Steve Tinney gingerly holds a small clay tablet on which a simple wedge mark is repeated many times. “This we call ‘second day at school,’ ” he jokes.
The cuneiform tablet, circa 1750 B.C., is an ancient copybook, demonstrating a student’s early efforts to master the world’s first written language. Tinney, associate curator of the Babylonian section of the University of Pennsylvania Museum of Archaeology and Anthropology, has selected the Sumerian school exercise from among the treasures displayed on what he calls “the dessert cart.”
Even after 13 years at Penn, the 42-year-old British-born Tinney still enjoys the show-and-tell. “This is a beautiful tablet,” he says of another artifact with unusually clear markings. “It’s a bunch of medical prescriptions, one of our very famous texts.” The complexity of the signs ties it to an older period, about 2100 to 2200 B.C. Cuneiform, which refers to “wedge-shaped” writing, began as pictographs, with pictures representing words. Over time, the signs evolved into more abstract and stylized forms, even as the writing system itself grew more flexible and complicated.
Outside Tinney’s office is the Tablet Room: file drawers filled with some 30,000 neatly catalogued cuneiform tablets, and fragments of tablets, many of them painstakingly joined together like pieces of a jigsaw puzzle. Thanks to late-19th-century excavations at Nippur, in present-day Iraq, Penn has the largest existing trove of Sumerian literary texts. The most celebrated example is the Epic of Gilgamesh, the tale of a king’s search for immortality that includes a precursor of the biblical flood story.
Drawing on its voluminous archive of cuneiform writing, the museum is nearing the culmination of a three-decade-long intellectual quest: the creation of the first modern Sumerian dictionary. So modern, in fact, that it is debuting as an Internet-based tool with links to new libraries of digitized, transliterated Sumerian texts. “This is not just dictionary writing,” says Tinney, the project’s director. “It’s also software engineering.” Still to come: a CD-ROM, print spin-offs, and even computer-assisted translation of Sumerian text.
Launched in 1976 with a National Endowment for the Humanities grant, the Pennsylvania Sumerian Dictionary Project modeled itself on a similar scholarly endeavor: the Chicago Assyrian Dictionary (CAD). This work-in-progress commenced in 1921 at the Oriental Institute of the University of Chicago. “We’re the old ladies of the process,” says the CAD’s director, Martha T. Roth. Under Sumerologist Åke Sjöberg, the Penn project proceeded at a comparable scholarly pace, publishing an exhaustively researched volume for the letter B in 1984, and following it with three volumes for A.
In 2002, Tinney, as fluent in the computer-based XML language as he is in Sumerian, took over as director. In the relatively short time after his arrival, he and his team have produced the cyberspace dictionary. “I realized that if we started pushing and pushed hard for a couple of years, we’d actually do this,” he says. “And I’ve been working maniacally ever since.”
This summer marked a watershed. For the first time, a preliminary version of the dictionary became available online. To find, for example, the word for “goat,” click on http://psd.museum.upenn.edu and open the Dictionary Search page. As much a cultural encyclopedia as a dictionary, the site will do more than supply translations. It will tell you, Tinney says, “the different colors of goat…, the different things that are done with goat, and…the cuts of goat meat—huge amounts of information.”
If the Penn project had stuck to print, “I’d be dead by the time it’s finished,” says 61-year-old Jerry Cooper, W. W. Spence professor of Semitic Languages at Johns Hopkins University. He notes that the Internet dictionary will make available a vast amount of material previously unpublished or found only in scattered sources.
By opening the material to scholars in emerging countries and to specialists in other fields, “the dictionary will revolutionize the way the field works,” says Piotr Michalowski, professor of Near Eastern Studies at the University of Michigan in Ann Arbor. “Because it is electronic and will be continuously updated, it will be a living, eternally morphing resource that will never go stale. It will also make the teaching of Sumerian much easier.”
Going digital is “a brave and bold move” that may serve as a model for other disciplines, Michalowski adds. “In this case, the field is actually showing others how to proceed.”
In their day, more than 5,000 years ago, the Sumerians, too, showed the way, inventing what is commonly believed to be the world’s first writing system (preceding and perhaps inspiring Egyptian hieroglyphics). “What I’ve learned is [that] the worst mistake we can make is to think that they’re primitive people,” says Erle Leichty, emeritus Clark Professor of Assyriology at Penn. “They are highly sophisticated from the first time they write.”
Between the Tigris and Euphrates Rivers in ancient Mesopotamia, the Sumerians founded rivalrous cities, with names like Ur, Kish, and Uruk. They developed mathematics and astronomy, and probably began writing to record commercial transactions. Tinney says the first surviving texts, from Uruk, date from as early as 3500 B.C. Sumerian was replaced by Akkadian as a spoken language around 2000 B.C., but the written language remained integral to an elite scribal education for centuries afterwards. It became a class marker, much as Latin later was for Europeans. Cuneiform was eventually used to write some 15 languages, according to C.B.F. Walker’s Reading the Past: Cuneiform.
Leichty’s eyes light up when he talks about the origins of the field once called Assyriology. (Both Assyrian and Babylonian are now regarded as dialects of Akkadian.) “All of this was completely dead, there was no Rosetta Stone [the slab used to decipher Egyptian hieroglyphics] and these brilliant guys at the end of the nineteenth century deciphered it,” says Leichty.
The key to translating Sumerian was the discovery at Behistun, in present-day western Iran, of cuneiform royal inscriptions in Old Persian, Akkadian, and Elamite. Two competing amateur scholars saw that Old Persian was alphabetic, and they used ancient Greek sources on Persian kings to infer the proper names in the inscription. When they defined the letters for “king,” says Leichty, “that gave them enough” to crack the code—“and then [in the 1840s] they had their Rosetta Stone.”
It took the scholars about 15 years to read the Akkadian, Leichty says. “Their problem was they were trying to read it either logographically [with signs representing words] or syllabically, but not a combination of the two. Then an Irish priest named Hincks got the bright idea that it was combined…. And, of course, once they got the royal inscriptions, they were able to get into old dictionaries.” The many bilingual Akkadian-Sumerian texts in turn opened the door to Sumerian.
Except that, as Tinney is fond of pointing out, many of the Akkadian translations—recorded as Sumerian was dying out—turned out to be wrong. “Our position is what is determinative of the meaning is context, and not the Akkadian translation,” Tinney says.
And that’s where the university museum’s bounty of literary texts, from about 1750 B.C., proved so useful. These days, Tinney notes, projects such as the Cuneiform Digital Library Initiative at UCLA and a similar effort at Oxford University are making it less imperative to handle original tablets.
Leichty first proposed the idea of a Sumerian dictionary in the early 1970s to his colleague, Sjöberg, at a party. A veteran of the CAD, Leichty believed that Penn needed a similar prestige-enhancing project. It turns out that he was preaching to the choir. For years, Sjöberg had been making cards of Sumerian words, complete with text references. Today, his thousands of cards—with not a single typo or white-out, Leichty says, because the perfectionistic Sjöberg discarded any mistakes—still fill file drawers in the library next door to the Tablet Room. It is noteworthy that there are no English translations; Sjöberg, now retired and living in Sweden, didn’t need them.
Since 1976, the NEH has funded the Pennsylvania Sumerian Dictionary Project at regular two- and three-year intervals, including a $302,000 grant for 2004-06. A current museum capital campaign aims at raising a $3 million endowment to fund dictionary research in perpetuity.
It takes a certain obsessiveness to be a Sumerologist or a lexicographer, let alone a Sumerian lexicographer. “It swallows your soul,” Tinney says.
The difficulties of Sumerian are overwhelming, even for experts. The language is what’s known as an isolate, with no known relatives (though the words “Tigris” and “Euphrates” and possibly “Eden” and “abyss” have Sumerian antecedents). There are 600 commonly used signs, which include words, syllables, and grammatical markers, and another 1,400 or so rarer signs. Many have multiple meanings; for example, “ka” means “mouth,” “voice,” and “to speak,” and is also used to represent other words starting with the “ka” sound. And both signs and meanings changed over time.
“Frustration’s a daily fact,” says Tinney. “You just learn to live with it. You have to make frustration your friend.”
As a language, he says, “Sumerian is wonderfully flexible. It can do anything…. What the years of experience teach you [is] you get used to the patterns. There are some really pathological cases. The ‘ne’ sign is really bad: It can be a noun or a verb, it’s the word for fire, it’s the word for burn, or it can be a verb prefix, or it can be a verb suffix, or it can be a noun suffix.”
Dictionary-makers face special problems, one of which is distinguishing words from sentences. Philip Jones, editor of the Sumerian dictionary, says that when he makes word lists, he finds himself “waking up in the middle of night saying, ‘Should that go with this, or should it go with that?’ ” Then there are the grammatical conundrums. “There are as many Sumerian grammars as there are Sumerian grammarians,” he says.
To complete a preliminary dictionary, Tinney, Jones, and their colleagues have had to set aside some of their perfectionism. “With this sort of lexicography, it can take weeks to figure out a single word,” Tinney says. “And one of the things that we’ve done in going to the online, breadth-first paradigm is to accept that we’re not going to solve all the problems from the get-go. What we’re trying to produce between now and 2006 is what I’m calling ‘the end of the beginning.’ It will be all the data, or almost all the data, and all the words and all the preliminary definitions. So we can then do the real painstaking work of deciphering the bits that remain really undeciphered.”
Meanwhile, Tinney expects colleagues to challenge Penn’s interpretations. Disagreement is “inevitable, because we do not understand Sumerian fully yet, that’s a simple fact. So a dictionary like this is a portrait, it’s a perspective. It’s not the truth.”
“This is an evolving, explosive field,” says Leichty, explaining why he believes the dictionary will never really be finished. Thousands of tablets remain buried in Iraq, and even with the war obstructing excavations, “there are about twenty-five thousand tablets a year coming out of the ground. The tablets are indestructible, they get scattered all over the place, and they’re all full of knowledge.
“While admittedly most of them are economic texts, receipts for sheep and things like that, it doesn’t make any difference,” Leichty says. “You can pick up a tablet like that, and it may have a brand-new word in it. Because that particular sheep might have a weird thing wrong with it—and suddenly you have a new word.”
Julia M. Klein is a cultural reporter and critic in Philadelphia. 
