John Barnes - Directive 51

If she looked any more wholesome and innocent, Ysabel thought , I’d just sell her to a pimp for resale to a Japanese tourist with a rape-a-librarian fetish. Dammit, where’s a good pimp when you need one?

But she said, “Yeah. So I’m going to turn in now and get up at dawn. It won’t be scary or dangerous or anything if you’d like to come along.”

“That would be so awesome.”

“Okay, so,” Ysabel said, “why don’t you grab first showers?”

“Eahh, I kind of want to catch the news. You can.”

It would look too weird to insist, so Ysabel shrugged. Anyway, it might be her last shower ever with unlimited hot water. She’d dress and take off while Little Miss Forgettable was in the shower. Nothing easier.

Ysabel was just toweling off when she saw the doorknob start to turn slowly, soundlessly, most of the way around and then return to its normal position. She reached for the little lock button. The door slapped her arm back and the base of a lamp caught her under the jaw.

Room spinning. Skull screaming like a bad smoke detector. What? She was—

The girl jammed the lamp into Ysabel’s naked belly, knocking the wind out of her, and let it crash to the floor, slamming Ysabel’s foot. She grabbed Ysabel’s long hair, wrapped it in a fist that pressed against her neck, and forced her head back and down, dragging her backward by the hair into the main room.

Ysabel’s feet slid and wobbled till she lost her balance completely and fell sideways on her ass on the hotel-room carpet.

The girl punched her, hard, in the cheek. “Roll over.”

She did. As she realized that the girl was tying her hands behind her back, she caught phrases from the American TV news: “believed to be,” “wanted for questioning,” and “Yuma.”

Oh, crap, and because they were in this big American Plaztatic Tower of a hotel, there would be someone with good English at the front desk, who would take “I need the police right away” seriously.

The carpet ground against her face, everything hurt, and the girl said, “Hey, is this one of those adventures you’re so patronizing about having had? You’re right, they’re fun .”

After all the excitement, Arnie was stuck, as ever, with giving his presentation at a time when it had to be an anticlimax. “Let’s start with what I usually do, okay? So you’ll see why it was I found what I found.

“Semiotics is the study of how signs mean—how one thing stands for another or how a message connects to its meaning. Like an oncoming car flashes its lights at you on the highway, what do you know? Something wrong ahead, cop or accident or animals on the road, so you slow down and pay attention. Statistical semiotics is about how populations of signs function as signs. Like it’s close to night, or you’re close to a tunnel, and in the other lane you see ten cars in a row with their lights on, so you know it’s dark ahead, and you turn your lights on—and if other people do that, it becomes a message to other oncoming drivers.”

“But they weren’t sending a message,” Heather said, playing the role to hurry Arnie to the point. “They just had their lights on, and you saw them.”

“Many messages aren’t sent by anyone,” Arnie said, “even though they’re perceived and received. A deer doesn’t leave tracks or scent because it’s trying to tell hunters or dogs where to find it. If there is an intention, that’s expressed through yet more signs, which might or might not be important. So it doesn’t matter that the other drivers weren’t trying to say anything to you. And it doesn’t matter that any one car had its lights on or off—one car could just be a forgetful person or one of those cars that doesn’t give you a choice. But a population of signs formed a message. One swallow does not a summer make, but five thousand swallows and a million green buds on the trees and forty Memorial Day sales in the stores does.

“Those messages that no one intends to send, that are sent by a lot of different sources collectively, are called system artifacts. Like people doing the wave in a stadium; the wave isn’t any one person, but it’s visible to everyone. Like one color of scrunchy being reserved for the popular girls in a school; nobody makes it up, everyone just knows. Surprising numbers of fads and fashions have no originator for any practical purpose.”

“So you guess at fads and trends like the pattern recognizers?” Colonel Green asked. “And you’re studying how Daybreak is a fad?”

“No.” Arnie had an expression that amused Heather every time she saw it; it was the same one she’d seen on an astronomer who had been asked to cast a horoscope. “The pattern recognizers and trendspotters just know a lot about fads in the past, and they watch the news and the social media and look for things that look like what happened before; it’s all about their feeling and intuition.”

“The sort of thing I do,” Crittenden put in. “Highly believable and I like to think insightful, but nothing the historian at Charlemagne’s court couldn’t have done.”

“At least it’s entertaining,” Arnie said, “but forgive my pointing out it’s not science, and it can’t tell you anything you don’t already know on some level. What I do is describe in numbers the whole huge network of communications—everyone and everything, be it person, bot, book, web site, accident, whatever creates signs, and the signs they send, and everything that interprets the signs and the secondary signs that they send to each other about them. There are numbers and geometry to express all the ways the messages and their sources and targets are similar, different, parallel, whatever. That results in huge data structures, many terabytes even for the most elementary problems, which is why no one did this before supercomputers. Then, with a very fast computer, we use wavelet shrinkage—that’s a statistical method for estimating fractals if you’re up on your math—to find patterns that are persisting.

“Or to oversimplify and use an analogy—which is what Dr. Crittenden does, and why he’s easier to understand than I am—the pattern recognizers look at the clouds, and say ‘That’s a horsie, I feel good about horsies, I guess it won’t rain for the picnic.’ I teach the computer to look at trillions of pictures of clouds and notice that puffy ones with flat bottoms are associated with lightning and hail—even if I’ve never seen a cloud before.”

“Or a horsie,” Green said, grinning. “I’ve got grandkids. I can relate to horsies.”

“Or a horsie,” Arnie said. “I don’t find horsies, I find the pattern by which people learn to look for horsies—and I also find some people talking to some other people are more likely to call them horsies, and under what circumstances, and maybe construct a relationship that associates with the prefix grand- in other relationships. And I can do that in languages I don’t speak. So statistical semiotic analysis shows me brand-new patterns, things that haven’t been perceived before relating in ways that people haven’t named before; from there we can work out the tests and methods for detecting following those patterns if they’re important and persistent.

“Usually these patterns that fall out of the math are just what we call an idea pump—a person or organization that just repeats a message and encourages interested people to repeat it—and those are intentional and single-minded. Like the pattern of beer commercials having pretty girls and occurring during football games in North America.