I didn’t, and now I’m doomed to live with the knowledge that I didn’t say three simple words that would have saved her life. I’m doomed to know that I didn’t take her to court in an attempt to have her committed, that I didn’t try hard enough to sneak medicine into her food, that I didn’t handcuff her to the bed whenever her carving got out of control. There are literally dozens of actions that could have prevented her from dy-ng, all things that I did not do.

Marianne Engel believed that she had killed me seven hundred years ago, in an act of kindness, but that story was fiction. The reality is that I killed her in this lifetime: not with kindness, but through inaction. While she believed that she was freeing herself from the shackles of her penitent hearts, I knew better. I am not schizophrenic. And still, I remained quiet. Ineffective. Murderous.

I face this fact for a few moments each day, but that’s all I can stand. Sometimes I even try to write it down before it slips away, but usually my hand begins to shake before I can get the words out. It never takes me long to start lying again, trying to convince myself that Marianne Engel’s imaginary past was legitimate simply because she believed it so deeply. Everyone’s past, I try to rationalize, is nothing more than the collection of memories they choose to remember. But in my heart, I know this is just a defense mechanism that I manufacture simply so I can go on living with myself.

All I had to say was “Marianne, come back.”

The word paleography comes from the Greek palais (old) and graphia (script), so it is not surprising that paleographers study ancient writings. They classify manuscripts by examining the lettering (size, slant, pen movement) as well as the writing materials (papyrus or parchment, scroll or codex, type of ink). Good paleographers can determine the number of writers who worked on a manuscript, can assess their skill, and often can even assign the manuscript to a specific region. With religious writings, they can sometimes identify not only a specific scriptorium but even a particular scribe.

Not long ago I engaged the services of two of the world’s foremost paleographers: one an expert in medieval German documents and the other an expert in medieval Italian documents. I hired them to look at the items that I had found, in addition to the cash, in the safety deposit box.

Two copies of Inferno, both handwritten but by different hands: the first in Italian and the other in German. Both appeared, to my untrained eye, to be hundreds of years old.

Before I would tell either paleographer what I wanted examined, I made them sign strict nondisclosure agreements. Both men found my request unusual, almost humorous, but consented. Professional curiosity, one supposes. But when I presented the manuscripts, both men realized in an instant that they’d been handed something exceptional. The Italian blurted an excited profanity, while the German’s mouth twitched at the corners. I assumed a pose of complete ignorance regarding the origins of the books, saying nothing about where I’d acquired them.

Because Inferno was immediately popular with readers, it is one of the most common works to survive in copies from the fourteenth century. The Italian paleographer had little doubt that my copy was among the very earliest, perhaps made within a decade of the first publication. He begged that I allow him to confirm his findings with other experts, but I declined his request.

The German was not as quick to assign an age to the translation, partially because his initial examinations provided some bewildering contradictions. First, he wondered how a manuscript so remarkably well-preserved had gone unnoticed for so long. Second, it appeared that a single hand had penned the entire work, which was highly unusual for such a long document. Third, whoever had produced the book was exceptionally skilled. Not only was the script beautifully formed, but the translation itself was better than most, if not all, modern ones. But it was the fourth point that was most puzzling: the physical attributes of the manuscript-parchment, ink, lettering-suggested that it had been produced in the Rhine area of Germany, perhaps as early as the first half of the 1300s. If this was true-though it hardly could be-then my manuscript predated any known German translation of Inferno by several centuries. “So you see, I simply must be mistaken.” He trembled. “I must be! Unless…unless…”

The German requested permission to perform radiocarbon dating on both the parchment and the ink. When I granted it, he had such a look of orgasmic joy on his face that I was afraid he might pass out. “Danke, danke schön, ich danke Ihnen vielmals!”

When the tests were completed and the parchment was dated to 1335, plus or minus twenty years, the German’s mood stepped a notch higher. “This is a discovery that is so far beyond anything that I…that I…” He couldn’t even find the words for his flabbergasted delight; the translation had been made within decades of Dante’s original Italian. I decided that it would not hurt to allow further research, and I even gave the German a push in a certain direction: I suggested that he might want to focus his investigation on the scriptorium at Engelthal. The German’s mouth twitched again, and he went back to his work.

When he contacted me some weeks later, he seemed to have finally accepted that he was investigating an impossible document. Yes, he confirmed, the work gave many indications of having been done at Engelthal. And yes, the copying was highly indicative of a particular scribe whose work was well known in the years circa 1310 to 1325. In fact, this scribe had always posed a minor mystery to scholars of German mysticism: her literary fingerprints were on a huge number of documents, her talent exceeding that of any of her peers, and yet her name could not be found anywhere. Such a secret could only have been kept by a coordinated effort between the prioress and the armarius of the time but, as Engelthal was otherwise proud of its literary reputation, the great question was: what was it about this particular nun that required such secrecy?

The German’s mustache was positively dancing as he spoke of all this but, he admitted, some points contradicted the Engelthal hypothesis. The parchment was of a different quality than that found in the monastery’s other documents, and the inks seemed to be of a different chemical composition. So while the workmanship suggested that it came from Engelthal, the German explained, the physical materials did not. And-need he even add this?-Engelthal would almost certainly have had nothing to do with Dante’s great poem. “It was not their particular milieu, if you understand what I mean. Not only was it in Italian, but entirely blasphemous for its time.”

The German asked, somewhat sheepishly, whether I had any more “hints” for him. As it turned out, I did. I suggested that he might now want to divert his attention from Engelthal to the city of Mainz, paying attention to privately produced books from the mid-1320s. The scribe, I said, might have written under the name of Marianne. The German’s bushy eyebrows furrowed under the weight of this new information and he begged to know how I could offer such specific suggestions. I said it was just a hunch.

He spent the better part of a month seeking out manuscripts that matched my parameters. He called often, sometimes to update me on progress but usually to complain that the confidentiality agreement was holding him back. “Do you have any idea how difficult it is to request such documents when I can’t explain why I need them? Do you think I can just go to the library and check out books from the fourteenth century?”