust' but under rahasum I, with the idiomatic meaning 'to hurry', as accepted now in CAD R 72b, 2. Note finally that Mayer (2003: 232) has now furnished another case of OB use of rahasum I with the meaning 'to hurry': Eidem and Laess0e 2001:43:12. 8. Cf., e.g., von Soden 1966; 1968; 1977; Tadmor 1982; and Greenfield 1984: 263. Here, I purposely use the term 'import', because the verb rahasum II is fully integrated into Akk. and cannot be called a "loanword™ or even "loan adaptation" (for these terms, see Mankowski 2000: 167ff.). 9. Durand 1988: 181-92. See also 2000: 582. Most recently this term was discussed in 2004: 189-91. 10. Not in Streck 2000.

708

NATHAN WASSERMAN

suggest that rihsum, the nominal derivative of rahasum III,11 corresponds etymologically and semantically to Heb. mxi? and mx» 'celebration, festive assembly' (HALOT 871-72). Semantically, rihsum in the Mari texts matches Heb. mxv and niXS perfectly, as the biblical references prove. Both rihsum and mx» denote an assembly summoned for a political or religious purpose (see esp. Joel 1:14, 2:15; 2 Kgs 10:20-22). The crucial role of parleying in this kind of crowded assembly is clearly conveyed byjer 9:1-5. Epigraphically, this etymology is based on the OB writing convention, used especially in the Mari texts, according to which the original consonants ', c, h, h, h, and g are represented with the H-series (that is, the signs /ah/, /ih/, /uh/, and /ha/, /hi/, /hu/). 12 Thus, ri-ih-su-um in the Mari letters is the cuneiform-writing realization of ri(sum. Turning to phonology, one observes that this proposal assumes that metathesis occurs between the Heb. and Akk. (in fact Amor.) corresponding roots: *CSR and *RCS, respectively. Metathesis of this sort is known in other cases as well. Note the following examples (alphabetized by the Akk. lemmata, with permutation of root radicals indicated):13 1. Akk. diparum vs. Heb. TD1? 'torch' (1-2-3 vs. 3-2-1 );14 2. Akk. esenum 'to smell (an odor)' vs. Heb. mnx, JITO* 'stench' or Aram. 'stinking fluid' (1-2-3 vs. 2-1-3);15

11. Not to be confused with the nominal form rihsu (from rahasum I 'to wash') in MA texts, designating a specific stage in processing grain, meaning 'washed' grains (discussed by Postgate 2006). 12. Streck 2000: 246, §2.171: "/V wird durch die H- oder 0-Reihe wiedergegeben." In §§2.171-2.173, different verbal roots with /c/ are listed (e.g., *CDB, *CDR, *CLY, *CMN, *CQB, *CZZ, *BCL, *YDC, *YPC, *YSC), in all of which the /'/ is graphically represented with signs from the H-series. 13. Zohari 1991 is the sole monograph dedicated to metathesis in Hebrew that is known to me. This work, although helpful in its collection of material, should be treated critically and with great caution. 14. AHw 172b; CAD D 156-57; HALOT 533b. In my opinion, the century-old proposition in Zimmern 1917 (accepted by Segert 1962 but not mentioned in AHw or in Mankowski 2000), according to which there is an etymological connection between Akk. diparum and Heb. TO1?, is convincing and should be preferred to the more recent suggestion of connecting Heb. TQ1? with Greektaxujra!;(cf. HALOT 533b with earlier lit.). For the /r/-/l/ consonant alternation, see Lipiriski 1997: 134-36 (§§17.3-17.6) and Segert 1962. 15. AHw 252a; CAD E 344-45; HALOT 1018-19.

On rahasum I, II, III and on Akk. rihsum = Hebrew

709

3. Akk. hallurum 'peas' vs. Heb. Vnn/D'Vnn 'weed in held and fruit garden' (1-2-3 vs. 1-3-2);16 4. Akk. lahrum vs. Heb. Vm 'ewe' (1-2-3 vs. 3-2-1); 17 5. Akk. lamattum < *lamantum vs. Heb. rtej 'ant' (1-2-3 vs. 3-2-1);18 6. Akk. lurmum/lurinum/nurmumvs. Heb. |W1 'pomegranate' (1-2-3 vs. 2-3-1); 19 7. Akk. pasdnum/pasamum/pesenum 'to cover up, to veil' vs. Heb. |SS* 'to hide, to conceal oneself (and cf. Heb. «]3S* 'to wrap around, wind around' said of turban or a garment; 1-2-3 vs. 2-1-3 or 2-3-1);20 8. Akk. simmiltum vs. Heb. aVlO 'stepped ramp, flight of steps' (1-2-3 vs. 1-3-2).21 Thus, though not inclusive in any way, this list prompts two observations: different kinds of permutation of root radicals exist when one compares Akk. roots with their West-Semitic cognates.22 The permutation that took place in Akk. *RCS vs. Heb. *
16. AHw 313a; CDA 102b; CAD H 47-48 (cf. esp. note at the end); HALOT 351b (for Vnri j71Paa in Zeph 2:9, see Greenfield 1982). 17. AHw 528a; CAD L 42-44; HALOT 1216a-b. 18. AHw 533a; CAD L 67b; HALOT 701a. Lamattum and namalum (AHw 725a; CAD N/1 208a) are both rare synonyms of kulbabum, the more common word for 'ant'(CAD K 501-2). 19. AHw 564-65, 804-5; CAD L 255-56; CAD N/2 345ff.; HALOT 1241-42. 20. AHw 840a, 838a, 856b; CDA 268a-b; HALOT 1039b and 1049a. Note the WestSem. verb sapanum 'to hide' (CAD S 96b) without the metathesis, attested in EA 147: 10. Cf. Grave 1982. 21. AHw 1045; HALOT 757-58; and Cohen 1997: 22. Cf. Mankowski 2000: 114-18. 22. The fact that these permutations in the root radicals take place proves that the corresponding Akk. and Heb. roots are true cognates and not words that were borrowed from one language to another, because, as shown in Mankowski (2000: 159): "Akk. r, I and n are preserved in all positions in direct BH borrowings." 23. Regarding the phenomenon of metathesis in Semitic, Lipiriski (1997: 193, §27.13) mentions that "there is little doubt that one of the consonants involved in many cases [of metathesis] is either I or r, i.e. one of the two 'liquids.'"

710

NATHAN WASSERMAN

On rahasum I, II, III and on Akk. rihsum = Hebrew mxs?

tendency, however, is not a strict condition, as other cases of metathesis (e.g., Akk. dispum vs. Heb. HOT 'honey') prove. Summing up, Akk. rihsum 'gathering' or 'parley' is etymologically and semantically connected to Heb. mX27 and mxs 'assembly'. The Akk. root *RCS, realized in cuneiform writing as rajiasum, is of Amorite origin; hence the restricted attestation of rahasum III and of rihsum in the Mari corpus only. The verb rahasum II 'to trust' entered Akkadian from Aramaic, a fact that explains its limited appearance in the late chronolects of Akkadian. The verb rahasum I 'to wash' is originally Akkadian, common Semitic in fact, and therefore documented in all periods of Akkadian. The following table outlines the distribution of rahasum I, II, and III in the various Akkadian chronolects.

<

1

I MB/MA

Akkadian rahasum I, II, III |Z CO" « O Z

+

+

+

+

-

+

-

+

-

-

^ Verb

Root

*RHE> (common Semitic, original Akk.) rahasum I 'to wash' rahasum II 'to trust' *RHS (Aram, import) rahasum III 'to gather' *RCS ~ *CSR (Amor, import)

(Man)

Bibliography Cohen, C. 1997 The Literary Motif of "Jacob's Ladder" (Gen 28:12). Pp. 15-26 inHadassha Shy jubilee Book, ed. Y. Ben Tolila. Beer-sheva: Ben-Gurion University Press. Durand,J.-M. 1988 Annexe: Le rihsum des Haneens. Pp. 181-92 in Archives epistolaires de Mari I, ed. J.-M. Durand. ARM 26/1. Paris: Editions Recherche sur les Civilization. 2000 Documents epistolaires du palais de Mari, vol. 3. LAPO 18. Paris: Cerf. 2004 Peuplement et societes a 1'epoque amorrie (I): Les clans bensim'alites. Pp. 111-97 in Nomades et sedentaires dans le proche-orient ancien: Compte rendu de la XLVIe Rencontre Assyriologique Internationale, Paris, 10-13 juillet 2000, ed. C. Nicolle. Amurru 3. Paris: Editions Recherche sur les Civilization. Eidem, J., and Laess0e, J. 2001 The Shemshara Archives 1: The Letters. Historisk-filosofiske Skrifter 23. Copenhagen: Royal Danish Academy of Sciences and Letters.

711

Fales, F. M. 1984 Assyro-aramaic: Three Notes. Or 53: 66-71. Grave, C. 1982 Northwest Semitic sapanu in a Break-up of an Egyptian Stereotype Phrase in EA 147. Or 51: 161-82. Greenfield,]. C. 1960 "Le bain des brebis": Another Example and a Query. Or 29: 98-102. 1982 A Hapax Legomenon: Vnn j?tt>a». Pp. 79-82 in Studies injudaica, Karaitica and Islamica Presented to Dr. Leon Nemoy, ed. S. Brunswick et al. Ramat Gan: Bar-Han University Press. 1984 Ana urduti kabasu = 132?V 1P3D. StOR 55: 257-63. 1992 The Verbs for Washing in Aramaic. Pp. 1588-94 in Semitic Studies in Honor ofWolfLeslau on the Occasion of His Eighty-Fifth Birthday, November 14th, 1991, ed. A. S. Kaye. 2 vols. Wiesbaden: Harrassowitz. Heimpel, W. 2003 Letters to the King of Mari: A New Translation, with Historical Introduction, Notes, and Commentary. Mesopotamian Civilizations 12. Winona Lake, IN: Eisenbrauns. Leslau, W. 1968 Observations on Semitic Cognates in Ugaritic. Or 37: 347-66. Lipiriski, E. 1997 Semitic Languages: Outline of a Comparative Grammar. OLA 80. Leuven: Peelers. Mankowski, P. V 2000 Akkadian Loanwords in Biblical Hebrew. HSS 47. Winona Lake, IN: Eisenbrauns. Mayer, W. 2003 Akkadische Lexikographie: CAD R. Or 72: 231-42. Postgate, N. 2006 Rihsu, /lisnu and sihtu, but Not Bulgur. NABU 2006, no. 12. Segert, S. 1962 Zur etymologic von Lappid 'Fackel'. ZAW 74: 323-24. Soden, W. von 1947 Rahasu I und II. Or 16: 456-57. 1966 Aramaische Worter in neuassyrischen und neu- und spatbabylonischen Texten: Ein Vorberichte I (aga-*mus). Or 35: 1-20. 1968 Aramaische Worter in neuassyrischen und neu- und spatbabylonischen Texten: Ein Vorberichte II (n-z und Nachtrage). Or 37: 261-71. 1977 Aramaische Worter in neuassyrischen und neu- und spatbabylonischen Texten: Ein Vorberichte III. Or 46: 183-97. Streck, M. P. 2000 Das amurritische Onomastikon der altbabylonischen Zeit I: Die Amurriter, die onomastische Forschung, Orthographie und Phonologic, Nominalmorphologie. AOAT 271/1. Minister: Ugarit Verlag.

712

NATHAN WASSERMAN

Tadmor, H. 1982 The Aramaization of Assyria: Aspects of Western Impact. Pp. 449-70 in Mesopotamien und seine Nachbarn: Politische und kulturelle Wechselbeziehungen im alten Vorderasien vom 4. bis 1 Jahrtausend v. Chr. Akten des XXVe Rencontre Assyriologique Internationale, Berlin, 3. bis 7.]uli 1978, ed. H. Nissen andj. Renger. Berliner Beitrage zum Vorderen Orient 1. Berlin: Reimer. Zimmern, H. 1917 Akkadische Fremdworter als Beweis fiir babylonischen Kultureinfluss. Leipzig: Leipzig University Press. Zohari, M. 1991 rma»n nstZO nniX 'Vsai nTTllX 'VlD^ (The Metathesis and Dual Forms in the Hebrew Language). Jerusalem: Carmel.