Rationality- From AI to Zombies

Home > Science > Rationality- From AI to Zombies > Page 7
Rationality- From AI to Zombies Page 7

by Eliezer Yudkowsky


  Here is the secret of deliberate rationality—this whole process is not magic, and you can understand it. You can understand how you see your shoelaces. You can think about which sort of thinking processes will create beliefs which mirror reality, and which thinking processes will not.

  Mice can see, but they can’t understand seeing. You can understand seeing, and because of that, you can do things that mice cannot do. Take a moment to marvel at this, for it is indeed marvelous.

  Mice see, but they don’t know they have visual cortexes, so they can’t correct for optical illusions. A mouse lives in a mental world that includes cats, holes, cheese and mousetraps—but not mouse brains. Their camera does not take pictures of its own lens. But we, as humans, can look at a seemingly bizarre image, and realize that part of what we’re seeing is the lens itself. You don’t always have to believe your own eyes, but you have to realize that you have eyes—you must have distinct mental buckets for the map and the territory, for the senses and reality. Lest you think this a trivial ability, remember how rare it is in the animal kingdom.

  The whole idea of Science is, simply, reflective reasoning about a more reliable process for making the contents of your mind mirror the contents of the world. It is the sort of thing mice would never invent. Pondering this business of “performing replicable experiments to falsify theories,” we can see why it works. Science is not a separate magisterium, far away from real life and the understanding of ordinary mortals. Science is not something that only applies to the inside of laboratories. Science, itself, is an understandable process-in-the-world that correlates brains with reality.

  Science makes sense, when you think about it. But mice can’t think about thinking, which is why they don’t have Science. One should not overlook the wonder of this—or the potential power it bestows on us as individuals, not just scientific societies.

  Admittedly, understanding the engine of thought may be a little more complicated than understanding a steam engine—but it is not a fundamentally different task.

  Once upon a time, I went to EFNet’s #philosophy chatroom to ask, “Do you believe a nuclear war will occur in the next 20 years? If no, why not?” One person who answered the question said he didn’t expect a nuclear war for 100 years, because “All of the players involved in decisions regarding nuclear war are not interested right now.” “But why extend that out for 100 years?” I asked. “Pure hope,” was his reply.

  Reflecting on this whole thought process, we can see why the thought of nuclear war makes the person unhappy, and we can see how his brain therefore rejects the belief. But if you imagine a billion worlds—Everett branches, or Tegmark duplicates1—this thought process will not systematically correlate optimists to branches in which no nuclear war occurs. (Some clever fellow is bound to say, “Ah, but since I have hope, I’ll work a little harder at my job, pump up the global economy, and thus help to prevent countries from sliding into the angry and hopeless state where nuclear war is a possibility. So the two events are related after all.” At this point, we have to drag in Bayes’s Theorem and measure the relationship quantitatively. Your optimistic nature cannot have that large an effect on the world; it cannot, of itself, decrease the probability of nuclear war by 20%, or however much your optimistic nature shifted your beliefs. Shifting your beliefs by a large amount, due to an event that only slightly increases your chance of being right, will still mess up your mapping.)

  To ask which beliefs make you happy is to turn inward, not outward—it tells you something about yourself, but it is not evidence entangled with the environment. I have nothing against happiness, but it should follow from your picture of the world, rather than tampering with the mental paintbrushes.

  If you can see this—if you can see that hope is shifting your first-order thoughts by too large a degree—if you can understand your mind as a mapping engine that has flaws—then you can apply a reflective correction. The brain is a flawed lens through which to see reality. This is true of both mouse brains and human brains. But a human brain is a flawed lens that can understand its own flaws—its systematic errors, its biases—and apply second-order corrections to them. This, in practice, makes the lens far more powerful. Not perfect, but far more powerful.

  *

  1. Max Tegmark, “Parallel Universes,” in Science and Ultimate Reality: Quantum Theory, Cosmology, and Complexity, ed. John D. Barrow, Paul C. W. Davies, and Charles L. Harper Jr. (New York: Cambridge University Press, 2004), 459–491.

  Part B

  Fake Beliefs

  11

  Making Beliefs Pay Rent (in Anticipated Experiences)

  Thus begins the ancient parable:

  If a tree falls in a forest and no one hears it, does it make a sound? One says, “Yes it does, for it makes vibrations in the air.” Another says, “No it does not, for there is no auditory processing in any brain.”

  Suppose that, after the tree falls, the two walk into the forest together. Will one expect to see the tree fallen to the right, and the other expect to see the tree fallen to the left? Suppose that before the tree falls, the two leave a sound recorder next to the tree. Would one, playing back the recorder, expect to hear something different from the other? Suppose they attach an electroencephalograph to any brain in the world; would one expect to see a different trace than the other? Though the two argue, one saying “No,” and the other saying “Yes,” they do not anticipate any different experiences. The two think they have different models of the world, but they have no difference with respect to what they expect will happen to them.

  It’s tempting to try to eliminate this mistake class by insisting that the only legitimate kind of belief is an anticipation of sensory experience. But the world does, in fact, contain much that is not sensed directly. We don’t see the atoms underlying the brick, but the atoms are in fact there. There is a floor beneath your feet, but you don’t experience the floor directly; you see the light reflected from the floor, or rather, you see what your retina and visual cortex have processed of that light. To infer the floor from seeing the floor is to step back into the unseen causes of experience. It may seem like a very short and direct step, but it is still a step.

  You stand on top of a tall building, next to a grandfather clock with an hour, minute, and ticking second hand. In your hand is a bowling ball, and you drop it off the roof. On which tick of the clock will you hear the crash of the bowling ball hitting the ground?

  To answer precisely, you must use beliefs like Earth’s gravity is 9.8 meters per second per second, and This building is around 120 meters tall. These beliefs are not wordless anticipations of a sensory experience; they are verbal-ish, propositional. It probably does not exaggerate much to describe these two beliefs as sentences made out of words. But these two beliefs have an inferential consequence that is a direct sensory anticipation—if the clock’s second hand is on the 12 numeral when you drop the ball, you anticipate seeing it on the 1 numeral when you hear the crash five seconds later. To anticipate sensory experiences as precisely as possible, we must process beliefs that are not anticipations of sensory experience.

  It is a great strength of Homo sapiens that we can, better than any other species in the world, learn to model the unseen. It is also one of our great weak points. Humans often believe in things that are not only unseen but unreal.

  The same brain that builds a network of inferred causes behind sensory experience can also build a network of causes that is not connected to sensory experience, or poorly connected. Alchemists believed that phlogiston caused fire—we could oversimply their minds by drawing a little node labeled “Phlogiston,” and an arrow from this node to their sensory experience of a crackling campfire—but this belief yielded no advance predictions; the link from phlogiston to experience was always configured after the experience, rather than constraining the experience in advance. Or suppose your postmodern English professor teaches you that the famous writer Wulky Wilkinsen is actually a “post-utopian.” What does this m
ean you should expect from his books? Nothing. The belief, if you can call it that, doesn’t connect to sensory experience at all. But you had better remember the propositional assertion that “Wulky Wilkinsen” has the “post-utopian” attribute, so you can regurgitate it on the upcoming quiz. Likewise if “post-utopians” show “colonial alienation”; if the quiz asks whether Wulky Wilkinsen shows colonial alienation, you’d better answer yes. The beliefs are connected to each other, though still not connected to any anticipated experience.

  We can build up whole networks of beliefs that are connected only to each other—call these “floating” beliefs. It is a uniquely human flaw among animal species, a perversion of Homo sapiens’s ability to build more general and flexible belief networks.

  The rationalist virtue of empiricism consists of constantly asking which experiences our beliefs predict—or better yet, prohibit. Do you believe that phlogiston is the cause of fire? Then what do you expect to see happen, because of that? Do you believe that Wulky Wilkinsen is a post-utopian? Then what do you expect to see because of that? No, not “colonial alienation”; what experience will happen to you? Do you believe that if a tree falls in the forest, and no one hears it, it still makes a sound? Then what experience must therefore befall you?

  It is even better to ask: what experience must not happen to you? Do you believe that élan vital explains the mysterious aliveness of living beings? Then what does this belief not allow to happen—what would definitely falsify this belief? A null answer means that your belief does not constrain experience; it permits anything to happen to you. It floats.

  When you argue a seemingly factual question, always keep in mind which difference of anticipation you are arguing about. If you can’t find the difference of anticipation, you’re probably arguing about labels in your belief network—or even worse, floating beliefs, barnacles on your network. If you don’t know what experiences are implied by Wulky Wilkinsen being a post-utopian, you can go on arguing forever.

  Above all, don’t ask what to believe—ask what to anticipate. Every question of belief should flow from a question of anticipation, and that question of anticipation should be the center of the inquiry. Every guess of belief should begin by flowing to a specific guess of anticipation, and should continue to pay rent in future anticipations. If a belief turns deadbeat, evict it.

  *

  12

  A Fable of Science and Politics

  In the time of the Roman Empire, civic life was divided between the Blue and Green factions. The Blues and the Greens murdered each other in single combats, in ambushes, in group battles, in riots. Procopius said of the warring factions: “So there grows up in them against their fellow men a hostility which has no cause, and at no time does it cease or disappear, for it gives place neither to the ties of marriage nor of relationship nor of friendship, and the case is the same even though those who differ with respect to these colors be brothers or any other kin.”1 Edward Gibbon wrote: “The support of a faction became necessary to every candidate for civil or ecclesiastical honors.”2

  Who were the Blues and the Greens? They were sports fans—the partisans of the blue and green chariot-racing teams.

  Imagine a future society that flees into a vast underground network of caverns and seals the entrances. We shall not specify whether they flee disease, war, or radiation; we shall suppose the first Undergrounders manage to grow food, find water, recycle air, make light, and survive, and that their descendants thrive and eventually form cities. Of the world above, there are only legends written on scraps of paper; and one of these scraps of paper describes the sky, a vast open space of air above a great unbounded floor. The sky is cerulean in color, and contains strange floating objects like enormous tufts of white cotton. But the meaning of the word “cerulean” is controversial; some say that it refers to the color known as “blue,” and others that it refers to the color known as “green.”

  In the early days of the underground society, the Blues and Greens contested with open violence; but today, truce prevails—a peace born of a growing sense of pointlessness. Cultural mores have changed; there is a large and prosperous middle class that has grown up with effective law enforcement and become unaccustomed to violence. The schools provide some sense of historical perspective; how long the battle between Blues and Greens continued, how many died, how little changed as a result. Minds have been laid open to the strange new philosophy that people are people, whether they be Blue or Green.

  The conflict has not vanished. Society is still divided along Blue and Green lines, and there is a “Blue” and a “Green” position on almost every contemporary issue of political or cultural importance. The Blues advocate taxes on individual incomes, the Greens advocate taxes on merchant sales; the Blues advocate stricter marriage laws, while the Greens wish to make it easier to obtain divorces; the Blues take their support from the heart of city areas, while the more distant farmers and watersellers tend to be Green; the Blues believe that the Earth is a huge spherical rock at the center of the universe, the Greens that it is a huge flat rock circling some other object called a Sun. Not every Blue or every Green citizen takes the “Blue” or “Green” position on every issue, but it would be rare to find a city merchant who believed the sky was blue, and yet advocated an individual tax and freer marriage laws.

  The Underground is still polarized; an uneasy peace. A few folk genuinely think that Blues and Greens should be friends, and it is now common for a Green to patronize a Blue shop, or for a Blue to visit a Green tavern. Yet from a truce originally born of exhaustion, there is a quietly growing spirit of tolerance, even friendship.

  One day, the Underground is shaken by a minor earthquake. A sightseeing party of six is caught in the tremblor while looking at the ruins of ancient dwellings in the upper caverns. They feel the brief movement of the rock under their feet, and one of the tourists trips and scrapes her knee. The party decides to turn back, fearing further earthquakes. On their way back, one person catches a whiff of something strange in the air, a scent coming from a long-unused passageway. Ignoring the well-meant cautions of fellow travellers, the person borrows a powered lantern and walks into the passageway. The stone corridor wends upward . . . and upward . . . and finally terminates in a hole carved out of the world, a place where all stone ends. Distance, endless distance, stretches away into forever; a gathering space to hold a thousand cities. Unimaginably far above, too bright to look at directly, a searing spark casts light over all visible space, the naked filament of some huge light bulb. In the air, hanging unsupported, are great incomprehensible tufts of white cotton. And the vast glowing ceiling above . . . the color . . . is . . .

  Now history branches, depending on which member of the sightseeing party decided to follow the corridor to the surface.

  Aditya the Blue stood under the blue forever, and slowly smiled. It was not a pleasant smile. There was hatred, and wounded pride; it recalled every argument she’d ever had with a Green, every rivalry, every contested promotion. “You were right all along,” the sky whispered down at her, “and now you can prove it.” For a moment Aditya stood there, absorbing the message, glorying in it, and then she turned back to the stone corridor to tell the world. As Aditya walked, she curled her hand into a clenched fist. “The truce,” she said, “is over.”

  Barron the Green stared incomprehendingly at the chaos of colors for long seconds. Understanding, when it came, drove a pile-driver punch into the pit of his stomach. Tears started from his eyes. Barron thought of the Massacre of Cathay, where a Blue army had massacred every citizen of a Green town, including children; he thought of the ancient Blue general, Annas Rell, who had declared Greens “a pit of disease; a pestilence to be cleansed”; he thought of the glints of hatred he’d seen in Blue eyes and something inside him cracked. “How can you be on their side?” Barron screamed at the sky, and then he began to weep; because he knew, standing under the malevolent blue glare, that the universe had always been a place of evil.
>
  Charles the Blue considered the blue ceiling, taken aback. As a professor in a mixed college, Charles had carefully emphasized that Blue and Green viewpoints were equally valid and deserving of tolerance: The sky was a metaphysical construct, and cerulean a color that could be seen in more than one way. Briefly, Charles wondered whether a Green, standing in this place, might not see a green ceiling above; or if perhaps the ceiling would be green at this time tomorrow; but he couldn’t stake the continued survival of civilization on that. This was merely a natural phenomenon of some kind, having nothing to do with moral philosophy or society . . . but one that might be readily misinterpreted, Charles feared. Charles sighed, and turned to go back into the corridor. Tomorrow he would come back alone and block off the passageway.

  Daria, once Green, tried to breathe amid the ashes of her world. I will not flinch, Daria told herself, I will not look away. She had been Green all her life, and now she must be Blue. Her friends, her family, would turn from her. Speak the truth, even if your voice trembles, her father had told her; but her father was dead now, and her mother would never understand. Daria stared down the calm blue gaze of the sky, trying to accept it, and finally her breathing quietened. I was wrong, she said to herself mournfully; it’s not so complicated, after all. She would find new friends, and perhaps her family would forgive her . . . or, she wondered with a tinge of hope, rise to this same test, standing underneath this same sky? “The sky is blue,” Daria said experimentally, and nothing dire happened to her; but she couldn’t bring herself to smile. Daria the Blue exhaled sadly, and went back into the world, wondering what she would say.

 

‹ Prev