Nicolas Perony: Puppies! Now that I've got your attention, complexity theory

譯者: Hao Li 審譯者: Becky Ma

Science,

科學，

science has allowed us to know so much

科學讓我們了解了非常多

about the far reaches of the universe,

關於浩瀚宇宙的深奧知識。

which is at the same time tremendously important

與此同時，宇宙是又是極為的重要

and extremely remote,

並且極為的遙遠，

and yet much, much closer,

然而同時， 它又比很多我們並不明白的事物

much more directly related to us,

離我們更近，

there are many things we don't really understand.

跟我們更加直接相關。

And one of them is the extraordinary

我們不明白的事情之一

social complexity of the animals around us,

就是我們周圍這些動物的社會複合性。

and today I want to tell you a few stories

今天我想給你們講幾個

of animal complexity.

關於動物的複合性的故事。

But first, what do we call complexity?

首先，什麼被我們稱為複合性呢？

What is complex?

複合是什麼？

Well, complex is not complicated.

複合並不是複雜。

Something complicated comprises many small parts,

複雜的事物是由很多小部分所組成的，

all different, and each of them

每一部分都各不相同，而且每一部分都

has its own precise role in the machinery.

在這個體系中有其自身的確切作用。

On the opposite, a complex system

與之相反，這是一個複合的系統

is made of many, many similar parts,

由很多很多類似的部件所組成，

and it is their interaction

而且他們之間的相互影響

that produces a globally coherent behavior.

才形成了一種宏觀上一致的行為。

Complex systems have many interacting parts

複合系統含有很多相互動的的元素，

which behave according to simple, individual rules,

它們根據簡單的、 個體的規則行動，

and this results in emergent properties.

這就導致了新特徵的出現。

The behavior of the system as a whole

系統作為一個整體的行為

cannot be predicted

是無法僅僅根據

from the individual rules only.

個體規則推測出來的。

As Aristotle wrote,

正如亞里斯多德寫道：

the whole is greater than the sum of its parts.

「整體大於其各部分的總和。」

But from Aristotle, let's move onto

但是讓我們從亞里斯多德轉到

a more concrete example of complex systems.

複合系統的一個更具體的例子吧。

These are Scottish terriers.

這裡有幾條蘇格蘭梗犬。

In the beginning, the system is disorganized.

開始時，該系統是混亂無章的。

Then comes a perturbation: milk.

現在來點干擾： 牛奶。

Every individual starts pushing in one direction

每個個體開始向同一個方向移動，

and this is what happens.

然後就會變成這樣。

The pinwheel is an emergent property

風車式的運動是一個新出現的特徵，

of the interactions between puppies

它源於小狗們之間的相互作用，

whose only rule is to try to keep access to the milk

小狗們唯一的規則是盡量搆得著牛奶，

and therefore to push in a random direction.

因此牠們就向一個隨機的方向移動。

So it's all about finding the simple rules

所以就是要找到，

from which complexity emerges.

從簡單規則中展現出的複合性。

I call this simplifying complexity,

我把這稱之為簡化複合性，

and it's what we do at the chair of systems design

這就是我們在蘇黎世聯邦理工學院系統設計院

at ETH Zurich.

所做的事。

We collect data on animal populations,

我們收集關於動物族群的數據，

analyze complex patterns, try to explain them.

分析複合模式，並且嘗試去解釋它們。

It requires physicists who work with biologists,

這需要物理學家、生物學家、

with mathematicians and computer scientists,

還有數學家和計算機科學家共同合作，

and it is their interaction that produces

運用跨學科領域的研究能力，

cross-boundary competence

to solve these problems.

來解決這些問題。

So again, the whole is greater

再一次，這樣一來整體是

than the sum of the parts.

比各個部分的總和要大。

In a way, collaboration

從某種程度上說

合作是複合系統的另一個例子。

is another example of a complex system.

And you may be asking yourself

你可能會問自己，

which side I'm on, biology or physics?

我是在哪一個學科分支上面呢， 生物學家還是物理學家呢？

In fact, it's a little different,

事實上，都不是。

and to explain, I need to tell you

要解釋這一點，我得告訴你

a short story about myself.

關於我自己的一個小故事。

When I was a child,

當我還是個孩子的時候，

I loved to build stuff, to create complicated machines.

我喜歡打造東西，創建複雜的機器。

So I set out to study electrical engineering

所以我決定去學電器工程

and robotics,

和機器人工程。

and my end-of-studies project

我的畢業設計

was about building a robot called ER-1 --

是造一個名叫 ER-1 的機器人，

it looked like this—

看起來就像這樣。

that would collect information from its environment

它可以從其環境中收集信息，

and proceed to follow a white line on the ground.

然後沿著地面上的白線前進。

It was very, very complicated,

這個機器人非常非常複雜，

but it worked beautifully in our test room,

但它在我們的測試室運作的很好。

and on demo day, professors had assembled to grade the project.

到了演示的那天，教授們聚集在一起要給我們的項目打分。

So we took ER-1 to the evaluation room.

於是我們把 ER-1 帶到評估室。

It turned out, the light in that room

結果，那間房間裡的燈光

was slightly different.

稍有不同，

The robot's vision system got confused.

結果機器人的視覺系統就被弄糊塗了。

At the first bend in the line,

在白線第一個轉彎的地方，

it left its course, and crashed into a wall.

它就偏離了路線，然後撞到了牆上。

We had spent weeks building it,

我們花了好幾個星期來打造它，

and all it took to destroy it

結果使它毀於一旦的

was a subtle change in the color of the light

竟然是房間裡燈光顏色的

in the room.

微妙的變化。

That's when I realized that

這使我意識到，

the more complicated you make a machine,

你把一台機器造得越複雜，

the more likely that it will fail

它就越​​有可能會由於

due to something absolutely unexpected.

某個絕對意想到不的因素而失敗。

And I decided that, in fact,

於是我決定，其實，

I didn't really want to create complicated stuff.

我並不想創建複雜的東西。

I wanted to understand complexity,

我想要了解的是複合性，

the complexity of the world around us

我們周圍的世界的複合性，

and especially in the animal kingdom.

尤其在動物王國裡。

Which brings us to bats.

這就得說到蝙蝠。

Bechstein's bats are a common species of European bats.

彼氏鼠耳蝠是一種常見的歐洲蝙蝠。

They are very social animals.

他們是非常具有社會性的動物。

Mostly they roost, or sleep, together.

通常他們棲息或睡在一起，

And they live in maternity colonies,

而且他們生活在母系的群落裡。

which means that every spring,

這就意味著每年春天，

the females meet after the winter hibernation,

雌蝙蝠在冬眠後聚在一起，

and they stay together for about six months

然後一起生活大約六個月

to rear their young,

來養育後代。

and they all carry a very small chip,

他們身上都攜帶了很小的芯片，

which means that every time one of them

這樣每次牠們其中之一

enters one of these specially equipped bat boxes,

進入一個這種具有特殊裝備的蝙蝠盒，

we know where she is,

我們就知道她在哪裡

and more importantly,

更重要的是，

we know with whom she is.

我們知道她和誰在一起。

So I study roosting associations in bats,

所以我研究的是蝙蝠的棲息聯繫，

and this is what it looks like.

看起來就像這樣：

During the day, the bats roost

白天的時候，蝙蝠

in a number of sub-groups in different boxes.

分成很多小組棲息在不同的蝙蝠盒裡。

It could be that on one day,

有可能某一天，

the colony is split between two boxes,

整個群落分在兩個盒子裡，

but on another day,

但是另一天，

it could be together in a single box,

牠們可能一起聚在一個盒子裡，

or split between three or more boxes,

或者分到三個或或更多個盒子裡，

and that all seems rather erratic, really.

這一切看起來確實相當混亂。

It's called fission-fusion dynamics,

這被稱為裂變融合動態，

the property for an animal group

指一群動物

of regularly splitting and merging

定期拆分和合併

into different subgroups.

到不同的子組裡的特性。

So what we do is take all these data

所以我們做的是收集來自不同日子的

from all these different days

所有這些數據，

and pool them together

並把它們集中在一起

to extract a long-term association pattern

來提取一個長期的關聯模式。

by applying techniques with network analysis

通過應用技術與網絡分析，

to get a complete picture

我們可以獲得一份

of the social structure of the colony.

關於整個群落社會結構的完整圖片。

Okay? So that's what this picture looks like.

對吧？這張圖片看起來是這樣子。

In this network, all the circles

在這個網絡中，所有的圈

are nodes, individual bats,

都是每一個節點，是一個蝙蝠個體，

and the lines between them

而它們之間的連線

are social bonds, associations between individuals.

是社會紐帶，是個體之間的關聯。

It turns out this is a very interesting picture.

事實證明這是張非常有趣的圖片。

This bat colony is organized

整個蝙蝠群

in two different communities

被分成兩個不同的社區，

which cannot be predicted

這是從日常的裂變未來動態中預測來的。

from the daily fission-fusion dynamics.

We call them cryptic social units.

我們把它稱為隱性社會單位。

Even more interesting, in fact:

更有趣的是，

Every year, around October,

每年 10 月前後，

the colony splits up,

種群分散開了，

and all bats hibernate separately,

所有蝙蝠都分開冬眠。

but year after year,

但是年復一年，

when the bats come together again in the spring,

當蝙蝠在春天再一次聚在一起的時候，

the communities stay the same.

兩個群體保持不變。

So these bats remember their friends

所以這些蝙蝠可以記住他們的朋友

for a really long time.

並且記住很長時間。

With a brain the size of a peanut,

雖然，蝙蝠的大腦只有花生粒大小，

they maintain individualized,

他們卻保持個性化、

long-term social bonds,

長期的社會紐帶，

We didn't know that was possible.

在此之前，我們認為這是不可能的。

We knew that primates

我們知道靈長類動物，

and elephants and dolphins could do that,

還有大象和海豚能做到這一點，

but compared to bats, they have huge brains.

但與蝙蝠相比，牠們的大腦大多了。

So how could it be

所以，蝙蝠的大腦怎麼可能

that the bats maintain this complex,

會讓蝙蝠保持這個複合的、

stable social structure

穩定的社會結構，

with such limited cognitive abilities?

而又在如此有限的認知能力的情況下呢？

And this is where complexity brings an answer.

在這裡複合性就給予了一個解答。

To understand this system,

要了解這個系統，

we built a computer model of roosting,

我們建立了一個棲息的計算機模型。

based on simple, individual rules,

基於簡單的、 個體的規則，

and simulated thousands and thousands of days

然後虛擬的蝙蝠群裡模擬

in the virtual bat colony.

成千上萬的日子。

It's a mathematical model,

這是一個數學模型，

but it's not complicated.

但它並不複雜。

What the model told us is that, in a nutshell,

簡而言之，該模型告訴我們的是，

each bat knows a few other colony members

每個蝙蝠認識種群少數的其他幾個成員，

as her friends, and is just slightly more likely

把牠們當作朋友，而且只是稍微更有可能

to roost in a box with them.

在與牠們棲息在同一個盒子裡。

Simple, individual rules.

簡單的、個體的規則。

This is all it takes to explain

這些就足以解釋

the social complexity of these bats.

這些蝙蝠的社會複合性。

But it gets better.

還有更有趣的呢。

Between 2010 and 2011,

2010 年和 2011 年之間，

the colony lost more than two thirds of its members,

種群失​​去了三分之二的成員，

probably due to the very cold winter.

可能因為那年寒冷的冬天。

The next spring, it didn't form two communities

第二年春天，沒有像每年那樣

like every year,

形成兩個社群，

which may have led the whole colony to die

那樣就可能會導致整個種群的死亡，

because it had become too small.

因為它變得太小了。

Instead, it formed a single, cohesive social unit,

取而代之的是，它形成了一個單一、 緊密的社會單位，

which allowed the colony to survive that season

這就使種群在那一季度中倖存了下來，

and thrive again in the next two years.

並在隨後的兩年裡再次蓬勃發展。

What we know is that the bats

我們知道那些蝙蝠

are not aware that their colony is doing this.

並不知道牠們的種群在這樣做。

All they do is follow simple association rules,

牠們所做的只是遵循簡單的關聯規則，

and from this simplicity

而從這簡單性中，

emerges social complexity

體現出了社會複合性

which allows the colony to be resilient

而這使得種群重新振作，

against dramatic changes in the population structure.

同時，抵禦群體結構的巨大變化。

And I find this incredible.

我覺得這令人難以置信。

Now I want to tell you another story,

現在我想告訴你另一個故事，

but for this we have to travel from Europe

但我們必須從歐洲

to the Kalahari Desert in South Africa.

到南非的卡拉哈里沙漠去。

This is where meerkats live.

貓鼬就住在那裡。

I'm sure you know meerkats.

我確信你們知道貓鼬。

They're fascinating creatures.

牠們是很有趣的生物。

They live in groups with a very strict social hierarchy.

牠們生活在具有非常嚴格的社會等級制度的群體裡。

There is one dominant pair,

群體裡有一對地位最高，

and many subordinates,

還有很多下屬，

some acting as sentinels,

有的充當哨兵，

some acting as babysitters,

有的充當保姆，

some teaching pups, and so on.

還有的教習幼兒，等等。

What we do is put very small GPS collars

我們所做的就是把很小的 GPS 項圈

on these animals

放在這些動物身上

to study how they move together,

來研究他們是如何一起行動的，

and what this has to do with their social structure.

同時研究這與牠們社會結構的關係。

And there's a very interesting example

這裡有一個關於貓鼬的集體活動行為

of collective movement in meerkats.

非常有趣的例子。

In the middle of the reserve which they live in

在牠們居住的保護區的中央

lies a road.

有一條馬路。

On this road there are cars, so it's dangerous.

這條路上有車，所以是很危險的。

But the meerkats have to cross it

但貓鼬們必須穿過去，

to get from one feeding place to another.

才能從一個獵食地到達下一個。

So we asked, how exactly do they do this?

我們想，牠們到底是怎麼做到的呢？

We found that the dominant female

我們發現，地位最高的雌貓鼬，

is mostly the one who leads the group to the road,

多數時候領著整個群體來到馬路邊，

but when it comes to crossing it, crossing the road,

但是到要橫穿馬路的時候，

she gives way to the subordinates,

她就讓路給她的下屬，

a manner of saying,

似乎用這種方式在說：

"Go ahead, tell me if it's safe."

「去吧，告訴我是不是安全。」

What I didn't know, in fact,

事實上，我不知道

was what rules in their behavior the meerkats follow

這些貓鼬遵循了什麼樣的行為規則

for this change at the edge of the group to happen

可以使群體的邊緣發生這樣的變化。

and if simple rules were sufficient to explain it.

是不是用簡單的規則就足以解釋它。

So I built a model, a model of simulated meerkats

因此，我建了一個模型。

一個模擬貓鼬橫穿模擬馬路的模型。

crossing a simulated road.

It's a simplistic model.

這是一個非常簡化的模型。

Moving meerkats are like random particles

正在移動的貓鼬就像隨機質點一樣，

whose unique rule is one of alignment.

其唯一的規則就是排成一列。

They simply move together.

牠們只是一起行動。

When these particles get to the road,

當這些粒子到達馬路時，

they sense some kind of obstacle,

牠們感覺到某種障礙，

and they bounce against it.

然後牠們撞了上去。

The only difference

那隻地位最高的雌貓鼬，

between the dominant female, here in red,

這里紅色的這只，

and the other individuals,

和其他個體的唯一差別，

is that for her, the height of the obstacle,

就是對牠來說障礙物的高度，

which is in fact the risk perceived from the road,

也就是察覺到馬路上的風險，

is just slightly higher,

稍稍大了一點兒。

and this tiny difference

這個個體運動規則中的

in the individual's rule of movement

小小的區別

is sufficient to explain what we observe,

足以解釋我們所觀察到的現象，

that the dominant female

也就是地位最高的雌貓鼬，

leads her group to the road

領著她的群體到馬路邊，

and then gives way to the others

然後讓路給其他貓鼬，

for them to cross first.

讓牠們先過馬路。

George Box, who was an English statistician,

喬治·鮑克斯是一個英國的統計學家，

once wrote, "All models are false,

曾經寫道：「所有的模型是虛構的，

but some models are useful."

但有些模型是有用的。」

And in fact, this model is obviously false,

事實上，這個模型顯然是假的，

because in reality, meerkats are anything but random particles.

因為在現實中貓鼬不是什麼隨機粒子。

But it's also useful,

但是這個模型也挺有用的，

because it tells us that extreme simplicity

因為它告訴我們這種極端簡單、

in movement rules at the individual level

處於個體層面的運動規則，

can result in a great deal of complexity

到了群體的層面上

at the level of the group.

就可以導致很大的複合性。

So again, that's simplifying complexity.

再說一次，這就是簡化複合性。

I would like to conclude

最後我要說的是

on what this means for the whole species.

這對整個物種意味著什麼。

When the dominant female

當地位最高的雌貓鼬

gives way to a subordinate,

給下屬讓路，

it's not out of courtesy.

牠不是出於禮貌。

In fact, the dominant female

事實上，地位最高的雌貓鼬

is extremely important for the cohesion of the group.

對這個群體的凝聚力來說非常重要。

If she dies on the road, the whole group is at risk.

如果她死在馬路上，​​整個集團將處於危險之中。

So this behavior of risk avoidance

所以這種風險規避的行為

is a very old evolutionary response.

是一個非常古老的進化反應。

These meerkats are replicating an evolved tactic

這些貓鼬只是重複進化而來的策略。

that is thousands of generations old,

這個策略已經歷經成千上萬代了，

and they're adapting it to a modern risk,

牠們只是用它來抵禦新風險，

in this case a road built by humans.

在這裡新風險就是人類所建的一條路。

They adapt very simple rules,

它們改進了非常簡單的規則，

and the resulting complex behavior

而由此產生的複合行為

allows them to resist human encroachment

使牠們可以抵禦人類

into their natural habitat.

對於牠們自然棲息地的入侵。

In the end,

最後，

it may be bats which change their social structure

這也許是蝙蝠通過改變牠們的社會結構

in response to a population crash,

去應對群體的災難；

or it may be meerkats

也許是貓鼬

who show a novel adaptation to a human road,

對人造馬路表現出的新適應方式；

or it may be another species.

或者也可能是另一個物種。

My message here -- and it's not a complicated one,

我想傳達的不是一個複雜的問題，

but a simple one of wonder and hope --

而是一個簡單驚嘆與希望。

my message here is that animals

我想要表達的是

動物所顯示出非同尋常的社會複合性，

show extraordinary social complexity,

and this allows them to adapt

讓牠們能夠適應

and respond to changes in their environment.

並對牠們的環境變化作出回應。

In three words, in the animal kingdom,

用三個詞來說就是，在動物王國中，

simplicity leads to complexity

簡單性導致複合性，

which leads to resilience.

複合性又導致適應力。

Thank you.

謝謝。

(Applause)

（掌聲）

Dania Gerhardt: Thank you very much, Nicolas,

達尼亞格·哈特：尼古拉斯，非常感謝你

for this great start. Little bit nervous?

帶來這麼好的一個開場。有點緊張麼？

Nicolas Perony: I'm okay, thanks.

尼古拉斯·普萊尼：我還好，謝謝。

DG: Okay, great. I'm sure a lot of people in the audience

達尼亞·格哈特：好的，很好。我確信觀眾席中的許多人

somehow tried to make associations

試圖以某種方式給動物行為建立聯繫。

between the animals you were talking about --

正如你所談到的蝙蝠和貓鼬，甚至是人類。

the bats, meerkats -- and humans.

You brought some examples:

你帶來了幾個例子：

The females are the social ones,

雌性是上流社會中的個體，

the females are the dominant ones,

雌性還是有統治地位的個體，

I'm not sure who thinks how.

我不確定誰會去思考為什麼會這樣。

But is it okay to do these associations?

但是，在人與動物之間做這樣的關聯妥當麼？

Are there stereotypes you can confirm in this regard

你有沒有一些在這方面可以確定的理論？

that can be valid across all species?

是否這樣的關聯在所有的物種之間都是可行的呢？

NP: Well, I would say there are also

尼古拉斯·普萊尼： 我想說，確實也有

counter-examples to these stereotypes.

這些理論的反例。

For examples, in sea horses or in koalas, in fact,

舉個例子，在海馬或者無尾熊之中，其實，

it is the males who take care of the young always.

總是由雄性來照顧他們的幼仔的。

And the lesson is that it's often difficult,

我得到的經驗是，有時這是有些困難、

and sometimes even a bit dangerous,

並且時常有一點危險

to draw parallels between humans and animals.

把人類與動物進行類比。

So that's it.

這也就是我的回答了。

DG: Okay. Thank you very much for this great start.

達尼亞·格哈特：好的。感謝你如此精彩的開場。

Thank you, Nicolas Perony.

尼古拉斯·普萊尼，謝謝你。