Revisiting bala.

• Author: Dahl, Jacob L.
• Position: Provincial Taxation and the Ur III State - Critical essay

The Ur III period (ca. 2100-2000 B.C., according to the conventional middle chronology) is well known for its massive bureaucracy, which has left us with tens of thousands of cuneiform documents. The brief period when the kings in Ur seem to have exercised absolute power and to have controlled nearly every aspect of economic life, essentially spanning the forty-five years between the adoption of Shulgi's administrative reforms (sometime after Shulgi 21) and the demise of dynastic supremacy during the early years of Ibbi-Suen, is one of the best documented periods from the ancient world. From this brief period we have detailed information on all kinds of mundane topics, in particular from the old provincial centers Girsu and Umma, from royal administrative centers such as Drehem, and from such newly discovered "rural estates" as Garshana. Nevertheless, vital aspects of the social, the economic, and above all the political system, still escape our understanding. Specialists have agreed for decades that one of the most vexing questions for understanding Ur III economic and political history addresses a term very common in accounts, namely Sumerian bala.

The recent publication of a revised 1999 Harvard dissertation proposes to assist us in understanding this important administrative term, which as a verbal stem means simply "to cross over," often found in documents describing, for instance, the transfer of living beings or goods from one river bank to the other. There seems already in the ED IIIa period to be an idiomatic usage of the term to describe a passage of time, thus "period," and when speaking of rule, a "term of office," as bala has indeed been described since the first Assyriological treatments of the term. Tonia M. Sharlach, the author of the book under review, has made it her task to define the problematic economic, political, and perhaps religious system encompassed by the term bala during the Ur III period. In her introduction she gives a brief outline of the history of the period and presents her hypothesis about the bala system. In the following chapters she describes the bala contribution of the Lagash and Umma provinces. Each of these chapters is split into two comparable sections: one concerning animals, and one concerning all other contributions to the bala. A brief conclusion follows. The book is appended with a number of charts, indices, and about 150 never-before-published texts.

While a review of the terminology surrounding bala must be welcomed, Provincial Taxation is burdened by an insufficient and occasionally illogical understanding of the nature of third-millennium accounting practices and of the structure of the Ur III empire, and by a disquieting number of editorial errors. It is, in the end, a valuable compendium of data that will assist specialists in their ongoing debate on the nature of bala, but due to its unacceptable typographic and citation errors and its limited control of the primary documents it is based on, the book cannot be recommended for the non-specialist or casual reader.

Let us start with the errors in formal presentation that lead to a certain inaccessibility. Most troubling is the author's imperfect system of referencing primary sources. When the author quotes from the primary sources, she never cites the line of the tablet she refers to. This may not be a problem when quoting documents with only one entry, rarely longer than five to eight lines; however, when using data and arguing on the basis of multicolumn documents the exact location of a reference is not only needed, it is critical to following and testing the author's argument.

There are many cross-references in the book under review, but few where the precise location of the reference is actually given. Statements such as "as we have already seen," "as we have discussed," etc., are found throughout the book and the reader is sent on an unnecessary search for the reference. We note, for example, p. 90, where the author writes: "For instance in the b al a account RA 4.78, which we have already examined,..." and find that it was briefly mentioned on p. 81. (This is difficult enough, but in addition the references are often incorrect. In this case the text was actually published in RA 5, not RA 4, and is generally known under the reference RTC 305. As is general in Assyriological literature, reference should be to the first reliable publication.) A similar problem exists with regard to the texts published in Provincial Taxation (mainly from the British Museum and the Yale Babylonian Collection). These texts are given publication numbers (see the concordance on pp. 271-74), but the author never gives these text numbers when she cites them in her main text, but instead cites museum numbers, thus sending the reader searching through the concordance.

Following a very brief introduction to the period, Sharlach approaches her topic with a short philological and etymological discussion of the term bala (p. 16). She defines the sources broadly (p. 17) to include texts without the key-term bala, as long as these refer to transfers of goods from the provinces to the crown, and asks the question, "... what were the financial obligations of the provinces to the crown?" A discussion of the most important previous studies of bala follows, giving due attention to the works of Hallo (1960) and Steinkeller (1987), and to a lesser extent those of Maeda (1994) and Sallaberger (1999). Sharlach presents what she finds to be the most important aspects of these works (p. 20). Hallo's view is described as one that reveals bala to represent a system of forced contributions to the cultic capital; Steinkeller's bala is the hallmark of a redistributive system; and Sallaberger's bala system is one of entitlements. Sharlach then suggests that the most obvious translation of bala is "tax," and to back up this definition she quotes Webster's dictionary (p. 20). In the end she argues that the bala incorporated aspects of all four systems, but that the word "tax" best covers all aspects (p. 21).

In the following chapter, the author sets out to describe the bala of the Umma province, but first cursorily describes the province itself. One of the key points of the present book is that the bala contributions of both Umma and Lagash consisted of much more than just animals, and that their cereal contributions can be estimated to have represented a significant fraction of the total production of the provinces. The author's assessment of the total production of Umma domain land (first adequately discussed in Maekawa 1987) is compared to a barley notation in the "debits" section of the text (Gomi-Sato) SNAT 536 (= BM 106178; see p. 31). The barley notation in that text is then used to estimate an approximate percentage of the total Umma domain barley production as bala contribution or tax. (The actual percentage that Sharlach claims from this text [44%] is not given before p. 69.) It must be noted, however, that SNAT 536 does not mention bala, and that it in no way conforms to the format we would expect of a true administrative document. Rather, it has all the features of a bookkeeping school text of the sort described by Englund (2004: 39, n. 22). The text has no colophon, standard terminology is missing, and the text calculates neither a "deficit" nor a "surplus." Conveniently, the "tax" owed by Umma (expressed as a percentage), is according to the analysis of this text almost identical to one later established for Girsu (see p. 69). Note however, that the author uses a yield ratio that is almost certainly incorrect when she calculates Umma barley production (30-34 or perhaps even 40 gur per [bur.sub.3]). Further, Sharlach establishes the size of the domain land of Umma on the basis of an unpublished article by Steinkeller (n. 26), suggesting that the area under cultivation was almost twice the size given in any published sources, most importantly those used in the study of Maekawa (1987). It is also important to note that the key text for Umma barley production, AAICAB 1, pl. 50, Ashm. 1912-1143, is not used by the author. The other key text, MCS 6, 83 (BM 105334), is used only indirectly by citation of the article of Maekawa (1987), the results of which are basically rejected in this book.

Even though SNAT 536 is evidently an artificial school text, it does preserve some features of an account. (See Sharlach's comments to the position of the technical term summarizing the "debits" section of an account in n. 32; cf. my comments, below to p. 116, on the structure of an Ur III account.) It seems ill considered when the author writes that SNAT 536 "lists a total of 10,200 gur of cereal products (8700 gur of barley, 1200 gur of coarse flour, 300 gur of flour)" (p. 33). The text does have a total of 8700 gur of barley in its "debits" section, and some notations of flour with their work-day equivalents in the following section. But to handle these numbers together makes no more sense than adding the raw products entering a factory to the finished products leaving it and then claiming that the total represents the full production of that installation.

When discussing the bala of Umma, the author is inevitably confronted with the hundreds of texts containing the subscript [sa.sub.3] bala-a, "within the bala." These texts have been studied by Goetze (1948), Hallo (1960), and most recently Maeda (1995), all three of whom concluded that this subscript essentially refers to the time of the year when the delivery was made, rather than the ultimate purpose or final destination of the delivery. Sharlach disagrees with this conclusion, writing (pp. 41-42) "But the weight of the evidence provided by the 800 or so [sag.sub.4] bala-a tablets that I have studied does not support this assertion. It is clear that [sag.sub.4] bala-a does not refer primarily to a period of time; rather it means 'within the bala-(payments)'--that is, the phrase refers...