Writing immediately after the fourth expedition to Nippur, Hermann Hilprecht, in characteristically combative tone, recalled the comments of Sir Austen Henry Layard, who had carried out less than two weeks of trial excavations at the site in 1851:
“On the whole I am much inclined to question whether extensive excavations carried on at Niffer [Nippur] would produce any very important or interesting results,” was the verdict of the great explorer [Layard] at the middle of the last century. “More than sixty thousand cuneiform tablets so far rescued from the archives of Nippur, temple library definitely located, and a large pre-Sargonic gate discovered below the desert,” was another message which fifty years later the present writer [Hilprecht] could dispatch to the committee of the Philadelphia expedition from the same mounds of Nuffar.
The success of the Philadelphia expeditions’ excavations at Nippur allows us to overlook Hilprecht’s crowing. Focusing on the tablets alone, the four expeditions discovered several collections of tablets which to this day are foundational for the study of Mesopotamian civilization.
The figures reported by Hilprecht later in the same work are impressive: the first expedition (1888-89) located a mere 2,000 texts; the second (1889-90) about 8,000; the third (1893-96) 21,000; and the fourth (1898-1900) 23,000. Though the alert reader will notice that there is a discrepancy of at least 6,000 texts between the two figures given by Hilprecht in the same volume, we must remember that counting tablets on this scale is not an exact science. Both the identification of joins among excavated fragments and the question of whether one is counting only the larger pieces or every single inscribed crumb make the number of tablets in a collection a somewhat fluid quantity, even without allowing for the natural exaggerations which may result from the excitement of the moment.
Even now, it is not possible to put an exact number on the final count of inscribed objects from the first four expeditions to Nippur. The tablets ended up in three main groups, in Philadelphia, Istanbul, and Jena, with a few texts in Berkeley, CA, and isolated fragments in the Louvre and the British Museum. Thus, in the electronic catalog of the Penn Museum's Babylonian Section there are 19,403 items from Nippur at the time of writing, but there are certainly unregistered joins—and even still-unidentified joins—which will in due course lower that figure by several hundred. In a similar catalog of the Jena collection, 3,000 pieces are recorded. The balance of the finds is in Istanbul, for which no catalog is available. According to scholars who were able to peruse the Istanbul collections in the mid-twentieth century, the total number of Nippur tablets in Istanbul is in the order of 25,000, which would fit with Hilprecht’s lower total of about 54,000.
Several factors have conspired over the ages to rob the Nippur tablets of their archaeological context. The first was ancient building activities at the site, which removed entire archives from their primary context and relocated them elsewhere. For example, one group of tablets recording precious commodities for the rebuilding of a temple about 2300 BCE was used as fill to level a road constructed around 1300 BCE. More havoc was wrought during the second century BCE by the Parthians, who carried out extensive excavations for the construction of large brick buildings, resulting in substantial disruptions of the primary contexts. A second factor is the absence of archaeological context for the early Nippur finds: while many tablets remained in their original contexts, the nineteenth-century approach to excavation involved minimal recording of findspots. A third factor is perhaps more frustrating: although the tablet finds were recorded in the day-books of the excavations, and the tablets were sent to Philadelphia with packing slips indicating the days on which the tablets were found, Hilprecht discarded these slips in the frenzy of unpacking.
Hilprecht, at least, was certain of one of the ancient contexts: his postulated “Temple Library”—the name he gave to a mound that yielded thousands of learning exercises, word lists, and literary texts. We now know that these texts almost certainly came from private houses in which young scribes were trained for administrative duties. All the same, Hilprecht's general sketch of this important subgroup of the Nippur finds is on the whole useful and accurate, as are his sketches of a number of other text groups. The learning exercises date to the end of the eighteenth century BCE and allow us to reconstruct the curriculum. At the pinnacle of their education, some of the trainee scribes wrote literary texts in the Sumerian language. These tablets form the basis for our entire modern understanding of Sumerian literature, a corpus of about six hundred compositions consisting of myths, heroic narratives (including the Sumerian versions of Gilgamesh tales), hymns to kings and gods, literary letters between kings or directed to gods, and various kinds of scribal compositions ranging from proverbs to dialogues, debates, and short tales.
We owe the emergence of this literature from obscurity to the pioneering work of two of Hilprecht’s later successors to the Clark Chair at Penn, Samuel Noah Kramer and Ake Waldemar Sjöberg. Kramer traveled extensively to Istanbul and Jena and scoured the drawers of the Penn collection, identifying fragments that join within and between the collections. He organizing the tablets and fragments according to the types of literary compositions they recorded. He reconstructed and translated many of these compositions for the first time, publishing them fearlessly and prolifically. Sjöberg took Kramer’s work one step further, creating the Pennsylvania Sumerian Dictionary Project, which continues at Penn to the present day and is creating the first exhaustive dictionary of the Sumerian language. Indeed, the presence at Penn of the bulk of known Sumerian literary tablets was one of the principal reasons for founding the Dictionary in the Babylonian Section.
Other important groups of texts from Nippur are administrative in nature. Besides the archive of temple equipment mentioned above, the Kassite texts and the Murashu archives are most worth mentioning. The former consists of various archives from the latter half of the second millennium BC, and are particularly associated with institutional organizations in Nippur, including the estate of the governor. Some of these records are tiny tablets covering a few allotments of rations, while others are huge ledgers consolidating smaller records, and still others, several hundred in number, are letters between individuals giving insights into the daily life of the city.
The Murashu archives, by contrast, are the private records of a family of businessmen dating to the fifth century BCE—most of the dated tablets come from the reigns of the Achaemenid kings, Artaxerxes I and Darius II. The range of activities recorded covers the leasing and renting of agricultural land, tax collection, short-term loans secured by fields and orchards, work contracts and lawsuits. Although most of the activity is carried out close to Nippur, members of the family traveled to Babylon and even to Susa (in modern Iran) in connection with their business. Written at a time when the local language, Akkadian, was probably already no longer spoken, the Murashu documents include a significant number of notes in Aramaic, added to the tablets in ink or in a very few cases scratched into the clay with a stylus. The broad connections of Nippur in this period are also indicated by the personal names in the texts, which include both Iranian and, most famously, Hebrew names, the latter principally identified by their containing the divine name Yahweh.