JOVANA
Library Glossary Getting Started Three Levels Fields How it works Mission
Join the mission
Back to the library
Cybernetics 1948

Cybernetics: or Control and Communication in the Animal and the Machine

Norbert Wiener

Machine, animal, and mind all steer by feeding their results back as causes.

Choose your version
In depth · the introduction

A ship's steersman doesn't plot the perfect course once and walk away — he watches, sees how far he's drifted, and nudges the wheel, again and again. Wiener saw that machines, animals, and even societies all steer themselves the same way.

The idea, unpacked

At the heart of the book is one loop, called feedback. A system measures how far it is from where it wants to be, and uses that gap to decide its next move — then measures again, and corrects again. A thermostat does it with temperature; your hand does it with the distance to a cup; a government does it with the gap between a policy and its result.

Wiener's radical claim was that this loop — not the flesh or the metal it's made of — is what produces purposeful behaviour. A machine that corrects itself toward a goal is doing the same thing as a living creature reaching for one. From that single idea he built a science, and named it cybernetics, from the Greek word for "steersman."

Where it came from

Norbert Wiener was a former child prodigy and an MIT mathematician. During the Second World War he worked on aiming anti-aircraft guns at planes that dodged and weaved — which meant predicting where a moving, self-correcting target would go next. That problem, of a system reacting to its own past, never let him go.

Afterwards he pulled the threads together with two collaborators — the Mexican physiologist Arturo Rosenblueth and the engineer Julian Bigelow — with much of the thinking done in Mexico City. In 1948 he published the result as a difficult, wide-ranging book, and gave the new field its name. To his own surprise, it became a bestseller.

Why it mattered

Before Wiener, an engineer building a controller, a physiologist studying a reflex, and a telephone engineer fighting noise on a line had no shared language. Cybernetics gave them one — feedback, signal, noise, information — and showed their problems were versions of the same problem. It also delivered a hard practical lesson: feedback that corrects too aggressively, or acts on information that arrived too late, doesn't merely fail — it swings into wild oscillation. Wiener noticed that this is exactly what a damaged nervous system does when a reaching hand shakes around its target.

A shower you can't get right

Think of an unfamiliar shower. The water's too cold, so you crank the tap hot; but the pipe is long, so nothing changes — you crank it further; then a wall of scalding water arrives, so you slam it back, and a moment later you're freezing again. You're a feedback loop with too much gain and too much delay, hunting back and forth and never settling. Tame the delay and ease your corrections, and you glide smoothly to the right temperature. That swing between overshoot and calm is the whole drama of the tool below.

A plot of a feedback loop steering toward a target line. Two sliders set the correction strength (gain) and the feedback delay. With gentle gain and no delay the curve rises smoothly and settles on the target; raise the gain or add delay and it overshoots, oscillates, and finally swings wider and wider out of control.

Where it sits

1948 was a founding year for the information age. The same year, Claude Shannon published his theory of information (also in this Library), and three years earlier John von Neumann had set down the design of the stored-program computer. Cybernetics was the third pillar — the one about control and purpose. Its name eventually faded as its children grew up — control theory, AI, robotics, the whole study of self-regulating systems — but the prefix "cyber-" in words like cyberspace still carries the Greek steersman inside it.

The original document
Original source text
Norbert Wiener · The Technology Press (MIT) & John Wiley / Hermann & Cie, Paris · 1948
Introduction · naming the field
We have decided to call the entire field of control and communication theory, whether in the machine or in the animal, by the name Cybernetics, which we form from the Greek κυβερνήτης or steersman.
Wiener chose the word to honour its ancestry: the first significant paper on feedback mechanisms was James Clerk Maxwell's 1868 study of the governor, and "governor" is itself a Latin corruption of the same Greek root, while the steering engines of a ship were among the earliest and best-developed feedback devices.
On the boundary between sciences
the most fruitful areas for the growth of the sciences were those which had been neglected as a no-man's land between the various established fields.
The book grew out of Wiener's wartime work on predicting the motion of a manoeuvring target, and out of years of conversation with the physiologist Arturo Rosenblueth; its method is to treat control engineering, neurophysiology, and communication as one subject.
Information and entropy
Just as the amount of information in a system is a measure of its degree of organization, so the entropy of a system is a measure of its degree of disorganization; and the one is simply the negative of the other.
From this identification Wiener develops messages, noise, filtering, and prediction as a single theory, and turns to feedback as the mechanism by which a system uses such information to govern its own behaviour — stable when its corrections are measured, oscillatory and finally unstable when they are too strong or arrive too late.
[ … ]
The full book runs to eight chapters — on time series, groups and statistical mechanics, feedback and oscillation, computing machines and the nervous system, Gestalt and the universals, and information, language and society — and is available complete at the source below.
Norbert Wiener · Cambridge, Massachusetts · 1948