Confusing Emergence Causing Emergent Confusion

It might be just an urban legend, but I have read that it was Charles Darwin with his Origin of Species, who has inspired Ludwig Boltzmann, to approach the study of gases as populations of molecules, using statistics to show some emergent or high-level properties and variables that these systems exhibit.

When Dan Dennett writes about Darwin’s Dangerous Idea, he argues, that some interesting phenomena may arise, evolve, emerge, develop from interacting simple processes, without the intervention of some higher order intelligence. We have already seen in Chaos game, that we are able to consistently draw IFS fractal pictures using random choice in the process, Darwin shows us how the species have evolved from simple organisms, other biologists show us, how life evolved from organic matter. Denett speaks even of universal acid, that evolution is can be thought of as a universal process, a  universal acid that “eats away” things that Occam’s Razor would have cut and sometimes eats away even the container, meaning that it changes our frame of thinking, our view of the specific phenomenon.

If this emergence thing is something universal, then we must have some pretty good definition, something that can explain in sufficient detail, so that we can unambiguosly explain it, or perhaps even design several “emergence detectors” at least for simple and very specific cases.

Nope, not so.  Well most definitions start from assumption that we recognize an emergence, when we see one, and then lament on its properties, relationships between parts and whole, upward and downward causations and other pretty nebulous terms disguised as a theory. The current definition shows emergence as a property of  self-organizing systems: “the arising of novel and coherent structures, patterns and properties during the process of self-organization in complex systems“, although physicists could show emergent properties of gases, like pressure and temperature without referring to self-organization, just by reducing the all the unnecessary variables and so increasing just the predictive efficiency, without referring to self-organization.  For the more interesing cases of the emergence, tha self-organization seems to be the key definition. Physicists, including the ones working in the field established by Boltzmann, are trained to be preciser, less ambiguous and even to exorcise the demons of external influences. They say that we can speak of self-organization, when we have a (relatively isolated) system, when left to itself (without external stimulation),  its statistical complexity increases with time. If we could exclude external causes of st. complexity increase, this is not just necessary but a sufficient condition for self-organization. And using the above definition, we can then really speak of emergence.

It is interesting though, that the complexity measure of self-organizing system increases, meaning, that we need more information to describe such a system, after it has self-organized itself. We speak of organization, which normally means increasing order, but here paradoxically this “new order” requires a more complex description than the previous “disorder”.

What we have discussed here is what is also being called weak emergence. Some say, that there is also strong emergence, which cannot be directly explained using only the behavior of the system on a micro level. Well as most systems are not isolated from other influences, this might be just a problem of scope, of a view. Stronger emergence might be just an expansion in scope of the observed system, where external influences interact and play an important role in the non-linear process of formation of a stable observed phenomenon. At the moment I cannot think of anything, neither have I read of any strongly emergent property, that would require some additional magic apart from the scope extension.

Darwin’s idea is dangerous, its scope is much broader than the evolution of living organisms, and even traditions, customs, societies, or perhaps Chinese restaurants, it spreads as I have tried to describe here into physics, computation (also with genetic algorithms). Understanding all of its implications might require sometimes to abandon some historical concepts, even those part of common sense and base the new understanding on science. And yet there is room for magic, for insight, for particular view, for choosing just the right scope, for something to emerge, to be able to see, to grasp, to understand. Yes, it hides in the point of view, in the look, in the gaze.

I really liked this illustration of Douglas Hofstadter in his Gödel, Escher, Bach: