Cybernetics is often encountered through its descendants before the field itself is understood. The word can suggest robots, artificial intelligence, command systems, or an abstract theory in which everything becomes a network. Those associations are not entirely accidental, but they are poor starting points. At its core, cybernetics asks a more practical question: how does a system continue to steer when the world refuses to stay still?
A thermostat is the familiar example because it makes the basic loop visible. A target is established. The present condition is sensed. The difference between the two matters. Action changes the condition, the result is sensed again, and the next action depends on what happened. The important feature is not the device. It is the recursion. Output returns as information that can alter future output.
Feedback is not merely response
Ordinary language often uses feedback to mean a comment delivered after the fact: praise, criticism, a rating, a review. Cybernetically, feedback is more demanding. Information about the consequences of action must re-enter the process in a way that can affect what happens next. A system that acts, receives consequences, and remains structurally unable to change is not regulating itself in any meaningful sense.
This is why error becomes useful. Error is not a moral category. It is discrepancy: the difference between a current condition and a condition that matters to the system. A boat drifting off course, a body moving outside a viable range, a practice producing the opposite of its stated aim - each creates information. The quality of regulation depends partly on whether that information can be detected and partly on whether the system has enough possible responses to do anything with it.
Control is not the same as domination
The word control creates another predictable misunderstanding. In ordinary speech, control easily becomes synonymous with force: one thing imposing its will on another. Cybernetic control is usually subtler. It is closer to steering. The point is not to eliminate disturbance but to remain viable through disturbance. A skilled driver does not command the road to become straight. A living organism does not demand that the environment stop changing. Regulation is the continuing work of maintaining a workable relation under conditions that cannot be completely controlled.
That distinction matters for human systems because fantasies of perfect control tend to destroy the very adaptability that real control requires. If the environment presents more variation than the regulator can recognize or answer, regulation fails. This is the practical force behind Ashby's concern with variety: a regulator needs enough range to meet the range of conditions that matter. Rigidity can look like control right up until the world changes.
Stability and adaptation are different achievements
Homeostasis is sometimes mistaken for stillness. It is better understood as dynamic stability. A body maintains temperature through continual adjustment; balance exists because change is happening, not because change has stopped. But there are situations in which preserving the old pattern is no longer enough. Adaptation requires a deeper revision - a change in strategy, organization, or response capacity because the previous form of regulation no longer fits the conditions.
This distinction is one reason cybernetics becomes useful far beyond machines. Human practices, institutions, habits, and environments all contain loops in which action changes conditions and conditions change future action. The analogy has limits, and those limits matter. People are not thermostats. Meaning, interpretation, embodiment, identity, and social relation make human regulation far more complex. But the cybernetic question survives the increase in complexity: what information returns, what difference is detected, what can change, and what happens when the regulator itself becomes part of the problem?
That is the value of cybernetics in this book: not a total explanation of human life, and not a machine metaphor imposed on experience, but a disciplined language for examining how direction survives contact with consequence.