Chapter 21 Chapter 4 "Do you really believe this set?"

Brian Arthur is usually not nervous about giving academic presentations.But the economics conference held at the Santa Fe Institute at the time was unusual. Before he got to Santa Fe he felt something big was going on. "When Arrow stopped me on campus that day, I heard that John Reed, Phil Anderson, and Mari Gell-Mann were the masterminds behind the economics conference that later became the Santa Fe Institute The director called me again, and it became clear: this meeting was considered a milestone by the people at the Santa Fe Institute." As the organizers of the meeting, Arrow and Anderson set the duration for ten full days. That's a pretty long session by the standards of an academic conference.George Cowan is set to give a press conference on the final day of the conference, which will also see John Reid in person. (Indeed, Anderson's readiness to attend the meeting is a testament to this. Seven months earlier, in February 1987, all the condensed matter physicists in the world were desperate for the discovery that the crude, but relatively mild boiling point of liquefied nitrogen, minus 321 degrees Fahrenheit, Anderson, like many other theoretical physicists, has been working day and night to figure out how these "high-temperature" superconductors achieve this effect.)

When Arthur arrived at the institute at the end of August, two weeks before the economics conference, and saw the roster of attendees, he knew that his real opportunity to present his research had come.Of course, he had long admired the names of Arrow and Anderson.He also knew his Stanford colleague Tom Sargent.Tom is often cited as a contender for the Nobel Prize for his work on the close relationship between "reasonable" private-sector economic decisions and the economic environment created by government.Also participating in the meeting were Hollis Chenery, an emeritus professor at Harvard University and former director of the World Bank Research Center; Larry Summers, a prodigy from Harvard University; University, Jose Scheinkman, who pioneered the use of chaos theory in economics, and Belgian physicist David Ruelle.The roster of delegates included almost twenty names, all of them scholars of this order.

He could feel his adrenaline levels starting to rise. "I realized that this might be a pivotal moment for me, an opportunity to present my concept of increasing returns to a group of people from whom I desperately wanted support. I felt instinctively that physicists would be very receptive My point of view, but I don't know exactly what they will say about my view, or what Arrow will think. Although the economists present are of high standard, they are mainly in normal economic theory. people with a reputation, so I have absolutely no idea how they're going to react to my views. There are no indications I can follow. I don't know what tone I should set for my speech, will I be Slamming, as it sometimes happens, maybe a friendly atmosphere."

So, as the meeting start date, Tuesday, September 8, approached, Arthur spent less and less time walking and talking with Stuart Kaufman and more and more time perfecting his on the speech.He said: "Tai chi teaches you to absorb the attack and counter quickly. I think I may need to learn this. The best way to make yourself stand firm under the artillery attack is to practice martial arts in slow motion. Because every time you Throwing a punch, you can think of it as conveying a certain point to the audience." The meeting opened at 9 am in the monastery's chapel-turned-meeting room.Participants sit around two rows of long foldable tables.Sunlight poured through the stained glass windows into the meeting room, as usual.

After Anderson gave a brief introduction about his hopes for the conference's main topic of discussion, Arthur stood up to give his first formal lecture, entitled: "Self-reinforcing Mechanisms in Economics."When he started the talk, he felt for some reason Arrow's expression becoming very concerned, as if he was worried that this guy Arthur was going to present a very weird picture of economics to the physicists.Anderson said: "I know that Arthur's ability to express is very strong. From the perspective of the institute, there are many aspects of economics that I want to develop, but I have not formed any knowledge level, and I have nothing to lose. .If this experiment fails, it fails."

But whether Anderson's concern was conscious or not, it was clear that he still had his strictly analytical instincts that morning, if only more so.Arthur's speech was designed to attract the attention of physicists from the beginning.When he uses terms like "self-reinforcing mechanisms," he explains that he's basically talking about nonlinear phenomena in economics... "Wait," interrupted Anderson, "what exactly do you mean by non-linear? Are you saying that all economic phenomena are non-linear?" Well, so to speak.Arthur replied.To use mathematical precision, the general assumption of diminishing returns corresponds to a "second-order" nonlinear economic equation in which the economy always tends toward equilibrium and stability.Instead, he focused on "third-order" nonlinearities, factors that drive some aspect of the economy away from equilibrium, which engineers call "positive feedback."

Arthur's answer seemed to satisfy Anderson.He could see Anderson, Paines, and other physicists nodding in agreement with many of his points.Increasing returns, positive feedback, nonlinear equations, these are very familiar things to physicists. Halfway through the morning meeting, Anderson raised his hand and asked, "Isn't economics a lot like spin glass?" Arrow understandably chimed in, "What's a spin glass?" It just so happens that Arthur has read a lot of books on condensed matter physics over the years, and he knows very well what spin glass is.The term refers to a group of magnetic substances that have no practical value, but whose theoretical properties fascinate physicists.Anderson has studied the substance since its discovery in the 1960s and has co-authored several important papers in the field.Like the more familiar magnetic substances such as iron, spin glasses are mainly composed of metal atoms whose electrons have the property of a purely rotational motion called "spin".Its spin, like iron, can cause each atom to generate a tiny magnetic field, and the magnetic force emitted by these fields causes the spin of neighboring atoms.But unlike iron, the interaction forces of the atoms in a spin glass don't cause all the spins to join up with each other, creating large-scale magnetic field effects, unlike the ones we see on compasses and refrigerators. magnet.

In contrast, the magnetism of spin glasses is entirely random, a state that physicists call "glass." (The nature of the atomic bonds in the pane window glass is similarly arbitrary. In fact, it is technically arbitrary whether ordinary glass can be called a solid or a special viscous liquid.) Not to mention, the atomic-level disorder of the glass means that spin glasses are a complex mixture of positive and negative feedback, in which each atom tries to form a parallel spin with its neighbours, and with the others Atoms rotate in opposite directions.Generally speaking, this is simply not a state that can be maintained continuously.Every atom will always have some resistance to aligning itself with neighboring atoms with which it does not want to align itself, but in the same sense the spins can be arranged in such a wide variety of ways that for any atom That said, this resistance is within a reasonable tolerable range.Physicists call this situation "partial equilibrium."

Yes, Arthur agrees with the economics analogy to spin glass.In this sense, spin glass is a good metaphor for the economy. "The economy is naturally a mixture of positive feedback and negative feedback, and the economy will produce a lot of natural states, or multiple equilibrium states." This is exactly what he has been trying to describe in the economics of increasing returns. Then Arthur saw the physicist nodding more often in agreement.Hey, the economics are not bad."Brian and I really felt the same way," Anderson said. "We were really impressed with Arthur's speech."

And so Arthur's talk went on for two full hours: locking, interdependence of pathways, QWERTY keyboards and possible inefficiencies, and the origins of Silicon Valley."While I was speaking, the physicists kept nodding and smiling. But every ten minutes or so, Arrow would say, 'Wait,' and beg me to elaborate more, or explain why he Disagree with me. He wanted to figure out exactly every step of my reasoning. When I started to formulate precise theorems, he and the few economists present wanted to see accurate evidence. This delayed my presentation , but it makes my argument more airtight."

Arthur finally sat down exhausted, knowing that his future in increasing returns research was promising.Arthur said: "My point of view was legitimized that morning. It wasn't me who convinced Arrow and others, but physicists who convinced economists that what I did was important to them. It's as important as bread and butter. Physicists are basically saying: 'This guy knows what he's talking about, don't worry you economists'” Maybe it's just his guesswork.But to Arthur, Arrow seemed visibly relaxed. If Arthur's speech gave physicists the impression that the minds of physicists and economists were on the same channel, they were quick to correct that perception. During the first two or three days of the conference, Arrow and Anderson invited several economists to give general lectures on conventional neoclassical economics, since physicists' knowledge of economics was limited to the level of undergraduate economics textbooks . "We're all very interested in this," Anderson said.For physicists, economic theory has always been an intellectual hobby. “We’d love to learn something about that.” Indeed, physicists were simply overwhelmed by the mathematical talents of economists when the difficult principles, laws and evidence were displayed on the screen through the projector.They felt both admiration and horror.They produced the rebellious view of traditional economics that Arthur and many other economists have uttered over the years.A young physicist said he remembered shaking his head in disbelief and saying: "These theories are too perfect. Economists seem to be so intoxicated with their mathematical formulas that they see nothing but trees." Nowhere to see the forest. Economists spend a lot of time trying to integrate mathematics into economics, and I think they may completely forget what these mathematical models are for, what these models are, or whether the underlying assumptions are Any sense. In many cases, all that's needed is common sense. Maybe if they both had low IQs, the models they made would be better." Of course, physicists have no problem with mathematics itself.Physics, more than the most thoroughly mathematicized sciences, has the aid of mathematical tools.But what most economists don't know, and would be surprised to find out, is that physicists have been relatively casual about mathematics. "Physicists use a little bit of rigor, a little bit of instinct, and also do some calculations on the back of the envelope. So their styles are really very different," says Arrow, who remembers feeling the same way when he discovered it himself. very shocked.The reason for this is that physicists always establish their hypotheses and theories through empirical evidence. "I don't know what the situation is like in research fields like relativity theory, but the general trend in physics is that you do some calculations first, and then you get data to prove it through experiments. So the lack of rigor in theory doesn't really matter." Not a very serious problem, the errors of the theory will always be corrected by the experiment. But in economics, we cannot obtain the data to prove the quality of the theory, and it is impossible to obtain the data of economics like physicists obtain the data .Our research has to start from a very small base, so we have to make sure that every step of the reasoning in theory is accurate." This is a very legitimate reason.But economists do therefore pay little attention to the evidence that does exist.Economists are still discouraged by the point made by physicists.For example, from time to time someone asks the question, "How big are the non-economic influences? Like political motivation in OPEC oil prices and mass psychology in the stock market." The more unscientific, tangled set of questions is the answer to the question: "These non-economic factors really don't matter," or "These factors are important, but they are hard to explain economically." to deal with", or "These problems are not always difficult to deal with, in fact, in some special cases, we are using economic methods to deal with these problems", or "We don't need to pay attention to these non-economic factors at all. , because these factors will dissolve themselves in the economic effect". And then there's this theory of "reasonable expectations."Arthur remembered being asked during his first day's presentation, "Is economics simpler than physics?" Arthur replied: "In a sense it is. We call the particles 'actors' - like banks, companies, consumers, governments, etc. These actors interact with each other like Particles interact with each other. It’s just that in economics, we don’t usually think about spatial scales, which makes economics much simpler than physics.” But he went on to say that there is still a big difference between economics and physics. "The particles of economics are smart, but the particles of physics are dumb." In physics, the elementary particles have no history, no experience, no goals, no worries and hopes for the future.It simply exists.This is why physicists are free to talk about the "laws of the universe".Arthur said that the physicist's particles reacted blindly and completely obediently to external forces.And in economics, "our particles anticipate ahead of time, trying to figure out how other particles will react if they take a certain action. No matter how you simulate it, our particles act based on Expectations and strategic considerations. And that’s where economics gets really hard.” Immediately, he said, he saw all the physicists in the room sitting up straight. "The problems of economics are not simple. Economics is like their physics, but this science has two interesting strange words: strategy and expectation." Unfortunately, economists' conventional interpretation of the expectation problem, that is, one in which expectations are based entirely on rationality, has set off firestorms among physicists.The fact that perfectly rational agents can indeed make perfectly correct predictions, that is, assuming that humans are prescient about everything, and therefore make choices that apply well into the future, humans predict them with unerring rationality All consequences implied by the action to be taken.This way you can safely say that humans in any situation will act in the most beneficial way based on the information they get.Of course, humans make mistakes in certain situations, such as oil shocks, technological revolutions, political intervention in bank interest rates, and other non-economic surprises, but humans are smart and quick enough to adjust One's own actions can therefore always keep the economy in a rolling equilibrium state, and supply will always be exactly equal to demand. The only problem, of course, is that humans are neither perfectly rational nor 100 percent correct in predicting the future, as physicists have long railed against.What's more, as several physicists have pointed out, even if you assume that humans are perfectly rational, and then assume that humans can make perfectly correct predictions about the future, there are theoretical loopholes.In nonlinear systems—and economics is certainly nonlinear—chaos theory tells us that even the smallest uncertainty in your internal environment, known to you, tends to have irreversible consequences, perhaps only momentarily, Your predictions may turn out to be nonsense. "Physicists have been running on us," Arthur said. "They're very surprised by the assumptions made by economists. The test of the assumption is not from real life, but whether the assumption is consistent with the prevailing view in the field of economics." Surprised. I saw Anderson leaning back in his chair with a smile on his face and asking, "'You really believe this? '" The cornered economists replied: "But it helps us solve some problems. Without making these assumptions, nothing can be done." Physicists would immediately retort: ​​"But what problem does that help you solve? If this assumption doesn't fit reality, you still can't really solve the problem." Economists aren't known for their modesty, and the economists at the Santa Fe Institute wouldn't be economists if they weren't enraged by all these attacks.They could complain about the shortcomings of economics in their own circles, after all, Arrow had purposely brought educated skeptics of conventional economics to the conference.But who wants to hear criticism of economics from outsiders?While every economist tried to listen and be polite enough to get the meeting going, emotionally they were markedly resistant: "What can physics do to help us? Are you guys that smart?" Of course, physicists are not known for their modesty.In fact, to many people outside of physics, the word that comes to mind for a physicist is "intolerable arrogance".This is not an intentional attitude of physicists, nor is it due to their personality, but more like the unconscious sense of superiority revealed by the British aristocracy.Indeed, in the minds of physicists, they are the aristocrats of science.They have acquired this sense of nobility in countless subtle and not-subtle ways from their first physics course: they are the heirs of Newton, Maxwell, Einstein and Bohr.Physics is the solidest, purest, most robust science, so if the economists show a tough attitude at Santa Fe, they will immediately fight tough, just like economists As Larry Summers quipped: Take an "I'm Tarzan, you're Jane" attitude: "Give us three weeks to master economics, and we'll show you what to do right." Eugenia Singer, who represented Reed at the meeting, has been concerned about the potential for conflict over such egotism on both sides."I was terrified that if this 'Tarzan' effect was triggered, our entire research program would be killed in the test tube before it was even ready," she recalls. At first, things seemed to be heading in that direction. "Most economists sit on one side of the table and most physicists sit on the other. I'm horrified by this phenomenon." Every now and then she takes Paines and Cowan aside and says, The two sides sit a little closer?" But this situation has not been improved. The danger of the potential total incommunication of economists and physicists was no less than a nightmare for George Cowan.Not because the Institute might lose Citibank funding if the meeting fails, but the meeting is the most convincing evidence of the Santa Fe Institute's philosophy.Two years ago, at the startup workshop, they brought together heroes from all walks of life for a weekend.Now, two groups of completely different and very proud people are put together for a 10-day meeting to study substantive issues together."We're trying to create an academic community that never existed before," Cowan said. "Maybe it won't work, maybe they have nothing to talk to each other, maybe it's just a heated debate between the two sides." It's not an unfounded concern, and late Santa Fe workshops have occasionally seen attendees rattling each other.But in September 1987, the gods of interdisciplinary research decided to smile again.Anderson and Arrow had done their best to bring people into the meeting who could listen and talk.Despite the underlying anger between the two parties, the participants eventually found that they had a lot to talk about.In fact, in retrospect, the two sides began to reach a consensus in a very short period of time. Of course, this was even more true for Arthur, who reached an agreement with the physicists in only half a day.