TEDxUIUC
Christoph Adami: Finding life we can't imagine
克里斯 艾達米: 尋找難以想像的生命
Filmed:
Readability: 3.7
652,149 views
如果外星生物並不像我們所熟悉的生命形態，那麼我們要如何去尋找它們的蹤跡呢? 在TED於伊利諾大學主辦的演講中，克里斯艾達米將告訴大家他是如何透過人造生命 --- 即具備自我複製的電腦程式，來找到一個指標，一個超脫我們對生命所認知的生命指標。
Christoph Adami - Artificial life researcher
Christoph Adami works on the nature of life and evolution, trying to define life in a way that is as free as possible from our preconceptions. Full bio
Christoph Adami works on the nature of life and evolution, trying to define life in a way that is as free as possible from our preconceptions. Full bio
Double-click the English transcript below to play the video.
00:15
So I have a strange career.
0
0
2000
我的職業很奇怪。
00:17
I know it because people come up to me, like colleagues,
1
2000
3000
這麼講是因為常有人這樣告訴我，例如我的同事
00:20
and say, "Chris, you have a strange career."
2
5000
2000
都會說:「克里斯，你的職業很奇怪耶。」
00:22
(Laughter)
3
7000
2000
(觀眾笑聲)
00:24
And I can see their point,
4
9000
2000
其實我理解他們的意思,
00:26
because I started my career
5
11000
2000
因為我一開始當過
00:28
as a theoretical nuclear physicist.
6
13000
2000
理論核子物理學家。
00:30
And I was thinking about quarks and gluons
7
15000
2000
那時我成天想的都是夸克和膠子,
00:32
and heavy ion collisions,
8
17000
2000
還有重離子的撞擊,
00:34
and I was only 14 years old.
9
19000
2000
那時我才14歲而已。
00:36
No, no, I wasn't 14 years old.
10
21000
3000
不，不對，不是14歲那年的事。
00:40
But after that,
11
25000
2000
不過在那之後
00:42
I actually had my own lab
12
27000
2000
我有了一間專屬的實驗室,
00:44
in the computational neuroscience department,
13
29000
2000
就在計算神經科學系那邊,
00:46
and I wasn't doing any neuroscience.
14
31000
2000
但是我完全都沒有做神經科學的研究。
00:48
Later, I would work on evolutionary genetics,
15
33000
3000
後來我開始研究演化基因,
00:51
and I would work on systems biology.
16
36000
2000
接著便是系統生物學。
00:53
But I'm going to tell you about something else today.
17
38000
3000
不過以上這些跟我今天要講的主題一點關係也沒有。
00:56
I'm going to tell you
18
41000
2000
我要講的是
00:58
about how I learned something about life.
19
43000
2000
我如何探悉到生命的一些東西。
01:00
And I was actually a rocket scientist.
20
45000
4000
我以前當過火箭專家。
01:04
I wasn't really a rocket scientist,
21
49000
2000
但嚴格上來說我不算是真正的火箭專家,
01:06
but I was working
22
51000
2000
只不過我曾經在
01:08
at the Jet Propulsion Laboratory
23
53000
2000
位於陽光普照的加州的太空總署的
01:10
in sunny California where it's warm;
24
55000
3000
噴射推進實驗室工作過;
01:13
whereas now I'm in the mid-West,
25
58000
2000
而我現在在中西部,
01:15
and it's cold.
26
60000
2000
氣候很寒冷。
01:17
But it was an exciting experience.
27
62000
3000
不過這是一個很有趣的經驗。
01:20
One day a NASA manager comes into my office,
28
65000
3000
有一天NASA主管走進我的辦公室,
01:23
sits down and says,
29
68000
3000
坐下後說:
01:26
"Can you please tell us,
30
71000
2000
「請你告訴我們，
01:28
how do we look for life outside Earth?"
31
73000
2000
如何能尋找到外太空的生命?」
01:30
And that came as a surprise to me,
32
75000
2000
我當時很驚訝,
01:32
because I was actually hired
33
77000
2000
因為我當初是被請來
01:34
to work on quantum computation.
34
79000
2000
進行量子計算研究的。
01:36
Yet, I had a very good answer.
35
81000
2000
然而我回答得很漂亮。
01:38
I said, "I have no idea."
36
83000
3000
我答:「我一點也不知道。」
01:41
And he told me, "Biosignatures,
37
86000
3000
接著他對我說:「生命跡象,
01:44
we need to look for a biosignature."
38
89000
2000
我們必須找出生命跡象。」
01:46
And I said, "What is that?"
39
91000
2000
我問他:「那是什麼?」
01:48
And he said, "It's any measurable phenomenon
40
93000
2000
他說:「生命跡象就是能讓我們能
01:50
that allows us to indicate
41
95000
2000
辨識出任何可量化
01:52
the presence of life."
42
97000
2000
生命的存在的現象。」
01:54
And I said, "Really?
43
99000
2000
我說:「真的嗎?
01:56
Because isn't that easy?
44
101000
2000
真的有這麼簡單嗎?
01:58
I mean, we have life.
45
103000
2000
我是說，我們有生命。
02:00
Can't you apply a definition,
46
105000
2000
但你能為生命下一個
02:02
like for example, a Supreme Court-like definition of life?"
47
107000
4000
類似最高法院般的終極定義嗎?」
02:06
And then I thought about it a little bit, and I said,
48
111000
2000
我再想了想, 然後說,
02:08
"Well, is it really that easy?
49
113000
2000
「就只有這樣而已嗎?
02:10
Because, yes, if you see something like this,
50
115000
3000
沒錯，如果你看到這個,
02:13
then all right, fine, I'm going to call it life --
51
118000
2000
毫無疑問，我會稱它為生命--
02:15
no doubt about it.
52
120000
2000
這是無庸置疑的。
02:17
But here's something."
53
122000
2000
但如果換成這個。」
02:19
And he goes, "Right, that's life too. I know that."
54
124000
3000
他說:「沒錯，這個也是生命。我很確定。」
02:22
Except, if you think life is also defined
55
127000
2000
可是倘若你認為得生命
02:24
by things that die,
56
129000
2000
是由會死亡的物體來定義,
02:26
you're not in luck with this thing,
57
131000
2000
那你就無法解釋這個東西,
02:28
because that's actually a very strange organism.
58
133000
2000
因為這是一個相當奇怪的有機體。
02:30
It grows up into the adult stage like that
59
135000
2000
當它進入成年期的時候就像這樣,
02:32
and then goes through a Benjamin Button phase,
60
137000
3000
然後就像班傑明的奇幻旅程一樣
02:35
and actually goes backwards and backwards
61
140000
2000
不斷退化,
02:37
until it's like a little embryo again,
62
142000
2000
直到胚胎為止,
02:39
and then actually grows back up, and back down and back up -- sort of yo-yo --
63
144000
3000
接著又長回來，再長大 -- 像溜溜球一樣過程循環 --
02:42
and it never dies.
64
147000
2000
而永生不死。
02:44
So it's actually life,
65
149000
2000
這也算是生命的一種,
02:46
but it's actually not
66
151000
2000
只不過它不是
02:48
as we thought life would be.
67
153000
3000
我們一般所認知的型態。
02:51
And then you see something like that.
68
156000
2000
再來你如果看到這個。
02:53
And he was like, "My God, what kind of a life form is that?"
69
158000
2000
他問:「天啊，這到底是什麼樣的生命形態呢?」
02:55
Anyone know?
70
160000
2000
有人知道嗎?
02:57
It's actually not life, it's a crystal.
71
162000
3000
其實這不算是生命，這是一種結晶體。
03:00
So once you start looking and looking
72
165000
2000
所以當你觀察的東西
03:02
at smaller and smaller things --
73
167000
2000
越來越小時--
03:04
so this particular person
74
169000
2000
有位老兄
03:06
wrote a whole article and said, "Hey, these are bacteria."
75
171000
3000
花了一整篇文章的篇幅只為了傳達一件事:「嗨, 這是細菌。」
03:09
Except, if you look a little bit closer,
76
174000
2000
但如果你靠近一點觀察
03:11
you see, in fact, that this thing is way too small to be anything like that.
77
176000
3000
你會發現，事實上這個物體已經比細菌還要小。
03:14
So he was convinced,
78
179000
2000
於是他被說服了,
03:16
but, in fact, most people aren't.
79
181000
2000
可是大部分的人還是不相信。
03:18
And then, of course,
80
183000
2000
當然,
03:20
NASA also had a big announcement,
81
185000
2000
NASA做了一個重大的宣布,
03:22
and President Clinton gave a press conference,
82
187000
2000
此外前總統柯林頓也召開記者會,
03:24
about this amazing discovery
83
189000
2000
宣布在火星的隕石裡
03:26
of life in a Martian meteorite.
84
191000
3000
發現有生命的存在。
03:29
Except that nowadays, it's heavily disputed.
85
194000
4000
但是現今這個說法受到嚴重的質疑。
03:33
If you take the lesson of all these pictures,
86
198000
3000
如果你仔細地研究這些照片,
03:36
then you realize, well actually maybe it's not that easy.
87
201000
2000
就會發覺區別生命並沒有那麼簡單。
03:38
Maybe I do need
88
203000
2000
也許我需要
03:40
a definition of life
89
205000
2000
一個生命的定義
03:42
in order to make that kind of distinction.
90
207000
2000
才能夠來做區別。
03:44
So can life be defined?
91
209000
2000
生命能被定義嗎?
03:46
Well how would you go about it?
92
211000
2000
你會如何著手?
03:48
Well of course,
93
213000
2000
當然
03:50
you'd go to Encyclopedia Britannica and open at L.
94
215000
2000
你會從大英百科的L開始查起。
03:52
No, of course you don't do that; you put it somewhere in Google.
95
217000
3000
不，你當然不會那樣做; 你會用Google搜尋。
03:55
And then you might get something.
96
220000
3000
然後你或會得到一些資料。
03:58
And what you might get --
97
223000
2000
接著把你搜尋到的 --
04:00
and anything that actually refers to things that we are used to,
98
225000
2000
所有我們習以為常的觀念
04:02
you throw away.
99
227000
2000
拋諸腦後。
04:04
And then you might come up with something like this.
100
229000
2000
然後你可能會得到這段
04:06
And it says something complicated
101
231000
2000
複雜的解釋,
04:08
with lots and lots of concepts.
102
233000
2000
裡頭包括許許多多的概念。
04:10
Who on Earth would write something
103
235000
2000
到底有誰會寫出
04:12
as convoluted and complex
104
237000
2000
這麼人費解,複雜
04:14
and inane?
105
239000
3000
又空洞的東西?
04:17
Oh, it's actually a really, really, important set of concepts.
106
242000
4000
但是這段定義確實涵蓋了一堆非常重要的概念。
04:21
So I'm highlighting just a few words
107
246000
3000
我標出了幾個關鍵字眼,
04:24
and saying definitions like that
108
249000
2000
這類的定義
04:26
rely on things that are not based
109
251000
2000
不是基於
04:28
on amino acids or leaves
110
253000
3000
胺基酸或葉子
04:31
or anything that we are used to,
111
256000
2000
或者我們熟悉的東西,
04:33
but in fact on processes only.
112
258000
2000
而是只基於過程。
04:35
And if you take a look at that,
113
260000
2000
如果你仔細看這段話的出處,
04:37
this was actually in a book that I wrote that deals with artificial life.
114
262000
3000
就知道是從我寫的一本有關人造生命的書而來。
04:40
And that explains why
115
265000
2000
這說明了
04:42
that NASA manager was actually in my office to begin with.
116
267000
3000
那位NASA主管來辦公室找我的原因。
04:45
Because the idea was that, with concepts like that,
117
270000
3000
因為用這樣的想法與概念,
04:48
maybe we can actually manufacture
118
273000
2000
我們也許能創造
04:50
a form of life.
119
275000
2000
一個生命的形式。
04:52
And so if you go and ask yourself,
120
277000
3000
如果你反問自己
04:55
"What on Earth is artificial life?",
121
280000
2000
「到底什麼是人工生命?」
04:57
let me give you a whirlwind tour
122
282000
2000
就讓我快速地帶你認識
04:59
of how all this stuff came about.
123
284000
2000
人工生命的由來。
05:01
And it started out quite a while ago
124
286000
3000
這是好幾年前發生的,
05:04
when someone wrote
125
289000
2000
有人寫了早期史上
05:06
one of the first successful computer viruses.
126
291000
2000
上最具破壞力的電腦病毒。
05:08
And for those of you who aren't old enough,
127
293000
3000
對年紀較輕的觀眾來說,
05:11
you have no idea how this infection was working --
128
296000
3000
你們可能不清楚這種病毒是從哪裡散播開來的...
05:14
namely, through these floppy disks.
129
299000
2000
就是從這種磁碟片傳染的。
05:16
But the interesting thing about these computer virus infections
130
301000
3000
不過這種電腦中毒有趣的地方
05:19
was that, if you look at the rate
131
304000
2000
可以從電腦的
05:21
at which the infection worked,
132
306000
2000
感染速率來看,
05:23
they show this spiky behavior
133
308000
2000
這張圖表反映出的上下波動
05:25
that you're used to from a flu virus.
134
310000
3000
跟一般的流感病毒沒有兩樣。
05:28
And it is in fact due to this arms race
135
313000
2000
事實上因為駭客
05:30
between hackers and operating system designers
136
315000
3000
和作業系統開發人員之間發生的爭奪戰,
05:33
that things go back and forth.
137
318000
2000
而使結果反反復複。
05:35
And the result is kind of a tree of life
138
320000
2000
這張電腦病毒的關係圖
05:37
of these viruses,
139
322000
2000
便成樹狀展開,
05:39
a phylogeny that looks very much
140
324000
3000
一個看似我們熟悉的生命發展史,
05:42
like the type of life that we're used to, at least on the viral level.
141
327000
3000
至少從病毒的層面來看是如此。
05:45
So is that life? Not as far as I'm concerned.
142
330000
3000
病毒能算是生命嗎? 我可不這麼認為。
05:48
Why? Because these things don't evolve by themselves.
143
333000
3000
怎麼說呢? 因為它們無法自行演化。
05:51
In fact, they have hackers writing them.
144
336000
2000
事實上，電腦病毒是駭客寫出來的。
05:53
But the idea was taken very quickly a little bit further
145
338000
4000
但是這個想法不久就有了一點進展,
05:57
when a scientist working at the Scientific Institute decided,
146
342000
3000
有一個在科學研究所的科學家決定,
06:00
"Why don't we try to package these little viruses
147
345000
3000
「我們為何不把這些電腦病毒
06:03
in artificial worlds inside of the computer
148
348000
2000
放進電腦的虛擬世界,
06:05
and let them evolve?"
149
350000
2000
讓它們自行演化?」
06:07
And this was Steen Rasmussen.
150
352000
2000
這位科學家就是斯蒂恩•拉斯穆森。
06:09
And he designed this system, but it really didn't work,
151
354000
2000
他設計了這套系統，不過沒效,
06:11
because his viruses were constantly destroying each other.
152
356000
3000
因為他的病毒會不斷自相殘殺。
06:14
But there was another scientist who had been watching this, an ecologist.
153
359000
3000
但當時還有一位科學家對這件事情很關心，是一名生態學者。
06:17
And he went home and says, "I know how to fix this."
154
362000
3000
他回了家說:「我知道怎麼解決。」
06:20
And he wrote the Tierra system,
155
365000
2000
他寫出Tierra系統,
06:22
and, in my book, is in fact one of the first
156
367000
3000
根據我書裡寫的，Tierra正是最早出現的
06:25
truly artificial living systems --
157
370000
2000
人造生命系統之一--
06:27
except for the fact that these programs didn't really grow in complexity.
158
372000
3000
只不過這些程式沒有真正複雜性的成長。
06:30
So having seen this work, worked a little bit on this,
159
375000
3000
看過這個成果之後，我自己也做了一點研究,
06:33
this is where I came in.
160
378000
2000
而我的研究就從此展開。
06:35
And I decided to create a system
161
380000
2000
我決定創造一個系統,
06:37
that has all the properties that are necessary
162
382000
2000
該系統必須滿足
06:39
to see the evolution of complexity,
163
384000
3000
複雜演化的所有必要條件,
06:42
more and more complex problems constantly evolving.
164
387000
3000
有越來越多複雜的問題持續在演變。
06:45
And of course, since I really don't know how to write code, I had help in this.
165
390000
3000
當然，由於我不會編碼，所以我找了槍手。
06:48
I had two undergraduate students
166
393000
2000
我請到了兩位
06:50
at California Institute of Technology that worked with me.
167
395000
3000
在加州理工學院與我共事的大學生。
06:53
That's Charles Offria on the left, Titus Brown on the right.
168
398000
3000
左邊的是查爾斯•奧佛瑞亞，右邊這位是提多•布朗。
06:56
They are now actually respectable professors
169
401000
3000
他們如今都是在密西根州立大學
06:59
at Michigan State University,
170
404000
2000
備受尊崇的教授了,
07:01
but I can assure you, back in the day,
171
406000
2000
但我可以向你保證, 在當時
07:03
we were not a respectable team.
172
408000
2000
我們並不是可受尊敬的團隊。
07:05
And I'm really happy that no photo survives
173
410000
2000
我很慶幸我們三人形影不離的合照,
07:07
of the three of us anywhere close together.
174
412000
3000
一張都沒有留下。
07:10
But what is this system like?
175
415000
2000
這個系統是什麼樣子?
07:12
Well I can't really go into the details,
176
417000
3000
我不方便探討細節,
07:15
but what you see here is some of the entrails.
177
420000
2000
不過我可以給你們看一點內部的構造。
07:17
But what I wanted to focus on
178
422000
2000
我著重的是
07:19
is this type of population structure.
179
424000
2000
這種族群結構圖。
07:21
There's about 10,000 programs sitting here.
180
426000
3000
這裡大約有一萬個程式。
07:24
And all different strains are colored in different colors.
181
429000
3000
每個種類都用不同顏色來分類。
07:27
And as you see here, there are groups that are growing on top of each other,
182
432000
3000
你會發現族群間會相互掩蓋,
07:30
because they are spreading.
183
435000
2000
因為它們散播開來了。
07:32
Any time there is a program
184
437000
2000
不論何時都有一個程式
07:34
that's better at surviving in this world,
185
439000
2000
較能夠在虛擬世界中存活下來,
07:36
due to whatever mutation it has acquired,
186
441000
2000
因為經過突變的過程,
07:38
it is going to spread over the others and drive the others to extinction.
187
443000
3000
這個程式將會蓋過其它群體甚至把它們趕盡殺絕。
07:41
So I'm going to show you a movie where you're going to see that kind of dynamic.
188
446000
3000
在我接下來要放的影片裡你們可以觀察到這樣的變化。
07:44
And these kinds of experiments are started
189
449000
3000
這個實驗是從我們
07:47
with programs that we wrote ourselves.
190
452000
2000
自行開發的程式進行的。
07:49
We write our own stuff, replicate it,
191
454000
2000
我們寫了程式, 然後進行複製,
07:51
and are very proud of ourselves.
192
456000
2000
我們對此感到非常驕傲。
07:53
And we put them in, and what you see immediately
193
458000
3000
我們把程式放到系統裡,
07:56
is that there are waves and waves of innovation.
194
461000
3000
就成了你現在看到不斷變動的波形。
07:59
By the way, this is highly accelerated,
195
464000
2000
順便提一下，這段影片是加快播放,
08:01
so it's like a thousand generations a second.
196
466000
2000
所以變化的速度大約是一秒衍生一千次。
08:03
But immediately the system goes like,
197
468000
2000
很快系統就有了改變,
08:05
"What kind of dumb piece of code was this?
198
470000
2000
「這究竟是什麼蠢代碼呢?
08:07
This can be improved upon in so many ways
199
472000
2000
這可以藉由很多種方法
08:09
so quickly."
200
474000
2000
快速獲得改善。」
08:11
So you see waves of new types
201
476000
2000
你可以看到新種類
08:13
taking over the other types.
202
478000
2000
取代其它種類的過程。
08:15
And this type of activity goes on for quite awhile,
203
480000
3000
這樣子的過程會持續一段時間,
08:18
until the main easy things have been acquired by these programs.
204
483000
4000
直到這些程式把大多數單純的結構納入為止。
08:22
And then you see sort of like a stasis coming on
205
487000
4000
接下來系統會面臨停滯期,
08:26
where the system essentially waits
206
491000
2000
系統在等待一個
08:28
for a new type of innovation, like this one,
207
493000
3000
全新的轉變，就像這樣。
08:31
which is going to spread
208
496000
2000
它將會覆蓋
08:33
over all the other innovations that were before
209
498000
2000
先前所有的變化
08:35
and is erasing the genes that it had before,
210
500000
3000
並且消滅之前所有的基因,
08:38
until a new type of higher level of complexity has been achieved.
211
503000
4000
直到系統演化到更具複雜性的層面。
08:42
And this process goes on and on and on.
212
507000
3000
這個過程會不斷重複上演。
08:45
So what we see here
213
510000
2000
所以我們在這看到的
08:47
is a system that lives
214
512000
2000
就是一個與我們熟悉的
08:49
in very much the way we're used to life [going.]
215
514000
2000
生命形式雷同的系統。
08:51
But what the NASA people had asked me really
216
516000
4000
但NASA官員問我的是
08:55
was, "Do these guys
217
520000
2000
「這些玩意兒
08:57
have a biosignature?
218
522000
2000
有生命跡象嗎?」
08:59
Can we measure this type of life?
219
524000
2000
我們可不可以衡量這樣的生命形式?
09:01
Because if we can,
220
526000
2000
因為如果我們可以,
09:03
maybe we have a chance of actually discovering life somewhere else
221
528000
3000
也許我們就能以客觀角度
09:06
without being biased
222
531000
2000
證實其它星球有生命存在,
09:08
by things like amino acids."
223
533000
2000
而不需靠胺基酸來判別。」
09:10
So I said, "Well, perhaps we should construct
224
535000
3000
我說:「我們必須建立一個
09:13
a biosignature
225
538000
2000
生命跡象,
09:15
based on life as a universal process.
226
540000
3000
並假設所有生命都會經歷一個共通的過程。
09:18
In fact, it should perhaps make use
227
543000
2000
實際上，這必須應用我
09:20
of the concepts that I developed
228
545000
2000
開發的概念來達成,
09:22
just in order to sort of capture
229
547000
2000
得以了解
09:24
what a simple living system might be."
230
549000
2000
一個簡單的生命體系如何運作。」
09:26
And the thing I came up with --
231
551000
2000
為了解釋我想到的方法--
09:28
I have to first give you an introduction about the idea,
232
553000
4000
首先我得介紹一個概念,
09:32
and maybe that would be a meaning detector,
233
557000
3000
或許這概念是個存在探測器,
09:35
rather than a life detector.
234
560000
3000
而不是生命探測器。
09:38
And the way we would do that --
235
563000
2000
我們的做法就是--
09:40
I would like to find out how I can distinguish
236
565000
2000
先辨認出一段文字,
09:42
text that was written by a million monkeys,
237
567000
2000
是由一百萬隻猴子聯合寫出來的,
09:44
as opposed to text that [is] in our books.
238
569000
3000
還是從我們平常看的書籍中節錄出來的。
09:47
And I would like to do it in such a way
239
572000
2000
我會這樣處理,
09:49
that I don't actually have to be able to read the language,
240
574000
2000
我不需要看懂這段文字的語言,
09:51
because I'm sure I won't be able to.
241
576000
2000
因為我知道我根本辦不到。
09:53
As long as I know that there's some sort of alphabet.
242
578000
2000
但只要我可以認出其中有的是字母。
09:55
So here would be a frequency plot
243
580000
3000
這是一張次數分配圖,
09:58
of how often you find
244
583000
2000
告訴你在這段
10:00
each of the 26 letters of the alphabet
245
585000
2000
由猴群隨機寫出來的文字裡
10:02
in a text written by random monkeys.
246
587000
3000
其中26個字母出現的次數。
10:05
And obviously each of these letters
247
590000
2000
顯然這些字母
10:07
comes off about roughly equally frequent.
248
592000
2000
出現的頻率大約相等。
10:09
But if you now look at the same distribution in English texts,
249
594000
4000
但是如果你看到的是一段英文段落的字母次數分配,
10:13
it looks like that.
250
598000
2000
就會長成這樣。
10:15
And I'm telling you, this is very robust across English texts.
251
600000
3000
而且這種現象在英文裡非常明顯。
10:18
And if I look at French texts, it looks a little bit different,
252
603000
2000
如果是法語就會不太一樣,
10:20
or Italian or German.
253
605000
2000
甚至是義大利文或德文。
10:22
They all have their own type of frequency distribution,
254
607000
3000
各種語言都有獨特的次數分配模式,
10:25
but it's robust.
255
610000
2000
但是結果都很一致。
10:27
It doesn't matter whether it writes about politics or about science.
256
612000
3000
不管內容是有關政治或科學。
10:30
It doesn't matter whether it's a poem
257
615000
3000
還是一首詩,
10:33
or whether it's a mathematical text.
258
618000
3000
甚至是一段數學式子。
10:36
It's a robust signature,
259
621000
2000
都有一個明顯的特徵,
10:38
and it's very stable.
260
623000
2000
而且非常穩定。
10:40
As long as our books are written in English --
261
625000
2000
只要我們的書籍是用英文寫的--
10:42
because people are rewriting them and recopying them --
262
627000
3000
因為人們會不斷重寫並抄寫書籍--
10:45
it's going to be there.
263
630000
2000
就會產生這個特徵。
10:47
So that inspired me to think about,
264
632000
2000
這讓我想到
10:49
well, what if I try to use this idea
265
634000
3000
如果我用這個概念,
10:52
in order, not to detect random texts
266
637000
2000
不是為了要從有意義的文章中
10:54
from texts with meaning,
267
639000
2000
挑出雜亂無章的文字,
10:56
but rather detect the fact that there is meaning
268
641000
4000
而是探測
11:00
in the biomolecules that make up life.
269
645000
2000
形成生命體的生物分子特徵。
11:02
But first I have to ask:
270
647000
2000
但首先我有個問題:
11:04
what are these building blocks, like the alphabet, elements that I showed you?
271
649000
3000
這些組成的基本單位是什麼? 就像我剛給你們看的字母。
11:07
Well it turns out, we have many different alternatives
272
652000
3000
事實證明，我們有很多種選擇
11:10
for such a set of building blocks.
273
655000
2000
可用來做為構成的基礎。
11:12
We could use amino acids,
274
657000
2000
我們能利用胺基酸,
11:14
we could use nucleic acids, carboxylic acids, fatty acids.
275
659000
3000
核酸、羧酸或不飽和脂肪酸。
11:17
In fact, chemistry's extremely rich, and our body uses a lot of them.
276
662000
3000
事實上化學物質存在相當廣泛，我們人體就充滿許多化學物質。
11:20
So that we actually, to test this idea,
277
665000
3000
所以,為了試驗這個想法,
11:23
first took a look at amino acids and some other carboxylic acids.
278
668000
3000
我們研究了胺基酸和其他的羧酸。
11:26
And here's the result.
279
671000
2000
這就是結果。
11:28
Here is, in fact, what you get
280
673000
3000
事實上,
11:31
if you, for example, look at the distribution of amino acids
281
676000
3000
譬如, 如果你觀察一個彗星或星際空間,
11:34
on a comet or in interstellar space
282
679000
3000
或者一個實驗室裡
11:37
or, in fact, in a laboratory,
283
682000
2000
所充斥的胺基酸,
11:39
where you made very sure that in your primordial soup
284
684000
2000
但必須保證在原生湯裡
11:41
that there is not living stuff in there.
285
686000
2000
沒有任何生命存在。
11:43
What you find is mostly glycine and then alanine
286
688000
3000
你能找到的大部分是甘氨酸和丙氨酸,
11:46
and there's some trace elements of the other ones.
287
691000
3000
還有一些其它的元素。
11:49
That is also very robust --
288
694000
3000
這個結果也相當明顯--
11:52
what you find in systems like Earth
289
697000
3000
你可以在地球的生態系統裡
11:55
where there are amino acids,
290
700000
2000
找到胺基酸
11:57
but there is no life.
291
702000
2000
但是沒有生命。
11:59
But suppose you take some dirt
292
704000
2000
但假設你採集一些土壤
12:01
and dig through it
293
706000
2000
在裡面找尋一番
12:03
and then put it into these spectrometers,
294
708000
3000
放到光譜儀下,
12:06
because there's bacteria all over the place;
295
711000
2000
因為土壤佈滿了細菌;
12:08
or you take water anywhere on Earth,
296
713000
2000
或者是你取地球上任何一處的水,
12:10
because it's teaming with life,
297
715000
2000
因為水裡富含生命,
12:12
and you make the same analysis;
298
717000
2000
然後你做一樣的分析;
12:14
the spectrum looks completely different.
299
719000
2000
光譜結果會截然不同。
12:16
Of course, there is still glycine and alanine,
300
721000
4000
當然結果仍然含有甘氨酸和丙氨酸,
12:20
but in fact, there are these heavy elements, these heavy amino acids,
301
725000
3000
但是更重要的因素是大量的胺基酸,
12:23
that are being produced
302
728000
2000
因而大量產生,
12:25
because these are valuable to the organism.
303
730000
2000
因為胺基酸對有機體而言非常重要。
12:27
And some other ones
304
732000
2000
而那些二十個以外
12:29
that are not used in the set of 20,
305
734000
2000
的沒被用的,
12:31
they will not appear at all
306
736000
2000
在任何情況下，
12:33
in any type of concentration.
307
738000
2000
則毫無出現的可能。
12:35
So this also turns out to be extremely robust.
308
740000
2000
這個結果極為顯著。
12:37
It doesn't matter what kind of sediment you're using to grind up,
309
742000
3000
不管你是要研磨哪種沙土,
12:40
whether it's bacteria or any other plants or animals.
310
745000
3000
不管是細菌或是動植物。
12:43
Anywhere there's life,
311
748000
2000
到處都有生命存在,
12:45
you're going to have this distribution,
312
750000
2000
你會得到這個分配圖,
12:47
as opposed to that distribution.
313
752000
2000
而不是無生物的分配圖。
12:49
And it is detectable not just in amino acids.
314
754000
3000
不光是胺基酸可被探测。
12:52
Now you could ask:
315
757000
2000
這時你可能會問:
12:54
well, what about these Avidians?
316
759000
2000
那Avidians呢?
12:56
The Avidians being the denizens of this computer world
317
761000
4000
Avidians是存在電腦世界裡的產物,
13:00
where they are perfectly happy replicating and growing in complexity.
318
765000
3000
它們在那快樂地繁殖成長。
13:03
So this is the distribution that you get
319
768000
3000
這就是Avida的分配圖,
13:06
if, in fact, there is no life.
320
771000
2000
假設Avida裡沒有生命存在。
13:08
They have about 28 of these instructions.
321
773000
3000
圖裡有28種指令。
13:11
And if you have a system where they're being replaced one by the other,
322
776000
3000
而且你如果可以創造一個供這些指令相互取代的系統,
13:14
it's like the monkeys writing on a typewriter.
323
779000
2000
彷彿是猴群在打字機上亂打字。
13:16
Each of these instructions appears
324
781000
3000
則每一種指令
13:19
with roughly the equal frequency.
325
784000
3000
所出現的頻率會大約相等。
13:22
But if you now take a set of replicating guys
326
787000
4000
但是如果是剛在影片裡出現的
13:26
like in the video that you saw,
327
791000
2000
會複製的玩意兒,
13:28
it looks like this.
328
793000
2000
看起來會像這樣。
13:30
So there are some instructions
329
795000
2000
有部分的指令
13:32
that are extremely valuable to these organisms,
330
797000
2000
對於有機體相當重要,
13:34
and their frequency is going to be high.
331
799000
3000
所以這些指令的出現頻率相對會很高。
13:37
And there's actually some instructions
332
802000
2000
不過也有一些指令
13:39
that you only use once, if ever.
333
804000
2000
只出現過一次。
13:41
So they are either poisonous
334
806000
2000
它們不是有毒
13:43
or really should be used at less of a level than random.
335
808000
4000
不然就是使用上必須低於隨機的水平。
13:47
In this case, the frequency is lower.
336
812000
3000
這種情況下出現頻率會比較低。
13:50
And so now we can see, is that really a robust signature?
337
815000
3000
那麼我們現在所看到的算是一個明顯的指標嗎?
13:53
I can tell you indeed it is,
338
818000
2000
我可以告訴你的確是，
13:55
because this type of spectrum, just like what you've seen in books,
339
820000
3000
因為這種分配型態，如同你剛看到的書,
13:58
and just like what you've seen in amino acids,
340
823000
2000
還有胺基酸的例子,
14:00
it doesn't really matter how you change the environment, it's very robust;
341
825000
3000
不管你怎麼改變環境，特徵就是這麼明顯;
14:03
it's going to reflect the environment.
342
828000
2000
並且會反映出環境的特色。
14:05
So I'm going to show you now a little experiment that we did.
343
830000
2000
我現在要給你們看一個我們做的實驗。
14:07
And I have to explain to you,
344
832000
2000
我得解釋一下,
14:09
the top of this graph
345
834000
2000
這張圖表的上方
14:11
shows you that frequency distribution that I talked about.
346
836000
3000
指的是我剛講到的頻率分配。
14:14
Here, in fact, that's the lifeless environment
347
839000
3000
事實上，這是個無生命的環境,
14:17
where each instruction occurs
348
842000
2000
每種指令出現的頻率
14:19
at an equal frequency.
349
844000
2000
都相等。
14:21
And below there, I show, in fact,
350
846000
3000
下面的圖表
14:24
the mutation rate in the environment.
351
849000
3000
代表環境的突變率。
14:27
And I'm starting this at a mutation rate that is so high
352
852000
3000
我把一開始的突變率調得很高,
14:30
that, even if you would drop
353
855000
2000
高到就算你
14:32
a replicating program
354
857000
2000
放入一個會
14:34
that would otherwise happily grow up
355
859000
2000
快速成長的複製程式,
14:36
to fill the entire world,
356
861000
2000
然後佈滿整個空間,
14:38
if you drop it in, it gets mutated to death immediately.
357
863000
4000
當你把程式放進去時，它會立刻突變至死。
14:42
So there is no life possible
358
867000
2000
在這種突變率之下
14:44
at that type of mutation rate.
359
869000
3000
沒有任何生命能夠存活。
14:47
But then I'm going to slowly turn down the heat, so to speak,
360
872000
4000
但是接下來我要把突變率降低,
14:51
and then there's this viability threshold
361
876000
2000
到一個適當的程度,
14:53
where now it would be possible
362
878000
2000
如此一來就有一個複製體
14:55
for a replicator to actually live.
363
880000
2000
能夠存活。
14:57
And indeed, we're going to be dropping these guys
364
882000
3000
然後我們要把這些玩意兒
15:00
into that soup all the time.
365
885000
2000
放到原生湯裡。
15:02
So let's see what that looks like.
366
887000
2000
看看會發生什麼事。
15:04
So first, nothing, nothing, nothing.
367
889000
3000
一開始沒有事情發生。一點都沒有。
15:07
Too hot, too hot.
368
892000
2000
然後發生劇烈的變化。
15:09
Now the viability threshold is reached,
369
894000
3000
現在已經達到可行數值
15:12
and the frequency distribution
370
897000
2000
以及頻率分配。
15:14
has dramatically changed and, in fact, stabilizes.
371
899000
3000
一開始發生劇烈變化, 然後, 事實上, 緩和下來。
15:17
And now what I did there
372
902000
2000
接著我就很邪惡的
15:19
is, I was being nasty, I just turned up the heat again and again.
373
904000
3000
把溫度一再再調高。
15:22
And of course, it reaches the viability threshold.
374
907000
3000
當然它就能達到可行數值。
15:25
And I'm just showing this to you again because it's so nice.
375
910000
3000
這實在是太讚了,所以我要再放一次給你們看。
15:28
You hit the viability threshold.
376
913000
2000
一旦降到可行數值
15:30
The distribution changes to "alive!"
377
915000
2000
分配就變成「有生命的, 萬歲!」。
15:32
And then, once you hit the threshold
378
917000
3000
當點突變率
15:35
where the mutation rate is so high
379
920000
2000
回升到可行數值
15:37
that you cannot self-reproduce,
380
922000
2000
就不能自我複製,
15:39
you cannot copy the information
381
924000
3000
也不能將信息
15:42
forward to your offspring
382
927000
2000
毫無錯誤地
15:44
without making so many mistakes
383
929000
2000
傳給後代,
15:46
that your ability to replicate vanishes.
384
931000
3000
因此複製能力就消失了。
15:49
And then that signature is lost.
385
934000
3000
代表生命指標也消失了。
15:52
What do we learn from that?
386
937000
2000
我們從中學到了什麼?
15:54
Well, I think we learn a number of things from that.
387
939000
4000
我認為我們學到了幾個重點。
15:58
One of them is,
388
943000
2000
第一,
16:00
if we are able to think about life
389
945000
3000
如果我們能夠
16:03
in abstract terms --
390
948000
2000
以抽象的定義來思考生命--
16:05
and we're not talking about things like plants,
391
950000
2000
我們不提植物,
16:07
and we're not talking about amino acids,
392
952000
2000
不提胺基酸,
16:09
and we're not talking about bacteria,
393
954000
2000
也不提細菌,
16:11
but we think in terms of processes --
394
956000
2000
而我們思考的是過程--
16:13
then we could start to think about life,
395
958000
3000
如此一來就能探討生命
16:16
not as something that is so special to Earth,
396
961000
2000
不限於地球特有的生命,
16:18
but that, in fact, could exist anywhere.
397
963000
3000
而是, 任何地方都可能有生物的存在。
16:21
Because it really only has to do
398
966000
2000
因為它只跟這些
16:23
with these concepts of information,
399
968000
2000
儲存訊息的
16:25
of storing information
400
970000
2000
概念有關,
16:27
within physical substrates --
401
972000
2000
在物質基底之內--
16:29
anything: bits, nucleic acids,
402
974000
2000
可以是任何東西: 位元組、核酸,
16:31
anything that's an alphabet --
403
976000
2000
任何可當成跟字母一樣的單位--
16:33
and make sure that there's some process
404
978000
2000
並且確保能有一個
16:35
so that this information can be stored
405
980000
2000
遠比你預期的時間還要長,
16:37
for much longer than you would expect
406
982000
2000
不被時間比例影響信息衰退,
16:39
the time scales for the deterioration of information.
407
984000
4000
可讓信息儲存起來的程序。
16:43
And if you can do that,
408
988000
2000
如果條件都滿足,
16:45
then you have life.
409
990000
2000
就算是生命。
16:47
So the first thing that we learn
410
992000
2000
所以我們學到的第一件事就是
16:49
is that it is possible to define life
411
994000
3000
生命可以單獨
16:52
in terms of processes alone,
412
997000
3000
依照程序來定義,
16:55
without referring at all
413
1000000
2000
無須借助
16:57
to the type of things that we hold dear,
414
1002000
2000
其它我們重視的東西,
16:59
as far as the type of life on Earth is.
415
1004000
3000
例如地球的生命型式。
17:02
And that in a sense removes us again,
416
1007000
3000
這個結論又一次
17:05
like all of our scientific discoveries, or many of them --
417
1010000
3000
就像所有的科學發現一樣--
17:08
it's this continuous dethroning of man --
418
1013000
2000
再度證明了
17:10
of how we think we're special because we're alive.
419
1015000
3000
我們的存在並沒有那麼獨特。
17:13
Well we can make life. We can make life in the computer.
420
1018000
3000
我們能夠製造生命，在電腦裡頭製造生命。
17:16
Granted, it's limited,
421
1021000
2000
當然, 這的確是有限的,
17:18
but we have learned what it takes
422
1023000
3000
不過我們藉此了解到
17:21
in order to actually construct it.
423
1026000
2000
架構生命的要素。
17:23
And once we have that,
424
1028000
3000
一旦我們有了這些要素,
17:26
then it is not such a difficult task anymore
425
1031000
3000
創造生命便不再是難事,
17:29
to say, if we understand the fundamental processes
426
1034000
4000
如果我們能掌握基本的過程
17:33
that do not refer to any particular substrate,
427
1038000
3000
而不透過任何特殊基底,
17:36
then we can go out
428
1041000
2000
我們就能走出現有的框架
17:38
and try other worlds,
429
1043000
2000
探索地球以外的世界,
17:40
figure out what kind of chemical alphabets might there be,
430
1045000
4000
了解那裏會有什麼樣的化學元素符號,
17:44
figure enough about the normal chemistry,
431
1049000
2000
認識普遍的化學物質,
17:46
the geochemistry of the planet,
432
1051000
3000
還有該星球的地球科學,
17:49
so that we know what this distribution would look like
433
1054000
2000
如此一來我們便能了解
17:51
in the absence of life,
434
1056000
2000
沒有生命存在的分配型態會如何呈現,
17:53
and then look for large deviations from this --
435
1058000
3000
然後利用分配型態找到一些特例--
17:56
this thing sticking out, which says,
436
1061000
3000
因為偏離值能凸顯出:
17:59
"This chemical really shouldn't be there."
437
1064000
2000
「這個化學物質不應該在那裏出現」。
18:01
Now we don't know that there's life then,
438
1066000
2000
我們現在不確定那裏是否有生命存在,
18:03
but we could say,
439
1068000
2000
但至少,
18:05
"Well at least I'm going to have to take a look very precisely at this chemical
440
1070000
3000
「我會仔細研究這個化學物質
18:08
and see where it comes from."
441
1073000
2000
辨識出它是從哪而來」。
18:10
And that might be our chance
442
1075000
2000
這也許會成為
18:12
of actually discovering life
443
1077000
2000
我們發現新生命的機會,
18:14
when we cannot visibly see it.
444
1079000
2000
即使我們不能看見生命的形體。
18:16
And so that's really the only take-home message
445
1081000
3000
這便是我要給你們唯一的
18:19
that I have for you.
446
1084000
2000
重要結論。
18:21
Life can be less mysterious
447
1086000
2000
當我們知道其他星球也存在生命,
18:23
than we make it out to be
448
1088000
2000
生命就沒有
18:25
when we try to think about how it would be on other planets.
449
1090000
4000
我們想像中的那般神祕了。
18:29
And if we remove the mystery of life,
450
1094000
3000
如果我們能夠揭開生命神祕的面紗,
18:32
then I think it is a little bit easier
451
1097000
3000
那對我來說,
18:35
for us to think about how we live,
452
1100000
2000
要思考我們如何生存,
18:37
and how perhaps we're not as special as we always think we are.
453
1102000
3000
以及我們不是那麼獨特這類的議題，就不再是難事。
18:40
And I'm going to leave you with that.
454
1105000
2000
我要把這部分留給你們去想。
18:42
And thank you very much.
455
1107000
2000
謝謝大家。
18:44
(Applause)
456
1109000
2000
(鼓掌聲)
ABOUT THE SPEAKER
Christoph Adami - Artificial life researcherChristoph Adami works on the nature of life and evolution, trying to define life in a way that is as free as possible from our preconceptions.
Why you should listen
Christoph Adami researches the nature of living systems, using 'artificial life' -- small, self-replicating computer programs. His main research focus is Darwinian evolution, which he studies at different levels of organization (from simple molecules to brains). He has pioneered theapplication of methods from information theory to the study of evolution, and designed the "Avida" system that launched the use of digital life as a tool for investigating basic questions in evolutionary biology.
He is Professor of Applied Life Sciences at the Keck Graduate Institute in Claremont, CA, and a Visiting Professor at the BEACON Center for the Study of Evolution in Action at Michigan State University. He obtained his PhD in theoretical physics from the State University of New York at Stony Brook.
Christoph Adami | Speaker | TED.com