Journal of Cuneiform Studies Volume 61
2009 Editor Piotr Michalowski, University of Michigan, Ann Arbor Associate Edito...
62 downloads
1525 Views
30MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
Journal of Cuneiform Studies Volume 61
2009 Editor Piotr Michalowski, University of Michigan, Ann Arbor Associate Editors Gary Beckman, University of Michigan, Ann Arbor Elizabeth Carter, University of California, Los Angeles Piotr Steinkeller, Harvard University Matthew W. Stolper, The Oriental Institute, University of Chicago Managing Editor Billie Jean Collins, Emory University
CONTENTS Daniel Potts, Bevel-Rim Bowls and Bakeries: Evidence and Explanations from Iran and the IndoIranian Borderlands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1
Wolfgang Heimpel, The Location of Madga . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
25
Eva von Dassow, Naram-Sîn of Uruk: A New King in an Old Shoebox . . . . . . . . . . . . . . . . . . . . . . . . . . . .
63
Anne Kilmer and Jeremie Peterson, More Old Babylonian Music-Instruction Fragments from Nippur . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
93
Jerome Colburn, A New Interpretation of the Nippur Music-Instruction Fragments . . . . . . . . . . . . .
97
Jeanette C. Fincke, Zu den akkadischen Hemerologien aus {attusa (CTH 546), Teil I. Eine Hemerologie für das „Rufen von Klagen“ (sigû sasû) und das „Reinigen seines Gewandes“ (subat-su ubbubu): KUB 4, 46 (+) KUB 43, 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 Philip C. Schmitz, Archaic Greek Names in a Neo-Assyrian Cuneiform Tablet from Tarsus. . . . . . . 127 Texts and Fragments Barbara Böck, Three New Sources of Mussuåu. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
BEVEL-RIM BOWLS AND BAKERIES: EVIDENCE AND EXPLANATIONS FROM IRAN AND THE INDO-IRANIAN BORDERLANDS Daniel Potts (The Universitij of Sydnexj)
The handmade (Karlsbeck 1980; Chazan and Lehner 1990; 25) or moldmade (Balfet 1980: 78, Miller 1981; 128) ves.sel known as the "bevel-rim" or "bevelled-rim" bowl (hereafter BRB) is (haracterized by a coarse, chaff-tempered, highly porous fabric, fired at a low temperature. With its distinctive, often sloppily indented (bevelled) rim and rough exterior (fig. 1), the BRB is easily identified and, once seen, rarely mi.staken for anything else. Although considered a characteristic Mesopotamian ceramic ¡eiffossil of the mid- to late-fourth millennium lie:, the first BRBs ever published were actually discovered in Iran, M Susa, during tbe seasons of 1897/98 and 1898/99 (de Morgan 1900:figs.91, 118, 121). in the winter of 1902/3 at least one complete BRB, later displayed in tbe Louvre, was recovered by Gautier and Lampre at Tepe Musiyan (Burton Brown 1946: 36). The first examples published from a Mesopotamian site were those found at Tell Abu Shahrein (ancient Eridu) in 1918 (Campbell Thompson 1920; tigs. 3.4 and 4.10). In 1925/26 .six BRBs were found at Jamdat Nasr (Mackay 1931: pi. 67.22-23), prompting Ernest Mackay to observe, '"The combination of beveled rim with a rough appearance should be of use in dating other sites where they migbt be found" fMackay 1931: 250). Henri Frankfort obviously concurred, for a year later be cited BRBs—a "rough bowl with thick walls, beveled at the rim"—among the diagnostic shapes of the Uruk period in his classic study of the "Sumerian problem" (Frankfort 1932: 17, n. 3). In 1928 more BRBs from Susa were published (Allotte de la Füye, Cumont, and de Mecquenem 1928; 102.fig.1.4) and during the next few years BRBs were recorded in Assyria for the first time during the British Museum's excavations (seasons of 1929/30-1931/32) at Nineveh (Campbell Thompson and 1 lutchinson 1931: 104; Campbell Thompson and Hamilton 1932: 88; Campbell Thompson and Maliowan 1933: 168). The fact that many BRBs were found upside down at Nineveh in the vicinity of the later Isbtar temple reminded their excavators of much later, similarly upturned Aramaic incantation bowls at Nippur. On analogy with these, Campbell Thompson and bis assistants suggested that BRBs had functioned as votive bowls. BRBs were recorded on the Iranian Plateau for the first time in 1933 during the excavation of Tepe Sialk (Ghirshman 1938: pi. 26.7b; Amiet 1985: 196 and fig. 1.S20) and by 1942 their presence or absence was being cited by D. E, McCown as a significant chronological marker in bis interpretation of Susas stratigraphy (McCown 1942; 43, 44). A year later R. de Mecquenem suggested that BRBs functioned as the markers of infant burials at Susa (de Mecquenem 1943; 13). In 1946 T. Burton-Brown compared BRBs to Predynastic and Old Kingdom Egyptian bread pots (Burton-Brown 1946: 36-37), in which he 1
JCSHI (2(HW)
DANIEL POTTS
Map of sub-regions, identified by Roman numerals, in which BRBs have been recorded (see Table 1 for complete listing of sites by name within each sub-region). was followed a year later by E. Baumgartel (1947: 93). Neither, however, explicitly suggested that BRBs had been used for baking bread (this is, however, implied by Schmidt 1982: 317), and even the formal parallel was dismissed by H, Kantor (Kantor 1954: 6; Hennessy 1967: 39). In 1952, P Delougaz published a thorough investigation into the problem of their function. Rejecting the Nineveh teams proposal that they had been used as votive offerings, Delougaz suggested instead that their "porosity, shape, and size would have been well suited' to "processes of food preparation such ... as the separation of whey from curds" (Delougaz 1952; 128). Discussion of function languished somewhat over the next decade as tbe emphasis shifted again to the utility of BHBs as chronological markers, Following on from McCown s observations about the presence of BRBs at Susa, R. H. Dyson Jr., included bevel-rim bowls among tbose ceramic indicators that denoted the spread of what he termed the "Uruk-Jamdat Nasr-related horizon'" in Iran in the second edition of Chwtiologiefi in Old World Archaeol- BRBs, with variations partit tilar to the lower and upper levels (Le Brun 1971: 177). Similarly, at Tepe Yahya it has been observed that, "Besides the classic Mesopotamian variety, both a smaller and a taller variety have been discovered bere which bave parallels at Susa" (Lamberg-Karlovsky and Tosi 1973; 36; Potts 1977; 28, n. 30). More recently, B Helwing has noted, "a clear typological distinction between early shallow forms tbat tend to be made from a loamy cbaff tempered clay, and of a later high, narrow conical form made from a clay with cbaff-and-giit temper is possible and chronologically significant" (Helwing 2005: 54, n. 17). Witbout a detailed analysis of all of tbe C'"* dates from Umk Mesopotamian, and contemporary sites in Iran, Syria, and Anatolia, it is impossible to defend the view tbat the shorter or "proto-BRB" is chronologically earlv eiiotigh to stipport the view tbat the form originated outside of Mesopotamia. It may just be a variant that falls chronologically within the earlier period of BRB use in the Near East. Whether or not the BRB originated in Mesopotamia, it has been considered a quintessentially Mesopotamian artifact for nearly a century Yet if BRBs were made in hundreds of places outside of Me.sopotamia (if we count Syria and Anatolia as well as Iran), should we continue to consider tbem Mesopotamian? As B, Helwing noted several years ago, "Bevelled rim bowls bave long been considered a marker for the Uruk culttue, until closer examination of assemblages from Northern Syria and Soutbeastem Turkey revealed that BRBs can occur alongside otbei"wise clearly indigenous assemblages... and they equally can occur on tbe Iranian plateau witbin strictly indigenous assemblages" (Helwing 2005: 54, n. 17), It is time to rethink our approach to BRBs and to stop looking at tbem as non-indigenous, intrusive elements in tbe many local ceramic traditions in which they appear. In this regard, the study of ancient religion provides us with an obvious analog}- that may be instru( tive, Many originally Mesopotamian deities, including Adad and bis consort Sbala, Inanna, KI (Earth), Nabu, Nana, Ninbursag, and Sin, were worshipped in sotithwestern Iran during tbe second millennium BC: (Potts 1999, in press), Some of these, such as Adad, were still worshipped during tbe Achaemenid period (Koch 1977: 110-11). W, F, M. Henkelman bas made the point, bowever, that the worship of some of these deities in Iran, including Adad, is attested over sucb a great span of time, that it is incorrect to view such deities as Mesopotamian or Babylonian when talking about tbe Iranian context (Henkelman 2006: 240). After fifteen hundred years of veneration, Adad's presence at Persepolis can hardly be considered evidence of tbe worship of a "foreign" deity. If anything, tbe Persians of the fifth century BC may have tbougbt of Adad as an Elamite deity, so ancient wa.s bis worship in the region, but certainly not as Me.si)potamian. A similar sort of logic may help us to understand the cultural cbaracter of tbe BRB. In this case, it is not tbe use and assimilation of BRBs over millennia that transformed tbem from something Mesopotamian into something local, but rather tbe fact tbat they appear to have been made and used in sucb a variety of non-Mesopotamian locales by non-Mesopotamians—since it seems inconceivable tbat Sumerian or Susian enclaves lurk beneatb tbe surface of every site on which BRBs have been found—tbat tbey must be viewed as part of the local cultural répertoire. This being the case, it seems logical to go one step further and to suggest tbat the tises to wbicb BRBs may have been put on tbe Iranian Plateau or in Pakistani Balucbistan were not necessarily the same as those assumed in tbe Mesopotamian beartland. Even if Nissen, Johnson, Englund, and others are correct in interpreting the BRB as a ration bowl in Mesopotamia, it seems difficult to extend tbis interpretation to Iran and Pakistan, where tbe small numbers of BRBs found on many sites where they appear would seem to argue against their having
BEVEL-RIM BOWLS AND BAKERIES
13
functioned in a ration distribution system, let alone "a regionwide pattern of labour recruitment" (Rothman 2004; 101), On the other hand, tme must ask whether their peculiar fabric and shape wotild have been reproduced over and over again if the BRB tlid not have some strong functional raison d'être? Multi-functionality was first seriously argued by A. Le Brun (Le Brun 1980: 66), and some years later K. Abdi suggested that the dramatic socio-economic developments tïf the Uruk period required a theap, easy-to-make, multi-functional container for a variet\' of daily domestic uses, a situation comparable to the increasing demand for cheap packing material ior exports during the Industrial Revolution of the nineteenth century, and the paper plates and Dixie c ups of the contemporary western societies. (Ahdi 1999; 223) But while many ceramic forms in antiquity were probably multi-functional, it seems difficult to believe that any number of other forms on hand at Tal-i Iblis, Tepe Ozbaki, or Susa couldn't have filled the need fttr multi-functionality. A Culinary Change and a Culture of Emulation in the Fourth Millennium IM;--* If one is looking for a single-function explanation, t)ther than the ration bow!, then, following Schmidt, Millard, Ghazan, antl Lehner, the analogy with Egyptian "pot bread" vessels seems tt) provide the "most convincing explanation for the function of such vessels ... as moulds for cooking leavened bread" (Wengrow 2001; 171), It is nt)table that, although a great deal t)f attention has been paid to early cereal domestication in the ancient Near East, far less attention has been paid to the grinding of harvested grains into flour (Landsberger 1922; Stol 1979) and the making of bread (Währen 1967; Grégoire 1999). As J.-R Grégoire bas noted, "Flat bread constituted the staple diet, but leavened bread had been knt)wn since the Neolithic period, as evidenced by cupola ovens, which coexisted in the Near East as early as from the sixth millennium BGE with cylindrical ovens (taiinr). While the former are suitable for leavened bread, the latter are more appropriate for the breaking of flat bread. The great bakeries used mainly cupola ovens" (1999; 255, citing Währen 1967: 11; Barrelet 1974; Brotnberger 1974; Crawford 1981). According to Pliny (Natural ///.siori/18.71; Chazan and Lehner 1990: 31), both grountl bitter vetch (Vicia ervilia) and chick pea {Cicer arientinum) could be used as leavening agents. Bt)th of these pulses were widely available in the ancient Near East (Stol 1985; Renfrew 1985). The "great bakeries" of which Grégoire wrote were, of course, the huge establishments attested in cuneiform sources. Not very many sites in Iran would have had grinding and baking establishments on a par with those known at Nippur or Umma, but Susa, Tal-e Malyan and Choga Mish—where ca 250,000 BRBs were found in just two seasons of excavation—may well have been baking bread on an industrial scale. Those loaves ma\- well have heen distributed as compensation for labor petformed, just as Rothman, Zagarell, and tbe Mesopotamian ration proponents have suggested. Elsewhere, however, it is equally possible tbat what spread was not a labor system with bread as rations or state/city-state coordinated merchant colonies, but a taste for leavened bread (whatever its Neolithic antecedents)— tentatively identified in contemporary proto-cuneiform text as GUG2;i (Englund 1998: 180, n. 417)— and a new t>'pe of baking technology using easy-to-fashion, locally produced BRBs. In discussing the comparative evidence from Egypt, Chazan and Lehner closely compared the technique of baking in thick-walled, Eg\ptian ¡x-dja bowls (Jacqnet-Gordon 1981 ft)r a detailed presentation of Egyptian bread pots) and in the thinner-walled BRB, suggesting that whereas hedja bowls were in fact portable ovens that were heated and then filled with dough, BRBs were probably filled with dough and then placed in an oven for baking. Importantly, the apparently crude fabric of the BRB, they argued, "can be explained as a response to the uneven and rapid heating to which these vessels were exposed ,.. The more open a ceramic fabric, the more able it is to absorb thermal shock" (Chazan and Lehner 1990: 30). Were
14
DANIEL POTTS
BRBs a kind of "baking tin" in which leavened bread was produced using a similar method to that documented in Egypt? Did non-Mesopotamian palates adopt a Mesopotamian mode of baking in the midfourth millennium that saw the spread of the BRB from the Mediterranean and Anatolia all the way to Pakistan? Was serious élite emulation involved, or just a taste for a new type of bread? The disappearance of tbe BRB, of course, requires an explanation as well Is the bread-baking explanation weakened by the fact that BRBs stopped being made after ca, 3000 BC? If one thinks of Nissens original ration hypothesis, then the answer to this que.stion must surely be "no," After all, the disappearance of the BRB in Mesopotamia did not mark the end of the ration system tbere; therefore, there is no reason to believe tbat the end of the BRB-using phase marked the end of eating leavened bread. Here an observation on bread shape may be relevant. As Grégoire noted, according to Wahren's research, "Loaves dated t(t the third millennium were made from barley, emmer, or wheat flour and were round, concave, or triangular, or even ball- and ring-shaped" (Grégoire 1999: 255), This suggests increasing diversification in bread-baking occurred after tbe fourth millennium, and one could suggest that after an initial phase in which the BRB was used as the main form for baking leavened bread, a greater number of ceramic forms came to be employed to produce loaves of a wide variety of shapes. East of Mesopotamia, and for that matter in those other parts of Western Asia where BRBs had been used, the local evolution of the baker's craft may well have resulted in the modifaction or invention of indigenous ceramic forms that came to replace the BRB as the baking of leavened bread became culturally internalized. Such a scenario would thus account both for the hundreds of thousands of BRBs found at sites like Ghoga Mish, where great bakeries may well have catered to hundreds if not thousands of dependent laborers, and small sites like Wezmeh Gave, where the odd sherd of a BRB may simply reflect the baking of unleavened bread in a fashion initially borrowed from Mesopotamia or Khuzestan. Acknowledgments As noted above, finding out where BRBs have been found is easier said than done, and for their help with some of the newer Itanian discoveries, not all of which are yet published, I would like particularly to thank Abbas Moghaddam (Iranian Genter of Archaeological Research) and Kamyar Abdi (Dartmouth Gollege). I must thank Abbas Moghaddam as well for permission to use the photo of a BRB in a kiln at Tall-e Abu Chizan, published here. Finally, for information and comments on an earlier draft of this paper I am particularly grateful to G G. Lamberg-Kariovsky and Richard Meadow (Harvard University); G A, Pétrie (Cambridge University); L. R Weeks (University of Nottingham); and J. AlvarezMon (University of Sydney).
BEVEL-RIM BOWLS AND BAKERIES
15
Table 1. Occurrences of BRBs at 107 archaeological sites in 19 snh-regions of Iran and Pakistan. Site/Surve\
Reference
I. Tcbriu 1. Tepe Mamorin 2. Wavan 3. Maríil Tepe
near/under new Imam Kbomeini hit Airport flKIA) soutb of Tebran on tbe way to IKIA pai1 of the Tepe Ozbaki complex; sherds from •"a number of beveled-rim bowls" were recorded
Abdi 1999: 84 Abdi 1999: 84 and pers. eomm. Majidzadeb 2000:fig.8.1; 2001/2: 3
4. Tepe Gbabristan
period IV: "about seventy beveled-rim bowl sherds and one complete "'Groben Rlümenfopf"
Majidzadeh 1976: 108,199
5. Tepe Sialk
periods
Gbirshman 1938: pi. 26.7b 1S.2O|, pi. 90 [S.371;D\'son 1965: 223.225; Amiet 1985: figs. 1-2; ilelwing 2005a: 54, n. 17
6. Ad.sman
"so ist auch ji't/t in der Sialk tV-Zcit eine geringe Anzahl von Formen vertreten, die zwar lokal in der Fertigung ist, deren Prototypen jedocb in der mesopolamischen Unikkultur zu sucben sind. Dazu geboren (ilockenlopTe, die nun wesentlich steilwandiger sind als die älteren Fxeinpiare"
Chegini et al, 2004: 215; Helwing 2()05b: 175
7. Cnidin Tepe
period V building ¡icriod VI: "At (iodin outsitie of the Oval Enclosure, in tiiiisp parts of the town occupied in late Period VI, one Hnds trom tbe Late Uruk assemblage only tbe bc^velled rim bowls and eoarse ware trays"
Weiss and Young 1975 Levine and Young 1986: 40
8-10.survey
"Bevelled rim bowls and eoarse ware trays have ... been found on tbe surf'aee of three otber sites ¡other tban Godin Tepe] in the Kangavar valley, but not in any quantity 1 BRB on ¡in uiuiamed .site "sberds of beveled-rim bowis have been picked up at Giyan itself (University Mu.seum collection)" "quantities of beveled-rim bowls at Deshiivar (sic) not many miles from Giyan"" "beveled rim bowls have been found on survey in sufiieient quantities on at least tbree sites (Md. 30,101, and 167)"
Levine and Young 1986: 40
17, Chogha Gavaneb
•"In onr surf'aee pick-up ;it Choghii Gavaneh we found stray pieces of Beveled-Rini Rowl"; in Step Trench I, "il dense deposit of discarded Uruk pottery, including Bevel-Rim Rowls"
Abdi 2001: 5
18. WezmebCave 19. survey
one sherd Abdi 2003: 424 "No Hulailan site ... yielded more than four Reveled- Henrickson 1983: 456. citing Mortensen rim bowls" 1976: 45 "A few bevel rim bowls and a po.ssible drooping spout Goff 1971: fig. 5.19; Wrigbt 1987: 147 [Goff 1971: figures ¡9.23;Mortensen 1975:figures7,8) on one of the campsites and one of the village site.s of the Ihihiihin Plain are the sole possible indications of the Late Uruk Period."
II. Qazvin
1 U.C. Plateau
IV (;. Western Zagros Kangavar)
11.survey 12. Tepe Giyan Kermansbab) 13. Tepe Debsavar IMahidasht) 14-16. survey
Islamabad)
eh Liiran 21. Tepe Farukhabad
20 BRBs
Wright t981 : pi. i ie-f. Tables 27 and C4
22. Tepe Musiyan
complete bowl displayed in the Louvre before WW III
Burton Brown 1946; 3fi
23. Susa 24. Choga Mish
indeterminutf but large number ea. 250,000 in the 3rd and 4th seasons alone
25-78. survey
"Bevel rim bowls apparently occur at all Middle and Late Uruk sites." of vvhith there are at least 54
inter alUil,r\inm 1971, 1978. 1980 Delougaz and Kantor 1996: 49,fig.8, 111. 83.T-V Jobnson 1973: 58 and Table 18
79-80. KS-1508,1617 81. Tall-e Abu Chizan
unstated number
Moghaddam and Miri 2003:fig.12.2-3 Moghaddam 2007
82. Tal-i Ghazir ÍRH-1) 83. Tepe Moravache/ RH-6 84. Tepe Bayamun/ RH-32
unstated number
Caldwell 1957-71:figs.18, 27
unstated number
Wright and Carter 2003: 76
unstated number
Wrighl and C;ai1er 2003: 81
85. Qaleh-ye Toi
2 BRBs
unpubli.shed;.sefn 17.11.2002
86. Tepe Sabz'aii Zabarjad
"a eoncentration of standard beveled ritn bowl Wright 1979:67 sherds ... covering an oval area oriented noiiheast.southwest, perpendicular to the oval summit of the mound" "five complete examples from the pit in Unit B as welt Wright 1979: 71, fig. 25b-c, Tables 7-8 as bv 15 rim-to-base seetions"
87 BZ.86/1 87. Arjan
possible proto-BRB I BRB (figs. 1-2)
Dittmannl984:52;ef.66 unpublisbed; seen 15.11.2002
89. Tul-i Boland Aloiii/ K25
BHB found by Zagarell: "many" on 2002 survey
(Lordegan)
90. Oaleh Gelli (Ll)
4 BRBs (figs. 3-4)
(Shahr-e Kord)
91. Sharak (SIO) 92. S17
"large numbers of beveled rim bowls" "large numbers of beveled rim bowls
Zagarell 1978:136 and unpublisbed; seen 19.11.2002 by Potts, Roustaei. Weeks, and Pétrie unpublished; seen 19,11.2002 by Potts, Roustaei, Weeks and Pétrie Zagarell 1989: 291, fig. 17.6.3 Zagarell 1989:291
93. Tol-e Nurabad 94. Tol-e Spid
41 BRBs recovered in 2003 excavations 23 BRBs recovered in 2003 excavations, more in 2007
95. Tappeh Mohammad Kazemi/MS47
1 BRB
VI Siisia
VII. Mianab/Gargar
VIII. Ram Ilormviz
IX. Oaleh-ye Toi X.Izeh
XI. Behbehan/Zuhreh
XII. Bakhtiyari mountains (Kliana Mirza)
XIII, Mania.sani
fig
Weeks et al. 2006:fig.3.100-102 Pétrie et a!. 2006: fig 4.73; Pétrie et al. 2007; Zeidi, McCall, and Khosrowzadeb 2006:fig.6.15 |MSP 19481 ZeidI, McCall, and Kbosrowzadeh 2006: 6.15 |MSP 1786)
BEVEL-RIM BOWLS AND BAKERIES
Region
Site/Survpv
17
Remarks
XIV Mai-v Daslit 96. Tal-e Malyan 97. Tal-e Kureh 98-100. survey
indeterminate but significant number (TUV; ABC, H5) Nicholas 1990: 56-57; Siminer 2003: 46-47; Miller and Sumner 2003: Table 2 at least 326 BRBs in both Terminal Lapui and Banesii Aiden 2003; 196 and Table DI levels at least 3 nut of 42 sites with Banesh diagnostics had hiuniner 2Ü03: 199 and Table E2 BRBs, a fiiiihcr 18 had dnubtfiit Banesh presence
\VB;mlSir 101. Tal-i Iblis
102-103. survey
jK'rind IV: "The Mesopotamian variety of beveled rim Chase, Caldwell and Fehérvári 1967; 184 Imwl period VI: "In a 5 m test pit (No, 111 200 m SSVV "occur rather sparingly"; Catdwell 1967: of tlie edge of the mound the first 20 cm level 38 andfig.39 lower rontained 61 beveled rim howl fragments, 4 trough .spouts und other sherds reminiscent of Sialk IV" 5 BRBs at Tal-e Khomi, 2 at Tal-e Dashtekar Alire?,a Khosrowzadeb pers, comm,; 2005
104, Tepe Langar
30 Itm southeast of Kennan, unstated number of BRBs Lamberg-Karlovsky 1968: 167
105, Tepe Yahya
indeterminate number
Potts 2001: figs. 2.19-20
106,Mathoutabad ((', 1 km E of Konar Sandal South)
"abundant fragments", "about ]3% of the whole ceramic assemblage, and their fragments come by the hundreds"
Vidale 2007 and pers. comm.
107 Miri Qalat
indeterminate number
Besenval 1994; 521; 1997: Bg, 18
.WI. Kerman XVli, Sdgluin Wlll.Jiroft
MX. Maknm
18
DANIEL POTTS References
Abdi, K, 1999 2001 Alden, J. 2003
The Bevcled-Rim Bowl: Function and Distribution, Pp. 64-85 The hanian World: Essays on Iranian Art and Archamlo^ii Fn'sented to Ezat O. Negfihlxni, ed. A, Alizadeh, Y. Majidzadeh, and S, M, Sliahmirzadi, Tehran: Iran University Press. (Persian. Knglisli summary pp, 222-23), Archaeological Research in the Islamabad Plain; Report on the First and Second Seasons, 1998-1999, Iranian Journal of Archaeology and History 13/2-14/1: 3-5 (English abstract). Appendix D. Excavations at Tal-e Kureh. Pp. 107-98 in Early Urban Life in the Ixind of Anshan: Excavations at Tal-E Malyan in the Highlands of Iran. ed. W. M. Siimner. Philadelphia: University Museum Monoy;raph 117,
Algaze, G, 1989 The Uruk E.xpansion: Cross-Cultmal Exchange in Early Mesopotamian Civilization. Current Arühro]X)logy 30: S7\-mS. 2005 The Uruk World Sy.'iteni 2nd edition. Chicago; University of Chicago Press. Allotte de la Fuye, F M,; Cumont, F.; and de Mecquenem, R. 1928 Numismati(fue, Epigraphie grecque. Céramique élamite. Paris; Mémoires de la Mission Archéologique en Perse 20. Amiet, P 1985 La période IV de Tépé Sialk reconsidérée, Pp. 293-312 in De î'îndus aux Balkan.'^, recueil Jean Deshayes, ed. J.-L. Hiiot, M. Yon, and Y. Calvet. Paris: Éditions Recherche sur les Civilisation.s. Balfet, H, 1980 A propos du métier de l'argile: Exemple de dialogue entre ethnologie et archéologie. Pp. 71-82 in Varchéologie de l'Iraq du début de l'cpŒiue néolithique à 333 avant notre ère, ed. M,-L. Barrelet Paris: CNRS, Barrelet, M.-L 1974 Dispositifs à feu et t uisson des aliments à Ur, Nippur et Uruk. Paléotient 2; 243-300. Baumgartel, E. 1947 The Cultures of Prehistoric Egypt. Vol, 1. Oxford: Oxford Universit\' Pre.ss, Beale, TW, 1978 Beveled Rim Bowls and Their Implications for Change and Economic Organization in the Later Fourth Millennium HC./ÍVES37: 289-313, Besenval, R. 1994 Le peuplement de l'ancienne Gí'drosie, de la protohistoire à la période islamique: Travanx archéologiques récents dans le Makran pakistanais. Comptes rendus de ¡Académie des hi.'icriptions et Udleslettres 1994: 513-35. 1997 The Chronology of Ancient Occupation in Mukran. Pp, 199-215 in South Asian .Archaeology 1995, ed. R, Allchin and B, AlUhin, Delhi: Oxford and IBH, Blackman, M. j . 19S1 The Mineralogical and Chemical Analy.sis of Banesh Period Ceramics from Tal-i Malyan, Iran. Pp, 7-20 in Scientific Studies in Ancient Ceramics, ed. M, J, Hughes. London: British Museum Occasional Studies 19. Bromberger, C, 1974 Fosses à cuisson dans le Proche-Orient actuel: Bilan de quelques observations ethnographiques. Pa/eorícíií 2:301-10, Buccellati, G. 1990 Sait at the Dawn of History: The Case of the Bevelled-Rim Bowls. Pp. 17-40 in Resurrecting the Past A Joint Tribute to Adncin Bounni, ed, P Matthiae, M, van Loon, and H. Weiss, Leiden: Uitgaven van het Nedeiiands Histoiisch-Archaeologisch Instituut te Istanbul. Buchanan, B, 1967 The Prehistoric Stamp Seal: A Reconsideration of Some Old Excavations./AOS 87; 525-40. Burton-Brown, T, 1946 Studies in Third Millennium History. London: Luzac, 1951 Excavations in Azerbaijan, 1948. London: John Murray.
BEVEL-RIM BOWLS AND BAKERIES
19
Biittcrlin. P 2003 Les temps ¡mHu-urlxiius da Mésa¡} ; Mesopotamia in the Late Fifth and Fourth Millennia BC. Journal of Archaeological Research 12: 75-119. Schmidt, K. 1982 Zur Verwendung der mesopotamischen "Glockentöpfe," Arcliäologisches Korresjxfíidenzblatt 12: 317-19. Senior, L, M., and Biniie, D. P, III, 1995 Accurately Estimating Vessel Volume from Profile Illustrations, American Antiquity 60; 319-34. Stein,G. 1999 Rethinking World Systems: Diasporas. Colonies, and Interaction in Uruk Mesopotamia Tucson: University of Arizona Press, Stol, M. 1979 On Trees, Mountains and Millstones in the Ancient Near East Leiden; Mededelingen en Verhandelingen van het Vooraziatisch Cenootschap Ex Oriente Lux 21, 1985 Bean.s, Peas, Lentils and Vetches in Akkadian Texts, BSA 2: 127-39, Sumner, W. M, 2003 Early Urban Life in the Land of Anshan: Excavations at Tal-e Malyan in fhe Highlands of Iran, Philadelphia: University Museum Monograpli 117, Vidale, M, 2007 The Graveyard of Mathoutabad. Unpublished paper delivered at the Mkldle Asian Intercultural Space (MAIS) conference, Ravenna, Italy, 7 July, Wahren, M, 1967 Brot und Gebäck im. Leben und Glauben des Alten Orient Bern; Schweizerisches Archiv für Brot- und Gebäckkunde, Weeks, L. R,; Alizadeh, K. S,; Niakan, L,; Aiamdari, K,; Khosrowzadeh, A,; and Zeidi, M, 2006 Excavations at Tol-e Nurabad, Pp. 31-88 in The Mamasani Archaeological Project Stage One: A Report on the First Two Seasons of the ICAR-University of Sydney Expedition to the Mamosani District. Fars Province. Iran, ed. D, T, Potts and K, Roustaei, Tehran: Archaeological Report Monograph Series 10, Weiss, K, and Young, T C, Jr 1975 The Merchants of Susa: Godin V and Plateau-Lowland Relations in the Late Fourth Millennium BC, Iran 13: 1-17,
BEVEL-RIM BOWLS AND BAKERIES
23
Wengrow, D, 2001 The Evolution of Simplicit> ; Aesthetic Labour and Social Change in the Neolithic Near East, World Archaeology SU: 168-88, Wright,! i. T 1979 .Archaeological Investigations in Northeastern Xuzestan, 1976, Ann Arbor: Museum of Anthropology Technical Reports 10, 1981 An Earl\ Town on the Deh Luran Plain; Excavations at Tepe Farukhabad, Ann Arbor: Museum of Anthrtjpology Memoir 13. 1987 TheSusiana Hinterlands during the Era of Primary State Fonnation. Pp, 141-55 in The Archaeology of Western ¡ran, ed. F 1 foie. Washington, DC: Smithsonian Institution. Wright, H. T. and Carter, E, 2003 Archaeological Surve\ on the Western Ram Hormiiz Plain, 1969, Pp, 61-82 in yeki hud. yeki nabud: Essays on the Archaeology of ¡ran in Honor of William M. Sumner, ed. N. F. Miller and K. Abdi. Los ,'\ngeles: Cotsen Institute of Archaeology Monograph 48. Zagarell, A, 1978 The Role of Highland Pastoralism in the Development of Iranian Citdlization iProto- and Prehistoric ¡ran). Inauguraldissertation, Freie Universität Berlin, 1986 Trade, Women, Class, and Society in Ancient Western Asia. Current Anthro^MÀogy 27: 415-30. 1989 Pastoralism and the Early State in Greater Mesopotamia. Pp, 280-301 in Archaeological Thought in America, ed. C-, C. Lamberg-Karlovsky, Cambridge: Cambridge University Press. Zeidi, M,; McCÀill, B.; and Kliosiouzatleb, A, 2006 Survey of Dasht-e Rostam-e Yek and Dasht-e Rostam-e Do, Pp. 147-68 in The Mamasani Archaeological Project Stage Oiw: A Re¡K)rt on the First Two Seasons of the ¡CAR-Uniœrsity of Sydney ExjK'dition to the Mamasani District. Fars Proviïice, Iran, ed, D, X Potts and K, Roustaei. Tehran: Archaeological Report Monograph Series 10.
THE LOCATION OE MADGA Wolfgang Heim,pel, University of California, Berkeley
1. The Thesis' F. Thureau-Dangin published in SAKl 176 XVIII a seal legend whose first three lines he read as H u - u n - n i - n i pa-te-si Ki-mas'" sakkanak Ma-ad-qa'""Hunnini, governor of Kimash, general of Madqa." The identification of "Ma-ad-qa" with the locality Madga, which is attested in Ur III and earlier periods, led to the conviction that, just like Kimash, it was located in the area east of the Tigris. Marvin Powell, collating the seal legend, found that "Ma-ad-qa'''" actually reads ma-at Elam'''. His new reading was quoted in RIA s.v. Kimas.^ Therewith the trans-Tigridian localization of Madga lost its rationale, and a new search for its localization is required. In his inscriptions, Gudea mentions Madga as source of bitumen and as the location of the ordeal river. The Old Babylonian letters from Mari show that bitumen from Hit was an economic necessity for southern Mesopotamia. Abi-Mekim, an emissary of Zimri-Lim of Mari to Hammu-rabi of Babylon, reports in ARM 26 468:20'-23' on negotiations about the city of Hit, whose possession both kings claimed. After much diplomatic back-and-forth, the matter came to a head when Hammu-rabi declared: "If it were just a word, why would I need Id? The means (of transportation) of your land is donkeys and carts; the means of (transportation of) this land is boats. I do need that city for the bitumen and the asphalt. If it were not for that, why would I need that city?" In the words of S. Lackenbacher, who published this text, "Les raisons qu'il (Hammurabi) donne sont intéressantes car elles ne sont ni politiques ni stratégiques mais d'ordre économique." This economic necessity did not start with Hammu-rabi. It existed as long as boats of the region were sealed with bitumen. The Mad letters also document the river ordeal in Hit, to which people came from as far as Aleppo and Karkamish in the northwest and Elam in the southeast. The double coincidence of Madga and Hit as principal source of bitumen and as location of a river ordeal indicates that they were two names of the same place. It is the aim of this article to substantiate this thesis.
1. Abbreviations follow the CAD. Ur III sources are quoted according to the Database of Neo-Sumerian Texts (BDTNS). Erlenmeyer 152 is the text treated in Englund 2003. In dates of Ur III texts AS, S, and SS stand for Amar-Suen, Shulgi, and Shu-Suen. Additional abbreviations are: BPOA: Biblioteca del Próximo Oriente Antiguo, Consejo Superior de Investigaciones Científicas: Madrid; CUSAS: Cornell University Studies in Assyriology and Sumerology. 2. For the seal legend see now RIME 3/2.6.1. Powell's collation may have led Pomponio and d'Agostino to understand the mention of Madga in Nisaba 7 10 as a "small town in the province of Lagash-Girsu."
25
JCS 61 (2009)
26
WOLFGANG HEIMPEL 2. A Closer Look at the Gudea Inscriptions
Cylinder A XVI 7-12 má ha-ù-na má na lu-a ésir a ba-al ésir IGI.ÉSIR im bábbar-ra hur-sag Má-ad-ga-ta níg-gurji má se gana deg-a-giny Gù-dé-a en ''Nin-gír-su-ra im-ma-na-ús
Boats (loaded) with h a-um earth, boats (loaded) with numerous limestones, with water-scooper bitumen, IGI.ÉSIR bitumen, gypsum from the "mountains"^ of Madga, goods as if it were grain from the fields, let Gudea for lord Ningirsu follow one another.
The translations of Falkenstein (1953: 153), Jacobsen (1987: 407), and Edzard (1997: 79) suggest that the bitumen was transported in types of boats called ha^una and nalua. Averbeck (1987: 646) proposed translating "boat-(loads) of ha^una-stones and nalu^a-stones (along with) ... bitumen ...," which agrees with the later view of Falkenstein (1966, 51): "Aus dem 'Gebirge von Madga' besorgte Gudea sich noch 'h a ^ u m Erde, zahlreiche na-Steine, abal Asphalt, IGI.ÉSIR-Asphalt und Gips' und liess all das auf Schiffen nach Girsu transportieren." More detail is found in the parallel passage of Statue B: Statue B VI51-63 Ma-ad-ga'" hur-sag Í7 lú-ru-da-ta ésir ([ ]) gú RÉG 214 im-ta-eji ki-sá é-ninnu-ka mu-ni-dù im ha-um im-ta-e]^]^
hur-sag Bar-ME-ta na4 na lu-a má gal-gal-a im-mi-si-si úr é-ninnu-ka mu-na-ni-gur
From Madga, (from) the "mountains" of the ordeal river, came down ( ) bitumen,... talents. He constructed with it the foundation terrace of House Fifty. Ha^um earth came down. From the mountains of Bar-ME boats with numerous limestone slabs, large boats, he filled with them. He installed them around the base of House Fifty.
The surface of line 53 is damaged. There is not enough space for ésir a ba-al ésir IGI.ÉSIR or even one of these. The sign after gú is unique; Thureau-Dangin listed it according to its form as REG 214, noting "composé de HI et KASKAL?" It can hardly be a number, which would be entered before the measure. Steible (1991: 169) assumes that gú RÉG 214 describes the bitumen, and so does Edzard (1997: 35) suggesting a possible translation "a myriad(?) of talents of bitumen."
3. Steinkeller (2007) convincingly argued the difference between kur and hur-sag as that of a single mountain as opposed to that of a mountain landscape. His standard translation "mountain range" is misleading in the case of the environment of Madga, where we see hills rather than mountains. For southern Mesopotamians anything that was not flat alluvium or a hill created by a ruined settlement, which was called dug, was apparently a mountain. The case of hur-sag Í7 lú-ru-da may be exceptional because the very site of the ordeal is called in later texts hurmnu, indicating a single feature.
THE LOCATION OF MADGA
27
The boatload of ha-ù-na of the cylinder inscription is called an earth (im). The source of "numerous limestones" is given as "the mountains of Bar-ME." There are limestone cliffs and limestone sills forming rapids along the middle course of the Euphrates, where the stone was quarried. For example, ARM 26 292 includes mention of a slab of stone for a stele (na4 na-re-e-em/na-ra-am) of an unusually long 21 m that had been cut in Sa Hiddan, which was opposite Hindanu, not far downstream from Mari. The area of Hit is devoid of possible limestone quarries. The "numerous limestones" would have come from father upstream. Lines 3-12 of the same column of Statue B mention two areas where limestone slabs were cut for use as stele. Many fragments of such stele are preserved, and are discussed by Suter (2000). Umanum and Menua are unrealized. Básala is etymologically connected with Bishri of Jebel Bishri if indeed the spellings Bajj-sal-la and Ba-s/sa-ar are variants."* Statue B VI3-12 U-ma-núm From Umanum, hur-sag Me-nu-a-ta the mountains of Menua, Bajj-sal-la from Básala, hur-sag Mar-TU-ta the mountains of Amorites, na4 na gal large limestone slabs im-ta-ej]^ came down. na-rû-a-sè Into stele mu-dím he made them. kisal é-ninnu-ka In the courtyard of House Fifty mu-na-ni-rú he erected them for him (Ningirsu). Why is there a gap between tbe listing of middle Eupbratian limestone quarries and of Madga, if the latter was the older name of Hit? A closer look at the entire section on building materials for the construction for Gudea's building shows the following sequence: Section on Ruilding Materials, Statue B V 23-VI63 Material
lumber lumber limestone for stele limestone for stele alabaster copper ebony gold gold Halub wood bitumen "numerous limestones"
Source Amanus area of Ebla Umanum Básala Tidanum Abullat of Kimash Meluhha Hahhum Meluhha Gubin Madga
Bar-ME
Place in text V28 V33
VI 3 VI 5 VI 13 VI 21 VI 26 VI 34 VI 39 VI 45 VI 51 VI 59
4. For a difFerent interpretation of this passage and of the cited place names, see Marchesi 2006; 16. I will discuss his proposal elsewhere.
28
WOLFGANG HEIMPEL
The sequence from Mount Amanus to the area of Ebla, and then to Jebel Bishri makes geographical sense, moving from the far west to the Euphrates, and down the river to an area not far upstream from Hit. The alluvium farther south is of no concern, and the view changes east, first to the source of alabaster, which should be Assyria, named after an Amorite tribal group, and then to the source of copper in Abullat of Kimash, which is, according to a new thesis of Potts, in the "Tiyari mountains, north of Amadiyah."^ Next are imports from the Persian Gulf, which would have come up the Tigris to Gu'aba. Last are Madga and Bar-ME. I cannot find a convincing reason for this arrangement, but it does not shake my belief that Madga is the older name of Hit. 3. The Name "Madga " Madga is first attested in the list of geographical names from Abu Salabikh, line 40, where it appears as Má-ga'^' (OíP99 73). This is definitely an early writing of Madga, as indicated by the variation between Ma-ga*"' and Ma-da-ga""' in the list of geographical names from Ebla (MEE 230 37). In documents, the place name is first attested in Old Akkadian administrative records from Girsu, as in ETC 235, which records the expenditure of "5 kors of barley, Madga boat" (5 se gur A-ga-dè'" má Ma-ad-ga'''), or jRTC253 "3 Ma-ad-ga" in a list of workers. In Ur III sources, the name is spelled Má-ad-ga, Ma-ad-ga, Má-da-ga, and Má-ad-da-ga, and misspelled as Ma-da-ad-ga in SATl 16,MVN12 124, and also, perhaps, in jRoc/ie.ster 216, often without the place determinative. In MVN 12 88 Madga is designated as a "city" (uru). There is a slight chance that it was called a "land" (kur) in Hirose 344 (see 8.2 below). A personal name is Ma-da-ga dumu Mu-da-gál in TEL 241: iv 9-10. So far, no reference is found where the geographical name was spelled Ma-da-ga in Ur III texts. In the sequence má Daa-ga in Or 47-49 249: 80, BPOA 1 458, and SAT 3 1213, Da-a-ga is a personal name. That is also likely the case of má Da-ga in JCS 2 (1948) 191, NBG 3221. All other references of the sequence má d a g a known to me come from contexts that indicate the geographical name Madga. The spelling Má-ad-da-ga indicates a pronunciation Mad(d)aga, the spelling Ma-ad-ga a pronunciation Madga. The name is not attested in texts after the Ur III period; it was supplanted with Id, It, or Ida.** 4. Trips to Madga from Umma Documents from Umma record the assignment of workers for trips to Madga. All these journeys were round trips, which is occasionally explicitly stated. The description in UTI 5 3147 is "gone to Madga, returned from Madga, and boat unloaded" (Má-da-ga-as gen-a Má-da-ga-ta gur-ra ù má ba-al-la). The load is described in MVN 14 310, unfortunately with the general term "things of
5. Daniel Potts in press. The localization would imply that Elam stretched far to the northwest at the time of its governor Hunini, which poses a problem. 6. The Old Babylonian name appears in letters from Mari. It was logographically written with the name of the local god as ''ÍD''', syllabically I-ID''' in ARM 26 503: 8 and I-da*^' in ARM 4 17: 7 as well as in the personal names l-din-''l-da and Síl-lí-I-d|a|. I expect -da to be -id. However, Durand, who collated the text, informed me that he had not seen the wedges that would allow the reading of -id. He also drew my attention to his index on Id (Durand 277: 699) where he states: "Le ville de Hit est tantôt écrite phonétiquement i-da, tantôt parle signe idéogrammatique id, ou parla séquence id-da, série de jeux graphies à partir de l'akkadien idduin, 'bitume,' Idda étant la forme féminine amorrite; la ville ne devait pas être dénommée à partir du terme sumérien 'fleuve.'" Durand does not remark on the form I-ID in AñM26 503: 8. That form can be the Sumerian word for river. It shares the lack of loss of final consonants with Isnunak, the pronounciation of Esnuna in Mad. The name of the inhabitants was spelled i-ta-jii/i. I-ID may have actually rendered I-it.
THE LOCATION OF MADGA
29
the load" (nig gú'-na).'' In AAICAB 1/1 Ashm. 1924-650 the load is identified as hitumen: "gone to Madga, returned from Madga, and in Umma bitumen hoat unloaded" (Má-da-ga-as g e n - n a Máda-ga-ta g u r - r a ù Umma'''-a má ésir ba-al-la). The duration of a trip is given in records of S 46 and 48 as 90 days, corresponding to three months of the administrative calendar. In AS 3, two records give sixty and one record seventy-five days. The three records of AS 7 give seventy, sixty, and fifty-five days. In AS 9, it was seventy-five and sixty days, and in SS 2 seventy, sixty-six, and sixty days. Only one trip was made in a given year. Different numbers of days of one and the same trip reflect the amount of barley and processed foods given to the supervisors who managed the livelihood of their crew. For example, one of the sources for the trip in SS 2 is an entry in Erlenmeyer 152, the balanced account of the scribe Lu-Shara for SS 2, which could not have been written before the end of the year. The entry reads, "three boys for seventy days, their wages** are for 210 days, gone to Madga, sealed tablet of Lugal-itida." BPOA 1 1314 records the duration of the same trip for the same three "boys" as sixty-six days. MVN 14 310 and UTI 5 3147, referring to the same trip, expressly mention the return from Madga and the unloading in Umma, for which only sixty days are given. The length of employment of workers for a trip may have varied according to whether or not they were engaged in unloading the bitumen. It may also reflect unidentified or unrecognized accounting procedures. I suspect that the different numbers of days in this and other groups of texts, which are tabulated below, were fractions of thirty-day months, seventy-five days being two and a half months, seventy days two and a third months, and sixty-six days two and one fifth months. Such time fractions were easily converted into fractions of pay. The single source for the trip SS 6 to 7 gives the duration of the trip as "from SS 6 XII 20 to SS 7 II 25," that is eighty-six days considering that SS 6 had thirteen months in Umma. The eighty-six days do not represent two months and a fraction of a month and should represent the actual duration of the trip and the unloading. The text records the assignment of thirty-five workers of ten agricultural supervisors. Trips from Umma. Agricultural supervisors (nu-bànda gU4) are asterisked (*), other work supervisors marked with a circumflex C^). Text
Date
Days
S 46
90
Syracuse 442
S 46
90
Nisaba 9 284
S 46
90
M VN 2129
S 46
90
BPOA 2 2655
S 46
90
Workers
Notations, supervisors, responsibles, sealers 2
unsealed gîri L u g a l - n i s a g e 2 unsealed giri L u g a l - n i s a g e 2
unsealed
giri L u g a l - n i s a g e 2 unsealed ugula Lugal-gigire* gîri L u g a l - n i s a g e 1 unsealed giri L u g a l - n i s a g e
7. The sign g u' is copied and read as "k u n" by the editors. The term may designate various "things" coming from Madga, not just varieties of bitumen, but also, for example, gypsum. "Things of a load" is also the technical term for a commodity tax (see CUSAS 5 7.5). 8. For the translation "wage" of á see CUSAS 5 2.18.1.1.
30
WOLFGANG HEIMPEL
Text
Date
Days
Workers
Notations, supervisors, responsibles, sealers
BPOA 2 2288
S 48
90
1
ugula g4 sealed by scribe Lugal-Emahe
SAT 2 686
ASI
60
1
"60 gurus U4 1-sè" ugula Lugal-gU4-e* sealed by scribe Abbagina
ßliV 5 272:156-59
AS 3
60
1 Má-da-ga-as gen-na year account of Lugal-gude* sealed tablet (kisib) of Abbagina
SAT 2 747
AS 3
60
1
Má-da-ga gen-a ugula Da-DU-mu* sealed by scribe Abbagina
AUCT3 299
AS 3
75
2
iti215 sealed by scribe Abbagina
Princeton 340
AS 7
70
1
ugula Basa sealed by Lugal-itida
BPOA 2 2227
AS 7
60
2 sealed by tbe scribe Inim-Sbara and tablet to be replaced by tablet sealed by (his father) Lugal-itida
Princeton 366
AS 7
55
1
ugula Lugalmu-ma ag'^ giri Lugal-itida sealed by (his son) Inim-Sbara
Î7TI3 1612
AS 9
75
3
Má-da-ga gen-a ugula Lugal-itida* sealed by Lugal-itida
UTI 5 3047
AS 9
75
1 sà-gU4
Princeton 335
AS 9
75
3
ugula Lugal-nisag-e sealed by Lugal-itida
Princeton 338
AS 9
75
1
ugula Da-DU-mu* sealed by Lugal-itida
Princeton 337
AS 9
60
BPOA 2 2107
SS2
70
2 ugula Ur-mes* sealed by Lugal-itida 1 ugula Sharakam'^ sealed by Lugal-itida
BPOA 1 1314
SS2
66
3
CDLJ 2003:1 I V 32-35
SS2
70
3 whole-year account sealed tablet of Lugal-itida
MVN 14 310
SS2
60
2 "gone to M, goods unloaded in Umma" ugula Ur-sigs'^ sealed by Lugal-itida
ugula Ur-é-nun-na* sealed by Lugal-itida
Má-da-ga gen-a ugula 17-pa-è' sealed by Lugal-itida
THE LOCATION OF MADGA
UTI5 3147
SS 2
60
AAICAB 1/1 Ashm. 1924- SS 6-7 650
86
MVN 18 439
68
31
5 "gone, returned, boat unloaded" ugula Abbagina sealed by Lugal-itida 35 "gone, returned, unloaded bitumen in Umma" multiple overseers, all agricultural supervisors sealed by scribe Abbagina 1 sealed by Lugal-itida
The First Leg of a Trip from Umma to Madga SAT 2 858 and 880, dated AS 5, record the assignment of workers for twenty-seven days of transport work. They treat the same undertaking in slightly different versions, forming a pair of a sealed and an unsealed tablet with the same subject matter.® SAT 2 880 was written and sealed by the scribe Abbagina, son of Lugal-magure, and would have been forwarded to the appropriate provincial administrator. SAT 2 858 is the unsealed copy that Abbagina wrote and gave to the overseer Lugal-magure (no relation). The discrepancy of the number of workers must be an error, and would have become a problem for Abbagina when the two tablets were viewed together at the occasion of composing the annual account in which the record of the work assignment was included. The Pair of Texts SAT 2 858 and 880 Line 1 2 3 4 5 6 7 8 9 10 11 12 13 14
858 3 gurus U4 3-sè nig gú-na kas-dé-a má-a gar U4 5-sè má Má-da-ga gid-da [U4] 6-sè má lá gis kése tir ''Nin-ildum-ma-ta Umma'"-sé má diri-ga ù má ba-al-la U4 5-sè im-dug-a é-da-na aka U4 3-sè kas-dé-a gub-ba U4 5-sè en-gaba-DU-ta Káb-su'*'-s[e] gi gid-]da má ba-al]-la ugula Lugal-[má-gur8-re]
880 2 gurus U4 3-sè nig gú-na kas-dé-a má-a gar-ra U4 5-sè má Má-da-ga gíd-da U4 6-sè má lá-a gis kése-r[á] tir ''Nin-ildum-ma-t[a] Umma'"-sé má diri-g[a] ù Umma'"-[a] m[á] b[a-al-la] U4 5-sè [im-dug-a] é-de4-na [aka] U4 3-sè [kas-dé-a gub-ba] U4 5-sè gi [gíd-da] en-gaba-D[U-ta Káb-su] ""-sé ù Umma'"-a [má ba-al-la] ugula Lugal-má-[gur8-re]
For three days, the workers put the commodity tax (nig gú-na) destined for a reception (kas-dé-a) in Nippur on a boat.'" For the next five days, they towed "a Madga boat" (má Má-da-ga), and for the 9. The translitération is based on photos taken by E. Salgues, for which I thank her and B. Foster. 10. For the reception, literally "poured beer" (kas dé-a), of the governor of Nippur see Sallaberger 1993,1; 35-36,144-45; 1993, II: Table 42. It happened during month VIII and could last into month IX. Receptions of the king in Nippur are also attested as work assignments, for example in CUSAS 3 221 and 240. They happened during month IV The timing of the reception of the governor fits the present texts (see section 7).
32
WOLFGANG HEIMPEL
next six days they tied wood to be towed behind a boat at the forest of Nin-ilduma (má là-a gis kése-rá tir ''Nin-ildum-ma), and floated it back to Umma. They also made adobe walls in PuzrisbDagan, served at the reception in Nippur, and transported reeds. As is not uncommon, the listing of tasks is not in strictly chronological order. I suggest the following scenario: The "Madga boat" was a boat bound for Madga. It was loaded with goods for Madga, but there was enough room to take the items that were destined for the reception in Nippur. The five days of towing the boat would have taken them to Ka-sahara. Upon having reached the Euphrates, they might have changed boats, unloading tbe goods for the reception onto a boat going downstream to Nippur or tbey took tbe Madga-bound boat to Nippur, wbere it could have taken on board more goods bound for Madga. The workers served for three days at the reception in Nippur and made adobe walls in nearby Milehouse for five days. They returned, walking overland to the forest of Nin-ilduma where they helped floating wood to Umma. The last assignment was a difFerent trip in the vicinity of Umma. 5. Pay for Workers Assigned to Madga Trips in Texts from Girsu Texts from Girsu provide information on the pay of workers going to Madga. If my interpretation is not mistaken, over three hundred workers participated in a single trip. The texts do not inform on the tasks of the workers. The transport alone, that is towing the boats upstream and floating tbem downstream, bardly justify the large number of workers and the duration of the trips. They must have been needed for the task of quarrying bitumen. 5.1 Orient 16 135 Tbis is tbe record of an inspection of dockworkers from all three districts of the province dated S 46 IX 7." Thirty-three men were assigned to a trip to Madga (Ma-ad-ga-sè). Three groups of eleven men were formed, each including a foreman called "big brother" (se s-gal). In the first group were three shipwrights (m á - g í n), including the foreman, one soldier (à g a : ú s), five dock workers (lu m a r sa), one "reed-boat person" (lii má g i), and one mat weaver (ad-kub4). The occupations of the workers of the second group are lost in a break. The third group consisted of two "old ones" (su-gÍ4), "old hands" in light of the fact that one of the two served as foreman, four carpenters (nagar), four mat weavers, and one'^ "reed-boat person."
11. The left edge of the tablet is inscribed. BDTNS and CDLI transliterate lugal-''inana U4-da su ba-a-da-ni-ti [ ] su-su-dam. The last line of the last column is "Year Kimash" (mu K i-mas'"'), the shortest possible version of the year name. Entire year names or the last parts were occasionally written on the left edge. The element U4-da is connectable with the element U4 1 -a in the standard full version of the year name. Further, Gomi's copy allows one to read the last two signs as ba-hul. This is then a version of the full name of S 46, albeit one not registered by Sykes (1973) or Sigrist and Gomi (1991): Ki-mas""' lugal ""Inana U4-da su ba-a-da-ni-ti [ ] ba-hul. The verbal form may be understood to include-ni-in causative function and -da- as abilitative. Inana would have enabled the king. "Through Inana the king was able to take Kimash into the hand [and Hurti?] (I[GI.DIB hu-urs-ti'']') was destroyed." The lack of^ an agentive is bothersome. Still, tbe conclusion that Inana was credited with the military success that gave the year the name seems inevitable. No other date formula of an LJr III king credits a god for military success, nor is a trace of it found in the known inscriptions of Shulgi. The name of regnal year 45 breaks the long list of tersely formulated year names. Perhaps long and varied year-name formulations blossomed during the very last years of Shulgi s rule. 12. Copy has "2."
THE LOCATION OF MADGA
33
5.2 Records of barley and flour expenditures for paying the workers who participated in the expedition of S 46 Tbe texts name the disburser and recipient of barley and flour and state the amount and its use as "Madga food" (sà-gal M á-ad-g a"") for tbe month of the date of the text. Tbe texts were inscribed on tablets with a sealed envelope and on unsealed tablets. Tbe seals were those of tbe recipient. Tbe unsealed tablets may have lost tbeir envelope, or they were copies retained by the disburser or tbe recipient. The tablets witb a sealed envelope went to tbe administration. Tbe barley and flour totals of tbe texts are divisible by sixty, wbicb is tbe standard ration of male workers. It was received by tbeir supervisors. They are relatively well documented in Nisaha 7 10 and associated texts, wbicb are treated below. Some of tbem appear to be foremen. DU.DU.NI, for example, was supervisor of workers of tbe type called "seized children" (dumu dabs-ba) and a "seized child" himself. The barley for bis crew was received on bis behalf by someone else, presumably because be was with bis crew on expedition to Madga. Other supervisors were not part of tbeir crew,'^ so, for example, the "overseer (and) dock scribe" (ugula dub-sar mar-sa) Esb^am.'"* He acted as disburser in his capacity of administrator of the dock and as recipient in his capacity of overseer'^ The context shows that tbe workers' rations were not simply moved by tbeir supervisors from the provincial administration to tbe worker The receipts were dated to months when tbe workers were on assignment to Madga. The administration paid supervisors the rations for tbeir workers on a montbly basis whether the workers were present or not. It was up to a supervisor to guarantee tbe livelihood of his crew during tbe entire trip. We do not know bow tbat was organized and what tbe workers actually received. It seems absurd tbat a worker would take along 180 liters of raw barley, wbicb he would masticate, grind, cook, bake, or barter for readily edible food during tbe trip. More likely, tbe workers took along flour, cooked soup, and ate dates furnisbed by their supervisors at the outset of their assignment and financed by tbeir monthly rations.'® In several records of receipts of barley for workers, tbe formula sa Madga appears {MVN 12 88, 120, 454; TCTI 2 8598, 3593). If understood literally as "in Madga," one migbt entertain tbe idea tbat there existed grain stores in Madga from wbich the pay of workers was drawn. Transport of barley to Madga is not recorded in tbe extant documentation. It is therefore more likely tbat the formula is used, just like Akkadian .sa, in a wide range of meanings. In the quoted texts it probably identifies Madga as tbe destination of tbe work assignment. The workers were "boatmen" (má-lah5), "seized children" (dumu dabs-ba), and "workers" (eren). Tbey received the standard ration of sixty liters of barley, including the "seized children." Tbe literal meaning of tbis term, which is typically found in texts from Girsu, indicates orphaned cbildren wbo were "seized," that is, taken in charge as workers when they were able to carry a full workload. By tbe time they were registered as recipients of rations, they were adults. The thirty-three men assigned to a trip of Madga at the inspection of S 46 IX 7, wbicb is recorded in Orient 16 135 (see 5.1), would be among tbem, specifically among, or in addition to, the ninety-nine "assistants" of MVN 12 93. 13. Overseers did not count as one of their crew. Both foremen and overseers were called "overseer" (ugula), so only the context reveals, or does not reveal, the difference. 14. Or was he overseer of the scribes of the dock? 15. Es-àm, presumably pronounced Esh^am and meaning "He-is-(child number) three," occasionally misunderstood and transliterated "3-àm," was not exclusively concerned with workers. MTBM 319, from S 47, and MVN 22 201, from SS 6, show that his responsibilities included dealing with bitumen. 16. For flour, oil, and dates put on Madga-bound boats see 5.4.3 to 5.4.5.
34
WOLFGANG HFIMPFL
Receipts of barley and flour for workers on mission to Madga by tbeir supervisors in S 46. Numbers enclosed in < > in the column "for" result from dividing tbe amount by rations of sixty liters; d. U.M = dumu Ur-Mama, d. L.U = dumu Lugal-usumgal. Text
Month Amount
for
from
Received by
MVN 12 34 ñA 62,15 24 MVÍV12 88
IX X
dumu dab5-ba éren dumu dab5-b a dumu dab5-ba 41 má-lah5 99 ses-tab-ba dumu dab5-lDa éren éren má-lah5 éren éren
Esh^am Nabasa Nabasa Esh^am Nabasa
Ur-kisala Esh^am Ur-kisala Ur-kisala for KA.KA Ur-mes sabra
MVA^12 103
X X
MViV 12 93
X
MVIV 5 148
MVIV 12 127
X X XI XI XI
SAT 1 16
XI
MVIV 12 78 MVIV 12 124 MVIV 12 120
I.O.O 7.3.0 O.I.O I.O.O 8.1.0 19.4.0 6.0.0 5.2.0 8.0.0 6.2.0 5.4.0 71.0
Rochester 216 XI MVIV2 17 XII
dabin se se zi se se se se se se se se
2.3.0 se 2.4.0 se
/ 1 Q xl"^
éren'**
basa Nabasa Nabasa
Ur-kisala for DU.DU.NI Lu-digira Ur-mes d. U.M. for Izu Ur-sag.TAR Ur-mes d. L.U. Nabasa Ur-kisala Nabasa ù "DUB"-Èr-ra for Ur-sag.TAR DU.DU.NI Ur-sag.TAR Lu-digira Nabasa Ur-kisala
5.3 Records from S 47 Nisaba 7 10, a "balanced account of Madga barley" (nig-kasy-ak se Má-ad-da-ga) is a summary text based on all receipts from the next year. It was compiled by Ur-Igalima, son of Atu, wbo was a "scribe of the dock" (dub-sar mar-sa).'^ The total of expenditures was 237.4.5 = 71,390 liters of barley According to A10N31,173 BM 14325, 49.3.2 3 = 14,903 liters were "deficit" (lá-i), which means that they had not been spent. Spent were 56,487 liters. Assuming an average ration of sixty liters of barley and a trip of ninety days, the amount sufficed for the cost of almost 314 workers. The text is divided into three sections. They may represent the three districts of the province. Nisaba 7 10 Amount of barley received sections supervisor
I
II
III
identifications of supervisors
Ur-sag.TAR Ur-è%igir
7.0.0 1.1.0
7.0.0 -
5.1.0 má-1 abg (CT5 17750 II 16-17; MVÍV8 179 XI 31) dumu L u g a l - ù s a r (CT5 17750II23-24)
17. The editor Sigrist (Sigrist and Gomi 1991) transliterates sà-gal ma-da ma-da-[ad]-ga-sè. 18. Envelope: sà-gal éren Ma-ga'''-sè. Tablet; sà-gal Ma-ad-da-ga-sè. 19. His title is known from MVN 8 179 IV 12-13 and other texts.
ga-túg. I assume the text reads sà-gal
THE LOCATION OF MADGA
Ur-ki-sal4-la
DU-(ú-)DU-NI Ur-mes Ur-mes Ur-ki-sal4-la Lú-dingir-ra Ka5-a-mu Es-àm Es-a-bi Gu-za-na Na-ba-sag
15.0.0
71.0 34.2.0 8.0.0 6.0.0 13.2.0 5.2.0 8.0.1 5
6.3.2 11.0.3 5 — —
I-SU4
erased A-kal-la Lú-si Totals
-
— — •
—
—
—
—
—
—
—
9.1.0
70.0 5.0.3 4.0.0
—
—
132.3.1
5.0.0 71.4 11.4.0 4.1.0 4.4.0 — —
35
d u m u d a b 5 - b a ( C r 5 17750 V 5-6) u g u l a d u m u dabs-ba (MVN 12 88) d u m u L u g a l - p a - è ( T C T i 2 3904) d u m u dab5-ba(CT5 17750118) dumu Lugal-usumgal dumu Ur-Ma-ma d u m u dab5-ba(CT5 17750II4) u g u l a d u m u Ur-^'^gigir (Rochester 216) d u m u H é - t i (CT5 17750 II 15) u g u l a d u b - s a r lu m a r - s a (rLB3, 64) dub-sar mar-sa {HSS 4 3)
13.2.2 5 15.2.2 n a g a r ( C T 5 17750) 2.0.0 d u b - s a r m a r - s a (BAOM 2, 25 21) 3.3.0 m á -1 a hs (CT 5 17750 VI 4-5) —
6.4.0 m á -1 a hg {MVN 8 179) 2.2.2
23.0.3 68.1.2 5
5.4 Other Years The documentation for earlier and later years is spotty. 5.4.1 Nisaba 13 48 is an undated balanced account of "Madga barley" for payment of 150 liters for eacb of sixteen "boys" (gurus). That would have been fifty liters per month for three months, or sixty liters for two and a half months. The supervisors of the workers mentioned in Nisaba 13 48 were Ur-Igalima, Esh^am, and Ur-sag.TAR, all known from the records of trips to Madga from Girsu in S 46 and S 47 5.4.2 ASJ 3, 54 3, from S 36, is a seed and fodder text of the household of Ig-Alima. The first item in incidental expenses is recorded in IV 11-13: 11 g u r u s 0.3.0-ta, [se-b] i 6.3 gur, x x Má-ad-ga'''.^° The amount of barley was good for a trip of three montbs of the eleven workers for each of whom a ration of sixty liters was budgeted. 5.4.3 CT 10, 44 BM 18962, from S 43, is a balanced account of grain and flour. One item of expenditures is 10,675 liters of ... flour for Madga (35.3.0 la 5 s 11 a z 1 KA Má-da-ga), whicb was received by tbe "flour scribe Ur-Nungal" (dub Ur-'^Nun-gal d u b - s a r zi). According to MVN 12 206 of S 47 II, ten liters of flour was tbe flour ration in addition to the sixty-liter barley ration for "assistants, boatmen" (ses-tab-ba ma-lahs), who were the kind of workers assigned to trips to Madga. Assuming a trip of tbree months, 10,675 liters of flour would have been suflicient for 356 workers, disregarding tbe subtracted five liters. It compares well with the 314 workers of the trip of S 47. The possibility that the flour was used for the purchase of bitumen iu Madga is discussed iu section
20. The trace.s in Maekawa's copy do not agree with sà-gal or se-ba. 21. The term zî KA is treated by M. Such-Gutierrez in Sefarad 63 (2003) 393-410. The reading of KA, and therewith the literal meaning, remains unknown. References cited on pp. 403-4 show that it could be used as rations of workers.
36
WOLFGANG HEIMPEL
5.4.4 HSS 4 3 is the final balanced account of sesame oil and lard from AS I.^^ An amount of seventy liters of sesame oil and thirty-seven liters of lard is identified as "Madga expenditure" (zi-ga Ma-adga'^')- If it was used as pay for workers on assignment to Madga, if they received one-sixth liter of oil per month as did male household slaves in Garshana,^'' and if sesame oil and lard were treated as equivalent in value, the amount would have been good for 214 workers on a three-month trip, or 321 workers on a two-month trip. The last number agrees with the two-month trip scheduled for AS 1 in Umma and the 314 and 356 workers calculated for trips in S 43 and S 47. 5.4.5 MVN 5 157, from AS 4 X, records that Lu-''lgi-ma-sè received oil for boatmen going to Madga. He was probably the boatman of TCTI2 3981 of AS 6 XL According to ITT 2 3237 of the same date, he operated—or owned—a boat identified with his name. He also received food "for Madga" in SS 2 IX (see 8.20 below). E. Sollberger, who edited the text, includes MVN 5 157 among the "texts in very rough copies (some not in Pinches's hand)." The first line reads "twenty' boys, each one half liter of oil." According to the oil ration of slave workers of one-sixth liter per month mentioned in 5.4.4, one half liter was good for three months. The total of oil would be 30 x 1/2 = 15 liters or 20 x 1/2 = 10 liters, which corresponds to the copy of the second line: l-bi 0.0.1 lugal. 5.4.6 TCTI 2 3449, from AS 6 X, is a case tablet. The envelope was sealed by Nabasa. The text records the receipt of 600 liters of barley for workers on assignment to Madga. In the text on the tablet, the workers are called "smith workers" (éren simug'), on the tablet "boatmen" (má-[lah5]). They appear to have been smiths by profession who served as boatmen on assignment, belonging to the group of workers called "assistants, boatmen" in CT 10 44 (BM 18962 [5.4.3[). If they went for two and a half months, and if they earned a ration of sixty liters per month, the barley would have sufficed for four of them. 5.4.7 QqTabCun 314, from AS 9 X, records the receipt of 3,600 liters of barley for twenty "boys" earning 180 liters each as "food of Madga workers" (sà-gal é r e n Má-ad-da-ga). The barley came from the bishop (saga) of the household of the god Gisbare. The 180 liters per person indicate a trip of three months of workers receiving a sixty-liter ration. 5.4.8 MVN 12 454, from SS 2 XI, records the receipt of 7,810 or 8,110 liters of barley flour as pay for workers hired for loading'' bitumen onto a boat operated—or owned—by a certain Lu-Nanshe.^'' The colophon consists of the notation "in/of Madga" (sa Má-ad-da-ga'), which appears to designate the place where, the work was done.^^ The text combines unusual or unique elements compared with the extant documentation: the naming of the operator—or owner—of the boat, the hiring of workers, and the loading of bitumen in Madga. The pay came from Ur-Nanshe and was received by Ur-Dumuzi. I am unable to identify the three persons, nor am I able to find information on the relationship between Lu-Nanshe and the provincial government, or information on the presence of locals that could be hired in Madga. The amount of barley is divisible by five liters, which was the standard low wage of hired male workers, but the resulting numbers are only divisible by two, so the loading would have been done by 781 or 811 workers for two days. 5.4.9 Rochester 202, from SS 2, records the receipt by a certain Shu-Eshtar of 950 liters of barley as food for Madga workers (sà-gal éren Má-da-ga) that was disbursed by a certain Lukala. The amount of barley could represent nineteen rations of fifty liters for a single month. The workers were "seized children" (dumu dabs-ba). The text is cataloged by the editor Sigrist and entered into the databases BDTNS and CDLI as coming from Umma. Lukala is attested as the disbursing official in 22. The date can be restored according to SAT 1 279 as demonstrated by Maekawa, AS] 20 (1998) 93. 23. CUSAS 5 2.18.3.6. 24. Gomi's copy of tbe number of kors must be read 26 or 27. Tbe work description is not fully preserved. Gomi s copy does not agree witb unloading, wbicb would bemá ba-al-la, and suggests loading, wbicb would be má-a gar-ra. 25. Tbe copy sbows a remaining single wedge of tbe sign after -d a and a partially obscured indented sign, wbicb could be g a or ta. Tbe suggested reading is wbat can be expected.
THE LOCATION OF MADGA
37
many texts from Umma at the time, but also in a few texts from Girsu.^® The term "seized children" is very common in texts from Girsu, but is sparsely attested in texts from Umma.^^ 5.4.10 MVN 12 457, from SS 3, records the receipt of twelve kors of barley as food for eighteen "boys" going to Madga and earning 200 liters each. If they received 60 liters per month, their trip would have been scheduled for a hundred days. If the trip was scheduled for three months, their monthly ration was 62 2/3 liters. 5.4.11 The small unsealed undated tablet BCT 2 179 reads "420 liters of barley ... -kilu, 420 liters Lugal-dalla, 300 barley Nina-za^a, five liters of lard. They received it for the persons of a boat bound for Madga" (1.2.0 se gur si-sá, KU/ma/ur-ki-lú, 1.2.0 Lugal-dalla, 1.0.0 se Ning-na-za-a, 5 sila í-sáh, double ruling, lu má Ma-ad-ga-sè, su ba-ti-és). KU/ma/ur-ki-lú and Ning-naza-a are attested only here. An "overseer" (ugula) L u g a l - d a l l a is attested in texts from Girsu {CT 10 16 [BM 12921 II 17], AS 4) and as foreman of dockworkers in DAS 68 in SS 2. He could well be the Lugal-dalla of BCT 2 179. The amounts are divisible by sixty, but the result of the division are the prime numbers seven and five, so only one month pay for nineteen ration recipients can be meant. The normal amount of oil rations was one-sixth liter per worker per month, so the five liters were a comparably generous rate. 6. Pay for Workers Assigned to Madga Trips in Texts from Umma The documentation is sparse. At the least it informs us about dates of trips to Madga from Umma in AS 5 and 6 that are not documented in the texts treated in section four. 6.1 SAT 2 851, from AS 5, records that "a receipt of Arad over 2,190 liters of barley, food of persons of Madga had been transferred from Abbagina by Lugina (7.1.3 se gur lugal, sa-gal lu Má-da-ga, kisib Arad, ki Ab-ba-gi-na-ta, Lu-gi-na ba-an-dib').^^ The formulation allows a literal interpretation of the "persons of Madga" as inhabitants of Madga who had come to Umma. More likely is a loose formulation for Ummaites going to Madga. The amount of barley divided by standard rations and attested lengths of trips does not yield integers. 6.2 UTI 5 3496, from AS 6, lists expenditures of barley that were to be placed on the account of the supervisor of granaries (a-gù ka-gury gá-gá-dam). One entry is "1,920 liters of Madaga barley, Abbagina son of A-ri-bi." The amount of barley would pay sixty-liter rations for sixteen workers for a trip of two months. 6.3 AAICAB 1/1 Ashm. 1911-170, from SS 2 X-XI, is a receipt sealed by Agu, son of Lugal-Emahe, of 180 liters of barley designated as barley rations of Madga (se-ba Má-da-ga) for two named workers. Ninety liters is an unusually high ration. 26. TCTl 2 2593 and TEL 29. 27. Certainly from Umma is Heisserer 10 and perhaps Hirose 358. Interestingly, AUCTS 492 from Umma mentions a "lieutenant of the seized children of Girsu." 28. The reader suggested correcting the transliteration d a hs to d i b. E. Salgues collated this text and the same phrase in SAT 2 851, 1120, and SAT 3 1493. In all cases, the sign is dih, in SAT 2 851 in the form that was in use in Umma, which was recognized by Steinkeller (1989: 285). Surely, dabs in the colophon of Nisaba 11, 14 has also to be corrected.
38
WOLFGANG HEIMPEL
6.4 Syracuse 446, undated, records amounts of barley flour received by supervisors called "overseers" (ugula) as "food for Madga" (sà-gal Má-da-ga), The personal name U-dag-ga is found in texts from Umma. It surely means "(the one from) Udaga" a location in the province of Umma. In BPOA 1 530 and AnOr 1 63 of S 46, the overseers Lugalmu-ma^ag and Aradmu are found among other overseers known from Umma texts. All amounts are divisible by sixty. If that was tbe flour ration for all workers under tbe supervision of tbe overseers, tbe amount of 240 liters would have lasted for a two-month trip for two workers, and the emended total would have been tbe pay for tbirty-six workers altogether If it was a supplementary flour ration of ten liters, as attested in MVN 12 206 quoted under 4.3, the emended total would have been pay for 216 workers. Tbe numbers as copied do not add up. Liters of barley flour Supervisor 1,320 Lugal-é-mab-e 240 Lugal-mu-ma-ág 240 Lugal-má-gur3-re 240 A-kal-la ses É-a-lú-bi 240 Lti-'^Sára 240 Ù-dag-ga "1.4 1" = 240'' Arad-mu 240 Ur- second rate sandals pairs of ordinary sandals ... leather sacks swelled sheep bides dyed hide prime'white hide (treated with) lard
8.13 UTI 4 2786 (Umma), a sealed tablet of the same date, records the receipt of leather wraps from Akala. They were destined for vessels going to various destinations and to Madga. The formulation of the text is difficult to understand. I give a tentative translation of lines six to eleven. 5 ka nig si la 2-ta ba-gar five (dyed sheep hides) were placed on the opening of (vessels) of two liters each Ur-^'Ba-ú-se forUr-Baba ka í-gára zú-lum (and) were wrapped over the openings ba-an-kése roots in Uruk. and more speciHctilly its connection to Kiillab. To tbe elements Falkenstein adduced i'or tbis argument one might add that when ANam restored the wall of Uruk, he described it as "the ancient work of divine Gilgamesb" (RIME 4.4,6,4.11. 5-8; see Falkenstein 1963; 18). Botb Sin-ka5ids roots in Unik and his acquisition of rule over the Amnänum, Falkenstein concluded, were es.sential for his attainment of tbe kingsbip of Uruk. B\ contrast, Micbalowski has argued that Sîn-kSsids use of the title "king of the Amnänum" signifies legitimization of bis claim to tbe throne on tbe basis of his membership in the royal lineage of Ihe Amnänum tribe (1983; 241). 2H. It bas generally been recognized that such "tariff's" did not correspond to reality, but served to articulate in specific terms the kings claim tbat his reign was characterized by abundance and pr'.L. 1923 Excavations at Ur of the Cha\óees. Anti/jitnries Journal o: 311-33, witb pis. 24-34. 1925 Tlie Excavations at Ur, 1924-1925. Antiquaries Journal 5: 347-402, with pis. 31-48. 1932 Excavations at Ur, 1931-2. A/ií/í/í/«rí(^.s;(j(íníü/12; 355-92. 1939 Ur Excavations, Vol. 5; The Ziggurat and Its Surroundings. London; Oxford University Press. 1954 Excavations at Ur A Record of Twelve Years' Work London; Benn.
MORE OLD BABYLONIAN MUSIC-INSTRUCTION FRAGMENTS FROM NIPPUR Anne Kilmer (University of California., Berkeley) ]eremie Feterson (University of Fennsylvania Museum. Philadelphia)
Twelve years have elapsed since the Ia.st publication of cuneiform texts from Nippur containing music instructions ÍKilmer and Tinney 1996 and 1997), Now, thanks to Jeremie Peterson, two more Old Babylonian fragments have been joined to N3354 + N3355, These fragments, N7745 and N7679, are presented here (in new hand copies by Peterson) together with N3354 -i- N3355. In general, no one has doubted that these texts repre.sent some form of musical instructions, but wbether they are actual tuning instructions to real performing musicians tuning their instruments in preparation for a performance, or whether they are more in tbe nature of educational exercises in the Edubba, one ( annot be certain. In bis article in this issue, Jerome Colburn suggests that tbese texts can display the music itself and can be rendered into modem notation by means of very different interpretations and applications of some key terms, especiall\ genniini, zennum, and siljpu. For our part, we have maintained the same method of understanding these instructions as relating to tuning as it was presented in Kilmer and Tinney (1996 and 1997). Infig.2 of Kilmer and Tinney (1996: 55), Janet Smith rendered the musical notation under the assumption that the scale was ascending. If we were to redo that illustration today, we would assume a descending scale in accordance with newer and widely accepted interpretations stemming from the work of Theo Krispijn (2002: 472) on the understanding of the rubric NU.SU in the Old Babylonian retuning text U.7/8Ü not as a Sumerian expression meaning no further" but rather as Akkadian ÍÍIÍ-.SÍÍ-/Í|ÍÍ1 meaning "tightening" to describe the tuning procedures of the foregoing paragraph (Gurney 1994: 101-2). As in our previous two articles, we offer no exact translation of all tbe Akkadian words in the text, but rather have placed in the right margin the type of musical information that each line seems to provide. Part I. The Nippur Fragments N3354 + N3355 + N7745 + N7679 Transliteration and "translation" Obv. Col. i
1. Also in Kilmer and Tinney 1996. 93
JCS 61 (2009)
ANNE KILMER AND JEREMIE PETERSON
94 2' 3' 4'
|-/ÍÍ/-)Í;?Í
-u]nv
5' 6' interval, tune
7'
9'
'w'^ ge-\en-nu'-um' \s]a-na-nam ze-en-uu-uiu
mode, tune & test ?, tnne
10'
(iá-ab-li-\tum\ ge-en-nu-um ù ze-en-nu-utn
mode, test & tune
IV 12' 13'
^A-ha^-mi'Uiit zcen-nu ii ge-en-nu-\uiii\ I \-nani ze-en-nu-u\m] [ |-an'' ^ze^-en-hut-inii^
4th string, tune & test ?., tune ?, tune
Obv. Col. ii r ze-^en-nu-um «^^ \ge-en-nu-mn] 2' i-sa-ar-tim ze-en-h}u^-\u7n\ 3' ze-en-nu-um ^(V ge-^en^-[nu-um\ 4' ze-en-nu-\um\ 5' 6' T
9' 10'
pr^ \ [x x) \ ri-hi uli-ri-im ze-eln-iiu-mn] ki-if-mu-um z[e-en-nu-um\ [ff\e-h'7-^-du-itm ^,e-\en-iiu-u
Hymn Title 6th string, tune int./mode, tune interval, test
[xl -'x^"3i/ sí-hi-ip pi-t[Í7n]
?, paired mode
[x- x]' 'x^ ze-e7}-7iu-\u7n r ' 1 1
1/
1
1
1
'-^6 " e7't
II
ÍÍ(í"///ít ze
en
?, tune test & tune 5th string, test int./mode, tune
e7i-bu-b{u-u7n |
int./mode, [tune/test|
Rev. col. iii 2' 'rí^-|/jí ulj-ri (or: re-\bu-tum 3' 4'
6' 7'
nu
uin
i-.ki-a\r-tum ze-en-7iu-um\
12'
14'
tune & test int./mode, tune tune 6; test tune
[[ 'xxxxxx^
6th string for; int.), |tune/test] int., |tune/test|
MORE OLD BABYLONIAN MUSIC-INSTRUGTION FRAGMENTS FROM NIPPUR 8'
{old 1') si-hi-ip qá-ah-li-tim \ |
paired mode, |tune?¡
9' 10'
(old 2') ^qá-ab^-li-tumze-[en-nu-um] (old 3') \ze'en-nu-u]m ù ge-\en-nu-um]
int./mode, tune tune & test
11'
(old 4') [ I ^ze-en'^'lnu'ttni]
[—], tune
95
Rev. col, iv r
[ \^x^ z.e'e\n-nii-tim\
{
2' 3'
I 1 '.V.V.V.V' i/;l/-ín-'í./;N".v'l
? int./mode, |tune/testl
4'
\u\h'-ru-um ^ge-cn-uu^-u ui
9th string, test
5'
\sa-n]a-nam i-sa-*ar-tini ze^-en-n[u-um\
?, intVmode, tune
6'
\ki^-it'-ni u-uin gc-en-nn-ti ni\
int./mode, test
T
\xxx ^zelge^-en-nu-u lu
tune/test
8'
[ I-'urn'
?
]
|,tüne
Commentary to N3354 -i- New Fragments and Joins The general sense of and vocabulary in the new fragments clearly belong to our recognizable music term,s (including isqu for isqu "throwstick/lot" and enbûbu for emhûbii "reed pipe"), but several unfamiliar and unprecedented terms also appear: Obv, col. i, 10b: [H\a-na-nam\ relate to sun "to equal/rival"? Rev. col. iii, 4': If urram, "morning, daytime" is a valid reading, then compare sent "morning" as the name of a musical interval. Rev. col. iv, 5': ] sa'-na]-nam; compare obv. i 10. These terms listed above join, tben, the earlier list of obscure terms, as noted in Kilmer and Tinney {1996:53). For the suggested meanings of zeîmwn and gennwn as "tune" and "test," see the discussion in Kilmer and Tinney (1996: 53). For a very different interpretation of these two terms as well as tbe term siljpu in these Nippur fragments and in Nabnitu 32, where we have assumed that sihpu indicated a paired tuning related to the seven "standard" tunings, see the contribution of Colburn in this issue. Kilmer and Peterson leave it to the musicologists to tackle these new interpretations. Pan II. AddeiiHuni to UM 29-15-357 + N3020^ Thanks to the new Old Babylonian Tuning/Retvming text fragment published in l'£T Vl/3 899, a corrected reading can be made for tbe interval/scale name previously read as nis GABA-ri-im (also read as nls mihrim) and translated as "rise of the duplicate" (or as "ri.se of the antiphon"). On L^ET VI/3 899, where the interval nU gßharhn is expected (Krispijn and Mirelman in press), the text writes ni-i& tU'uh-ri-im. Therefore we can now read GABA as iuh and render niH tuhrim rise of tbe tuhriinf (a part 2, See Kilmer and Tinney 1996: 49; 1997,
96
ANNE KILMER AND JEREMTE PETERSON
of the foot, see CAD). Exactly what this means remains obscure and needs discussion. Therefore, in the UM text's line 9' (Kilmer atid Tinney 1996; 52), we now read (with collation by Peterson) ni-is tuulj'\ri-im] instead of |.vxi hii-id^ ai-u6-|.v .\i; or we may restore |fíi'/íí-í/j| hii-is^ íM-i/¿i-|?7-im¡, following Colburn (in this issue).
N7C71
Fig. 1. N3354 + N3355 + N7745 + N7679 obverse and reverse (new joins marked). References Gurney, O. 1994 Babylonian Music Again. Iraq 56; 101-6. Kilmer, A., and Tinney, S. 1996 Old Bah\lonian Music Instruction Texts./CS48: 49-56. 1997 Gortection to Kilmer/Tinney "Old Babylonian Music Instruction Texts./CS48 (1996)."/CS49; 118. Krispijn, T 2002 Musik in Keilschrift; Beiträge zur altorientalische Musikforschung 2. Pp. 466-79 in Studien zur Musikarchäologie III, ed. E. Hickmann, A D. Kilmer, and R. Eichmann. Orient-Archäologie, Band 10. Rahden/ Westf,; Marie T.,eidorf, Krispijn, T, , and Mirelnian,S. In press The Old Babylonian Tuning Text UET VI/3 899. Iraq.
A NEW INTERPRETATION OF THE NIPPUR MUSIC-INSTRUCTION FRAGMENTS Jerome Colhurn (Champaign, IL)
As Anne Kilmer (to whom go my deepest thanks for encouraging me to present this article) has stated, the Old Babylonian music instruction fragments from Nippur "are difficult to piece together in a meaningful musical context"' when interpreted as instructions for tuning up a stringed instrument. The puzzle persists even with the two new pieces she and Jeremie Peterson present in the preceding article. In the discussion that follows, the abbreviation NMI 1 (for Nippur Music Instructions 1) refers to N3354 + N3355 + N7745 + N7679; NMI 2 denotes UM 29-15-357; and NMI 3 denotes N3020. A summary of the relevant cuneiform theoretical music documents, for those who may not be familiar with the foregoing literature, is given in the Appendix to this article. The Appendix also discusses the rationale for the staff notation that is adopted here. The term "instructions," it must be acknowledged, is a misnomer if understood in the sense of instructions that a musician could read and follow or a teachers instructions to pupils. As Piotr Michalowski points out in a forthcoming article, the OB scribe and musician received different educations, and a musician would not be able to read these texts.^ If they are, as they appear to be, written descriptions of musicians' actions related to particular compositions, they were prepared for purposes and under circumstances unknown to us. The term "instructions" in this article is nevertheless retained in continuity with the previous publications. Obstacles to the Interpretation of the Fragments as Tuning Instructions Interpreting these texts as tuning instructions entails an inherent ambiguity. If a term with sehpum, such as sehep pltim, means a scale,'' then its occurrence in a "tune" step {zennuvi)* of a tuning protocol can be understood as an instruction to tune multiple strings to that scale; in a "test" step {gennum)^ it would mean that multiple intervals in the scale are to be tested. Likewise, the name of one of the thirds/sixths, such as serdûm, can refer only to that string set. The name of a fifth/fourth, however, such 1. A. D. Kilmer and S. Tinney, "Old Babylonian Music Instruction Texts," JCS 48 (1996) 54. 2. P. Michalowski, "Learning Music: Schooling, Apprenticeship, and Gender in Early Mesopotamia," in Musicians and the Tradition of Literature in the Ancient Near East, ed. R. Pruzsinszky and D. Shehata (Vienna: LIT, in press). I offer my deepest thanks to Piotr Michalowski for accepting this work, for much necessary criticism, and for sharing with me a draft of his forthcoming paper. 3. R. L. Crocker and A. D. Kilmer, "The Fragmentary Music Text from Nippur," Iraq 46 (1984) 81-85. 4. Kilmer and Tinney, "Old Babylonian Music Instruction Texts," 54. 5. Ibid.
97
JCS 61 (2009)
98
JEROME COLBURN
as isartum, can mean either tbe string set itself or tbe corresponding scale. If tbe term means tbe string set, tben tbe instructions cannot call for any scale tbat is not a sehpum. If tbe term means the scale, however, tbe fiftbs/fourtbs cannot be named in tbe tuning process, in contrast to what is actually described in tbe known "Tuning Text" UETVll 74^ (and now UET YI/S 899).^ Richard Dumbrill proposes tbat in tbese texts isartum and so fortb always mean tbe string sets, wbereas sehep isartim and so fortb mean tbe scales described in tbe Tuning Text.^ Tbe otber texts we bave tbat name scales, bowever—tbe Tuning Text, tbe Assyrian song catalog KAR 158,^^ and tbe music from Ugarit'"—do not use tbe word sehpum. In actual use, tbe terms are inconsistently and problematically appled. One would expect tbe terms for (rough) tuning step and for testing (or fine-tuning) step each to apply to its own specific category of entities. Gennum as "test" should certainly be applied only to string sets or {sehpü of) scales, and tbat is how it is applied in most instances, but it is applied to a single string in NMI 1 i' 12', ii' 12', iv' 4'. Against wbat is a single string to be tested? Zennum is usually applied to a sehpum or a string set, but it is applied to the fourth string in NMI 1 i' 4' and NMI 2 ii' 9'. In reference to wbat is the fourth string to be tuned? Operations are wasted. On NMI 2 iii' 2'-4', one is told, '^serdûm, test. Isartum, tune, tune, and test." Wby is it necessary to (rougb) tune tbe isartum scale twice—in the process destroying the relationship established witb tbe preceding fine-tuning of tbe serdûm string set? On tbe same tablet, ii' 7'-8', tbe instruction to test tbe qablltum occurs before tbe instruction to (rougb) tune it; likewise NMI 1 i' 11'. On NMI 1 ii' 2'-4' one is told, "Isartum, tune, tune and test, tune." This requires tbree complete (rougb) retunings, apparently; again, wby was one not enougb? Altbougb the material is fragmentary, tbese paradoxes arise in wbat is preserved, not as a result of tbe lack of material. Tbese difficulties are so substantial tbat a reexamination of tbe problem and its controlling assumptions seems justified. The Basis for the Interpretation of the Fragments as Tuning Protocols Tbe interpretation of tbe Nippur fragments publisbed in 1996 as tuning instructions rests on Richard Crocker s identification of tbe sehpum terms as names of scales." These terms were discovered by Aaron Sbaffer in a Middle Babylonian version of tbe lexical text Nabnitu XXXII.'^ In tbat text, a list of tbe names of the strings themselves is followed by tbe names of tbe corresponding fiftbs/fourtbs or scales, eacb one followed by its sehpum {isartum, sehep isartim, kitmum, sehep kitmim, and so on). Sehpum is attested otberwise (and later) in tbe meanings "stretcb, extent" (already found in a stock pbrase sehep AN u KI in OB Mari'^); "sweeping attack"; "covering"; "inner bark of tbe kiskanû tree"; "prone position" (CAD S 238-39). It is derived from sahäpum "to cover, overwhelm, spread over"; "put a cover on, cover over"; "turn over (?), upside down (?), lay flat, lay (?)
6. O. R. Gurney, "An Old Babylonian Treatise on the Tuning of the Harp," Iraq 30 (1968) 229-53. 7. S. Mirelman and T J. H. Krispijn, "The Old Babylonian Tuning Text UET VI/3 899," Iraq (forthcoming). 8. R. Dumbrill, The Archaeomusicology of the Ancient Near East (Victoria, BC: Trafford, 2005), 75. 9. A. D. Kilmer, "The Strings of Musical Instruments: Their Names, Numbers, and Significance," in Studies in Honor of Benno Landsherger, AS 16 (Chicago: University of Chicago, 1965), 267-68. 10. H. C. Cüterbock, "Musical Notation in Ugarit," RA 64 (1970) 45-52. 11. Crocker and Kilmer, "The Fragmentary Music Text from Nippur," 84-85. 12. A. Shaffer, "A New Musical Term in Ancient Mesopotamian Music," Iraq 43 (1981) 79-83. 13. D. Charpin, "Les malheurs d'un scribe, ou de l'inutilité du sumérien loin de Nippur," in Nippur at the Centennial: Papers Read at the 3Sè Rencontre Assyriologique Internationale, Philadelphia, 1989, ed. M. dej. Ellis (Philadelphia: Samuel Noah Kramer Fund, 1992), 9, line 12.
THE NIPPUR MUSIC-INSTRUCTION FRACMENTS
99
hricks" {CAD S 30-36). In view of the meaning "turn over (?), upside down (?)," Shaffer suggested that sehpum might mean the "inversion" of the interval in its modern musical sense: the interval produced by transposing one of the pitches of the original interval by an octave to make the upper pitch the lower and vice versa. Sehpum cannot mean "inversion" in that sense, however, because the Tuning Text and the list of string pairs CBS 10996''' treat fifths and fourths, which are inversions of each other, identically; because sehpum terms existed for string pairs that could not be inverted on a nine-stringed instrument, as in sehep kitmim, where kitmum is the string pair (3, 6); and because sehpum terms existed only for the fifths/fourths/ scales, even though some of the thirds and sixths can also be inverted on a nine-stringed instrument. Crocker, in reply to Shaffer, argued that because the string set or scale names in Nabnitu XXXII appear in the same sequence as they do in the Assyrian song catalog and the second series of the Tuning Text, where they denote scales, rather than in the sequence in which they appear in CBS 10996, the names must have referred to scales in Nabnitu XXXII as well, and therefore the sehpû must also have been scales. Crocker further suggested, on the analogy of the plagal modes of Cregorian and Byzantine chant but without reference to the cuneiform evidence, that each sehpum covered a lower pitch range than the corresponding scale named without sehpum. Tuning such a "paired mode" alongside the ordinary scale (as called for by the Nippur fragments interpreted as tuning instructions) would require as many as twelve strings.'^ The document that names the sehpü, however, names only nine strings,'® which is curious if the theory it represents required more. When the Nippur fragments were subsequently brought to light, the presence of names of scales, as the sehpû were believed to be, among the names of strings and string sets following an apparent hymn title compelled the interpretation of these fragments as instructions for tuning the instrument to those scales and adjusting particular intervals to play the named hymn. In view of the aforementioned ditficulties with that interpretation, however, it may be worthwhile at this point to reconsider the claim that the sehpû must have been scales. Sehpû If the terms in the second section of Nabnitu XXXII, including the sehpû, do not refer to scales, they must refer to string sets. For there to be a reason to list the sehpû, they must have been distinct from any of the other known string sets. There must also have been a regular relationship between them and the fifths/fourths to which they are said to belong. Mathias Bielitz has suggested (on other grounds than those presented here) that the sehpü could be string pairs derived from the fifths/fourths by inward displacement of fingers, reducing the widths of the intervals."^ When such a change is applied to a pair of strings spanning a fourth, the result is a pair of adjacent strings, sounding a second, such as (4,5) from kitmum (3,6). Such string pairs fulfill both of the requirements set forth in the preceding paragraph. As seconds they are not identical to the fifths/ fourths or thirds/sixths known from CBS 10996, and they are regularly derivable from fourths. Under this hypothesis, the sehpû would relate to the corresponding fifths/fourths in the same way as titur isartim (3,5), the "bridge of the isartum]' relates to isartum (2,6).
14. Kilmer, "The Strings of Musical Instruments." 15. Kilmer and Tinney, "Old Babylonian Music Instruction Texts," 54. 16. Crocker and Kilmer, "The Fragmentary Music Text from Nippur," 82. 17. M. Bielitz, Über die babylonischen theoretischen Texte zur Musik: Zu den Grenzen der Anwendung des antiken Tonsystems. 2., erweiterte Auflage für HeiDok (Neckargemünd: Männeies, 2002), 55-56.
100
JEROME COLBURN
Treble in front, bass in back —*^*—
—©—
3"
«.•>
*•>
sehep isarti
isartu ©—
—e*i—
kitmu sehep kitmi 1
nid qabli
sehep nid qabli
embQbu
i >ö i
1— n —
s*^ 1— e —
sehep Ê;mbûbi
®"
cv
Tï
—e —e
nis tuhri
pîtu
sehep pTti "TV©
—©*^—
sehep nis tuhri
'
©•
1
1
qablîtu sehep qablTti
Fig, 1, Relationships of the sehpü with the corresponding fifths/fourths.
In playing a third such as titur isartim on a lyre or harp with the fingers of one hand, the fingers touch strings 3 and 5 while the hand arches over string 4 without touching it—a pattern for which "bridge" is an appropriate term. In playing a second, the fingers touch both adjacent strings with no intervening untouched string. The meaning of sehpum as "extent, covering, overlay" fits such a pattern, without need to invoke the dramatic ideas of overwhelming or overturning that the parent verb may suggest. Four of the string pairs listed in CBS 10996 appear as fourths: nid qablim (1, 4), qablltum (2, 5), kitmum (3, 6), and pltum (4, 7). The sehpû of these pairs are easily formed directly: (2, 3), (3, 4), (4, 5), and (5, 6). Isartum, on the other hand, is described as a fifth (2, 6) in CBS 10996, yet sehep isartim is listed in Nabnitu XXXII. How can such a sehpum be formed? As discussed in the Appendix, the Tuning Text shows that a string set such as isartum included not merely the two strings mentioned in CBS 10996 but also the octaves of those strings available on a given instrument, such as string 9 along with string 2 on a nine-stringed instrument. With string 9 included, the fourth (6, 9) also represents isartum, and the sehpum of that fourth is (7, 8). On a nine-stringed instrument all of the seconds except (1,2), and its octaves at (8, 9), lie within fourths of which they are the sehpü, and all of the fifths/fourths except embübum (3, 7) may appear as fourths from which sehpû can be formed. Defining (1,2) and (8, 9) as sehep embübim fits the logic that defines the other sehpü. If there were a tenth string, it would be the octave of the third, and (8, 9) would be the sehpum of (7, *10), although it is not clear that such a statement would be made in OB times. The sehpü as seconds are shown, in the staff notation described in the Appendix to this chapter, as formed from the fourths in traditional order in fig, 1 and by themselves in order of pitch infig.2. Some Objections If the sehpü are string sets, the question arises why they do not appear in the known string set catalogs: CBS 10996 and the left column of the Tuning Text, which lists the thirds, fourths, fifths, and sixths that involve each string.'* If the sehpü are not interpreted as seconds, however, this question does not go away but takes a different form: Why do the catalogs not include terms for the seconds? The answer would have to be that the authors of the catalogs were interested only in consonant intervals,'® not in the dissonant seconds. That answer still makes sense even if names for the seconds turn out to exist.
18. R L. Crocker, "Remarks on the Tuning Text UET VII 74 (U. 7/80)," Or 47 (1978) 99-104. 19. A. Mann (ed.). The Study of Counterpoint from Johann Joseph Fux'sGradus ad Parnassum (New York: Norton, 1965), 20.
THE NIPPUR MUSIC-INSTRUCTION FRAGMENTS
101
Treble in front, bass in back
sehep embubi sehep md qabli sehep qabliti sehep kitmi sehep piti sehep nîs tuhri sehep isarti Fig. 2. The sehpü in order of pitch. One could also ask, if the second section of Nabnitu XXXII is a list of string sets rather than of scales, why it does not list the thirds/sixths. A possible answer is that it may have. There is enough room in the missing portions of the column for a listing of the thirds below the fifths and fourths and their sehpü. A column could be at least thirty-eight lines deep;^° the nine strings, the summary line, the line for pismu, and the seven fifths/fourths and their sehpü take up twenty-five lines, leaving at least thirteen in which to fit entries for the seven thirds/sixths in a separate section. MaZrutu., Pismu, and Isartu in the NA Fragment Crocker's argument that the sequence of names in Nabnitu XXXII requires it to be a sequence of scales remains to be dealt with. The second section of Nabnitu XXXII in the NB version begins with a term pismu that does not appear in the MB version, followed by isartu, the first term in the list of fifths/fourths or scales. Because of its occurrence in this text and the premise that the section is a list of scales, pismu has been understood to be a scale as well.^' The same sequence pismu, isartu found in the later Nabnitu XXXII also occurs on BM 65217 + 66616, a Neo-Assyrian fragment that also contains a broken commentary after each term.-^ In this text, pismu is preceded by the previously unknown term maZrutu, which, by association, has been taken to be yet another scale. If this list is a list of scales, one would expect the commentary on each item to mention tightening and loosening strings, making intervals "clear," and perhaps testing or fine-tuning. These topics are not found, at least in the preserved portions of the commentary. Instead, what remains and can be understood of the commentary focuses on strings and fingers. Reading these passages, without an a priori assumption that they must refer to scales, suggests that they are instructions for placing the hands to play strings and string sets (which would be the same in any scale). If so, they militate against the contention that the sequence pismu, isartu, and so forth, found in Nabnitu XXXII, necessarily designates names of scales. As one example, the most nearly intact commentary, that for maZrutu, "the sound of whose strings to play together, one with the second not brought near,"^^ reads like a term for playing two strings as some type of dichord. If so, it is further evidence that, as Kilmer proposed over thirty years ago to an incredulous musicological community,^"* the musicians of the cuneiform world made use of the sound of simultaneous different pitches. 20. I. Finkel and M. Civil, Materialen zum sumerischen Lexikon XVI: The Series SIG7.ALAN = Nahnltti (Rome: Pontifical Biblical Institute, 1982), 254. 21. A. D. Kilmer, "How the Mesopotamians Did or Did Not Express the Concept of the Octave," in Studien zur Musikarchäologie V, ed. E. Hickmann and R. Eichmann (Rahden-Westf.: Marie Leidorf, 2004), 276. 22. A. D. Kilmer, "A Music Tablet from Sippar (?): BM 65217 + 66616," Iraq 46 (1984) 69-80. 23. Kihner, "A Music Tablet from Sippar (?)," 73, rev. ISa-h. 24. A. D. Kilmer, "The Cult Song with Music from Ancient Ugarit: Another Interpretation," ñA 68 (1974) 69-82.
102
JEROME COLBURN
The description of the next term, pismu, is more broken, but it also mentions strings, fingers, and rûtu, a Neo-Assyrian term defined as "half-cubit, span" {CAD R 438-39). When the fingers are fully abducted, the distance from the tip of the thumb to the tip of the little finger is indeed about half a cubit (distance from elbow to tip of extended fingers).^^ The term could thus refer to the fingers stretched out in the same way across the strings of a lyre or harp. Theo Krispijn connects pismu with pasämu "to veil,"^® and in such a position the fingers do spread like a veil over the strings. Musical Use of the Seconds If the sehpû are string sets, the Nippur fragments become sequences of string set names interspersed with unknown technical terms and other words. Such a pattern looks like a description of performance rather than a tuning procedure—in other words, like what we would call music. Kilmer and Dumbrill have each previously suggested that the fragments are descriptions of performance, but both interpreted the sehpû as scales.^^ Thus NMI 1 ii' 6'-9' gave, for Kilmer, a starting pitch, two string pairs as the "intervals" used (as she interpreted zennum at the time), and the scale, from which the complete performance was to be developed somewhat as in an Indian râga. When the fragments containing gennum were added, Kilmer replaced this interpretation with the tuning interpretation discussed earlier.^^ For Dumbrill, lines 6'-8' give the music to be played, while line 9' names the scale as a kind of colophon. If so, a sequence of only three string sets would represent an entire song, which does not seem likely. The present interpretation, as entirely a description of playing a sequence of string sets, avoids these difficulties. Leaving aside for a moment the meaning of the technical terms zennum and gennum, it becomes possible to discuss at least the dichord sequences. For a full description, it would be necessary to know what scales were used, but that information is not preserved explicitly, though it can be inferred as described in the following section. Nevertheless, some features of the music can be discussed independently of the scale. NMI 1 ii' 6'-9' shows a progression rebi uhrim (string 6), kitmum (strings 3 and 6), serdûm (4,6), sehep pltim (5, 6 by the interpretation of sehpum advanced here). In Western terms this is oblique motion:^^ stepwise motion through strings 3, 4, and 5 against a drone or pedal point on string 6. Similar oblique motion occurs in NMI 2 ii' 2 ' - i r , with the drone on the fourth string this time: nidi qablim (1, 4), titur qablttim (2, 4), and sehep qablltim (3, 4). A unison, abanûm (string 4), leads to a second, sehep kitmim (4, 5), by oblique motion in ii' 9 ' - i r . The neat fit of the seconds obtained by the present interpretation of sehpum into the motion patterns in these passages supports this interpretation. It also shows that the Old Babylonian composers did not use these dissonances haphazardly but prepared them melodically, somewhat as in the Western counterpoint tradition.^" A sehpum is followed by its corresponding fourth {sehep qablttim, qablltum) in NMI 1 iii' 8'-9' and in NMI 2 ii' 5'-7'. This may have been a convention for resolving the dissonance. 25. Author's personal observation. 26. T. J. H. Krispijn, "Musik in Keilschrift: Beiträge zur altorientalistischen Musikforschung 2" in Studien zur Musikarchäologie III, ed. E. Hickmann, A. D. Kilmer, and R. Eichmann (Rahden-Westf.: Marie Leidorf, 2002), 465-79. 27. A. D. Kilmer, "Musical Practice in Nippur," in Nippur at the Centennial: Papers Read at the 35è Rencontre Assyriologique Internationale, Philadelphia, 1989, ed. M. de Jong Ellis (Philadelphia: Samuel Noah Kramer Fund, 1992), 101-12; Dumbrill, Archaeomusicology of the Ancient Near East, 86-88. 28. Kilmer and Tinney, "Old Babylonian Music Instruction Texts," 53. 29. Mann, The Study of Counterpoint, 22. 30. See, e.g., Mann, The Study of Counterpoint, 41.
THE NIPPUR MUSIC-INSTRUCTION FRACMENTS
103
Identification of the Tuning and the Modality Under the assumption that the composers did not use the diminished fifth/augmented fourth in a scale (called "not clear" in the Tuning Text), as is the case at Ugarit,''' one can set bounds on what the scales could be. If a fifth/fourth appears in a passage, the scale in which that fifth/fourth is not clear is ruled out. If the practice of following a sehpum with its corresponding fourth was a general convention, that fourth should also be added to the list of fifths/fourths used. NMI 1 i' contains kitmum (ruling out the emhübum scale) and qablltum (ruling out the isartum scale). Column ii' contains isartum (ruling out the kitmum scale), kitmum, embübum (ruling out the pltum scale), and sehep pltim (if followed by pltum, ruling out the nld qablim scale). The only possible scales for column ii' (whether one includes the whole column or only the column from line 10' on) are therefore isartum, nls tuhrim, and qablltum, but if the two columns belong to the same composition and therefore use the same scale (or if they belong to different compositions that were grouped together by the scale), the only possibilities are nls tuhrim and qablltum, surprisingly far, in terms of steps along the second series in the Tuning Text, from the "normal" isartum. On the reverse, pltum, kitmum, isartum, and qablltum are found, ruling out nld qablim, em.bûbum, kitmum, and isartum and leaving pltum, nls tuhrim, and qablltum. The reverse may not belong to the same composition as the obverse. NMI 2 i' may contain nld qablim and isartum, but the column is too damaged to be certain. In column ii', qablltum and nld qablim are found, ruling out the isartum and nls tuhrim scales. In NMI 2 iii', isartum and kitmum occur as in NMI 1 ii', ruling out the kitmum and embübum. scales, and nls tuhrim (see Kilmer and Peterson, this volume) would rule out qablltum. No fifths/fourths are certain on NMI 3. The dichord sequences from NMI 1 are notated in fig. 3. For NMI 1 i' and ii', taken to be from the same composition, the nts tuhrim scale is used, in which five strings are flattened relative to isartum. The other passages are taken to be in pltum (three flats relative to isartum) because it is the closest scale to isartum (in tuning steps) that permits the fifths/fourths found in those passages. The nls tuhrim scale would also be possible for NMI 1 iii' and iv'. The beginning of ii' 10', which follows sehep pltim, is restored as [pi-tu-u]m according to the proposed dissonance resolution convention, and the notes are consequently bracketed. The restoration in iii' 2' is taken to be re-[bu-tum] rather than re-[bi uh-ri-im]. String set sequences from NMI 2 are notated in fig. 4. NMI 2 iii' has some musical similarities to NMI 1 i'-ii' and may belong to the same work, so it is notated in nls tuhrim as well The following restorations are suggested to col. iii'. In line 6', "^x^ (ending in a vertical)-dti-Mm should be a string set name by its position, but the only known string set name ending in -du-um is serdûm, so [s]e-^ery-du-um is emended here. In line 9' there is enough space before ni-is tu-uh-r[i-im] to restore either some hint syllables (see p. 105) or [sé-he-ep], and the latter (which may have resolved to nls tuhrim) is restored here. In line 8' [re]-^bi uh^-ri-im is at least not ruled out by the traces on the copy. In line 7' there is room for a word of about four signs, perhaps [hamsum], [gennum], or [zennum], of which the former was chosen. All these restorations are bracketed in the figure. In NMI 1 ii' and NMI 2 iii', the third and the fifth above the drone are heard, which to our ears creates a sense of modality and a tonal center on the 6th string; with the instrument tuned to the nls tuhrim scale the modality is minor, with a sharp 6th degree (although that degree, on string 1, is not actually found here).
31. A. D. Kilmer, "Musik. A," § 5.3, Group B, RIA 8 (1997).
104
JEROME COLBURN
obv. i 1 ^
Z
Z+G s,Z
obv. ii' 2'-14' Z,Z+G,Z Z
G+Z Z+G s?,Z
Z
Z+G,Z G
G
*.•>
}' ^
>—§--^
J
-J,—'•—n
_lí—i.>" [ o ]
e Lipit-Estar
rev. iii'2'-3' Z?
[
8'-10' ?
x.zu rev. iv' 3'-6' ? G
Z,Z+G Z
-O-
IS5I -©-
8 ]
Fig. 3. String set sequences from NMI 1 (N 3354+N 3355+N 7745+N 7679). Z, zennum,; G, gennum; s, sananam. obv. i i ' 2 ' - l l ' Z,Z :^
Ö
Z Tï
G+Z Z
rev. iii' 2'-6' G Z,Z+G Z
G Z
;
-©-
"XT"
-©-
kun?.na
in.sa
Fig. 4. String set sequences from NMI 2 (UM 29-15-357).
Zennum and Gennum The meanings "tune" and "test" attributed to the terms zennum and gennum are a consequence of the premise that the Nippur fragments are tuning instructions.^^ If, as argued here, the Nippur fragments describe musical performance, these terms must instead represent different ways of playing a string set, for which there are boundless possibilities. The information these terms represented must have been very important to the performance, however, because they are so frequent in the texts. In support of the interpretation of zennum and gennum as "tune" and "test," Kilmer connected these terms to Sumerian zi and ge-(en), found in OB Proto-Lú 622-627^^ and in Shulgi Hymn B.^"* In the latter text, at line 160, zi "to raise" is paired with su "to lower," but it is not certain what is being raised and lowered. At line 171 of Shulgi B, ge-na occurs as the alternative to gid-i "tighten" (sharpen a string) and tu-lu "loosen" (flatten a string). Although this use of gen in a tuning context appears to support the meaning "test" for gennum, it must be noted that the most logical possibility for a term contrasted with sharpening and flattening is not fine-tuning, which on the lyre would be done with the same physical manipulations as rough tuning, but simply keeping the pitch unchanged, consistent with one meaning of gen: "(to be) permanent."^^ A successful test, after all, means that no further adjustments are necessary. A term for keeping the pitch unchanged could easily find use in performance as well as in the tuning context of Shulgi B. It would be an appropriate description for sounding the same note or dichord on the instrument for some length of time.
32. 33. 34. 35.
Kilmer and Tinney, "Old Babylonian Music Instruction Texts," 53. Kilmer and Tinney, "OB Music Instruction Texts," 54; Kilmer, RIA "Musik. A," § 5.2. Krispijn, "Musik in Keilschrift 2," 466-67. ePSD s.v. gin [establish]. Online: http://psd.museum.upenn.edu/.
THE NIPPUR MUSIC-INSTRUCTION FRACMENTS
105
Zennum is not as easily explained philologically The final consonant of the root written with zi in the meaning "rise" is not n but g,^** and there does not appear to be a Sumerian root *zen. If related to Sumerian z i, the term must have been reworked to resemble gennum. As Michalowski has pointed out (personal communication, July 2008), all the other known musical terms in these fragments are Akkadian rather than Sumerian, so Akkadian interpretations for these terms should be sought as well. Etymology alone, however, cannot answer the present questions. The occurrence of a term that is not a musical one but is found in numerous poetic compositions— the royal name Lipit-Estar—in a description of musical performance on a stringed instrument suggests that the fragments pertain to the accompaniment to singing.*^^ In such a case the most important information for the instrumentalist, after the sequence of string sets, would be how to synchronize the accompaniment with the vocals. It is likely that zennum and gennum played a role in that task. There are so many uncertainties and variables, however, that any attempt to ascertain the meaning of these terms more precisely from the materials at hand quickly becomes a layering of conjecture upon conjecture. The primary variable is the identity of the sung texts themselves. The possible clues are the words, or parts of words, that are found in the musical instructions but do not belong to the known body of musical technical terms. Most prominent of these words is the name Lipit-Estar at NMI 1 ii' 5', taken from the outset to be tbe incipit of a song.''^ The only known poetic text (or supposed poetic text) that begins with this name, however, is the one commonly referred to as Lipit-Estar Hymn B (Li B),'^^ which is now understood to be a scribal school exercise that would not be set to music.""* One might consider as an alternative the possibility that the name was written there not because the composition began or ended there but because the name, as part of the sung text, was to be sung at that point in the musical performance. This hypothesis makes it unnecessary for the name to be the incipit of the text and therefore multiplies the possibilities for the identity of the text, but there is, naturally, no evidence that such a procedure was ever followed. In several lines scattered through these instructions, mysterious syllables that do not belong to any of the known musical terms are written before the names of the string sets. The temptation arises to view these as hint syllables from the sung text, placed to aid in synchronizing accompaniment with vocals. The beginning of NMI 1 ii' 9', [xJ.i^xlZU (in which the sign immediately before ZU ends in two stacked verticals), occurs on only the fourth line following the name Lipit-Estar, from which it is separated by two occurrences of zennum and one of gennum. On the assumption that the name was to be sung where it is written in the musical instructions, might it be possible to identify the sung text by searching the known literature for an instance in which -zu is placed fairly closely after the name Lipit-Estar and then, from the distance between them, compute approximately how much musical time zennum and gennum could have taken? Such a search can be made on Lipit-Estar hymns A through E and H thanks to the online publication of these texts."" Unfortunately, the search turns up only five instances of ZU within three lines or so of the king's name, and four of them are in Li B (lines 25-27, 28-29, 33-34, and 57-58)."^ The fifth, on the other hand, is in Li E 2 (a line that is at the end of the sag4-ba-tuku section, which is followed 36. 37. 38. 39. 40. 41. 42.
ePSD s.v. zig |rise]. Online: http://psd.museum.upenn.edu/. A. D. Kilmer and M. Civil, "OB Musical Instructions Relating to Hymnody," JCS 38 (1986) 96. Ibid. H. L. J. Vanstiphout, "Lipit-Estar's Praise in the Edubba," JCS 30 (1978) 33-67. Michalowski, "Learning Music." ETCSL. Online: http://etcsl.orinst.ox.ac.uk/. Ibid., c.2.5.5.2.
106
JEROME COLBURN
by the third and fourth ki-ru-gú; that is, slightly more than halfway through the complete hymn),*"^ where the king's name is followed immediately by ú-a-zu. If that is the text that is meant, at least part of the two instances of zennum and one gennum would almost certainly have to have been played as instrumental solo of unknown duration, and nothing can be inferred about the time significance of zennum and gennum. It is, of course, equally likely that the sung text was not one of those now known, that the unexplained syllables were not part of the sung text, or both, again leaving us without a basis for inferences. èananam The new fragments added to NMI 1 by Kilmer and Peterson in the preceding article introduce a new term sananam, which by its position between gennum and zennum in i' 10' and before isartum in iii' 5', could refer to a string set or be another playing direction. Kilmer and Peterson suggest that it is connected with sanänum "to become equal, to rival, to match, to claim equality, to defy," "to reach the same height" (CAD S I 366-370). One is tempted to suppose that what was "equal" was the octave doubling of a string,'*'' and therefore that sananam meant a pair of strings an octave apart, either (1,8) or (2,9) on a nine-stringed instrument. Michalowski points out (personal communication, July 2008), however, that the accusative case form of sananam contrasts to the nominative used everywhere else for names of string sets. (There are a few instances of i-sa-ar-TIM without preceding noun or preposition, which might be read i-sa-ar-tumg) As an accusative, sananam would have an adverbial meaning, "equally" or the like, and would have to modify the sense of the sequence of string sets and zennum/ gennum terms. One possibility is that sananam simply signals a repeat; the non-mimated, and therefore perhaps plural, zennu u gennu in i' 9' and i' 12', lines preceding sananam, may be significant. Conclusions Interpreted in detail as tuning protocols, the Nippur fragments present severe practical difficulties in terms of ambiguous instructions and sometimes senseless and self-defeating sequences of actions. These difficulties do not arise when they are interpreted as descriptions of musical performance, even though many questions of detail remain. The interpretation of sehpum as a pair of adjacent strings between those of the corresponding fourth after which the sehpum is named is consistent with the musical progressions shown in the fragments and with the otherwise attested meaning of this term. CBS 1766, recently identified as pertaining to music, provides support for this interpretation. A discussion is in preparation. That zennum and gennum had some rhythmic meaning appears likely, but at present nothing reliable can be inferred from the evidence available. New studies and new material are to be welcomed. Further data will also be necessary to determine the meaning of the new term sananam. If the instrumental music from these tablets is the accompaniment, what might the singing have sounded like? Imagine trying to infer the vocal melody for a jazz standard one has never heard, given only a chord chart. The present task may be somewhat simpler, because it is likely that the vocal line generally matched one or the other of the pitches heard from the strings. For a singer to offer a different note than those played on the strings, there would have to be some way of relating that third note to the other two—that is, a general theory of harmony, applicable between the instrument and the voice 43. Ibid., c.2.5.5.5. 44. For a similar semantic process, see Kilmer, "How the Mesopotamians Did or Did Not Express the Concept of the Octave," 277.
THE NIPPUR MUSIC-INSTRUCTION FRAGMENTS
107
and capable of relating more than two pitches at a time. There is no evidence of such a theory from the OB period. It is, however, possible that the singing involved melodic ornamentation practices that we cannot infer from the documentation we have. On the other hand, the vocal line could be much simpler Might it be that the "drone" represents the actual vocal line, somewhat similar to the reciting notes in various types of chant? A discussion of the implications of the present work for the music from Ugarit, which this music does not much resemble and from which it is separated by several centuries, is in preparation. Appendix The following is a summary of the terminology encountered in the study of cuneiform musical texts and an explanation of the use of staff notation to represent the musical material discussed here. The first section of the lexical text Nabnitu XXXII consists of a list of nine strings: 1. í/udmú "first" 2. samüsu "second" 3. salsu qatnu "third thin" 4. Ea-bânû "Ea-creator," written abanû in the present fragments 5. hamsu "fifth" 6. rebi uhri "fourth behind" 7. saisi uhri "third behind" 8. sini uhri "second behind" 9. uhru "behind" CBS 10996,''^ a Neo-Babylonian list primarily of key numbers for mathematical operations,'"' contains a list of the following pairs of string numbers. They are known to be string numbers because they are preceded by the logogram SA "string" and because they are also identified by the corresponding names from the preceding string list in Nabnitu XXXII. 1,5: nls tuhri; 5, 7: sëru 2, 6: isartu; 1, 6: salsatu 3, 7: embûbu; 2, 7 rebûtu 4, 1: nîd qabli; 1, 3: isqu 5, 2: qablltu; 2, 4: titur qabllti 6, 3: kitmu; 3, 5: titur isarti 7 4: pitu; 4, 6: serdû The strings in each pair are either two, three, four, or five strings apart. The eighth and ninth strings are not used in this document, and where one would expect them, the first and second strings appear instead (as in salsatu, where after 1, 5, 5, 7, and 2, 6 we would expect 6, 8, but 1, 6 appears), suggesting a heptatonic scale."*^ This hypothesis was confirmed by the discovery of the Tuning Text,"*** in which, in response to the condition isartu or qablltu is not "clear," the musician is instructed to adjust both string 2 and string 9, from which it can be inferred that those two strings are separated by an octave. The instructions in the Tuning Text can be understood physically and musically if the heptatonic scale is diatonic, like our do, re, mi,... scale or the white keys on a piano. In that case, making the (2,6) string pair sound "clear" would mean to tune the pair to a perfect fifth (the interval between do and sol. 45. 46. 47. 48.
Kilmer, "The Strings of Musical Instruments." A. D. Kilmer, "Two New Lists of Key Numbers for Mathematical Operations," Or 29 (1960) 273-308. Kilmer, "The Strings of Musical Instruments," 265, 266. Gurney, "An OB Treatise on the Tuning of the Harp."
108
JEROME COLBURN
representing a frequency ratio of 3:2). Making the (5, 2) pair "clear" would mean to tune to a perfect fourth {do and fa, representing a frequency ratio of 4:3).^^ The requirement to adjust both strings 2 and 9 to make isartu clear also shows that, for the author of the Tuning Text, the concept of isartu was not restricted to strings 2 and 6, which define isartu in CBS 10996, but included string 9 as well. Likewise, qablltu included string 9 as well as strings 2 and 5. Evidently, the term "string pair" is too restrictive for these entities. We can dare to generalize that each name from CBS 10996 comprised not only the minimal two strings named in that text but also all the octaves of those strings available on a given instrument. As a result, the term "string set" is used in this article rather than "string pair." The term "set" also includes sets that contain only one member—that is, single strings—and it is used in that expanded meaning in this article as well. A "sequence of string sets" therefore may include any combination of single strings and string "pairs." In light of the foregoing considerations, in the present article the string sets defined in CBS 10996 as the pairs (1,5), (2,6), (3, 7), (4,1), (5, 2), (6, 3), (7, 4) are referred to collectively as "fifths/fourths." The other string sets listed in CBS 10996 are referred to as "thirds/sixths." It should be noted that the sevenstringed minimal representation CBS 10996 is a later simplification, not only because it was written later than the Tuning Text but because the rearmost string it mentions, string 7, is still called saisi uhri as if there were two more strings behind it. Below the list of strings, the first column of Nabnitu XXXII contains the beginning of a list of the fifths/fourths, each one followed by its sehpum: isartum, sehep isartim; kitmum, sehep kitmim, and so forth, as further discussed in the text of this article. The instrument (or category of instruments) described by these two texts is variously referred to in modern publications as a harp or a lyre. That is, it comprises more than a few strings, and each string simply produces one pitch that is not altered by fingering techniques. Without any intent of opening a detailed organological discussion beyond the scope of this article,"" the term "lyre" is occasionally used here as a convention, but in most instances the word "instrument" is used instead. In every diatonic scale, one of the fifths/fourths cannot be perfect. Called an augmented fourth, a diminished fifth, or a tritone, the offending interval lies between fa and ti in the major scale, or between F and B on the white keys. This fact is essential to interpreting the Tuning Text. In each step, one starts with a lyre in a state named after one of the fifths/fourths; observes that another one of the fifths/fourths is not "clear"; and adjusts one of the strings in the not-clear fifth/fourth, leaving the lyre in a new state described by the name of another fifth/fourth. The modern counterpart of the procedure is changing the F-B interval to a perfect fourth by making the F sharp or the Bflat.The steps proceed in one direction in the first series of the Tuning Text and in the opposite direction in the second series. The state called isartum, "normal," is the endpoint of the first series and the beginning of the second. In the isartum state, the not-clear fifth/fourth is (5, 2). One can therefore represent these tuning states (called "scales" in this article) by modern key signatures, using the white-key "natural" scale to represent the Babylonian "normal" scale. If the first string of the lyre was the lowest in pitch, then the second string corresponds to F, the fifth string to B, and the second series of the Tuning Text, which begins by changing the second string in "normal," corresponds to adding sharps beginning with F#. However, converging lines of evidence indicate that the first string was the highest in pitch, not the lowest.^' In
49. D. Wulstan, "The Tuning of the Babylonian Harp," Iraq 30 (1968) 215-28. 50. See now Michalowski, "Learning Music." 51. R Vitale, "La musique suméro-accadienne: gamme et notation musicale," UF 14 (1982) 241-63; 0. R Gurney, "Babylonian Music Again," Iraq 46 (1994) 101-6; see also A. D. Kilmer, "Continuity and Change in the Ancient Mesopotamian Terminology for Music and Musical Instruments," in Studien zur Musikarchäologie II, ed. E. Hickmann, I. Laufs, and R Eichmann (RahdenWestf.: Marie Leidorf, 2000), 114, 116.
THE NIPPUR MUSIC-INSTRUCTION FRAGMENTS
109
(a) Strings: Nabnitu XXXIIfirstsection 1 2 3 4 ©
n
331 ISSZ
qudmû
samusu salsu qatnu abanû
(b) Fifths/Fourths: CBS 10996 1,5 2,6
hamsu
3,7
4,1
-©-©-
331 32:
-©-
-©-
351 331
nis tuhri
isartu
embubu
nid qabli
(c) Thirds/Sixths: CBS 10996 5,7 6,1 -e-
rebi uhri saisi uhri sini uhri
7,2
5,2*
6,3
7,4
-©-
331
331
qablitu 2,4
1,3
uhru
kitmu
pitu
3,5
4,6
331 ZS31
seru
salsatu
rebûtu
isqu
(d) Tunings (Scales): UET VII74, VI/3 899 isartu kitmu
embubu
331 331
-©-©-
isartu uncl. loosen 6
nîd qabli 33:
pitu unclear loosen 4
pitu »O-
kitmu uncl. loosen 3
nîs tuhri 1?rr
serdû
331
-©-
qablîtu unclear loosen 2,9
titur qabliti titur isarti
,
33: 331
nid qabli unclear loosen 1,8
embubu uncl. loosen 7 ,
qablîtu 35:
nîs tuhri unclear
Fig. 5. Strings, string pairs, and tunings represented in staff notation on the baritone clef.
that case the second string corresponds to B, the fifth string to F, and the second series begins by flattening B. The five lines and four spaces on a modern staff are attractive as a representation of the nine strings of the lyre. If the first string was highest, the top line of the staff must represent C, which could be indicated by a C clef on the top line or by an F clef on the middle line, one line lower than its position when it is used as the ordinary bass clef. The latter, known as a baritone clef, is supported by the software the present author is using and therefore is used here. The strings, string sets, and tunings discussed here are summarized in this staff notation infig.5. The staff notation, of course, remains misleading in that we do not know what the actual pitches of the strings were, but the representation is valid for the relationships between the pitches. Under this representation the instrument is a transposing instrument, with the amount of transposition unknown.
z u DEN AKKADISCHEN HEMEROLOGIEN AUS HATTUSA {CTH 546), TEIL I. EINE HEMEROLOGIE FÜR DAS „RUFEN VON KLAGEN" {SIGÛ SASÛ) UND DAS „REINIGEN SEINES GEWANDES" {SUBÄT-SU UBBUBU): KUB 4, 46 (+) KUB 43, 1 Jeanette C. Finche {Leiden}
Die hethitischeu Gelehrten in Hattusa zeichnen sieb durch eine modern anmutende Haltung zu den geistigen Errungenschaften ihrer Epoche aus, denn sie beschäftigten sich nicht nur mit ihrer eigenen Kultur sondern auch mit derjenigen der angrenzenden Länder, Ein besonderes Augenmerk wurde dabei auf Mesopotamien gelegt, weil hier die beiden Dialekte des Akkadischen—Bab\loniscb und Assyrisch—gesprochen wurden, welches zur damaligeu Zeit als Ungua franca im ganzeu Vorderen Orient bis hin nach Ägypten aligemein anerkannt war. Um mit anderen Ländern in Kontakt zu treten, war das Beherrschen des Akkadischen unverzichtbar. Somit mußten sowohl diejenigen Kaufleute, welche mit Mesopotamien und Syrien Handel trieben, als auch hohe Offiziere, viele Diplomaten, internationale Kuriere und eine bestimmte Gruppe von Schreibern diese Sprache nicht nin- sprechen, sondern zum Teil auch schreiben und lesen können. Ihre Ausbildung erfolgte in ihrem Heimatland—wohl zunächst hauptsächlich in der hethitischen Hauptstadt—nach mesopotamischem Vorbild, indem Texte des babytonischen Schul-Curriculums im Unterricht durchgenommen und anschließend entweder ab- oder nach Diktat bzw. aus dem Gedäcbtnis niedergeschrieben wurden'. Gelegentlich wurden auch Übersetzungen
Diese Untersiichnng ist im Rahmen des von Wilfred H. van Soldt, Leiden, geleiteten Projektes Transfer of Knowledge in a Cuiwifiirm Culture entstandeii, das von der Nedcriandse Organisa tic i:<x)r Wí'fL'nsc¡ia¡)}x.'tijk Onàvrzoi-k (NWO) finanziell und in Zusiinimeiiarheit mit dem Leiden ¡uHlitute jor Area Studies (LIAS) orftani.siert wird. Nicole Pleitee, Würzburg, .sei an dieser Stelle für ibre krilisclie Durchsiebt des Manu.skriptes gedankt. In die.sem Artikel verwendete Abkürzungen, die AHiv oder CAD nicht verzeichnen, sind fcilgende; ßB^^O = Berliner Beiträge zum Vorderen Orient (Berlin: Dietrich Reimer Verlag, 1987-); CTH = Emmanuel P. Laroehe, Calaliigue des textes hittites [Paris; Éditions Klincksieck. 1971); Emar VI.4 = D. Arnaud, Reclierclies au ¡xiijs d'Astata. Emar Vi/-í. Textes de la bibliothhfue: transcriptions et traductions, Mission Archéologique de Meskéné-Emar {Paris: Recherche sur les Civilisations, 1987); HZL = C Rüster und E. Neu. Hethitisches Zcichenlexikon {Wiesbaden: Harrassowitz, 1989), 1, Für das babvlonischt' Sehul-Curriculum und die damit verbundenen Lehrmethoden vgl. zuletzt umfassend P D. Gesehe, Schulunterricht in Babylonien im ersten Jahrlausend v. C/ir, AOAT275 {Münster: Ugarit-Verlag, 2001).
111
JCS Hl {2(K>9)
112
JEANETTE C. FINCKE
ins Hethitische angefertigt und entweder dem akkadi.schen oder sumerischen Text der Vorlage in einer zu.sätzlichen Kolumne beigefügt" oder auf einer gesonderten Tafel festgehalten^. Einige Werke der mesopotamischen Literatur sind unter den in Hatttisa aufgefundenen Tontafelu ausschließlich in der hethiti.schen Übersetzung bzw. Adaption überliefert'', andere hethitischsprachige Werke zeigen einen eindeutigen mesopotamischen Hintergrund'', Während der althethitischen Zeit wurden sogar hethitische Erzählungen in akkadischer Sprache niedergeschrieben, ohne daü sich eine hethitiscbe Version— sofern eine solche existierte—erhalten hat*^. Die ersten Lehrer des Akkadischen' kamen wahrscheinlich aus dem mit dem hethitischen Reich eng verbundenen Syrien**, möglicherweise auch aus Mesopotamien^ bzw. Babylonien'" selbst. Die von ihnen ausgebildeten Hethiter übernahmen dann später selbst
2. Dies ist besonders bei einif^en lexikalischen Listen (z. B. CTH 300, 301 und 302) sowie bei einigen Konfipositionen der Weisheitslitpratnr {CTI! 315 und 316), Hvmnen {CTH 314), Gebeten (C77Í 312), Ritualen' (CTH 792) und Texten aus dem Bereicb der Divination [CTll "333, Ü47 549 und 552) der Fall. 3. Dies findet sich /. B. beim Cilgames-Epos ICTII341 ) sowie bei vielen Omentexten, \%]. hierfür z. B. K. K. Riemschneider, DU' akkadischen und hethitischen Omenlextc aus Bagazköy, DBH 12 (Dresden: Techni.sche Universität Dresden, 2004) (unveränderte Publikation des vorläutigen Manuskripts seiner Habilitationsschrift aus dem Jahre 1973'; vgl. hierfür SMEA 46/2 |2004| 216Anm. 2). 4. Vgl. z. B. das mr tóm/ian-Epos [CTH 310) (vgl. hierzu zuletzt E. Rieken, J > r hetbitische .sar-tom/iäri-Text: arcbaisch oder archaisierend?", in Akten des IV internationalen Kongresses für Hethitologie Würzhwg, 4.-8. Oktnlx-r 7999, Hrsg. G. Wilhelm. StBoT 45 IWie.sbiiden; Ilarrassowitz, 2001|, 576-85), die Erzählung um Nariîm-Sîn in Anatolien (CTH 311 ) und der AtnmihäsisMytbos {CTH 347). 5. In diesem Zusammenhang sind die aus Babylonien importierten und ins Hethitische übersetzten Rituale zu nennen, wie z. B. das Krs;itzkönigs!itual (vgl. H. M. Kümmel, Ersalzritnale für den lietintischen König. StBoT 3 ¡Wiesbaden: Harrassowitz, 1967), das ..Rituel contre l'insomnie" (so CTH 432; vgl. Pb. 11. J. llouwink ten Gate in SÍÍÍ and Sanction in israA and Meso^xitamia: A Comparative Study, Hrsg. K. van der Toom. Studia semitica neerlandica 21 [As.sen: Van Gorcum, 1985|, 125-33, und G. Beckman, „A Hittite Ritual for Depre.s.sion {CTH 432)" in Tahiilaria Hcthacorum: Hethitische Beiträge Silvhi Konak zum 65. Geburtstag, Hrsg. D. Groddek nnd M. Zorman, Dre.sdner Beiträge zur Hethitologie 25 ¡Wiesbaden: Harrassowitz, 2()07|, 69-81) sowie die kító/í-Hituale (vgl. G. Beckman, Jrhe '"/«(/»/f-Ritiiiil" from Bogazköy {CTH 718)". in Receiit Develojnm'uts in Hittite Archacotogy and History. Pa}iers in Memory of Hans C Cüterbock, Hrsg. K. A. Yener inid H. A. Hotfner Ir. |W^inon¡i Uike, IN: Ki sen brau ns, 20021,35-41). 6. Vgl. z. B. die althethitische historiographische Erzäbinng von der Belagerung von ürsu in akkadiscber Sprache, vgl, hierfür zuletzt G. Beckman, „The Siege of Ursu Text {CTH 7)", ICS 47 (1995) 23-34 Das Feblen einer entsprechenden betbitiscben Version kann aber auch zufällig sein. 7. Mit den mtiglichen Traditionswegen akkadischer Literatur nach Hattusa haben sich mehrere Wissenschaftler be.schäftigt, vgl. 7,. B. G Beckman, „Mesopotamians and Mesopotamian Learning at Hattusa", jCS 35 (1983) 97-114; A. Arthi, ..Hethitische Mantik und ihre Beziehungen zur mesojxitami.schen Manlik", in Meso}X)tamieii und seine Nachbarn Hrsg. H. Kühne, H. J. Nissen und J. Renger. BBVO I {- CRRAI 25) (Berlin: Reimer, 1987), 279-93; G Wilhelm, Mcdizinisclw Omina aus Hatlusa ht akkadischer Sprache, StBoT 36 (Wiesbaden: Harrassovi itz, 1994), 1-18. Den Fokus auf die Paläographie derjenigen akkiidiscli geschriebenen Texte, welche nicht zur St hiilausbildung gehören, wie z. B. die akkadisehcn Fassungen helhitisfber Staat.svcrträge oder Briefe, legt J. Klinger, „Wer lehrte die Hethiter das Schreiben?", in ///. Vlunlararasi iiiütoloji kongresi bildirleri Corum ¡6-22 Eyliil 1996 - Acts üf tbe UIrd International Congress of Hitlitoh)gy Çorum. 16-22 September ¡996, Hrsg. S. Alp und A. Süel (Ankara, 1998), 365-75; S. 370-72 weist er nach, daii einer der filihesten datierbaren Texte der Hethiter in akkadischer Sprache, derjenige akkadiscbe Brief yattnsilis !.. welcher im syrischen Tikunani gefunden wnrde, nicht von einem Hethiter geschrieben wurde, Daraus schliefet er. daft es zu jener Zeit noch keine Spezialisten für das Abfassen akkadischer Texte in Hattusa gab. 8. Der Eintluli syri.siber Scbreiber auf das im hethitischen Königreich gefundene Schrifttum geht mindestens bis ins 17, Jb. V. Chr. zurück, als die Hethiter die Keilschrift übernahmen, vgl. Beckman, jCS 35 (1983) 100; vgl. aber auch K. Hecker, „Zur Herkunft der hethitischen Keilscbriff, SCCNll 8 (1996) 291-303, der die syriscben Zeichenformen bereits im /räiii/ii-zeitlichen Anatolien nachweisen kann. Wäbrend der mittelhethitischen Zeit läEt sich ein erneuler Import von Schrifttum ans Syrien nachweisen, vgl. Beckman, jCS 35, 102-11. 9. Vgl. hierfür die hetbitiscben Übersetzungen mesopotamiscber Epen und Hymnen, die auf altbabylonische Vorlagen zurückgehen müssen, vgl. bierfür Beckman,^CS 35 (1983J 100-102. Diese Vorlagen könnten zwar von syrischen Gelehrten nach IJattu.sa mitgebracht worden sein, ein direkter Kontakt von Hethitern mit mesopotamiscben Spezialisten sollte ahcr nicht au.sgeschtiLssen werden. 10. Klinger, Acts of tbe Hlrd International Congress of hittitolngy Çorum (1996), 374, kommt zu dem Ergebnis, daß die „bei den Hethitern gebräuchliche Keiiscbrift ... auf eine Fonn der altbabylonischen Kursi\'f zurüi k|geht|. die älter ist als die im nordsyrischen Raum zur Zeit Hattusilis I. gebräuchliche". Das bedeutet, dafe im Verlauf der althethittschen Zeit eine neue Gruppe von Lehrern wohl direkt aus Babylonien zu den Hethitern gekommen sein muß.
DEN ARKADISCHEN HEMEROLOGIEN AUS HATTUSA (CTH 546). TEIL 1.
113
den Unterricht in der hethitischen Hauptstadt. Andere Lehrer kamen in der Eolgezeit direkt aus Mesopotamien" oder aus dem hurritischen Kulturraum'" wie z. B. aus dem Großreich Mittani, dem Land Nuhasso'' oder dem stark htirritisch beeinflußten und im Südosten des hethitischen Großreiches gelegenen Land Kizzuwatna''', Einige der akkadischen Texte aus Hattusa wurden von den Lehrern seihst geschriehen, andere von ihren Schülern, Weil Akkadisch-Lehrer zu unterschiedlichen Zeiten und aus unterschiedlichen Orten nach Hattusa kamen' ' und an den Sc hulen der verschiedenen Länder jeweils eine andere Ausprägung der Keilst hrift gelehrt wurde, finden sich unter den akkadischen Tafeln aus Hattusa Texte mit unterschiedlichem Duktus'*^: spät-alt- (bzw. archaisierender) und mittelbabylonischer Duktus, assyrischer Duktus des 14. Jh.s v. Chr., mittanischer und assyro-mittanischer Duktus'^ sowie spät-alt-, mittel- und junghethitisoher Duktus'^ Im Zuge ihrer Beschäftigung mit dem babylonischen Schrifttum studierten die Hethiter auch die in Mesopotamien übliche Art der Divination, indem sie diese sowohl in der Originalsprache lasen und tradierten, als auch ins Hethitische übersetzten'"'. Während die Hethiter in der Regel provozierte Diviiiationstechniken durt hfiihrten, waren in Mesopotam ¡sehen die un provozierten Methoden von größerer Bedeutung, sofern man die Menge an Schriftzeugnissen als Hinweis auf das Ansehen innerhalb der Gesellschaft intei-pretieren daif. Das bedeutet, daß die Hethiter neben der Eingeweideschau verstärkt das Verhalten von Tieren und anderen Objekten in einer speziell arrangierten oder definierten Umgebung beobachteten und nach vorher festgelegten Vorgaben als positiv oder negativ deuteten^". Diejenigen
11. Die Anwesenheit meso[X)tamischer Schreiber in Hattusa kann für die niittelhethitis(he Zeit nachgewiesen werden; währtMid der Croßreitliszeil arbeiteten mehrere assyrische und bahylonische Experten in der hethitischen Hanptstadt, vgl. Bet-kniim, JCS 35 ÍI983) 102-lÜ, hes. 108 (Liste mit den Namen der Spezialisten aus den entsprechenden Liindern), 12. Vgl. in diesem Zusammenhang z. B. die in HattuSa aufgefnndenen Omentexte in hunitischer Sprache, vgl. hierfür zuletzt S. de Marti no,/J/i" umnÜHchen 7Í!if(',ChS 1/7 (Roma: Bfinsijinori, 1992). Bei den Eingeweideschaucimina venvenden die hnnitisehen Texte sumeri.sfhe termuii tevhuicu während die vergleichbaren hethitischen Texte diese dnreh hurritisehe ersetzen. Darans läKt sich ein Traditionswefi von Mesopotamien nher die Hurriter zu den Hethitern nachzeichnen, vgl. de Martino, Die Texk', 3-4. Vjrl. ferner J. KHiifier, „Die hurritisehe Tradition in Haltusa und das Corpus hui ritischer Texte", in Kiiltin^ Altonetitaüscfie Studien für Volkert Haas zum 65. Geiur/stog, Hrsg. T. Richter et ai, (Saarbrücken: Saarbnit ker Dnickerei, 2001), 197-208. 13. Vgl, z. B. die beiden Tafeln mit hethitischen Omina KVB8, 29 (Mondomina] und KBa 10, 7 (Eingeweidesehan: KI.CUB), die beide von Schreibern mit hurritischem Namen aus Nnhasie gesehtieben wurden, was vermuten läßt, daK sie selbst den Text zuvor aus dem Akkadischen ins Hethitisehe übersetzt hatten. Oh die in llatttisa anwesenden Gelehrten aus Nuhaä5e auch Akkadiseh unterriehteten, läßt sich anham! der Belegla^e natürlich nicht nachweisen. 14. Vgl hierfür aueh J, L. Miller, .SÍHÍ/ÍCT in theOiights. Dc-vchpiiwut and liiteqm'tatitm of the Kizzmtxitna Rituals, StBoT 46 (Wiesbaden: Harrassowitz, 20Ü4), und R. StrauR, Rciriigtinguriten aus Kizzuicatna (Berlin: de Griiyter^ 2006). 15. Einen Überblick über die zeitliche Verteilung der einzelnen Textgenera akkadisi her Lileratur in IJattu.sa gibt Beckman, ;CS35 (19831 97-114. mit u-eiterer Literatur. 16. Vgl.J. Klinger, ..Zur Paläographie akkadisehsprachiger Texte aus Hattusa", in Hittite Studies in Honor of Hairy A. Hoffner ¡r. mi the Occasion of his fíStli Hirthday, Hrsg. G. Bt^ckman I't al. (Winona Lake, IN: Eisenbranns. 2003), 238-48, der sieh vorwiegend mit den akkadisch gescbriehenen Staatsveilrägen aus Hattusa be.sehäftigt und herausgefunden hat, daß z. B. einige der Abschriften von Staat.sverträgen Suppiluliumas I. in akkadischer Sprache aus demjenigen Land stammen, mit welehem der Vertrag geschlossen wurde, nnd damit die Schreibertradition des jeweiligen Vertragsiandes widerspiegelt. 17. Für die Unterscheidung von assyro-mittanischem, niittanischem und as.syri.schem Duktus des 14. Jh. vgl. die Paläogfapbie von D. Sehwempr. Akkadische Ritiiahaiis Hattusa, THcth 23 (Heidelberg: Winter, 199S), 17-39. 18. Vgl. hierfür die Zeichenlisten in C. Rüster nnd E. Neu, Ufiliitischt's Zeicheulcxikon (Wiesbaden: Harrassowitz, 1989); C. Rüster. Helliilisclic K('il.\chrift,-Pal(iogra})hie, StBoT 2(1 (Wiesbaden: Harrassowitz, 1972); F. Starke, Die kcilNchriftliiu^ischcii Texte in Ihiiscimtl StBoT 30 (Wiesbaden: Harrassowitz, 1985), 59, 82.110.142. 219. 302 (vgl. auch besonders S. 21-31). 19. Beekman,yCS35 (1983) 101, hält alle akkadischon divinatorisehen Texte ans HattuSa fin" ursprünglich direkt aus Mesopotamien importiert. Die.se Einschätzung kann nach den inzwi.schen erfolgten Stndien der divinatorisehen Texte so nicht mehr aufreiht erhalten werden, vgl. hierfür z. B. oben Anm. 12 (für Eingeweide-Omina) sowie Wilhelm, StBoT 36 (1994), 2-3, 5 (für diagnosUsch-prognostis( he Omina) nnd U. Koch-Westen holz, „Mesopotamian Astrology at Hattusas" in Die Rollt' der Astronomie in den Kulturen Mesojiotamicns. Hrsg. H. D. Galtcr. Cirazer Morgen land i sehe Studien 3 (Graz: mi-Drvick- & Verlag.sgesellschaft, 1993). 232-35 (für asttologische Omina). 20. Vgl. hierfür die Übersieht liber die entsprechenden Techniken von Th. van den Hont, ..Omina (Omens). B, Bei den Hethitern", RIA 10 (Berlin: de Gruvter, 2003-2005), 88-90.
T
114
JEANETTE C. FINCKE
Texte, welche derartige Verfahren beschreiben, werden in der Hetbitologie als Orakeltexte oder mantische Texte bezeichnet. In Mesopotamien wurden hingegen eher 7Atfällig beobachtete Erscheinungen am Himmel und auf der Erde als ominöse Zeichen betrachtet. Weil in Mesopotamien jede.s ominöse Zeichen als von den Göttern hervorgerufen galt, mußte jedes einzelne Zeichen eine ganz spezielle Bedeutung haben und konnte nicht einfach als positiv oder negativ gewertet werden. Die korrekte Interpretation dieser Zeichen war Aufgabe von Spezialisten, die ihr im Laufe der Zeit durch Erfahrung immer stärker anwachsendes Wissen in Handbüchern zusammenstellten. Als Eorm für ihre Aufzeichnungen wählten die Spezialisten den Konditionalsatz, bei dem das zu beobachtende ominöse Zeichen im mit summa, „wenn'\ eingeleiteten Vordersatz (Protasis), nnd die entsprechende Vorhersage im nachfolgenden Hauptsatz (Apodosis) steht"'. Mit den Omentexten eng verwandt sind die Hemerologien^l Während die Omentexte ominöse Zeichen beschreiben und hinsichtlich ihrer Bedeutung für zukünftige Geschehnisse deuten, stehen in den Hemerologien Ge- tmd Verbote für bestimmte Handlungen an einzelnen Tagen des Jahres, die gegebenenfalls von Vorhersagen über die persönliche Konsequenz bei Nichtbeachtung der Vorschrift begleitet werden. Die Verbalformen in den Protasen der Omentexte stehen demnach in der 3. Person oder im Stativ, während diejenigen der Hemerologien bei den Vorschriften nnd Verboten im Prekativ oder Imperativ bzw. Prohibitiv, bei den Vorhersagen hingegen in der 3. Person oder im Stativ stehen. Besonders fragmentarisch erhaltene Texte beider Gruppen können demnach auf den ersten Blick identisch erscheinen. Der Unterschied läßt sich oft nur bei denjenigen Texten erkennen, bei denen die Verben nicbt logographisch, sondern syllabi.sch geschrieben werden. Dieser Umstand sowie die Tatsache, daß in beiden Textgattungen Vorhersagen über die Zukunft von Individuen enthalten sind bzw. sein können, hat dazu geführt, daß sowohl Omentexte als auch Hemerologien zur Gruppe der divinatori.scben Texte gezählt werden. Entsprechend hat E. Laroche^^ alle akkadischen und hethitischen Omentexte, Hemerologien sowie die hethitischen Orakeltexte im Kapitel „Divination" unter den Nummern CTH 531560 zusammengestellt und nach Inhalt geordnet. Die seit Erscheinen von CTH publizierten Texte aus iJattusa wurden jeweils den von Laroche etablierten CTH-Nummern zugeordnet. Mit zwei Eragmenten aus der Gruppe der divinatorischen Texte aus Hattu.sa, die beide von derselben Tafel mit einer Hemerologie .stammen, beschäftigt sich die vorliegende Studie: KUB4, 46 (Bo 2196) ist das Fragment einer akkadi.schen Hemerologie (CTH 546.5), die sich auf das „Rufen von Klagen" (vgl. Z. 13: si-gu lil-^si^) bezieht. Die Anordnung des Textes auf der Tafel ist außergewöhnlich, denn die Zeilen sind über je vier Kolumnen hinweg ge.scbrieben; In der ersten steht i-na, in der zweiten der Monat, in der dritten der Tag (genannt wird nur der 15. Tag), und in der vierten Kolumne, von der nur der linke Abschnitt mit den Anfängen der Sätze erhalten ist, folgen die Ver- und Gebote bzw. Vorbersagen. Auf dem als KVB 43, 1 (Bo 1787) publizierten akkadischen Omenfragment (CTH 560,1). bei dem der Ansatz des oberen Randes erkennbar ist, sind drei Kolumnen vollständig erhalten: Die linke enthält Anweisungen und Vorhersagen, gefolgt von einer Kolumne, in welcher i-na steht, an die sich eine weitere anschließt, in welcher der Monat steht. Von der ersten Kolumne des Stückes ist am linken Rand des Fragmentes nur der rechte Abschnitt erhalten. Dennoch ist genug Text zu lesen, um das Stück als Hemerologie zu bestimmen und als vorrangig behandeltes Thema das „Rufen von Klagen" (vgl. Z. 1,3: [sigû U] Isi) und das „Reinigen seines Gewandes" (vgl. Z. 1, 3, 5: TÚG-sú DADAG) zu erkennen. 21. Für die Gleichförmigkeit des FomiuLirs dieser Omensätze mit dein jen i^cn der ¡ikkudischen Gesetzestexlp und dem dieser Übereinstininiung míifííicherweise zugrundeliegenden Q'danken vgl. zuletzt j. ('. Fincke, „Onniiit, die göttlichen „Gesetze" der Divination", j£0/. 40 (2006-2007) 131-47. 22. Einen ersten Überblic k iihev diese Textgruppe bietet R. Labat, „Hemerolf^ien". RIA 4 (Berlin: de Gruyter, 1972-1975), 317-23. 23. E. Laroche, CTH (1971).
z u DEN AKKADISCHEN HEMEROLOGIEN AUS HATTUSA {CTH 546), TEIL T.
115
Der gleichartige Aufbau beider Fragmente—bei KUB 43,1 feblt nur diejenige Kolumne, in welcher der Tag bezeichnet wird—sowie die Überschneidungen in Her behandelti'n Thematik macht die Zusammengehörigkeit zur selben Tafel wahrscheinlich, obwohl die Kopien der einzelnen Stücke auf den ersten Blick keine Übereinstimmung erkennen lassen: KUB 4, 46 ist von Ernst E Weidner in einer stark sy.stematisieiien Art kopiert worden, während KUB 43,1—aus der Hand von Ka.spar K. Riemschneider— auf den ersten Blick in einem anderen Duktus geschrieben scheint. Von beiden StiU ken ist der Fundort luibekannt, .so dais die Möglichkeit, über diesen eine Zusammengehörigkeit zu bestimmen, ausscheidet, Weil nur für KUB 4, 46 ein Photo zur Verfügung stand (vgl. die Online-Datenbank mit der ,JConkordanz der hethitischen Texte" des Hethitologie Portal Mai7iz ¡online; http://www,hethport.uni-wuerzburg,de/ hethkonk/|). hat Jared Miller im Sommer 2008 freundlic'he!"vteise XL'ß43, 1 in Ankara für mich angesehen tind photographiert. Für seine Hilfshereit.schaft möchte leb ihm auf diesem Wege ganz herzlich danken. Der Vergleich beider Photos ergibt, dais die Oberfläche von KUB 4,46 leicht abgerieben ist, während die Keile auf dem Fragment KUB 43, 1 noch scharf eingetieft sichtbar sind, wodurch auf den ersten Blick der Eindruck zweier unterschiedlicher Handschriften entsteht. Daß beide Fragmente dennoch von derselben Tafel stammen, zeigt sich unter anderem daran, daß auf beiden Fragmenten die Linien für die Koiumnentrenner nicht parallel, sondern in zum Teil recht stark abweichenden Winkeln zueinander gezogen winden. Die Linien scheinen zudem nicht in einem Arbeitss( hritt mit einem Lineal oder einer Schnur erzeugt worden zu sein, denn sie sind zum oberen Rand beider Fragmente hin stark gebogen. Bei dem erhaltenen Abschnitt dürfte es sich um die Vorderseite der Tafel handeln, denn der Schreiber hat die Kohminen in der linken Tafelhälfte (= KUB 4, 46) noch einwandfrei konzipiert, so daß die Keile direkt neben den Kolumnentrennern beginnen, während er auf der rechten Hälfte {= KUB 43,1) in Platznot geriet und für diejenige Kolumne, welche das Wort ina enthalten sollte, zu wenig Platz vorgesehen hat, wodiin h die Keile von i- bereits direkt auf dem Koiumnentrenner beginnen mußten. Um diese Eigentümlichkeiten der Tafel deutlich zu machen, habe ich mit Hilfe des von Jarcd Miller aufgenommenen Photos und demjenigen der Online-Datenbank die beigefügte Kopie angefertigt, KUBA, 43 {+) KUB4Q,l ist Bruchstück einer akkadi.schen Hemerologie, die sich in der linken Kolumne auf Has Rufen von Klagen in Verbindung mit Hem Reinigen Hes Gewandes bezieht; das Thema der rechten Kolumne läßt sich nicht ermitteln, Bevor auf den Inhalt dieser Hemerologie näher eingegangen wird, soUeu zunächst Orthographie und Paläographie Her Fragmente in Hinblick auf die Herkunft Her Tafel untersucht werHen. Orthographie Bei Hen Vorhersagen für das Individuum vei*wendct Her Schreiber von KUB4. 46 ( + ) KUB4S, 1 Hie besonHers in divinatorischen Texten aus Hattusa, Ugarit und Emar zu findende Schreibung ZA für atnclu, \ Das bedeutet, daß Her Schreiber dieser Tafel aus dem Bereich dieser drei Kulturzentren
24. Dem neubahvlonischen .^Berlin Vocabulary VAT 244" zufolge ist ZA neben dem eiiifathen sKiikiechten Keil SANTAK das im Ladfl Snhi—SC. dem ö.stlicK an Muri anschlieBenden Land am miltleren Euphrat zwiischen Rapiqti und Hindiinii—geläufifie Logogramm für ..Mann", vgl. Rs. iv 3f.; ZA | MIN (sv. a-nte-lu) EME SUH.A, 4; m-an-tak DIS | MIN (sc. a-mi-lü) EME SÜH.A (vgl. G. A. Reisner, ZA 9 |1894| 163). Die Gleichung von ZA mit amelu findet .sich auch in der Li.ste LÚ = Sa Tafel I Z. 14 (ZA | a-mi-lu), vgl. MSL XII 93, in der Serie Ea = nâqn Tafel I Z. 20 (za-a \ ZA | |MIN| {sc. za-zu-u) \ a-nii-Ui), hier hereits in der mittelassyri.scheii Version, vgl. MSL XIV 176, sowie in Proto-Ea Zeile 167 {za-a \ ZA], vgl. MSL XIV 38, und im Syllahar S'' Tafel I Z. 7 lza~a I ZA | a-mHu). vgl. M.SMII 9fi, Fiir die Verwendung von ZA fiir akkad. amëlu, .,Mann", in Hattusa vgl. ferner den im mittelbabylonisc'hen Duktus geschriebenen Text mit physiognomischen Omina KUB 37, 210, den im assyrischen Duktus des 14, Jh.s v. Chr. grsrh rieben en physiognomischeri Ümentext AfO Beih. 3 Nr. 65 (Bo 6263) und das physiognomische Omenfragment KBo 46, 269 sowie das wohl
116
JEANETTE C. FINCKE
stammt bzw. hier seine Ausbildung zum Schreiber genossen hat^^ Das Fragment KUB 4, 46 zeichnet sich ferner dadurch aus, daß der Schreiber das Wort sigû mit dem Zeichen SI be^^innen läßt und damit nicht der hethitischen Konvention, sondern der korrekten mesopotamischen Schreibweise folgt. Daft es sich bei diesem Fragment nicht um ein Importstück aus Babylonien oder Assyrien bzw. um eine von einem Babylonier oder Assyrer gescbriebene Tafel handelt, läßt sich daran erkennen, daß beim Prohibitiv der 3. Person von sasû die fiir bethitische Scbreiber charakteristische St hreibung der s-haltigen Silben mit s-haltigen Keilschriftzeichen vorhanden ist {lä isem, „er soll nicht rufen"', in der Schreibung NU i-,sii-î,S".si]); bei den anderen hemerologischen Texten aus Hattusa zeigt sieb diese Besonderheit hingegen häufiger beim Nomen sigû als bei der Verbalform von sasû (vgl. KUB 4,44 Vs. 7: si-gu-u und 10: ',si-gH'-|i/|, vgl. aber auch KUB 4, 45 Vs, 5': si-gu-u ú-si+is-sú und 8': si-gii-u ÍÍ-,SÍÍ-I-Í,S-.SÍÍ]). Der Schreiber dürfte eine Vorlage gehabt baben, von der er den Anfang dieser speziellen Phrase korrekt abgeschrieben, sich dann aber auf sein Gedächtnis verlassen hat, wodurch sich die für ihn geläufigere Schreibung einschleichen konnte. Paläographie Die fragliche Tafel ist kein Importstück aus Babylonien, wie sich an der Form des Zeichens TUG im Fragment KUB4S, 1 erkennen läßt; Dieses Zeichen besteht aus insgesamt vier horizontalen Keilen, von denen die mittleren beiden etwas nacb rechts eingerückt sind, und einem abschließenden vertikalen Keil. Die entsprechende alt- oder mittelbabylonische Zeichenform hätte in.sgesamt fünf horizontale Keile baben müssen (vgl. Labat, Manuel d^'pigraphie akkadienne [Paris: Geuthner, ''1988|, Nr, 254), Die Form von TUG zeigt vielmehr die typische hetbitisc he Ausprägung der älteren Zeit (vgl. HZL Nr, 212, erstes Beispiel). Auch das Zeichen KI des Fragmentes KUB 4.46 Vs. i 10 läßt sich mit dem althethitischen Zeichen des Zidanza-Vertrages der Ilethifisciwn KeiLschrift-Paiäographie von C, Rüster (StBoT20 [1972], 192) vergleichen. Das Zeichen HI/DUG in KUB áS, 1 präsentiert mit der Sequenz von vier schrägen über einem schrägen Keil eine Variante, die der Keilschrift-Paläographie zufolge sowohl in der althethitischen als auch in der späten Großreichszeit Verwendung fand (vgl. StBet.sbeschwönmgen", StP Ser. Maior 5 (Rome: Ponteticio Instituto Hihlico,1976), 111-13. 30. VAT 7445 {KUB IV 47) Rs. 13-15: a-na i-du-u NL' i-du-u a-na ]Xi-Uir ar-ni-i\a\ (14) a-na na-sà-ali GlO-ia a-na su¡)-sur ma»íí-í)-í|íi| 115) si-