Emergence

[curses]

I saw an episode of Nova the other day that was dealing with the concept of emergence in both organic and inorganic situations. The show explained the basic principle of emergence well, but I'm particularly interested in reading further into how it works within large groups of organisms. Would anyone happen to have a good place for a beginner to the concept to start reading?

[The Lone Ranger]

Emergence is very-much an idea rather than a developed field of inquiry. As such, it's a very new concept in some ways, and there isn't that much that has been written about the general phenomenon, to my knowledge. Rather, it's something that is studied on a "one-on-one" basis. That is, physicists study how patterns can arise from seemingly-random inputs to form things like weather systems on planets. Chemists study how combining different atoms results in the emergence of molecules with entirely different properties. Biologists study how individual organisms' behavioral patterns result in herd behavior, and so on and so on.


Stuart Kauffman has written a couple of popular books on the general subject of emergent properties, and that might be a good place to start. In The Origins of Order: Self-Organization and Selection in Evolution, Kauffman argues that self-organization may play an important role in providing the "raw material" with which natural selection works. Another of his books, At Home in the Universe: The Search for the Laws of Self-Organization and Complexity, is more general, and looks at how order often spontaneously arises from seemingly-chaotic processes.

Cheers,

Michael

[davidm]

How does subjective mind, qualia, emerge from brain states? Is qualia emergent on physical brain states? If not, what is it? :dunno:

[curses]

Thanks, TLR. That's exactly the sort of thing I was looking for. What are the advantages of studying it on a singular basis versus the whole phenomenon? I guess a better way to ask that is why hasn't it really been studied as a whole until very recently?

[The Lone Ranger]

It has only been fairly recently that people like Kauffman have been claiming that there are general "rules" that apply to emergent phenomena, and that those same rules can be applied to physical, chemical, meteorlogical, biological systems, etc. If that turns out to be correct, a very wide range of phenomena could be explained by a few relatively simple rules.

It's a fascinating field of inquiry, and by no means is everyone convinced that it really is possible to distill general rules that apply to self-organization in such widely different systems. Only time will tell, I suppose.


How does subjective mind, qualia, emerge from brain states? Is qualia emergent on physical brain states? If not, what is it? :dunno:

I must confess that I'm not entirely certain what your question is. At the moment, all we can really say is that very little is known about how "the mind" emerges from physical brain states. Neurophysiologists are pretty-much unanimous that it does, if for no other reason than that physical changes in the brain (whether by introduction of foreign chemicals, stimulation by electrodes, whatever) undeniably affect mental states, but the mechanisms by which "the mind" is generated as a function of brain state are still poorly-understood at best.

Some people (e.g. Susan Blackmore) argue that consciousness is essentially an illusion anyway -- that consciousness is simply an after the fact illusion created by the brain to "justify" our behaviors. There has been some fascinating work that seems to point in that direction (such as studies of "split brain" patients), but it's still a quite controversial claim.

Really, for all we know about the physical functioning of the brain, how it generates consciousness is still pretty-much a mystery.

Cheers,

Michael

[curses]

It has only been fairly recently that people like Kauffman have been claiming that there are general "rules" that apply to emergent phenomena, and that those same rules can be applied to physical, chemical, meteorlogical, biological systems, etc. If that turns out to be correct, a very wide range of phenomena could be explained by a few relatively simple rules.

It's a fascinating field of inquiry, and by no means is everyone convinced that it really is possible to distill general rules that apply to self-organization in such widely different systems. Only time will tell, I suppose.
Interesting. Is there anyone doing work that directly opposes the idea of emergence?

[The Lone Ranger]

Interesting. Is there anyone doing work that directly opposes the idea of emergence?

Not to my knowledge, really. I get the impression that a lot of working scientists think it's more or less a "fad" rather than a branch of investigation that has any likelihood of panning out. As such, they're content to investigate emergence in their own fields while waiting for people like Kauffman to finally realize how silly they've been.

Time will tell, I suppose.


Personally, I wouldn't be surprised to learn that there are a few relatively simple rules that govern most forms of emergent behavior in complex systems. I do think that some devotees of chaos theory behave in a somewhat "evangelical" manner (as some critics have pointed out), convinced with something approaching a fanatic's zeal that they've uncovered some deep principles that -- when fully understood -- will allow us to understand, well, just-about everything about the behavior of complex systems. I do think they tend to overstate the applicability of their discoveries.


Nonetheless, they have a point. Many complex systems display emergent properties, and some very complex-seeming behavior can be explained with just a few very simple rules. I watched a computer program that simulated the behavior of a swarm of flies once, for instance, and it had only two rules governing the behavior of the virtual flies. It was downright spooky how well it mimicked the behavior of actual flies.

For those who are interested in such things, try a web search for "Artificial Life" computer programs. The granddaddy of them all is John Conway's Game of Life (http://www.bitstorm.org/gameoflife/). What this and other "Cellular Automata" programs demonstrate is that astonishingly complex (and sometimes very orderly) behavior can arise from systems that have only a handful of very simple rules.

One such program, Tierra (http://life.ou.edu/tierra/) is actually used to study the process of biological evolution. It's an evolution simulator, and when you let it run, the digital creatures routinely "invent" sex, parasitism, etc. -- and in short, "evolve" in a way that's remarkably like the way living organisms evolve.

In Tierra, which was developed by Thomas Ray, an ecologists, computer programs compete for CPU time in much the same way that living organisms must compete for food and living space. The codes of the programs can spontaneously and randomly mutate, making some better competitors and others worse competitors. Those that are successful can replicate and so come to dominate the digital landscape. As I mentioned, when the simulation is run, the programs often behave in a remarkably lifelike manner. Different programs will evolve to swap code with each other and recombine their codes (the digital equivalent of sex); some programs will evolve to "steal" code from others, "killing" them in the process (the digital equivalent of predators attacking prey); some programs will cast off bits of their code and thus become entirely dependent upon other programs for their own self-replication (they become parasitic upon other programs).

Notice that the "rules" are very simple, and are deliberately set up to mimic what happens in living organisms. The individual programs must compete for resources (in this case, CPU time), just as individual organisms must compete for resources. Those that are unsuccessful at getting resources will "die" and won't pass on their code; those that are successful get copied and so survive into the next generation. Whenever they reproduce, there's a chance of a "mutation" occurring; that mutation may be detrimental, it may be completely neutral -- or it may be beneficial. Just like mutations in the genes of living organisms.



Fascinating stuff!



Cheers,

Michael

[JoeP]

Emergence is very-much an idea rather than a developed field of inquiry. As such, it's a very new concept in some ways, and there isn't that much that has been written about the general phenomenon, to my knowledge. Rather, it's something that is studied on a "one-on-one" basis.

...

I feel strangely disappointed. I vaguely thought of "emergence" as linked to, even arising from, work on chaos - at least, non-linear dynamical systems - that produces oddities like Feigenbaum's constant popping up in period-doubling bifurcations in all kinds of different systems.

Is there any link between studying emergence and studying chaotic dynamical systems, or should I stop drinking while reading popular science books?

[JoeP]

... I'm particularly interested in reading further into how it works within large groups of organisms.Like the membership of :ff:? :chin:

[The Lone Ranger]

I feel strangely disappointed. I vaguely thought of "emergence" as linked to, even arising from, work on chaos - at least, non-linear dynamical systems - that produces oddities like Feigenbaum's constant popping up in period-doubling bifurcations in all kinds of different systems.

Is there any link between studying emergence and studying chaotic dynamical systems, or should I stop drinking while reading popular science books?

Oh, there's a strong linkage between the concept of "emergence" and chaos theory. A good deal of chaos theory deals with the sorts of order that spontaneously emerge in systems that are "on the edge of chaos."

Chaos theory was immensely popular for a time, and its proponents made lots of grandiose claims. To listen to some of them, they were on the verge of explaining -- well -- just about everything, from the formation of galaxies to the way the fronds are distributed on the stems of ferns. But while it's true that certain mathematical relationships crop up with surprising regularity in nature, it seems that perhaps the "chaos evangelists" (as some call them) were just a mite overenthusiastic in their initial predictions.

Still, the study of how complex, orderly systems spontaneously arise is a fascinating field. That it didn't turn out to be quite so easy to predict the behavior of such systems as early theoreticians thought it might turn out to be certainly doesn't mean that there aren't general patterns that we might discover and learn to predict.


The attitude toward such notions differs between the various fields of science. Physicists and chemists have long been comfortable with the notion of "self-organizing complexity," because they see that sort of thing all the time in their work. So, the notion that complex systems can spontaneously self-organize hasn't been much of a stretch for them.

Biologists, on the other hand, tend to be wary of such ideas, since a lot of them associate such ideas (when applied to biological systems) as somehow implying that living systems are infused with some sort of elan vitale, an idea that was abandoned in the biological sciences some time ago. Some biologists have been very dismissive of Stuart Kauffman's work, thinking that he's trying to overthrow evolutionary theory. Some think that he's insisting that natural selection has nothing to do with evolution, and that biological diversity is simply another example of "self-organized complexity."

Of course, that's not what Kauffman is saying. He's simply pointing out that emergence may provide some of the variation with which natural selection works. He has also pointed out that from what we're learning of complex, self-organizing systems, given the conditions on the early Earth, the spontaneous arising of living cells may well have been inevitable.

The biologists who are most comfortable with the idea of self-organized complexity are probably ecologists. Like physicists and chemists, ecologists are quite familiar with the fact that complex systems can be self-organizing -- after all, there are few things more complex than ecosystems, and they're self-organizing. So, much of the early work in chaos theory was done by mathematically-inclined ecologists such as Robert May.


I doubt anyone would be surprised to know that much of the early work in chaos theory was done by a meteorologist, one Edward Lorenz. If anyone understands that tiny variations in the initial conditions can lead to vastly different outcomes when you're dealing with complex systems, it's a meteorologist.

Cheers,

Michael

[curses]

I checked a copy of The Origins of Order: Self-Organization and Selection in Evolution out from the library yesterday, I'm sure some more questions will be forthcoming.

[But]

I don't understand why there should be anything controversial about the idea that all complex behaviour in the universe arises from very simple rules. Everything that is observable is created by quantum theories of the fundamental forces and their mathematical relationships.

[JoeP]

:burns:
I shall continue drinking and reading popular science books.

[JoeP]

I doubt anyone would be surprised to know that much of the early work in chaos theory was done by a meteorologist, one Edward Lorenz.
Count me surprised. I had no idea he was a meteorologist. I vaguely assumed the Lorenz (http://en.wikipedia.org/wiki/Edward_Lorenz) of the attractor (http://en.wikipedia.org/wiki/Lorenz_attractor) was the Loren[t]z (http://en.wikipedia.org/wiki/Hendrik_Lorentz) of the relativistic contraction (http://en.wikipedia.org/wiki/Lorenz_contraction). :blush:

[JoeP]

I don't understand why there should be anything controversial about the idea that all complex behaviour in the universe arises from very simple rules. Everything that is observable is created by quantum theories of the fundamental forces and their mathematical relationships.

But ... But ... sure, we can see that complex systems have arisen. The issue, if we can call it that, is whether the fact that complex behaviour arises from simple rules should be expected, or whether it implies a hidden hand, a designer.

Religious aspects aside, humans are wired to find intention in order, are we not? It's that wiring that leads to surprise that order could arise without intention.

[But]

In that case, I have a solution that makes everybody happy:

1/sqrt(2) * ( |there is a god> + |there is no god> )

[The Lone Ranger]

I don't understand why there should be anything controversial about the idea that all complex behaviour in the universe arises from very simple rules. Everything that is observable is created by quantum theories of the fundamental forces and their mathematical relationships.

It's not so much that it's surprising that complex behavior arises from very simple rules as that the same rules might apply to lots of different and apparently unrelated systems. That's something that a lot of people are still finding difficult to believe. And indeed, it seems that the early adherent of Chaos Theory may have been a bit premature in their hope that it would soon be possible to explain all sorts of complex emergent behaviors with just a few very general rules.



Religious aspects aside, humans are wired to find intention in order, are we not? It's that wiring that leads to surprise that order could arise without intention.
There's something to that. Many people have pointed out that humans do seem to have an innate tendency to see "intention" in order. Think of all the people who insist that a perfectly ordinary mesa on Mars was deliberately sculpted by aliens to look like a face, for instance. And, of course, Creationists simply refuse to accept that all you need for adaptive evolution to occur is 1.) competition, 2.) reproduction, 3.) heredity, and 4.) the occasional mutation that affects fitness.

Regardless, it is somewhat surprising to a lot of people that highly complex and orderly systems can and do spontaneously arise from chaos.

Cheers,

Michael

[BDS]

It's not so much that it's surprising that complex behavior arises from very simple rules as that the same rules might apply to lots of different and apparently unrelated systems. That's something that a lot of people are still finding difficult to believe. And indeed, it seems that the early adherent of Chaos Theory may have been a bit premature in their hope that it would soon be possible to explain all sorts of complex emergent behaviors with just a few very general rules.


That's the problem -- it's not that complex, emergent behaviors will NEVER be explicable by a few general rules. It's just that WE don't know how to do it.

Reductionist explanations (the opposite of "emergent explanations") are very attractive -- but often simplistic in terms of their explanatory or predictive value. A classic example would be a biological "explanation" for incest taboos. True -- when close relatives have children there is slightly increased chance of recessive-trait birth defects. However, incest taboos vary from culture to culture; often a person is prohibited from marrying one type of relative who is equally closely related to that person by blood as another relative whom one MUST marry.

The more accurate "explanation" only "emerges" when one looks at the culture as a whole.

However, this is not to say that we can't eventually explain or predict everything in terms of physics. We just can't do it yet.