Выбрать главу

The dice remain scrupulously fair.

Po-kwai begins the third phase, another measurement of correlations within her brain. I can understand Lui’s impatience with these inward-looking experiments—but at the same time, I can appreciate, more than ever, ASR’s reasons for proceeding cautiously. I may know for a fact that all kinds of macroscopic feats are possible, but I’m thrashing around in the dark trying to master them, and taking huge risks in the process. Left to themselves, ASR might take ten years before they try anything similar—but when they do, they’ll be in complete control; they’ll know precisely what they’re doing.

I think: maybe they’re the best people to explore the true Ensemble’s mysteries, after all. Slowly, methodically, rigorously, respectfully…

Po-kwai is successful on the second day; she seems pleased, but not surprised, by this. She’s clearly gaining confidence in her skills with the mod, despite the obscurity of the operational details. How long before this growing sense of assurance, of control, invades her dreams—and shuts me out?

I sit in the anteroom, watching the simulated dice rise and fall automatically, ten times a minute, hour after hour. I keep my real vision fixed on the dice, while holding two windows in my mind’s eye: the Hypernova menu, and an interface to an analysis program—a modified, miniature version of the ion experiment software, smuggled to me by Lui in a two-second RedNet handshake.

Smearing ON.

The dice are tossed.

Smearing OFF.

Enter results.

Primed, I could do this indefinitely, without the slightest change in mood. Deprimed, I slide from bursts of enthusiasm into grey tedium, then screaming boredom, then stretches of merciful automatism—from which I emerge more frustrated than ever. All of which may be helpfuclass="underline" whatever my differences upon smearing, it’s hard to believe that I’m not unanimous in wishing to cut short this mind-numbing procedure—and the only way to do that is by succeeding.

Or is it? I can hold my virtual selves to ransom only if, after each collapse, I remain in control—and the truth is, I have no way of knowing what the eigenstate mod will be used for: to choose the state of the dice, or to choose my own state of mind. At the next collapse, I might find that a state has been selected in which I’ve simply given up on the experiment… or given up on the true Ensemble. Every time I smear, all the rules of the game are being thrown into the air, alongside the dice. I can only hope that they’re harder to sway.

I pocket the dice generator seconds before Lee Hing-cheung arrives to relieve me. The program in my head—running much more slowly under von Neumann than it would on any decent hardware—scours the accumulated data with ever more sophisticated and obscure tests in the hope of detecting an effect, but spits out its final, unsurprising conclusion as I step off the homebound train:

[null hypothesis unchallenged.] I turn up for duty expecting to find that Po-kwai has been granted a rest day, but my orders are to report to Room 619. When I get there, Lee explains. ‘She says it doesn’t tire her any more; there’s no reason to hold up the work.’

I stand guard with single-minded vigilance, as if to compensate for my nocturnal dereliction. I blank out the chatter of the scientists, and suppress any sense of anticipation. P3 distils me into a pure observer—wired to respond in an instant to any contingency, but until that moment, utterly passive.

When Po-kwai emerges from the ion room, an hour later, they call it a day. In the elevator, heading for the restaurant, I ask, ‘How’s it going?’

‘Good. We’ve had useful data all afternoon.’

‘Already?’

She nods happily. ‘I think I’ve crossed some kind of threshold; everything’s just getting easier and easier. Well… you know what I mean. I do nothing, as always. I take no credit—but it certainly looks like the smeared Po-kwai has finally mastered Ensemble.’

For a moment, I’m tempted to ask her to repeat what she said, but there’s no need; I heard her perfectly, and the meaning is unambiguous. And if she’s never named the mod before, no doubt she was explicitly instructed not to—by Leung, perhaps—with sufficient emphasis for the message to sink in more fully than all the other ‘security bullshit’.

I see no reason to admonish her for the slip.

I sit through dinner with infinite patience, nodding politely while Po-kwai complains about how boring the food has become.

I sit in the anteroom, listening to her moving about the apartment, wondering what difference, if any, this information will make.

At one a.m. I deprime, and my joy is no longer constrained. The true Ensemble is the mod named Ensemble—and this perfect equation, this electrifying symmetry, is the final confirmation of everything I believe. A revelation, yes—but in retrospect it seems impossible that it could have been otherwise. And what greater inspiration could I hope for, to guide and encourage the virtual selves who remain loyal to my mission?

I take out the dice generator, invoke the mods, begin.

The dice fall at random, again and again, but I’m not discouraged. My smeared self can’t be expected to perform instant miracles, however fervently he’s pursuing the task… least of all when I annihilate him by collapsing, every six seconds, and he has to begin again, picking up the threads from whatever hologrammic traces of his experience are preserved in my brain.

Must I collapse so often—after every throw? It’s true that Po-kwai succeeded with this approach—and collapsing after each ion would have given her the simplest possible goaclass="underline" amplifying one of just two possibilities. Her task and mine aren’t identical, though; Ensemble is in her skull, not mine. Maybe I need to smear for a longer time, to generate versions of myself capable of influencing the mod. How long was I smeared when Karen appeared, unbidden? I have no way of knowing; the process was out of my control.

Now, that’s no longer true.

I tick the ON switch.

On the table beside me, the dice generator sends the images of the cubes spinning into the air. They look almost solid—even glinting convincingly as they pretend to catch the ambient light—and they fall to the surface with a faint simulated click.

Snake’s eyes, two ones—my target.

I twitchily suppress the by now instinctive third step of the routine, and, leaving the Hypernova menu untouched, enter this first result into the analysis program—thinking: each time I do this, von Neumann will smear into multiple versions, with copies of the program which have been fed every possible combination of results so far. I don’t have to think about individual throws; all I have to do is choose an eigenstate in which the analysis program eventually declares success. Surely I can manage a task as simple as that—with the help of the true Ensemble.

Snake’s eyes for a second time. And a third.

What if I collapsed right now, before the program gives a verdict? What will this have been—a fluke? A coincidence? A rare—but insignificant—run of good luck? Or am I already witnessing the proof that I will remain smeared beyond that point?

Snake’s eyes, for a fourth time. At one chance in thirty-six each toss, the probability of a run of four or more—just once in all the thirty thousand tosses, the ten nights’ worth of data that I have so far—is already down to 1.7 per cent.

A fifth time… at 0.048 per cent. Having crossed its arbitrary one per cent threshold, the program starts flashing messages of triumph.