“First-order cybernetics” is a label applied after the fact. The founders never used it; to them it was simply cybernetics. The term came into use in the 1970s, when Heinz von Foerster distinguished the cybernetics of observed systems (first order) from the cybernetics of observing systems (second order).
First-order cybernetics treats the system as something out there. Picture an observer standing outside it, notebook in hand, identifying inputs, outputs and goals, and asking a single question: how does this system keep itself on track?
The core concepts
Negative feedback. The system measures the gap between where it is and where it wants to be, then acts to close it. Think of how you adjust the shower until the water feels right. This simple idea explains how purposeful behaviour can arise in a machine.
Homeostasis. A system keeps its essential variables (temperature, blood sugar, cash flow) within limits that allow it to survive. Ashby’s homeostat showed a machine that could find a new stable state after being disturbed.
Ultrastability. Ashby’s term for a system that, when its normal feedback fails, changes its own internal structure until it finds a configuration that works. This is a simple model of learning.
Requisite variety. Ashby’s law: a regulator can only control a system if it can produce as many distinct responses as there are disturbances to counter. “Only variety can absorb variety.” If you have ever managed a team and faced more problems than you had options, you have met this law, whether or not anyone told you its name.
The good regulator. Roger Conant and Ross Ashby proved in 1970 that “every good regulator of a system must be a model of that system.” To control something well, you need a working model of it.
The black box. You can learn how a system behaves by systematically varying inputs and watching outputs, without opening it.
Where it was used
First-order cybernetics proved its worth almost at once, in a remarkable range of places.
- Control engineering: autopilots, industrial process control, guided weapons.
- Neuroscience: models of reflexes, the brain as a feedback controller, early neural networks.
- Management: Stafford Beer’s Viable System Model, which describes the functions any organisation needs to survive.
- Economics and planning: input-output models, Soviet and Chilean attempts at cybernetic planning.
- Ecology: population dynamics and later Gaia theory.
The limits
First-order cybernetics assumes the observer can stand outside and see the system clearly. For a thermostat, that works perfectly well. For a family, a city or an economy, a few awkward questions start to appear:
- Who sets the goal? The thermostat’s set point is chosen by someone. In social systems, choosing goals is political.
- Who draws the boundary? What counts as the system and what counts as environment depends on the purpose of the person describing it.
- What happens when the system can read the description? People change behaviour when they know they are being measured.
These questions led to second-order cybernetics. First-order thinking remains alive and well, though, and for many engineering and organisational problems it is still exactly the right tool. The toolkit on this site uses it directly.
Further reading
- W. Ross Ashby, An Introduction to Cybernetics (1956)
- Roger Conant and W. Ross Ashby, “Every Good Regulator of a System Must Be a Model of That System” (1970)
- Stafford Beer, Brain of the Firm (1972)