经典力学：决定论的宇宙蓝图

The structure of classical mechanics (经典力学: 描述宏观物体运动的物理理论) implies that if you knew the position and velocity, not just of one particle, but of every particle in the universe, then the laws of physics would determine what happens next, at the next moment and the next moment, and infinitely far into the future, and for that matter, indefinitely far into the past. This might bother you a little bit if you think, "Wait a minute, I'm a person, I'm a human being, I have the ability to make choices. I'm not determined by the laws of physics." Both scientists and philosophers have thought about that, and they still don't agree on what the right way to think about it is. I'm Sean Carroll, a physicist and philosopher at Johns Hopkins University, host of the "Mindscape Podcast," and author of "The Biggest Ideas in the Universe" series, including "Space, Time, and Motion" and "Quanta and Fields." 经典力学 (Classical Mechanics: 描述宏观物体运动的物理理论) 的结构表明，如果你知道宇宙中所有粒子——不仅仅是一个粒子——的位置和速度，那么物理定律就能决定接下来会发生什么，在下一个时刻，再下一个时刻，直至无限远的未来，甚至也能无限回溯到过去。如果你想，“等等，我是一个人，一个人类，我有能力做出选择。我不是由物理定律决定的。”这可能会让你有点困扰。科学家和哲学家都思考过这个问题，但他们仍然未能就如何正确看待它达成一致。我是肖恩·卡罗尔，约翰霍普金斯大学的物理学家和哲学家，《心景播客》的主持人，也是多本书的作者，最新作品是“宇宙中的大思想”系列，包括《空间、时间与运动》和《量子与场》。

物理学家的简化之道：球形奶牛的启示

Is reality just a clockwork machine? I like to say that physics is hard because physics is easy. By that I mean, when we think about physics as students, we took classes, we read books, and it was hard because there's all this new stuff, all these ideas, all these equations that we don't come across in our everyday lives. But the reason those ideas are hard and those equations are there is because physicists have an amazingly successful technique: to take all the messy world around us with all its peculiarities and specificities and boil it down to really simple systems. You might imagine or remember taking physics courses where there were frictionless surfaces and pendula that rocked back and forth perfectly. We are always idealizing, always imagining there are no complications, and then we put them back in. That's a strategy that would completely fail if you tried to do it for psychology, biology, or political science. But for physics, it turns out to work incredibly well. There's a joke that physicists like to tell each other—they don't like to tell it to others because it's not very funny. But the idea is that a dairy farmer wants to improve his yield, getting more milk out of the cows, and for some reason, he approaches a physicist and says, "Could you look at my farm and tell me how to get more milk out of the cows?" The physicist thinks about it, comes back with a sheaf of calculations, and says, "Okay, first, imagine a spherical cow." The idea—I know it's not that funny—but the idea is that the first thing the physicist is going to do is try to imagine a simpler situation. Real cows are not spherical. It would be a very different dairy farm if the cows were spherical, but you can calculate the volume and metabolic rate of the cow much more easily if it were spherical. The joke, of course, is that doesn't work in dairy farming, but it works really well when you're considering a spherical universe, a spherical solar system, or a spherical atom.

现实仅仅是一台发条机器吗？我喜欢说物理学之所以难，是因为它很简单。我的意思是，当我们作为学生学习物理时，我们上课、读书，觉得它很难，因为有那么多新概念、新思想、新方程式，这些东西在我们的日常生活中从未遇到。但这些思想之所以难，这些方程式之所以存在，是因为物理学家拥有一项极其成功的技术：将我们周围混乱、充满怪异和特殊性的世界，提炼成非常非常简单的系统。你可能还记得，在物理课上，有无摩擦力的表面，有完美来回摆动的摆。我们总是在理想化，总是假设没有复杂因素，然后再把它们加回去。如果你尝试将这种策略用于心理学、生物学或政治学，它会彻底失败。但对于物理学来说，它却出奇地有效。物理学家之间有个笑话，他们不喜欢讲给别人听，因为不怎么好笑。但这个想法是，一个奶农想提高产量，让奶牛产更多的奶，出于某种原因，他找到一位物理学家，说：“你能看看我的农场，告诉我如何让奶牛产更多奶吗？”物理学家思考了一会儿，带着一叠计算结果回来，说：“好的，首先，想象一头球形奶牛。”这个想法——我知道它不怎么好笑——但重点是，物理学家要做的第一件事就是尝试设想一个更简单的情况。真正的奶牛不是球形的。如果奶牛是球形的，那将是一个非常不同的奶牛场，但如果它是球形的，你可以更容易地计算奶牛的体积和代谢率。当然，这个笑话的寓意是，这在奶牛养殖中行不通，但当你考虑一个球形宇宙、球形太阳系或球形原子时，它却非常有效。

牛顿的革命：从亚里士多德到力与运动

The first really huge revolution in physics was the existence of classical mechanics, handed down by Isaac Newton and others. It took a while; Newton was building on the shoulders of giants. But before Newton, there was Aristotle. Aristotle said that things have natural places they want to be and natural ways they want to move. Newton, however, said something completely different. He stated that if something is not acted on by a force, it will continue in a straight line at a constant velocity forever. And if it is acted on by a force, he could tell you how it would move, with an equation to do that.

物理学中第一次真正巨大的革命是经典力学的出现，它由艾萨克·牛顿及其他人传承下来。这花了一段时间；牛顿是站在巨人的肩膀上。但在牛顿之前，有亚里士多德。亚里士多德认为事物有它们想去的“自然位置”，有它们想运动的“自然方式”。而牛顿则提出了完全不同的观点。他说，如果一个物体不受力的作用，它将永远以恒定的速度沿直线运动。如果它受到力的作用，他就能告诉你它会如何运动，他有一个方程可以做到这一点。

台球：一个理想化的决定论系统

Physicists like to simplify things a great deal. Billiards, the pool game, is pretty close to being simple. It's not exactly, because when you hear those balls click against each other, that sound is giving off energy and it's somewhat wasteful. But in principle, if you had no friction, no sound, and no air resistance, the balls bouncing around the pool table—let's also imagine there are no pockets, so the balls can just bounce off the edges of the table forever—they would go forever; they wouldn't stop, right? The energy contained in the system remains constant. And the laws of physics, as Laplace points out, suffice to predict exactly what the balls are going to do at every moment, given what they are doing right now. So, to the extent that it's okay to ignore friction, noise, and things like that, not only is it true that if you imagine hitting the balls and watching them move, you could predict exactly what's going to happen on the basis of the laws of physics. If somehow you could take a snapshot later in their motion, so you know both where they are and how fast they're moving—maybe a little clip of a movie—then the laws of physics would let you go backwards and reverse engineer what exact configuration the balls were in. We don't perceive that in our everyday world because the world is full of noise, dissipation, and air resistance and things like that. But in the pristine, perfect world of imagined classical mechanics, the past and future work equally well; you can go from any one moment to any other moment.

物理学家非常喜欢简化事物。台球，也就是我们玩的撞球游戏，就非常接近简单。它并非完全简单，因为你知道，当你听到球相互撞击时，那声音会散发能量，这有点浪费。但原则上，如果没有摩擦力、没有声音、没有空气阻力，台球桌上的球会一直弹跳——我们再假设没有袋口，所以球可以永远在桌边弹跳。它们会永远运动下去，不会停止，对吧？系统中所包含的能量保持不变。正如拉普拉斯指出的，物理定律足以精确预测球在任何时刻会做什么，只要知道它们现在在做什么。因此，在可以忽略摩擦力、噪音之类因素的程度上，不仅如果你想象击打这些球并观察它们的运动，你能够根据物理定律精确预测将会发生什么。如果某种方式，你能在它们运动的后期拍下一张快照，从而知道它们的位置和速度——也许是一小段电影片段——那么物理定律就能让你回溯，反向推导出这些球当时处于什么精确配置。我们在日常世界中感受不到这一点，因为世界充满了噪音、耗散和空气阻力等因素。但在想象中的经典力学那纯净、完美的世界里，过去和未来同样有效；你可以从任何一个时刻推断到任何其他时刻。

拉普拉斯妖与发条宇宙范式

It's interesting that Newton came up with the framework of classical mechanics in the 1600s, and people were very excited—physicists, mathematicians, philosophers. They didn't really have physicists at the time; they were all considered natural philosophers. But they worked on it, thinking about the motions of the planets and things like that. The implications of this idea are profound for how we think about what physics is and what physics tells us. This wasn't fully realized until Pierre-Simon Laplace, over a hundred years after Newton. The structure of classical mechanics implies that if you knew the position and velocity, not just of one particle, but of every particle in the universe, and you knew the laws of physics and had infinite calculational abilities—none of these are at all plausible, but we're imagining them right now—then the laws of physics would determine what happens next, at the next and the next moment, and infinitely far into the future, and for that matter, indefinitely far into the past. So, according to classical mechanics, you can take any one moment in the history of the universe, and the information contained in what is going on at that moment is sufficient to fix what will happen at every other moment in history. Laplace, who was quite imaginative about these things, put it in terms of a metaphor. He said, "Imagine a vast intelligence." Later, commentators dubbed it Laplace's demon (拉普拉斯妖: 一个假想的智能体，如果它能知晓宇宙中所有粒子在某一时刻的位置和动量，就能预测宇宙的未来和回溯过去). He didn't call it that; he was famously an atheist and didn't like to talk about demons. But to this demon, this vast intelligence who could know everything about the universe at any one moment, Laplace says the past and future are an open book. You would know everything because what happens now fixes the entirety of space and time. This is the idea that the laws of physics fix what's going to happen, in principle, precisely and exactly, if you know what's happening right now. So this became known as the clockwork universe paradigm (发条宇宙范式: 一种哲学观点，认为宇宙像一个巨大的机械装置，其所有事件都由物理定律预先决定). The universe clicks along in perfect accord with the laws of physics forever.

有趣的是，牛顿在17世纪提出了经典力学的框架，人们对此非常兴奋——物理学家、数学家、哲学家。当时他们还没有真正意义上的物理学家，都被认为是自然哲学家，但他们致力于此，思考行星的运动等等。这个想法的含义对于我们如何思考物理学是什么、物理学告诉我们什么，都具有深远的影响。这直到牛顿之后一百多年，皮埃尔-西蒙·拉普拉斯才完全认识到。经典力学的结构意味着，如果你知道宇宙中所有粒子——不仅仅是一个粒子——的位置和速度，并且你知道物理定律，还拥有无限的计算能力——所有这些现在都不可信，但我们正在想象——那么物理定律就能决定接下来会发生什么，在下一个时刻，再下一个时刻，直至无限远的未来，甚至也能无限回溯到过去。因此，根据经典力学，你可以选取宇宙历史上的任何一个时刻，该时刻所发生事件中包含的信息足以确定历史上所有其他时刻将会发生什么。拉普拉斯对这些事情颇具想象力，他用一个比喻来描述。他说：“想象一个巨大的智能体。”后来，评论家们称之为拉普拉斯妖 (Laplace's demon: 一个假想的智能体，如果它能知晓宇宙中所有粒子在某一时刻的位置和动量，就能预测宇宙的未来和回溯过去)。他自己没有这样称呼，他是一位著名的无神论者，不喜欢谈论魔鬼。但拉普拉斯对这个能知晓宇宙在任何一个时刻一切的“恶魔”，这个巨大的智能体说，过去和未来对它来说是一本打开的书。你将无所不知，因为现在发生的一切决定了整个时空。这个想法，即如果你知道现在正在发生什么，物理定律原则上就能精确无误地决定接下来会发生什么。因此，这被称为发条宇宙范式 (Clockwork Universe Paradigm: 一种哲学观点，认为宇宙像一个巨大的机械装置，其所有事件都由物理定律预先决定)。宇宙永远按照物理定律完美地运转。

决定论、自由意志与兼容论

Now, this might bother you a little bit if you think, "Wait a minute, I'm a person, I'm a human being, I have the ability to make choices. I'm not determined by the laws of physics." Both scientists and philosophers have thought about that, and they still don't agree on what the right way to think about it is. But the favorite way to think about it is the following: In principle, if you knew exactly everything that was going on in the universe, you could predict the future. Now, classical mechanics isn't quite right. Eventually, we're going to talk about quantum mechanics, so that's another thing you have to keep in mind. But to the approximation that classical mechanics is good, you are determined in what is going to happen. But guess what? You don't know all the positions of all the atoms and all the molecules that make up you. Indeed, you literally cannot know them because the memory storage capacity to know all that would be at least as big as your brain, if not bigger. And if you made your brain bigger, now you just have more molecules to keep track of. It is impossible to actually have a real Laplace's demon in the universe. It's just a thought experiment to make vivid the implications of determinism. So philosophers have settled on what they decided to call compatibilism (兼容论: 一种哲学立场，认为自由意志与决定论是相互兼容的) in the sense that, on the one hand, the deep-down microscopic laws of physics are perfectly deterministic—or they're not if you're in quantum mechanics, but they're pretty deterministic anyway. But since you don't know it, you should be asking yourself, "What is the best I can do? What is the best way that I can try to understand human beings given the vastly incomplete information I have?" I know about my friends' personalities, their predilections, and their traits, but I don't know every neuron in their brain. And under those circumstances, you will model, you will think about a fellow human being or about yourself as an agent capable of making choices. Everyone does that, and that's the right thing to do because you are not Laplace's demon.

现在，如果你想，“等等，我是一个人，一个人类，我有能力做出选择。我不是由物理定律决定的。”这可能会让你有点困扰。科学家和哲学家都思考过这个问题，但他们仍然未能就如何正确看待它达成一致。但最受欢迎的思考方式是这样的：原则上，如果你确切地知道宇宙中发生的一切，你就能预测未来。现在，经典力学并非完全正确。最终，我们还会讨论量子力学，所以那是你需要记住的另一件事。但就经典力学的良好近似而言，你所发生的一切是被决定的。但你猜怎么着？你不知道构成你的所有原子和分子的所有位置。事实上，你根本不可能知道它们，因为存储所有这些信息的记忆容量至少会和你的大脑一样大，甚至更大。如果你把大脑做得更大，那么你只是有了更多的分子需要追踪。宇宙中不可能真正存在一个真实的拉普拉斯妖。它只是一个思想实验，旨在生动地说明决定论的含义。因此，哲学家们达成了一种共识，他们称之为兼容论 (Compatibilism: 一种哲学立场，认为自由意志与决定论是相互兼容的)，其含义是，一方面，深层次的微观物理定律是完全决定性的——或者如果你处于量子力学中则不是，但无论如何它们都相当具有决定性。但由于你并不知道这些，你应该问自己：“我能做到最好的是什么？在拥有大量不完整信息的情况下，我能以何种最佳方式去理解人类？”我了解我朋友的个性、他们的倾向和特质，但我不知道他们大脑中的每一个神经元。在这种情况下，你会将一个同类或你自己建模，将其视为一个能够做出选择的行动者。每个人都这样做，这是正确的做法，因为你不是拉普拉斯妖。