Emergence and Emergent Properties in Science

George Ellis

Strong and weak emergence

‘Emergence is when you put simple things together and new behaviours emerge. Can you explain them all in terms of the lower level? Weak emergence is where you can. Strong emergence is where you can’t. I believe both exist.'

Transcript

Print

STRONG AND WEAK EMERGENCE

David: Can I ask about… various people have mentioned to us, they’ve said there’s weak emergence and strong emergence. What are they and what’s the difference?

GE: Okay, emergence is when you put simple things together to create much more complicated things which have got behaviour which is completely different from the simple things which you put together. So new kinds of behaviours emerge at the higher level structures you create out of these simple things. And so just as an example: each of us are made out of atoms, of protons, neutrons, electrons, but what we get at the level of the body, the mind, is completely different from what you had at lower levels.

Roughly speaking, weak emergence is where new powers come in but you can explain them in terms of the lower level, and strong emergence is where you can’t.

David: Right. Which one do you believe in? You believe in strong emergence?

GE: I believe both exist. For instance, I just believe, as I have made very clear, the mind can emerge and that requires top-down causation and you can explain all of the properties of the brain in terms of the physics of the molecules, the neurons and all the rest of it, but you can’t explain the ideas in there. And so those ideas are strong emergence. And so, in my view point, it’s absolutely clear: strong emergence occurs and has physical causes.

David: And so that phrase that Rosenberg uses, where he says the physical facts fix everything?

GE: It’s not true. But, you see, what is true is once these ideas have been incorporated in the brain and we learn them, then they get built into the connection strengths in the neuron. Once that has happened, then it is reasonably plausible that a totally identical brain with totally identical strengths and totally identical excitations at that instant might experience the same thoughts. But the point is, you can’t get to that state without top-down causation.

I believe that as complexity arises, higher level rules come into being which were not implied by the lower level. Just as a very, very simple example. Physics talks about interaction between particles, forces, electromagnetic and strong and weakened direction. A little bit up, you get bacteria and amebae where the rules that come in there are Darwinian selection.

David: You’ve got life.

GE: Yes. Physics says nothing about Darwinian selection. A new principle has come into being.

David: Is that a level or emergence then?

GE: Yes.

David: So you have physics and chemistry, and then at some point they manage to create something which is then life, which has its own rules.

GE: It has its own rules, and those rules are governed by information. One of the things which comes into existence is information, which didn’t exist at the lower level. But the rules by which the information acts come into existence.

David: Right, yes, DNA, for example.

GE: And at a higher level abstract thought comes into existence, and all the possibilities that thought allows come into existence.

David: Right, so then in my world you’ve got physics and chemistry, which gets more and more complicated until it creates an emergent level of life.

GE: Yes.

David: And then natural selection is building more and more complicated things until eventually it creates minds which can see or touch that level of ideas.

GE: But these are what are called the major transitions in the evolutionary history. They’re not transitions just of… They’re transitions of quality. The quality of what happens changes completely. And I think what is very important here is the following: my scientific colleagues listen to me and say, ‘Never-the-less, I believe if you knew everything about the universe at the time of the Big Bang, you could predict everything that happens today.’

David: No, do they really say that?

GE: Well, yes.

David: I thought that went out a couple of hundred years ago. Do people really say that?

GE: I’ve had a Facebook interchange with a very, very interesting colleague who claimed this a couple of months ago.

David: But that’s untenable. Surely then he is saying the reason that you would say the answer to 2+2 is 4 has got nothing to do with mathematics. It’s because some quark and electron headed out from the Big Bang bumped, and it was always going to make you say 4.

GE: That’s right.

David: That’s crackers, isn’t it?

GE: I’m just completing this book on top-down causation in which I had some derogatory comments about this. I said it was ‘ludicrous’, or some word like that, and the reviewer of the book said, ‘How can you use the word “ludicrous”’? You’re a scientist.’

David: But it is ludicrous, don’t you think?

GE: It is ludicrous. The idea that everything that we would be saying, the idea that the theory of relativity and the Battle of Waterloo was written into the Big Bang, is just ludicrous.

Peter Atkins

Is emergence just mumbo-jumbo?

‘I don’t see that there is any real difference between an emergent property like temperature, which I claim can be understood in terms of the behaviour of individual molecules, and the emergent property of being a butterfly.'

Transcript

Print

Is emergence just mumbo-jumbo?

David: Can I ask you…? We’ve been using the word ‘science’: what are the qualities of science? I mean, I’m thinking of reductionism, determinism, those things. When you think of science, is it that kind of reductionistic science, or…?

PA: I think the core of science is imagination in alliance with honesty. Just to be honest without being imaginative means that you’re not doing very good science; being imaginative without being honest means that you’re being a poet.

David: That’s the poets struck off the list!

PA: I think the combination of the two is really the core of being an effective, interesting scientist.

David: Okay.

PA: Then I think you then have the distinction between reductionism and emergence.

David: Yes, what’s…?

PA: I think the easy part of science is reductionism: dismembering the butterfly in order to find its atoms. But the difficult part is reassembling the simple ideas that you’ve developed and creating the butterfly out of the atoms and seeing that an idea of beauty emerges from that.

David: Yes, when you’ve reduced it down to its bits, you understand how all the bits work. But even if you do understand that and you put all those bits back together, if all you understand are the rules for the individual bits, you’ll have left out the emergent rules. Once it becomes the whole system, once it becomes the butterfly, there are new rules which emerge at the level of the butterfly and have very little to do with the level of its atoms. Does that work for you or is that mumbo-jumbo?

PA: That’s mumbo-jumbo. I mean, all defeatism is mumbo-jumbo, basically.

David: Why is it defeatism?

PA: Because it suggests that there’s something beyond your understanding of the entities that make up the butterfly.

David: Why should that be a defeat, though?

PA: Well, because we hope through the scientific method that we will understand absolutely everything. It’s a wonderful challenge that the world is out there in all its complexity. We little humans are tinkering with understanding little bits of it, but we’re gradually migrating towards an understanding of the whole, and that includes understanding how the individual entities collaborate into producing a system. I mean, it’s a real challenge.

David: Yes, and in its own way, a very powerful and beautiful one. I mean, everything we’ve got scientifically is from that method. But I don’t see why it’s a defeat to say in addition to the rules that govern the individual parts there could be entirely scientific rules which operate at the level of the system.

PA: Yes, but we’ve got no need for them at the moment. There are rules that operate only when you have congregations of atoms and molecules for which it would be meaningless to talk about the property of a single atom: temperature. What do you mean by ‘the temperature’ when you’ve got a single atom? It can’t mean anything.

But what you mean by temperature is the distribution of myriad molecules over the available energy levels following a particular form. Now, I can understand that. You might say that temperature, therefore, is an emergent property. And in a sense it is, but I also understand it in terms of what is going on at the level of a collection of individual molecules. So I don’t see that there is any real difference between an emergent property like temperature, which I claim can be understood in terms of the behaviour of individual molecules, and the emergent property of being a butterfly.

Ard: Really? So how about…?

PA: I mean, I’m an optimist. And I can’t stand philosophers because they’re all pessimists! So don’t be defeatist! Be optimistic!

David: When we talked to George Ellis and others who are keen on emergence, one of the reasons they’re keen on it is, they say, ‘Look, with reductionism everything is caused from the lowest level.’ Whereas, with emergence, they say you then get what they call downward causation, so the system itself then has an effect on its parts. Some things in the universe are not caused by just the rules which guide the atoms and the chemicals…

PA: Well, that’s nonsense, isn’t it?

David: Is it?

PA: I think it must be. It’s a mechanical universe. It might be a quantum mechanical universe, but effect follows cause even in a quantum universe.

David: Yes, and they’re saying sometimes the cause is at the higher level and causes an effect lower down; that it’s not all cause coming up from the atoms.

PA: Well, I can see that if I have a particular thought that it might make me sweat through embarrassment, for example.

David: Yes. That’s precisely it.

PA: So there is a kind of downward…

David: Right, so they’re saying an emergent level of the universe is the level of thought.

PA: Yes.

David: And that once it’s in the universe, then a whole new set of rules emerge which have causative power: they cause things to happen. So suddenly it’s not just that the atoms cause a reaction in your kidneys which make you think that you’re thirsty and that you have a thought. You see what I mean? It’s a radical challenge to reductionism, isn’t it? They think it is, and I’m inclined to agree it is.

PA: Science fiction writers often play with ideas like that as well. It’s quite interesting to see the way that writers of fiction think.

David: Yes, but I’m asking you as a scientist whether you think there’s something to it. Because you said, ‘Well, I could see an idea…’ It seemed like you were entertaining this notion might be right.

PA: Only to amuse you!

David: Okay, that’s me put down. But not to amuse me, what do you think of it? Because people like Noble and Ellis and others, they say, ‘Look, this is the future of science. This is a revolution. Yes, there’s causation, reductionist science, but there’s this other kind: the downward causation from emergence.’ They see this as a real scientific endeavour.

PA: Yes. Well, obviously, because it’s a scientific endeavour, you shouldn’t dismiss it out of hand. You should explore whether there are consequences, whether it’s testable and whether there is any… what we loosely regard as truth.

David: Loosely regard?

PA: And if there is even the hint of evidence for it, then it’s worth pursuing, because science doesn’t dismiss out of hand, or at least shouldn’t dismiss out of hand.

David: What’s your feeling about it as a scientist?

PA: What does my gut say, rather than my brain?

David: Yes.

PA: Worth thinking about, but not worth investing in.

Ard: That’s your investment advice? Sell!

Alex Rosenberg

What science can’t explain, yet

‘You should treat emergent properties as a signal, or a flag, that indicates a domain in which science has some work to do. People say, “Reduce this. I dare you.” And science eventually finds a way to do it.’

Transcript

Print

WHAT SCIENCE CAN’T EXPLAIN, YET

David: Reductionism, determinism: do they, when you put them together… A lot of people say, ‘Ah, what that means is we see ourselves as machines’, which, I think, people are quite happy to look at an ant and go, ‘Yeah, it’s a little machine with legs’. Is that what those two parts of modern science force us to think?

AR: Well, yes, unless you’re going to put too much weight on a machine, because a machine is an artefact, and an artefact requires a designer, and the designer in our case was Mother Nature operating through the principle of blind variation and natural selection.

Ard: I think there’s no ambiguity. We believe that the machines…

AR: But the machine includes our brain and therefore our minds, and we’re descendant, all of us, from the same struc…

David: Okay.

AR: That’s one of the key ideas of Darwin’s discoveries.

Ard: So maybe someone would say… So we’re animals. I think nobody dis…

AR: Without a divine spark.

Ard: Without a divine spark. We’re animals, but there’s nothing… If we’re looking in that category, there’s nothing beyond the fact that we’re animals. We’re no different from animals.

AR: Yes, right.

David: But unlike a lot of animals, we have thoughts.

AR: Yes, unlike a lot of animals, but like a lot of animals.

David: Yes, but there’s a qualitative difference in the thinking which we are capable of than most of the rest. So in other words…

AR: I think I just…

David: Yes, we’re animals. But something has emerged in our evolution…

AR: I just don’t think I’m going to grant that. If you want to use the words ‘emergent’ or ‘irreducibly complex’ or ‘valuative’, then I think these are placeholders for questions on the agenda of science. At this point we don’t have a good handle on the details of the answer to this question that I favour. And you, holding an alternative view, have probably even less grounds for confidence.

Ard: So David doesn’t believe in God at all?

AR: No, but he does believe in the existence of emergent properties, shall we say.

David: Well, I think that’s…

Ard: Do you think that’s something he should let go of, emergent properties?

AR: Yes.

David: Why?

AR: Alternatively, I think you should treat emergent properties as a signal, or a flag, that indicates a domain, or a terrain, in which science has some work to do.

David: Could there not be a proper science which did include emergence, which would be scientific? Why couldn’t that be? I feel open-minded about it.

AR: My confidence here is based on induction, and I could be wrong, because inductive arguments are arguments that are not truth-preserving. They’re arguments from a finite body of data to expectations about the future.

David: Yes, it’s been true up to now, and so…

AR: When I look at the history of science, I look at the history of people drawing lines in the sand and challenging science to transcend them, and science successively doing it. And the most famous example is this: Kant, the great German philosopher who wrote The Critique of Pure Reason proved, or purported to prove in that book, that the only possible way to conceive of the universe was in the way that Newton had. He was a total determinist and mechanist and reductionist about the physical world. But then he said, ‘There will never be a Newton for the blade of grass.’ And what he meant by that was when it comes to the biological domain, we will have to employ teleological or purpose-driven theory, because the properties of life and of the means-and-ends economy of biological nature are impervious to physical explanation. And 22 years later in Shropshire, England, the Newton for the blade of grass was born.

David: Mm-hm.

AR: Charles Darwin. Okay? And it’s a perfect example of how people say, ‘Reduce this. I dare you.’ And science eventually finds a way to do it.

Denis Noble

Emergent properties in biology

‘What is the function of the heart? What is the function of the leg? You cannot answer that question if you say, “It’s just a bunch of molecules interacting.” You’ve got to address, therefore, the question, what is its purpose?'

Transcript

Print

emergent properties in biology

David: Can I ask you to define…? Tell me what emergence is and how it’s different from the reductionist programme.

DN: I think that what emerges is functionality. Now that’s a difficult one, I know, but you see, remember, I am a physiologist, so what does a physiologist ask as his main question? It is, ‘What is the function of the heart? What is the function of the liver? What is the function of the leg?’ And you cannot answer that question if you say, ‘It’s just a bunch of molecules interacting.’ You’ve got to address, therefore, the question: what is its purpose?

David: What’s it for?

DN: What’s it for? I can’t avoid that. Now, I know that people who see all of this from a purely molecular point of view will say, ‘But it’s just blind chance that has produced all of this, therefore there is no function.’ But that’s to make the mistake of thinking that because the components don’t have a function individually, that therefore the system doesn’t have a function.

David: So is emergence concentrating on the system level?

DN: Yes.

David: And saying what about the system, though?

DN: Saying that it serves a purpose because that function has emerged. Now, it might have emerged originally from random action. I don’t deny the possibility that life has arisen by purely random chemical reactions initially. I don’t know the answer to that. It’s a question, and it may be right. But once you have got a system that constrains the parts, it takes over.

David: The system takes over?

DN: Yes, the system takes over, because you’ve got… well, in technical terms, you’ve got what the mathematicians would call an attractor: you’ve got something that constrains the components to, as it were, always go to a particular state.

Ard: What you’re saying is if you have an oscillator, like in your heart, and you drill down to the molecular level and you see a protein moving back and forth, and you say, ‘Why is that protein moving back and forth?’, well, there’s no molecular reason for it to move back and forth. The reason is this collective system level that’s channelling information downwards into the cell and not upwards.

DN: Exactly so. And those proteins are proteins that open and close.

Ard: Okay.

DN: And the opening and the closing is controlled by the cell property.

David: So the components build something, but once they’ve built that something…

DN: It takes over.

David: …that something then has an effect on the components that built it from then on?

DN: Exactly so, yes.

Ard: And so if you ignore the fact that the top-level description is acting on the lower-level…

David: Then you’re missing...

Ard: Then you’re missing the point.

Ard: So you were a physiologist, and you had what people thought were hare-brained ideas about trying to use a computer to calculate the heart, and they didn’t let you in. Is that right?

DN: That’s correct, and for a very interesting reason: they were thinking totally in reductionist terms.

DN: The question put to me was, ‘Where in your equations is the oscillator?’ They couldn’t therefore see that the oscillator was going to emerge from the interactions, and if you separated those components of the model out, they wouldn’t oscillate. So it’s simple. They didn’t understand that.

Ard: So is the story that they let you in only in the morning, or something?

DN: Yes. You remember in those days you had one computer in the whole of London. You had to queue up to use it. So it was running 24 hours a day. I was given between two o’clock and four o’clock, am.

David: That’s slightly unfriendly, isn’t it?

DN: Well I was thought to be the person who was least likely to produce a successful computation. Six months later I got a paper in Nature.

David: But at the time you yourself were a reductionist like everybody else?

DN: Yes, absolutely. That means I couldn’t explain to them what I’ve just said to you. I should have said, ‘It will emerge from the equations.’ Instead, I got a pencil out and scribbled on a bit of paper, in the way that physiologists sometimes do, ‘I think this will interact with that, which will interact with that, and, hopefully, something may come out of this that helps me.’ But I didn’t have the language in those days to say to them, ‘This will be an emergent property.’

So I wasn’t able to reply to these computer experts that the phenomenon of rhythm would emerge from the equations representing the proteins in the cell. If I’d done that, I might have got success more quickly.

David: I think they’d have thrown you out.

DN: Precisely! I think they’d say, ‘What on earth is this about?’ But I wouldn’t even have been able to formulate it like that, and yet it is a downward process of causation in which the properties of the cell constrain those proteins to behave differently from what they would do if they were in a Petri dish.

David: I had imagined that when we talked about levels of emergence, that it was just, sort of, life, and then, maybe, thought. Are those levels of emergence, or are you saying there are lots more?

DN: Yes, I think they are, but I also think that we have to add many more. Just think of any cell in your body. That kind of cell probably took at least a billion years to emerge, because it is unbelievably complicated. It’s made of many different organisms coming together. There’s been a process of what we call symbiogenesis. Your mitochondria in those cells, which are the energy factories, those were originally bacteria.

David: But each of those levels… Do they then bring a whole new set of rules that come with that level?

DN: Yes.

David: Which are rules which weren’t there before?

DN: Exactly so. Because that bacterium is no longer free. It’s no longer free in the sense that it’s now constrained by the system of which it has become part, and, therefore, it’s no longer a bacterium.

David: Right. So is natural selection an emergent rule?

DN: I think that evolution has evolved. I mean, the process of evolution has evolved and there have been many mechanisms at different stages.

David: It makes this link, then, between emergence and the whole discussion about whether you can have genuine novelty in the world, in the universe.

DN: Yes.

David: Or whether the universe came into being with its rules and that’s it. Now we’re just working out the consequences.

DN: Yes.

David: You are saying, if I’ve understood you, because of emergence that, creation is a more open thing?

DN: Precisely, yes.

Ard: And new rules are being created…

DN: …as we go along. Once a system has emerged that constrains the components, it is self-maintaining.

David: Self-maintaining, right. I think you’re saying more than that. It’s self-maintaining but then can generate new emergence?

DN: Exactly so. Yes, yes.

But to people who say, you know, ‘How could all that happen?’, part of my answer would be, first of all, it had a long time in which it could happen. And second, the idea that it happened entirely by blind chance is the wrong way to describe it.

Blind chance might well have been something at the beginning, I don't know. Frankly, I’ve no idea what happened at the very beginning. I don’t think anybody else does either. But once you have the emergence of a property that constrains the rest of the parts, you have the potential for further novelty.

For example, one of those cells can interact with another cell. One can eat the other, or the other way round, or they can come together to combine. Those are new things that couldn’t have happened before the emergence of the cell in the first place. So, yes, each stage creates the potential for new properties.

David: So when you get life emerging out of chemistry, you have something genuinely new and it brings with it its own new rules?

DN: Yes, exactly.

David: So if you knew everything about chemistry, would you be able to predict everything about life?

DN: I think the best way to put that answer is actually to refer to chemistry itself. Let’s take the water molecule. It’s got oxygen; it’s got two hydrogens, and we know the properties of oxygen and the two hydrogens. From that alone you’d find it very difficult to predict the properties of water.

Now, the theoretical chemist will reply to that and say, ‘I know enough about quantum mechanics. I know enough about the movement of electrons around atoms, that I can show how, when you produce the interaction between oxygen and the two hydrogens, I get the properties of water. I could even think in terms of computing that entirely from quantum mechanical descriptions.’

My reply to the theoretical chemist was to say, ‘But you’ve already introduced the interactions in saying that you will compute what happens when those two hydrogens and oxygens come together.’ Because although you’re saying that from your quantum mechanical equations, you can, in principle calculate the behaviour and the emergence of water, what you’re failing to understand is it’s only because you have introduced those constraining factors, which are the relationships of things coming together.

David: But did you realise how big an idea, or as fertile an idea it has been in your life? Because it has been a major… I mean, in some sense it set the course of your life.

DN: It set the course.

David: When did you realise that that was the power of the idea?

DN: I think these kinds of realisations take 20, 30, 40 years, because it upsets everything. Even when I came to write the little book, The Music of Life, nearly ten years ago, I wasn’t there yet, and that process is continuing. But there was still a process to occur which enabled me to start answering questions: where does the meaning come here? Because life is a meaningful process, it has to be. You have to address that question. I don’t think there’s any way of avoiding it.

Ben Okri

Writers, narrative and emergence

'There's a great nervousness with poets and novelists about this emergent idea. I mean, a lot of writers want to be in control of their material. They want to say, "I shaped this."'

Transcript

Print

WRITERS, NARRATIVE AND EMERGENCE

David: Do you think when you write narrative, or when you read great narrative, is there some discovering of something rather than just completely creating it?

BO: Yes, I think it's a wonderful image and idea, and I think you do discover it. It emerges through the rigor of, for want of a better word, the mathematics of your story. Sometimes a narrative or a poem goes somewhere you did not entirely intend it to go because of its own inner logic. And this could be also the emergence of the theme – not just where the story goes – actually what it is about. You discover it in the process of doing it. It's quite terrifying; quite wonderful.

Ard: So is it that sometimes the narrative surprises you? Does it do something you didn't expect?

BO: Yes, it surprises you because the narrative is not an external thing: narrative is what emerges. I think it's two things really: it's what emerges from you, because every story you tell is drawing upon the deeper, as well as the superficial, aspects of your mind, your psyche, all of what makes you.

So you might start very deliberately, but because you've started, all sorts of unsuspected lower, deeper, subterranean elements come into play. People who write can't get over that fact that you start with something quite simple and something turns up that you hadn’t… There's no way you could've dreamt that it'll emerge, and then you have to shape that.

Ard: You use that word ‘emerge’, and I thought that was interesting. Scientists talk about ‘emergent phenomena’. So you put a lot of things together, a lot of individual little units, and sometimes, from the collective, something emerges that was a surprise.

BO: That sounds like a lot of story-tellers.

Ard: Okay, yeah.

David: But the scientists are very, very nervous about it. They're really not sure whether it's true, and they're not sure if they even want it to be true.

Ard: Some scientists love it.

David: Some do, but there's a great nervousness about this emergent idea in science.

BO: Well, there's a great nervousness with poets and novelists about this emergent idea. I mean, a lot of writers want to be in control of their material. They want to say, ‘I shaped this. This came from me. This was a deliberate act.’

They don't want to be surprised by Anna Karenina. That's the last thing. I like to be. I work to be surprised wide awake. And I can see why the emergent idea you speak of is worrying to scientists, because it has to do with the element of control and the element of objectivity, the element of truth. But I think it's part of the wonder of what we do.