I first examined the Aramaic papyri in the Brooklyn Museum in the fall of 1969. One of the documents that most fascinated me turned out to be the “Document of Wifehood” (perhaps more colloquially, the marriage contract) by which Anani, a Jewish temple official, took Tamet, an Egyptian slave, as his wife. She was given to him by her slavemaster Meshullam, with whom Anani made the contract. As I was turning the document this way and that, from front (recto) to back (verso) , conservator Kenneth Linsner, who was with me, pointed to some holes at the bottom of the back of the document.

“What’s this writing?” he said.

“What writing?” I looked more closely. There it was. On either side of one of the holes were a few Aramaic letters. “No one has ever noticed any writing there [at the bottom of the back of the document].” By “no one,” I was of course referring to Emil G. Kraeling because he was the only scholar who had seen these documents. It was he who had published them along with excellent photographs. But the published photograph of this document did not show the bottom of the back (verso), presumably to save space, only the top of the back, showing a single line of text at the top.

Unfortunately, most of the line of writing at the bottom of the back did not survive, as indicated by the good size hole in the middle of the line. I could see one word at the beginning (spr), which means document (in other contexts, book) and the beginning of another word. On the other side of the hole were two other letters: mt. I immediately recognized these as the last two letters in Tamet’s name. (Remember that Aramaic as well as Hebrew is written without vowels.)

Here was the endorsement of the document—that is, the title that identified it when it was rolled and folded up. Important legal documents were stored by first rolling them up and pressing them flat. On the bottom of the back, which remained exposed by this rolling and flattening process, the scribe would write the “endorsement” that identified the document. The document was then reduced to a third of itself by folding in the two ends of the flattened document, still leaving part of the endorsement exposed. In this condition, the document was wrapped with string and a daub of clay placed over the knot to secure it. Into the soft clay, the seal of the owner or scribe was impressed.

Endorsements began with the word spr—document—and followed a fixed formula, including a space after the word spr for a seal. The first letter of the second word in this endorsement was also preserved: It was the first letter of the word for wifehood. Then came the hole. After the hole were the last two letters in the name of Tamet. Based on these clues, as well as the size of the hole and the contents of the document on the other side, I was able to reconstruct the endorsement with considerable confidence: “Document of wi[fehood which Anani wrote for Ta]met.”

Kraeling knew as well as I did that documents like this have endorsements. But he published an entirely different endorsement to this document. According to Kraeling, the endorsement of this document read: “Tamet brought in to Anani in her hand silver, 1 karsh, 5 shekels.” Kraeling of course recognized that this language was not the usual form of an endorsement. He explained this allegedly stark deviation from the usual form of an endorsement in this way: “The ‘endorsement’ appearing on the outside of the roll is usually a memorandum enabling the owner of a number of sealed rolls to know what each is about. Sometimes, however, it has a supplementary recording function, as here.”

This is an entirely ad hoc explanation. No other example of such a “supplementary recording” has ever been found.

What Kraeling thought was the endorsement was the single line at the top of the back of the document. Kraeling was a marvelous editor of the papyri, and we may easily forgive him for this lapse. None of the other documents in the archive had any text on the back of the document, except the endorsement. So when Kraeling saw the single line of text at the top of the back of the document, he assumed it was the endorsement. What he failed to notice—or perhaps, to be more accurate, failed to appreciate—was that the endorsements on the other documents were always at the bottom of the back of the document, never at the top. Had he appreciated this fact, he would surely have looked at the bottom of the back and noticed the letters that Linsner called to my attention as I turned the document over.

What, then, was the line at the top of the back? It was simply the end of the text that ran over onto the back. It is clear from the erasures and changes in the main text, as well as the different inks used (determined by microscopic examination and confirmed by infrared photographs) that this document was heavily negotiated. Among the issues was the amount of Tamet’s dowry. Obviously, there had been very considerable haggling over it, meagre though it was. Finally, at the end of the negotiation, Meshullam apparently gave in and agreed to pay Anani an additional karsh and five shekels—in cash! This is attested by the runover line at the top of the back of the document, an addition to Tamet’s dowry that came in the process of the haggling that continued down through the final stages of the redaction of the document.

Unfortunately, no one could catch Kraeling’s error from the publication since the published photographs show only the alleged endorsement and not the rest of the verso (back), which was supposedly blank.