First-order cybernetics

Term from Heinz von Foerster, 1970s, describing work from the 1940s on

The cybernetics of observed systems: the original programme of the 1940s and 1950s, which studied feedback, control and communication in systems treated as if seen from outside.

First-order cybernetics is the name given, in retrospect, to the cybernetics of the 1940s and 1950s: the work of Norbert Wiener, W. Ross Ashby, Warren McCulloch, Claude Shannon and the Macy Conferences. The term was introduced by Heinz von Foerster in the 1970s to contrast it with second-order cybernetics.

What it studied

First-order cybernetics treated systems (thermostats, guided missiles, nervous systems, organisations) as objects to be described and controlled by an observer standing outside them. Its core ideas were feedback, information, regulation, stability and adaptation. Its typical tools were mathematical models, servomechanisms and machines such as Ashby’s homeostat and W. Grey Walter’s tortoises.

Legacy

Much of first-order cybernetics passed into engineering, computing and the sciences under other names: control theory, information theory, operations research, artificial intelligence and neural networks. Its picture of the observer outside the system is the one second-order cybernetics set out to revise. The two are best seen as complementary: second-order cybernetics keeps first-order models and asks who builds them, for what purpose, and with what consequences.

Further reading

  • Norbert Wiener, Cybernetics (1948)
  • W. Ross Ashby, An Introduction to Cybernetics (1956)

Mentioned in