Authors: Robert Kroese
“
The effect on a human being is minimal. You probably wouldn’t even feel it. But it will disrupt your central nervous system slightly for about a millisecond. Just long enough to slow down your motor response a little for a few seconds afterward. So that if you were to, say, toss a coin during that period, the result might be different than if you hadn’t been hit by the randomizer.”
I
’m looking over the device. It sure doesn’t look like much. “So Tali had one of these on her. That’s how she manipulated the coin toss.”
“
Yes,” says Heller.
Something is still bugging me.
“Isn’t this thing, the randomizer, part of the deterministic system too? How is using this device any different than trying to stop the event by calling the police?”
He smiles, as if anticipating the question.
“Very good. You’re exactly right. The magic of the randomizer isn’t in the electromagnetic pulse it sends out, but in the way it determines whether to send the pulse. You see, inside the box is a small amount of a radioactive isotope. Don’t worry, it isn’t dangerous. The isotope emits a particle about once every two seconds, and there’s a component that detects the particles as they are emitted. The device has a counter that is always set to either one or zero. If, during a given second, a particle is detected, then the device sets the counter to one. If it doesn’t detect a particle, the counter is set to zero. The randomizer will only emit a pulse if the trigger is pushed while the counter is at one. If it’s at zero, the randomizer will do nothing.”
I
’m now thoroughly confused. “So it doesn’t work half the time?”
“
The probability isn’t actually fifty/fifty, for reasons that are difficult to explain, but for our purposes, let’s say that it doesn’t emit a pulse half the time. That isn’t the same as it not working half the time. Although it’s also true that the interference doesn’t work half the time. It’s very complicated.”
Yes, it is. Heller proceeds to explain it, but it doesn
’t really penetrate. I’m going to do my best to reconstruct his explanation from my memory, later reading in his book, and other research (mostly Wikipedia). Again, you can skip this part if you don’t really give a shit how the randomizer works.
SKIP THIS PART
Essentially the randomizer works a lot like a real-life Schrödinger’s cat. In Schrödinger’s thought experiment, the cat is either alive or dead because of something that happens at the subatomic scale. Since there is a fifty percent chance of a particle being emitted over a certain time frame, the cat has a fifty percent chance of being alive and a fifty percent chance of being dead. The essence of the experiment is the transference of quantum indeterminacy from the subatomic level to a macro level. And along with this indeterminacy comes true randomness: it is impossible to predict whether a particle will be emitted, and therefore the cat’s status is non-deterministic, which is to say that it’s objectively random.
In fact
, I found at least one company online that sells devices similar to Heller’s randomizer (without the electromagnetic pulse part, of course). You can buy a device that will plug into a USB port on your computer that will generate truly random numbers based on quantum phenomena. There are other services that use atmospheric noise to generate random numbers, but although these numbers are random for all practical purposes, they are not
truly
random. Numbers produced from atmospheric noise are produced by the same deterministic system that Heller is trying to get around. Of course, unless you’re a quantum physicist doing some really wacky things (like Heller), there’s absolutely no reason you’d need true randomness. If atmospheric noise isn’t random enough for you, you’re either a genius or a nut. Possibly both.
Anyway, the point is that because of its reliance on quantum phenomena, the behavior of Heller
’s randomizer is truly random. There’s a fifty percent chance that when you push the button, the device will generate an electromagnetic pulse, and no one, not even Ananke, can predict whether a pulse will be generated. Now let’s say that a particular event will only happen if a certain coin toss comes up heads. Tali shows up at the location of the coin toss and pushes the button, thereby injecting randomness into the situation. If a pulse is emitted, the coin will come up tails. If a pulse is not emitted, the coin will come up heads. Ananke has no way of knowing which will happen.
The obvious question is:
“Why not make the pulse fire every time you push the button? Don’t you have a better chance of interfering with Ananke’s intentions if you do your best to interfere with the coin toss every time?” Paradoxically, the answer is no. This is because Ananke will attempt to anticipate any action you take to thwart her, and if you interfere with the coin toss every time, she will simply invert the required result. It will turn out that instead of heads being required to bring about the event, tails was required – and by altering the result, you have just brought about the event you were trying to prevent. If you act in a predictable, deterministic fashion, you’re on Ananke’s turf. She’ll win every time. Heller uses this illustration in his book:
Think of a boxer, Dan, who is much better at striking with his right arm than his left. Dan is facing a skilled opponent named Tim, who knows that Dan’s right is his best weapon. Tim will anticipate that Dan will try to use his right arm at every opportunity and constantly block every attack. As a result of Tim’s blocking, Dan often resorts to using his weaker arm because Tim isn’t expecting it. Of course, Tim will quickly catch on to what Dan is doing and begin to block Tim’s left as well. What you end up with is an endless feedback loop where Tim is attempting to anticipate Dan’s actions and Dan is attempting to anticipate Tim’s reactions. If Dan is skilled enough, he may occasionally land a punch despite Tim’s blocking, not because he always attacks with his stronger arm but because Tim is unable to predict what Dan will do.
The point is that, paradoxically, in order to exert one’s will against Ananke, one must cede some power to her. Only by failing to interfere half of the time do you have any chance of preventing her from realizing her intentions. I think this gets back to Heller’s Yin-Yang comments in his book. Pure volition becomes pure determinism: if you try to exert your own will against Ananke one hundred percent of the time, you will fail one hundred percent of the time. It’s a counterintuitive concept, but it makes perfect sense in a perverse sort of way.
Now Heller
’s strange statement starts to make sense as well:
It doesn’t emit a pulse half the time. That isn’t the same as it not working half the time. Although it’s also true that the interference doesn’t work half the time.
What he means is that half of the time Ananke is unable to anticipate the result of the coin toss. She guesses wrong half the time because she has no way of knowing whether a pulse will occur. And it’s important to understand that what causes this uncertainty is not the pulse itself, but the
possibility that a pulse will occur
. It’s like the boxer example: sometimes Dan strikes with his right and sometimes with his left. What allows him to occasionally get through Tim’s defenses is not his overpowering right arm; it’s the
uncertainty
of whether he will strike with his left or his right. Similarly, what allowed Tali to occasionally thwart Ananke’s intentions was the uncertainty of whether she would interfere with the coin toss or not. Heller tells me that Tali has been successful in preventing the event roughly fifty percent of the time whether or not the pulse occurred. In other words, half of the time that Tali interfered, she did so by
not interfering
. To distinguish these two sorts of interference, Heller uses the word “tamper” to describe the introduction of quantum randomness into a situation, and he uses the word “interfere” to describe physically altering the coin toss result. So you can tamper by randomly interfering or not interfering. In the cases where Tali tampered but didn’t interfere, what Tali essentially did is fool Ananke into thinking that she
would
interfere, which caused Ananke to reverse the required coin toss result. So just as Ananke could cause Tali to bring about the very event she was trying to prevent,
Tali could do the same thing to Ananke
. And that brings us back to free will, determinism, Yin-Yang, and all that. Which is the cause and which is the effect? Who is pushing whom?
And then there
’s Heller’s statement that the probability of a pulse being emitted isn’t actually fifty/fifty. He explained this too, but I’m still not sure I understand it. I think the problem is that even when the pulse is emitted, it doesn’t invert the result of the coin toss one hundred percent of the time: either the effect on the central nervous system isn’t enough to change the result, or it’s so great that it causes the number of rotations of the coin to change by an even number, so the result is the same as it would have been if there had been no interference. So the odds of a pulse occurring have to be jacked up a bit above fifty percent to even the odds of the interference altering the result. Something like that, anyway.
OK, START READING AGAIN HERE
“So,” I ask him, “you’ve injected randomness into the coin toss, but the toss itself was going to happen either way, right? That means it’s part of the deterministic system. And if that’s true, why doesn’t Ananke just stop the toss from happening in the first place?”
“
Ananke isn’t omnipotent. There are limits to what she can do. Every event is dependent on every other event, and some of those events are predictable – that is, predictable to
us
. And that gives us some power. How hard she fights to make a particular event occur is a function of how badly she wants that event to occur, and she clearly cares more about some events than others. She allows us to prevent some minor events so that she can focus on the ones she really cares about.”
He turns back to the computer and hits refresh.
“Hmm,” he says. “Hayward’s mean PDC is up to two point nine. And it’s got a crux at three twenty-six, right at the epicenter.”
The clock on the wall says two forty. That means the crux is in forty-six minutes. Hayward is a good half hour from here.
“Is two point nine high?”
“
If Tali were here, she’d want to head to the crux location, just in case.”
“
So are you going to go?”
He shakes his head.
“I don’t tamper. Not anymore.”
“
Why not? If that event happens, people are going to die. You could save them.”
“
I wish it were that simple,” he says wearily. “I used to think Tali and I were doing some good, but these days … I just don’t know. Trying to alter the future, it messes with your head. She fails half the time, you know. Tali, I mean. More than half the time, because sometimes she doesn’t get to the crux point in time. And when she does get there and successfully interferes, and the event occurs anyway – well, in a very real sense she was the
cause
of the event. It wouldn’t have happened if she hadn’t interfered.”
“
But she couldn’t know that. And overall, the effect of her interference is positive, isn’t it?”
“
I think that’s another question you can’t look too closely at. Think of it this way. As you said, although Ananke can’t know the result of the toss, she does know that a toss is going to occur – and if you intend to interfere with the toss, she knows that too. And that means she knows that there’s a fifty percent chance that you will prevent the event from occurring. The effect of all this is that your attempt to prevent the event is factored into the probability of the event. In other words, an event that you intend to prevent is inherently less likely than an event that you don’t intend to prevent, and that decreased probability will show up as a lower PDC. This is another reason that only large-scale events are detectable through the system: in order for the PDC to be high enough to detect, the scale of the event has to be something like twice as great in order to compensate for the decreased probability. Another way to look at it is that by preventing some events from happening, you’re forcing Ananke to make the events that do occur to be worse than they otherwise would be. By saving some, you’re dooming others to die.”
“
That’s some pretty abstract reasoning. I mean, I think I see what you’re getting at, that somehow, no matter what you do, the scales come out even. But it seems to me that if you just look at the situation itself, it’s pretty clear that Tali is helping people. She certainly saved a lot of lives at the pier.”
“
Sure, that time it worked out OK, at least on a surface level. But most of the time it doesn’t. Most of the time Tali goes out, she comes back in tears.”
“
Why did she start tampering in the first place? Tali must have understood everything you’re telling me, right? She must have known that she wasn’t doing any real good.”
Heller shakes his head.
“I didn’t say that. I honestly don’t know whether the net result is positive or not. Like I said, I used to think Tali was doing some good. Now I realize the question may be impossible to answer, at least coming at it rationally. It’s like a Zen
koan
: if you concentrate hard enough on it, you’ll find yourself thinking about nothing. Or maybe like a Rorschach test: the value of tampering is what you bring to it. In any case, Tali couldn’t help it. It’s not in her nature to stand by and let bad things happen. It’s worn on her, though. I’m not sure she could do it much longer. In fact, it’s occurred to me that maybe she left because the whole thing became too much for her. Frankly, I don’t know how she did it as long as she did. Have you ever known someone who deals with death every day?”