ArDO: Yes we want Lebanon to be the Switzerland of the East and Beirut the Paris of the East

 

 | HomeArchives  | Links|

 

 Device  Articles  Politics  Language  Pictures  History/Culture  History 

HAYIM TADMOR (JERUSALEM)

ON THE ROLE OF ARAMAIC IN THE ASSYRIAN EMPIRE   

 

 

 

 

HAYIM TADMOR

Contemporary research has thrown increasing light on a highly significant facet of culture in the Assyrian Empire: the symbiosis of Aramaic and Akkadian[1]. A surprising breakthrough in this respect was the publication of the bilingual inscription, cuneiform and alphabetic, from Tell Fekheriye in Syria, incised on the statue of a ruler from Gozan, which is written in archaic style in Akkadian and Aramaic [2]. We can now state fairly confidently that bilingualism was current on the Western periphery of Assyria, the bulk of whose population consisted of Arameans, at the very least from the mid ninth century B.C.E. onward [3].

However, it was only some 120 years later, when the territories west of the Euphrates were conquered that this symbiosis was officially recognized and Aramaic became the second language of the empire, alongside Akkadian.[4]

1.     See: A. L. Oppenheim. Letters from Mesopotamia. Chicago 1967. pp. 42-48;.J Muffs. Studies in the Aramaic Legal Papyri from Elephantine, Leiden 1969. pp. 189-190; H. Tadmor, The Aramaization of Assyria, in: J. Nissen & J. Renger, Mesopotamien und Seine Nachbarn (XXV. Rencontre Assyriologique Internationale 1978). Berlin. 1982. pp. 449 -469; A. R. Millard. Iraq XLV (1983) [XXIX Rencontre Assyriologique Internationale. 1982). pp. 101-107; F.M. Fales, Aramaic Epigraphs on Clay Tablets of the New Assyrian period  (Studi Semitici. nuova serie 2). Rome. 1986. pp. 36-47 (henceforth. Hales.Epigraphs).

2.     See: A. Abu-Assaf. P Bourdreuil & A. R. Millard. La Statue de Tell Fekherye, Paris. 1982; J.A. Kautman. Maarav III (1982) pp. 137-175; R. Zadok. Tell Aviv IX (1982). pp. 117-129; J.C.  Greenfield & A. Shaffer. Iraq XLV (198.w). pp. 109- 116; idem. Anatolian Studies XXXIII (1983), pp. 123 129; idem. RB XLII (1985). pp. 45-59; F.M. Fales. Svria XL. (1984). pp. 233-250; idem, Epigraphs, pp. 40-43. 

3.     For an early interchange between Akkadian and Aramaic verbs ”to kill” (daku > qatalu) in the Assyrian scribal practice of the tenth century, see my note: Towards the Early History of qatalu, The Jewish Quarterly Review LXXVI (1985), pp. 51-54.

4.      This may be seen as a reversal of the practice current in the 2nd millennium B.C.E in the lands west of the Euphrates. Then it was a Western Akkadian that was served as the language of literacy and lingua franca throughout the whole area, including Egypt. The cuneiform archives of the royal chancelleries at el-Amarna, Urgarit, and Hattusha document this practice. The literary centers of Canaan (e.g. Hazor, Aphek), where the alphabetic script was already incipient use, yielded evidence of extensive use of Akkadian. For cuneiform texts from Hazor see:

             B. Landsberger  & H. Tadmor IEJ XIV (1954), pp. 201-218; W.W. Hallo & H. Tadmor,

             IEJ XXVII (1977), pp. 1-11, H. Tadmor, ibid, pp. 98-102: For those form Aphek see:

             A.F. Rainey, Tel Aviv II (1975), pp. 125-129; III (1976), pp. 137-140; D.I. Owen

             Tel Aviv VIII (1981), pp. 1-17;  

Assyrian reliefs beginning from the time of Tiglath-pileser III [5] provide nurnerous portrayals of a scribe writing on a tablet or a board,[6] side by side with another scribe writing on papyrus or a parchment scrolI.[7] Such scribes would record the loot taken in battle or count the number of enemy casualties.[8]

This pictorial rendition undoubtedly corresponds to the phrases "Assyrian scribe" (tupsharru Ashuraya) and "Aramaic scribe" (tupsharru Aramaya) that occur together in the various documents, referring to officials in the imperial service.[9] The "Aramaic scribe" was of particular importance in the western part of the empire, where the royal correspondence was conducted also, or according to some authorities primarily, in Aramaic.[10] As the Aramean elements in Assyria

             W. Hallo, ibid., pp.. 18-24. For the early alphabetic ostracon from 'Izbet Sartah, see M. Kochavi,

             Tel Aviv IV (1977). pp. 1 13. A, Demsky, ibid. pp. 14~27.

 

5.    See. eg,. R.D. Barnett & M. Falkner The scriptures of Tighlath-Pileser Ill, London. 1962, Pls.

 V-VI; G. R. Driver. Semitic Writing. London, 1948. Pls. 23B 24C; D.J. Wisernan, Iraq XVII

              (1955), Pl. III, 2.

6.     D.J. Wiseman (above. n. 5, pp. 8-13) has argued convincingly that the rectangular object the

             scribe is holding is not a clay tablet, but rather an ivory or a wooden board (in fact. a hinged

             diptych). covered with wax. 

7.     For textual evidence of writing on papyrus (niyaru and urbanu). see CAD, N, II, p. 201;

W. von Soden. AHw p. 1428. Writing on parchment is attested mainly from Neo-Babylonian documents; see CAD. M. I. p. 31 S.V. magallatu (an Aramaic loanword). In Neo-Assyrian documents, parchment (KUSH nayaru. CAD. N, II. p. 201) is rarely mentioned.

       8.   I. Madhloorn has suggested that the person holding a scroll was not a scribe but an artist, whose

             task was to sketch battle scenes that would later be carved on stone reliefs.

             (Acta Antiqua Academiae Scientiarium Hungaricaie XXII (1974). p. 385).

             I believe however that conclusive evidence against this thesis may be furnished by a fresco

            from Til Barsip, a major Assyrian administrative center in the western part of the Empire

            (F. Thureau-Dangin-M.Dunand.Til Barsib, Paris. 1936. Pl. pp. 54-55). It portrays two scribes

            (Fig. 1), one writing on a tablet and the other on a sheet of papyrus. standing behind

            three courtiers who are facing the king (only the lower part of his figure survives, with part

            of a  crouching lion). Since this fresco is concerned not with a military campaign in a

           foreign land,  but with a ceremony at the royal court, it stands to reason that the person

           holding the sheet of papyrus, standing beside the scribe with the tablet, is, not an artist but

           likewise a scribe, recording the royal instructions in Aramaic.

9.     See the evidence collected by J. Lewy, in Hebrew Union College Annual XXV (1954), pp.

      185-190. For the list of dignitaries. K. 4395. in which the tupsharru Ashuraya is followed by the

             tupsharru Ar(a)maya (p. 188 rr. 74), see now MSL, Xllr p. 239 V: 5-6.

10.  See: S Parpola, `Assyrian Royal Inscriptions and Neo-Assyrian Letters`, in F. M. Fales (ed.),

             Assyrian Royal Inscriptions: New Horizons, Rome, 1981, pp. 122-123. It is not surprising that the

             Neo-Assyrian Empire in the west needed Aramaic as a kind of lingua franca for its official

             dealings. After all, Assyria had inherited what had been the kingdom of Darnascus, with its

             existing administrative system; see B. Mazar. "The Aramean empire and its relations with Israel`,

             BA XXV (1962). pp. 111-112 (now in his "The Early Biblical Period: Historical Studies, Jerusalem

            1986, pp. 163-165).

gained ascendancy particularly as a result of mass deportations.[11] scribes in the

capitals of the empire were obliged to acquire proficiency in both scripts, cuneiform and alphabetic. Indeed, there are economic documents dating from the seventh century written on clay tablets in Akkadian with annotations [o…….] even summaries in Ararnaic.[12] To my mind. such documents were most likely written by a single scribe, fluent in both languages.[13] The same pertains also to the oracular queries on certain state matters, put before the Sun-God Shamash and the patron of extispicy. lt is stated there that very often the tablet of query was

accompanied by a slip of papyrus (niyaru, urbanu) which carried the name of the

person concerned  - the subject of the query-  inscribed, no doubt, in Aramaic.[14]

Through such bilingual scribes Akkadian absorbed not only Aramaic terms in many areas such as administration, literacy, and even warfare,]15] but also spelling conventions characteristic of alphabetic script.[16] In Babylonia, during the period of the Assyrian Empire and particularly under the Neo-Babylonian kings, there was even a special term, sepiru, borrowed from Aramaic. to denote the bilingual scribe.[17] This term was written sometimes phonetically (se-pi-ru) and sometime ideographically: (A. BAL, literally "one who converts, transposes." i.e. a person who reads a text in one language and translates it into another) [18] A related term is targummanu . "interpreter." already known in Old Assyrian and Western Akkadian of the second millennium B.C.E.; the corresponding ideogram is EME. BAL "one who converts speech."[19] This term apparently signifies a person who translates oral communications, in contrast to the sepiru, who is concerned with writing.[20]

Of special significance in this connection is the relief of Sargon II from his palace at Khorsabad which portrays the siege of a city in a hilly country (Fig. 2)."

11.  See: B Obed,. Mass Deportations and Deportees in the Neo-Assyrian Empire, Wiesbaden. 1978.

12.   See: S.J. Lieberrnan. BASOR 192 (1968), pp. 25-31; J. Naveh, The Development of the

              Ararnaic Scripts`, The Israel Academy of Sciences and Humanities, Proceedings, V/I (1970),

              pp. 16-17, and now in detail, Fales, Epigraphs.

13.   See: `Ararnaization of Assyria` (above. n 1); p. 453.

14.  See. J.A.  Knudtzon. Assyrische Gebete an den Sonnengott. Leipzig, 1983, Nos 50:4. 124:4, 131 r.2.

15.   See. `Ararnaization of Assyria` (above. n. 1), pp. 454-455.

16.  See: A. Poebel. Studies in Akkadian Grammar (Assyriological Studies, lX),(Chicago 1939.  pp. 60-64.

17.  Cf. J. Lewy (above, n. 9). pp. 191-199.

18.  Ibid, p 196, n. 108.

19.  See: I.J. Gelb ‘The Word for Dragoman in the Ancient Near East’, Glossa II (1968) pp. 93-104.

20.  See: J. lewy (above, n. 9), pp. 196, n. 108. 

21.  See: P. E. Botta & E. Flandin, Monuments de Nineve, Paris 1849, II, Pl. 145; IV, Pl. 180: 1-2 M El-Amin, Sumer IX (1953), p. 219-225.

As the late Y. Yadin observed, an officer, leaning out of the turret of a siege

Rnachine, holds a scroll in his hands, apparently appealing to the besieged inhabit-

ants to surrender (Fig. 3),[22] Yadin suggested that this scene recalls the biblical

description of Rab-shakeh,  the royal chiefcupbearer, who called upon the people

of Jerusalem to surrender before Sennacherib. his master. (II Kings 18: 17-35).

According to the cuneiform epigraph inscribed across the relief, the city be-

sieged by Sargon was "Pa(?)-za-shi, the fortified city of Mannea."[23] Hence, the

language used by the officer. who is holding the scroll and addressing the people

of the city, must have been Mannean. As, to the best of our knowledge, the

Manneans did not possess any script for their language, it stands to reason that the

scroll in the officer`s hands was inscribed in Aramaic, like any other scroll in the

hands of scribes on Assyrian reliefs. Such an officer might have been an "lnterpreter of Mannean, targummanu  sha Mannaya, a term attested in a Neo-Assyrian document.[24] Naturally. a person holding a text in Aramaic and translating it aloud. would be an

Assyrianized Mannean raised in Assyria as a hostage or a deportee.

By analogy, can one surmise here that Rab-shakeh too was reading from an

Aramaic scroll when delivering his message to the besieged population of

Jerusalem? The appeal of the Judean nobles to Rah-shakeh. “Please, speak to

your servants in Aramaic, for we understand it; do not speak to us in the language

of Judah in the hearing of the people (standing) on the wall" (ll Kings 18:26)

indicates that they expected the envoy of the Assyrian king to address them in

Aramaic, the customary language of diplomatic negotiations in the West. Rab-

Shake, however, had a surprise in store for them: he harangued the people on

the ramparts of Jerusalem directly, speaking in the vernacular.

The remarkable features of the Rab-Shake narrative in the Bible are not only

the ability of an Assyrian envoy to deliver an eloquent speech in the Judean

tongue, but also the very appearance of the kings chief cupbearer as the royal

spokesman in time of war. I know of no other case of the chief cupbearer par-

ticipating in a military delegation, together with the Tartan (Tartanu, the Viceroy)

and the Rab-saris (rab-sha-reshi, originally the chief eunuch. later the commander in

chief).[25] In addition. at that period both these courtiers outranked Rab-shakeh.

 22. The Art of Warfare in Biblical Lands JerusaIern~Ramat-Gan, 1963, , pp. 320 and 425.

 

23. See: C.B.F. Walker, ‘The Epigraphs` in P.Albenda, The Palace of Sargon , King of Assyria,

      Paris, 1986, p. 112.

 24. See: C.H.W Johns, Assyrian Deeds and Documents,, ll, Carnbridge, 1901, No. 865: 6-8.;

 25. It is regrettable that in the past many scholars failed to identify properly the titles ‘Rab-Shakeh`

and ‘Rab-saris` in ll Kings I8, both Akkadian Ioanwords. A common error was that the title written ideographically LU GALSAG stands for ‘Rab~shakeh’. Examples of this misconception

may still be found in the standard collections of Near Eastern texts in translation. Nowadays

 Why then. of the three members of the Assyrian delegation was the chiefcup

bearer chosen to conduct the negotiations with King Hezekiahs representatives.

As a possible explanation of this choice, I would suggest that Rab-shakeh, a [….]

alone of the Assyrian entourage, was fluent in Judean and capable of delivering a

convincing propaganda speech in that language. It is inconceivable that the

cupbearer of Sennacherib, who was certainly not an expert in Semitics.

have acquired this fluency. if he were not a Westerner in origin: an Israelite,

Moabite or Ammonite.[27] Highly-placed officials at the Assyrian court, v[…]

names betray an origin west of the Euphrates. are mentioned in numerous monuments

from the eighth century onward.[28] In a late Babylonian tradition. Ahikar

the hero of an Aramaic narrative, was ummanu (i.e. the counselor and scribe of

King Esarhaddon,[29] and Nehemiah, another Westerner, attained the position

"the kings cupbearer" in the court of Artaxerxes l (Neh. 1:11). Such ap[…]

ments were surely no exception in the history of royal courts. Indeed, they were

typical of Assyria. the only one among the empires of the ancient Near East

which the language of the conquered. forcefully acculturated, ultimately

vailed over the language of their imperial masters.*

      there can be no doubt that this title was pronounced rab-sha-reshi, whereas the usual ideographical

     writing for the chief cupbearer (rab-sha-qe-e), was LU GAL BI.LUL. More in my discussion about

     these two titles in C.L. Meyers and M. O’Connor (eds.), Eassays in Honor of David Noel […]

     man. Philadelphia. 1983. pp. 279-295.

 26. Cf. my comments in the Hebrew Encyclopaedia Biblica , vol. VII, col. 324, and a rnore recent […]

      Cohen, Israel Oriental Studies IX (1979), pp. 32-47.

 27.  As suggested to me by my colleague. Prof.  J. Naveh. `I`he rnain document in Moabite is st[…]

       Mesha monument. For Ammonite see now K. P. Jackson, The Ammonite Language of the [….]

       Age (Harvard Semitic Studies, XXVII). Cambridge. Mass., 1983.

 28. See; Oded (above. n. ll), pp. 105-107.

 29  See: J.J A. van Dijk, in Uruk Vorläufiger Bericht XVIII, Berlin, 1962, pp. 44-45; J. C. Greenfield

      in: Hommages a A. Dupont-Sommer. Paris. 1971, pp. 50-51.

·         The same topic is treated in a paper which appears now in Hebrew (Eretz Israel XX [1989][….]

249-252). The present English version has been modified and revised. It is a privilege to publish

             it in the Anniversary Volume for H.I.H. Takahito Mikasa. historian of the Ancient Near East

             and patron of scholarship.

 

The articles published on this site represent the opinion of their writers and not the opinion of the webmasters.