“Well, again, yes and no. Not really the genetic code in our genes proper, but this code in a tiny part of the other 95 percent of our DNA on the X chromosome,” Jenna said.
“I see,” I said, though I didn’t yet, at least not fully. “And Glen’s death?”
“He phoned to tell me he had completed the final translation of the code, had words up on his screen… and when I came over, he was dead,” she started sobbing. “I think those words killed him.”
“OK,” I poured more soda in her cup. So now I knew something: either she had killed her boyfriend, and cooked up this story to throw us off track, or there was something genuinely strange going on here. The coroner’s extensive autopsy had already found no demonstrable cause for the sudden massive failure of all of Chaleff’s systems that had killed him—“looks like everything just blew at once for no apparent reason,” Dave told me—so we knew Chaleff hadn’t just died of your common heart attack or stroke while he was doing his research. It was something more. Like something had reached in and turned off—or on—some master switch, as Dave had said yesterday. The question was who—and what. And the what was not only what reached in, but what was the switch?
I could see why the lieutenant was thinking homicide. In cases like this— cases involving dead young bodies— the cause of death was all too often murder. Barring tragedies like AIDS, young bodies don’t very often expire on their own.
Now it might shock the public to hear this, but in many ways murder is the forensic scientist’s best friend. Once the cops get a confession of murder, however inarticulate, it points to the facts, and we can use it, working backwards, to piece together a detailed description of the death and its circumstances. Reverse engineering is always easier than working from the ground up.
But truthfully, I hated confessed murder as the cause of death, always resisted it as the explanation until impossible to do so. Not only for the obvious moral reasons—I’m as glad as the next guy to find a bit less depravity in the human species wherever I can—but because, well, I savor the thrill of an investigation in which I don’t know the final conclusion beforehand, in which science leads to the cause of death rather than vice versa.
And I’d learned the hard way that some kind of nefarious intervention, something worse than mere murder, always loomed as a possibility when research scientists were involved. I’m not talking about dressing up a lover’s quarrel or cutthroat professional competition with a fatal malfunction in a laboratory like they do in the movies. I’ve had experience with things much worse. The public had no idea.
But what was the agent of death here?
Words on a screen?
They made sense neither as a weapon nor a lethally malfunctioning piece of equipment.
“You have any idea what those words were?” I asked.
She looked up at me and her eyes refocused, as if my voice had pulled her away from some contemplation deep and distant. “No,” she said. “The screen was blank when I arrived.”
So now I knew she was probably lying about at least one thing.
Some of her facts were easy to verify. Jenna had been telling the truth about Chaleff’s last call to her apartment. And there was no sign of anyone entering Chaleff’s apartment between the time of that call, and the time Jenna arrived, about forty-five minutes later, when she said she’d found Chaleff dead.
Her story about the special section of the human genome project took more work to confirm. She’d told me the MIT Media Lab had a piece of the research action. Nic Negroponte, head of the lab, was an old friend of mine. He didn’t know much about that part of the project, but put me in touch with an associate who did.
“Ralph Hertzberg here,” the voice on the phone said. “Nic told me to expect your call.”
“Great,” I said. “Let me start by trying to explain to you what I think you’re working on—what I understand and what I don’t—and you tell me where I’m wrong.”
“Shoot.”
“OK,” I said. “DNA is commonly said to be a genetic language, but that’s not quite right. It’s really a recipe for the construction of other proteins into cells that have specific properties—heart cells, brain cells, and so forth in humans—cells and organs and systems that come into being during gestation.”
“That’s right,” Hertzberg said.
“OK,” I said. “So in fact, DNA isn’t really a language at all—it’s really an arrangement of proteins that causes other proteins to develop in a certain way, into heart cells, etc. So really DNA is a catalyst for the development of living organisms. But we say as a shorthand that it’s a code or a set of instructions. Am I on the right track here?”
“Very much so.”
“Good," I said. "All right, then. So tell me this: How do we get from DNA, which isn’t really a language— or is only a language in a metaphoric sense—how do we get from that to this chromosomal material which Jenna Katen says Glen Chaleff was able to read on his screen?”
Hertzberg sighed. “Not very easily, but I’ll try to explain. First, you have to understand that there’s lots of protein material associated with chromosomes that we have no idea what the function is. Not everything there is just genes. In fact, most isn’t. Some material we’ve identified as seeming to have a catalyst function for the genes themselves—sort of meta-catalysts—some seem to control timing of genetic instruction of other proteins in ways we’re just beginning to fathom. But most of this extra genetic material is still a mystery to us.”
Right, the so-called junk DNA, I thought. “And the, uh, the linguistic material on the 8 percent of X chromosomes is, was, in the mystery area?”
“Yes.”
“Has this material been found only on human chromosomes?”
“So far, yes,” Hertzberg said. “Primate chromosomes were the first other place we looked—chimp and ape DNA is 99 percent the same as human—and we found nothing like it.” “Nothing that could generate words on a screen?”
“Look, let me be honest with you,” Hertzberg said. “I know what Jenna told you, but we don’t even know for sure that this binary chromosomal material can be converted into readable words. It seems transformable into a binary code, yes, but we have no way of really testing the accuracy of that transformation, since we have nothing precisely of this kind to measure it against. And we certainly don’t know for sure if that code can support actual words. What we get from that code at first is some sort of general protolanguage, strongly resembling Indo-European in its subject-predicate structure, and therefore recognizable as a real language to some researchers, I guess. And assuming that to be Indo-European, or proto-Indo-European, we can make rough translations into English, Sanskrit, what have you. But the results are extraordinarily speculative to say the least—I’d say the noise to signal ratio must be well over 40 percent in the final translation. Though that’s conjecture too—the actual distortion could be far more, or less, for that matter. Bottom line: We’re dealing with a hell of a lot of conjecture here. That’s why we haven’t published anything about this yet. It’s still in the very early stages of research. Most of our work is.”
“All right,” I said. “Let’s back up a little—I’m very much a layman when it comes to linguistics. What made you think in the first place that the binary transformation of chromosomal code yielded patterns that looked like Indo-European?”
“Ah,” Hertzberg said. “That was the relatively easy part. We already have ASCII table renderings of most known human languages, including many long extinct. ASCII and binary configurations are readily transformable into one another. So when we converted the binary chromosomal code to ASCn, its similarity to primal Indo-European in ASCII was noticed right away.”