Steve Grand


CREATION
Life and How to Make it

Phoenix 2001



pg 73

Keywords : emergence - interactions - When populations of interacting structures become arranged in certain configurations, and something new and surprising comes into existence, we call this an emergent phenomenon - Conway's Game of Life: glider - logical reasoning - prediction - we could, in principle, predict the existence of a glider from the rules of Life, but only by actually carrying these rules out, simulating the system - Matter is just one link in the chain of being . Atoms are no more real than societies or minds. Hardware is a subset of software - self-sustaining patterns in space and time - The interconnectedness of all things - self-organizing chains of cause and effect - 'problem' of how mind has influence over matter is spurious. Since the two are not distinct, the idea that one can affect the other should not be at all difficult to accept - purposeful action - cause and effect act in webs, not chains - think of the information acting upon the brain at least as much as we think of the brain acting on the information - gestalt: form, the whole that is greater than the sum of its parts - Life, we discover, is a loose coalition of selfmaintaining eddies in a flowing stream. When the whole becomes more than the sum of its parts something new and perfectly real comes into existence. This gestalt is not mysterious, but neither is it a figment of our imagination - the basic buildin blocks of cause and effect





EMERGENCE

When a relatively complex result arises out of simple interactions between members of a population, it is known as emergent behaviour.

The word ‘emergence' is quite controversial in some quarters. When populations of interacting structures become arranged in certain configurations, and something new and surprising comes into existence, we call this an emergent phenomenon.

Yet some scientists dispute that these phenomena are real at all, and think they are a product of our own desire to categorize things, or that they are a surprise to us only because we are not clever enough to have predicted their occurrence.

Is emergence really only ignorance? Unexpectedness is certainly a common feature of emergence. Nobody who was given the rules for Conway's Game of Life would immediately infer the existence of the glider. But is that perhaps because we are just not clever enough to see it? If we were smarter, would the existence of the glider be immediately obvious from the rules alone, and hence no surprise? If so, would the glider no longer deserve to be called emergent? I don't think so.

‘From a drop of water’, said the writer, ‘a logician could infer the possibility of an Atlantic or a Niagara without having seen or heard of one or the other. So all of life is a great chain, the nature of which is known wherever we are shown a single link of it’. Arthur Conan Doyle A Study in Scarlet

The quote from Sherlock Holmes suggests that we can infer the existence of something like a waterfall from any element in its chain of existence, such as a raindrop. This typically Victorian attitude, which expects logical reasoning to conquer all, is actually more hubris than fact.

Once we have seen Conway's glider we can easily work out how the basic rules led to its creation, so the existence of the glider could in principle be logically inferred from the rules alone - after all, it does not arise by magic. The fact that we fail to anticipate it is partly a problem of perspective: if I peer at you through a small hole in a fence when you are some distance away, I can see you but you cannot see me. Logic suffers from a similar oneway effect. Having seen the trick we can now work out how it was done because this exercise is a narrowingdown process, aiming towards a known goal. But going the other way is much harder. Starting with knowledge of the parts, it is more difficult to work out the behaviour of the whole because we do not know where we are heading.

Yet even with enough brainpower at our disposal, there is perhaps a more fundamental reason why the concept of emergence is justified. Certainly we could, in principle, predict the existence of a glider from the rules of Life, but only by actually carrying these rules out, simulating the system in our minds or in a more concrete form such as a computer program or an exercise on paper. In other words, to see the ultimate effect of the rules of Life, we are obliged to play out the game in our heads and see what happens. Just staring at the rules will not get us there. If we can predict the behaviour of the system only by running it, then this is no kind of inference at all - it is more like turning straight to the last page of a detective novel to find out whodunnit.

So emergence is genuinely surprising, but which things are truly emergent and which are not? Is breakfast an emergent phenomenon caused by the juxtaposition of eggs and bacon? Is soccer an emergent consequence of a group of men in shorts?

What is more, are emergent phenomena real things or are they just figments of our imagination? This is why I went to such lengths to disabuse you of the notion that matter is different from form. Matter is just one link in the chain of being . Atoms are no more real than societies or minds. Hardware is a subset of software. If we accept that an atom is still a real thing after all this, then we must give equal weight to something like a mind, because they are both made of the same non-stuff. Conway's glider is a real thing too. Just because it is 'immaterial' we should not regard it as any less real than a molecule, because both are self-sustaining patterns in space and time.

But we do have to be cautious about which things are really emergent phenomena, and which are artefacts that result from the way our brains perceive and classify the world. For something to be an emergent phenomenon I think it has to be persistent in its own right. For example, a business meeting is a thing in some limited sense but it does not contain the means for its own continuation, so it is not a thing in the striking way that a society is. Societies actively maintain themselves through laws, policing and wars, while business meetings simply drag on ... Neither is a soccer game an emergent phenomenon, because it does not assemble or maintain itself. A soccer match is instigated from outside, and if we do not count injury time or extra time it lasts for 90 minutes regardless of what happens on the pitch. Soccer as a sport, on the other hand, emerged out of nowhere and would probably come as a complete surprise to an alien biologist, given the basic rules of human behaviour to work from. There is no external referee waiting to blow his whistle when soccer has run its time as a social phenomenon, it continues because (in some way that I have never understood) it contains a mechanism for its own survival.

Business meetings and soccer matches are imposed from the top down, they are local consequences or manifestations of a larger system, whereas societies and sports emerge from the bottom up, as a global consequence of a population of smaller components. Yet to some degree this is just a matter of perspective. Even soccer and societies are persistent only because they form part of a bigger system, like eddies in a larger stream. This is true to a greater or lesser extent of most, if not all, emergent phenomena. Really, there is only one huge emergent phenomenon in existence, called the universe.

Inside this there are regions of smaller, more or less independent, interacting loops, of cause and effect that we think of as things in their own right. But usually these coherent regions remain so only by interacting with their context. We can think of minds, societies, atoms or ideas as discrete things, but we must always be mindful that they form part of a larger matrix.

The interconnectedness of all things

Sometimes the subtle interplay between distant parts of this giant superpattern can be surprising to those people who take a less holistic view of nature. If things like telepathy or precognition have any real basis, and I am not suggesting they do, then they might plausibly involve self-organizing chains of cause and effect, as if the universe as a whole were conspiring with itself to link up things that otherwise would not be linked, in the pursuance of some grand plan (or more likely a huge joke!). Maybe poetic justice is a real force in nature, it would certainly explain a lot about history if it were. Such ideas need not be so surprising.

We have seen that the long-debated 'problem' of how mind has influence over matter is spurious. Since the two are not distinct, the idea that one can affect the other should not be at all difficult to accept. We find ourselves perfectly happy to believe that our minds control matter, though. We have thoughts, and we implement those thoughts as purposeful action, and the world changes in consequence. If some portions of this superpattern can show purpose, then perhaps others can too. Much of this is mere speculation. Yet the notion of purposeful behaviour does bring up the general issue of what we mean by 'control' so before we look at feedback systems I'd like to make a few observations about this.

Perhaps the first thing to say is pretty obvious but all too frequently forgotten: cause and effect act in webs, not chains, There is no such thing as the prime cause of any particular circumstance; every cause is the effect of at least one other cause, and usually far more than one. For the want of a nail, the shoe was lost; for the want of a shoe, the horse was lost; for the want of a horse the battle was lost. But what caused the shoe to need a new nail? Why was the army short of horses? What started the battle?

When we draw up a family tree we generally show a small number of ancestors leading to a large number of descendants. Yet of course we could equally draw it the other way around, one child has two parents, four grandparents and eight great grandparents. Similarly individual causes lead to many effects, each of which causes other effects. But at the same time, every effect has many causes, each of which was the effect of other causes. The universe is one huge domino run, with many dominoes falling at once, and their paths crisscross so that the passage of one domino changes the future course of many others. To create and understand life we must remember this, and not be simplistic about cause and effect. !

Another mistake we often make is to think of control as something that is imposed from outside or from the top down. Control is really as much an effect as a cause. We shall come back to this when we look at free will, but it is worth bearing in mind that a general is controlled by his troops just as much as he controls them. In a feedback loop, like any loop, there is no start and no end.

This view that control is essentially synonymous with domination has also had consequences for the way people think about natural intelligence, and how neuroscientists interpret the workings of the human brain. Our metaphors for the operation of the brain are frequently drawn from the production line. We think of the brain as a glorified sausage machine, taking in information from the senses, processing it and regurgitating it in a different form, as thoughts or actions. The digital computer reinforces this idea because it is quite explicitly a machine that does to information what a sausage machine does to pork. Indeed, the brain was the original inspiration and metaphor for the development of the digital computer, and early computers were often described as 'giant brains'.

Unfortunately, neuroscientists have sometimes turned this analogy on its head, and based their models of brain function on the workings of the digital computer (for example by assuming that memory is separate and distinct from processing, as it is in a computer). This makes the whole metaphor dangerously self-reinforcing.

But there is a risk that by casting the brain in an active role as the processor of passive information we may be missing the point. Not all machines work in such an assertive and forceful way. Imagine a coin-sorting machine, for example (Figure 8). Such a machine might work by allowing coins to roll down a narrow slope between two vertical sheets of board. A series of holes of increasing size is cut into one of the boards, with a pocket behind each one. When coins of various denominations are rolled down the slope, each one will slide past the smaller holes but fall through the first one large enough to accommodate it. The coins enter in a random order and come out sorted by size. Yet the machine doesn't sort the coins, the coins sort themselves.

Such a simple, static device would probably not work in practice. Perhaps the coins would twist and get stuck in the smaller holes. In any case, we get a better analogy if we allow the machine to have moving parts - perhaps sprung flaps that send heavy coins one way and lighter coins another. The idea I am trying to get across is that such a machine is controlled by the coins coming into it, rather than it being the thing that controls the coins. I suspect we might get a better insight into the workings of the human brain if we take this as our metaphor, and think of the information acting upon the brain at least as much as we think of the brain acting on the information.

For example, imagine a signal from your senses entering your brain and following a certain path through all the neurones, as if it were a child running through a maze, to emerge eventually as a signal that triggers a muscle. Imagine too that the passage of each signal through the maze of neurones opens some doors and closes others, so that subsequent signals take a different route. The signals are controlling one another, using the brain as a kind of memory of their past behaviour. You shouldn't take this analogy too literally because it does have its pitfalls, but the point I'm trying to make is that control does not imply the domination of one thing over another, it is purely something that happens. The machine and the information can interact and cooperate, without either one of them being 'in control'. The image of the brain as a topdown commandandcontrol system which processes data in the way that a bureaucrat processes forms can be very misleading.

Another universe beckons

For the past few chapters I have been trying to give you a feeling for the gestalt: for form, for the whole that is greater than the sum of its parts. I have been encouraging you to take what might be an unfamiliar and perhaps disorienting view of the natural world. From this perspective, concepts that were previously clearcut and definite are now fuzzy and intertwined. Substance has been replaced by form. Chains have turned into webs, and masters have become slaves. Rather than a fixed cast of actors we see an endless Russian doll of levels of being.

Life, we discover, is a loose coalition of selfmaintaining eddies in a flowing stream. When the whole becomes more than the sum of its parts something new and perfectly real comes into existence. This gestalt is not mysterious, but neither is it a figment of our imagination. It is surprising, but that does not mean it is unknowable or that it cannot be created to order. Owen Holland and his colleagues created a small emergent phenomenon deliberately and with understanding. They didn't discover rules for making frisbee cemeteries by accident - they invented them. They did this by 'constructive tinkering', by developing a feel for how certain kinds of rules play out, and exploring these ideas in thought experiments before validating them with robots.

Frisbeepushing is very simple, but I believe that far more complex emergent phenomena can also be a product of human art, as well as a product of accident or evolution. To achieve this we need to adopt a different mode of thought from that used by the scientist, or indeed the artist. What is required is a mindset that society has tended to undervalue in the past: the mind of the inventor. We can engineer the gestalt and bring new persistent phenomena into being by design. We first need to collect natural phenomena, just as Victorian naturalists collected butterflies or beetles: we must note how rivulets form in the sand on a falling tide, how water flows out of a bath, how children inevitably get themselves into trouble. We should look into the souls of these phenomena and extract their essence. But unlike the scientist, who labels it and keeps it in a glass cabinet, we must take this essence and build new things with it. We should see how the same flows of cause and effect can be used in another context to create something new - how an idea can be stolen from here, and a clever trick from there, and folded together in a stepwise synthesis that brings something glorious into being.

Soon we shall try to identity the palette of basic colours from which the rich tapestry of the universe is woven - the basic building blocks of cause and effect. This will provide us with tools we need to create life for ourselves. But first I want to move away from our universe altogether and explore a parallel world called cyberspace because that is where our artificial life forms are going to be created.


Steve Grand :Creation





HOME      BOE     SAL     TEXTE