Can you tell when a system is about to tip?

我希望你思考一下，一个系统的行为是如何发生变化的。

I want you to think about how the behavior of a system can change.

特别是，我希望你思考一下这种变化是如何非常迅速地发生的。

Particularly, want you to think about how this change can happen very quickly.

比如说，经济是如何突然且不可预测地从繁荣走向萧条的？

For example, how does an economy go from boom to bust very suddenly and unpredictably?

换句话说，它从一种行为模式突然转变到另一种。

In other words, it tips from one type of behavior to another.

那么，这些临界点是什么呢？

So what are these tipping points?

它们真的那么不可预测吗？

And are they really so unpredictable?

今天，你将听到来自加州大学默塞德分校的认知与信息科学助理教授泰勒·马格蒂斯的分享，他剖析了复杂系统行为发生根本性变化的深层原因，并提出了一个问题：你能判断一个系统即将发生突变吗？

Today, you'll hear from Tyler Marghetis, Assistant Professor of Cognitive and Information Sciences at the University of California Merced, where he pulls apart the underlying reasons why a complex system's behavior can radically change, And where he poses the question, can you tell when a system is about to tip?

这里是《简化复杂性》播客，我们在这里探讨复杂系统的基本原理。

This is Simplifying Complexity, a podcast where we explore the underlying principles of complex systems.

有些系统看起来违背了我们对世界的理性认知，比如经济、生态，甚至你和我。

Systems that seem to defy our rational view of the world, like economies, ecologies, or even you or me.

我是法证工程师肖恩·布雷迪，也是你们的主持人。

I'm forensic engineer Sean Brady, and I'll be your host.

大家好。

Hello, everyone.

欢迎收听本期节目，也欢迎泰勒来到节目。

Welcome to the show and welcome Tyler to the show.

肖恩，谢谢你邀请我。

Thanks for having me, Sean.

那么泰勒，我们要谈论复杂系统思维中的一个关键方面，那就是临界点。

So Tyler, we're going to talk about a key aspect of complex systems thinking, is tipping points.

临界点的例子是什么呢？

What's an example of a tipping point?

它们无处不在。比如我们每天从睡眠到清醒的过渡：你会从完全与外界隔绝的状态，突然进入感知世界的状态，可能迷迷糊糊，可能需要咖啡，但你的思维和存在方式，已经和片刻之前截然不同。

They show up everywhere from our daily transition between sleeping and being awake, where you go from completely disconnected from the world to suddenly engaged, perceiving it, perhaps groggy, perhaps in need of coffee, but in a way fundamentally transformed from how you were thinking and being just moments before.

所以从这些日常的例子，一直到经济崩溃，或者从熊市突然转到牛市，又或者政治革命、政治崩塌这些情况。

So it goes from that mundane example all the way up to cases of economic collapse or sudden switches from a bear market to a bull market, examples of political revolution or political collapse.

这在生物学和生态系统里也体现得很明显。

And it even shows up in biology and things like ecosystems.

就拿湖泊生态系统来说，有些湖原本清澈见底，到处都是游来游去的鱼，有时候却会突然变得很浑浊，水里长满了藻类，让水变得很浑浊。

Like lake ecosystems, you can have a lake that goes from beautiful, clear, dominated by fish that are swimming around around to sometimes very suddenly switching to a really murky so called turbid state where there's all this algae.

基本上什么都看不清了。

You can't really see much.

在这些例子里，都有一个本来显得很稳定的状态。

And in all those cases, you have one state which seemed pretty stable.

你睡了几个小时，甚至好几个小时。

You were sleeping for hours and hours.

完全没理由觉得这种情况会突然结束。

No reason to suspect that that might end.

或者你已经习惯了某个经济状态，觉得它挺稳定的。

Or you have economic regime that seems pretty stable.

一切进展顺利。

Things are going well.

或者你有一个这样的湖泊系统。

Or you have this lake system.

它看起来很好。

It looks great.

你知道，那些漂亮的鱼，清澈的水。

You know, all those nice fish, clear water.

然后突然之间，你遇到了一个转变。

And then somewhat suddenly, you get a switch.

你会经历一个剧烈的过渡，转向你可能不想要、或者可能想要、或者只是出乎意料的其他状态。

You get this tumbling transition to something else that you might not want or maybe you do want or maybe it's just unexpected.

但无论如何，它看起来和你之前看到的根本不同。

But either way, it seems fundamentally different from what you were seeing before.

这些就是临界点，有时也称为从一个状态到另一个状态的关键转变。

And those are tipping points or sometimes called critical transitions from one state to another.

那我们说的状态是什么意思呢？

And what do we mean by a state when we're talking about that?

其实状态这个词听起来很静态。

So the word state sounds really static.

对吧？

Right?

我一想到状态，就想到某种不变的东西。

It sort of when I think of a state, I think of something that's not changing.

但我们指的不是那个。

That's not what we mean here.

我们说的更像“动态模式”这个词，虽然听起来很科学，但我的意思就是系统以一种可靠、可预测且反复出现的方式运转。

We're talking more about I like the word dynamical regime, which, you know, sounds science y, but all I mean by it is that the system is operating in a reliable, perhaps predictable, and recurring way.

比如在经济领域，股市当然会有波动。

So in the case of the economy, sure, you're having fluctuations in the stock market.

但如果处于稳定模式或稳定状态，随着时间的推移，你会看到类似的动态。

But if it's a stable regime or stable state, you're gonna have similar dynamics over time.

所以也许股价大致上都在逐渐上涨。

So maybe stock prices are gradually increasing pretty much across the board.

说到睡眠，我们的大脑在睡眠期间其实非常活跃。

In the case of sleep, our brains are incredibly active during sleep.

我们在学习。

We're learning.

有时在练习。

We're sometimes practicing.

在想象。

We're imagining.

在做梦。

We're dreaming.

这些都在我们大脑里发生。

That's all happening inside of our brain.

所以虽然有很多变化发生，但它的运作方式在时间上基本上是一致的。

So there's lots of change happening, but it's working in more or less the same way over time.

所以当你谈论一个系统，尤其是复杂系统中的状态或体制时，它就像是一种运行模式，在一段时间内大致保持相同的动态变化。

And so when you talk about a state or regime in a system of any sort, in a complex system in particular, it's kind of like a mode of operation where more or less you have the same kind of dynamics happening over time.

这就是为什么对我们人类来说这么令人困惑，对吧？

This is why it's so confusing for us human beings, isn't it?

因为我们观察到这种状态是稳定且良好的，我们甚至能写出方程来描述它，然后我们就以为它会永远这样持续下去，但突然之间就不是了，不是吗？

Because we observe that state, it's nice and stable, we can even write equations to describe that state, and then we think it's going to go on like that forever and ever, and then suddenly it doesn't, does it?

它发生了突变。

It tips.

问题不在于它打乱了我们对世界那种线性的因果看法。

The issue here isn't that it screws around what our sort of linear cause and effect view of the world.

我的意思是，只要有一个观察者站在旁边，大家总会理所当然地认为事情只能是这样发展的。

I mean, everyone always assumes insofar as there is an observer standing by, you assume that this is the only way it's gonna be.

而且这理由也很充分。

And with good reason.

对吧？

Right?

通常我们都有这样的观察经验，觉得事情本来就是这样运作的。

Often you have a history of observation that this seems to be the way things work.

如果你看看那些伟大帝国或大型社会制度的历史，这一点就尤其明显了。

So this is especially true if you look at the history of great empires or great social systems.

我是说，真的很难想象在某一个时间点，也许人类、语言或者文化会完全消失，彻底从历史上抹去。

I mean, it's really difficult to imagine that at some point, maybe that people, that language, that culture might not exist at all, that it could be lost entirely to history.

作为人类，我们特别擅长预测。

As humans, we're really great predictors.

这也是我们的一种能力，就是不断地预测未来。

That's sort of part of our powers that we're constantly predicting the future.

有时候我们会判断失误，过度预测了。

Sometimes we misfire on that and we sort of over predict.

我们总以为眼前发生的事会一直持续下去。

We think that what's happening is gonna keep going.

而当你遇到那些非线性变化时，就像你说的，突然之间，我们的预期就被彻底打乱了。

And when you get those nonlinear changes, as you said, all of a sudden, it really screws with our expectations.

我喜欢那个冰块的例子。

I love the example of the ice cube.

你知道，零下五度的冰块，加一度变成零下四度，它还是冰块。

You know, you get the ice cube at minus five degrees and you add a degree and it goes to minus four and it's still an ice cube.

然后到零下三度，它还是冰块。

Then you go to minus three and it's still an ice cube.

到零下二度，它还是冰块。

And you go to minus two and it's still an ice cube.

你只是在线性地升高温度。

You're just linearly increasing the temperature.

但一旦超过零度，情况就变了。

But once you hit above zero, you change date.

从冰块变成一滩水，这让我们感到惊讶。

You go from that ice cube to the puddle of water, which is surprising to us.

不是因为我们对冰有多了解，而是因为我们通常不会指望其他系统，比如经济，也会这样。

Not because we understand ice, but in general that we don't expect that from other systems like the economy.

更棘手的是，有时候状态或体制已经变了，但表面上还看不出来。

And the trickier thing even is that sometimes you have a change in the regime, a change in the state, but it doesn't show up yet.

对吧？

Right?

这些变化需要时间才能显现出来，我们才能理解。

These things take time to show up in a way that we could understand.

就拿冰块那个例子来说。

So think of the ice cube case.

温度可能已经超过了临界点，就是从固态变成液态的那个点。

You can have the temperature above that critical point where you go from frozen to so from a solid state to a liquid state.

但整块冰并不会瞬间变成一滩水。

You don't have instantaneous transformation of the entire solid ice cube into a puddle.

它是慢慢发生的，逐渐变化的。

It happens fairly slowly and gradually.

但你不知道的是，在系统背后的那些主导方程里，其实已经翻过去了。

But unbeknownst to you, underlying that in the sort of governing equations of that system, it's already tipped over.

它已经进入了那个状态，没错，不会再变回固态冰了。

It's already in the regime in the state where, yeah, you're not gonna get solid ice anymore.

最终，所有的东西都会变成水。

Eventually, everything is gonna turn to water.

社会也是如此。

And that's true for societies too.

气候系统也是这样。

That's true for climate systems.

个人也一样。

That's true for individuals.

有时候我们已经越过了临界点，但自己还没意识到。

Sometimes we're already tipped over, but we don't even know it yet.

只是需要一点时间，我们能看到的东西才会跟上那个根本性的、近乎颠覆性的巨大变化——那些变化其实已经发生在系统内部，也就是驱动整个系统的引擎里。

It just takes a bit of time for the things that we can actually observe to catch up with the fundamental, almost transformational, radical change that's happened under the hood, so to speak, in the engine that's driving the system.

这其实是复杂系统的一个基本概念，对吧？我们知道它们是非线性的，也知道它们的行为方式非常难以预测。

This is a fundamental concept, isn't it, of complex systems that you're, you know, we know they're nonlinear, we know that they can behave in ways that are very, very hard to predict.

确实如此，就像你说的，这适用于各种不同的系统，但本质上它们都是一样的，因为我们有一群相互作用的适应性主体在各自运作，然后就出现了异常行为。

And this really holds true, as you say, for a whole range of different systems, but fundamentally they're all the same in the sense that we have groups of interacting adaptive agents doing their thing and you get unusual behaviour.

如果我们深入探讨，为什么会发生这种情况？

If we sort of put into it, why does this happen?

系统中导致这些临界点发生的潜在过程是什么？

What's the underlying process that's happening in the system that is causing these sorts of tips?

对。

Yeah.

回答这个问题有不同方式。

There's different ways of answering that.

对吧？

Right?

一种方式是用图像来说明。

And so one is with a picture.

经典的临界点图示就是杯中的球。

So the classic picture of these kinds of tipping points is the ball in a cup.

这个想法是想象一个平面，上面可能有两个凹陷。

The idea here is to imagine some sort of flat surface that maybe has two divots in it.

如果你把一颗弹珠之类的球放在那个平面上，它会掉进其中一个凹陷里。

And so if you were to drop a ball, say a marble onto that surface, it would fall into one of those divots.

它会卡在那里。

It would get stuck there.

即使你稍微晃一晃桌子让球来回弹跳，只要那个凹陷足够深，也就是山谷够深，它就会待在里面。你可以想象，你会觉得球本来就该这样运作。

And even if you're sort of shaking the table a bit so the ball is bouncing around, if that divot is deep enough, if you have a deep enough little valley, it'll stay in there, you can imagine that you might come to expect that that's just the way the ball works.

球会一直待在那个谷底，直到它越过边缘掉进旁边的另一个凹陷处，然后突然之间，它就到了桌面的另一部分，也就是表面的另一个区域。

It's always gonna be in that valley until it goes over the lip into the adjoining divot, and all of a sudden, it's in a different part of the table, different part of the surface.

这是很多复杂系统科学家用来理解状态或体制之间关键转变的直观图像。

That's the intuitive image that a lot of complex system scientists use to make sense of these kinds of critical transitions between states, between regimes.

所以这就是“杯子里的球”这个类比。

So this is the ball in a cup analogy.

你可以想象，其中一个凹陷代表牛市，另一个代表熊市；或者一个是睡眠状态，另一个是清醒状态。

You can imagine maybe, you know, one of those divots is a bull market and the other is a bear market, or one of those divots is sleeping, or the other one is awake.

那么随之而来的问题就是，为什么这类系统往往会表现出存在不同‘势阱’、能作为吸引子的状态。

The question then that that raises, of course, is why do these kinds of systems tend to behave as if there are these different basins that can act as attractors on their surface.

你提到过一个原因：当你有很多不同的微小组件相互耦合、彼此互动时，你对某个其他变量的连续线性变化可能会产生突然的响应。

And you pointed out one reason is that when you have a bunch of different small components all coupled up with each other and interacting, you can get this abrupt response to continuous linear change in some other variable.

所以，只要你有同质元素彼此连接，它们之间就能产生大量的正反馈，从而获得极强的韧性。

So whenever you have homogeneous elements that are connected with each other, so you can get a lot of positive feedback between them, you can get incredible resilience.

比如冰块的情况，只要温度低于零摄氏度，即使温度发生很大变化，它的状态其实并不会改变太多。

So that, you know, in the case of the ice cube, you're changing the temperature a lot as long as it's below zero Celsius, and it's not really changing very much.

同样地，你可以想象一个人站着的样子。

Similarly, you can think of someone who's standing.

对吧？

Right?

比如我站着，然后你推我一下。

So I'm standing up, and you're pushing me.

我会左右摇晃着保持平衡。

I'm sort of weaving back and forth.

也许风吹到我身上。

Maybe the wind blows on me.

我完全没事。

I'm like, totally great.

这是复杂系统具有韧性的一个绝佳例子。

That's a fantastic example of the kind of resilience that complex systems can have.

或者罗马帝国。

Or the Roman empire.

周围发生了那么多糟糕的事情，但它却持续了很长时间。

All these horrible things are happening around it, and it sort of stuck around for a really long time.

但突然之间，那些负责维持韧性的小部分都翻转到另一种状态，你就看到了灾难性的崩溃。

But then all of a sudden, all the little parts that are responsible for a lot of that resilience flip into the other state, and you get this catastrophic collapse.

对。

Yeah.

是什么在驱动这个过程？

What's driving that?

那是正反馈吗？

Is that positive feedback?

关键的变化因素是什么？

What's the key sort of thing that's changing there?

其中一个因素是正反馈。

One element there is positive feedback.

也就是说，你变化得越多，就越会促使更多的变化。

So that's where as you change more, that encourages more change.

所以一旦过了某个临界点，就会发生失控的转变。

So you get this runaway transition that happens once you go past a particular critical point.

不过并非所有系统都会这样。

So not all systems show this.

很多复杂系统可以很平稳地退化。

You have a lot of complex systems that can degrade quite gracefully.

你能再详细说说这个吗？

So can you expand on that a little bit more?

举个例子，人类在对话中就是这样。

So one example is humans in conversation.

对吧？

Right?

那是一个非常坚韧的系统。

That's a really resilient system.

即使背景音量增大，我们也能很好地适应，不会出现对话突然崩溃的那种情况。

You can increase the background volume, and we're really good at adapting to that so that you don't get this, like, sudden collapse where, like, conversations fall apart.

这个系统极其坚韧，背景噪音增加时只是稍微难一点，再难一点，但不会出现像湖泊生态系统那样突然从清澈状态翻转为浑浊状态的灾难性崩溃。

It's an incredibly resilient system that gets a little bit harder as you get background noise and a little bit harder and a little bit harder, but you don't get this really catastrophic collapse that you sometimes see with, like, a lake ecosystem, for instance, where it switches really suddenly from that nice clear state to a turbid state.

确实有些系统能够优雅地应对控制参数的变化、外部噪音、扰动或系统受到冲击。

So you do have some systems that are able to respond gracefully to some change in some control parameter or outside noise or perturbations or the system being shaken.

但也有另一些系统，当发生退化、温度变化、噪音增加或某个关键变量改变时，它看起来完全不在乎。

But you have other systems that as you have that breakdown, or the temperature is changing, or it's getting noisier, or some variable of interest is changing, it looks like the system doesn't care at all.

一切如常，直到轰然崩塌。

It's business as usual until, boom, it breaks.

我们大概知道哪些类型的系统会出现这种灾难性的崩溃，这跟它们内部的连接方式有关。

And we know a bit about which kinds of systems show that kind of catastrophic collapse, and it has to do with the way they're wired up on the inside.

所以，如果一个系统由大量相同的部件组成，也就是结构均质，而且所有部件都相互连接、彼此影响，那它看起来会非常稳固。

So if you have a system where it's made up of lots of parts that are the same, so they're homogeneous, and they're all connected with each other, so they're all influencing each other, those can look incredibly resilient.

你可以不断试探这类系统的极限，它似乎完全不受影响，直到你越过某个临界点。

You could push that kind of system around, and it looks like it just does not care what you're doing until you go too far.

这时候会发生什么呢？因为所有部件都一样，又互相作用，一旦有一个先崩了，接着第二个、第三个也跟着倒，突然之间就像雪崩一样，全部都被拖垮了。

And what happens is, because they're all the same, all the parts are the same, and they're all interacting with each other, When one flips and another one flips and a third, all of a sudden, it's sort of an avalanche, and they all get pulled over.

在高度互联的银行体系里，你也能看到这种现象。

You see this a bit with hyper connected banks.

二零零八年的那场金融危机就是例子，当时这些银行彼此之间的关联度极高。

This was the great financial crisis in two thousand and eight, is that all these banks were super connected with each other.

所以它们景气的时候，大家都跟着好。

So when they were doing well, they were all doing well together.

看起来一片大好，没人会想到会发生我们亲眼目睹的那种全球性经济危机，因为我们之前一直觉得这些银行能扛住股市的各种风浪，直到它们真的扛不住了为止。

It looked fantastic, and no one thought that you could have a global economic crisis on the scale that we saw because we had this history of seeing these banks being incredibly resilient to all the storms of the stock market until they weren't.

然后我们就看到了银行接连倒闭，像多米诺骨牌一样，直到今天还有影响。

And then we just had this collapsing domino of banks, and we have fallout from that today.

那是2008、2009年的事了，从美国开始，然后蔓延到全球。

So that was, you know, back in 2008, 2009, starting in The US and spreading around the world.

这就是一个例子，说明如果系统高度同质化，所有部分都相互连接，会发生什么。

That's an example of what happens if you have a really homogeneous system where all the parts are connected to each other.

看起来一段时间内很好，但总是存在灾难性临界点的危险，在那个情况下，我们真的不希望系统变成那样。

Looks great for a while, but there's always this danger of catastrophic tipping points into, in that case, a system that we really were not happy with.

它从那个稳定点翻转，但最终会回到另一个稳定点，大概就是我们现在的状态，它找到了某种平衡——用“平衡”这个词绝对不准确，但它重新找到了稳定。

It tips from that stable point, but eventually comes back to another stable point, which presumably where we are now that it sort of finds that equilibrium is absolutely the wrong word, but it finds that stability again.

它总得去某个地方。

It has to go somewhere.

很多这样的系统都有多个可以存在的状态。

And a lot of these systems will have multiple regimes that they can inhabit.

这叫做亚稳定性。

So that's called metastability.

对吧？

Right?

所以它不是完全锁死的，而是处于这个状态中。

So it's not completely locked in, but it's in this regime.

它大致上是稳定的。

It's more or less stable.

然后就像一个小球在一个碗里的图像，你可以有好多不同的谷底。

And then with this image of a ball in a cup, you could have a whole bunch of different valleys.

你不知道那些谷底在哪里。

You don't know where they are.

你不知道它们存在于什么地方。

You don't know where they exist.

但是如果你被弹出其中一个谷底，小球就会一直滚下去，直到找到下一个谷底，然后它就会停在那里。

But if you get knocked out of one, the ball's gonna roll down till it finds the bottom of the next valley, and it'll stay there.

所以，很可能经济现在就在另一个这样的谷底里呢，而且这种临界转变几乎在任何规模上都是这样的。

So presumably, the economy is in another one of those valleys, and that's gonna be true for just about any scale at which you get these kinds of critical transitions.

在人类尺度上，我最喜欢的例子之一是宗教信仰或政治立场的转变。

What you one of my favorite examples on a human scale is religious conversions or political conversions.

在这些情况下，人们极度投入于特定的信念体系、社会群体或行为模式，看起来似乎永远都不会改变。

These cases where people are hyper invested in a particular set of beliefs, in a particular social community, in a particular set of behaviors, and it looks like it's never gonna change.

从内部感受来看，确实也是这样。

It feels like that from the inside too.

然而，有时你会看到这种彻底的转变，比如有人从极右翼变成极左翼，或者反过来；从虔诚的基督徒变成公开的无神论者，反之亦然。

And yet, you can sometimes get these radical transformations where people go from being far right to far left or vice versa, from being a devout Christian to an avowed atheist and vice versa.

所以你说得对，当一个系统被踢出这些非常稳定的状态——也就是这些稳定的动力学状态时，它会非常努力地寻找其他可以稳定下来的地方。

And so you're right that when a system gets knocked out of one of these really stable states, one of these stable dynamical regimes, it works really hard to find somewhere else where it could stabilize.

会有一些明显的信号，对吧？

There's telltale signals, isn't there?

我们稍后的节目里会再详细聊聊这个。

And we're going to talk about this in a later episode in little more detail.

不过有几点你可以留意一下，它们确实能给你一些预警，至少对某些系统是有效的。

But there's a couple of things you can look out for that do give you some warnings, don't give some of the systems.

那你能讲讲临界减速吗？

So can you talk about critical slowing down?

那是什么？

What's that?

我想先说说韧性。

I guess I'll say a word about resilience first.

韧性就是系统承受冲击的能力。

So what it means for a system to be resilient, which is just the ability of a system to get knocked around.

也许它会被打乱一点，但只要有足够的时间，它就会回到正常运作的状态。

Maybe it gets knocked off course a bit, but if you give it enough time, it returns back to that standard operating state.

你可以想象一下宗教皈依的情况，比如某人原本信仰虔诚，但生活中发生了可怕的事情。

So you can imagine in the case of a religious conversion, maybe someone has a devout faith, Something horrible happened to their life.

这真的动摇了他们的根本。

It really sort of shakes them to their core.

他们可能开始产生怀疑。

They might start having doubts.

但过了几个月，他们又慢慢回到了信仰那个稳定的核心。

But given a few months, they sort of return to that stable core of their faith.

或者是一个经济体。

Or an economy.

对吧？

Right?

经济可能会受到一些冲击。

You might have some shock to the economy.

比如银行突然倒闭，或者某个地方爆发了战争。

Maybe a bank suddenly goes bankrupt or a war breaks out somewhere.

这会在经济系统中引发连锁反应，但只要给足时间，它还是会回到稳定的运行状态。

And that leads to ripples through the economic system, but given enough time, and it returns back to that stable operating state.

这就是一个韧性系统的例子。

So that's an example of a resilient system.

有意思的是，从科学角度来看，我们可以量化一个系统的韧性程度。

And the cool thing is, from a scientific perspective, is you can quantify how resilient a system is.

而且有几种不同的方法可以做到这一点。

And there's a couple different ways of doing this.

一种方法是，如果你能——通常我们做不到——但想象你从上方俯瞰，像上帝视角一样。

So one, if you can somehow and we normally can't do this, but imagine you have the view from above, you're sort of all seeing.

如果你能确定那个球在布满山谷和杯状凹陷的桌面上的位置，你就能测量出那个盆地的宽度。

If you could figure out where that ball is in that table with all the valleys, all that cups, you could, for instance, measure how wide is that basin.

一个非常非常宽的盆地意味着你可以把系统推离轨道很远，但它仍然会留在那个山谷里。

A really, really wide basin means that you can knock the system far off course, but it's still gonna be inside that valley.

给它足够的时间，它就会滚回来。

Give it enough time, it's gonna roll back.

或者你可以看看墙壁有多陡。

Or you could look at how steep the walls are.

这意味着你可以用力敲打它，它只会撞到墙壁然后弹回来。

That means you could really sort of knock it hard, it's just gonna bounce right off that wall and come back.

当然，对于大多数系统，我们无法从经济突然变成那种完美、全知的角度，知道球在布满山谷的桌子上的位置。

Now, of course, for most systems, we're not able to go from say, the economy to suddenly know in this like beautiful, perfect, omniscient way where the ball is on the table of valleys.

我的意思是，如果我能做到那一点，我早就成万亿富翁了。

I mean, if I could do that, I would be a trillionaire.

真的。

Truly.

可惜，我们还没有这样的工具。

Unfortunately, we don't have tools for that yet.

所以我们只能靠比较粗糙的工具，但有时候确实管用。

So we're left with pretty coarse tools, but they do sometimes work.

大家想出的一个办法，就是看一个系统被推离原轨之后，要多久才能回到原来那个稳定的运行状态。

And one of the tricks that people have settled upon is looking at how long it takes a system once it's been knocked off course to return back to that core stable operating dynamics.

这个就叫恢复率。

And that's the rate of return.

在韧性强、抗打击的系统中，这个恢复过程非常快。

And in a resilient system, that happens really quickly.

而在脆弱、不抗打击的系统中，就需要更长的时间才能恢复。

In an unresilient system, it takes longer for that to happen.

而在一个逐渐失去韧性的系统中，恢复速度就会慢下来。

And in a system that's gradually losing resilience, that slows down.

所以你会从一个被冲击后能飞快恢复的系统，变成一个恢复时间越来越长的系统。

So you go from a system that really really quickly returns from being knocked around to one that takes longer and longer and longer.

就是说，你可以想想人慢慢变老，再从受伤中恢复这样的情况。

I mean, you can think of people aging and recovering from an injury.

所以我们的身体就是超强韧性的系统。

So our bodies are hyper resilient systems.

我以前也当过国家队摔跤手。

I was a national team wrestler at some point.

那时候我经常把肩膀弄脱臼，脖子也扭伤。

I would completely tear my shoulder out of whack and crank my neck.

但到了第二天早上，我居然又能回到垫子上，跟那些想打我撂倒的壮汉们硬碰硬。

And by the next morning, somehow, I was back on the mat banging around with these mean ogres who were trying to beat me up.

现在可不敢这样了。

That is not the case for me anymore.

我已经三十好几了，要是睡姿不对，我得花两周才能恢复到最佳状态。

I'm well into my thirties, and if I sleep funny, it takes me two weeks to get back to optimal performance.

这就是一个典型的例子，说明韧性在逐渐下降。

That is a clear example of a gradual loss of resilience.

你不可能去经济体系或生态系统里，给它来一次重击，然后看它多久能恢复，因为那样不道德，我们不想去干扰经济。

You can't go into an economy or an ecosystem and knock it really hard and see how long it takes to come back, because that would be unethical because we don't wanna be messing with the economy.

所以人们发明了一些很酷的数学工具，专门用来观察系统自然波动的情况。

And so there are some cool mathematical tools that people have developed that just look at how the system is naturally bouncing around.

任何系统总会有一些噪音。

There's always gonna be some noise in a system.

你可以观察系统是如何自然变化的。

You could look at how the system is naturally varying.

然后根据这些数据，推断系统从微小波动中恢复的速度。

Try to use that to infer how quickly it's returning from small little fluctuations.

这就是一种通过观察系统自我稳定的速度来捕捉韧性的思路。

And that's this idea of trying to capture resilience by looking at how quickly the system is able to stabilize itself.

而且那个球回到桌面的时间越长，按照你那个球在桌上的比喻，我猜它只是越来越接近山谷的边缘，这意味着它离下一个山谷越来越近，它就会掉进去，变成完全不同的状态。

And the longer it takes for that ball, to come back to your analogy with the ball at the table to come back, I presume it's just getting further and further up the sides of the valley, which just means it's getting closer to the next valley over where it's gonna tip into and be quite different.

这基本上就是你想测量的东西吗？

Is that essentially what you're trying to measure in inverted commas?

所以发生这种情况的一个原因可能是，那个很深的盆地变得越来越浅了。

So one reason why that might be happening is that that really deep basin is becoming increasingly shallow.

想象一个球在一个非常非常深的碗里，你去敲它，它很快就会弹回来。

So imagine a ball that's in a really really deep bowl and you hit it, it's really gonna quickly fall back.

现在在同样大小的开口但平坦得多的盘子上做同样的事。

Now do the same thing on a plate with the same size opening, but it's much more flat.

你敲那个球，如果它还能滚回来，花的时间肯定比在很深很深的碗里要长得多。

You hit that ball, if it rolls back at all, it's gonna take considerably longer than you would see in a really, really deep ball.

而随着那个盆地越来越浅，那种状态也越来越难以维持。

And as that basin becomes increasingly shallow, that state becomes increasingly untenable.

它不稳定。

It's unstable.

在某些情况下，它可能会变得完全平坦，以至于那个状态对系统来说根本不再是一个选项。

And in some cases, it might become completely flat so that that state doesn't even exist anymore as an option for the system.

比如冰块那个例子，一旦温度超过某个值，水保持固态就不再是一个可行的选择了。

And that's what you have in some for instance, the ice cube case, once you get above a certain temperature, it just isn't an available option for the water to remain solid.

那如果有人跟我争论复杂系统思维和临界点的作用，他们可能会说，这有什么意义呢？

So what are the key sort of criticisms of, if I get into an argument with people about complex systems thinking and the role of tipping points is people go, what's the point?

对吧？

Right?

你有这些临界点，它们很难预测，你可能会进入一个不同的状态，更糟的是，你可能会进入几个不同的状态。

So you've got these tipping points, they're really hard to predict, you could tip into a different state, even worse than that, you could tip into several possible different states.

所以如果你是个普通人，想要理解这些，知道我的状态可能会在某个时候改变，这有什么意义呢？

So if you're a lay person trying to digest this, what's the point in knowing that, well, my state could change at some point.

我无法准确预测什么时候，也无法准确预测它可能会漂移到什么其他状态。

I can't predict exactly when, and I can't predict exactly what other state it could actually drift into.

我们不是预言家，但至少我们能很好地检测出韧性在下降，变化可能即将到来。

We're not prophets, but we can do a pretty good job, actually, at at least detecting that resilience is dropping, that a change might come.

而且，你可以决定采取干预措施。

And listen, then you could decide, for instance, to intervene.

如果你有证据表明某个系统正在失去韧性，你不需要被动地坐视不管。

If you have evidence that a system is losing resilience, you don't need to passively stand by and let that happen.

其实，我们在医学上就是这么做的。

I mean, this is what we do with medicine all the time.

对吧？

Right?

我们知道，随着人变老，他们的韧性会越来越差。

We know that as people get older, they become increasingly less resilient.

他们对受伤、生病的抵抗力会下降。

They lose resilience to injury, to illness.

那我们怎么办呢？

And so what do we do?

我们建议人们锻炼身体。

We tell people to exercise.

我们会建议人们多休息。

We tell people to rest.

我们会采取各种措施，来增强我们身体和免疫系统的恢复力。

We do all these things that we know increase the resilience of our bodies, of our immune systems.

所以，当我们发现某个系统正在失去这种恢复力时，这就应该成为一个采取行动的信号。

And so when we can tell that a system is losing resilience, that could be a call to action.

目前有迹象表明，我们的气候系统现在可能正面临这种情况。

There's some evidence that that might be happening to our climate system right now.

所以不稳定性在不断增加。

So you have increasing instability.

我们正经历越来越多的大规模极端天气事件。

We have these major weather events that are becoming more common.

人们把这些现象视为气候系统韧性下降的证据，甚至可能突然切换到另一种状态。

People have interpreted that as evidence of a climate system that's losing resilience, that might be able to actually switch into a new regime.

一想到可能存在某种未知的、也许更炎热或更干燥的新状态，就觉得挺可怕的。

That's really kind of scary to think that there's some unknown other regime, maybe much hotter, maybe much drier.

我们真的不知道。

We don't really know.

所以，这可能是一个行动号召，如果我们想维持当前的状态，就要尽一切努力增强这个状态的韧性。

And so that could be a call to action to try to do everything we can to increase the resilience of the state we're currently in if we want to maintain that state.

另一方面，这也为鼓励转变打开了可能性，如果我们希望发生转变的话。

On the other hand, this opens the possibility of encouraging a switch if we want one to happen.

所以你可以把创造力看作是一个人从一个状态切换到另一个状态，从一个模式切换到另一个模式的例子。

So you can think of cases of creativity as a case where someone is switching from one regime to another, from one state to another.

他们处于一种状态，对世界的理解可能和其他人一样，比较普通，然后他们可以切换到另一种状态，对世界有全新的、彻底不同的看法。

So they're they're in one state where they understand the world in maybe a really sort of mundane way that everyone else has, and then they could switch into another state where they have a radically reconfigured vision of the world.

他们可能是艺术家。

They're an artist.

可能是科学家。

They're a scientist.

也可能是试图治愈癌症的生物医学研究员。

They're a biomedical researcher trying to cure cancer.

如果我们能察觉到某人的信念或信仰正在失去稳定性，而且我们希望他们能转向一种更开阔、更扩展的世界观，那么我们就可以决定：现在是时候给他们一点推动力了。可以通过某种方式激励他们，或者让他们感到极度疲惫，从而进一步削弱其原有信念的稳定性，进而对这一系统施加干预，促使它跨越临界点，转向我们期望的、比之前更好的状态。

If we're able to tell that someone's convictions, someone's beliefs are losing resilience, and we want them to tip over into some new expanded way of understanding the world, all of a sudden, we can decide, okay, now is the moment to give them a bit of a bump, inspire them in some way, make them really exhausted so they further decrease the resilience, intervene in some way in the system to bump it over to something that we hope is gonna be better than where it was before.

非常感谢你跟我们分享关于状态和临界点的见解。

Thank you very much for talking to us about states and tipping points.

谢谢，肖恩。

Thanks, Sean.

这是我的荣幸。

My pleasure.

感谢收听《简化复杂性》，在这里我们将与来自世界各地的专家一起探讨复杂科学的核心概念。

Thanks for listening to Simplifying Complexity, where we look at the key concepts of complexity science with expert minds from across the world.

比如涌现、自组织、适应、网络、缩放、临界点等等，还有很多其他内容。

Concepts like emergence, self organization, adaptation, networks, scaling, tipping points, and much more.

本期播客由 Brady Heywood 和 Wavelength Creative 制作。

This podcast was produced by Brady Heywood and Wavelength Creative.

为了确保不错过任何一期节目，请在你的播客应用中订阅或关注本节目。

To make sure you don't miss an episode, be sure to subscribe to or follow the show in your podcast app.

我是肖恩·布雷迪，我们下期节目再见。

I'm Sean Brady, and I'll see you in our next episode.