Re: Emergence Seminar--British Emergence
Posted by
Russ Abbott on
Sep 15, 2009; 4:18am
URL: http://friam.383.s1.nabble.com/Emergence-Seminar-British-Emergence-tp3645669p3646966.html
Owen,
Here's how I would start.
I'm not scientist enough to know what 'configuration physics' or 'configuration chemistry' means. My guess is that it means something like a structured collection of matter where the structure itself is important. One of my friends likes to talk about that sort of thing as global constraints. I think that's a fine way of expressing it, when one understands global as referring to the entity being structured and not the world at large.
I would nominate that concept--i.e., the ability to create a structured entity from unstructured components--as the commonality among "emergent" phenomena. (That's why I like the notion of level of abstraction as representative of emergence.)
That raises a few questions.
- What are the possible "binding forces" that can be used to create structure? (My answer is that there are two categories of binding forces: static and dynamic. The static ones are the forces of physics. They produce emergent phenomena like chemistry as Roger said. The dynamic ones are much more open and depend on the entities being organized. They produce emergent phenomena like biological and social entities.)
- How do those binding forces work? (My answer is that the static ones work according to the laws of physics. For the dynamic ones it is much more difficult to find a useful generalization since again it depends on the entities being structured.)
- Where does the energy come from that powers those forces. (My answer is that for static forces, the energy is standard physics. Static entities exist at equilibrium in energy wells. For dynamic entities the energy is continually imported from outside. That's why they are "far from equilibrium." They must import energy to keep themselves together.)
- Finally, what holds levels of abstraction together within software? (My answer is that software is subsidized. It runs without having to worry about the energy it uses. Consequently software confuses us because it hides the energy issue. One can build anything one can think of in software using the mechanisms for construction built into (and on top of) the programming language one is using.)
-- Russ
On Mon, Sep 14, 2009 at 8:43 PM, Owen Densmore
<[hidden email]> wrote:
[This is an email I sent to the reading group. It's title was:
Emergence, Chaos Envy, and Formalization of Complexity
I think that, rather than worrying about the existing concepts of emergence, we would be far better off looking at the history of Chaos and how they achieved amazing results in a short time, and how we could similarly attempt formalization of complexity. One idea is to simply look at the "edge of chaos" idea in more detail, thus placing complexity as a field within chaos.]
Nick has started a seminar on Emergence based on the book of that name by Bedau and Humphreys. This got me to thinking about the core problem of Complexity: its lack of a core definition, along with lack of any success in formalizing it.
Chaos found itself in a similar position: the Lorenz equations for very simple weather modeling had quirks which were difficult to grasp. Years passed with many arguing that Lorenz was a dummy: he didn't understand error calculations, nor did he understand the limitations of computation.
Many folks sided with Lorenz, siting similar phenomena such as turbulent flow, the logistics map, and the three body problem. All had one thing in common: divergence. I.e. two points near each other would find themselves at a near random distance from each other after short periods of time.
See: http://en.wikipedia.org/wiki/Chaos_theory
Complexity similarly arose from observations such as sand-pile formation, flocking, ant foraging, and so on. Their commonality, however, was not divergence but convergence, not chaos but order. Typically this is coined "emergence".
I would like to propose an attempt to do what Poincare, Feigenbaum, Layapunov and others have done for Chaos, but for Complexity.
Nick has hit the nail on the head, I think, in choosing Emergence as the core similarity across the spectrum of phenomena we call "complex".
The success of Chaos was to find a few, very constrained areas of divergence and formalize them into a mathematical framework. Initial success brought the Rosetta stone: the Lyapunov exponent: a scalar metric for identifying chaotic systems.
It seems to me that a goal of understanding emergence is to formalize it, hoping for the same result Chaos had. I'd be fine limiting our scope to ABM, simply because it has a hope of being bounded .. thus simple enough for success.
You see why I included Chaos Envy?
-- Owen
============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
lectures, archives, unsubscribe, maps at
http://www.friam.org
============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
lectures, archives, unsubscribe, maps at
http://www.friam.org