Biology Reference
In-Depth Information
work on 'Principles of the Self-Organizing Dynamic System' in 1947, served as
major stimuli for the rash of conferences on self-organizing systems organized
by the Office of Naval Research (ONR) in the 1950s. All of these discussions
and investigations were conducted with an explicitly engineering aim, namely,
the design and construction of systems that could organize themselves, grow
themselves, and perhaps even reproduce themselves.
1
As Ashby and von Foerster
formulated it, the crucial property for the realization of these goals lay in the
relations between components; out of such relations functionality or purpose
would spontaneously emerge. Frank Rosenblatt's
Perceptron
project embodied
this vision in its most ambitious form, but despite enthusiastic support in its
early stages from the ONR, development of the Perceptron did not continue
long enough or far enough to produce anything like a live issue. By the 1960s,
support had dried up and the spirit of artificial intelligence, organized around
very different principles, had moved elsewhere. Some of Ashby's arguments
survived, but in other areas of engineering - most notably, in control theory.
The next important mutation in this history comes with the emergence of
studies of nonlinear dynamical systems in physics and mathematics in the late
1970s and 1980s. Here again, the term self-organization surfaces, but now
with a further shift in meaning. For Ilya Prigogine, the term referred to the
emergence of 'dissipative structures' in systems far from equilibrium and low
in entropy. Given a sufficiently large flux of free energy in such systems,
one can expect to see the spontaneous emergence of such striking phenomena
as eddies, vortices, or Bénard rolls and cells. At roughly the same time, the
study of nonlinear dynamical systems by mathematicians was yielding similar
insights about these same phenomena, although here they were described in
terms of stable attractors and limit cycles. Either way, self-organization now
referred to the production of stable patterns observed in physical, and sometimes
in biological, systems governed by nonlinear dynamics. Enthusiasm for such
analyses ran high, generating some extraordinary expectations. As Paul Davies
wrote, 'Mathematically we can now see how nonlinearity in far-from-equilibrium
systems can induce matter [here quoting Charles Bennett (1986)] to “transcend
the clod-like nature it would manifest at equilibrium, and behave instead in
dramatic and unforeseen ways, molding itself for example into thunderstorms,
people and umbrellas” '. (1989, pp. 111)
The late 1980s brought a noteworthy (and indeed, much noted) addition to
these ideas, namely, Per Bak's notion of 'self-organized criticality' (SOC) (Bak
et al., 1987).
2
Bak and Sneppen (1993) developed a model for evolutionary
1
In a 1958 article entitled 'Electronic “Brain” Teaches Itself ', Rosenblatt is quoted as saying, 'In principle, it
would be possible to build Perceptrons that could reproduce themselves on an assembly line and which would
be “conscious” of their existence.' (New York Times, 13 July 1958, Section 4, p. 9).
2
In 1996, Bak reported that, since the coining of the phrase SOC in 1987, 'more than 2000 papers have been
written on the subject, making' this initial paper 'the most cited in physics ' (Bak, 1995).
